S. Hrg. 104-462

EARNED INCOME TAX CREDIT

Y 4. S 74/9: S. HRG. 104-462

Earned Incone Tax Credit. S.Hrg. 10..

HEARINGS

BEFORE THE

COMMITTEE ON GOVERNMENTAL AFFAIRS UNITED STATES SENATE

ONE HUNDRED FOURTH CONGRESS

FIRST SESSION

APRIL 4, 1995 ADMINISTRATION OF THE EARNED INCOME TAX CREDIT

APRIL 5, 1995 DESIGN AND EFFECTIVENESS OF THE EARNED INCOME TAX CREDIT

Printed for the use of the Committee on Governmental Affairs

U.S. GOVERNMENT

WASHINGTON

/ 8n '"«*

°St°^uhr

)N : 1996 "'**«s*^2> D$t

For sale by the U.S. Government Printing Office

Superintendent of Documents, Congressional Sales Office, Washington, DC 20402

ISBN 0-16-052749-X

S. Hrg. 104-462

EARNED INCOME TAX CREDIT

Y 4. G 74/9: S. HRG. 104-462

Earned Incone Tax Credit. S.Hrg. 10..

HEARINGS

BEFORE THE

COMMITTEE ON GOVERNMENTAL AFFAIRS UNITED STATES SENATE

ONE HUNDRED FOURTH CONGRESS

FIRST SESSION

APRIL 4, 1995 ADMINISTRATION OF THE EARNED INCOME TAX CREDIT

APRIL 5, 1995 DESIGN AND EFFECTIVENESS OF THE EARNED INCOME TAX CREDIT

Printed for the use of the Committee on Governmental Affairs

U.S. GOVERNMENT PRINTT^5*»K£iCTrOC'^//^ ^C?fj/ WASHINGTON : 1996

For sale by the U.S. Government Printing Office

Superintendent of Documents, Congressional Sales Office, Washington, DC 20402

ISBN 0-16-052749-X

COMMITTEE ON GOVERNMENTAL AFFAIRS

WILLIAM V. ROTH, Jr., Delaware, Chairman TED STEVENS, Alaska SAM NUNN, Georgia

WILLIAM S. COHEN, Maine JOHN GLENN, Ohio

FRED THOMPSON, Tennessee CARL LEVIN, Michigan

THAD COCHRAN, Mississippi DAVID PRYOR, Arkansas

CHARLES E. GRASSLEY, Iowa JOSEPH I. LIEBERMAN, Connecticut

JOHN McCAIN, Arizona DANIEL K. AKAKA, Hawaii

BOB SMITH, New Hampshire BYRON L. DORGAN, North Dakota

Franklin G. Polk, Staff Director and Chief Counsel Mark S. Mullet, Professional Staff Leonard Weiss, Minority Staff Director Michal Sue Prosser, Chief Clerk

(II)

CONTENTS

Opening statements: Page

Senator Roth 1, 65

Senator Glenn 4, 67

WITNESSES

Tuesday, April 4, 1995

Hon. Margaret Milner Richardson, Commissioner, Internal Revenue Service; accompanied by Phil Brand, Chief Compliance Officer; Ted F. Brown, Re- fund Fraud Executive; Michael P. Dolan, Deputy Commissioner; and Judy Van Alfen, Chief, Taxpayer Service 15

Lynda D. Willis, Associate Director for Tax Policy and Administration, U.S. General Accounting Office; accompanied by Michael Brostek, Assistant Di- rector, Tax Policy and Administration Issues; and Jennie Stathis, Director, Tax Policy and Administration Issues 34

Richard M. Hersch, Former President, Quik Tax Dollars, Inc 42

Dan Stein, Executive Director, Federation for American Immigration Reform (FAIR) .' 49

Daniel B. Grunberg, Ph.D., Vice President of Technology, Jackson-Hewitt and Associates 52

Finn M. W. Caspersen, Chairman of the Board, Beneficial Corporation 54

Wednesday, April 5, 1995

Hon. Don Nickles, U.S. Senator from the State of Oklahoma 76

Hon. Leslie Samuels, Assistant Secretary for Tax Policy, U.S. Department

of the Treasury 85

George Yin, Professor of Law, University of Virginia School of Law 102

Donald R. Huston, Tax Preparer, Dover, DE .' 104

Deborah Walker, Chair, Tax Executive Committee, American Institute of

Certified Public Accountants 107

Ismael Gonzalez, Esquire, (former IRS Attorney who now represents tax

fraud defendants) 110

Marvin H. Kosters, Resident Scholar, American Enterprise Institute 115

John Karl Scholz, Department of Economics, La Follette Institute of Public Affairs, and the Institute for Research on Poverty, University of Wisconsin- Madison 117

Robert Greenstein, Executive Director, Center on Budget and Policy Prior- ities 120

Alphabetical List of Witnesses

Caspersen, Finn M. W.:

Testimony 54

Prepared statement 152

Gonzalez, Ismael:

Testimony 110

Prepared statement : 263

Greenstein, Robert:

Testimony 120

Prepared statement 289

Grunberg, Daniel B.:

Testimony 52

Prepared statement 149

(III)

IV

Page

Hersch, Richard M.:

Testimony 42

Prepared statement 141

Huston, Donald R.:

Testimony 104

Prepared statement 248

Kosters, Marvin H.:

Testimony 115

Prepared statement 268

Nickles, Hon. Don:

Testimony 76

Prepared statement 206

Richardson, Hon. Margaret Milner:

Testimony 15

Prepared statement 131

Samuels, Hon. Leslie:

Testimony 85

Prepared statement (Apr. 4, 1995) 210

Prepared statement (Apr. 5, 1995) 228

Scholz, John Karl:

Testimony 117

Prepared statement 277

Stein, Dan:

Testimony 49

Prepared statement 143

Walker, Deborah:

Testimony 107

Prepared statement 251

Willis, Lynda D.:

Testimony 34

Prepared statement 136

Yin, George:

Testimony 102

Prepared statement 236

APPENDIX

Prepared statements of witnesses in order of appearance 131

GAO Report entitled "Earned Income Credit, Targeting to the Working Poor,"

GAO/GGD-95-122BR, March 1995 164

Articles from the Omaha World Herald submitted by Mr. Stein 148

Questions from Senator Glenn for GAO 311

Questions from Senator Glenn for IRS 312

ADMINISTRATION OF THE EARNED INCOME TAX CREDIT

TUESDAY, APRIL 4, 1995

U.S. Senate, Committee on Governmental Affairs,

Washington, DC. The Committee met, pursuant to notice, at 10:07 a.m., in room SD-342, Dirksen Senate Office Building, Hon. William V. Roth, Chairman of the Committee, presiding. Present: Senators Roth and Glenn.

OPENING STATEMENT OF SENATOR ROTH

Chairman Roth. The Committee will please be in order.

Today, our Committee is meeting to discuss the earned income tax credit. This Committee, as part of its mandate, is charged with searching out waste, fraud, and abuse within the Federal Govern- ment. Last year, we held a hearing on "High Risk and Emerging Fraud in the IRS." We learned a great deal about fraud and waste in the IRS tax-filing system.

GAO has indicated to me that something more than 95 percent of the dollars involved in electronic filing fraud in the tax filing system relates to the earned income tax credit. The GAO has also reported consistent IRS figures over the past 13 years showing fraud and error rates in the following amounts: 1982, 29 percent; 1985, 39 percent; 1988, 34 percent; and in 1994, 24 percent. These fraud and error rates show a consistent pattern of lost taxpayer money, and I believe it is important for this Committee to address this issue.

In May 1993, I wrote to IRS Commissioner Richardson, asking for more detailed information on the earned income tax credit. A constituent of mine, who also is a tax preparer, had written me, saying that he had seen massive fraud regarding the earned in- come tax credit. In the IRS letter to me, it was stated that their estimate, the IRS's estimate, of non-compliance in the earned in- come tax credit area was between 30 and 40 percent nationally.

Knowing that the earned income tax credit is the fastest growing entitlement in the Federal Government and that it will grow more than five-fold in real terms between 1988 and 1996, it is clear that the potential risk to the American taxpayer in lost revenue from this program has grown dramatically since 1988. I think the chart up there shows very well the growth rate of this program from 1975, where the number of families involved was roughly a million or so, it grows to almost 19 million families by the end of the pe-

(l)

riod, 1996. And, of course, likewise, the dollar amount has gone up very substantially.

We learned that the program grew five-fold, in real terms, be- tween 1988 and 1996. As I said, the potential risk to the American taxpayer in lost revenue from this program has, indeed, grown dra- matically. If one takes this 30 to 40 percent error and fraud rate times the roughly $120 billion we will be spending on this program in the next 5 years, there is a potential loss in this period of 5 years of about $40 billion, so these are, indeed, very huge sums that we are talking about.

As a result of these findings, I asked the General Accounting Of- fice to conduct a comprehensive investigation of the earned income tax credit. In addition to the fraud and error rate problems, we began to realize that much of the money spent in the EITC pro- gram seems to be going to middle-class Americans, as opposed to its intended beneficiaries, low-income working class Americans. In fact, middle-class Americans earning as much as $28,000 a year will still be eligible for some earned income tax credit next year.

Despite great hopes that the program was encouraging work, we learned that much of the program actually discourages work. In fact, there is little work requirement at all in the EITC, but only an earnings test. In other words, while one must have earned in- come to be eligible for the EITC, one need not work a specific num- ber of days. So, if in 1994, for example, one worked a single day and earned $5,000 as an entertainer, and that was all one earned that year, one would still be eligible for the EITC, if he or she met the other requirements.

Finally, before I introduce our witness, I was having breakfast down in Virginia and I was amazed to pick up the Roanoke Vir- ginia Times and find this article about "Inmates Profit in Prison". It turns out that the story is about a man, Jimmy Blankenship, who concocted false tax returns and wage statements and is now serving time in the Bland Correctional Center. But after he went to prison, he actually expanded his operation and started recruiting fellow inmates. He created fake Form 1040s, showing that the pris- oners were working when they were actually in prison, and the IRS actually sent them refunds of anywhere from $2,000 to $4,200. Ac- cording to the newspaper article, Jim is called the "H&R Cell- block".

It is with this background, I believe, that the Committee must undertake these 2 days of hearings on the earned income tax credit to determine the extent of fraud, waste, and abuse and learn more about the targeting and the success of this program in meeting its worthy and lofty goals of encouraging work and lifting working Americans out of poverty.

[The prepared statement of Senator Roth follows:!

PREPARED STATEMENT OF SENATOR ROTH— NEWS RELEASE

ROTH PUTS EITC UNDER MICROSCOPE; FINDS BILLIONS OF DOLLARS LOST TO FRAUD

Clinton's Much-Touted Tax Credit Program is Boon for Rip-Off Artists

WASHINGTON— Governmental Affairs Committee Chairman William V. Roth, Jr. (R-DE) Tuesday began two days of hearings looking into the problems of waste

and fraud in the Earned Income Tax Credit program, the fastest growing income transfer program in the Federal Government. His opening statement follows:

Today the Committee on Governmental Affairs is meeting to discuss the earned income tax credit. This Committee, as part of its mandate, is charged with searching out fraud, waste and abuse within the Federal Government. Last year, this Commit- tee held a hearing on "High Risks and Emerging Fraud in the IRS." We learned a great deal about fraud and waste in the IRS tax filing system. GAO has indicated to me that something more than 95% of the dollars involved in electronic filing fraud in the tax filing system relates to the earned income tax credit.

The GAO has also reported consistent IRS figures over the past 13 years showing fraud and error rates in the following amounts:

1982 29%

1985 39%

1988 34%

1994 24%

These fraud and error rates show a consistent pattern of lost taxpayer moneys, and it is important for this Committee to address this issue.

In May of 1993 I wrote to IRS Commissioner Richardson asking for more detailed information on the earned income tax credit. A constituent of mine — a tax pre- parer— indicated he had seen massive fraud regarding the earned income tax credit. The IRS letter to me stated that their estimate of non-compliance in the earned in- come tax credit area was "between 30% and 40% nationally." Knowing that the earned income tax credit is the fastest growing entitlement in the Federal Govern- ment, and that it will grow more than five-fold in real terms between 1988 and 1996, it is clear that the potential risk to the American taxpayer in lost revenue from this program has grown dramatically since 1988. One could take this 30% to 40% error and fraud rate times the roughly $120 billion we will be spending on this program in the next 5 years, and arrive at a potential loss to the taxpayers of about $40 billion.

As a result of these findings, I asked the General Accounting Office (GAO) to con- duct a comprehensive investigation of the earned income tax credit. In addition to the fraud and error rate problems, we began to realize that much of the money spent in the EITC program seemed to be going to "middle-class" Americans, as op- posed to its intended beneficiaries — "low-income working class" Americans. In fact, middle-class Americans earning as much as $28,000 will still be eligible for the earned income tax credit next year.

And despite great hopes that the program was encouraging work, we learned that much of the program actually discourages work, and in fact, there is little work re- quirement at all in the EITC, but only an earnings test. In other words, while one must have earned income to be eligible for the EITC, one need not work a specific number of hours. So, if in 1994 one worked a single day and earned $5,000 as an entertainer, one would still be eligible for the EITC, if he met the other require- ments.

Finally, before I introduce our witness, let me bring to the Committee's attention yesterday's headline in the Roanoke Virginia Times. It reads "Inmates Profit in Pris- on; But IRS Takes Dim View of Tax-Fraud Scheme." The story is about a man, Jimmy Blankenship, who concocted false tax returns and wage statements, and is now serving time in Bland Correctional Center. After he went to prison, he actually expanded his operation, and started recruiting fellow inmates. He created fake Form 1040s showing that the prisoners were working, when they were actually in prison. The IRS actually sent them refunds of anywhere from $2,000 to $4,200. They call him "H&R Cellblock."

So, it is with this background that I believe this Committee must undertake these two days of hearings on the earned income tax credit to determine the extent of fraud, waste and abuse, and to learn more about the "targeting" and the success of this program in meeting its worthy and lofty goals of encouraging work and lift- ing working Americans out of poverty.

Senator Roth. It is now my pleasure to introduce our first speak- er

Senator Glenn. I have an opening statement.

Chairman ROTH. Yes, the minority does have rights. [Laughter.]

Senator Glenn. We have wondered occasionally, this year. [Laughter.]

Chairman Roth. Senator Glenn.

OPENING STATEMENT OF SENATOR GLENN

Senator Glenn. Thank you, Mr. Chairman.

I am glad we are having this hearing this morning, and I would note the sub-headline on that, the smaller headline on that article says, "But IRS Takes Dim View of Tax Fraud Scheme", which I would presume would be our case.

Chairman ROTH. And true of our first witness.

Ms. Richardson. Absolutely.

Senator Glenn. The earned income tax credit is one of the most important programs the IRS oversees. The less fortunate in our so- ciety too often find themselves fighting to feed their families or pay their bills and stay off public assistance. This helps those who would otherwise be on welfare, who have minimal jobs, and as they are trying to work their way up to a higher position of some kind, this helps them out.

The vast majority of these individuals are not some sort of crooks or tax cheats. They desperately want to work and the tax code should not make public assistance more attractive than work. Working families who receive the credit are not getting some sort of a handout. They are working very hard for each paycheck and need a little assistance to move up to a higher level.

I know we are going to hear about some problems with the pro- gram, but I hope that everyone here agrees with its goal, and that is to make work pay. Let us not tear the credit apart. Let us make it work better.

Thanks to President Clinton's expansion of the EITC, in just my home State of Ohio, more than half-a-million working families in Ohio are getting a little extra back from their paychecks. There are lots of stories we can tell on this. I wish we had time to tell all the stories, just of our Ohio families, but let me share just one with you.

Brenda Manders is a divorced mother of a 3-year-old in Colum- bus. She has received the credit for 3 years. Brenda, who works for legal services and has been in training to become a legal secretary, this year received a total refund of $2,740. This was very fortunate, because after a separation from her husband, Brenda and her child were left with nowhere to live. Faced with homelessness, she was able to use her tax credit to pay a security deposit and rent an apartment for her and for her child. Without it, Brenda and her child may well have wound up on the street.

While we are at it, we had a letter, also, Mr. Chairman, from a success story in your home State of Delaware. Mindy Heck of Wil- mington is a divorced mother of three. When she was married, she was able to work at home and care for her kids. Now, she is work- ing for Goodwill and is receiving the advance tax credit in her pay- check, which she uses to pay for child care. She has also used her credit to make mortgage payments so that she would not lose her family home.

That is what the earned income tax credit is all about. Don't get me wrong, it is not just one big fairy tale. There are a lot of prob- lems with the program, and as with so many tax provisions, there has been too much fraud and too much abuse, we would all agree. We sought to fix some of these in the 1993 reconciliation bill. Those corrections did not take effect until this past tax year, so we will

have to keep a close eye on how well they work. I think this soon, it is difficult to say how they really will work.

I also want to point out that the very same 1993 bill specifically mandated the phased-in expansion of the EITC. We have here in the Omnibus Budget Reconciliation Act of 1993, it is phased in year after year after year as to what the percentages will be and how the thing phases in. So this is not some run amok, runaway program. The phase-in is going exactly as it was planned to go by laws we passed right here on Capitol Hill. We are the ones who set the phase-in, as outlined in that Act.

As soon as the expansion is completely phased in, after it has gone through the several steps, then the program will drop down to its normal low growth rate. In fact, the credit will decline as a percentage of the GDP after 1997. I wish we could say that about many of the special tax breaks that we hear about being proposed in the Congress these days. As the 1993 changes were being made here in Congress, I don't recall any proposals on the other side of the aisle to phase it in more gradually.

I think it is important to note that the EITC has been a biparti- san program, all the way through. Its origins go back to Governor Ronald Reagan in California, and President Nixon first proposed it in Washington. Its first major expansion came under President Ronald Reagan in 1986. It was expanded under President Bush in 1990 and expanded again under President Clinton in 1993, so it has enjoyed broad bipartisan support.

I was hoping to ask Les Samuels from the Treasury Department about some of these issues today, and I am sorry he will not be here at the hearing to discuss such matters of policy. We thought he was going to be here today, but they decided to change his testi- mony to later. We do have his remarks available at the press table so you can compare his policy statement there with other testimony today, and I think everyone will find them to be very insightful.1

Finally, I think we should be very clear that the problems we will be addressing with the EITC are really part of a larger picture. This is one little part. How do we narrow the gap between what individuals and corporations owe and what they actually pay? We have to narrow that gap in order to uphold public confidence in the tax system.

The IRS currently is at a crossroads, and their success depends on what we in Congress will do this year. The IRS is plagued and has been plagued by an antiquated, inefficient computer system, and whether we are talking about browsing, filing fraud, or just collecting what is owed, it is crucial that the tax system moderniza- tion, the TSM program, be kept on schedule and properly imple- mented.

The IRS must also continue to ensure that taxpayer records are kept confidential. That is why I am today introducing legislation called the Taxpayer Browsing Protection Act, to make computer voyeurism or browsing a criminal offense, and I will be introducing that later today.

I was shocked to learn at our hearings last summer how some overly-curious IRS employees have called up the returns of friends

Prepared statement for April 4 hearing submitted by Les Samuels appears on page 210.

or neighbors, even in some areas, celebrities, and, unfortunately, there seems to be a technical loophole in our ability to deter and punish such outrageous transgressions. My legislation will close that loophole by making browsing a criminal offense and giving taxpayer records the protection that they deserve.

In closing, I want to point out that, last week, we held a hearing on GAO. One of the matters of concern to the Committee and to the Chairman was the frequent delay in receiving GAO reports. It is my understanding that in this particular case, the original target date given GAO was to be this June. However, given the timeliness of today's hearings, GAO was asked and was able to speed up their work to meet the Committee's needs, so they are getting this one in well ahead of schedule. In the wake of our GAO hearing, I thought that should be also noted this morning.

I look forward to the testimony, Mr. Chairman, and I am glad we are having this hearing.

[The opening prepared statement and news release of Senator Glenn follow:!

OPENING STATEMENT OF SENATOR JOHN GLENN

Good Morning. Let me start by thanking Senator Roth for holding this hearing. I believe that the Earned Income Tax Credit (EITC) is one of the most important programs the IRS oversees. The less fortunate in our society too often find them- selves fighting to feed their families, pay their bills and to stay off of public assist- ance. The vast majority of these individuals desperately want to work, and the Tax Code should not make public assistance more attractive than work. The working families who receive the credit are not getting some sort of handout — they are work- ing very hard for each paycheck.

Now, I know we are going to hear about some problems with the program, but I hope that everyone here agrees with its goal — to make work pay. Let's not tear the Credit apart. Let's make it work better.

Thanks to President Clinton's expansion of the EITC, more than half a million working families in Ohio are getting a little extra back from their paychecks. I wish we had time to tell the stories of all of those Ohio families — but let me just share one with you:

Brenda Manders, a divorced mother of a three-year old in Columbus, has received the Credit for 3 years. Brenda, who works for Legal Services and has been in training to become a legal secretary, this year received a total refund of $2,740. This was very fortunate, because after a separation from her husband, Brenda and her child were left with nowhere to live. Faced with homelessness, she was able to use her tax credit to pay a security de- posit and rent an apartment for her and her child. Without it, Brenda and her child may well have wound up on the street.

And while I'm at it, Mr. Chairman, I am proud to share with you a success story right in your home State of Delaware:

Mindy Heck of Wilmington is a divorced mother of three. When she was married, she was able to work at home and care for her kids. Now, she is working for Goodwill and is receiving the advance tax credit in her pay- check which she uses to pay for child care. She has also used her credit to make mortgage payments so that she would not lose her family home.

That's what the Earned Income Credit is all about. Now don't get me wrong. It's not just one big fairy tale. There are some problems with the program. As with so many tax provisions, there has been too much fraud and abuse.

We sought to fix some of these in the 1993 Reconciliation Bill. Those corrections did not take effect until this past tax year, so we will have to keep a close eye on how well they work.

I also want to point out that the very same 1993 Bill specifically mandated the phased-in expansion of the EITC. This is not a program out of control; its growth was planned very deliberately. As soon as the expansion is completely phased in, the program will drop down to its normal low growth rate: In fact, the Credit will

decline as a percentage of the GDP after 1997. I wish we could say that about the many special tax breaks currently proposed over in the House.

Now as the 1993 changes were being made here in Congress, I don't recall any proposals on the other side of the aisle to phase it in more gradually. In fact, I think it's important to note that the EITC has been a bipartisan program. Its origins go back to Governor Ronald Reagan in California. President Nixon first proposed it in Washington. And its first major expansion came under President Ronald Reagan in 1986.

You know, I was hoping to ask Les Samuels from the Treasury Department about some of these issues today. I'm disappointed that he was not allowed to participate in today's hearing — to discuss such matters of policy — though until late yesterday he thought he was welcome. But we have his remarks available at the press table. I think everyone will find them to be very insightful.

Finally, I think we should be very clear that the problems we will be addressing with the EITC are really part of a larger picture — how do we narrow the gap be- tween what individuals and corporations owe and what they actually pay. We have to narrow that gap in order to uphold public confidence in the tax system.

The IRS is currently at a crossroads, and their success depends on what we in Congress will do this year. IRS is plagued by an antiquated and inefficient computer system. Whether we're talking about browsing, filing fraud, or just collecting what's owed, it's crucial that the Tax Systems Modernization (TSM) program be kept on schedule and properly implemented.

The IRS must also continue to ensure that taxpayer records are kept confidential. That is why I am today introducing legislation — The Taxpayer Browsing Protection Act — to make to make computer "voyeurism" or "browsing" a criminal offense.

I was shocked to learn at our hearings last summer how some overly-curious IRS employees have called up the returns of friends, neighbors, and even in some cases, celebrities. Unfortunately, there seems to be a technical "loophole" in our ability to deter — and punish — such outrageous transgressions. My legislation will close that loophole by making browsing a criminal offense and giving taxpayer records the pro- tection they deserve.

In closing, I want to point out that last week we held a hearing on GAO. One of the matters of concern to the Chairman was the frequent delays in receiving GAO reports. It is my understanding that, in this particular case, the original target date given GAO was to be this June. However, given the timeliness of today's hearing, GAO was asked — and was able — to speed up their work to meet the Committee's need. I do believe, in the wake of our GAO hearing, that this should be noted, and I'm sure, appreciated by the Chairman.

I look forward to our hearing this morning, and to the testimony of our witnesses. Thank you.

[The news release follows:]

NEWS RELEASE FROM SENATOR GLENN

GLENN BILL TARGETS IRS BROWSING

'Taxpayer Browsing Protection Act" To Make Snooping Through Tax Files

A Crime

Washington, DC — Calling browsing through taxpayer files by IRS employees a se- rious breach of trust, Senator John Glenn (D-Ohio) today introduced the "Taxpayer Browsing Protection Act of 1995" to make such actions a crime. The proposal is a result of investigations, overseen by Glenn, that uncovered Internal Revenue Service employees who used certain computer codes to gain unauthorized access to confiden- tial taxpayer files. However, after further investigation, it was determined that a technical "loophole" exists and this activity is not considered a crime. Glenn's legis- lation will now make browsing a crime punishable by up to one year in jail and a $1,000 fine. Also, government employees convicted of this crime would lose their jobs.

"Investigations undertaken by my Governmental Affairs Staff, the GAO and the IRS discovered that widespread browsing through taxpayer records has occurred at the IRS," Glenn said. "How much of this was done with the intent to defraud the government and how much was done as simply 'prurient window peeping' is unclear. However, what is clear is this is a serious breach of trust. As Americans sit down to fill out their 1994 returns, they need to be assured that this most personal infor- mation will remain private and confidential."

"Up to now, though, employees caught simply 'snooping1 through these files couldn't be charged with a crime — they only faced agency discipline. My legislation changes that and makes browsing a criminal offense, with penalties and the threat of job loss. We've got to eliminate this activity — we cannot tolerate this invasion of privacy."

"Snooping" by IRS employees was first uncovered by Glenn in the summer of 1993. Last year, Glenn revealed that 1,300 IRS employees had been investigated na- tionwide since 1989 for improperly accessing taxpayer accounts. Often this entailed IRS employees gaining unauthorized access to the files of friends, neighbors, rel- atives or celebrities. Currently, if the information obtained from browsing is not shared with unauthorized individuals or altered, the action is not considered a crime.

"If you browse and share that information with someone — it's a crime. If you browse files in order to change taxpayer return information — it's a crime. Now, under my legislation any sort of unauthorized access of taxpayer accounts will con- stitute a crime. Employees should know that if you browse, you could go to jail, be fined and lose your job."

Glenn said he has worked with the IRS to craft the legislation and welcomed their support. "I'm glad that IRS realizes the severity of this problem and has cooperated in finding ways to eliminate it. Making browsing a criminal offense — and ensuring those who are convicted get fired immediately — is one way to help preserve the sanctity and privacy of taxpayer records in the mass computer age. I look forward to quick Congressional consideration of my legislation," Glenn concluded.

[Articles follow:]

1,300 IRS Workers Accused

Of Snooping at Tax:'Retuf ns

Employees Used Computers to Peek at Friends' Files

By Stephen Barr

Washington Posl Sl.irt Writer

More than 1,300 employees of the Internal Revenue Service have been investigated or dis- ciplined for using government computers to browse through tax returns of friends, relatives and neighbors, Sen. John Glenn (D-Ohio) said yesterday.

Glenn, chairman of the Senate Governmental Affairs Committee, said that of the 1,300 em- ployees nationwide that the IRS has investigated since 1989 more than" 500 cases occurred in the last 10 months.

The nationwide figures on IRS employee snooping in tax files were compiled following Glenn's disclosure in August 1993 that almost 370 employees in the agency's Southeast region had improperly examined taxpayer accounts.

That disclosure prompted outrage from sev- eral senators and concern from tax specialists, who said they were surprised that so many workers could be accused of abusing their posi- tions.

"How much of this was prelude to fraud, and how much was just prurient window peeping is

difficult to say," Glenn said in a statement. Call- ing the nationwide figures "disturbing," Glenn said the' IRS has a moral and legal obligation to protect the privacy of taxpayer records.

Glenn said that more than 420 of the IRS cas- es have resulted in some type of discipline, with several hundred more cases under investigation.

Glenn's committee will discuss the new data on IRS snooping at a hearing scheduled for today and also will look at the potential for fraud in- volving taxpayers who file inflated or phony, tax refunds. Officials from the departments of Edu- cation and Housing and Urban Development also will testify on whether any of their programs face financial risk because of fraud.

IRS Commissioner Margaret Milner Richard- son, in testimony prepared for the hearing, said the agency has not only disciplined employees but also undertaken several "preventive meas- ures" to ensure that IRS workers know and un- derstand that they are responsible for protecting taxpayer privacy.

The agency's Integrated Data Retrieval Sys- tem, which is used to locate and adjust taxpayer accounts, handles about 1.2 billion transactions a See IRS, A7, Col. 1

10

Glenn- Calls IRS Figures 'Disturbing' j

IRS, From Al

year, she said. About 56,000 IRS em- ployees have access to that system out of a work force of 115,000.

The commissioner also noted that she has appointed IRS's first 'prlv;icy advocate," Robert Veeder, formerly of the Office of Management and Budg- et, who will oversee efforts aimed at safeguarding taxpayer accounts.

Since last year's hearing by Glenn, the IRS his urged stiffer and more uniform' discipline for workers caught browsing. :The agency also has in- stalled an Electronic Audit Research Log to monitor abuse.

Glenn learned about IRS employee snooping alter the General Account- ing Office .reported that the "IRS did not adequately control access authori- ty given to computer support person- nel or adequately monitor employee accesses to taxpayer data."

IRS documents released by Glenn last year showed that the agency had been investigating regional employees for about three years. A small number of the cases involved workers who looked up the tax accounts of celebri- ties. At least one employee was re- ported to have altered about 200 ac- counts, and received kickbacks from inflated refund checks.

In his statement yesterday, Glenn expressed renewed concern about the rising level of taxpayer fraud detected by IRS. New data show that 53,100 cases' of paper or electronic filing fraud have been caught in the first five months of this year, he said. As a re- sult, Glenn and Richardson said, IRS expects 1994 fraud cases to double over last year's tally of 77,000 detec- ted cases.

In particular, Glenn said, electronic filing fraud has increased in recent years — growing from 6,000 returns detected in 1991 to 13,000 in 1992 to nearly 26,000 last year.

The IRS failed to stop refunds of about $25 million on returns iater de- tected as fraudulent last year, Glenn said. "However," he said, "no one knows what the actual fraud is be- cause so- much fraud goes undetec- ted."

The electronic filing system allows taxpayers to take their returns to au- • thorized preparers or professional fil- ers for transmission via computer to. the IRS.

Richardson described several schemes in her testimony, including one involving two owners of an in- come tax preparation firm in Salinas, Calif. The owners obtained the names and Social Security numbers of area agricultural workers from their cli- ents' payroll records. Unknown to the workers, the two preparers submitted more than 200 tax returns claiming more than $165,000 in bogus refunds. IRS uncovered the scheme, and the two preparers were jailed.

The IRS, Richardson warned, may be hampered in its efforts to combat fraud and thwart employee snooping because congressional appropriations committees have recommended cuts of about $400 million in next year's funding for Tax Systems Moderniza- tion (TSM), the agency's multibillion- doilar project to redesign and upgrade old computer systems.

The appropriations cutback, if en- acted, "may require us to stop all of the major hardware acquisition we had planned" and "completely rethink" the TSM project, Richardson said.

11

I.K.S. Staff Is Cited in Snoopings

1,300 Workers Have Been Investigated

nyR(iiicRri).iii:usimYJr.

cd 369 employees In icglon that il suspected of misusing lis Integrated Data Retrieval Syslcm. I'rlvncy Protected by l.nw Although Ihc agency has brood

discharges, 30 suspensions. 68 repri- mands, 26 admonishments and 18 counseling sessions. The other 203

WASHINGTON. July 18

iiilcrii.il 51 ol Hie

knowlcdgcd, however, lhal In cases there were "inconsls- in how people were punished

|Mi"i:5 10 snoop on (axpnycts. Gov- einmenl ollicials disclosed today. .

In most instances, the violations appear to have involved browsing by nil ions employees win, were interest- ed, lor example, in the linnnclal standing ol li i.-j.ils. neighlrais. ene- mies, potential Inlaws, siockbiokcis celebrities and louncr spouses.

In about a ihiid of the cases em- ployees have been subjected In sanc- tions tanging from counseling to dis- charge, wiib several I hed cases '.

still unresolved. Most uf ihc rest ol :

the Invcsllgations concluded lhal the

employees wcic engaged In olfici.il

business.

Some Involve Fraud

Some employees were said to hnvc been guilly only ol misguided at- tempts in hcl,. I, lends cut through II S. red tape or Intel prel often con- tusing I, It S. communications

A lew cases, however. InvolvcJ

fraudulent refunds and klckl'.acks lu the employees. Some indictments are said to have resulted.

now much ol lids was prelude In fraud and liow much was just pruri- ent window-peeping Is difficult to say." said Scnalor John Glenn, Hie Ohio Dcmoci at who heads ihc Senate Government Affairs Committee. Din he called the findings "disturbing " adding (hat the l.tt.S. had a nioi al and legal obligation for confidentiality "when Americans sit down and pro- vide the Government Willi Ihcir most personal Information."

Hie committee lias scheduled a bearing for Tuesday at which em- ployee browsing and a g, owing prob- lem ol taxpayer Iiaud. pai tlcuTorly Involving electronic filing, will be dis-

c1;V.ed-.?0"1c °'""! <,clalls °r »'<»i

will be discussed wcic lelcascd today by (he committee and the I.H.S.

In n statement prepared for her appearance, Margaret Mllncr Rich- ardson, Commissioner of Internal ?"="'„„• "IJll'"'>»c agency detect- en 53 100 returns fraudulently claim-

It Ai ' r?:.""l.!.T.,J" ",'""ds dl" '"B ">c

li .-uidi.lcnl

about twice a

77.SOO dc ,

"•■la sharp rise I attempts al fraud. Ion or some comblnai

The new figure ol 1,300 investiga- tions lor the entile country came as no surprise, some tax specialists and .

"I'm not at all shocked by lhat." said Michael J. Murphy, executive! director of Ihc Tax Executives Insll-: tutc, a corporate group, who served I as ucpuly commissioner from 1987 to 1002. While declaring that even one case ol unauthorized search was a

Privacy vs. Modernisation j

The challenge, specialists said. Is to ni-'ilaln privacy even as the system Is modernized to allow employees to bring taxpayer ac

lions" a year, the agency said.

"Is it Important? Is It serious? Yes , II Is." commented Lawrence D. Glbbs, a former commissioner now with Ihc low firm ol Miller ft Cheval- ier. "Am 1 alarmed? No." he added.. "Out It is a wake-up call." I

In addition to the penalty, guide, I Commissioner Richardson said she . had appointed Robert Veedcr, whom she described as a specialist hi the Privacy Act and Freedom ol Inlor- mation Act al the OHIcc of Manage- ment and Uudgcl. as the I.R.S.'s lirsl

Doth lormcr officials said they sup- ported the commissioner's disinclina- tion to inform taxpayers whose files had been improperly browsed lhal lids had oecuncd. "I lliink lhal Is Just creating an administrative night- mare," Mr. Murphy said, allowing lor the possible exception in a case where ', somebody's liability had been changed.

Muchol the increase In liling Iroud,

Scnalor Glenu and the I.R.S. agree. involves the earned income lax crcd- 1. il. a miiltihllllon-dollar program that; can result In low-income workers re- J ccivlng Government checks lor lhou-1 sands of dollars. The committee reported that 20,937.' electronic liling fraud had, Ihc lirst lour months-

been detccled

i were caught bclore

12

Tax fraud on rise by crooks using electronic filing

WASHINGTON (AP) - Electron- ic filing, which lets Impatient r»~ {iir:J-<re«ors ilp their tax returns to ♦he Internal Revenue Service, wax used by crooks to rap the govern- ment for $16 million this vear.

The 1R8 ha« far mora difficulty catching fraudulonl refund claims from electronic returns than from traditional paper returns. .

9oa John Glenn (D., 0,), who made the statistics public yester- day, suggested that electronic filing may have to be suspended while the p oh the problem, lleve that the agency currently has the capacity or tech- nology to catch most refund schemes," said Senator Glenn, chairman of tho Senate Governmen- tal Affairs Committee. "Tliore must be hotter controls on electronic III- Ing."

l'» said ha wants the electronic filing program added lo the govern- ment's list ol "high-risk" programs vulnerable lo abuse, ami wants U\3 Commissioner Margaret Richardson to come up with a plan to curb electronic filing fraud before next year.

IRS spokesman Frank Koith aald, "We have made significant progress tn dealing with refund fraud, both on paper and In the electronic arena. We remain convinced that the elec- tronic filing program Is an Impor- tant part of our i ' the IRS.'

plans to modernize

Mr. JCeltn confirmed sUtlsttci provided by Senator Glenn's staff that showed a difference between the early detection ralos of phony iclund claims filed on paper and 'jiose filed electronically.

During the first four months of this year, the IRS found thai paper Isi returns seeking $65 S million In refunds were fraudulent. The agen- cy discovered the fraud- early enough to prevent lis. I million of that from going out to the cheaters.

For electronic filings, the IR8 found $44.1 million In fraudulent refunds, of which $29.1 million was spotted before refunds went out.

During that four montlie, tlw 1118 approved almost $(2 billion of re- funds.

In 1MJ, electronic filers tried to .dupe the IRS for $64 million In undeserved refunds. The agency spotted the questionable refunds In time to proven', $!t million from Solng otit, but the checks for anothor tU million were mailed.

That compared with $32 million worth of falsely claimed rotunda ou paper filings, of which $7z.» million was detected In time to prevent payment.

Under electronic filing, available nationally since 1090, authorized preparer! and commercial compa- res can file returns for larpayers

Glenn: Bottsr control* naedod.

via computer, cutting the rotund time from sis or seven wocks to two weekn.

Jennie Stalhla, a tax expert for Congress's General Accounting Of- fice, suggested earlier this year that electronic returns are pro- cessed so quickly that It la difficult for the IRS to block questionable claims,

Mr. Keith sild electronic filing may, In fact, make It easier In the long run to prevent refund fraud. The figures cited by Senator Glenn's staff, he added, "may not delect the extent to which we prevented fraud from ever getting Into the system this year."

In remarks prepared for delivery today to tho ssnator's committee, Ms. Richardson said the IRS has tightened tha BtandardB that new companies must meet before being allowed to file electronic returns. " For 1006, they will have to submit to fingerprinting for criminal back- ground checks and give authoriia- lion for credit checks, she said.

Additional fraud-fighting power Is planned when IRS computers are modernized, but the fate of that upgrade li uncertain as congressio- nal committees look for ways to cut fedoral Spending.

For the budget year that begins In Octobor, tha IRS sought $089 million to upgrade Its computers. The House cut that by $617 million, the Senate by $386 million a conlereuce committee now Is negotiating what the final amount should be.

The IRS alio has entored Into a partnership with Los Alamos Na- tional Laboratory to tap the brain- power and computer expertise that built tha atomic bomb.

/•

*J

13

|:|lM

:■/'

01 the IflS' 1 15,000 employees, 56,000 hnvc access to computerized tax records. Since 1909, 1,300 have been Investigated. 01 360 potential violations in llic Southeast region from 1909-93:

1G5 resulted In r.omo corrective nclion 203 cases were found

17 discharges 36 suspensions 60 reprimands lQ

26 odmonlshrnents 10 counseling sessions

In I business

Saorcn: IHS

a n.r»..i. usa toomt

COVER STOP;

*e©iiifirms the worst ffear'

Ac lion taken against 420 employees, but virtually 'no controls' on access

Hull's supprr.nl lo lie lor olfici ilcnUnl. 1 hose probes i esiilicd «(MRS workers.

Hy llili Montague nnrl Phillip Rorin USA I ODAY

C3T0N - nip, I no! lie watching, I lie snooping,. Revenue Service

.100 IHS employees liavc .■ctt iuvcstiealed lire past five •nis on suspicion ol linprnp- nxpnyer Tiles — information il use only nuu slikliy ennft- in disciplinary action against

;.!-.< pn . John

■pint; is dirncull t

S.IV5

. employees -

( omiiilllee i

Thcpiobli II"- agency's liri.lioi) workforce - have access lo what's c.-'llcil the Inlcgtated t>;it.i Retrieval System, or IDRS, Ihe computer syslcin thai liamlics i ollcciloii ami sloragc ol lax payer ihforninlion.

Employees arc supposed to use llic system in answer tax- payers' questions, or help lliein resolve problems. Dot lor sonic IRS works's. Hip system has piovulcd a way (n satisfy

14

COVER STORY

IRS promised privacy '20 years ago'

Continued from 1A

Evidence of IRS privacy abuses was revealed by the Senate Govern- mental Affairs Comtnltlce last Au- gust. Then, the IRS said allegations of tax fraud by Its employees triggered n review of operations In the agen- cy's Southeast Region. That probe of employee use of computer records mcovercd 3G8 potential privacy vlo- atlons. So officials checked suspl- :lous patterns of computer use In the IRS' other six regions.

Meanwhile, the General Account- ing Office, the watchdog arm of Con- gress, conducted lis own study. Highlights of that study:

► Of 308 employees Implicated in the IRS' Southeast Region, 203 were cleared of wrongdoing. Of the re- maining 1C5 workers disciplined for misconduct, 17 were fired. Another 30 were suspended. The rest re- ceived either reprimands, admonish- ments or mandatory counseling. The IRS says It ain't provide a similar breakdown of the 1,300 cases uncov- ered In lis national probe.

►- The IRS' ethics office concluded that in at least 83 of the cases In the Southeast region, snoopers were treated too leniently.

► According to a 1992 Internal IRS audit, some agency employees sub- mitted fraudulent tax rerurns, then used their computer access to moni- lor the IRS' review of those returns. Others used the computer to Issue fraudulent refunds to friends and as- sociates. Criminal Indictments have been brought against some.

h IRS data security systems In- clude "virtually no controls ... to limit what employees can do once (hey are authorized access," the GAO says. The IRS has only a limited ability to trace Improper attempts by employees to read computer files, GAO auditors say. '

► Each year, Uie IRS computer system handles more than 1.2 billion requests for information from IRS employees. That volume of traffic, coupled with the IRS" nrlmillvn secu-

rity systems, make It difficult if not Impossible to detect abuses.

Congressional leaders say there's need for n crackdown.

"The IRS' disregard of taxpayer rights confirms Uie worst fear that the American people have about Uie IRS," says Sen. David Pryor, D-Ark., chairman of the Senate IRS over- sight subcommittee. "This Illegal and offensive activity must stop and It's clear Congress must ncL"

Federal law already sets strict criminal penalties for unauthorized disclosure of taxpayer Information by IRS employees. Vlolators,can be punished by a fine of up to $5,000 or five years In prison.

Uut Llicre are no criminal penal- lies Just for snooping. Some experts say they may be needed. "At some point, you have to start whacking people," says Sheldon Cohen, a tax attorney and former IRS commis- sioner.

IRS spokesman Henry Holmes says the IRS hasn't decided yet whether to notify taxpayers who have been victims of employee snooping.

After last year's revelations, the IRS updated Its disciplinary guide- lines, and Ls developing a training program on computer security. The agency also named Robert Veeder, an expert on computer privacy, to Uie new post of privacy advocate.

IRS critics wonder how serious Uie agency really Is about cracking down. The new disciplinary guide- lines list snooping as No. 20 on a list of 27 employee offenses — Just after "prohibited personal practices," and Just before "oUier 'offenses."

"The IRS promised to give us pri- vacy 20 years ago, and Uiey sUII haven't delivered,'" says David Bumham, a fonner InvcsUgatlve rc- poiler for 77ie New York Times and auUior of a 1989 book, A Law Unto Use!/, about the IRS.

One big problem: Too many em- ployees have access to Uie computer system that handles returns. Agency officials sav that's necessary because

so many IRS employees deal wlUi taxpayers and their problems.

"We don't want to curiall Uie ablil ty of our people to do Uielr Jobs, which Is to help taxpayers," says Holmes.

But outside experts say poor man ngement Is a more likely explana tlon. "Clearly, loo many people have access lo sensitive Information," says Walter Goldberg, a senior manager wllh Ernst & Young, a major ac- counUng firm.

Ironically, the IRS' antiquated computer system Is one of Uie agen- cy's few obstacles to employee snooping.

Currently, taxpayers mall Uielr re- turns to one of 10 regional service centers for processing, where Uie ln- formaUon Is reviewed and then typed Into computers. Computer tapes containing those records are shipped to Uie IRS' main computer center in MarUnsburg, W. Va.

Because Uie computers In Uie ser- vice centers aren't linked lo each other, or to Uie MarUnsburg comput- er, IRS employees using Uie IDRS system can only access data filed by taxpayers In their region. And only some types of information can be re- trieved. '

However, Uie IRS ls In Uie middle of an $8 billion computer systems up- grade. Eventually, opUcal character readers will be used to scan tax re- turns Into Uirec main computers, hooked together In a naUonal net- work Uiat links Uie 10 IRS service centers, eight regional offices and 65 district offices.

An earlier IRS modernlzaUon plan was knocked down by Congress In Uie late 1970s, In part because of pri- vacy concerns. IRS officials say Ute new system will help protect priva- cy, by giving agency managers high- tech tools to watch employees. But critics remain skepUcal.

Says Burnham: "The only way we are going to deal with this problem Is If we have some kind of outside pri- vacy watchdog group to keep Uie pressure on."

15

Chairman Roth. Thank you, Senator Glenn.

I think we are in agreement as to the purpose and intent of the program, which was to help struggling families get off welfare without fostering dependency and inviting abuse. Our concern, of course, is that when, over a period of many years, the waste, fraud, and abuse consistently shows a pattern of 25 to 40 percent, I am sure it is a matter of concern to all of us. The question is, how can this situation be remedied?

You are absolutely right, this program has grown under both Re- publican and Democratic administrations. The most significant growth, of course, has come in the last couple of years, as shown in the chart there. As I said, we will have, next year, roughly 19 million families securing EITC out of a total of 117 million families filing income tax, so it is about one family out of six that is benefit- ing, presumably, from EITC. That is, in part, the good news.

The bad news, of course, is that 25 to 40 percent who are getting funds that they do not deserve. That is the bad news.

Mr. Samuels, as originally scheduled, will be here tomorrow. That was the original schedule, and we, too, are looking forward to his testimony.

In the meantime, it is a pleasure to welcome you here, Ms. Rich- ardson. I know your interest, your concern about this problem. We look forward to hearing you. Would you please tell us who accom- panies you?

TESTIMONY OF THE HON. MARGARET MILNER RICHARDSON,1 COMMISSIONER, INTERNAL REVENUE SERVICE; ACCOM- PANIED BY PHIL BRAND, CHIEF COMPLIANCE OFFICER; TED F. BROWN, REFUND FRAUD EXECUTIVE; MICHAEL P. DOLAN, DEPUTY COMMISSIONER; AND JUDY VAN ALFEN, CHIEF, TAXPAYER SERVICE

Ms. Richardson. Thank you, Mr. Chairman and Senator Glenn.

I have with me today Michael Dolan, who is the Deputy Commis- sioner, who is on my left and your right, and Ted Brown, who is our Refund Fraud Executive. I also have with me Judy Van Alfen, who is the Chief of Taxpayer Service, and Phil Brand, who is our Chief Compliance Officer.

I have a longer written statement which I would like to have submitted for the record, with your permission, and then I would like to summarize it for you.

Chairman ROTH. Without objection.

Ms. Richardson. We appreciate the opportunity to be here today to discuss our efforts to stop refund fraud, and I think, Senator Roth, you pointed to an article in the Virginia paper, which, I think, highlights that we do take a dim view of fraud, if that is what the subheading said. I would certainly agree with that.

I think that also highlights the fact that refund fraud is much broader than the earned income tax credit, because in that particu- lar case, I believe, we had 39 returns filed and only two of them touched on the earned income tax credit. The vast majority were just plain old fraud.

lThe prepared statement of Ms. Richardson appears on page 131.

16

Shortly after I became Commissioner, which was just a little less than 2 years ago, I recognized the need for us to step up our fraud detection efforts. Since that time, we have had to take many steps to ensure that fraud detection receives the highest priority. The IRS is and will remain committed to detecting and preventing at- tempts to undermine our tax systems by those who are unwilling to comply with the tax laws.

This administration is strongly committed to making work pay and to lifting workers out of poverty in the most efficient and ad- ministrable manner possible. In order to improve its effectiveness, in 1993, the administration supported a proposal to simplify the earned income tax credit by eliminating the two supplemental cred- its for health insurance coverage and for taxpayers with children under 1 year of age. The proposal was enacted as part of the Omni- bus Budget Reconciliation Act of 1993.

Then, in 1994 in the GATT legislation, the EITC was denied to nonresident aliens and also to prisoners.

Taxpayers are required to provide a taxpayer identification num- ber for each earned income tax credit qualifying child, regardless of age. Also, now members of the armed forces who are stationed abroad are eligible for the credit, and the Department of Defense is required to report both the IRS and military personnel non- taxable earned income that is paid during the year that is included in computing the earned income tax credit.

The administration has submitted additional legislative propos- als as part of the budget this year, the fiscal 1996 budget, which is still pending final action. The budget includes a provision that would deny the earned income tax credit to undocumented workers, and we would be authorized to use simpler and more efficient pro- cedures when taxpayers claiming the EITC fail to supply valid So- cial Security numbers.

We do hope that Congress will act on these proposals, but before we consider significant other changes to the program, we hope that we can wait to observe the effects of the recent legislation as well as these enhanced compliance efforts that we will be talking about in a few minutes.

The IRS is like financial services businesses, credit card compa- nies, insurance companies, and we are challenged daily by individ- uals who are trying to cheat the system. We have a significant ad- ditional challenge, however. We cannot screen all of our customers based on their credit history or other information. We have to take all comers.

This morning, I want to share with you what the IRS has learned about refund fraud, describe for you our actions during the current filing system to stop such fraud, and then to discuss with you our future fraud prevention plans.

Understanding the fraud schemes confronting the IRS is essen- tial to planning the most effective methods to detect and prevent fraud. This is difficult, because many fraud perpetrators think cre- atively, they relish devising complex schemes, and they adapt con- tinuously to new fraud controls.

Before this Committee last July, I reported on three filing season fraud studies that the IRS planned to provide more comprehensive analysis of the characteristics and extent of refund fraud.

17

The first study involved a small sample of 1,000 returns that were filed electronically during January of 1994, which had claimed the earned income tax credit. The preliminary results of that study, as we reported to you last July, showed that roughly 35 to 45 per- cent of the 1.3 million returns that had been filed electronically claiming the EITC during that 2-week period contained errors that required adjustments, both up and down, in the amount of the earned income tax credit claimed. About half

Chairman ROTH. Do you have a breakdown? When you say both up and down, what percentage were

Ms. Richardson. We do, and I will provide that for the record.

Chairman ROTH. Could you summarize what that is? What per- cent were under?

Ms. Richardson. I believe about 6 percent were under.

Chairman Roth. Six percent were under?

Ms. Richardson. About 6 percent were under.

Chairman ROTH. And 94 percent were over, then?

Ms. Richardson. I am sorry, about 29 percent were over.

Chairman Roth. Twenty-nine percent were over?

Ms. Richardson. Right. About half were unintentional errors and about half appeared to be the result of some intentional mis- representations. I think, perhaps, somewhere between half and a third, or about a third, maybe, were in what we would call the fraud category.

Taxpayer characteristics that we gleaned from the study, though, aided in the development of additional fraud controls, which we im- plemented this year. The final analysis of the study should be available sometime in the next month, and we will be delighted to share those results with this Committee.

The second study, which we conducted in February of 1994, was to determine whether refunds from the electronic filing system were being received by taxpayers as issued. It involved 2,200 tax- payers whose returns had been filed electronically by electronic re- turn originators, or ERO's. That study has also been completed, and in only a handful of those cases did there appear to be evi- dence that the ERO's were keeping part of the taxpayers' refund.

Our third study is currently underway. It involves a statistically valid random sample of approximately 2,000 refund returns filed electronically and on paper claiming the EITC that had been filed throughout this 1995 filing season. The results from the third study are going to be used to expand our understanding of issues identified during the first study. The field work and analysis of the results will be completed later this fall and we will also be pleased to share those results with the Committee.

I have learned that to maintain effective fraud prevention strate- gies which were perfectly satisfactory today, it may be useless to- morrow. For that reason, the study of refund returns filed during this filing season currently underway will provide valuable infor- mation for planning our strategy for next year's filing season and beyond.

As part of our continuing efforts to prevent and detect fraud, we have developed and implemented numerous systemic verifications and enhancements for the 1995 filing season. In addition, signifi- cant resources are being directed to preventing and detecting ques-

18

tionable and fraudulent refund claims before those refunds are paid.

These initiatives include increased verification of taxpayer Social Security numbers, additional checks of returns claiming certain credits, ERO suitability checks, and increased ERO monitoring, as well as additional compliance resources devoted to fraud detection and prevention.

An important part of our strategy includes delaying all refunds that, as a result of the computer analysis and fraud identification profiles, appear to be erroneous or, perhaps, fraudulent. This addi- tional time for review, in most cases, up to 8 weeks, has helped and will help us detect fraud schemes, including duplicate uses of Social Security numbers.

Internal studies and the report of an outside expert confirmed that fraud was being perpetrated through the use of incorrect and invalid Social Security numbers. As a result, during this filing sea- son, we are devoting substantial resources to ensuring that tax- payers claiming refunds use the proper Social Security number. A correct, valid Social Security number must be provided for the tax- payer, the spouse, and dependents before an electronically-filed re- turn will be accepted.

So far this filing season, over 3.9 million occurrences — and, I want to underscore, not returns — of either missing, invalid, or du- plicate Social Security numbers have been identified on electroni- cally-filed returns, resulting in the affected returns being rejected.

But our checks of Social Security numbers are not limited to elec- tronically-filed returns. We are also checking

Chairman Roth. Would you explain, for the record, exactly what you mean by 3.9 million occurrences?

Ms. Richardson. Not returns?

Chairman ROTH. Not returns.

Ms. Richardson. There may be situations where a taxpayer, a spouse, and two or three children are listed on a tax return, maybe even more dependents, and those Social Security numbers, if each one were wrong, you might have four, five or six occurrences but only one tax return.

Chairman Roth. So it would be only one return, but there could be more than one error in one return, is what you are saying?

Ms. Richardson. That is correct. Or, perhaps, they have tried to file a return with names of dependents and a spouse but no Social Security number, so that would be three or four occurrences.

Chairman Roth. But you did find that there are 3.9 million er- rors, occurrences

Ms. Richardson. Involving Social Security numbers.

Chairman ROTH [continuing]. Involving Social Security numbers?

Ms. Richardson. With returns that people attempted to file elec- tronically.

Chairman Roth. And these are all electronic?

Ms. Richardson. Those were all in the electronically-filed re- turns.

Chairman ROTH. Is that much higher than on paper?

Ms. Richardson. I am not sure we have the precise numbers right now for the paper, do we, Ted?

19

Mr. Brown. They are not as high, because the ELF season is much earlier. We get most of the ELF returns through mid-Feb- ruary and our paper peak is just now beginning to build, toward April 17.

Chairman Roth. Please proceed.

Ms. Richardson. The failure to provide a valid Social Security number results in a delay until we can resolve the matter.

In one service center, we identified over 400 uses of the same in- valid Social Security number. One preparer prepared returns using this phony Social Security number over 400 times, 64 times as a Social Security number for the primary taxpayer, 113 times as a Social Security number for the children being claimed for the EITC, and 261 times as a Social Security number for children being claimed as dependents. Because we do check for duplicate So- cial Security numbers, in this one situation, we have already de- layed refunding over $380,000.

We spent a lot of time both before and during the filing season urging taxpayers to use correct Social Security numbers on their tax returns for themselves and their dependents. We emphasized the importance of accurate Social Security numbers this filing sea- son by including a message to that effect on the cover of all tax packages and through many public service announcements.

While our verification of Social Security numbers may cause delays for legitimate taxpayers this year, once the Social Security number problems are corrected, these taxpayers should not experi- ence delays in future years because of Social Security number prob- lems.

Our studies of fraud also found fraud related to credit, such as the earned income tax credit, and motor fuel excise tax credits. As a result, during this filing season, we are performing additional checks on returns claiming these credits to ensure that only those taxpayers who are entitled to them receive them. Refunds are being delayed on some returns to allow us additional time to verify claims prior to issuing the refunds.

Because of the additional time needed to complete the review, in some cases, taxpayers who claim credits may initially receive their refunds of withheld income taxes, followed by a separate refund check for the credits. When a refund is delayed, a notice is sent ex- plaining the reason and that, generally, the refund will be sent within 8 weeks from the date of the notice.

Our fraud screens are designed to detect suspicious returns. However, some taxpayers who have filed complete and accurate re- turns have also had their refunds delayed. We regret any inconven- ience that this has caused them, but despite what you have heard and what you may hear later this morning, we have received many positive responses to our fraud prevention efforts this filing season.

I believe that most taxpayers understand that the IRS needs the additional time to verify the accuracy of refunds claimed to main- tain the integrity of the tax system and to make certain that those who are not entitled to refunds do not get them and that those who are entitled do. Taxpayers who have legitimate hardships as a re- sult of the refund delay have been and will continue to be helped through our problem resolution program.

20

At the beginning of this filing season, we estimated that 82 mil- lion individual refund returns would be filed, and up to 8 percent of these refunds could be delayed with the new screens and filters. Through March 24 of this year, almost 41 million refunds have been issued. A little over 38 million were issued for the full amount of the refund, 2.6 million were partial refunds, and only 2.6 million refunds have been delayed in their entirety. These numbers are consistent with our estimates and we continue to project that ap- proximately 8 percent of the total refunds will be delayed.

I thought I might give you some examples of the potentially fraudulent schemes we have already detected this filing system.

In one scheme, 73 paper returns prepared by a Virginia preparer were filed at our Philadelphia and Austin service centers, each of which claimed large deductions on Schedule A. Because of our new automated detection systems, an additional 200 electronically-filed returns by the same preparer were identified, again, with large de- ductions on the Schedule A.

In another scheme, we had numerous returns with a Schedule C, or the self-employed schedule, with net income in the range of $8,000 to $10,000 that claimed head of household filing status. These returns also claimed the full earned income tax credit. Since most of the taxpayers had similar or identical surnames and lived at the same address or a similar address, we slowed those returns down. To date, we have identified about 112 returns with over $200,000 in refunds claimed.

We are going to work, as I have mentioned to this Committee be- fore, with the Department of Justice and with the U.S. Attorneys to continue to actively pursue prosecution of criminal violations. Last year, in fiscal year 1994, 51 return preparers were convicted of fraudulent refund schemes and were sentenced to an average prison term of 20 months. Thus far this year, 14 preparers have been convicted of fraudulent refund schemes and have been sen- tenced to an average prison term of 21.5 months.

We have also learned that some

Chairman Roth. If I might just interrupt to make sure I under- stand, you really have two types of problems, as I see it. You have the individual who may deliberately or through lack of understand- ing make an error in his return. That is part of the problem. But it is also true that there are some major scams

Ms. Richardson. Correct.

Chairman ROTH [continuing] . Where hundreds, if not thousands, of forms are prepared, in some cases by so-called tax preparers, and these involve hundreds of thousands, if not millions, of dollars. Is that correct?

Ms. Richardson. That is correct, also.

Chairman Roth. Thank you.

Ms. Richardson. We have also learned that some of the elec- tronic return originators have been responsible. These are not peo- ple who are technically preparers but who just send the returns in. They have been responsible for initiating or aiding with a large amount of fraud.

Chairman Roth. And they are a part of the larger scams, or are they?

21

Ms. Richardson. They can be, or they can also do individual sit- uations as well.

Chairman ROTH. They also do individual?

Ms. Richardson. But as part of our fraud prevention efforts this filing season, we have instituted fingerprint and credit checks on new ERO applicants to better ensure that only appropriate and re- sponsible individuals participate in electronic filing. Of the 33,000 applications this year that had to undergo suitability checks, 1,500 applicants were rejected because of failure to meet our more strin- gent admission requirements.

Another effort we have undertaken throughout the filing season is enhanced monitoring of these electronic return originators in our various district offices. So far, these visits have proved very suc- cessful. Through the beginning of March, we had conducted over 3,600 monitoring visits, resulting in the suspension of 103 ERO's from the program, and we have issued warnings to an additional 303.

Another fraud prevention step that we took this filing system was to eliminate the direct deposit indicator, or the DDL In the past, this indicator has signaled an electronic return originator that a taxpayer's refund would not be reduced to satisfy another government debt. Although the DDI had been used by lenders to determine whether to issue refund anticipation loans, the IRS has no involvement with those loans.

Our experience over the last few years with electronic filing and the DDI showed that refund fraud schemes were assisted by the availability of refund anticipation loans. Thus, we decided to cancel the indicator. Lenders are still free to make refund anticipation loans, based on their usual lending criteria.

We recognize that the vast majority of practitioners and ERO's are interested in maintaining the integrity of our tax system. They recognize their responsibility to prepare, file, or transmit correct in- formation to the IRS. However, when we identify those few who do abuse the authority of their position by committing fraud or who fail to adhere to our program guidelines, we will remove them from the program and pursue criminal enforcement to the fullest extent we can.

This filing season, our criminal investigation division is using new technology to aid in the detection of refund fraud schemes, both paper schemes and schemes on electronically-filed returns. To identify these sophisticated schemes, we are using the Los Alamos National Laboratory to design software to detect anomalies and match patterns in large data sets. New anomaly detection pattern recognition tools were developed and they are actually being tested at our Cincinnati service center this filing season.

In addition to enhancing our systemic filters to detect more ques- tionable refund claims this season, we have substantially increased the enforcement resources dedicated to identifying fraud schemes, as well as examining questionable refunds. Criminal investigation resources in our questionable refund detection teams were in- creased by 11 percent and examination resources were increased by 277 percent, with over 1,700 staff years being devoted to curbing abuses and fraud.

22

As of March 24 of this year, we have identified 623 electronic fil- ing schemes. In connection with these schemes, we have delayed about $15 million claimed on over 7,000 returns. We have also identified 632 paper return schemes involving over 4,300 returns and have delayed refunds claimed of $8 million.

In past filing seasons, refunds on many returns involved in g po- tentially fraudulent scheme would be issued before the scheme was detected. The delay of questionable refunds this filing season has provided additional time to identify fraudulent claims and to select questionable claims for examination before the refunds are paid.

Although detection and prosecution of refund fraud are impor- tant, it is virtually impossible to prosecute every instance of fraud. Thus, we have to continue to build our barriers to fraud. Our goal is to prevent fraudulent returns from entering the system.

One of the biggest challenges in meeting this goal is to install the up-front fraud control that will effectively detect and prevent fraudulent refund claims from entering the system.

A number of the initiatives and systemic enhancements installed last filing season and in the current filing season were designed to stop fraud involving the earned income tax credit. We have early indications that the fraud control initiatives put in place this filing season are reducing the number of fraudulent claims involving the EITC, thus making the EITC unattractive to the fraudsters and preserving the credit for those who have earned it. However, when that filing season study I mentioned earlier is completed, we will be able to measure EITC compliance more precisely, and we would like to work with the Committee when we get those results.

Through the same systemic filters and detection efforts that are addressing EITC fraud, we are making strides in stopping fraud in other areas, such as the motor fuel excise tax credits. For example, we found one return was filed this season for someone purporting to be a self-employed beautician, claiming the tax-free use of over 42,000 gallons of gasoline, generating a proposed refund of $6,000. On this return and many others like it, we have stopped the re- funds.

One way that taxpayers who are eligible to receive the earned in- come tax credit can avoid potential refund delays is through the advanced earned income tax credit. Workers who qualify for the AEITC can get up to $105 a month in their paychecks, whether they are paid weekly, biweekly, monthly, or whatever, and they can do it by filling out a very simple form and providing it to their em- ployers.

The IRS and the Treasury Department are working to make more taxpayers aware of this option. For example, last year, we sent over 14 million EITC recipients information about the ad- vanced earned income tax credit. The Treasury Department has worked with corporate CEO's to gain their support and to solicit other CEO's and national organizations to provide information about the advanced earned income tax credit. The IRS and the Small Business Administration are also cosponsoring employer seminars in strategic locations throughout the filing season to train employers how to compute and provide the AEITC for their em- ployees.

23

So far, we think the preliminary results show that more tax- payers have opted for the AEITC than in all of 1993.

Mr. Chairman and Senator Glenn, I can assure you that we are committed to stopping all fraud in the tax system and we are going to continue our efforts to ensure that only those hard-working Americans who are eligible for the EITC receive it.

Although we are still in the midst of our current filing system, under the direction of Mr. Brown, our filing fraud executive, we have already begun planning our fraud prevention strategy for next year. We are gaining valuable information this filing season on which to base the modification and refinement of our current strat- egies. Based on the results, we will revise the standards used to screen ERO's and adapt the systemic screens used to detect fraud during the filing season, and, if necessary, we will put in place new filters.

While we continue to enhance our detection and our prevention efforts, the real key to improving our ability to detect fraudulent refundable credit schemes is our tax system's modernization pro- gram. Without modern equipment and software, applying expert systems analysis to large data bases is virtually impossible. Tax systems modernization will not only provide the computing power and capacity needed to apply sophisticated fraud detection systems, it will also provide us with more timely access to information.

Mr. Chairman, now, more than ever, we need to find a way to provide and assure a stable funding vehicle for the completion of tax systems modernization. As you indicated in your opening state- ment, fiscal year 1996 is a pivotal year for us as we continue our plans to acquire and implement major new systems. What happens to our fiscal year 1996 budget will impact the tax administration of the future and really affect our ability to administer the system.

As I stated earlier, fraud is a dynamic, constantly changing phe- nomenon. Prevention and deterrence are clearly the keys to con- trolling it. Even if we are successful in our current efforts to elimi- nate all fraud, our job will not be done. In our experience, when one avenue of fraud is shut down, the fraudsters clearly migrate to other, more accessible avenues.

As I have stated before, in some instances, it may be necessary to delay questionable claims for refunds while they are carefully scrutinized, and perhaps even to pay interest on those refunds rather than risk allowing fraudulent claims. We will continue to re- main vigilant in our fight against fraud.

Recognizing the importance of increasing compliance and includ- ing fraud prevention efforts, last year, Congress supported, with your help, additional funding of a 5-year compliance initiative that was going to cost $405 million a year, and we put in place a plan to generate an additional $9.2 billion in those 5 years which would be used for deficit reduction.

Unfortunately, less than half of the first year into that initiative, and it is already showing positive results, late last week, the possi- bility was raised that $100 million of the $405 million compliance initiative would be rescinded for this year.

I understand that Congress has to make difficult funding choices in these challenging times, but some cuts that might appear to produce a short-term benefit may not actually do so. Because every

24

$1 spent on compliance produces an average of $5 in revenue, the rescission proposal is not good business. It also will impact on our ability to continue our filing fraud efforts this filing season, and, if made permanent, certainly beyond this filing season.

At the urging of Congress, we have tried to change the way we do business, including enhancing our ability to prevent and detect fraud. A change in our funding, which could possibly not take effect until the last quarter of this year, really will impact on our ability to continue this program.

Mr. Chairman, that concludes my statement, and Mr. Brown and others and I will be happy to answer any questions you have.

Chairman Roth. Thank you very much, Ms. Richardson.

I do not question and I do not doubt your sincerity in seeking to address this problem of waste, fraud, and abuse. One of my con- cerns is the fact that, down through the years, we have had many other directors appear before us and make the same statement, and I am sure that they are all equally sincere and serious about it.

But what is so bothersome is that here we have, in the earned income tax credit, the fastest-growing program, I guess, on the books, with a tremendous percentage of waste, fraud, and abuse. I mentioned the figures from the General Accounting Office, show- ing that down through the years, beginning way back in 1982, the figure was 29 percent waste, and in 1985, 39 percent, in 1988, it was 34 percent, and in 1994, 24 percent. So we see this sort of on- coming figure.

My first question is, what do you estimate in 1994 will be the loss of taxpayers' money because of fraud and noncompliance in the earned income tax credit?

Ms. Richardson. I am not sure that we are really in a position to make that estimate. I mentioned the filing season study that we have underway. When those results are in, which will be at the end of the filing season, we will be delighted to share those with you. It will probably be early in the fall.

The estimates that were made last year, I believe, by the Treas- ury Department task force, were that it could be $1 to $5 billion a year. I would like

Chairman Roth. It could be more than $5 billion?

Ms. Richardson. One to $5 billion.

Chairman Roth. One to $5 billion a year?

Ms. Richardson. What I would hope, and what I hope our num- bers will show, is that our fraud prevention efforts and the filters we employed for this year will cause the numbers to be lower.

Chairman Roth. But that is a significant figure as of last year, $1 to $5 billion.

Ms. Richardson. That was the estimate that the Treasury task force had.

Chairman ROTH. You mentioned making a study involving a small sample of 1,000 returns during January 1994, and the pre- liminary results of the study show that 35 to 45 percent of the 1.3 million returns with EITC filed electronically through January 28 did contain errors, both up and down, in the amount of EITC claimed.

Ms. Richardson. Right.

25

Chairman Roth. That is a pretty substantial number at this time, at least if that preliminary finding is correct, would you

agree?

Ms. Richardson. Yes. I think Mr. Brown worked on that study and might be able to give you some more specific information. But I think when we complete the study from this year, we will have a much better sample, because we are doing both paper and elec- tronically-filed returns throughout the filing season.

Chairman ROTH. This study was just on electronic returns, wasn't it?

Ms. Richardson. It was just electronic, and it was just for, I be- lieve, a 2-week period in January. We are now doing a random se- lection throughout the filing season of both paper and electroni- cally-filed returns, I think 1,000 paper and 1,000

Mr. Brown. It will be between 1,600 and 2,000 returns this year.

Ms. Richardson. And when those numbers are in, I think we will have much better data that the Commission can use.

Chairman Roth. I wasn't clear from your earlier answer on the claims. As you said, the adjustments had to be made both up and down. You said up were 6 percent, is that correct?

Mr. Brown. Six percent of that roughly 35, 45 percent range. That is not 6 percent of

Chairman ROTH. I see. Six percent of the

Ms. Richardson. Of the 35 to 45 percent.

Chairman Roth [continuing] . Thirty-five to 45 percent had to be adjusted upwards. In other words, the taxpayer claim was too low?

Ms. Richardson. Correct.

Chairman ROTH. But 29 percent were the opposite, they were too high?

Mr. Brown. That is correct.

Chairman Roth. It is my understanding that you are trying to verify EITC claimants' eligibility before processing returns, but you cannot verify all eligibility requirements or criteria before sending refunds to taxpayers. My question is, what criteria are you using, then, in holding up the refunds on returns for 6 to 8 weeks and how many of those held up are not verified when they are finally released?

Ms. Richardson. I might say, and I will let Mr. Brown elabo- rate, but we are not really in a position to discuss exactly why and what criteria we are using because that involves the law enforce- ment information. If people knew why we were holding it up, they wouldn't claim that and they would try something else, so we have been very reluctant to discuss any of our screens and filters.

Chairman Roth. I certainly wouldn't want you to disclose any- thing. Can you answer it in a general manner?

Mr. Brown. I think the important thing, Senator Roth, is, first of all, we are not verifying all the EITC claims. The two categories that we are looking at are ones where there is a Social Security number problem. That obviously raises a question about the de- pendency or the parent.

The second group is a category that we styled as high risk, and it is the subcomponents of what determines high risk that I would prefer not to disclose, but in that high risk category, based on prior

26

years' results of audits and our research, those are the categories that are also being delayed until we can do some verification.

Chairman Roth. It is the area of EITC where most of the fraud and abuse is arising, is that correct?

Mr. Brown. We find fraud using just

Chairman ROTH. You find it everywhere, but percentage-wise, what percent involves the EITC?

Mr. Brown. The percentage of our fraud, and I think that is the important thing to remember, is you are only talking about what we know, but in that category, EITC is a large component. But we also find fraud in

Chairman Roth. Can you quantify that number?

Mr. Brown. Of the electronically-filed returns, it is in the 90- plus percent, but, also, a high percentage of EITC filers file elec- tronically, so it may not make it statistically.

Chairman ROTH. Ninety percent? Would you explain again what you mean by the figure, 90 percent?

Mr. Brown. Of the detected fraud, last year

Chairman ROTH. Ninety percent of them are EITC?

Mr. Brown. They involve the EITC as a component.

Ms. Richardson. Of the electronic.

Mr. Brown [continuing] . Of the electronically-filed returns that we identified as fraudulent.

Ms. Richardson. But not 90 percent of the electronically-filed re- turns.

Mr. Brown. Correct, of the detected

Chairman ROTH. Of those that are

Ms. Richardson. That were detected.

Chairman Roth. Let me ask you this question, Ms. Richardson, hypothetically. In determining your eligibility for the earned in- come tax credit, the amount of assets you have are of no signifi- cance. In other words, as long as your earned income is within the figures, the fact that you have a $1 million home or other assets doesn't disqualify you.

For example, I understand there is a case where a banker di- vorces his wife, settles the divorce, leaving her with the house, a large portfolio of growth stocks and tax-exempt bonds and the two children. I don't know whether that is a break or not. [Laughter.]

Although he does not pay alimony, he does pay $5,000 a month in child support, or $60,000 a year. He then hires her to work in one of his banks on a part-time basis for $10,000 per year. I under- stand this is an actual case and it was done to maximize tax sav- ings.

But it is true that the assets you hold in and of themselves do not disqualify you for this benefit, so you could be a multi-million- aire in assets and as long as it didn't bring in a certain disqualify- ing income, you would still be eligible for EITC. Is that correct?

Ms. Richardson. That is correct, Mr. Chairman. It is geared to your earned income as well as the number of qualifying depend- ents.

Chairman ROTH. I understand one of the proposals that is made from time to time is to disqualify one for these benefits if you have assets above a certain amount. One of the problems, I understand, or objections to that approach, is that it would be very complex for

27

IRS to have to determine what the assets of an individual are, and that is one of the reasons for not going in that direction. Do you care to comment?

Ms. Richardson. I guess I am not aware of what specific ideas have been discussed in the past, but I do know that when there are items that are on the face of the return or on schedules with the return, or where we can verify them through third-party reporting, we find it much easier to administer. When you are talking about assets and those issues, we do not have any independent way of verifying that in the course of processing a tax return, so it would make it more difficult for us to administer, I believe.

Chairman ROTH. And yet, it is pretty hard to justify to the tax- payer that he is being taxed to transfer funds to an individual who is, indeed, very wealthy, so that is an inherent problem of the pro- gram, isn't it?

Ms. Richardson. And I think that is something that, I believe, you said you will be getting into tomorrow, some of the policy con- siderations.

Chairman Roth. My time is up. I have some additional questions for the record.

Senator Glenn. Thank you, Mr. Chairman.

I apparently have been singled out a little bit because we held a hearing last summer, and I, apparently, from the letters that we have gotten from back home in Ohio, and from Dallas, for some reason or other, as the one who is to be blamed if checks are late or things are being held up or these loans don't get back in or they aren't refunded to people in a particular time period.

In effect, I am being criticized for helping ferret out fraud, I guess, and I am going to wear that one like a badge, although we have received quite a number of letters. It can't be just accidental. I don't know where they are coming from, but somebody has sort of an organized program because we held hearings last summer and criticized some of this and brought some of it to light. Appar- ently, now you have gone into a real program to try and address this problem, so you can share some of this with me. I will forward some of the letters over to you.

Ms. Richardson. I have received quite a few that I will be happy to share with you. I suspect they read exactly the same way yours do.

Senator Glenn. They are probably sending carbon copies to both of us, I guess.

Ms. Richardson. I will say, though, I think there has been an orchestrated effort on the part of some people who are not cor- porate Robin Hoods who would have you believe that they are — I think they are probably more concerned about their bottom line than they are about their clients. I think if they were more con- cerned about the people who are entitled to the earned income tax credit, they would make sure they were aware of the advance op- tion, might suggest that people lower their withholding so they aren't claiming such a large refund or a dependent on a large re- fund, and perhaps even lowering some of the rates.

Senator GLENN. I haven't understood, when we are trying to get down to people that are, in effect, cheating the system, when we are trying to get that out

28

Ms. Richardson. Absolutely.

Senator GLENN. I am glad to accept criticism like that when we are trying to get to the bottom of this.

Ms. Richardson. And I think the vast majority of the American people do support these efforts, because I think they are honest and law abiding and I think they understand they pay for the peo- ple who do cheat the system.

Senator Glenn. The point I was going to make was the truly honest taxpayers don't have anything to fear from this initiative. It is the people who are trying to cheat the system, and those are the ones you are trying to get on.

According to GAO, you have taken several steps in 1995 to re- duce fraud. Criminal and credit history checks on those newly ap- plying to file taxpayers returns electronically, do you want to com- ment on that one?

Ms. Richardson. Yes, we did. I believe we had about 33,000 new applicants this year. We did subject them to fingerprint and credit checks and we did find some who were not suitable, based on those criteria.

Senator Glenn. You are doing more comprehensive tests for missing or invalid Social Security numbers. You mentioned that one, I think, a little while ago.

Ms. Richardson. Right.

Senator Glenn. You are delaying refunds, primarily to the EITC claimants, to allow additional time to identify problematic returns, especially those using the same Social Security number that has been used on other tax returns. You mentioned that one a little while ago.

You are being attacked for inconveniencing or delaying some of the tax refunds or the payments to these people, while at the same time, we are allowing fraud that we should be taking out.

Let me ask, how much of the fraud here is by tax preparers? Do we have that broken down as to people who prepare their own re- turns as opposed to tax preparers?

Ms. RICHARDSON. No, we do not, Senator Glenn, and I think we want to be very careful, because, as I said in my statement, both my oral statement and my written statement, the vast majority of the tax return preparers, as well as the electronic return origina- tors, are honest and they have a vested interest in our getting the fraud out of the system, too.

Senator Glenn. Sure.

Ms. Richardson. Some of the people that purport to be tax re- turn preparers are setting up shop just for the filing season in va- cant lots, if you will, in some cases. They are really out there to prey on the suspecting taxpayers.

Senator Glenn. What is your overall error rate that you find, just in all taxes, the paper returns as well as electronic and so on?

Ms. Richardson. We estimate that the error rate for electroni- cally-filed returns is really less than half of one percent. In the paper processing, it is somewhere between 15 and 19 percent.

Senator Glenn. Between 15 and 19 percent?

Ms. Richardson. Yes, 15 and 19 percent.

Senator Glenn. Of all paper returns?

Ms. Richardson. On the paper returns, there are errors.

29

Senator GLENN. There are errors in them?

Ms. Richardson. Yes.

Senator Glenn. How many of those, then, would be fraud cases, where people are really trying to get away with something, as op- posed to just honest errors or small mistakes?

Ms. Richardson. I don't know. I will have to get that number for you. I am not sure that we have it. I don't have it with me today, if we have it broken out that way.

Senator GLENN. I certainly understand the fact that banks and other lenders have felt a pinch because the IRS no longer guaran- tees a refund is valid and will, in fact, be deposited directly to them, in effect, meaning that such loans are not secured by a gov- ernment guarantee. It could lead to higher charges or fees, of course. As I understand it, you are getting a lot of complaints from those people. Is that correct?

Ms. Richardson. I have heard quite a few complaints, yes. I be- lieve several banks decided not to continue in the business.

Senator Glenn. I haven't understood why you should be the guarantor of these things when they make the loans, supposedly have their own loan rules, and yet you are supposed to be the guar- antor of this and send the check to them previously, correct?

Ms. Richardson. That is what they would like us to do.

Senator Glenn. You are not doing it now.

Ms. RichaRdson. I don't understand why I should do that, ei- ther, nor should the American taxpayers be the guarantors, it seems to me. I have studied banking. I believe bankers are paid to assess the risk.

Senator GLENN. You are doing a number of things here that, you hope, will have an impact this year on straightening a lot of this out and catching a lot of the fraud involved. Longer term, will the tax system modernization program, if properly implemented, help to alleviate some of the delays caused by EITC compliance efforts?

Ms. Richardson. Most definitely. As I think I have said before this Committee several times and in many forums, fraud is some- thing that we always have to be alert to. But if we have more cur- rent access to information and are able to process it more quickly, take advantage of the fraud detection capabilities of a more modern system, we will go a long way towards eliminating it from our sys- tem completely.

Senator Glenn. One of the things that we talked about last sum- mer when we had our hearing were the violations, as I saw it, of browsing and people getting into accounts that they had no busi- ness in because they weren't working those accounts legitimately at IRS. You have taken some steps to stop that kind of browsing.

What has been the impact of IRS's revised penalty guide and what you call the EARL system, electronic audit research log, the computerized search for browsing patterns? Is that having a good impact now?

Ms. RICHARDSON. It is having a very positive impact, and we are reporting to this Committee on a monthly basis of what steps we have taken and their impact. I would be delighted to get you that up-to-date information.

30

We also have a sign-on warning, before anyone logs onto the computer, that reminds them of what their obligations are and what the penalties are.

Senator Glenn. I am introducing legislation today that I had dis- cussed with you earlier to close that loophole, because right now, there is a loophole there where browsing is not as much against the law as some other things, so we are going to tighten that one.

Ms. Richardson. There is some ambiguity about whether there are criminal penalties available, and we appreciate your support.

Senator Glenn. Thank you. On the browsing cases, how many new cases have you found, say, this year or in the last 6 months? Do you have figures on that?

Ms. Richardson. I would like to give those to you for the record, if I could, just to make sure I give you the precise number.

Senator Glenn. That would be fine. Have you caught some peo- ple there and have any of them been fired for their browsing ef- forts?

Ms. Richardson. We have. We have dismissed people and we also have people who are being prosecuted, and we will get you the precise numbers.

Senator Glenn. If you could get that, we would like that for the record.

Ms. Richardson. I also want to emphasize, as I have before, that we will not tolerate even one person abusing taxpayers' rights of privacy and confidentiality, but it is also a very small number, a very small percentage, I think less than one percent of the people, that we have ever detected.

Senator Glenn. Can you run through the accounts receivable? That is one we go through every time you come up before us, and I am very interested in that one. Accounts receivable in general, now, is up to what, $150-some billion? A lot of that, of course, is in bankruptcies, individual or corporate, but what is actually col- lectible out there, right now? We got you some additional FTE's, full-time employees, 5,000 or whatever it was, that we hope will go out.

Ms. Richardson. Correct.

Senator Glenn. The figures in the past have been that each em- ployee is worth about five or six times that person's keep, and so it is a good return.

Ms. Richardson. Correct.

Senator Glenn. Of the $25 or $30 billion that is out there and actually collectible, we estimate, in collectible funds, how are we doing on that? It makes everybody hopping mad when they think people are getting away with something and they have the money but just aren't paying it, and we don't go out and get it. What are you doing in that regard?

Ms. Richardson. Absolutely. Our collections are up for, I guess, this year and part of last year, and the numbers for the amount of that that are collectible, Mr. Dolan has right here at his finger- tips.

Mr. Dolan. You know, Senator Glenn, this is one of those sub- jects where we can get twisted and twisted on numbers. We are given the financial audit that we undergo with the General Ac- counting Office. Today, we would identify in a category we call

31

total active accounts receivable $88 billion. When we look at that, we net a few things out and come up with what we call a net active of $79 billion. We apply, with respect to the general accounting rules, what you would call an allowance for doubtful account, and basically, when we talk about the collectible, the part with direct intervention we feel reasonable certain is collectible, is today listed as $26.7 billion.

So in terms of where we would first target our collection assets, it would be against that. That is not to suggest that some other part of that allowance isn't ultimately reachable, but that is basi- cally what we would first direct our collection resources against.

Senator GLENN. And you are putting effort into that now? What new things are we doing in that area to go out and get it?

Mr. DOLAN. One of the things the Commissioner mentioned ear- lier is the $405 million initiative for this year that the Congress supported and is now subject to some scrutiny under the rescission. That thinking process really was heavily directed into both our automated collection sites and our field collection sites, and so a very significant portion of that $405 million initiative that we said would produce $9 billion over 5 years is directed exactly at this problem.

Senator Glenn. Mr. Chairman, my time is up, but I just hope in all the budget cutting efforts around here, if anything, we can expand the funds available to the IRS, because I think we get back far more than we spend, just in the examples that were given here just now. We get back about five times the keep of each new agent over there, and we can go out and get some of that near $30 billion.

We fuss over all sorts of little $10 million, $100 million here and there, which is big money, of course, but compared with what is out there to be brought back into the Treasury, and should be, and is collectible — it is not the big $100 billion-plus figure that is not col- lectible because of bankruptcies or whatever, this is money that is collectible if we go out and get it.

I just hope we can keep the tax system modernization on track and keep it well-funded so we can go out and get that money.

Ms. Richardson. I think it is also important to remember that you have to plan your strategies and you can't be subject to the va- garies of a stop-and-start process every few months. Some of those collection initiatives have taken a year or more to plan and we are now starting to realize the benefits of some of those and actually starting to realize some of the benefits of that compliance initiative. But the real benefits are going to come in the second, third, fourth, and fifth years.

Senator Glenn. Mr. Chairman, if I might just have an additional 30 seconds, I was just handed a note by staff back here, and I didn't know I was going to get into this subject this morning, but I understand we have an amendment that has been proposed by Senator Grassley that would cut $100 million off of your funding. It is a proposed $100 million reduction from tax law enforcement, which has the effect of reducing compliance activities.

I guess Senator Grassley was unhappy that the $405 million ap- propriated last year for new compliance initiative was provided out- side the discretionary caps. I am concerned we go outside those caps, also, but what it says here is that if this was a successful cut,

32

IRS would have to furlough most of your compliance staff for at least 10 days and there would be a corresponding loss of revenues. Is that a correct statement?

Ms. Richardson. That is correct.

Chairman Roth. I am sure that the Finance Committee will give careful consideration to your request, Ms. Richardson, and there is no question that, at least, in my judgment, that one of the best ways of addressing the problem of fraud, waste, and abuse is to uti- lize modern technology. I am a strong supporter of anything that can be done, because it does seem to me that that is a large share of the ultimate answer to the problems we are having in income tax returns.

That does bring me, Commissioner, to an article from Tax Notes which makes the statement that the IRS has rejected thousands of legitimate tax returns — this is a copy of the statement — that the IRS has rejected thousands of legitimate tax returns because the IRS computer filters are not sophisticated enough to match data with the Social Security Administration's records.

In the past, apparently, the IRS has maintained that errors and fraud were the sole reason for rejecting returns while Social Secu- rity seems to blame IRS for having a poor system to match these Social Security numbers. My legislation from last year, and now the Treasury proposal, would require this matching process.

Do you care to comment on the problem?

Ms. Richardson. I would. There was a further article, which I will be happy to provide you and your staff, in yesterday's Tax Notes which was a follow-on to that which corrected the record and also corrected the statements of the Social Security Administration. Their Deputy Commissioner for Systems, "contradicted a staff member and said there is no magic, because we all use the same data base."

The article you read, I believe, was essentially in error, and the processing errors that were

Chairman Roth. So it is a different

Ms. Richardson. The first article was not correct.

Chairman ROTH. But there is apparently a conflict between

Ms. Richardson. There is not conflict

Chairman Roth. I am sorry, what did you read from the staff?

Ms. Richardson. There is no conflict between us and the Social Security Administration.

Chairman Roth. At the management level. What about the staff?

Ms. Richardson. No one knows who made that comment, so we cannot verify it, but we have been assured that we all use the same data base and it has the same information in it. Each of the errors that were provided to us, we verified with them and did not have anything to do with our systems. I would be happy to give you that information.

Chairman Roth. The General Accounting Office has been critical of some of the technological developments in your organization.

Ms. Richardson. They have.

Chairman Roth. I would urge you to work with them, be- cause

Ms. Richardson. They have, and we are actually working very closely right now. Mr. Bowsher and I have a group looking to ad-

33

dress all of the tax systems modernization issues, and, I believe, by the end of this month, we will be reporting to you and to other Members of Congress about what we have found.

Chairman Roth. Let me ask you this question. The IRS has com- mitted significant resources to the issue of fraud in the EITC. How much of your fraud resources are devoted to the EITC?

Ms. Richardson. We don't actually break out by specific cat- egories of fraud. Last year, we had about 854 full-time equivalents working on fraud and fraud-related issues. This year, we have stepped up our efforts and we have about 2,000, just over 2,000 full-time equivalents that would be addressing all kinds of refund fraud.

.Chairman ROTH. But you have no estimate of what percentage are devoted to EITC?

Ms. Richardson. No, and it is really not broken out that way.

Chairman ROTH. Let me ask you an entirely different question. There is quite a significant marriage penalty in the program, is that not correct, in the EITC, in the sense, for example, you had two married people living together, having four children under age 18, and each of the married people have a job earning about $11,500 per year. How much would they increase their EITC bene- fits if they simply got a divorce, although they continued to live to- gether, and each parent took two children as a part of the divorce?

Ms. Richardson. I am not sure exactly the dollar amount, but you are right. The way the law is written, there is a reduction, I guess, in that case. If you had two single individuals, their earned incomes would be counted separately.

Chairman Roth. So it does encourage people to live together and not be married, I guess.

Turning to the issue of illegal aliens, are illegal aliens now eligi- ble for EITC?

Ms. Richardson. Under the way the law is written today, yes, sir, that is true.

Chairman Roth. Have any recommendations been made by the administration concerning that?

Ms. Richardson. Yes. I believe it was in the budget that was submitted. The fiscal year 1996 budget contained a change to pre- clude undocumented workers from claiming the earned income tax credit.

Chairman Roth. What number of illegal aliens are now securing the earned income tax credit? Do you have any estimate?

Ms. Richardson. I don't think we have any estimate. The earned income tax credit is really paid based on, in the way the law is written today, it is based on whether or not you earn income, not your citizenship status.

Chairman Roth. But there are a significant number. I have heard the figure as high as 166,000 illegal aliens are benefiting from the program. You have no information or figures on that, of any type?

Ms. Richardson. I don't, but we can check and see

Chairman Roth. Isn't that a serious problem?

Ms. Richardson. I believe there are 20 million families who are supposedly eligible, so that doesn't seem to be a large number. The way the law is written today, if you earn income, no matter what

34

your status, you are eligible for the earned income tax credit. In order to claim it, presumably, they are paying taxes, as well, so it is being claimed against income they earned.

Chairman Roth. The General Accounting Office estimates that 160,000 taxpayers out of about 8.7 million who filed paper returns were likely to be illegal aliens. Even though there are 8.7 million, if there are 166,000 filers, and if they get $1,000 each, that would be a pretty significant sum, wouldn't it? It would be about $166 million.

Mr. DOLAN. Senator Roth, if I might, I think that number most likely comes out of our need to assign a temporary number of some sort in order to process what is, in many instances, somebody who has earned wages, paid taxes, due perhaps a regular refund as well as his claiming under the current law of the EITC.

Chairman Roth. This is a matter we will be looking further into.

Ms. Richardson, I want to thank you and the two gentlemen for being here today. We may have some additional questions, so we will keep the record open for the next 2 days.

Ms. Richardson. I would be delighted.

Chairman ROTH. But thank you for being here. We look forward to working with you on trying to address what I consider an ex- traordinarily important problem.

Ms. Richardson. And thank you, Mr. Chairman and Senator Glenn. We look forward to working with you to make the system work for everybody. Thank you for your support.

Chairman Roth. Thank you.

For our next witness, I would like to introduce Lynda Willis and Mike Brostek from the General Accounting Office. They have been working for over a year to look at some of the waste and abuse problems that face us in the EITC program. I want to welcome you here today and say we look forward very much to your testimony.

Ms. Willis, it is indeed a pleasure to have you here. Do you want to introduce your companions?

TESTIMONY OF LYNDA D. WILLIS,1 ASSOCIATE DIRECTOR FOR TAX POLICY AND ADMINISTRATION, U.S. GENERAL AC- COUNTING OFFICE; ACCOMPANIED BY MICHAEL BROSTEK, ASSISTANT DIRECTOR, TAX POLICY AND ADMINISTRATION ISSUES; AND JENNIE STATHIS, DmECTOR, TAX POLICY AND ADMINISTRATION ISSUES

Ms. WILLIS. Yes, thank you, Mr. Chairman. I would just like to wait a moment until we get our storyboards up on the easel.

I believe we are about ready, Mr. Chairman.

Chairman Roth. Please proceed.

Ms. Willis. I have with me today Jennie Stathis, who is the Di- rector of Tax Policy and Administration Issues at GAO and Mike Brostek, who is our Assistant Director responsible for our ElC-re- lated work.

We are pleased to be here today to assist you in your on-going efforts to better ensure that only the working poor receive the earned income tax credit, or EIC. As you requested, Mr. Chairman, we are releasing today the second in a series of reports done for

1 The prepared statement of Ms. Willis appears on page 136.

35

you on the EIC. I ask your permission to have this report put in its entirety in the hearing record.1

Chairman Roth. Without objection.

Ms. Willis. My testimony today will briefly summarize the re- port, in which we address EIC noncompliance issues and options for measuring recipients' economic resources in determining eligi- bility for the credit.

Originally authorized in 1975, the EIC provides assistance to low-income working taxpayers to offset the impact of Social Secu- rity taxes and to encourage them to work. At various times, Con- gress has broadened EIC coverage and increased the credit amount to ensure it would not fall in purchasing power, to increase or maintain the progressivity of the tax system, and to better ensure that working individuals would have incomes above the poverty line.

As Figure 1 in my testimony illustrates, which is also the figure that is on our storyboard, with these changes, the overall cost of the EIC is expected to increase more than five-fold, in real terms, between 1988 and 1996, when EIC costs are estimated to exceed $24 billion.

Chairman Roth. So it goes from roughly $5 billion to $25 billion, is that what the chart shows?

Ms. Willis. It goes from roughly $4.4 billion in 1988 to roughly $25 billion in 1996, Senator, yes.

The most recent changes to the EIC in the Omnibus Budget Rec- onciliation Act of 1993 increased the maximum credit available and the income level at which individuals can qualify for the credit. For the first time, it also granted eligibility to certain low-income tax- payers without children.

As Figure 2 illustrates, the credit gradually phases in, plateaus at a maximum amount of $3,370 for a taxpayer with two qualifying children in 1996, and then phases out until it reaches zero at $27,000 in income.

Turning now to noncompliance, which is defined as erroneous EIC claims caused by negligence, mistakes, confusion

Chairman Roth. Could I interrupt you?

Ms. Willis. Sure.

Chairman ROTH. As long as we are going through this, I would like to ask a question. As I understand it, from $1 to $8,424, for every additional dollar I earn, I become eligible for an earned in- come tax credit of 40 cents?

Ms. Willis. That is correct.

Chairman Roth. So I, in effect, earn $1.40. So that is an incen- tive to get off the welfare roll, as I understand it.

Ms. Willis. Right.

Chairman ROTH. Then when you come up to the $8,424 figure, I receive the maximum amount that I can get under the program, which is $3,370. So whether I earn $8,424 or $10,999, I still get a $3,370 earned income credit?

Ms. Willis. That is correct.

Chairman Roth. But then, it begins to phase out.

Ms. Willis. Right.

1 The report of the General Accounting Office appears on page 164.

36

Chairman ROTH. So, in a sense, you begin to have what you might contend is a disincentive, because by earning more, you get less earned income tax credit, is that correct?

Ms. Willis. That is correct.

Chairman Roth. So every dollar I earn over $10,999, I lose 21 cents of my earned income tax credit?

Ms. Willis. Right.

Chairman ROTH. And that ultimately phases out at $27,000. So on this side, there is a definite incentive, but on the other side, you can argue that it does partially act as a disincentive.

Ms. Willis. Yes.

Senator Glenn. Mr. Chairman, by going to a higher employment, though, you are still phasing out.

Chairman Roth. That is right, but I am just pointing out, under this chart, what it appears to mean.

Senator Glenn. I know, but I don't see it as a disincentive. I think as long as people's overall income is going up, it is an incen- tive to get a better job, and you are being helped during that time period while you are expanding your earning power until you get out there to the $27,000. Then, you are out of the program.

Chairman Roth. That is the intent, certainly.

Senator Glenn. It would still be an incentive. It wouldn't be a disincentive, as I see it, for somebody to say, no, I don't want to get a better job. I don't want to earn more because I am going to have to lose this 20 percent. They still have an increase of 80 per- cent, if you want to put it on that basis. Is that correct?

Chairman Roth. Yes, if I understand it, their total income would be higher as it goes up.

Senator Glenn. Sure.

Chairman Roth. I think that is correct.

Ms. Willis. Right. Their total income with their earnings would be higher, but their credit would be reduced in the phase-out range.

Senator GLENN. As they go on to higher levels.

Ms. Willis. Absolutely.

Senator Glenn. That sounds good to me.

Chairman Roth. Please proceed.

Ms. Willis. Turning to noncompliance, EIC noncompliance has been and continues to be a problem. For example, compliance measurements done by the Internal Revenue Service in 1988, that you cited earlier, Mr. Chairman, estimated that about 42 percent of EIC recipients received too large a credit and about 35 percent of the total EIC paid out may have been awarded erroneously.

Although a current statistically valid measure of overall EIC compliance does not exist, the results of limited studies and of IRS efforts to enforce the EIC suggest that a significant compliance problem remains. An IRS study of electronically-filed returns dur- ing a 2-week period in 1994 found an estimated 29 percent of the returns claimed too much EIC. Thirteen percent of these returns were judged to have intentional errors, a surrogate measure for possible fraud.

This filing season, IRS has expanded its efforts to ensure EIC compliance. In doing so, it is using lessons learned from its 1994

37

study and enforcement experience to improve its systems to iden- tify possibly non-compliant returns.

Verifying Social Security number accuracy is key to IRS's efforts. IRS checks the accuracy of Social Security numbers, expanding their efforts this year to emphasize those used for dependents and ElC-qualifying children.

As of March 17, 1995, IRS verification procedures had identified nearly 4.1 million instances of problems with returns. These pri- marily involved returns that did not appear to contain valid Social Security numbers for dependents or qualifying children.

In addition, as of that same date, IRS had delayed refunds to at least 2.9 million EIC claimants for up to 8 weeks. This delay is in- tended to allow IRS staff time to identify any Social Security num- bers that have been used on more than one tax return. Commis- sioner Richardson referred to this effort earlier.

IRS identified duplicate Social Security numbers as a problem during the 1994 filing season. For the delayed returns, IRS gen- erally sends out the portion of any refund that was due to overpay- ment of taxes but withholds the EIC portion of the refund claimed.

Not surprisingly, with a large new initiative, IRS experienced some problems as it began checking for duplicate Social Security numbers. These problems included difficulties in constructing the data base to use in identifying the duplicate numbers, poorly-orga- nized computer listings that enforcement personnel found difficult to use, and cumbersome procedures for coordinating among IRS service centers.

IRS national office officials told us that initial problems with the duplicate Social Security number system had been overcome early in the filing season, but compliance personnel continue to report problems using the duplicate Social Security number data. We in- tend to continue monitoring this effort for you.

Although it is too early to assess the success of IRS's new or ex- panded enforcement initiatives, the steps taken seem to be focusing appropriately on current indicators of problematic returns.

Despite IRS's efforts to better verify EIC claimants' eligibility be- fore processing refunds, the IRS currently cannot verify all eligi- bility criteria before sending refunds to taxpayers. In the long run, sound enforcement of the earned income credit may require even better verifications of recipients' eligibility before refunds are made.

We have made several recommendations in the past that could help to make the EIC less of a problem for IRS and taxpayers. As discussed more fully in an appendix to this testimony, those rec- ommendations call for eliminating differences between the defini- tion of a qualifying child for EIC purposes and the definition of a dependent for purposes of claiming a dependency exemption; en- couraging the advanced payment option, whereby persons eligible for the EIC can choose to receive it in advance as part of their pay- checks; and moving toward timely computer matching of employer wage information with tax return data.

Turning now to ways for better measuring EIC filers' resources to determine eligibility, although the EIC is intended to provide as- sistance to the working poor, unlike certain welfare programs, tax- payer wealth is not taken directly into account in determining EIC

38

eligibility or the amount of credit received. EIC criteria also do not consider all types of income that taxpayers may receive.

At your request, we assessed the potential changes in overall EIC costs that might result from including a wealth test and a more comprehensive adjusted gross income test in determining eli- gibility. We also evaluated the administrative implications of ex- panding the eligibility criteria. Generally to facilitate administra- tion of the expanded eligibility criteria, we initially looked at items that are currently reported in some form on the individual's income tax return.

For the wealth test, we analyzed asset-derived income, such as taxable interest and dividends, tax-exempt interest, estate and trust income, rental income, and capital gains. For the expanded adjusted gross income test, we first analyzed the impacts of includ- ing non-taxed Social Security income, tax-exempt interest, and non- taxed pension distributions in the taxpayer's AGI. At your request, we subsequently added child support payments, which do not cur- rently appear on any IRS form, to the income items.

Based on our work, you requested that the Joint Committee on Taxation provide revenue estimates for various eligibility options that we reviewed. According to Joint Committee on Taxation esti- mates, denying the EIC to taxpayers whose income from wealth ex- ceeds a certain threshold could reduce program costs $318 to $971 million in fiscal year 1997, depending upon the design of the test.

The board that we have on the easel now, Mr. Chairman, lays out the options that we looked at, the income thresholds that the Joint Committee analyzed for you, and the estimated revenues from the various thresholds and accompanying options.

It should be noted, Mr. Chairman, that the revenue estimates provided by the Joint Committee would have to be reduced by any amount incorporated into the Act recently passed for the self-em- ployed health insurance deduction.

Expanding taxpayers' adjusted gross income to include non-taxed Social Security income, tax-exempt interest, and non-taxed pension distributions could yield $1.45 billion in the same period. Also, add- ing child support payments to the expanded adjusted gross income would increase 1997 revenue savings by another $686 million.

However, adding an indirect wealth test or an expanded adjusted gross income definition to the eligibility criteria would add to the EIC's complexity and administrative burden. Complexity has been a continuing EIC issue, because it can lead to increased errors and dissuade deserving taxpayers from claiming the credit.

Of the potential changes to EIC criteria, adding child support to taxpayers' adjusted gross income likely would cause the greatest complexity, because information on such income is not collected by IRS and systems may not exist to generate the information.

In addition, there are significant limitations in measuring poten- tial EIC recipients' actual wealth through the income reported on tax returns. For instance, such a test would not measure the value of taxpayer assets such as capital stock funds that yield little, if any, annual income. These limitations could raise concerns that taxpayers with similar wealth could be treated differently.

Turning now to the eligibility of illegal aliens, the Internal Reve- nue Code does not prohibit illegal aliens from receiving the EIC if

39

they meet the prescribed eligibility requirements. However, illegal aliens cannot be employed lawfully in the United States. Because the EIC is intended, in part, to encourage employment, it works at cross purposes with the prohibition on employment of illegal aliens.

Although no one knows how many illegal aliens may be claiming and receiving the EIC, IRS estimated that a minimum of 160,000 out of 8.7 million taxpayers who filed paper returns claiming the EIC in 1994 were likely to be illegal aliens. IRS expected most of these refunds to be denied because taxpayers could not support their claims by verifying that the dependents met the age, relation- ship, and residency requirements.

Some unknown portion of returns may also be filed by illegal aliens who use Social Security numbers belonging to other individ- uals. IRS's new enforcement efforts, if successfully implemented, should reduce the number of illegal aliens as well as U.S. citizens incorrectly receiving the EIC.

A Senate bill you introduced in 1994 and the administration's Tax Compliance Act of 1995 would deny the EIC to illegal aliens. The administration's proposal would require that all EIC recipients provide Social Security numbers that are valid for employment in the United States for themselves, for their spouses, if applicable, and for qualifying children. Because illegal aliens cannot qualify for Social Security numbers that are valid for employment in the Unit- ed States, they would not be able to receive the EIC.

The administration's proposal would permit IRS to use stream- lined procedures to enforce the requirement that EIC claimants have valid work-related Social Security numbers. The administra- tion estimates that requiring all EIC recipients to provide valid work-related Social Security numbers and using streamlined proce- dures to enforce this requirement would yield about $400 million in revenue savings in fiscal year 1997.

Mr. Chairman, that concludes my prepared statement. My col- leagues and I would welcome any questions you may have.

Chairman Roth. Thank you, Ms. Willis.

I note that you say, verifying Social Security number accuracy is key to the IRS's effort. One gets a dimension of that problem when you read, as your testimony provides, that as of March 17, 1995, IRS verification procedures had identified nearly 4.1 million in- stances of problems with returns.

Ms. Willis. Yes, sir.

Chairman Roth. That is a significant number. That would be out of what? We don't know how many returns had been filed by that time, do we?

Ms. Willis. And also, that number, Senator, is instances. As the Commissioner was relating earlier, there could be more than one instance on a particular return.

Chairman Roth. Yes, but there are 4.1 million instances where something could be wrong?

Ms. Willis. Yes, sir.

Chairman Roth. There could be more than one on a return.

In your testimony, you point out that if we include, for example, a wealth test, funds could be saved. In other words, if you have a wealthy individual, he might become ineligible for EITC, but it is extremely difficult to administer those kind of requirements — the

40

complexity of the program becomes substantially greater. Is that correct?

Ms. Willis. Adding a wealth test to the EIC would make it more complex. When we identified items in our options for the wealth test, one of the reasons that we selected items that were listed in some form on the return was to minimize, or at least reduce, the amount of administrative complexity that would be involved.

Chairman ROTH. Will you give some examples of that, again, of some of the things that you could include that wouldn't be too com- plex?

Mr. BROSTEK. Sir, the chart that we have here includes the items that we analyzed in our test.

Chairman Roth. Would you list those again?

Mr. Brostek. Yes. It is taxable interest, taxable dividends, tax- exempt interest, net estate and trust income, rental

Chairman Roth. So exactly what would you do? So everybody understands what you are talking about, what exactly would you provide?

Mr. Brostek. The way a test like this would work is the tax- payer would add up their income from these various sources and if that income exceeded some threshold, they would not be eligible.

Chairman Roth. They would no longer be eligible?

Mr. Brostek. So, for instance, we analyzed the options of a $1,000, $1,500, or a $2,500 threshold, and the estimates that are provided here are the ones that were provided by the Joint Tax Committee. If the taxpayer reported and added up tax-exempt, tax- able interest, and taxable dividends and the total of those exceeded $1,500, then the taxpayer wouldn't be eligible, and, according to the JCT estimates, about $505 million would be raised in fiscal year 1996.

Chairman Roth. That, of course, is partly what we did yester- day.

Mr. BROSTEK. Correct, and that is why these sums would have to be reduced by the Act that was passed yesterday.

Chairman ROTH. Let me ask you, how much have the taxpayers of this country lost from noncompliance in the EITC since inception of the program, based on past estimates of noncompliance in the program?

Ms. WILLIS. Mr. Chairman, recognizing that there is no com- prehensive study or estimate for all of the years since the program was put in place, if you apply a 30 percent figure, which is within the range of most of the studies that have been done, if you apply a 30 percent figure to the $81.6 billion that was paid out between 1976 and 1993, the 30 percent of that is $24.5 billion.

Chairman ROTH. Twenty-four-point-five billion dollars, in other words, didn't go to the working poor for whom it was intended but to those who were guilty of fraud or abuse of some sort?

Ms. Willis. Those returns that were non-compliant. It was not

Chairman Roth. Non-compliant. Could you estimate how much of the figure is due to fraud, as opposed to mistakes, confusion, and other problems with the tax law?

Ms. Willis. No, Senator, you really can't. The only number that we have that is a surrogate for fraud, the intentional noncompli-

41

ance from the 1994 study, indicated that about 12 percent of the dollars, or 13 percent of the returns, the noncompliance was inten- tional, but that is a very narrow study. We should have a much better figure on that after the 1995 study is done.

Chairman ROTH. But to date, using the 30 percent figure, you would estimate how much?

Ms. Willis. Through 1993, using the 30 percent figure, it would be $24.5 billion.

Chairman Roth. Twenty-four-point-five billion dollars. How much, based on past and current estimates, will the taxpayers lose in the next 5 years from noncompliance and fraud?

Ms. WILLIS. As problematic as applying our existing estimates to past payouts is, it is even more problematic for the future because of potential changes to the credit, changes in IRS's technology to identify noncompliance before the money is paid out. But we cal- culate that between 1994 and 1998, for that 5-year period, based on the President's budget, $122.7 billion will be paid out in the EIC program. Thirty percent, if we, again, use that figure from past studies, 30 percent of that number is $36.8 billion.

Chairman Roth. Whatever the actual figure is, it is a very sig- nificant sum, if past is prologue.

Ms. Willis. Yes. We think that noncompliance in the EIC pro- gram is a significant problem.

Chairman Roth. Let me ask you this. Out of the last 15 to 20 years that you have been looking at the error and fraud rate in the EITC, has there been a constant level of error rate in the program? Has it ever dropped below 25 percent?

Ms. Willis. The taxpayer compliance measurement program, which is what IRS uses to measure noncompliance, that has looked at the EIC, one was done in 1982 and the error rate there for the dollars involved was 29 percent. For 1985, the error rate was 39 percent. For 1988, the error rate was 34 percent. So for those three study years, the lowest error rate was 29 percent.

But I would stress that there have been a number of changes made to the program since then that

Chairman ROTH. Hopefully.

Ms. Willis [continuing]. Respond to some of the concerns that were raised in those studies.

Chairman Roth. Up until recently, IRS has generally paid EITC refunds to known illegal aliens. In fact, they have identified certain returns with 205(c) written in the space for a Social Security num- ber as one way to tell who illegal aliens might be. Why has IRS only recently begun stopping these claims, and how much of the taxpayers' money could be estimated to have been lost to EITC fil- ers and dependents who are illegal aliens claiming the EITC?

Mr. Brostek. Sir, if I could respond to that

Chairman Roth. Please.

Mr. Brostek. IRS has not actually begun stopping awards to il- legal aliens because they do consider themselves required by law to provide it as long as an illegal alien meets the eligibility criteria.

What IRS has done is they have identified, and this is the 160,000 figure that we are talking about, they have identified a universe of tax returns where that code was entered for a qualify- ing child. Because that code was entered and from their enforce-

42

ment experience, they have reason to believe that they have likely been filed by illegal aliens.

But that universe was not actually paid out. They stopped those returns so that they could correspond with the taxpayer and try to find out whether they were actually qualified to receive it, and IRS believes that they probably won't pay out most of that because the taxpayers will not come back with the qualifying information.

Chairman Roth. We may have additional questions, so we will leave the record open. We do appreciate your being here today.

Thank you, Ms. Willis, and other members of the panel.

Ms. Willis. Thank you, Mr. Chairman.

Chairman Roth. We will next call up Mr. Hersch, who was a former President of the defunct Quik Tax Dollars, Inc. Mr. Hersch was convicted of filing false tax claims, money laundering, and bank fraud.

Mr. Hersch, we are pleased to have you here today. We would ask you to introduce who is here with you. We, of course, expect you to answer the questions, if he is your attorney.

Mr. Feinberg. Good morning, Senator. May I introduce myself? I am Matthew Feinberg with the Boston law firm of Stegele and Feinberg and I am here, obviously, in a representative capacity. Mr. Hersch has been invited here, and, obviously, he will be ad- dressing you, Senator, and the panel.

Chairman Roth. Mr. Hersch, will you please proceed?

TESTIMONY OF RICHARD M. HERSCH,1 FORMER PRESIDENT, QUffi TAX DOLLARS, INC.

Mr. Hersch. Good morning, Mr. Chairman and Members of the Committee. I was invited here to discuss the ease with which fraud can be perpetrated on the IRS's electronic filing system. I am grate- ful for the opportunity to appear before you today regarding these problems and specifically the easy target that earned income tax credit makes for people like me who are tempted by its invitation for abuse.

I have some qualifications to speak on this issue. In 1993, I was indicted for filing false tax returns and for money laundering that I committed as a tax preparer for the tax years 1991 and 1992. I plead guilty to all these charges in January 1995, and I am cur- rently awaiting sentencing in the U.S. District Court in Boston.

I would like to describe briefly for you my background, my tax preparation business, and the nature of the schemes I used to de- fraud the Government.

I got into the tax preparation business because, in 1990, I loaned someone else money to start a tax business. He took off, leaving me with computers and equipment that he purchased and 1,000 tax re- turns to file. I had no license or other certification to prepare tax returns myself and no experience, but I quickly discovered that none of this was, indeed, needed, not even to prepare and file the 1,000 returns that my erstwhile business associate had left for me. Those returns had been sent to him by check cashing businesses in the Philadelphia area.

'The prepared statement of Mr. Hersch appears on page 141.

43

Before continuing with my story, I should like to briefly explain the nature of the check cashing business and their customers. The check cashing businesses that I dealt with the first year and in subsequent years were located in depressed inner-city neighbor- hoods. The customers were usually poor. The customers did not have bank accounts. They used check cashing businesses to cash their paychecks, welfare payments, to pay their bills, and to con- duct their other basic day-to-day transactions.

These businesses charge a high rate, often a high flat fee or a percentage of the check or the transaction, in order to handle their customers' business. As I learned, they also advertised and pro- vided tax preparation services. Those services were usually pro- vided by an outside and unlicensed preparer, like myself, who would receive customers for the check cashing businesses.

Returning to my story, I prepared and filed in 1990 all of the 1,000 returns that had been left by my business associate, and I did so honestly, relying on the information supplied by the cus- tomers to the tax preparation businesses. I received only my legiti- mate filing fee. In the course of reviewing the customers' tax infor- mation, however, I discovered clear evidence that many of them were lying to the IRS on their returns in order to take advantage of the EITC, and they were getting away with it, with no problem.

The customers were falsely overstating the number of their de- pendents, lying about their status as head of household, and claim- ing to have earned wages for work that they never did by falsifying the W-2 that they submitted to me. Maybe because there were so many people doing it, and maybe because of the small amount of money involved for each taxpayer, the IRS never seemed to care.

EITC could stand for "Easy Income for Tax Cheats". As I discov- ered, it was a particularly popular vehicle for fraud for several rea- sons. First, the EITC was a refundable credit that a taxpayer gets depending on how little they earned within a range that qualified them as low-income taxpayers.

Second, the EITC benefits low-income taxpayer to a certain ex- tent, only if they show dependents on their returns. The taxpayer claims some wages, but not too much, adds up two dependents, and thereby maximizes his refund, without adequate verification.

Third, the EITC was so well-known in the tax preparation busi- ness that it was promoted and check cashers heavily based much of their short-term lending business in the form of a refund antici- pation loan, or RAL.

My business was coming from check cashers and tax preparation companies. I assembled a network of check cashing companies in 25 states that contracted to have me prepare and file tax returns for their customers for the coming year. My business grew dramati- cally. My company filed a total of 9,000 returns in 1992 for the tax year 1991, which netted my customers approximately $18 million in total refunds. Of that total, I would guess that roughly half of the returns contained false information about dependents, wages, or filing status that allowed the customers to receive more money through the EITC than they were entitled to.

That year, in 1992, I recognized how easy it was. All I needed were fake names and Social Security numbers that I made up by using any combination of nine numbers. I never worried about the

44

IRS cross checking any of this information. I just made up fictitious returns for fictitious filers. I personally filed 200 fictitious returns that year, which netted me roughly $300,000 to $350,000 in re- funds.

In addition to that money which I made from my fictitious filers, I also made money processing the false returns given to me by real customers. For these customers, I, and often the tax preparation company who had sent me the customers, would keep a portion of the taxpayer's overstated EITC refund.

In 1993, business continued to boom. I formed a new company, Quik Tax Dollars, which was a joint venture between myself and a company called Monetary Management Corporation, which owned 130 check cashing businesses. I ran the day-to-day oper- ations of this tax preparation and filing company. That year, we filed 29,000 tax returns, resulting in roughly $50 million in re- funds. I would estimate that 40 percent of those refunds were based on falsely obtained EITC credits.

As with the year before, I made money by pocketing a portion of the overstated refunds received by real customers. I also profited by turning non-qualifying taxpayers into qualifying taxpayers enti- tled to a refund and pocketed the whole amount. I filed approxi- mately 250 to 300 of those returns that year, yielding roughly $500,000 in false refunds.

That same year, the rules of the game changed a little, although nothing I could not overcome. The IRS started cross checking Social Security numbers with the first four letters of a taxpayer's last name. Now, I could no longer just fabricate taxpayers' names and Social Security numbers. Instead, I simply took the first four let- ters of a customer's name, changed the last several letters, used the customer's actual Social Security number, inflated the wages, the number of dependents, and often changed the filing status, all to maximize the refund. In most cases, I would file the return and pocket the inflated amount after giving the customer his normal re- fund.

I have explained the substance of the tax fraud I was engaged in. I want to make it clear, however, that none of this would have been possible without the electronic filing program and refund an- ticipation loan practices of the tax preparation business. The RAL's brought the customers in the door and the RAL practices made it possible for me to take a cut of a real taxpayer's fraudulently in- flated EITC, or to bilk the Government for a phony return.

The RAL business works very simply. A customer looking to file a tax return and obtain the money quickly walks into a check cash- ing or tax preparing company, hands over whatever tax informa- tion he has, and, at the same time, applies for a RAL. The tax in- formation comes to me. My company prepares and files the tax re- turn electronically with the IRS.

The IRS receives the electronic return and usually acknowledges the return, which means only that the IRS has quickly scanned the return for very limited purposes. This acknowledgement does not mean that the return is accurate, but this acknowledgement has been sufficient to permit the check to be cut and the RAL to be is- sued.

45

I then had the authority to cut the loan check directly to the cus- tomer, netting out my fees, as well as the fees of the bank and the check cashing outlet. The amount of interest charged can get stag- gering in States that do not regulate these practices. I would some- times receive a portion of the tax preparation company's take as a kickback. Only a few large banks are really significant players in the RAL business.

Despite how straightforward my schemes were, I was only caught because several of my employees of my company became in- formants and went to the authorities. I am confident that if the employees had not turned me in, the IRS would never have caught on and that I would still be in business today.

This is really just a brief summary of the innumerable schemes opened by the IRS's electronic filing system. For example, last year, I discovered a widespread tax fraud scheme in Southern California involving fake W-2s and EITC, which I reported to the IRS. In all modesty, it would take several hours for me to share with you the virtually endless possibilities.

Thank you for your time today, Mr. Chairman and Members of the Committee. If you have any questions, I will be very happy to answer them.

Chairman Roth. Mr. Hersch, you conclude, if I understand what you are saying, that in your judgment, it is practically impossible to make this program fraud-proof?

Mr. HERSCH. Mr. Chairman, this program has to be ripped out and started from the bottom up, because what it has done, it has made Christmas in January, February, and March, every taxpayer that knows they can get earned income tax, period.

If you take the phone calls that my company had and most of the other tax preparation companies had in January, February, and March, everything is getting repossessed. In January, cars get re- possessed, homes get repossessed, power is turned off, water is turned off. It only happens in January, February, and March. It happens because everybody is looking to collect the extra dollars.

On the chart that you had up here before, a person earning $27,000 is in worse shape than a person earning $20,000 because he is going to pay more taxes and get less earned income tax credit. The person that earns $20,000 gets the maximum benefit, more dollars to himself.

I received a call from a woman, in one particular instance. She said, I want my earned income tax credit. I said, do you have any dependents? She said, yes, I have a brother. He is 36 years old. I said, you can't claim your brother to get earned income tax credit unless he is permanently disabled. She said, I will call you back on the phone. She called me back a half-hour later and said, he is permanently disabled now. I will sign for the paper. She got a $1,500 refund and it went through the IRS and she was never questioned.

That is the kind of fraud that is being perpetrated on the Gov- ernment today because of the earned income tax credit, and there have to be checks made. The young lady who was up here before was talking about the filing of the W-2s. What good is my em- ployer filing my W-2 in February when I have already filed my taxes electronically in January, using an upgraded return? The

46

Government doesn't find out about that until April, May, or June. The horse is out of the barn before the door is locked. And because it has been advertised so much and promoted so much, everybody knows about it. It is done easily in 40 or 50 percent of the cat- egories.

I know it should be done for the right people, but it is being done for people who have six children, who have a daughter that is 19 who has two children illegitimately. The father goes in in the morning and claims head of household. The wife goes in in the afternoon and claims head of household. They never get caught. They have four children, four legitimate Social Security numbers. The checks and balances have to be started at once and the tax system has to be revamped to save the money.

Chairman ROTH. Let us say you were the author of a new pro- gram. We do away with the existing EITC.

Mr. Hersch. I am not saying to do away with it. I am saying to close the loopholes.

Chairman Roth. What I want to know, if you were the one re- sponsible for closing the loopholes, what would you provide? What changes would you make?

Mr. Hersch. We heard this morning, you have to start with the electronic filing information numbers, the EFI numbers. The chair- woman this morning said that out of 33,000 applications last year, they turned down 1,500 because they didn't qualify. What about the other 112,000 that are out there right now? They are the same as they were 3 years ago or 4 years ago. They are not checking old ones; they check new ones.

So you have 33,000 new ones, of which 1,500 are no good, and you have 110,000 old ones that are added to the 33,000, so you now have 143,000 filers. That is the first place you start. I was a con- victed felon and ended up with an EFI number.

Chairman Roth. You ended up with what?

Mr. HERSCH. An electronic filing information number, by just ap- plying. I had a 3-year-old girl apply for one and get one. All you have to do is write your name on a piece of paper. There are no checks and balances.

Chairman Roth. Is it possible to create checks and balances?

Mr. Hersch. Absolutely. If I go to a, and I will use quotations, a "certified public accountant who is licensed," these people will not take a chance that an ordinary layman or tax preparer will. If you walk into a store that is put up on January 1 to file your tax re- turns until April 1, he is not a tax preparer. He can add 2 and 2 and has a computer in front of him. It shouldn't be done that way.

The whole licensing process should be started and should be checked, and that is the first way you stop it. I could go over hours and show you things that are done on a daily basis by people to collect this money.

Chairman Roth. Let me go back to, as I understand, what you said a few minutes ago. You obtained an electronic filing identifica- tion number, or "EFIN," I guess they call it?

Mr. Hersch. I had 105 EFI numbers.

Chairman Roth. You were able to obtain 105 EFI numbers? How did you do that, again?

47

Mr. Hersch. Just by filling out the forms and sending them in to the Government.

Chairman ROTH. And there was no effort made to check it? Did they know you were a felon?

Mr. HERSCH. It was on the application.

Chairman Roth. It was on the application, but you still obtained the number?

Mr. Hersch. Yes, sir.

Chairman Roth. You stated that a critical component of defraud- ing the Government was supplying the IRS with false information on a tax return about wages, dependents, and filing status.

Mr. Hersch. Sure.

Chairman ROTH. Why wasn't the IRS able to catch you by cross checking the false information that you provided?

Mr. HERSCH. I believe we have heard testimony that, up to this point, they have put in new filters and new processes, but they haven't gotten to the point yet where they can tell you that Mrs. Jones has four kids and one of Mrs. Jones' four children is 21 years old and she has two illegitimate children, and they can become fos- ter parents for those two children and claim earned income credit for those two children, also. So if the wife claims her own children and the husband claims the daughter's two children as foster chil- dren, they both earn earned income credit. They both can walk into the same place and get it done at the same time, rather than file "married filing jointly".

Chairman Roth. These scams are really big business, aren't they? I think you said your company filed, in 1993, some 29,000 tax returns?

Mr. Hersch. That is correct.

Chairman Roth. And of those, you filed approximately 400 re- turns that you made up out of whole cloth, using false names, is that correct?

Mr. HERSCH. By using false information.

Chairman Roth. False information?

Mr. Hersch. That I did myself, but out of the 29,000 returns, we issued 20,000 refunds. Half of those refunds, 10,000 of them, there was false EITC information on them from the people who filed them, from the people who gave them to the tax preparer.

Chairman Roth. Given how easy all this was, why didn't you file more false returns?

Mr. Hersch. Fear that the bank would come to a level of not being able to collect from the Government a percentage. Once you go over a certain percentage, the bank would never let you file in the following year. If you ran over 2 percent of your returns that they didn't collect, they wouldn't file, so I kept the returns in the range that I knew if they didn't collect those 400 out of the 20,000, it was going to be in the range of being able to get a license for the following year.

Chairman Roth. What percentage of a check cashing company's total income is derived from their tax preparation business during tax season, do you have any idea?

Mr. Hersch. I would estimate that probably somewhere in the range of 35 to 40 percent.

Chairman Roth. Thirty-five to 40 percent?

48

Mr. Hersch. And any tax preparers that open up for 4 months, you realize they get 100 percent of their income in that 4 months.

Chairman ROTH. Let me ask you, this illegal business, how many other organizations did you know that were involved in this in your community? Is this pretty general information, that there are these

organizations that will help you

. Mr. HERSCH. Absolutely, many. It hasn't been the job of the tax preparer, in my own words, myself, that if you walked in and you gave me identification and said, I have two children and here is their identification, I don't have to check it any further. I put it down on the paper. When you bring me the information, I am not the policeman who says, well, you really don't have those two kids, the same way as the woman who had a brother. I can't say she doesn't have a brother. If you want to sign that form, Senator, you sign the form. You will get the money.

So it is a case of there are many out there. There are a lot of large companies that try and watch the fraud as best they can, but remember, they have two eyes, and if your friend is behind that desk when I bring my stuff in, or I want you to prepare my tax return and I give you an envelope with a set of answers on it and say, this is what I want to get done, I will file your tax return.

But if anybody ever took a census in January, February, and March in the United States of America, they are going to find more dependents floating around this country than they have the other 9 months of the year.

Chairman Roth. Is it out on the street, pretty much, that EITC is fair game, that this is an easy way to

Mr. Hersch. You answered the question. It is as easy as apple pie. It doesn't take much to do anything, and it is on the street on a daily basis, and it is on the street because of the speed in which the Government has tried to help the people that need the money.

Chairman Roth. In other words, the purpose of EITC, of course, is to get money into the hands of the working poor as fast as pos- sible.

Mr. Hersch. Correct.

Chairman Roth. But the fact is, by filing electronically and get- ting a quick response from the Government, it opens the door to fraud. It makes it that much easier. Is that correct? Is that what you are saying?

Mr. Hersch. You have given the working poor an opportunity to have more money easily because Mrs. Jones has her family, she lives with another guy, she lives with two people, they don't want to file together, so you are giving them all the opportunity to get this money.

Chairman Roth. So, as we expand the program, making the ben- efits larger, we are providing

Mr. Hersch. More fraud is going to be perpetrated because they are going to get more money.

Chairman Roth. So you would predict that, instead of less fraud

Mr. Hersch. Unless it is stopped, there will be more.

Chairman Roth. But if the program expands

49

Mr. HERSCH. On this basis, if the program expands on this basis today, the hole is big enough to drive the Queen Mary through, and that is literally it.

Chairman Roth. I want to thank you, Mr. Hersch for being here today. We may have some additional questions. If so, we will sub- mit them in writing.

Mr. Hersch. Thank you.

Chairman ROTH. Thank you.

We now come to the final panel. I am pleased to welcome Finn Caspersen, who is Chairman of the Board of Beneficial Corpora- tion; Daniel Grunberg, Vice President of Technology, Jackson-Hew- itt; and Dan Stein, the Executive Director of the Federation for American Immigration Reform.

Gentlemen, I thought we would start out with Mr. Stein and then proceed to you, Dr. Grunberg, and finally to you, Mr. Caspersen. Welcome, gentlemen. It is nice to have you and see you again.

TESTIMONY OF DAN STEIN,1 EXECUTIVE DIRECTOR, FEDERATION FOR AMERICAN IMMIGRATION REFORM (FAIR)

Mr. Stein. You caught me off guard. Thank you very much, Mr. Chairman and Senator Glenn. Thank you very much for the oppor- tunity to be here.

My name is Dan Stein. I am the Executive Director of the na- tional public interest group FAIR, the Federation for American Im- migration Reform. FAIR is working very hard to try to end illegal immigration and develop workable legal immigration policies that serve the national interest.

The growing evidence of widespread use of the earned income tax credit program by illegal aliens is growing into a national scandal. It represents the vulcanization of immigration enforcement inter- ests within the Federal Government when the Internal Revenue Service puts an enormous priority on the question of tax collection without regard to the eligibility of a particular alien.

The institutional response of the Internal Revenue Service has, for some years, now, been that illegal aliens are eligible for the earned income tax credit. Mr. Chairman, we do not believe that is an accurate interpretation of the law. However, assuming it is, it does represent a potentially enormous tax burden for other tax- payers, legal taxpayers in the United States.

Data are difficult to accumulate on these issues, but we do know from the 1986 amnesty program that approximately 70 percent of the illegal aliens working in the United States are living in fami- lies whose income is so low that they are eligible for the earned in- come tax credit program. Therefore, we know that a very large per- centage of working illegal aliens or undocumented aliens are eligi- ble or would be eligible for the earned income tax credit.

The problem in enforcement relates to the vulcanization between the Internal Revenue Service, the Immigration and Naturalization Service, the Social Security Administration, and birth records, which are maintained separately by all 50 States.

lThe prepared statement of Mr. Stein appears on page 143.

50

The General Accounting Office, in discussing the program where 160,000 probable illegal aliens were identified, in fact, has simply asserted that the IRS is surmising, based on speculation, that those who did not follow up their claims after inquiry were prob- able illegal aliens and that they may not have been able to produce, at that point in time, valid Social Security numbers. But, in fact, the Internal Revenue Service did not now, nor would they under the administration's proposal, actually identify illegal aliens. They would simply be identifying people who would not be able to produce legitimate Social Security numbers.

To be very brief, because I know your time is short, the rec- ommendations that we have, Mr. Chairman, are we think that, first of all, Congress needs to restate the presumption that an ille- gal alien is not eligible for any Federal benefit, including the earned income tax credit, expressly by law. The position of the IRS, that unless Congress expressly and specifically excludes all classes of aliens, including those wfco have entered without inspection, doesn't seem to us a consistent reading with the entire Federal United States Code, which would include the Immigration and Na- tionality Act, as well as the Internal Revenue Code.

Moreover, we believe that Congress needs to encourage the es- tablishment of a national computer verification system to coordi- nate birth records and death records in all 50 States, which both the Social Security Administration, the INS, and the IRS can use to verify claims to citizenship as well as to verify alienage, not only in this context but in a variety of others. Barbara Jordan and her Commission on Immigration Reform have also suggested this solu- tion.

We also prefer the legislation that you proposed last year, Mr. Chairman, S. 2552, to the administration's proposed plan. Your bill would have statutorily defined illegal aliens, as well as certain non- immigrant and temporarily resident aliens, as expressly not eligi- ble for the earned income tax credit. The administration's proposed bills in this Congress, S. 453 and H.R. 891, tie eligibility to "work authorization", which is a new concept in the immigration law, and relies upon the alien presenting a legitimate Social Security num- ber.

Mr. Chairman, there are many, many ways that an illegal alien can obtain a legitimate Social Security number. I call your atten- tion to an article appended to my testimony, a very recent article from the Omaha World Herald,1 which shows how large, well-co- ordinated scams obtain legitimate birth certificates or produce phony birth certificates, which is now big business in this country, to obtain legitimate Social Security numbers.

Having worked on this issue for years, I know that the Social Se- curity Administration is very reluctant to be involved in verifying whether or not a person is, in fact, a citizen when they issue a number or verifying whether or not a person is in an alien status which is lawful or has remained in one for any period of time. The INS is developing a model program with the Social Security Ad- ministration to begin the verification process for Social Security

'The articles appears on page 147-148

51

numbers, but that effort is years and years away from being brought to any kind of workable on-line conclusion.

The mechanism chosen by the administration relies upon the in- eligible alien being unable to obtain a Social Security number. Fur- ther, the administration does not exclude those cases where an alien may have accumulated the earned income while in an ineli- gible work status, thereby providing yet another magnet for illegal aliens to come here, work illegally, and then expect to get these benefits later. This is so long as an alien has a Social Security number at the time the tax form is filed.

The main flaw, though, in the administration's proposal is that it fails to consider the vast variety of ways in which an alien ineli- gible to work can still obtain a Social Security number or the ways in which an alien may present a valid Social Security number on a tax return that was only issued by SSA for brief periods or work- authorized periods for only a temporary period of time.

Without mandating specific cooperation between INS, IRS, SSA, and the various State birth agencies in making these eligibility de- terminations, the use of fraudulently obtained valid Social Security numbers will mushroom under the administration's proposal. This means that the administration's bill would only deny earned in- come tax credits to aliens who had not obtained valid Social Secu- rity numbers, regardless of their actual immigration status.

Further, the administration would still permit a wage earner to claim a dependent or more who does not reside with the principal alien in the United States, nor does the administration exclude wages withheld when the alien was working out of status from the determination that a now-work-authorized alien may qualify for the EITC.

We believe all the above areas need to be addressed to ensure that illegal aliens do not look to the earned income tax credit pro- gram as yet another welfare rip-off program financed on the backs of U.S. taxpayers.

Lastly, we believe, and we are disappointed about the adminis- tration's proposal, the administration's proposal does not mandate more cooperation between the IRS and the INS in the enforcement of Federal immigration laws. The IRS should be required to notify the INS of the reported addresses of EITC claimants who are ei- ther unable to provide valid Social Security numbers or, after ver- ification with Social Security, cannot provide evidence of lawful res- idence. Unless aliens unlawfully present are convinced that tax fil- ing will raise the possibility of inquiry by the Immigration Service, this problem will continue to grow.

Mr. Chairman, under the immigration law for almost a century now, an alien may not come to this country for any reason if they are likely to become a public charge. The fact that the evidence of illegal aliens using the earned income tax credit is growing is part of the reason why there is a growing American backlash and con- cern about the cost and consequences of illegal immigration.

We hope that you will give this your serious attention. We ap- plaud the fine work that you have done on this issue and a variety of related issues in the past and would be happy to answer any questions you may have.

Chairman ROTH. Thank you, Mr. Stein.

52

Dr. Grunberg.

TESTIMONY OF DANIEL B. GRUNBERG,1 Ph.D., VICE PRESI- DENT OF TECHNOLOGY, JACKSON-HEWITT AND ASSOCIATES

Mr. Grunberg. Mr. Chairman and Members of the Committee, my name is Daniel Grunberg and I am Vice President of Tech- nology and a Director of Jackson-Hewitt Tax Service. We appre- ciate the opportunity to present our experiences during the current tax season.

This tax season, Jackson-Hewitt Tax Service operated 1,232 of- fices, both franchise and company-owned, in 44 States and the Dis- trict of Columbia. We will have prepared over 640,000 tax returns by April 30. We have been staunch supporters of the IRS's elec- tronic filing program since the beginning. In fact, we were awarded several IRS contracts to develop the test packets that all electronic filing companies must pass to participate in the electronic filing program. Over 89 percent of the returns we prepare are electroni- cally filed. We have offered free electronic filing in our offices since 1990.

The IRS has made tremendous strides in fighting fraud over the last 4 years, but this year, they humiliated and injured several mil- lion hard-working taxpayers who can least afford it. Let me tell you what happened in our offices.

On February 3, the IRS payments for refunds made by electronic deposits to customers' bank accounts were not made for the full re- fund. Twenty-eight percent of our customers only received the non- EIC portion of their refund. These partial payments were made on returns that had already been accepted by the IRS as having prop- er Social Security numbers and matching names.

During the next 3 days, we and every other tax preparer had a virtual revolt on our hands. Twenty-eight percent of the checks, or 45 percent of those with EITC, were for amounts far less than peo- ple were expecting, and the only proof we could offer that weren't stealing their money was a computer printout that we had hastily assembled for them. In numerous cases, police had to be called to our offices to calm taxpayers and even forestall riots.

The Associated Press reported that the GAO did a spot-check of IRS taxpayer assistance calls and found that only 13 percent of the calls were answered during the period January 30 to February 10. When our customers did manage to get through, they were told such things as, first, they did not receive their EIC because of er- rors their tax preparers made, or second, that their tax preparers had their money and they should go get it from them.

We have over 500 franchisees and every one of them has horror stories to tell. We have hundreds, if not thousands, of documented cases of evictions, repossessions, and other hardships all resulting from this non-payment of expected refunds.

The typical taxpayer affected had an average income of $11,600 with children and a job. They are below the poverty line. To some- one who has a weekly take-home pay of less than $200, the average EITC payment of $1,500 is almost 8 weeks' wages. This money is needed for them just to survive.

1 The prepared statement of Mr Grunberg appears on page 149.

53

Keep in mind that our large stack of stories represents only 1,232 electronic return originators. This is less than 5 percent of the total in the country, who all have similar stories to tell.

Our company prides itself on winning customers for life. If we had known, we would have told every customer applying for the EITC what was going to happen. The IRS did make an announce- ment in February in a mailing to tax preparers, after many thou- sands of customers and tax preparers had already been taken by surprise. We are being told now that if the IRS pays at all, it will be 8 weeks before a check will be mailed to the taxpayer's home. It has been more than 8 weeks and we have heard from very few people who have received their checks yet. The IRS simply did not adequately inform the public about the massive changes that were about to take place this February.

Many people choose electronic filing and have their money direct deposited because receiving checks in the mail in their neighbor- hood is unsafe. Paper checks also have a much higher incidence of fraud, both by fraudulent endorsement and by fraudulent claims of stolen checks.

If a taxpayer has been evicted from their apartment, how are they going to receive a questionnaire or check in the mail? Treas- ury check envelopes say, "Do not forward."

Most taxpayers getting the EITC have been getting it year after year. They expect and count on this money to survive. The IRS could have mailed questionnaires, done name and Social Security matches, cross references, or performed audits during the summer of 1994. This would not have disrupted so many lives as the meth- od they chose.

The IRS 1040 package mailed to the public in late December or early January says, "If you file a complete and accurate return, your refund will be issued within 21 days. You can also get the con- venience and safety of direct deposit." And, there is a footnote. "Some refunds may be temporarily delayed as a result of compli- ance reviews to ensure that the returns are accurate." Perhaps 7 million taxpayers deserve more than a footnote.

We have worked closely with the IRS to prevent fraud. Previous to this tax year, we were catching hundreds in our office after the IRS accepted their returns. Now, the IRS has fraud under control and we are seeing far fewer cases. With the name and Social Secu- rity number matches implemented by the IRS just this year, it has become exceedingly difficult to commit any type of organized wide- spread fraud.

The IRS has mailed questionnaires, Form 9598, to some tax- payers before they will pay the EITC. This questionnaire is another new form from the IRS. We predict that significant numbers will fail to be returned due to fear or ignorance, not fraud.

We understand the need for secrecy in the battle against fraud. However, this is not a case of a few people being audited. This was a campaign punishing 45 percent of the people because of a few fraudulent filers. Many taxpayers have told us that they will just file a paper return next year. They say, why bother filing electroni- cally if it is going to cause this much trouble? People are very re- luctant to do anything to bring them under the scrutiny of the IRS.

54

We believe that there will be another drop in the number of elec- tronically-filed returns next year if strong, immediate action is not taken. We recommend the following.

One, restore the direct deposit indicator. At the time of electronic filing, taxpayers need to know if, when, and how their refunds will be paid.

Two, additional information required to document the EITC should be asked for at the time the original return is filed. The change in IRS procedures may have been well-warranted, but the methods used brought hardship to many hard-working taxpayers. We believe the IRS needs electronic filing and that it can reduce cost and combat fraud at the same time. It should be a partnership between the IRS and private industry, not an antagonistic relation- ship.

We are willing to work with the IRS in any way possible to find a solution to the current situation. Thank you.

Chairman ROTH. Thank you.

Mr. Caspersen.

TESTIMONY OF FINN M.W. CASPERSEN,1 CHAIRMAN OF THE BOARD, BENEFICIAL CORPORATION

Mr. CASPERSEN. Mr. Chairman, Senator Glenn, my name is Finn Caspersen. I am Chairman and CEO of Beneficial Corporation. I thank you for this opportunity to discuss the earned income tax credit and our experience at Beneficial with fraud in connection with electronic filing.

Mr. Chairman, I applaud your Committee's continued efforts and the efforts of all other Members of this Committee to address fraud issues as you focus on ways to reduce the cost of Government.

As you know, one of the most significant initiatives designed to reduce the cost of administering the IRS is the systems tax service modernization program, which we have heard much about today. Certainly, thus far, in a decade-long progression, the most success- ful element of this modernization is electronic filing of tax returns.

Electronic filing eliminates the cost of storing millions of paper returns. It eliminates the transforming of paper returns into usable electronic data. It is relatively error-free. We have heard the fig- ures from the Commissioner of approximately 15 percent error rate when the IRS does it and less than a half-of-a-percent when it is done electronically. The cost of processing is significantly less when the preparers do it electronically than when the IRS does it by paper.

We have had extensive experience of electronic filing and the earned income tax credit. Our subsidiaries, the Delaware Bank, the Beneficial National Bank, makes more refund anticipation loans than any other bank in the country. We call them RAL's, as we have heard earlier, and they are made available only to those tax- payers who file their returns electronically and expect a refund.

Last year, almost 14 million taxpayers filed their returns elec- tronically. Of these 14 million, 9 million obtained the RAL, and there is no question that RAL was the driving force behind all elec- tronic filing.

1 The prepared statement of Mr. Caspersen appears on page 152.

55

Unfortunately, as we have heard, there is fraud and it is signifi- cant fraud. The IRS is to be commended for recognizing this prob- lem and taking steps to overcome it. We all hope, and, indeed, ex- pect that anti-fraud programs initiated by the IRS will substan- tially reduce the problem.

However, I would submit that some of the changes instituted over the past year, and you specifically asked that I address these questions, have adversely affected many honest taxpayers, their tax preparers, and RAL lenders, such as ourselves. These changes, while well intentioned, have resulted in too many taxpayers wait- ing far too long to receive their refunds, certainly, much longer than they have ever had to in the past.

This has been particularly true for the most needy taxpayers, the real taxpayer that the earned income tax credit was supposed to benefit, those who have claimed the EITC in their electronically- filed returns. Most of these taxpayers, granted, not all, but most, represent hard-working low-income families for whom the delay constitutes a very significant, substantial, hardship.

Many of these taxpayers are also our RAL customers. When a taxpayer obtains a RAL, the loan is repaid with the refund, which the taxpayer has instructed the IRS to send to their account at the lending bank in the form of a direct deposit, again, a savings to the IRS.

Earlier this year, we learned that certain taxpayers, numbering in the millions, who had claimed the EITC and obtained the RAL, the IRS would no longer direct deposit the EITC portion of the re- fund to the taxpayers' accounts at the lending bank. These tax- payers had certain criteria, which the Commissioner, obviously, for good and sound reasons, did not want to discuss this morning, but, I might add, these are rather gross criteria. There is nothing so- phisticated about the criteria utilized in these screens.

Instead, the refund would be sent directly to the taxpayer in the form of a paper check. This was contrary to what we expected, what other tax preparers expected, and what the taxpayer had in- structed, period.

While we thought and were advised that there might be some limited situations that this would occur, we did not anticipate, nor did the face-to-face discussions with the IRS reveal, the enormous magnitude of refunds to be sent by paper check. Because of this change in midstream, over 300 million of Beneficial National Bank's taxpayer loans are now at risk and may not be collected. This has nothing to do with fraud, I might add.

Faced with this prospect, Beneficial National Bank sued the IRS in order to seek a solution which would result in the IRS using di- rect deposit to the lender instead of paper checks to the taxpayer. As the suit moved forward, we concluded that the IRS did, in fact, believe in good faith that there was no viable solution during the current tax year. Moreover, the IRS assured Beneficial that it would make the necessary and appropriate changes to remedy the problem in the 1996 filing season.

For this reason, as well as our belief that communications prob- lems of the past, hopefully, are now behind us, we terminated the litigation.

56

Late last year, the IRS announced another policy change which has dramatically affected electronic filing, RAL lenders, and, ulti- mately, EITC recipients. Under it, the IRS no longer adheres to its former policy of notifying taxpayers and their RAL lenders when a refund was to be set for tax liens, child support payments, delin- quent Government-guaranteed loans, or other obligations.

This notification was important to RAL lenders as a means of controlling credit risk, because many of our customers do not other- wise qualify for traditional loan products. Remember, $11,000, $30,000 average income, EITC eligible, good faith, hard working — these are not gold card holders.

Prior to this change, we were able to make our product more widely available to people who needed our RAL products because the IRS, which, it is very important to note, is the only institution which has the information immediately available to it, advised us that the taxpayer had no Federal liens. This change in IRS policy resulted in increased costs of RAL's — we had far greater losses and then we had to raise the price — and decreased eligibility.

In effect, what the IRS did, and I think this is very important to understand — it has nothing to do with fraud — the IRS passed the cost of the deadbeat dad-type individual onto the honest tax- payer who wanted a RAL. They had information about who was a deadbeat dad, they wouldn't give it out, and they passed the cost on in increased fees to honest taxpayers.

Based on our experience, we estimate the IRS will experience at least a 25 percent decrease in electronic filing this year, a substan- tial setback for everyone concerned with the future of electronic fil- ing and tax service modernization.

As you have heard today, there was a tremendous outcry among the taxpayers. We received, over the last 2 months, over 4 million phone calls — as compared to under 400,000 in the previous year.

In spite of the long-range implications of these problems, we con- tinue to believe in the importance of the program and the benefits to be derived from it. We urge Congress to continue to support the IRS as it strives to implement tax service modernization in a meaningful way. Likewise, we completely agree with the impor- tance of deterring and detecting fraud in our tax system and be- lieve that fraud can and should be addressed without impacting honest taxpayers in the manner and to the extent they were im- pacted this year.

We believe that the key to the problem involves no cost to the Government. We believe that the key to the problem is for mem- bers of the electronic filing industry and the IRS to work together in a cooperative fashion. After all, we and our other banks have the same incentive as the IRS in reducing fraud in electronic filing. When we grant a RAL and a taxpayer does not receive a refund because of the fraud, we lose. Our bottom line is affected.

I emphasize this as strong as possible, that even if we have ap- proved a RAL, even if we have approved it, the IRS is under no obligation whatsoever to send a refund to the bank if it has deter- mined, in its normal processing, that the return is fraudulent or in error and the taxpayer is not entitled to the refund.

Senator Glenn, we neither ask nor receive any governmental guarantees. I think it is very important to understand that. Even

57

as the IRS improves its fraud detection capability to stop fraudu- lently claimed refunds, the brunt of any further fraud will continue to fall directly on the RAL lender, us, and that is the way it should be. We have no problem in that.

We have been and will be willing to put our money where our mouth is. I have personally offered to the IRS and to the Commis- sioner that we will bear the cost of any fraud and give her 6 months, give the Service 6 months to review that return, and if she finds fraud, not to pay, period, i.e., a 6-month free look. We believe our screens are that effective.

An even more important reason for the IRS electronic filing in- dustry to cooperate is that the industry has made and will continue to make significant contributions to fraud detection in electronic fil- ing, and I believe I have covered many of those in my submitted testimony. I think it is important, however, to emphasize two sin- gle points. When you have electronic filing with honest ERO's, that is the only time that you come face-to-face with a taxpayer, and there is a tremendous advantage in that.

Second, the banks are in the business of detecting fraud. At Ben- eficial, we make millions of loans to millions of customers. Many of these loans, indeed, millions, are made on a 60- to 90-second time frame because we have on-line connections to all the key fraud indicators. The IRS does not do this. We have offered and will continue to offer the IRS access to these exact same screens that we utilize here.

Mr. Chairman, Senator Glenn, I would like to conclude by em- phasizing that Beneficial and the other members of the industry are committed to concrete actions to reduce fraud. We are confident that an effective fraud prevention program can be implemented in a way that will not result in unnecessary hardship for taxpayers and particularly the deserving recipients of the EITC.

These goals can be achieved, however, only if the IRS and the electronic filing industry work together in the future. For this rea- son, I propose to you and urge you to support the formation of an effective public-private partnership between the Service and the in- dustry, with the goal of establishing a widely-utilized, secure, fraud-free, and affordable electronic filing system in this country at a minimum cost to the American taxpayer. Thank you.

Chairman Roth. Thank you.

Senator Glenn, I think, has a commitment, so I will call on him first.

Senator Glenn. Yes, I do. Thank you very much, Mr. Chairman. I thank you all for being here this morning.

Are there any other cases where we use Government refunds or Government checks deposited directly for purposes of, say, mort- gage or whatever?

Mr. CASPERSEN. The military has a military allotment program.

Senator Glenn. That is a voluntary allotment system.

Mr. Caspersen. It is a voluntary allotment system whereby they can provide lenders and/or dependents electronically-transmitted checks. It is a very successful program.

Senator Glenn. I am very familiar with that one.

What did you do in this case? Do you have the people sign such papers as they sign in the military, then, for their allotment

58

checks, or was this just a custom that the IRS got into of sending the checks directly back to you?

Mr. CASPERSEN. No, this was a specific revised procedure that covers this in detail, a specific form, as, of course, there is with al- most anything with the IRS, and then the customer signs that and authorizes and directs the IRS to do certain things.

Senator Glenn. Out of the New York Times on Wednesday, Feb- ruary 22, there was an article that referred to this and it said, "To a taxpayer expecting a $1,000 refund, a bank spokesman, Robert Wade said the bank charges a $30 electronic filing fee and a $59 loan fee. The loan costs equal an annual percentage rate of 250 percent, he said." I don't know what your response to that is. Aren't there any usury laws in Delaware?

Mr. Caspersen. There are usury laws in New York, too. I have two answers to that question.

Prior to the IRS changing the rules on the direct deposit indica- tor, we charged $29 per return. This is a fee, because the particular law that governs those, it is treated as an annual percentage, and we do that just for protection's sake.

I might compare this with the $1,000 Western Union fee, if you want to wire by Western Union to San Francisco or to Baltimore, it is set at $75. Indeed, the IRS fee for restructuring one's own obli- gations on an installment payment, in addition to all the interest they appropriately charge, they charge $45.

We started out with a fee of $40-some. We reduced it to $29. It was our goal to reduce that to under $20, but because the IRS would not share with us that information which they alone had, we could not do that this season. Our goal is continually to take ad- vantage and lower this, frankly, for the very selfish reason that the lower it gets, the bigger the market.

Senator GLENN. What was your average time of the loans being in effect?

Mr. Caspersen. This season, it is infinite.

Senator Glenn. Previously?

Mr. Caspersen. Previously, it would range from 10 to 9 months,

depending on the-

Senator GLENN. You had a guarantee, then, that the Government was going to pay it back directly to you.

Mr. Caspersen. Absolutely not. The Government, at any time, could look at that return and say, this is fraudulent. All the DDI, the so-called direct deposit indicator, did, and it is very important to understand this, all that it did was tell you there are no liens of a certain nature. It did not say that the Government was going to pay.

Senator Glenn. But most of those came back. What percentage of them were paid by the Government?

Mr. Caspersen. It ranged from 96 or 97 percent to the last year was 99.5 percent. I might add that the break point is around 2-plus percent. After 2 percent of not payments, you break even. You don't make anything. If it goes to 2.6 percent, you lose money.

Senator Glenn. A 96 percent return, though, guaranteed by the Government, is pretty good.

Mr. Caspersen. No, it isn't. A 96 percent return, given the size of the fee, means you have lost 1.5 percent on your total invest-

59

ment. We had $3.5 billion out last year. One-and-a-half percent of $3.5 billion is not something the shareholders like to see.

Senator GLENN. Your $29 fee you said you were charging, let us say, on a $1,000 loan, and the time it was out, what would be the average annual rate on that?

Mr. CASPERSEN. I am not sure what the APR is on that. This is a fee, and I might add, this

Senator Glenn. A fee is a percentage interest, as with anything else, just like it is if you buy real estate. The points are figured in percentages, the same way as a loan fee.

Mr. Caspersen. I will find that out and submit it to you.

Senator Glenn. Was it higher than 15 percent, do you think, on an annual rate?

Mr. Caspersen. If you take the days, absolutely. But the fact is, this is a fee that allows immediate access, and there is risk. And bear in mind, the risk is being realized this year. It was realized the first 3 years when we were in the red. Currently, we could be up to $300 million in the red.

Senator Glenn. But your risk wouldn't be on more than 3 or 4 percent of the loans, right?

Senator Glenn. Yes, but if you lose more than 2.4 percent of the loans, you have lost money, because the $29, on an average $1,400 loan, you can see the computations, you lose money. This is not a get-rich scam.

Senator Glenn. I just think using the Government as a guaran- teed payback is not

Mr. Caspersen. We are not getting guarantees, sir.

Senator Glenn. If these people are that reliable, why don't you just make your own loans to them?

Mr. Caspersen. They are not reliable. These are people

Senator Glenn. You want the Government to take the risk, then, while you make the loan and collect the interest.

Mr. Caspersen. No, sir. We are not asking the Government to take any risk. We are asking them to do all the fraud they want, do all the fraud investigations. All we ask is if they have a child support lien on their records, which they only have, if they have an IRS lien on their records, which they only have, that they tell us, in addition to what the credit bureau has

Senator Glenn. You have heard testimony here this morning, we don't have equipment to do all this checking that you are talking about doing and all this indexing back and forth. We are having trouble, in fact, we have legislation that has been submitted al- ready to cut the IRS back, and even the tax system modernization is going to be cut by about a fourth, I guess. We can't even keep up with that.

Mr. Caspersen. But the IRS

Senator Glenn. You people are free to run.

Mr. Caspersen. I am sorry?

Senator Glenn. You are free to run, pretty much as I see it. We can't check all these things you want us to check.

Mr. Caspersen. No, the IRS has this in their records and, up until this year, gave it out. We are offering, sir, the advantage of these modern screen systems that we use for millions of customers and we are offering it to the Service for nothing because it is in

60

our interest that this program be contained. It is in our bottom-line interest that fraud be kept to the absolute minimum. These are to people that deserve the income. We don't want the undeserving.

Senator GLENN. In a normal loan situation, you would check the background, right?

Mr. Caspersen. We do check the background.

Senator Glenn. You check the background?

Mr. CASPERSEN. Yes, we check the background and we go to the credit bureau. My point is, this is not a normal loan situation. This is something where the IRS has sole information.

Senator Glenn. But you apparently had extended 1.2 million loans this year. Did you check all those backgrounds?

Mr. Caspersen. We extended 2.8 million loans this year, to date, and we checked every background through the credit bureau on an on-line basis.

Senator Glenn. I just think the fees that you are getting here, if these figures, the percentages are right as reported, I would think you would be up against usury laws. You asked for a tem- porary restraining order and you lost. Are you appealing that case?

Mr. Caspersen. We withdrew the request, as I said in the testi- mony because it became apparent to us, and under the require- ments for the temporary restraining, that the IRS in good faith be- lieved they could not make the change during this tax season with- out endangering the process. They reconfirmed to us that they will make the change over the tax season. In the interim, as I testified, we have over $300 million at risk.

Senator Glenn. The question is whether it should ever have been put out there at risk to begin with. What other business are you involved in? Is this your major business?

Mr. Caspersen. No, sir. We have a $13 billion company. We are one of the largest consumer lenders in the United States, Canada, and the United Kingdom. We know how to make these loans. We know how to lend money. We know how to do fraud checks

Senator Glenn. If you check these people so well, though, then why can't you collect your own loans? Why not give back the money to the people who apply for the refund or apply for the credit, they get it, and it is up to every American to determine their own busi- ness arrangements. Why can't we do that with this, like every other

Mr. Caspersen. I couldn't agree with you more, sir. It is up to every American. If the people are voting with their feet and 7 mil- lion people asked for the RAL's last year, I think they are making a choice, and I don't think the Government should take action that raises the price and passes the cost of the fraudsters, as the IRS calls them, onto the honest taxpayer.

Senator Glenn. It is awful hard for me to see how we use the Government as a guaranteed loan payer, which is what you are doing by asking that the check be given to you, not to anybody else.

Mr. Caspersen. We are only asking that

Senator Glenn. The person doesn't have any choice in the mat- ter. It goes right directly to the bank

Mr. Caspersen. The person has all the choice in the matter. It is he or she that directs it. And we are not asking the Government to be a guarantor. We are telling the IRS, take your time, look

61

here, look there, look anywhere you want, take 6 months — 6 months, that is longer than paper returns — and then, if you find fraud, put it to us. We will take the risk.

Senator GLENN. But if your profit margins were required to be reduced to, say, 10 percent, so they stayed under usury rates, on an annual basis, whatever the length of your loan, but on an annualized basis, would that put you out of business?

Mr. Caspersen. The IRS is putting us out of business by

Senator Glenn. I am asking, would that put you out of business, if you had to accept rates like normal bank rates?

Mr. Caspersen. Sir, we are a normal bank.

Senator Glenn. I am asking, normal bank rates of 10, 12 per- cent, say, for a loan like this, could you operate on that or do you have to have these fantastic returns to stay in business?

Mr. Caspersen. You have to have this $29-level fee in order to stay in business.

Senator Glenn. Which comes out to what on an annualized basis for the length of your loan? That is what I asked you.

Mr. Caspersen. As I said, I will provide you with that figure. I don't have it.

Senator Glenn. But it would be well above the 15 or 20 percent, I presume.

Mr. Caspersen. It certainly would.

Senator Glenn. And you still don't think that is usury rates?

Mr. Caspersen. No.

Senator Glenn. Does the State of New York think it is usury rates?

Mr. Caspersen. Neither the Federal Reserve, which governs Beneficial National Bank, which is a national bank, -has given us in the OCC, has reviewed it and said it is not usurious.

Senator Glenn. I don't understand what usury rates are, then. If you get well above 15 or 20 percent and they are not usury rates, I don't know how they define it, then.

Mr. Caspersen. I am trying to use the analogy to what the IRS itself charges for a restructuring. I am trying to use the analogy for what Western Union charges for sending money. In addition to that, we bear the risk of fraud.

Senator Glenn. Western Union doesn't require that something coming from the Government goes directly into the Western Union account.

Mr. Caspersen. No, Western Union asks you for your check right there up front.

Senator Glenn. And the individual has the option of paying that at that point.

Mr. Caspersen. That is right, and the individual has the option of directing the check or not directing it at this time.

Senator GLENN. Thank you, Mr. Chairman.

Chairman Roth. Thank you, Senator Glenn.

As I understand what you are saying, Mr. Caspersen, what had been structured was a program that did make loans available to the working poor that otherwise have not been available, is that a fair statement?

Mr. Caspersen. That is correct, sir.

62

Chairman Roth. And the taxpayer does sign the form that au- thorizes IRS to make the payment to the bank, is that correct?

Mr. Caspersen. That is correct, yes.

Chairman Roth. As far as the DDI is concerned, of course, that only enables the Federal Government to ensure that the taxpayer doesn't owe other funds, such as student loans, so it merely clears that there are no other Government liens?

Mr. Caspersen. That is absolutely correct.

Chairman Roth. One of the advantages, if I understand what you are saying, is that this approach has also encouraged the use of electronic filing, which, in turn, does result in substantial sav- ings to the Government, not having to transmit from paper docu- ments to the Federal Government data bases, so that there is a savings there as part of the justification.

But the DDI doesn't in any way enable, as it works for the Fed- eral Government, to ensure that the Social Security number was accurate and represented that as a taxpayer, so that is where there has.been a problem from the IRS point of view, is that correct?

Mr. Caspersen. That is correct. The DDI is- entirely apart from the Social Security number problem, which is an entirely different problem, and, frankly, a serious problem.

Chairman Roth. Yes.

Mr. Stein, let me ask you a couple of questions. The hour is growing late and I know you are all probably hungry, as am I.

Mr. Stein, do you have any estimates as to how much the IRS has lost of the taxpayers' money as a result of paying EITC bene- fits to illegal aliens?

Mr. Stein. Mr. Chairman, that is a good question. The difficulty of obtaining the response goes directly to the fact that the IRS doesn't want to know who it is paying out the earned income tax credit to or whether they are here illegally or not. This institu- tional desire not to know leaves us only speculating.

If only half of the 160,000 of the ones identified in your GAO- requested report had been eligible for it, that would have been $200 million right there. If you assume that 70 percent of the 3.2 million individuals were in families where they were eligible for the earned income tax credit, you have to assume that the figure is cer- tainly tens of millions of dollars, maybe $100 million or more.

The most important thing to remember, however, are the trends. The ease of access to the program has created an enormous growth for those who are here illegally using the program, and I think of that, we can be sure.

Chairman Roth. I have one more question, Mr. Stein. As you know, in the 1993 budget legislation, there was a substantial in- crease in the EITC Can you give us any idea as to whether this dramatic increase in the EITC will encourage illegal immigration to this country?

Mr. Stein. Mr. Chairman, that is a good question, because the calculus or magnet that draws people here to work illegally in- cludes not only the perception of the availability of public education for children or various other Federal welfare programs, but also the tax structure, which includes the EITC.

Americans don't think much about the fact that, within the ille- gal alien networks, these kinds of programs are common knowl-

63

edge, and the dramatic increase in the size of the benefits, given the earned income required to qualify for the program, has to be a magnet that draws people from countries like Mexico to work here illegally and then file the claim.

Chairman Roth. Dr. Grunberg, do you think RAL's significantly encourage electronic filing, and so serve an important public serv- ice to the taxpayer and the IRS? Are there any ways we can en- courage this without losing the taxpayers' money to fraud with the program?

Mr. Grunberg. Yes, I believe the RAL's do encourage people to electronically file, but this year, due to all the changes, the num- ber, the percentage, for example, in our business, that were RAL's is down significantly. Less than 30 percent of our customers obtain a RAL and we still are filing about 89 percent of our returns elec- tronically. A large number of people will file electronically if they can still have the money direct deposited to their bank account or to a bank account that will provide them a check that can be hand- ed to them. The large percentage of our business was that product, as well, this year.

The customers really want to know when and where and how they are going to get their money, and then they are happy. They can say, OK, I can expect it in such-and-such a time frame and they can plan based on that information.

Chairman Roth. I think both you gentlemen heard Mr. Hersch's testimony. Mr. Grunberg, do you think that tax preparers should be licensed with the IRS, much like they are with California? Would that help?

Mr. Grunberg. We are in favor of any regulations that will stem fraud. Beneficial Bank and Jackson-Hewitt, the whole industry has an interest in cutting down fraud, for obvious reasons.

A number of the things that Mr. Hersch mentioned earlier today are no longer possible, in particular, this last year. This is the first year the IRS instituted Social Security number checks for depend- ents, so you can no longer make up dependents' names and Social Security numbers. You can't fabricate dependents to get extra earned income tax credit.

There were also schemes in previous years where a gang would get four or five dependents and use them repeatedly on many tax returns. That can no longer be done. This year, the IRS was check- ing to make sure that any given dependent could only be used once.

So a lot of the schemes that were possible in years past were stopped this year, and I think that the IRS has made great progress in these new checks, in particular, this year. We saw a lot of problems in our office with the rejected Social Security numbers that previous speakers were talking about. Those have been straightened out. If the return is going to be accepted by the IRS for electronic filing, their Social Security numbers have to be cor- rect and those have to be real people on the records of the IRS and SSA.

Chairman Roth. Mr. Caspersen, why do you think the IRS does not run the most basic fraud checks, like you do, on EITC returns?

Mr. Caspersen. The IRS does run basic fraud checks. Unfortu- nately, they are penalized by having equipment that is two to three decades old, literally falling apart and rusting in place. Their com-

64

puters would not be found in any private sector organization that was viable. Their programming is four or five generations old. They just can't do it.

Chairman Roth. Not to interrupt, but it is fascinating that you raise that, because, basically, the Federal Government is spending $25 billion a year on purportedly new technology. We have spent something like $200 billion in the last, I think it is, 10 years. GAO tells me — I have had a study made of this — that the problem is that the technology that we are buying isn't what we need to do the job.

Mr. Caspersen. You are absolutely right, Mr. Chairman. It real- ly is a difference between public and private sector here. If we make the mistakes, if we spend the money and don't get the re- sults, we are out of business. Unfortunately, they have gone down a lot of wrong roads, not intentionally, obviously, but they are not on the cutting edge, not even the cutting edge, they don't even have a knife. They can't do what is necessary.

We do all these fraud checks that I just outlined, check with the credit bureaus, check through 15 separate screens for something like 500 to 1,000 factors. We have created an industry-wide fraud service bureau. All these things, we can do within a matter of 30, 40 seconds, all on-line, and this is nothing unusual. I would love to say we are better than anybody else, but we are not.

You have the Visa, you have the Master Card, you have any of these massive card-issuing companies. They don't take 6 weeks and they are putting out their own money, just like the IRS is putting out their own money. They make decisions instantaneously based on the information, and they detect fraud, they have expert sys- tems.

These all can be done, and it is not at huge cost. This is not a high intellectual thing that you have to go to Los Alamos and farm it out. This is something, and all you have to do is go to Bank One, Chase Manhattan, whoever they want to. The technology is there. It can be expanded, it is not expensive, but they are reinventing the wheel.

The other problem is they have to reach out — we have the same interest — to industry that is there and work in cooperation. Then, I think, we can really cut the fraud. But, you can spend billions of dollars and it is not going to happen at this juncture.

Chairman Roth. The hour is late, and we could spend the after- noon. I am very concerned about bringing the Government into the 21st century. We are still dealing with the industrial age rather than the information age. There is no question but what, as I say, the General Accounting Office has been very critical of other agen- cies, including the IRS, in its updating of data and using modern techniques.

Gentlemen, thank you very much for your patience and for being here today. We may have some additional questions. If so, we will submit them in writing. Thank you very much.

Mr. Caspersen. Thank you, Mr. Chairman.

Mr. Grunberg. Thank you.

Mr. Stein. Thank you.

Chairman Roth. The Committee is adjourned.

[Whereupon, at 1:11 p.m., the Committee was adjourned.]

DESIGN AND EFFECTIVENESS OF THE EARNED INCOME TAX CREDIT

WEDNESDAY, APRIL 5, 1995

U.S. Senate, Committee on Governmental Affairs,

Washington, DC. The Committee met, pursuant to notice, at 10 a.m., in room SD- 342 Dirksen Senate Office Building, Hon. William V. Roth, Chair- man of the Committee, presiding. Present: Senators Roth and Glenn.

OPENING STATEMENT OF CHAIRMAN ROTH

Chairman Roth. The Committee will please be in order. This is the second day of hearings on the Earned Income Tax Credit. Yes- terday's hearing was concerned with the administration of the EITC, while today's hearing is focused on the effectiveness and de- sign of the Earned Income Tax Credit.

This Committee, as part of its mandate, is charged with search- ing out fraud, waste, and abuse within the Federal Government. My concern in this regard has been whether the Earned Income Tax Credit is serving the important goals that it was originally en- acted to serve. From the viewpoint of this Committee, it seems to me that we should review the effectiveness of this program in meeting its goals, and because the program was more than doubled in 1993, this is an excellent time to examine the program.

Yesterday we learned that taxpayers could lose as much as $37 billion over the next 5 years if the fraud and error rate is not re- duced in the Earned Income Tax Credit.

We also learned that taxpayers have already lost in the neighbor- hood of $25 billion since the program began. And I might add, that under other Federal programs, like AFDC and Food Stamps, if the error and fraud rate exceeds 5 percent of the program costs, Fed- eral sanctions kick in.

Yet, here we have a direct transfer program with an error and fraud rate of between 30 and 40 percent, and certainly there seems to have been little success reducing these problems over the last several years, 13 years.

Hopefully that will change. We have learned that a substantial amount of these benefits have been paid and are being paid to ille- gal aliens. And we were shown by the GAO that the cost of the pro- gram could be reduced substantially if we eliminated those with substantial assets and outside income from the EITC.

I have been working on these issues for some time, and fortu- nately the Treasury has taken a leading role in trying to solve

(65)

66

some of these problems. Clearly, fraud continues at an alarming rate, but the Treasury has made substantial efforts to stop benefits for illegal aliens, and has made budget proposals so that benefits would only go to truly needy people.

They have also adopted, in large part, my legislation requiring the matching of Social Security numbers to look for fraud and for illegal aliens claiming the credit.

Finally, I would ask that a copy of Gene Steuerle's recent article from Tax Notes be included in the record. In this article, he criti- cizes the administration for not designing a program that can be administered. He points out the incredible increase in incentives to cheat the tax system as a result of program changes. I regret that Mr. Steuerle's schedule did not permit him to be here today.

So it is with this background that this Committee will begin this day of hearings on the Earned Income Tax Credit to help us learn more about the targeting of the Earned Income Tax Credit and the success of this program in meeting its worthy and lofty goals of en- couraging work and lifting working Americans out of poverty.

[The prepared statement of Chairman Roth follows:]

PREPARED STATEMENT OF SENATOR ROTH— NEWS RELEASE Roth Examines Effectiveness of EITC in Second Day of Hearings

WASHINGTON — In a second day of hearings, Governmental Affairs Committee Chairman William V. Roth, Jr. (R-DE) Wednesday looked at the effectiveness and design of the Earned Income Tax Credit, the fastest growing income transfer pro- gram in the Federal Government. His opening statement follows:

This is the second day of hearings on the Earned Income Tax Credit. Yesterday's hearing was concerned with the "Administration of the Earned Income Tax Credit." This Committee, as part of its mandate, is charged with searching out fraud, waste and abuse within the Federal Government. My concern in this regard has been whether the Earned Income Tax Credit is serving the important goals that it was originally enacted to serve. From the viewpoint of this Committee, it seems to me that we should review the effectiveness of this program in meeting its goals, and because the program was more than doubled in 1993, this is an excellent time to examine the program.

Yesterday, we learned that taxpayers could lose as much a $37 billion over the next 5 years if the fraud and error rate is not reduced in the Earned Income Tax Credit. We also learned that taxpayers have already lost in the neighborhood of $25 billion since the program began. I might add that under other Federal programs, like AFDC and food stamps, if the error and fraud rate exceeds as little as 5 percent of the program costs, Federal sanctions kick in. Yet here we have a direct transfer program with an error and fraud rate between 30 and 40 percent and seem to have little success reducing those problems over the last 13 years. We learned that a sub- stantial amount of these benefits have been paid, and are being paid to illegal aliens. And we were shown by the General Accounting Office, that the cost of the program could be reduced substantially if we eliminated those with substantial as- sets and outside income from the EITC.

I have been working on these issues for some time, and fortunately the Treasury has taken a leading role in trying to solve some of these problems. Clearly, fraud continues at an alarming rate, but the Treasury has made substantial efforts to stop benefits for illegal aliens, and has made budget proposals so that benefits would only go to truly needy people. They have also adopted in large part my legislation requiring the matching of Social Security numbers to look for fraud and for illegal aliens claiming the credit.

Finally, I would ask that a copy of Gene Steuerle's recent article from Tax Notes be included in the Record. In this article, he criticizes the Administration for not designing a program that can be administered. He points out the incredible increase in incentives to cheat the tax system as a result of the Administration's program changes. I regret Mr. Steuerle's schedule did not permit him to testify today.

So, it is with this background that this Committee will begin this day of hearings on the earned income tax credit to help us learn more about the "targeting" of the

67

earned income tax credit and the success of this program in meeting its worthy and lofty goals of encouraging work and lifting working Americans out of poverty.

Chairman Roth. At this time, it is my pleasure to introduce our first witness, my fellow colleague. But before we do that, we will call upon Senator Glenn for his statement.

OPENING STATEMENT OF SENATOR JOHN GLENN

Senator Glenn. Thank you, Mr. Chairman.

Yesterday, I shared with you a couple of stories about working mothers who have relied on the EITC, Earned Income Tax Credit. The examples were Brenda Manders of Columbus, Ohio; and Mindy Heck of Wilmington, Delaware. Today there is another story. I wanted to briefly mention Zorida Hart, a single parent, who works as a switchboard operator at the Council for Economic Opportuni- ties in Cleveland.

She received a credit of $1,978, which she used to pay for day care. And she has put $900 of that credit in the bank to save it for a rainy day. And we wish the U.S. Congress were as penny-wise as Zorida is with the money that she has.

Over the past few days I have heard from a lot of Ohio parents who rely on the EITC to help them with child care so they can have a job. This is a program that is working for Ohioans, working for all the people across this country. At yesterday's hearings we heard several complaints about EITC and I would like to addres° these one at a time, because I suspect we might hear them again.

First, we heard about the problems with fraud — something we spent a great deal of time working on during my years as Chair- man and that Senator Roth is working on now, and it is a critical issue. And I think if we learned any lesson yesterday it is that we had better keep pushing to prepare the IRS for the 21st century through projects like the Tax System Modernization Initiative.

As Mr. Casperson from Beneficial stated yesterday, the IRS is seeking to crack down on fraud but it is hampered by antiquated systems. Yet, we hear a proposal is going to be made on the floor of the Senate over here to cut $100 million out of IRS efforts to modernize. It does not make any sense at all.

We need to change our antiquated systems to uphold public ^con- fidence, not only in the EITC but in our Tax Code generally. It is a program they have worked on for some time and it is just now beginning to really make sense. They are beginning to move ahead with it.

Second, next we heard that the EITC program is just simply out of control. Well, it is not. It just flat is not. The increases we have seen in the program have been mandated, very specifically, by Con- gress. It is roughly on schedule. We have scheduled increases by law, and we have phased them in several years at a time.

The first major increases in the program took place under Presi- dent Reagan. The second increase in the program was laid out under President Bush. And in 1993, under President Clinton, the Congress approved this bill which very specifically sets forth the years that the program increases will take place. I have a copy of the bill right here with me today and on page 432 of this it starts in on the phasing-in, year-by-year-by-year, of exactly how it is to phase in. That is the reason for the increases, not some exploding

68

cost. That was phased-in as part of this program that is laid out in law. That is how it was supposed to work.

If it was not working this way, then we probably should do away with the whole program because that is the way the program is supposed to cut in. So this exploding cost that we have titled on the first chart there, it may be exploding but it was exploding under the law as provided for it, the increases.

And after these increases, the program will return to its nor- mally low rate of growth. In fact, after 1997 it will grow at a rate less than the Gross Domestic Product, GDP. I am sure there are not many provisions you could say that about in the tax package currently under consideration in the House.

Third, some said this program discourages work. I would be sur- prised if this did not come up again today. I am looking forward to it coming up. I have heard from a lot of people who have de- scribed how this EITC has enabled them to work by helping them to pay for things like child care, transportation or work clothes.

But I have not heard from a single person who said they turned down over-time or turned down a promotion because it would affect their tax credit. Just give us a break on that. That just is not hap- pening.

Carol and Roy Wilmonts of Columbus, Ohio, received the EITC for the past 2 years. They use it to help pay bills, giving Carol the flexibility to care for their kids — Amber, Ashley, Autumn and Nich- olas. That is part of the equation some people ignore. Some use the EITC to pay for child care. Others use it to provide the financial flexibility for one spouse to care for the kids. We all talk about fam- ily values. We posture about them when we run for office. We talk about them in the Senate. We talk about them during hearings. And here is something that is working for family values, and we want to destroy it. I do not, but some people apparently would like to destroy it.

There is not a work disincentive for Carol. Roy received a pro- motion. He is now a manager at one of the Muffler King spots. He and Carol no longer receive a tax credit, but they are glad they got one when they really needed it.

This program catches people between when they are on full wel- fare and when they are on full employment and able to make their own way in the world completely. And that is the beauty of this program. It is a transition-type program. It helps people in that toughest time period.

Fourth, we also heard yesterday you get the credit even if you work just a tiny little bit. True. Well, what do you know, then you just get a tiny little bit of credit to go along with it.

It is not something where you work just a little bit and you are on the gravy train from there on. You get just a little bit of tax credit, because you were working just a little bit and your income was low.

Fifth, we heard that those with little earned income but a lot of interest income can take advantage of the system. Well, at the re- quest of the Clinton administration we are putting an end to that kind of thing, through language, in the self-employed health care deduction bill that the Chairman was much involved with.

69

Sixth, we heard about the so-called marriage penalty but no one bothered to mention that the EITC can also encourage marriages. Without the EITC there is a great deal of financial risk for a moth- er on AFDC who does not work and is considering marrying some- one with low earnings.

If she marries, she and her children will become ineligible for AFDC and also lose some of their food stamps, not to mention the loss of Medicaid. The EITC helps to offset those losses. By tying the knot and getting married, the couple will gain an EITC benefit of up to $2,157 if they have one child, and up to $3,564 if they have two or more children.

Now, this would, partially and in some States wholly, make up for the loss of AFDC benefits. Encouraging single mothers, who are on welfare, to marry into working families, that is very worthwhile in my book. One of the biggest problems we face as a Nation is the fact of the dissolution of the family structure. And anything we can do to maintain it, I think, is good.

Now, some people may view the amount of credit we are talking about here as of little consequence. But let me offer an illustration provided by Dan Grunberg in his testimony before our Committee yesterday.

For someone who has a weekly take-home pay of less than $200, the average EITC yearly payment of $1,500 is almost 8 weeks pay, and that makes a very big difference for people in those lower in- come areas.

So, Mr. Chairman, if we are taking a vote today count me as a supporter of the EITC. Can it be improved, and can IRS' following of this be improved? Of course. But we can sit all day with fancy charts and graphics and statistics, but nothing will substitute for the personal experiences of real people like Roy and Carol Wilmonts. They are working hard to get by, and they needed that little extra help that EITC offers, and they worked hard to get it.

This phases in to help people, and as the chart we had here yes- terday showed, it phases in to help people, and as they are able to improve their position it phases out as they no longer need it as their situation improves.

I think that is a good program. Thank you very much.

[The prepared statement of Senator Glenn follows:]

PREPARED STATEMENT OF SENATOR GLENN

Thank you, Mr. Chairman. Yesterday I shared with you a couple of stories about working mothers who have relied on the Earned Income Tax Credit — Brenda Manders of Columbus, Ohio an Mindy Heck from Wilmington, Delaware.

Today I'd like to tell you about Zorida Hart, a single parent, who works as a switchboard operator at the council for Economic Opportunities in Cleveland. She received a Credit of $1,978 which she used to pay for day care. And she's put $900 of that Credit in the bank to save it for a rainy day. I wish the U.S. Congress were as penny-wise as Zorida.

You know, over the past few days I've heard from a lot of Ohio parents who rely on the EITC to help them with child care so that they can have a job. This is a program that's working for Ohioans. More importantly, it's keeping Ohioans work- ing.

At yesterday's hearing, we heard several complaints about the EITC. I'd like to address these one at a time — because I suspect we might hear them again.

First, we heard about the problems with fraud — something we spent a great deal of time working on during my years as Chairman. This is a critical issue.

And I think, if we learned any lesson yesterday, it's that we had better keep push- ing to prepare the IRS for the twenty-first century through projects like the Tax

70

System Modernization (TSM) initative. As Mr. Caspersen from Beneficial stated yes- terday, the IRS is seeking to crack down on fraud but is hampered by antiquated systems. We need to change that to uphold public confidence not only in the EITC, but in our Tax Code generally.'

Next, we heard that the EITC program is simply out of control. Well, I'm here to tell you — it's not. The increases that we have seen in the program have been mandated very specifically by Congress. We've scheduled increases by law and phased them in several years at a time.

The first major increase in this program took place under President Reagan. The second was laid out by President Bush. And in 1993 under President Clinton, the Congress approved this bill which very specifically sets forth the years that the pro- gram increases will take place.

After these increases, the program will return to its normally low rate of growth. In fact, after 1997 it will grow at a rate less than the GDP. I'm sure there aren't many provisions you could say that about in the tax package currently under consid- eration in the House.

Third, then some said this program discourages work. I know this will be coming up again today. And I'm looking forward to it. I've heard from a lot of people who have described how the Credit has enabled them to work by helping them pay for things like child care, transportation or work clothes. But I haven t heard from any- one who turned down overtime or a promotion because it would affect their tax cred- it. Give me a break.

Carol and Roy Wilmonts of Columbus, Ohio received the EITC for the past 2 years. They used it to help pay bills giving Carol the flexibility to care for their kids — Amber, Ashley, Autumn and Nicholas. You see, that's part of the equation people ignore. Some use the EITC to pay for child care. Others use it to provide the financial flexibility for one spouse to stay at home and care for the kids. It's not a work disincentive for Carol. Roy received a promotion and is now manager at a Muffler King. He and Carol no longer receive a tax credit. But they're glad that they got one when they really need it most.

Fourth, we also heard yesterday that you get the credit even if you work just a tiny little bit. Well — what do you know — then you get just a tiny little credit.

Fifth, and we heard that those with little earned income but a lot of interest in- come can take advantage of the system. Well at the request of the Clinton Adminis- tration we're putting an end to that kind of thing through language in the self em- ployed health care deduction bill that the Chairman was so involved with.

Sixth, then we heard about the so-called "marriage penalty." But no one bothered to mention that the EITC can also encourage marriages. Without the EITC, there is a great deal of financial risk for a mother on AFDC who does not work and is considering marrying someone with low earnings. If she marries, she and her chil- dren will become ineligible for AFDC and also lose some of their food stamps — not to mention the loss of Medicaid.

The EITC helps to offset those losses. By tying the knot, the couple will gain an EITC benefit of up to $2,157 if they have one child and up to $3,564 if they have two or more children. This will partially, and in some States, wholly make up for the loss of AFDC benefits.

I'll tell you, encouraging single mothers who are on welfare to marry into working families is sure worthwhile in my book.

Now, some people may view the amount of credit that we're talking about here as of little consequence. But let me offer an illustration provided by Dan Grunberg in his testimony before our Committee yesterday. For someone who has a weekly take home pay of less than $200, the average EITC payment of $1,5000 is almost 8 weeks pay. That makes a big difference.

So, Mr. Chairman, if you're taking a vote today, count me as a supporter of the EITC. We can sit around here all day with fancy charts, graphics and statistics. But nothing will substitute for the personal experiences of real people like Roy and Carol Wilmonts. They're working hard to get by. They needed that little extra help that EITC offers. And they worked hard for it.

Thank you, Mr. Chairman.

71

72

8>ud|sissv »Joddfts A|;suDj awo?uj A|un;a$ p«u»«ia|ddns |

tuauiaiitag saXo|doi3 (Biap«i

1

mi

piDiipaw

sabD|Ud9J^(j|

74

75

76

Chairman Roth. Thank you, Senator Glenn. As I have indicated in my opening statement, there is no question but that a lot of good has been done, accomplished, by this program. And we sympathize and support the concept of programs that provide incentives to get off of welfare and into a meaningful job. The concern, and my con- cern, is that although we all have many, many anecdotes where we show the program working — it is a fact that the fraud and error rate is very substantial.

As was said by the GAO yesterday, the program could result in billions of dollars, as much as $37 billion, of loss, over the next 5 years. Now, it seems to me that that money that is being misspent, whether it is by error or by fraud or for whatever reason, and it can be a lot better utilized if it gets to those we are trying to help.

That is the purpose of these hearings — to try to find out what needs to be done to ensure that the hard-earned taxpayer dollar is being well utilized. It is my pleasure to

Senator Glenn. Mr. Chairman, if I might just make one 30-sec- ond comment here?

Chairman Roth. Sure.

Senator Glenn. One way to not go about correcting the problems is to cut IRS funding, and that is going to be proposed. I under- stand the cuts to be about one-fourth or one-third which makes them less able to ferret out fraud, waste and abuse. So I hope we can defeat that when it comes up on the floor of the Congress.

Chairman Roth. Let me say this, the good Senator has pointed that out a number of times.

Senator Glenn. I have, indeed.

Chairman Roth. And my concern there, again, is just the matter of spending dollars does not necessarily bring back returns to the IRS. As a matter of fact, the GAO has been very critical about the expenditures by IRS and other agencies for high-tech computer and other modern information processes, because it says this money is being wasted.

As a matter of fact, we are spending, as I said yesterday, some- thing like $25 billion a year for high-tech products and GAO says its not doing the job it was intended to. We have spent something like $200 billion, altogether, for computers and modern technology.

So I agree with you, as long as the money is well-spent and se- curing something in return for the taxpayer dollar, we ought to proceed. But I also would urge and insist that the mere expendi- ture of funds does not have a beneficial result unless well spent.

With that, we will proceed to our distinguished friend and col- league, Senator Nickles.

TESTIMONY OF HON. DON NICKLES,1 A U.S. SENATOR FROM THE STATE OF OKLAHOMA

Senator Nickles. Mr. Chairman and Senator Glenn, thank you, both, very much. My appearances before your Committee and my working with you on a lot of legislation maybe I should try to get on this Committee in the next Congress. I compliment both of you because you and your staffs have had a big load this year, probably the heaviest of any Committee before Congress.

'The prepared statement of Senator Nickles appears on page 206.

77

I particularly want to thank you for looking into the Earned In- come Tax Credit. I somewhat question the name of the Earned In- come Tax Credit, it is exploding in cost. I have a chart that shows the cost, one of the reasons why I first started looking into the pro- gram.

I also want to compliment you for your work to basically lead the effort to try and curb the abuse in this system. I remember reading an editorial which said — and I did not bring a copy of it, I will find it and include it for the record — but 40 percent of the District of Columbia might be eligible for the Earned Income Tax Credit.

And I was astounded by that statement, and I am shocked by these numbers. Mr. Chairman, when you look at the explosion of this program — a program that cost $5 billion in 1991, and more than doubled by 1994 up to $11 billion, and it is going to double again by 1996 when it is going to be about $20 billion — this is a program that is exploding totally, completely out of control.

And to have that kind of increase, at the same time where you are reading from a GAO report — and I know, Mr. Chairman, you have already alluded to it — but when it talks about 29 percent of the taxpayers claimed too large of an Earned Income Tax Credit, 22 percent claimed a credit that they are not entitled to, and 7 per- cent claimed more than they were entitled to, and $358 million was erroneously claimed out of about $1.5 billion claimed during that period — and now we are spending many billions more. And 13 per- cent of the taxpayers intentionally claimed too much EITC and these intentional claims accounted for about $183 million, about half of the $358 million and that was when the program was a much lower amount.

Now, we are doubling this program every couple of years, Mr. Chairman, it is out of control. And we need more oversight, and that is exactly what you and Senator Glenn are doing, and I com- pliment you for it.

Mr. Chairman, President Clinton, at the State of the Union ad- dress earlier this year, called the 1993 EITC expansion a working family tax cut.

He said, "We took the first step in 1993 with a working family tax cut with 15 million families with incomes under $27,000, a tax cut that this year will average about $1,000 per family." That was in his State of the Union address.

In fact, Mr. Chairman, the EITC is not a tax cut. It is the Fed- eral Government's fastest growing and most fraud-prone welfare program and we need to call it that. That is what it is.

It is a welfare program. It is an income redistribution program, and it is probably the most fraud-prone program that we have today. And to allow these kinds of expansions without trying to curb this growth, I think, is irresponsible.

The Earned Income Tax rate growth rates for the last 4 years, Mr. Chairman, are not inflation, not 10 percent. They are 55 per- cent, 18 percent, 22 percent, and 55 percent. You do not have any other program, that I am aware of in the Federal Government, that is growing at this kind of a rate, any other program.

People talk about health care costs and other program costs. I have quoted Medicaid costs for the last 4 years at 28, 29, 13, and 8 percent. I have ingrained those figures because they are so large.

78

They are not compounding near as fast as the Earned Income Tax Credit.

Amazingly enough, Mr. Chairman, this kind of shocked me. The EITC will, in Fiscal Year 1996, eclipse the cost of Federal Aid to Families with Dependent Children. Now think of that. It is going to be a bigger program, as far as dollars are concerned, than Fami- lies with Dependent Children.

With regard to EITC's original purpose of reducing the tax bur- den on working families — and that is the way this thing is always described, that is the way that President Clinton described it in the State of the Union — consider the following facts. That is 90 percent of the cost of EITC is a direct handout, not a tax reduction, it is a direct handout, or Federal outlays paid directly to individuals who have zero income tax liability. It is a negative income tax.

I remember Senator George McGovern, when he was running for President I believe in 1972, championed the idea of a negative in- come tax, that is exactly what this is. It is a handout. It is a gift, and it is prone to fraud.

Only 10 percent of the cost of the EITC is a tax refund, and so again, when people say, hey, it is a tax cut for working people, it is not. It is a handout. It is a gift. It is a program that is exploding in cost.

Mr. Chairman, I have an extensive statement and I will just ask you to include that in the record, but I will just make a couple of additional comments.

On the percentages of funds that are going to and lost to fraud and abuse, when you hear 30-some-odd percent may be wasted to fraud or abuse, then I think we have got a program that has run amok and we need to stop it before we have this further compounding of costs.

In Fiscal Year 1994, we are spending $11 billion and we are doing that with this kind of track record? When you find out that 30-some-odd percent is going to people who are misleading, not tell- ing the truth on the applications, or fraud, and yet, we are going to allow this program to compound and expand and basically dou- ble in 2 years.

In 1994 it is going to cost $11 billion, and in 1996 almost double that figure, to $20 billion? I do not think that is responsible. Why does the EITC attract such abuse? Because the EITC offers big cash checks.

The maximum credit for a multiple-child family is $3,114 in 1995. Although recipients can elect to collect the credit in equal paycheck installments, less than 1/1 of 1 percent choose to do so according to the GAO. Again, keep in mind, that means 99.5 per- cent say they want the lump sum. That is the way they are receiv- ing their money is in a lump sum, a much more fraud-abuse prone situation.

Mr. Chairman, Senator Glenn alluded to this, the IRS being over-worked, and they have come under some fire for delaying re- turns because they are trying to do some screening.

I compliment them for delaying the returns for trying to make sure that they would stop some of this. I have heard of stories where individuals were going through various sections of town try- ing to sign people up so that they can show people that they are

79

eligible for this program. I do not know what they are receiving from people for signing people up. I do not know if they are getting a percentage. I do not know what it is. But this is massive abuse and it has to be stopped. And it needs to be stopped before we allow these expansions.

In 1993, we made some big expansions. I know that President Clinton is proud of it, but I believe they are mistakes. How can you expand a program that has this kind of waste, fraud and abuse and allow it to double in 2 years?

I believe that the levels that we doubled in 1993 must be scaled back, and the top income eligibility levels, of $24,394 to $26,691, must be reduced to more closely reflect need.

Further, the expansion of EITC to taxpayers without children, in my opinion, should be repealed. And finally, we should examine other options, including proposals to fold this program into welfare block grants and to the States.

Mr. Chairman, this program needs radical change. The radical change that we gave it in 1993 was a massive expansion at tax- payers expense, and again, an open-ended entitlement program which is growing beyond belief. Senator Glenn, I hope that you are correct in the analysis that this program would level off at $25 or $26 billion.

But when I see a program grow from $5 billion in 1991 to $20 billion in 1996 I think we have a program that is out of control. And when I read the abuse statistics that GAO found in a~much smaller program I say it is time for Congress to act to reforfci this program. I compliment this Committee for having these hearings, and I pledge to you my assistance and willingness to work with you as a member of the Finance Committee to try and make some of these changes that are so called for. _

I thank you for letting me testify today.

Chairman Roth. I thank you, Senator Nickles.

Let me just ask a couple of questions. Your point, I think, is well taken that this is not a tax cut but, indeed, is a transfer of funds. No question about it.

Is it not true that this is the only refundable program in the Tax Code? In other words, to secure an Earned Income Tax Credit one need not pay any taxes at all to be eligible?

Senator NiCKLES. Well, you are exactly right. That is, 90 percent of the recipients last year paid no income tax, had zero Federal In- come Tax liability, and 90 percent of those people drawing these billions of dollars were receiving a negative income tax. They are receiving a check. It is just an income transfer program.

And 10 percent had some tax liability reduced.

Chairman Roth. That is right. There are a limited number that are getting a return of the taxes that would otherwise be due. But that is not critical to the program.

Is this the only refundable program in the Tax Code?

Senator NiCKLES. As far as I know, yes.

Chairman Roth. Now when you talk about fraud and error in the range of 30-40 percent, how does that compare to other pro- grams?

Is it not true that in many of the other programs, for example Food Stamps or AFDC, that if the State has an error and fraud

80

rate in excess of 5 percent, as compared to what we are talking 30- 40 percent here, sanctions are imposed on the State, is that cor- rect?

Senator Nickles. I believe the Senator is correct. And I might mention the fraud rate in this program is so far greater — I remem- ber some highlighted examples of waste in the Food Stamp pro- gram that brought about calls for some reform.

But the percentage of waste or abuse and fraud, and it was fraud, was much, much smaller in the Food Stamp program than it is alleged in this program. And my point being that was when GAO was checking, the program was much smaller. The report was done in October of 1994, so we were spending, I am guessing, in the few billions of dollars. Now, that we are talking about going up to $20-some billion I think it is just that much more open for abuse.

Chairman Roth. Is it not true that when you offer an incentive of a $1.40 cents for every dollar earned and it goes all the way up to roughly $28,000, you are providing a much greater incentive for fraud than has been the case in the past?

Senator Nickles. No question about it. And plus, you are talking about — and Mr. Chairman, I have some charts I will ask you to in- clude in your record — but you are talking about cash benefits, 90 percent of which are received in a lump sum of $3,000 this year, last year it was $2,500, the year before it was $1,400. So when you start talking about giving people a couple of thousand dollars, $3,000, then all of a sudden we find out, well, they start lying about children, how many children they have. That was part of the problem, whether they lie about their ages, that has happened.

I am not saying that every person has done it. But when you find a 30-some-odd percent fraud rate, when these dollar figures go up, Senator Glenn, I think that is one of the reasons you are going to see — when people realize they can get a lump sum of $3,000 and $3,500 in 1996—1 think you are going to find a lot of people trying to abuse the program.

If you can fill out a tax form and get a check from Uncle Sam for cash of $3,000 or $3,500, I think that is a lot of incentive for people to try to abuse the program.

Chairman Roth. Well, as a convicted felon said yesterday, when you can get that kind of return in a matter of 2 or 3 days from elec- tronically-filed returns, it does invite fraud.

But the question is, how can we correct that? That is the purpose of the hearing.

Senator Nickles. Well, Mr. Chairman, I appreciate you having the hearings because you are going to consider a lot of options. I do not consider myself an expert. I consider myself a person who is trying to reign in the total cost of Government. And I see a pro- gram that goes from $4 billion to $20 billion in a period of 6 years — in 1990 we are spending $4 billion and in 1996 we are going to spend $20 billion — that is not acceptable to me. I do not think we can afford it.

Chairman Roth. Did Congress vote for those increases?

Senator Nickles. Congress, in 1993, voted for a massive expan- sion because we said this program was originally entitled for fami- lies with children. Congress said, no, it would apply to parents

81

without children, in 1993. I think that was a major expansion of a program that Congress was not paying very much attention.

Chairman ROTH. It added 5 million.

Senator NlCKLES. It adds what, 5 million participants or people eligible.

Chairman Roth. That is right.

Senator NlCKLES. And a lot of other people are going to find themselves eligible or make themselves eligible for those kinds of benefits. When I mentioned, in 1996, that this program will eclipse the total cost of our Aid for Families with Dependent Children, I hope that sends a signal.

That is a welfare program that we have had for a long time to aid families with dependent children, and to turn this program into an entitlement for parents without children is a massive, massive expansion of a welfare program on this negative income tax.

I mentioned in my opening comment we call this the Earned In- come Tax Credit, that is probably as misleading of a title as one could come up with. Maybe we should just call it the negative in- come tax, or maybe we should call it the refundable low-income — well, it is not even a refundable tax credit. A tax credit assumes that you are going to receive credit for taxes paid, and you will not have to pay it. And then we are making it refundable and negative.

So it is a cash payment. Maybe we should call it cash payments for low-income families. I do not know, but we need to change it. It is not an earned — what is earned about it?

It is a massive welfare transfer program. We should call it what it is. And it is exploding in costs. Those cost figures should concern anybody if they really want to balance the budget. You cannot have programs compounding from $4 billion to $20 billion in 5 years and get anywhere close to a balanced budget.

I do not really think it is going to be leveling off. I think the more people find out that they are eligible for this, they are going to make sure that there are other members that might be eligible for it.

And I am afraid that is what is happening right now. That is one of the reasons why it is exploding so rapidly.

Chairman Roth. Senator Glenn?

Senator Glenn. Thank you, Mr. Chairman.

This explosion that is on the chart there and that you referred to and the whole thing being out of control, it is not out of control. This is the law we passed. And it is the law that is being — every- thing going on under this is under the law.

Now, people can lie about their number of children. They can lie about the wages they got and so on. But so do corporate people lie on their taxes sometimes and they get caught. And it is just as ille- gal for poor people to lie on their taxes as it is for the corporate giants to lie on their tax return also.

So what we have to do, it seems to me, is not throw out the good that this program does because there has been some abuse of it. We will have to correct the abuse. The Gregg proposal which is going to be put forward or I guess has already been, to cut IRS funding, I just cannot imagine anything that is going to increase corporate malfeasance as well as poor people's malfeasance. It does not make any sense.

82

And just at a time when we have the Tax System Modernization Initiative coming up and being ready to put into effect. This pro- jram started under President Nixon. It was expanded in 1986, under Packwood and Bradley's proposal, with Reagan support for this.

It was expanded again in 1990. And it was expanded again in 1993. But I think if every place we see in Government where there is some fraud, waste and abuse which this Committee — of all Com- mittees in the Congress — has done more to document through the years than any other Committee.

We have GAO studies in Committee files that are yards deep on things with the Pentagon and with contractors and with many phases in our society where there has been cheating against the Federal Government.

Now, all at once we want to take this thing that is provided by law, which is a good program if properly administered — and I want to cut down those rates, too, and the rates have come down some on the fraud and abuse — but it is a program that phases in — as we had the charts yesterday — and when the income gets up to a cer- tain point, starts to phase out.

As far as it being a welfare program, if there is no income, there is no EITC. It helps avoid what little people are earning, when they get employment, it helps avoid much of that going back to the Gov- ernment and putting people back on welfare again. That is the pur- pose of the program.

But if there is no income, there is no EITC. But what we have to do is cut the fraud, waste and abuse. You know, people are going to lie on things, whether they are a corporate executive or whether they ^re poor people, if they think it is to their advantage. They will take a chance on it, and sometimes they get caught.

And I want a higher percentage of people that get caught no matter what. No matter what, whether they are poor or rich or cor- porate or whatever they are we have to be setting up a tax system that can catch those people who are defrauding the Government and on one is going to argue that.

I am certainly not going to argue on the other side of that one. But a program like this that really is designed to help bridge this gap between straight welfare and when people are completely self- sufficient — they and their families — and that is what this was de- signed to do, was bridge that gap.

And do we have to make it work better? Why, of course we have to make it work better. But if we are saying that any program that has had some increase in costs, you know, I am sure we have oil depletion allowances that we should cut out of there. I know that would interest the Senator from Oklahoma. Is that something that has expanded through the years? Have other welfare programs ex- panded? Of course. Has education costs expanded? Of course. There are a lot of programs that have expanded in cost. Because they have expanded in costs does not mean that we automatically cut the programs.

We know that the programs do some good. I would submit the same thing here. The fraud that was pointed out yesterday, the Chairman alluded to earlier, that is a terrible thing. And we

83

caught that particular fellow. He has not been sentenced yet. He may be going to jail, we do not know yet.

But we want to set up a tax system and tax system moderniza- tion that will catch the rich, the poor, the corporate, the whoever that is supposed to be paying taxes and getting away with it, with fraud and abuse. And that applies to the poor people as well as to the other, but for every program where we wind up with some abuse we do not say that we eliminate the program because there is abuse in it.

We have banks that get robbed. Do we do away with banks? Oh, it would be a terrible thing, my goodness. We have some banks robbed, let us do away with banks. That would be silly and it is probably a silly example, I admit that. But it makes the point any- way of every time we have abuse we should catch the abuse and punish the abusers and that applies to the people on welfare or people on EITC that is not welfare, as I see it. But it bridges this time period from welfare to when they are self-sustaining in their income.

And if they are lying about things, let us catch them. And we do that with a better tax net to catch them out there. The Tax System Modernization, we are depending a lot on that, not by saying the whole program is bad.

If every time we find somebody that has been abusing the Gov- ernment in a major way, I would say we would have most of the defense contractors in this country in jail right now.

We say, no, we catch them, we fine them, and we do these things, and we try to set-up the new procurement legislation that says we are going to eliminate some of the possibilities of how they can get away with these things. That is the same thing we are doing here, and the same reason why I am glad that the Chairman called these hearings.

I want to correct the abuses as much as anybody else, but I think when we call for eliminating major parts of this program I think that is just wrong, because it is helping a lot of people. It is provid- ing a valuable bridge function between welfare and self-sustaining income, and I think it has done a lot of good.

That is the reason that four presidents — Nixon proposed it, Reagan favored the expansion as I recall, or Packwood and Bradley were the ones who proposed it here. It was expanded in 1990 again, and it was expanded again in 1993 not because it was not working, but because it was doing basically a good job. And then we say, well, but it has increased in cost. Well, that is exactly what we pass laws to provide for.

I think it has been doing basically a good job. But let us get the crooks that are out there abusing the system. That is the main thing, not throw out the baby with the bathwater.

Chairman ROTH. Our time is running out. Just let me make this final statement Basically the goals of the program are highly de- sirable, no question about it. We want to get people off of welfare into meaningful jobs and that was a basic purpose of this program.

But I have to tell you, as just one Senator, I find it very difficult to justify to my constituents any program, no matter how worth- while in purpose and goals, where you have a fraud and error rate in the range of 30 or 40 percent.

84

Now, hopefully that is going down and the whole purpose of these hearings are to try to find out by what ways and means we can make this a program that works, or at least have a replace- ment that will achieve the goals of this program.

So, yes, you can find fraud among any group, whether it is bank- ers or the rich. As somebody said yesterday, this program has made it possible for the poor to work the system. And unfortu- nately there may be some truth there.

But what we are seeking are solutions to the problem so that we have a program that does work.

Senator NlCKLES. I want to thank both of my colleagues for al- lowing me to testify on this hearing, on trying to find the solutions since we do have such abnormally high abuse. If we would have had this GAO report, Senator Glenn, I would venture to say that we would not have had the expansions that we had in 1993. That report came out in 1994. I do not think the big increases that passed in 1993 — they passed for a lot of different reasons — I did not support those.

I did not support the Tax Bill in 1993. I think there were some mistakes in that bill. I think this is one of those mistakes. My guess is if we had had the GAO report Congress would not have passed it. It would not have been included in part of the package.

I would urge this Committee to work with the Finance Commit- tee and enact these reforms early while the program is still below the $10 billion figure, before it becomes a $20 billion program.

Then it is going to be so entrenched it is going to be more dif- ficult to reform. Once these things start they are very difficult to corral or contain. So I would urge the Committee to think of re- forms like repealing some of the expansions that were made in 1993. Let us keep this to a program just with dependent children. That is a big change. Let us reduce the income eligibility. Right now, it goes up to $30-some-thousand and we do not have to have that increase.

The Finance Committee made a small step, and as Senator Glenn mentioned, the administration supported part of it as far as, well, wait a minute, should not we have some limitation on how much passive income they could have before they would still qual- ify for this program. We restrict it a little bit.

I think we can tighten this program a lot. I would just urge the Committee to help us to make these reforms now while they are more manageable, more makeable, than when we have a $20 bil- lion program. I think it is going to be very difficult to make.

Senator Glenn. If the Senator will yield for just a moment, though, there is also a proposal by the Republican Conference to cut GAO funds by one-fourth, even though they are in the process of coming down over several years to cut one-fourth, but you want to do it this year, and you will not have the same kind of GAO re- ports you referred to as being valuable in this particular case.

We are cutting some of these things that are just going to be dev- astating to our ability to administer what should go on in Govern- ment.

Chairman Roth. That is another issue. The private sector has shown you can do more with less, and some of us: think Govern-

85 ,

ment can do the same. But thank you, Senator Nickles. It is a pleasure to have you with us.

Now, indeed, it is my pleasure to introduce the Assistant Sec- retary of Tax Policy, Leslie Samuels. I thank you both for being here today and for your patience. We look forward very much to your testimony. Secretary Samuels?

TESTIMONY OF HON. LESLIE SAMUELS,1 ASSISTANT SEC- RETARY FOR TAX POLICY, U.S. DEPARTMENT OF THE TREAS- URY

Mr. Samuels. Chairman Roth and Members of the Committee, I would like to summarize my written statement and have my writ- ten statement inserted into the record.

Chairman Roth. Your full statement, as for all the witnesses today, will be included as if read.

Mr. Samuels. I am pleased to have the opportunity today to dis- cuss the goals, design, and effectiveness of the earned income tax credit, the EITC. The administration is strongly committed to the goals of the EITC, which are to make work pay and to lift workers out of poverty in the most efficient and administrable manner pos- sible. The administration's commitment to the EITC program for low-income working families — and I repeat, working families — has been demonstrated through more than a dozen legislative and ad- ministrative actions since early 1993.

Since the creation of the EITC in 1975, bipartisan support for the program and its goals has been growing. Republicans and Demo- crats alike have viewed the EITC as the best viable alternative to current welfare programs. With its message of "work pays," the EITC helps reduce dependency on welfare and increase reliance on jobs.

Congress has voted to significantly expand the EITC in three major pieces of legislation: first, with the Tax Reform Act of 1986 during President Reagan's administration; then with OBRA 1990 during President Bush's administration; and, finally, with OBRA 1993 during President Clinton's administration.

The EITC program, as expanded by three Presidents, has made a difference, as Secretary Rubin was recently reminded. Visiting a VITA site, the Secretary met Rhonda Clark, a mother from Mary- land. Talking of her experiences, Mrs. Clark said, "I enjoy working and I want to continue. The EIC gives me some of the help I need — to keep working, to stay independent, and to support my family. It is a help I cannot do without."

Mrs. Clark's experience provides a vivid example of how the EITC makes a real difference in people's lives by encouraging them to work and providing them with additional assistance.

The administration is committed to the EITC in advancing the following four key goals: One, to make work pay for those who might otherwise be on welfare; two, to ensure that an individual who works full-time throughout the year will not live in poverty; three, to target benefits to those with the greatest needs while minimizing distortions; and, four, to make it easier for eligible indi- viduals to claim the credit and for the IRS to verify their eligibility.

1 The prepared statement of Mr. Samuels appears on page 228.

86

I would like to briefly address each of these four goals.

First, for low-income families, the EITC makes work pay in two ways. Unlike many other assistance programs for low-income fami- lies, the EITC is limited to working families. Moreover, the credit amount initially increases rather than decreases for each addi- tional $1 of earnings. The positive link between the EITC and work can help offset the work disincentives created by other tax and transfer programs such as Social Security taxes and food stamp benefits. The EITC, with its positive credit rate on low earnings, is the only program designed to help offset the marginal tax rates imposed by these other programs.

A second goal is to ensure that a person who works at a full-time job for the entire year will not live in poverty. For most families dependent on a worker who earns the minimum wage, it takes both food stamps and the EITC to lift them out of poverty.

Third, the benefits of the EITC should be targeted to families with the greatest needs and those who can best be served by the positive incentives associated with the EITC.

As a consequence, the credit rate is highest at very low earnings levels where individuals are often making the critical choice be- tween work and welfare. Because larger families have greater needs than smaller families, taxpayers with two or more children are entitled to a larger EITC than taxpayers with one or no chil- dren.

Families with incomes slightly above the poverty level also re- quire assistance. Wages have stagnated for many workers and have declined markedly for low-wage workers. By providing the EITC to families with incomes of up to $28,524 in 1996, the program pro- vides protection from the effects of wage stagnation.

The fourth goal of the EITC is simplicity and verification. If eligi- bility rules are simple, taxpayers can more easily claim the EITC and avoid costly errors. With simple and verifiable eligibility rules, the IRS can also better ensure that the EITC is paid only to tax- payers who are eligible for the credit.

Simplicity is particularly important because eligible individuals can claim the EITC directly when they file their tax return. It has been estimated that between 80 and 86 percent of eligible persons claims the EITC in 1990.

The design of the EITC under current law reflects a balance among these four goals. The EITC program design is a direct result of actions taken by Congress in both prior and current administra- tions. Let me review the basic structure of the credit.

The EITC is a refundable tax credit available to low-income workers who have earned income and meet certain adjusted gross income thresholds. Because the credit is refundable, individuals can receive the full amount to which they are entitled, even if the amounts exceed their income tax liability. The amount of the credit increases significantly if an individual has only one or two qualify- ing children.

Let me note here, Senator Nickles mentioned that he thought that 90 percent of the credit is provided by way of outlays. Our sta- tistics show that 25 percent of the credit is a refund of income taxes and the balance is accounted in outlays. So our numbers are 75 percent outlays and 25 percent income tax refunds.

87

If you take the Social Security taxes that are attributable to a worker — that is the employer and the employee portion — and at- tribute that, you will find that the EITC offsets Social Security taxes as well as the income taxes with respect to working families.

To be eligible for the EITC, a taxpayer must reside in the United States for over 6 months. A child qualifies a filer for a larger EITC by meeting relationship, residency, and age tests, and these criteria are described on the board: income test, relationship test, resi- dency, age, and then there is a tiebreaker rule in the case where a child qualifies for the three relationship, residency, and age tests with respect to a working adult.

The credit is determined by multiplying an individual's earned income by a credit percentage. For a family with only one qualify- ing child, the credit percentage for 1996 is 34 percent. The credit amount increases as income increases up to the maximum thresh- old. For 1996, the threshold is estimated to be $6,340; therefore, if there is only one qualifying child, the maximum basic credit for 1996 is $2,156. And you can see on this board the amounts, the phase-in and phase-out range, which I will describe more in a minute.

The credit is reduced and eventually fully phased out once ad- justed gross income or, if greater, earned income exceeds a certain threshold. For 1996, the phase-out range is estimated to begin at $11,620. The phase-out is accomplished by reducing the credit by 15.98 percent of the excess of AGI or, if greater, earned income over this $11,620 threshold.

The credit is completely phased out and is no longer available to taxpayers with incomes above the end of the phase-out range. In 1996, this income level — we are talking about families with one child — is estimated to be $25,109. If there are two or more qualify- ing children, the credit percentage income thresholds and phase- out percentage are higher. In 1996, the credit percentage for fami- lies with two or more children will increase to 40 percent of the first $8,900, with a maximum credit of $3,560. The phase-out per- centage will be 21.06 percent, and the phase-out range will be ex- tended to $28,524.

Workers without children may claim the EITC if they are be- tween 25 and 64 years of age and are not claimed as a dependent on another taxpayer's return. For these workers, the basic credit is 7.65 percent of the first $4,230 of earned income.

Chairman Roth. Could I ask a question? What you are saying is that if an individual has two children, it phases out at, what, $28,000?

Mr. Samuels. It completely phases out.

Chairman Roth. Completely phases out. But up to that point, you do get some earned income tax credit?

Mr. Samuels. Right. It decreases. As you can see, the red line at the top is for two or more children in families with a working parent or parents, and it starts to phase out as the income is ex- tended. And I would say, Mr. Chairman, with any low-income pro- gram

Chairman Roth. That is the question I wanted to ask you. Do you consider $28,000 working poor or middle class, or how would you characterize them?

88

Mr. Samuels. Mr. Chairman, I would characterize this as a pro- gram that is designed for low-income families. It has, you can see on the left of this chart

Chairman ROTH. What would you define as middle class?

Mr. Samuels. Let me just explain.

Chairman Roth. Go ahead.

Mr. Samuels. Because I think you have to understand the dy- namics of how to design any low-income program. In our case — and this is not just this administration. This is the way the program has been designed at the beginning. You have to have a phase-out.

Chairman Roth. But this phase-out was extended very substan- tially, was not it, 2 years ago in the

Mr. Samuels. It was extended, and in order to avoid the high marginal tax rates

Chairman Roth. Before 1993, what was the phase-out then?

Mr. Samuels. The phase-out range was extended by $3,000.

Chairman Roth. To $25,000; is that correct?

Mr. Samuels. It would have been the point where we phased out the one-child family. That was what was extended in 1993. So if you look at the blue line, which is just about $25,000, that is where the phase-out ended before the OBRA 1993 expansion. And as I say, anytime you have a low-income program where you have an incentive to work, you have to have a phase-out. Otherwise, if you don't have a phase-out, you have a cliff, which I think everyone would view as unacceptable. So you need to have a phase-out which does not involve marginal tax rates on those in the phase-out range that are unacceptable. So in this particular case, Congress passed a phase-out range that ends at $28,500.

Senator Glenn. If the Chairman would yield, was

Chairman Roth. If I could just finish, and then I will be glad to yield.

What is the median income of a typical American family?

Mr. Samuels. For a four-person family, it is nearly $50,000, the median income.

Chairman Roth. Okay.

Senator Glenn. What is poverty level then for four?

Mr. Samuels. About $16,500.

Senator Glenn. The only question I was going to ask, and I don't know the answer to this: Was the old program about the gray line there and the new program that was put in in 1993 — is roughly the red line?

Mr. Samuels. For families with two or more children.

Senator Glenn. The red line would basically be the expansion of the program in 1993, then.

Mr. Samuels. For families with two or more children.

Senator Glenn. Okay. Thank you, Mr. Chairman.

Chairman Roth. Please proceed, Mr. Samuels.

Mr. Samuels. Mr. Chairman, for these three credits, the income thresholds for both the phase-in and phase-out ranges are adjusted for changes in the cost of living. There are two ways to receive the EITC. Individuals can claim the credit by completing a Schedule EIC when filing their tax return. Alternatively, individuals with qualifying children may elect to receive a portion of their EITC in advance by filing a Form W-5 with their employers. These individ-

89

uals are entitled to receive on an advanced basis up to 60 percent of the credit allowable for a family with one qualifying child.

As I mentioned at the outset, the administration and Congress have taken a number of legislative and administrative actions dur- ing the past 2 years to improve the effectiveness and administra- tion of the EITC. OBRA 1993 expands the EITC and makes the program more effective in achieving its policy objectives.

First, OBRA 1993 increased the returns from working for those outside the workforce and for other very-low-wage workers. For low-wage workers with two or more children, the EITC will fully offset the combined employee and employer portions of the Social Security taxes and the food stamp benefit reduction formula.

The OBRA 1993 expansion was also a critical step forward to- ward achieving the goal that a full-time worker should not live in poverty if he or she works throughout the year. In combination, a full-time minimum-wage job, food stamp benefits, and the EITC should lift a single parent with one or two children out of poverty. The OBRA 1993 expansion significantly closes the poverty gap. However, since the minimum wage has not kept pace with infla- tion, the job is not yet completed. And this is why the President has proposed that the minimum wage be increased over 2 years by 90 cents.

In OBRA 1993, we also balanced the goals of providing the need- iest families with sufficient income support while minimizing the marginal tax rates placed on families with higher, but still modest level of incomes. While the definitive effect of OBRA 1993 cannot be measured yet, we believe that the legislation will, on net, in- crease work effort.

While some workers with large families will face slightly higher marginal tax rates, they are unlikely to change their behavior much in response. These are individuals who are already very at- tached to the workforce. They cannot easily adjust their hours of work in response to small changes in tax rates. They need both their jobs and the EITC to meet their day-to-day needs, and most employers will not allow them the discretion to work fewer hours.

The effect of the higher marginal tax rates on some workers in the phase-out range will be outweighed by the effect of the increase in the credit rate. By making wo**k pay, the OBRA 1993 increase in the credit rate will encourage non-workers to enter the workforce and other low-income part-time workers to increase their hours of work.

Finally, OBRA 1993 simplified the eligible credit for the EITC beginning in 1994 by eliminating the two supplemental credits for health insurance coverage and for taxpayers with children under 1 year of age. Largely as a consequence of the repeal of these provi- sions, we have been able to great simplify Schedule EIC, and as you will see here, this is the new Schedule EIC, and it is much simpler for people to deal with than the old schedule as a result of the repeal of the two supplemental credits. And as I said, that repeal takes place in taxable years 1994. It was still in the law in 1993, so we haven't seen the benefits of this simplification yet re- flected in the compliance figures that I know this Committee has been reviewing.

90

The Uruguay Round legislation contains four provisions to im- prove compliance as well as the targeting of the EITC to those with the greatest need. First, non-resident aliens will not be entitled to the credit. In addition, effective for 1994, prisoners will not be eligi- ble for the EITC based on their earnings while incarcerated.

The legislation improves the administration of the EITC in two ways: By 1997, taxpayers will be required to provide taxpayer iden- tification numbers for all dependents and EITC-qualifying children. The Department of Defense is also required to provide military per- sonnel and the IRS with information regarding housing and sub- sistence allowance. As a result of the GATT legislation, the IRS will be able to verify eligibility of EITC claimants in a more effi- cient way because of these two requirements.

As Commissioner Richardson testified yesterday, the administra- tion has taken a number of steps to ensure that eligible individuals know about the EITC and the advance payment option. The admin- istration has also taken significant steps to ensure that only those who are eligible for the EITC receive it.

The administration included several proposals to improve the targeting and administration of the EITC in this year's budget sub- mission, and we are ready to work with Congress on these propos- als and hope that Congress promptly enacts them.

First, the EITC would be denied to taxpayers having more than $2,500 of taxable interest and dividends beginning in 1996. This cap, under the administration's proposal, would be indexed for in- flation after 1996. The proposal would improve the targeting of the EITC to families with greatest need, as taxpayers with significant interest and dividend income can draw upon the resources that produced this income to meet family needs.

The administration's proposal, with some modification, is includ- ing in the conference agreement for H.R. 831, a bill to extend and expand the 25-percent health insurance deduction for self-employed individuals that the Chairman has been very involved in.

The conference agreement lowers the investment income cap to $2,350 and expands the categories of income subject to that cap to include tax-exempt interest as well as net positive rents and royal- ties. The investment income cap would not be indexed. At the out- set, approximately 550,000 taxpayers would lose benefits under H.R. 831.

We are very concerned that the investment income cap is not in- dexed. We believe that this income cap should be indexed in the same manner as all other income parameters for the EITC and just as personal exemptions, standard deductions, and tax rate brackets are indexed. Without indexation, savings will be discouraged over time, and the number of persons affected will increase. By the year 2000, the number of affected taxpayers would increase from about 550,000, where we are now, to 650,000.

Our second budget proposal would improve targeting of the EITC. Only individuals who are authorized to work in the United States would be eligible for the EITC beginning in 1996. Taxpayers claiming the EITC would be required to provide a valid Social Se- curity number for themselves, their spouses, and their qualifying children. Social Security numbers would have to be valid for em- ployment purposes in the United States.

91

In addition, the IRS would be authorized to use simplified proce- dures to resolve questions about the validity of a Social Security number. Under this approach, taxpayers would have 60 days in which they could either provide a correct Social Security number if there was not a correct Social Security number on a tax return, or request that the IRS follow the current law deficiency proce- dures. If a taxpayer failed to respond within this period, the 60-day period, he or she would be required to refile the tax return with correct Social Security numbers in order to obtain the EITC.

In combination, these provisions would strengthen the IRS' abil- ity to detect and prevent erroneous refunds from being paid out. And I would urge that Congress act quickly on these proposals, and in particular the simplified procedure approach which we think would be very useful for the IRS in dealing with the compliance is- sues that are under discussion by this Committee.

In the administration's 1994 welfare reform proposal, H.R. 4605, States would be given additional flexibility with respect to the EITC by allowing four demonstration projects to determine the ef- fects of alternative methods of delivering advanced payment of the EITC. We hope Congress would act on this demonstration project proposal.

With respect to other suggestions to modify the design of the EITC program, the administration will evaluate them using the same criteria we apply to our own proposals. Does a proposal make work more attractive to those outside the workforce and to others with minimal ties to the workforce? Does the proposal reduce the poverty gap for families with full-time workers? Does the proposal improve the targeting of the EITC to the neediest in the least distortionary manner? And, finally, does the proposal make it easi- er for taxpayers to claim the EITC and, importantly, for the IRS to verify their eligibility before refunds are paid out?

We are concerned that many of the suggestions under discussion do not meet these criteria. For example, we would object to a res- toration of the pre- 1991 law which based eligibility for the EITC on dependent and head-of-household rules. These rules were very difficult for taxpayers to understand because they did not conform well to ordinary people's notions of caring for a child. The Bush ad- ministration worked together with Congress to modify and simplify these rules in 1990. Restoring the old rules would increase com- plexity and non-compliance.

We would also have concerns about lowering the investment in- come cap below the $2,350 figure in H.R. 831. While we do not want taxpayers with substantial investment assets to receive the EITC, we also do not want to discourage hard-working individuals from being able to accumulate sufficient capital for important in- vestments, such as a down payment on a home or a business.

We recognize that the targeting of the EITC to the neediest workers can also create work disincentives. As incomes increase above the beginning of the phase-out threshold, EITC benefits begin to decrease. As a consequence, the marginal tax rates for families of modest means increase. Among recipients in the phase- out range, the EITC could cause some individuals, primarily the spouses of other workers, to reduce the number of hours worked in response to high marginal tax rates. And this chart shows the pa-

92

rameters of the EITC for a family with two or more children, and it shows how you increase the EITC up to the plateau; and then the phase-out range, which is what we are discussing now, is set so that it does not create unduly high marginal rates for those in the phase-out range.

The EITC is similar to any benefit program which targets assist- ance to the very neediest families. We cannot target assistance to low-income families without causing marginal tax rates to increase for families with slightly higher incomes. However, we can seek to minimize such distortions.

In designing a program to make work pay, it is impossible to re- duce the marginal tax rates in the phase-out range without either increasing the marginal tax rates on other families or reducing the credit amount for most families. There is a tradeoff between the appropriate level of the EITC and the length of the phase-out range. A balance must be struck between the benefits of a higher credit rate for very-low-wage workers and the costs of either higher marginal tax rates for families in the phase-out range or a longer phase-out range.

There is, in short, no simple solution. Proponents of alternatives should demonstrate how their proposals result in a better balance of costs and benefits than current law. Last month, the Senate Re- publican Task Force on Entitlement Reform released a draft report which included a recommendation to reduce expenditures on the EITC by $27 billion. Under the task force's suggestion, the income thresholds for determining EITC benefits would no longer be in- dexed. This represents a reversal of one of the key provisions in the Tax Reform Act of 1986. As a result of legislative action in the 1980's, the Tax Code contains a number of provisions which are in- dexed for inflation each year. These include the personal exemp- tion, the standard deduction amount, and the width of the income tax brackets. It is unfair to suspend indexation on the one provi- sion which is solely targeted to low-income taxpayers.

Under current law, an estimated 21.1 million taxpayers will claim the EITC in 1996. If benefit thresholds are not adjusted for inflation since 1994, participation would shrink to 18.2 million tax- payers by the year 2000.

The task force justified these reductions on compliance grounds. However, there isn't a single compliance issue that is addressed by eliminating indexation. Instead, the suggestion denies eligibility for the EITC to millions of law-abiding, working taxpayers and reduces the benefits of millions of other who are playing by the rules.

The administration is committed to improving compliance with the EITC rules. Its actions in the last 2 years are clear evidence of this commitment. Consequently, the administration strongly op- poses the task force proposal which would reverse the long-stand- ing bipartisan support for a program that makes work pay.

Mr. Chairman, this concludes my remarks. Thank you for allow- ing me to testify today, and I would be pleased to answer any ques- tions.

Chairman Roth. Les, thank you for being here today. I want to say that I think there are some changes made by Treasury that are highly desirable and we hope they will reduce substantially the problem of compliance.

93

I know that my good friend and colleague has another hearing that he also wants to attend, so I am going to ask Senator Glenn to go first.

Senator Glenn. Thank you, Mr. Chairman. We have too many hearings scheduled at the same time, and I am supposed to be at another hearing. I appreciate your courtesy in letting me go first here this morning. I will try and be brief.

Do you agree with GAO's findings that 29 percent of EITC claim- ants erroneously received overpayments of the credit in 1994? Is that a valid figure?

Mr. Samuels. Senator Glenn, the GAO report is based on a sur- vey that the IRS undertook in January of 1994 with respect to the 1993 tax year. And that is a number, the preliminary number that we have seen. We are awaiting final numbers. We think that those numbers do not obviously reflect all the changes that we have pro- posed, some of which were included in OBRA 1993 and the GATT legislation and our administrative positions. So we acknowledge that there is a serious compliance issue, and we have been commit- ted to trying to solve that problem.

Senator Glenn. But what was the sampling size, do you know?

Mr. Samuels. It was about 1,000 electronically filed returns in January of 1994, and I think that one question about it is that it is just electronically filed returns, and it was very early in the fil- ing season. So it isn't necessarily indicative of what you would find for a full filing season for both electronic and paper returns.

Senator Glenn. Are you doing any further analysis of that, or is IRS doing any further sampling to get a broader scale look at it?

Mr. Samuels. Senator Glenn, in this filing season, the IRS is going to undertake a survey of both paper and electronically filed returns and analyze it. We would hope that that will be available in the fall once they have been able to analyze it, match, make sure they understand what the error rate is.

Senator Glenn. The 1,100 sample was out of how many returns total that are filed under EITC?

Mr. Samuels. There were about 15 million returns filed in 1994 with respect to the 1993 tax year.

Senator GLENN. I don't want to denigrate their 1,100 or so or whatever it was, because we use national polling now, and with 800 or 1,000 or so we determine what 250 million Americans are supposed to be thinking, saying, wearing, eating and so on. Polling is pretty sophisticated, but I sort of question whether this number would represent a fair shake at this — although I am not trying to say there isn't a lot of fraud and abuse. We know that. But these figures would show that it has come down from 40 percent a few years ago down to about 29, which isn't very good. We would like it to come down a lot faster than that, of course. But we are looking forward to getting some additional figures.

You mentioned the support test problem that we got away from with some legislation in the past. Would you expand on that, the impact on error rates if Congress would reinstate the support test? Would that be something that would get away from the problems we have had, or would it just complicate things so much more that it — I think you indicated you thought it might increase the problem instead of decrease it. Is that it?

94

Mr. Samuels. Senator Glenn, we believe that if we went back to the pre-OBRA 1990 law which had a dependency test based on the support that was provided by the working member of the family, that would increase the error rate, and we would strongly urge the Committee not to do that.

Senator Glenn. We have had a big increase now. In the mid- 1980's, EITC costs were running in the neighborhood of $2 billion a year. I believe that figure is correct. Even with inflation, if we get into that, by 1997 the EITC will cost the Federal Government one way or another about $29 billion a year.

What amount of that came out of OBRA, the Omnibus Budget Reconciliation Act of 1993? Was that the major increase? Let's say that $2 billion through the years would have increased anyway up to, say, four or five. Is that additional increase due to OBRA 1993?

Mr. Samuels. Senator Glenn, my understanding is that about $7 billion of the $29 billion is attributable to OBRA 1993, so that is seven out of 29. Ten billion dollars of the $29 billion is attributable to OBRA 1990. So the 1993 change is actually less than what was done in OBRA 1990.

Senator Glenn. So 10 of that came out of the Bush administra- tion and support for it here on the Hill, of course. Then the seven came out of the Clinton 1993.

Mr. Samuels. Right.

Senator Glenn. Okay. We keep hearing about marriage pen- alties. Doesn't EITC reward marriage for many low-income tax- payers instead of going the other way?

Mr. Samuels. Senator Glenn, yes, the EITC would reward mar- riage. When you have, for example, someone who is on welfare who marries a low-wage worker, they would be getting a marriage bonus. With any discussion of marriage penalties, I think it is ap- propriate to always take into account the fact that there are signifi- cant marriage bonuses in our tax system as well as in the EITC design.

By the way, that is not to say that we aren't concerned about marriage penalties, but you do have to look at both sides of it.

Senator Glenn. Do you have an estimate on how many families EITC has gotten above the poverty line or how many Americans have been affected by this beneficially? Let's say, how many got out of poverty, families and people, overall?

Mr. Samuels. My recollection is about 1.4 million families were taken out of poverty and raised above the poverty line.

Senator Glenn. I had some figures from somewhere — I think they are from the Census Bureau — but they indicated that EITC helped about 548,000 families to get above poverty in 1993 and about 1.8 million persons were lifted from poverty as a consequence of EITC. I think those are from the Census Bureau. Is that a valid figure, do you think?

Mr. Samuels. That is a figure that the Census Bureau has put together, and we have an analysis that looks at it in a slightly dif- ferent way, and our number is somewhat higher.

Senator Glenn. Now, looking at it the other way, is there a chicken-and-egg approach? Are employers paying lower wages be- cause they know Uncle Sam will come in and subsidize their pay- roll?

95

Mr. Samuels. Senator Glenn, we don't think that that is a mate- rial issue. It is difficult for employers to know what their employ- ees' income is, how many children they have. So we have not seen evidence that this is a problem with the program.

As I say, it is geared also to minimum-wage workers, in part to bring them out of poverty with food stamps and the EITC. And, ob- viously, a minimum- wage worker would not be subject to a reduc- tion of wages.

Senator Glenn. Some of the opposition to the EITC is because there is inherent fraud, and I want to root out fraud, and that is what much of this Committee's work has been on for the last sev- eral years. But to say we throw that out, like I said earlier, that is like saying that because people rob banks, we should close the banks down. It may be a crude analogy, but I think IRS' auditing and enforcement activities must be strengthened. The honest re- cipients of EITC assistance shouldn't be singled out because of problems with IRS enforcement and sleazy cheats of all stripes who try and game the system.

How does fraud in the EITC program compare to other pro- grams? Are there any recent estimates of the amount of, say, cor- porate tax fraud each year or other types of tax fraud?

Mr. Samuels. Senator Glenn, I think one general comment, and then let me answer the specific one. I think we are committed, the administration is committed, the IRS is committed to improving compliance with the EITC. That is why we have taken all these steps.

There is, however, a problem with compliance not just with EITC, but in other areas of the tax law. And I know the Commis- sioner has testified about the tax gap, the compliance gap. It, for example, is fairly high, and we have statistics. We can provide those to you but the problem is with compliance by businesses with income that is not subject to information reporting. And there was an op-ed piece in last Saturday's New York Times by an owner of a small business complaining that there was all this fraud and abuse going on in the restaurant industry.

Now, we do not suggest, no one would suggest that somehow or another we would change the tax benefits available to small busi- ness because there was a problem in the restaurant industry. From personal experience, what we should do is deal with the compliance issue and try to do our best to improve compliance. And that is what we have been trying to do with the EITC and that is what we are trying to do in other areas of the Tax Code.

Senator Glenn. Those are all the questions I have this morning and we may have some additional questions for you that we would like to submit for the record. Senator Roth, I appreciate your con- sideration in letting me go so that I can get over to my other hear- ing. Thank you, very much.

Chairman Roth. Thank you, Senator Glenn.

Is it not true that the most serious fraud problems faced by IRS today are in the area of EITC? And if not, what are the areas that you say are, percentage-wise, more serious?

Mr. Samuels. Mr. Chairman, it is difficult to kind of identify one are or another, one is more serious than the other. As I said to Senator Glenn, we are committed to reducing the error rate in the

96

EITC and to minimize, to the fullest extent we possibly can, the fraud and abuse that has been identified.

This is a program that rewards people who are working so we think it is a very important program. But if you look at other areas that are under-reported

Chairman Roth. Just a minute, let me

Mr. Samuels. Sure.

Chairman Roth [continuing]. Since I am asking the questions; you say "those that are working," Even though there is no work re- quirement beyond having earned income. So whether one works 1 hour, 1 day, 6 months or a year is not the relevant factor. It is what your earned income is, is that not correct?

Mr. Samuels. It is your earned income. And as you can see, as your earned income goes up

Chairman Roth. So you do have this ironic situation that if you earned, let us say, as a rock star $5,000 in 1 day and that is all you did, you would be eligible for earned income tax credit if you had children?

Mr. Samuels. As a theoretical matter that is a possibility.

Chairman Roth. And is that not really part of the problem that we are facing, that you rightfully talk about the need for simplicity in the program and not letting it become too complex. However, but by keeping it simple we get into these difficult situations where the program can be utilized by those that are not truly working poor.

Mr. Samuels. Mr. Chairman, if we thought that that was any- thing more than an extremely rare case, I would agree with you. But we do not think that example is anything other than an aber- ration. And I guess, at least for that rock star, one would hope that he would work more than 1 day a year.

Chairman Roth. But we are faced with the fact that there has been roughly 30-40 percent fraud, abuse and error for many dif- ferent reasons. You would agree that it is hard to justify a program that would have, on a continuing basis, that kind of error, fraud and abuse?

Mr. Samuels. Mr. Chairman, I would say that we should wait until

Chairman Roth. I am asking the question.

Mr. Samuels. Well, actually

Chairman Roth. In the past it has been what? The GAO has said roughly 30-40 percent. Would you agree that that is not satis- factory?

Mr. SAMUELS. Absolutely. We agree that that is not satisfactory and that is why we have been making all these proposals and im- plementing all these changes to make sure we can reduce the error rates. There are error rates and understatements of income in other areas of the tax law. The IRS is working hard to try to deal with that, but that does not suggest that we should take away the tax benefits from those groups who have a high under-reporting of cash income.

Chairman Roth. The indication, of course, is that the fraud and abuse error in this area is the most important problem faced right now. You are right. There are other areas of serious concern, but I think most people would agree that we cannot continue to have a program where it is estimated that as much as $37 billion in the

97

next 5 years may be misspent. It is not going to those that you and I want to help.

I think you would agree with that. But whether it turns out that the amount of misspent money is $37 billion or somewhat less, we do have a serious problem in this area, would you not agree?

Mr. Samuels. The administration believes that there has been a problem and that is why we have taken, as I said in my oral testi- mony, over a dozen steps to try and improve the program.

I think when people make the estimates, I do not know exactly how that estimate was calculated, but I think it was on the basis of a program that does not reflect the steps that have been taken. What I would urge Congress to do is to wait until our steps are taken.

Chairman Roth. Let me point out, you made the observation be- fore, that there are a number of steps being taken that should help. At the same time, yesterday, we had a great deal of testimony that there is a tremendous expansion in the program. An individual can receive as much as 40 cents on the dollar. This is going to invite even more fraud and abuse.

So you have both factors coming into play. As I said in the begin- ning of my remarks, I applaud many of those steps. I have been talking about them for the last year or so and hopefully they will help significantly.

At the same time, because of this doubling of the program; in- creasing it to a $1.40 for every additional dollar one earns within the limits, you also invite more fraud. The question we are trying to answer today is how do we correct that?

But let me go back, if I might, just for a moment to the question, of whether there are any other tax programs that have a non- compliance rate higher than EITC and, if so, which ones?

Mr. Samuels. Mr. Chairman, that is a question that we have ac- tually been trying to examine. And one of the interesting problems that we face is that we do not have information about compliance rates for other programs, other parts of the Tax Code current infor- mation that would allow us to give you a thoughtful response to that question.

The tax, as you know, we did the so-called TCME program in 1988, there was one scheduled for this year. I think that once that program is completed we can give you a much better answer that is not based on kind of anecdotal evidence.

We do know we have a very serious compliance with respect to under-reporting of income that is not reflected in the 1099 report- ing forms. That is a very serious problem and one in which the IRS is trying to address and I know Tax System Modernization Initia- tive will help do that.

Chairman Roth. One of the problems, as I understand it, of ad- ministering this program is that the problem the IRS normally faces is that the taxpayer understates his income. Now, we have exactly the opposite situation. That is where people have an incen- tive to overstate, to try to increase the Earned Income Tax Credit, so that many people say it is very difficult for the IRS to admin- ister this kind of a program because it really is a transfer of income and it is a different situation than normally the IRS faces, is that correct?

98

Mr. SAMUELS. Mr. Chairman, it is correct that if you overstate your income, if you are in the phase-in range, you can increase your credit.

We have been looking at that. To the extent the taxpayer pro- vides a W-2 the IRS can match and verify that. So we think that over time that aspect of the program which is what most taxpayers report, wage income, can be matched.

The area where there is the potential for abuse is in the schedule C which is the self-employed schedule. And that is something that the IRS is carefully looking at. To date, we do not have any evi- dence that there is any material problem but we are very con- cerned about that and are going to observe it very carefully to see whether there is any potential increase or any increase in non- compliance in that area.

I know that has been discussed. Some people have suggested that we do not give EITC to those who have earnings from self-em- ployment activities. I think, to date, we do not feel that we have enough evidence that there is so much noncompliance there that that is something that we should recommend.

We have a problem in saying that people, only employed workers should get the EITC, as opposed to self-employed. We think at this stage we should not be making those distinctions.

Chairman Roth. Let me interrupt you here. That is what con- cerns me. I understand the problem you face. But the issue is that time and again, we say we cannot afford to change the rules. Whether it is self-employment, or the marriage penalty or child welfare, or whatever, because doing do so makes the program so complicated that it becomes very difficult to administer.

On the other hand, you face the fact that if you do not correct it, then there can be these glaring examples that make the whole program appear beneficial to those who are not deserving. Just for example, take assets. You can have unlimited assets but if it does not bring in income it does not make you ineligible. Is that correct?

Mr. Samuels. That is correct.

Chairman Roth. And the same thing is true on payments for children. A person can be getting $60,000 a year, a spouse can be getting $60,000 a year for child care, but that does not count. And the reason we are told, in each of these cases that we should not change the rules, is because it makes it too complicated. And there is some merit. I am not saying that. But to me, we are on the horns of a dilemma. We have a program where you either have to let a lot of people escape through loopholes that raise serious questions about the effectiveness of the program, or you make it so complex that it becomes very difficult and very expensive to administer.

And yet, if we look at the record, for the last several years, we find that, at least in my judgment, the fraud, error, abuse rate is unacceptable. We cannot continue a program with this kind of error rate.

So we either have to find the means of correcting it or perhaps looking at some different approach. Because I think we all agree that a basic goal of Government is to get people off of welfare and various programs of that sort and into meaningful jobs. So there is no difference in intent or goals or purpose, but there is a serious question of method.

99

Even if the figure of $37 billion that will be misspent in the next several years is way off, let us say it is half of that. In these days of deficit, I do not think we can afford to have $18 billion, or what- ever the figure is, going to those who do not deserve it.

So, to me, we are genuinely on the horns of a dilemma as far as this program is concerned.

Mr. Samuels. Mr. Chairman, I would make one suggestion. That is that the administration, as you know, is committed to reducing the error rate and to improving compliance. And we have taken nu- merous steps with that goal in mind. I would say one thing, we ap- preciate congressional support for those steps. For example, the delay of refunds in order to match the Social Security numbers is a very important part of our program.

But I would hope that before Congress acts that they look at the facts of today's program and not rely on data that is out of date and does not reflect today's program.

So I think that we have recognized that there is an issue. We are attacking that issue. We think that what we are doing will make a difference. And we have been trying to target the program better. We think the proposals that we made, in OBRA 1993 and the Uru- guay Round legislation, the proposals that are in our budget, all go to better targeting and to improving compliance. And that we should wait to see whether we have been successful. We think we will be.

And as I say, one of the proposals in the budget which has not yet been acted on that we announced, which I know you support, is to deny EITC to undocumented workers. And a crucial part of that is also this simplified procedure for denying recipients EITC if they do not provide a valid Social Security number. We believe that you should have a valid Social Security number when you file a tax return. When you go to the bank and go to the ATM machine, you cannot get money out unless you put your valid number, ATM number, into the money dispensing machine.

We think that taxpayers should be asked to do the same thing with respect to their tax returns. We think we have designed a sys- tem in our proposal that will be fair to people because we do not want to, in any way, abridge taxpayer rights. We think we have a proposal which is a fair proposal that gives people time to re- spond. And if they do not respond in time, they have to refile. It is like putting your card back into the ATM machine.

So we think that when you look at all of these proposals that we have a very good program.

Chairman Roth. Let me go ahead, because I do have more ques- tions I want to ask you.

Knowing that other means-tested programs like AFDC and Food Stamps are subject to Federal sanctions if the error and fraud rate exceeds 5 percent, it seems clear to me that a 30 percent or more error rate is not acceptable. Even if we cut it in half, that is still 15 percent.

I find that that is unacceptable and I think unacceptable to the taxpayer. Would that suggest that more low-income people might be benefited if we took this program out of the tax law and distrib- uted benefits in a more efficient manner?

100

Mr. Samuels. Mr. Chairman, I think that one of the significant advantages of this program is that its goal is to reward work, as opposed to other Government transfer payment programs. It does not require people to wait in long lines at welfare offices, or wait and go through the, in some sense, embarrassment of applying for Food Stamps.

Chairman ROTH. But yet we face the fact that from 30-40 per- cent of the funds are going to the wrong people, are being misspent. Yes, some are benefiting and do have an incentive to work and work longer. But unfortunately, at the same time, you have this other problem. And then we come up again and you say, well, we cannot do anything about those problems. We cannot con- sider assets. We cannot consider child payments. We cannot con- sider that because it makes it too complicated.

Perhaps we might be better off using a bureaucracy that already exists and make direct payments.

Mr. Samuels. What I would just say is that one, there are two issues that we are talking about. One is the compliance issue. And we appreciate Congress' support in our efforts to increase compli- ance and reduce the error rater. And we have suggestions. We are willing to listen to other suggestions that anyone might have.

The second issue is the targeting issue. That is where you get into the questions of adding complexity. We think that the way the administration and Congress has dealt with the income test on in- vestment assets is a fair and appropriate way, subject to this issue about indexing. With respect to other suggestions, we think in terms of targeting, you have to look at the overall population, who you are trying to help — the people you want to take out of welfare and put them on work. So you say to yourself, do we really have a problem that deserves making 20 million taxpayers deal with extra lines on their tax returns?

For example, your question about the rock star who works 1 day. We do not think that that, in any way, is a material part of the population. It is a theoretical problem. It is not, in our view, a real problem.

Chairman Roth. But it is true that there is no incentive as far as work hours are concerned. I mean whether you work 1, 5, or 50 hours or even half a year it has no relevance. So you are not nec- essarily giving that incentive.

Mr. Samuels. What I would say is if you look at the phase-in range, we are giving an incentive for every extra dollar of work. We are talking about a population where every dollar is important. This is people who are trying to get out of poverty and onto a fair standing of living. These are people who are on welfare who we want to make work.

I think that is the focus of this program. We think that is why this program has had bipartisan support since its beginning and why it has been increased, because everybody agrees that that is the fundamental objective of the program and that the program makes sense. So we think this is a program, in terms of targeting, where we, with the proposals that we have had

Chairman ROTH. Let me — because time is passing and we have to move on — just let me make a couple of observations.

101

Number one, about the bipartisan support. I would point out that 2 years ago in the budget bill, that was a partisan vote. I mean there is no way you could say that was a bipartisan approval of the budget 2 years ago. All the Democrats voted with the President and all the Republicans voted against him.

So I think there is a serious question as to bipartisan support for the

Mr. Samuels. But I would also say that in OBRA 1993

Chairman ROTH. There is no question in the beginning— but it is, of course, it was 2 years ago when there was this huge

Mr. Samuels. No, when Senator Glenn asked me— just to go back— $29 billion is what the program is estimated I think in 1996, of that $29 billion, $7 billion came from 1993 and $10 billion came from 1990. That is that 1993 was not this balloon that some sug- gest occurred. This has been a progression, over time, recognizing that this is a good program and we know it will be able to work.

Chairman Roth. The fact is now that we have something like — how many millions of families are now on this program?

Mr. Samuels. We estimate that there will be 21 million.

Chairman Roth. So that 21 million families are going to be re- ceiving EITC out of what, 117 million?

Mr. Samuels. Right, there are about 112 million.

Chairman Roth. There is no question that in the past it has been bipartisan but in the last case it was otherwise.

Well, I appreciate very much your coming here today, Les, and we, as we have in the past, look forward to working with you. As I said, I think a number of the steps taken are highly desirable. Unfortunately, I don't think they go far enough to correct the prob- lem. We look forward to working with you.

Mr. Samuels. Thank you, Mr. Chairman. We appreciate the sup- port you have given to us and we really want to work together on this compliance issue and other proposals that may be suggested.

Thank you, very much.

Chairman Roth. Thank you.

I would now like to introduce the first panel. We are very pleased today to have Deborah Walker from the AICPA; and Don Huston, who is the one who first made me aware of this very serious prob- lem, he is a Delaware taxpayer. George Yin, who is a professor of law from the University of Virginia; and Mel Gonzalez, who is a former IRS attorney and now defends tax fraud cases.

As I said, Don Huston is a constituent of mine who has brought my attention to the dangerous level of fraud and abuse in the Earned Income Tax Credit, and I want to thank him publicly for his efforts and for his willingness to come forward, not only today, but to provide us with information that underscored the impor- tance of the problem.

I appreciate your taking the time out, I know, this is a very busy season as we are into the tax season.

Mr. Yin, why don't we start with you first, and please proceed.

Let me say, I think we have asked everybody to try to keep their statements roughly to 5 minutes. Your full statement will be in- cluded as if read.

102

TESTIMONY OF GEORGE YIN,1 PROFESSOR OF LAW, UNIVERSITY OF VIRGINIA SCHOOL OF LAW

Mr. Yin. Thank you, Mr. Chairman, for inviting me to testify today. In the interest of saving time, I will just summarize my statement. If you have before you my one-page summary I think you should be able to follow along pretty easily.

In my testimony I am going to address the following question: How can Congress achieve the general goals of the EITC program in a more efficient manner? You have heard testimony yesterday and today on the extent of noncompliance in the program. If Con- gress can ensure that only those entitled to the benefits in fact re- ceive them, the goals of the program can be achieved at signifi- cantly reduced costs.

I offer you, therefore, seven ideas to meet that objective. First, Congress should continue to evaluate whether the EITC program ought to be provided through the tax system rather than as a di- rect Government expenditure. This relates to one of the last ques- tions you raised with Secretary Samuels.

Most analysts believe that although the tax system promotes higher participation in a welfare- type program like the EITC, the potential for noncompliance is also greater. In other words, the tax system transfers benefits pretty effectively, both to those entitled to the benefit and, unfortunately, to those not so entitled.

Indeed, in some research I recently completed with several oth- ers, we tentatively arrived at exactly that conclusion. Compared to programs like AFDC and Food Stamps, the EITC program has a higher participation rate but also a higher noncompliance rate. Thus, if Congress is willing to sacrifice some participation in ex- change for improved compliance, it ought to reconsider whether the tax system is the appropriate vehicle to deliver the benefit.

Now, I realize that the notion of shifting to a direct expenditure program is probably the last thing on Congress' mind these days. But Congress should not be deceived into thinking that a program implemented through the tax system is somehow without adminis- trative costs. Recent events highlight all too clearly how the EITC program taxes the IRS' resources and potentially jeopardizes its much more important mission of collecting revenue in a fair and efficient manner.

Moreover, as you know, in an effort to control the Federal budg- et, as well as to reform programs providing benefits to low-income households, Congress is considering converting various entitlement programs to block grant programs to be administered by the States.

Yet, the EITC program, as currently designed, is an open-ended entitlement program administered by the Federal Government. Anyone meeting the conditions of the tax statute, and indeed, far too many who don't meet those conditions are entitled to the cash benefit.

It may well be that Congress wants to confer preferred status on the program and maintain its entitlement nature, but that judg- ment should be made affirmatively and not by default due to the fact that the program is a part of the tax system.

1 The prepared statement of Mr. Yin appears on page 236.

103

Congress should evaluate the program in the same manner as di- rect expenditure programs benefiting low-income households.

Second, if you wish to keep the program as part of the tax sys- tem, consider replacing much of it with a payroll tax exemption on a flat amount of wages. As you well know, the EITC originated, in part, as an effort to rebate to low-income workers the payroll taxes collected from them. Instead of collecting those payroll taxes and then trying to return those amounts to workers, in the form of an EITC payment, it would make much more sense simply to refrain from collecting the payroll taxes in the first place. Further compli- ance could be expected to be very high.

Now, to be sure, many in Congress might be fearful of tampering with the Social Security system. They might object to a proposal that appears to decouple the link between Social Security taxes and benefits. The reality, however, is that for low-income workers that link has already been decoupled. Such workers ostensibly pay pay- roll taxes and thereby, of course, become entitled to Social Security benefits even though the EITC payment completely reimburses them for their Social. Security contributions. They, in effect, pay no Social Security taxes, yet are entitled to receive Social Security benefits.

My proposal is simply to accomplish that exact same result in a direct fashion by not collecting the payroll taxes in the first in- stance.

Let me skip ahead to four additional suggestions to improve com- pliance without a major redesign of the program.

One is to reduce the size of the program and the amount of the benefit provided. Now, you have already heard testimony yesterday and today about the ability of taxpayers to overstate their income and, thereby, get a benefit that they would not be entitled to.

My point is basically that the IRS is not really able, and the tax laws are not designed, to identify those overstatements of income.

My second suggestion is to change the program so that the amount of the EITC can be more easily verified by the IRS. My idea here is to base the amount of the benefit on earnings reported on a W-2 form and, further, to delay any EITC until the IRS, in fact, receives a matching employer copy of the W-2. That would preclude not only basing any EITC on income that does not exist, but also on a fabricated W-2, which also appears to be a problem.

Third, I suggest

Chairman ROTH. Can I ask you a question? So, you are saying there is this program now where they pay in advance.

Mr. Yin. Yes, there is a program now where they get the credit in advance.

Chairman Roth. But you would recommend that — you don't think that is helpful?

Mr. Yin. I actually think, based on the GAO report of 1992, that that is a program even more susceptible to fraud than the lump sum payment. I think there are good reasons for the advanced pay- ment but if noncompliance is the concern I would not push that program.

My third suggestion is to require that a taxpayer reside with a qualifying child for the entire taxable year, rather than just half the taxable year, as is the case under current law.

104

There have been a number of instances, apparently, where either the wrong taxpayer or two taxpayers claim a particular child and where some taxpayers even claim fictitious children. My proposal doesn't do anything to curb the claiming of fictitious children but it would provide a more easily verifiable test for the IRS and would clear up some of the confusion in situations where a child lives in several different homes in the course of a year.

My final suggestion is to consider barring the claiming of the credit on electronically filed tax returns. Just as a burglar is going to burglarize the most vulnerable house on the street, those of a criminal mind are constantly casing the soft spots in our Federal tax and transfer system.

Surely, one of the most vulnerable points is the receipt of EITC money through an electronically filed return. The speed with which a fraudulent claim can be processed and awarded makes that transaction a particularly appealing one to criminals. I understand that the IRS, of course, has taken important steps during this fil- ing season to curb fraud in electronic filing, and Congress should probably await an evaluation of those efforts before barring an elec- tronic EITC claim altogether. But there is no strong policy reason to permit electronic EITC claims, particularly if doing so makes the system more susceptible to fraud.

I recognize the IRS and various businesses would probably all ob- ject to this suggestion. I think the short answer to those in the pri- vate sector is that they are not the intended beneficiaries of the EITC program or of electronic filing. As to the IRS, I think the bur- den is on them to establish sufficient control over the problem to justify continued availability of electronic filing in this area.

Mr. Chairman, that concludes my statement. I would be happy to entertain any questions.

Chairman ROTH. Let me ask you a quick question and we will go onto the others. By electronically filing, it does save a step, as I understand it, for the IRS is that if you file by paper then some- body has to put the information on the paper into the data bank. So it requires that additional step, but you don't think that is suffi- ciently important enough, or brings about enough savings to war- rant it?

Mr. Yin. I don't think if you are talking about a $37 billion prob- lem over 5 years, I would say, I want to have a better indication of what is the administrative savings on that step, as you describe it, from the IRS' standpoint.

Chairman ROTH. Thank you.

Mr. Huston.

TESTIMONY OF DONALD R. HUSTON,1 TAX PREPARER, DOVER, DELAWARE

Mr. HUSTON. Senator Roth, thank you for having me here and for allowing me to testify to provide practical information regarding the issues and problems tax preparers face because of the EITC. I have been preparing personal income tax returns for 13 years. The opportunity for fraud and misinterpretation involving the EITC has always existed. With the advent of electronic filing, the

1 The prepared statement of Mr. Huston appears on page 248.

105

rapid refund or refund anticipation loan, a new opportunity for fraud was available.

Single men were claiming their girlfriend's children, as was the girlfriend. Married couples were splitting to file independently as head of household with each claiming a child. Grandparents were claiming grandchildren because the parents were on welfare. Resi- dent and non-resident aliens were coming on the scene as part of organized groups.

The State Department of Health and Social Services, that is, each individual State, was coaching their welfare recipients how to get as much EITC as possible by telling the preparer the couple lived apart so that both husband and wife could claim the EITC. Non-existent dependents were appearing. Boyfriends would come to have their tax return prepared with a list of Social Security num- bers and names he could not pronounce, much less spell.

As a tax preparer, I do not function as an auditor or a policeman. I have an obligation to report obvious cases of attempted fraud and I must refuse service to anyone who would attempt or insist upon circumventing the law. I am also not supposed to decide whether an individual can support him or herself on such meager income and claim two dependent children without outside assistance.

During the 1993 tax season, for the tax year 1992, I estimated that about 85 percent of the returns I prepared involving EITC in- volved some manner of fraud. Tax year 1992 appeared to bring ev- eryone out of the woodwork because it was fast money and the word had spread from the previous year — the IRS was ill-prepared for this situation.

The IRS was, in fact, helping to perpetuate the problem by giving EITC even though it was not deserved. The IRS was not checking the returns to see if people were using the same addresses, same children, or even if the Social Security number for other than the primary taxpayer was legitimate.

Most of the returns involving EITC refunds were processed through electronic filing and the RAL. The filers did not care about the cost of this type of filing because it wasn't their money.

Even the people the IRS caught because they owed back taxes, back child support or federally-sponsored loans, received checks for the remainder of their refund directly from the IRS. The IRS was helping to perpetuate the fraud and preparers had to seek out the perpetrators to ask them to pay their preparation and processing fees. Some paid, some did not.

When President Clinton stated he was increasing the EITC I knew something had to be done. It was at this time that I con- tacted your office, in Dover, to explain the situation and my esti- mate of the amount of fraud that is involved in returns involving the EITC. The first feedback I received was a letter from you ex- plaining your course of action and a copy of a letter you had ad- dressed to the Treasury Department.

I also received additional letters from you and copies of responses from the Treasury Department of further investigations revealing more realistic estimates of the amount of fraud.

The feedback was very encouraging. The fraud, however, contin- ued in tax season 1994. Rapid refunds continued without abate- ment, however, the tax preparers in my office were trying to stem

106

the fraud as much as possible without placing themselves in an ad- versarial role.

The news media were also exposing the groups who were exploit- ing the EITC and the IRS' inability to curtail the fraud. This type of activity was also present in Delaware.

At the beginning of this tax season, there was a glimmer of hope as procedures and rules were tightened. More stringent procedures and checklists evolved to help tax preparers weed out fraudulent filers. Only one financial institution would be involved in RAL's in our area, mainly because the risk was too great.

In the past, the cost of electronic filing was $35 and with a bank fee of another $30. In 1995, the bank fee would vary from $35 to $95, depending upon the size of the refund. The biggest deterrent was the activity of the IRS and its new-found ability to cross-check Social Security numbers with names and dates of issue.

The withholding of EITC funds from the financial institution was probably the real clincher. The word spread fast and rapid refunds appeared to come to a screeching halt. I, for one, am thankful for the procedures the IRS has instituted this year. They were long overdue. We preparers need better guidelines for dealing with EITC filers, as whole.

I do not relish telling a filer he or she cannot claim their children as dependents because either the Social Security Administration or the Department of Health and Social Services is providing more than they are for their children.

I believe that if the Department of Health and Social Services had to report to the IRS the amounts provided to each welfare re- cipient it would also help stop any further fraud and provide an in- centive for people to get off the welfare rolls.

I also believe that minor children, recipients of Social Security, should be included in the EITC equation, and overall equation of dependency. The IRS, through your efforts and your staff, have made a significant difference this year in the abatement of fraudu- lent tax returns involving the EITC. Electronic filing is estimated to be down about 18 percent this year, as well as the tax prepara- tion business.

Chairman Roth. What was that figure again, down about what?

Mr. HUSTON. Down about 18 percent. There will be pressure brought to restore EITC funding for financial institutions involved in RAL's. We have an obligation, in some manner, to teach finan- cial responsibility and that cannot be done if filers are allowed to squander money in so frivolous a manner. In other words, just doling it out through the rapid refund situation.

However, they could go to a loan shark, as they have done in the past, so the devil you have and the devil you don't have.

Thank you for allowing me to testify before this committee. There are numerous incidents and anecdotes involving all manner of filers for EITC and unfortunately time does not allow me to re- late all of these to you.

I would be happy to answer any questions at this time.

Chairman Roth. Again, thank you for your contribution making my office, myself, in particular, aware of the seriousness of the problem.

107

What additional steps, if any, would you recommend be taken in the program today and could be the States be of any help here?

Mr. Huston. I think the welfare departments of the States or the Department of Heath and Social Services could help if they would put out a statement similar to what the Social Security Ad- ministration puts out and relay that to the IRS. Therefore, you would eliminate or help eliminate the possibility of someone else using those children other than the person who was intended to use them. That happens all the time, unfortunately. They can have legitimate Social Security numbers, come in with legitimate Social Security cards, but that may not be the right person who is really qualified to receive that EITC.

Chairman ROTH. Could the program, in your judgment, be better coordinated with the welfare programs?

Mr. Huston. Yes, sir. I think that could be but that is up to the States. Are the individual States willing to get into this.

Chairman ROTH. Assuming, that there is talk about, for example, putting a block grant for welfare into the States. Would it make sense to try to coordinate this program as a part of that reform?

Mr. Huston. I don't know, sir, because of the fact that within my State you have a problem with some of the case workers who are encouraging fraud. They are encouraging splits of the family unit. They are not — completely opposite to what was testified earlier — holding the family unit together, they are encouraging a split here so that they can get as much EITC out of it as possible.

Chairman Roth. Thank you, Mr. Huston.

Ms. Walker, it is, indeed, a pleasure to have you here today. We appreciate your patience. Please proceed.

TESTIMONY OF DEBORAH WALKER,1 CHAIR, TAX EXECUTIVE COMMITTEE, AMERICAN INSTITUTE OF CERTIFIED PUBLIC ACCOUNTANTS

Ms. Walker. Thank you, Mr. Chairman. The AICPA appreciates the opportunity to testify today on the Earned Income Tax Credit for individual taxpayers. I am Deborah Walker, Chair of the Tax Executive Committee of the American Institute of Certified Public Accounts, a national professional organization of CPAs with more than 320,000 members.

Many of our members are tax practitioners who collectively pre- pare income tax returns for millions of Americans. The AICPA urges that simplification of the tax system be a legislative priority, in particular, the Earned Income Tax Credit is in critical need of simplification. It, in fact, is used as an example of what should not be or what is not a simple Tax Code.

We strongly urge the Committee and Congress to rewrite the Earned Income Tax Credits to be understandable and usable by the taxpayers for whom the benefit is intended, low-income wage earn- ers.

This group of taxpayers generally lacks the resources to hire peo- ple that can deal with complex tax laws and is often not able to pay for tax preparation assistance.

'The prepared statement of Ms. Walker appears on page 251.

108

Chairman ROTH. Let me ask you a question there, because my understanding is that 50 percent of the working poor's taxes are done by professionals, or tax preparers, anyway.

Ms. Walker. Well, I am not sure what the statistics are but I think that about 50 percent of our working poor can't deal with the tax system and, therefore, have to go to a paid preparer. I mean the pieces that are in place, the parts of the tax law that apply to the working poor should be able to be dealt with by the working poor, and perhaps paid preparers, but certainly you would hope that more than half of them could cope with the tax system them- selves, at least the part that applies to them.

Chairman Roth. We certainly agree here, in principle, no ques- tion about it.

Ms. Walker. The AICPA welcomes proposed changes to make the credit more effective and offers several suggestions. Let's go back to when the Earned Income Tax Credit was enacted in 1975. Really the policy goal was to provide relief to low-income taxpayers from the regressive effect of Social Security taxes and improving worker incentives within that group.

According to the IRS it affects 15 million taxpayers. Over the last few years, the number one individual tax return error discovered by the IRS during processing has been the Earned Income Tax Credit, including the failure of eligible taxpayers to claim the credit and use of the wrong income figures.

Chairman ROTH. But I want to make sure — because I think that is a very important point. You are saying that the fraud and error abuse in the Earned Income Tax Credit is the most serious prob- lem.

Ms. WALKER. Right. And I heard the question you asked Sec- retary Samuels. I can't imagine that there is a more serious prob- lem. But you know, of course, as he probably rightfully answered, they haven't done studies and they may not be aware of other prob- lem areas. It is hard for me, when I look at the statistics in the GAO report, to believe that perhaps with the exception of the un- derground economy and not reporting of income, that there is an- other area with such serious problems. The under reporting prob- lem is the only other thing that comes to mind that could be so problematic.

Chairman Roth. But as we said earlier, here we have the prob- lem of over-reporting.

Ms. Walker. That is right, because the tax rate was raised and when that happened I can remember people sat around and said, watch, we are going to have over-reporting of income in order to claim this credit.

Chairman Roth. Please proceed.

Ms. Walker. The frequent changes over the past 20 years con- tribute greatly. There has been changes 10 times. It is a nightmare of eligibility tests and just 3 days ago it was changed one more time awaiting President Clinton's signature on the tax bill to pro- vide a deduction for health insurance payments for self-employed people.

I mean just looking at that we see as the problem — the complex- ity that comes about. We have got an Earned Income Tax Credit; and now we have defined that there is an abuse — people claiming

109-

the credit who have that there is interest and dividends. And what is happening is, we say, oh, well that is not right. If these people can save money and earn interest and dividends, then perhaps they are not so entitled to an earned income tax credit, so we make a little fix. Well, that little fix adds another worksheet, another complexity. And, in my mind, the Earned Income Tax Credit has gotten to the point where we should sit back and look at it, and stop with the little fixes.

And unfortunately, this example of H.R. 831, was really driven, I believe, by revenue-raising considerations. We needed to find some revenue, and here is a little fix. We will take care of this abuse and what we have got is this layering that makes the law too complicated for people to deal with.

My testimony outlines that the credit currently requires the tax- payers to consider nine eligibility requirements: the number of qualifying children; taking into account relationships, residency and age tests; taxpayers earned income, taxable and non-taxable; taxpayers adjusted gross income; thresholds, phase-outs, and vary- ing credit rates. The taxpayer has to complete a checklist, a work- sheet, another worksheet, a schedule with six lines and two col- umns, and then file Form 1040, rather than 1040-EZ. There are seven pages of instructions, 39 pages in an IRS publication describ- ing this credit.

The credit is determined by multiplying a credit rate by earned income up to a threshold; it is reduced by a phase-out rate. And when we look at all of this, if we cut right through it and say what would we recommend to change, we come down to definitions. I mean unless we are going to start with some major policy consider- ations, should this really come through the tax system and is there a better way to compensate for the regressivity of Social Security taxes, then let's work with Section 32 and concentration what would we do?

It would probably be a complete rewrite but I can tell you it would generally simplify definitions in the calculations. We would use definitions that already exist within the Tax Code as opposed to adding back things like non-taxable income and then we have to define what that is, and it is hard for the IRS to verify that.

Chairman Roth. If I understand what you are saying, if we are going to keep this program we ought to start fresh again, and re- write it, is that correct?

Ms. Walker. Yes. Take it back to the Senate Finance Committee and let's see if we can get this simpler? And my guess is that it can be drafted in a much simpler form, cost the same amount of money, and it will eliminate the confusion which I think — and I agree there is probably abuse from non-law abiding citizens — but I think that there is also unintentional abuse from people that just simply can't cope with the rules.

Chairman ROTH. Well, what about Mr. Yin's proposal that you exempt the first $10,000 in payroll taxes?

Ms. Walker. Yes. And as we go through we have some specific legislative recommendations which deal with Section 32. We then, in our testimony, mention other reforms to the system one of our suggestions is the same as his, exempt the first certain amount of taxable earned income from Social Security tax, which is what Mr.

110

Yin's proposal was. Yes, we support that. We have a similar pro- posal.

Chairman Roth. Please proceed.

Ms. Walker. The written testimony outlines these definitional changes. I will, in the interest of time, just mention them. The earned income definition, the qualifying child definition, we ques- tion whether the differentiation between one child and two children is enough of a differentiation to make any difference. And do we want to differentiate between people that have one child or two children and no children?

Regarding the definition of a child within the Tax Code — we have a dependent definition and an eligible child definition. We could equate all of those things and in many cases get to the same place where we are right now, which is providing through the tax law in- centives to people to go to work through the tax law.

The other big piece we would look at is to combine and extend the denial provisions. If you are in the AMT, you cannot claim the credit. If you have got foreign-earned income, you can't claim the credit. And that denial goes to the interest and dividend change that recently came in.

Again, we have this layering effect. There should be some broad categories that if this category applies, the credit is not available to you.

And finally do we really need an EITC table or can we just pro- vide a percentage rate that applies to taxable income or in certain cases, adjust the tax rates that are already there?

Another reform is to lower the refundable credit and we will eliminate the problem of over-stating income by equating it to the Social Security tax rates that are paid which is what the Earned Income Tax Credit was originally designed for. And then there is another reform — exempting the first certain amount of taxable earned income.

In closing let me just say, we certainly support measures to eliminate the complexity of this credit. We would welcome the op- portunity to assist in not having the Earned Income Tax Credit as an example of complexity to use in a Code that is far too complex. The AICPA would be glad to work with you in offering any other suggestions.

I would be glad to take your questions.

Chairman Roth. Thank you, very much, Ms. Walker.

I will now turn to Mr. Gonzalez.

Welcome and glad to have you here.

TESTIMONY OF ISMAEL GONZALEZ,1 ESQUIRE, (FORMER IRS ATTORNEY WHO NOW REPRESENTS TAX FRAUD DEFENDANTS)

Mr. GONZALEZ. Good afternoon, Mr. Chairman and Members of the Committee.

I am writing to you as a private practitioner specializing in tax fraud defense in New York City and as a former attorney with the Internal Revenue Service. But I am also speaking to you as a con- cerned citizen living in the greatest country in the world. Based on my past experience as a Special Assistant United States Attorney

JThe prepared statement of Mr. Gonzalez appears on page 263.

Ill

for the IRS prosecuting fraud cases and my current defense work I am of the firm belief that Congress should abolish the Earned In- come Tax Credit in its entirety.

I find this in my experience as a Government attorney and now, as a defense attorney that this whole EITC program is a vehicle for fraud, waste and abuse. In Hispanic areas of New York City, particularly in our Puerto Rican and Dominican neighborhoods, there is a great misconception of the purpose and use of the Earned Income Tax Credit program. Unfortunately a small minority of these people have seen earned income credit as an opportunity to rip off our Federal Treasury.

In 1992 and in 1993 an estimated $13.6 million in false claims for the Earned Income Tax Credit were filed by women on welfare with Spanish surnames in the New York City area.

These figures came out of the IRS Service Center in Holtsville, New York. A great majority of these women lived in upper Manhat- tan and in the Bronx. Being that a majority of my clients are His- panic, I am only aware of the effect of Earned Income Tax Credit in this particular community. I am by no means stating that His- panics are the sole cause of Earned Income Tax Credit fraud or they are to blame for the fraud in the system.

My views are that most Hispanics are hard-working and honest taxpayers. I represented two women who were charged with pre- paring false earned income credit schedules for welfare mothers, Anna Ortiz, and Miriam Peralta, which were the subject of news- paper articles in the New York Times and the Daily News, printed on February 17, 1994.1

Enclosed copies of the articles from the Times and Daily News paints my clients as somehow conning welfare mothers into believ- ing that they would get refunds of up to $1,400 if they filed earned income credit returns and pay my clients $25 to prepare these ficti- tious returns.

Both of my clients told every welfare mother that there was no guarantee that they would receive a refund through the Earned In- come Tax Credit because they were collecting public assistance.

Nevertheless, these mothers would take the chance and ask my clients, please prepare the returns. If I have a chance of getting $1,400 let me take the chance.

Many of these mothers would come to my clients saying that they heard about a million dollar "trust fund for poor people" that the Federal Government was administering that they could apply for. The fraud was so widespread in 1993, that one of my clients told myself and the U.S. Attorney's Office for the Southern District of New York, that she knew of one woman preparing so many false Earned Income Tax Credit returns that she had her business oper- ating out of 10 individual apartments in Manhattan.

Of course, what I think would facilitate this Earned Income Tax Credit feeding frenzy was the availability of electronic filing and rapid refunds. Other clients told me of situations where individuals without children would buy legitimate Social Security numbers from non-filing individuals with children so they could claim the Earned Income Tax Credit. Once they would receive their refund

1The articles appears on page 266-267.

112

checks, they would split the proceeds of the tax refund with the person who gave them the legitimate Social Security number, so there is a commission.

I understand that this credit goes up to $3,100 next year. In my opinion the Internal Revenue Service is not doing enough to spread the word in the inner-city Hispanic communities about the criminal ramifications of filing false returns and specifically how the tough and rigid the United States guidelines can act as a bludgeon on their heads.

By spreading the word, I mean by having Spanish-speaking em- ployees of the IRS serve as speakers in churches, town meetings, and at community functions. The only time members of the His- panic community ever see, hear or read about the IRS in their neighborhoods is when the IRS is involved in large currency sei- zures, or narcotics or money-laundering related enforcement in con- junction with the United States Drug Enforcement Administration.

In my opinion this is how they conceptualize the IRS. Service employees must spread the word in an Hispanic community that filing false tax returns, and in particular, filing a false claim for the Earned Income Tax Credit could land the offender in jail. If this is not done through community group meetings, use of the mass media, such as Spanish radio and newspapers, the Congress should altogether abolish the Earned Income Tax Credit.

I have given talks about Earned Income Tax Credit fraud on the Spanish radio station in New York City and have received numer- ous positive telephonic inquiries. However, I find that the Spanish press, the printed press in New York, is not very interested in this subject matter, as I have submitted two large articles in Spanish to one well-known Spanish newspaper for free publication but never received a response from the editors.

I believe that it would cost more tax dollars in stopping the fraud than it would in fulfilling the intended purpose of Congress, there- fore, I think it should be eliminated.

I want to thank you, Senator Roth and Members of your Commit- tee for giving me this opportunity to talk and I am available for questions at this time.

Chairman ROTH. Thank you, Mr. Gonzalez. I appreciate your coming. I know you were originally scheduled to come yesterday but conflicts prevented that. And I am delighted that you are able to be here today.

What you are saying to me is that in your judgment this pro- gram cannot be saved? That it invites fraud on the part of very, you might say, innocent people in the sense that they don't under- stand the program, and as a result, a scam artist can take advan- tage of them, is that correct?

Mr. Gonzalez. To a certain extent, sir. I think that also the problem is that a lot of people are unaware of the criminal rami- fications. They don't think, they think, well, if I get caught I will have to give the money back. They don't understand that there is something called the United States Sentencing Guidelines that these matters are handled in a United States District Court in which Federal judges can put them in jail.

They, when they think of having problems with the law in these particular neighborhoods they are thinking about other people that

113

have been convicted for let's say, possession of narcotics and they see them go to jail and they are out the next day.

There is not enough of an understanding, there is no knowledge out there with respect to the big problems that someone could de- velop, the criminal problems, the problems of having to go to jail if you get caught cheating on these returns.

Sure there may be some, the return preparers are probably aware of the problem, but there is not enough information out there. There has been misinformation about what are the criminal ramifications — how someone can go to jail.

Let me give you an example. You have this publication, 596, and it says right on the front, the Government may owe you money, in big capital letters with an exclamation point.

People see that and they say I understand this publication is written in Spanish. Let me see how I can get some money from the Government. But there is nothing in this pamphlet talking about how you could end up going to jail for 3 years if you are convicted for fraud on this particular schedule. If word were to get around that you can go to jail for putting in a fraudulent or an Earned In- come Tax Credit return, then people are going to think twice before they file this schedule.

But no one, in my opinion, no one seems to know about that part.

Chairman ROTH. So that is point No. 1, there needs to be better information out among the working poor.

Mr. Gonzalez. Absolutely.

Chairman Roth. And, if I understand your testimony, even if that were the case, it still invites fraud and the program is not sal- vageable. That if we want to do something to help the working poor, to get them into meaningful jobs, we really need to take an- other look, a new approach, is that what you are saying?

Mr. Gonzalez. Absolutely.

Chairman ROTH. Anybody else care to comment? Mr. Yin.

Mr. YlN. Well, I don't agree with that. I think that whether the program should be continued or not should be based on factors other than the issues that we are talking about today, that is the noncompliance issues. It seems to me you need to examine the in- centives and the disincentives that are created and if you feel, on balance, that there are not good incentives being created by the program, then that would be a reason to get rid of the program. But if you feel that the program is basically accomplishing a good end, it seems to me I would not get rid of the program. I would reform it in a very major way and I have laid out several sugges- tions that would modify the program quite considerably which I think would cure, for the most part, the compliance issues that you are concerned about.

Chairman Roth. Ms. Walker.

Ms. Walker. Let me make just one mention because there is a lot of talk about electronic filing and, of course, we realize that that can increase the abuse and the fraud that goes on. But I think we have to be very careful before we say that electronic filing is bad or inappropriate and maybe nobody is saying that. I hope not.

Because, in my mind, the IRS, in order to do its job efficiently, is going to have to, in the end, in the end being by the year 2000

114

or so, require electronic filing and make it very easy for taxpayers. We are moving into an era when we don't have to file by paper.

Chairman Roth. Well, theoretically isn't it possible, by collecting data banks and so forth and through electronic filing, you may be able to do a better job of uncovering fraud, although it raises other questions of privacy and so forth.

Ms. Walker. That is right. Assuming that the credit is based on verifiable information.

Chairman Roth. Mr. Yin.

Mr. Yin. Just to respond to Ms. Walker. I don't disagree with the fact that I think the way of the future is electronic filing. What I am simply saying is I would not lead with Earned Income Tax Credit claims in electronic filing, because that is a very vulnerable aspect of this way of the future. I would pursue electronic filing and encourage it in many other aspects, perhaps all other aspects of the tax system and let the EITC aspect of it be the end part, where once we have got the mechanisms in place where we can as- sure that the fraud and abuse and so forth won't take place, then I would allow EITC claims to be processed through electronic means.

So I am not disagreeing with the basic point about electronic fil- ing but I am just simply saying the order.

Chairman Roth. Yes. It is my understanding you are a member of the Treasury Fraud Task Force.

Mr. Yin. I am and, of course, I am not testifying on their behalf today. I am testifying on my own behalf.

Chairman ROTH. I understand. Mr. Huston.

Mr. HUSTON. Yes, sir. As far as electronic filing goes jt is really a salvation to the lower-income person who cannot afford enough money, up-front, to pay for a return done properly.

If you put these people back out on the street in a cottage indus- try you are liable to end up with anything. At least if they are going into the tax preparation offices throughout the country you know there is a half-way chance it is going to be correct, better than half-way chance it is going to be correct. And then the money is withheld from their refund check, the preparation fees and elec- tronic filing fees are withheld from that particular check.

What happens, however, if the IRS detects something wrong with the return is that they stop the rapid refund and they send the per- petrator a check directly. That is the problem there. It is with the IRS not with the preparers or the individuals. You can't put every- body in jail and there are too many poor people out there, you would be overflowing.

Chairman Roth. Time is passing and I want to thank each and every one of you for being here today. I think your testimony is ex- tremely helpful and we may have additional questions in the next few days and if we do we will submit them in writing.

Thank you ladies and gentlemen very much for being here today.

It is now my pleasure to introduce our next and final panel, a very distinguished group. They include Dr. Marvin Rosters from the American Enterprise Institute; Karl Scholz from the University of Wisconsin and Robert Greenstein from the Center on Budget and Policy Priority.

115

Gentlemen, it is a pleasure to welcome you here. Your full state- ments will be included as if read, and I would ask that each of you restrict your opening remarks to 5 minutes. Thank you very much.

And Dr. Kosters, why don't we start with you?

TESTIMONY OF MARVIN H. KOSTERS,1 RESIDENT SCHOLAR, AMERICAN ENTERPRISE INSTITUTE

Mr. Kosters. Thank you, Mr. Chairman. I am pleased to have this opportunity to discuss possible ways of improving the earned income tax credit. It has certainly become a major item in our in- come transfer programs for families with children. And there are problems there that deserve some attention.

There are three problem areas that I discuss in my prepared statement: How the EITC affects work incentives, how the mar- riage penalty arises and compliance issues. What I would like to do in my summary is to focus mainly on the work incentives ques- tion and a little bit on compliance.

The general point that I want to make is that problems in all three of these areas could be reduced a great deal if the EITC were smaller and less ambitious as an income transfer program. Now, much of the attraction of the credit comes from the constructive idea that it helps to make work pay and, therefore, that it provides a way of redistributing income without discouraging work and for some people, actually encouraging work.

There are questions about how well it achieves this, however. The first question, I think, is how are work incentives really af- fected by the current program? And the second is, could the overall impact be improved by changing its design somewhat? With regard to the first question, I think it is important to ask how families in different circumstances are affected. And what I have done is to look at families with children and consider that question.

And the first point to make, I think, is that some 60 percent of families with children are above the income cutoff and are essen- tially unaffected, except that many of them would pay more taxes, of course, to support the EITC. There are another 30 percent or so in the income range that would receive the earned income tax cred- it. They are eligible because they are in the right income range and someone is working or has earnings. That is a total, then, of some 90 percent. And I will come back to this 30 percent in a moment.

There are another 10 percent that are in the low income range, but have no one working. Now, this is the group presumably where it is hoped that enough would be encouraged to consider choosing work instead of not working in order for the EITC to have a net positive and favorable impact. The positive impact on this 10 per- cent, of course, needs to be weighed against the work incentives that operate on the 30 percent that are eligible for the credit and that have someone working.

Now, there are two ways in which the people in the income range with someone working are affected. One is that the credit availabil- ity means that there is somewhat less pressure for them to work. It reduces the pressure to work. And the other is that there is a tax rate, an implicit tax rate, where the EITC is phased out.

1 The prepared statement of Mr. Kosters appears on page 268.

116

That is to say, people — about two-thirds of this 30 percent — expe- rience a tax rate of about 21 percent if they have two children when the income tax credit is fully phased in next year. Now, I think most analysts would agree that the 30 percent that I referred to tend to be discouraged from work. They would tend to work less.

So the question that comes up is, what about those 10 percent? Now, Karl Scholz has looked at this very important and difficult issue to analyze. And he concludes that maybe, on balance, work might increase. But there are uncertainties about that. In any case, I think it is fair to say that the extent to which we should expect more hours of work by people on account of the EITC is really somewhat in question. It is likely to be relatively small. And we need to recognize that many people would be discouraged from work.

That brings us to the second question, of course: Could it be rede- signed in a way so that it would have more favorable incentives? I believe that it would have a better impact on work if it were less generous, if the income cutoff were lower, if there was a smaller maximum credit, and if it were phased out more quickly.

Under those circumstances, many fewer people would be discour- aged from work, and we would still have the 10 percent who would be encouraged, of course, potentially. Granted, we need to recognize that the work incentives effects would be weakened somewhat and that transfer effects would be smaller.

But I think we need to recognize that income cutoffs that are contemplated for next year that are now in place are quite high. For example, next year, for a family with two dependents, it is about $28,500. And if we compare that with median earnings — I have 1992 figures here — year-round, full-time workers across the country have a median income of $26,000. This is just people who work year-round and full-time. For all workers

Chairman ROTH. That is interesting, because when I asked the assistant secretary, I thought we got a figure of $50,000.

Mr. ROSTERS. Well, I think that figure refers to family income for a family of four. What I am talking about is the median earnings of individual workers. Many high income families have more than one worker. In fact, a large share of them do. That is the difference there. In fact, if you take all workers, not just those that work year-round, full-time, the median in 1992 was $18,400. And if you take adult workers, you will find that — and here I am talking about workers 25 years old and older — 58 percent of them earn less than $25,000 in 1992.

That raises questions about whether we should regard the cutoff levels where the EITC is phased out as appropriately describing the working poor. Now, there are other approaches one could use for a variety of purposes. One could use other mechanisms to re- duce the high marginal tax rates that people now face. But the most straightforward approach seems to me to be reducing the size of the EITC.

Now, let me skip the marriage penalty. A good deal has already been said about that, and I only want to say a couple of things about the compliance issue. It seems to me that it is an important issue because this is a difficult area to monitor and yet it needs to be monitored very carefully to avoid discrediting the IRS or the

117

whole EITC program. Of course, it is a function of the service to collect tax revenues in the first instance. And it is very important to maintain the integrity of the agency in that function.

It seems to me that when we think of expanding IRS compliance activities, we need to recognize that we already have programs pro- viding income transfers with large staffs who are involved in audit- ing and compliance. So we ought to ask ourselves whether it really makes sense to have a parallel and separate compliance activity, or whether instead somehow these income transfers ought to be ad- ministered by a common agency rather than each one having a sep- arate compliance program.

In summary, I think that there may be a role for the EITC. But I do not think that bigger is necessarily better. I think it has grown too large and that shrinking it would help the compliance problem, help the marriage penalty and help the work incentives issue, as well. By arguing that the EITC should be smaller, I do not mean to argue here that we should have fewer income transfers to lower income families.

But I think if we were to maintain the level of income transfers that we now have, it would be preferable to administer those through other agencies to a greater extent than we now do. I think we are placing too heavy a burden on the EITC with too little in the way of likely results in terms of work incentives, and that it would be better to reform these other transfer programs in ways that would help to encourage work, and to require work where that is appropriate, instead of maintaining the EITC at its current size.

Thank you, Mr. Chairman.

Chairman Roth. Dr. Rosters mentioned you by name, Mr. Scholz, so I will call on you next and then we will finish, if we may, with you, Mr. Greenstein.

TESTIMONY OF JOHN KARL SCHOLZ,1 ECONOMICS DEPART- MENT, LA FOLLETE INSTITUTE OF PUBLIC AFFAIRS AND THE INSTITUTE FOR RESEARCH ON POVERTY, UNP/ERSITY OF WISCONSIN-MADISON

Mr. SCHOLZ. Mr. Chairman, thank you for inviting me to testify on the effectiveness and the design of the earned income tax credit. The EITC plays a central role in policy as it affects the working poor. Over the past 20 years, there has been a striking change in the bottom of the Nation's earnings distribution, beginning in the 1970's, but accelerating over the 1980's.

In 1973, the median male without a high school diploma earned over $24,000 in 1989 dollars. By 1989, the median worker with the same level of education earned less than $15,000. The trend is nearly as dramatic for males with only a high school degree. The erosion of labor market opportunities for people with low levels of education has placed enormous strain on the Nation's anti-poverty programs.

Against this backdrop, the EITC has provided an important sup- plement to the earnings of low skilled workers. A family receives the EITC by filling out a tax return. But many low income families are not legally required to file. A married couple with two children,

The prepared statement of Mr. Scholz appears on page 277.

118

for example, is required to file a tax return in 1994 only if they had incomes above $16,000, though with an income considerably below that, the couple would be entitled to an EITC of over $3,000.

If the EITC is to be successful at meeting the objective of making work pay, families who are eligible for the credit should receive it. The best available evidence suggests that this indeed happens to a large extent. The EITC participation rate is around 80 to 86 per- cent. The high participation or take up rate of the credit is striking when compared to the AFDC participation rate of, say, 60 to 70 percent or the food stamp participation rate of 50 to 60 percent.

The United Kingdom has an EITC-like program called the family credit. It is administered through the transfer system and directed toward families with children. Official estimates place the partici- pation rate of the family credit at around 50 percent. Thus, both compared to other in-kind transfers in the U.S. and comparable work related benefits in the United Kingdom, the EITC gets high marks for reaching those who are eligible for the benefit.

The EITC has different labor supply incentives, as Marv men- tioned, depending on the taxpayer's income relative to the subsidy, the flat and the phase-out range of the credit. The subsidy or the phase-in range of the credit provides mixed incentives to work. The flat and claw back ranges provide an unambiguous incentive to re- duce hours of work.

Most EITC participants are in the flat or phase-out range of the credit, again as Marv mentioned. That raises the concern that the EITC may lead to a net reduction in the labor supplied by low-in- come households. In fact, the credit in aggregate is likely to reduce hours worked by workers. However, the economic significance of these effects is small. The estimates that I think are most reliable suggest that the average reduction in hours of work will be around 11 per year. This average reflects the mix of increasing hours by about 38 hours per year in the phase-in range and a reduction of 3 and 11 hours per year in the flat and claw back ranges of the credit, respectively.

Moreover, to the extent that taxpayers are unaware of the effect of the credit on after tax wages, both the positive effects of the credit in the subsidy range and the negative effects of the credit on taxpayers in the phase-out range is overstated. Thus, the best available empirical evidence suggests that the EITC has a small, but detrimental effect on the hours of workers.

The credit has an unambiguously positive labor market effect on the decision of whether or not to work. Moreover, these effects can be important, as there is widespread agreement among economists that the strongest empirical labor market effects of wages and hence the EITC is on participation rather than hours of work.

The empirical work on this topic shows that net wages positively and significantly affect labor market participation and, as an added bonus, negatively affect transfer program participation, particularly for single parents. This increase in the participation caused by the EITC is likely to offset or more than offset the reduction in hours among those who work.

So in aggregate, the evidence suggests that the EITC probably will increase hours worked by low income households, but the ag- gregate positive effect is likely to be small. At the same time, the

119

modest positive labor market effects of the EITC should be con- trasted to the detrimental effects on both labor market participa- tion and hours of work associated with other income transfer pro- grams.

The future viability of the EITC depends on the IRS being able to reduce non-compliance. Doing so requires two distinct tasks. First, programs must be put in place to detect accurately and re- port the degree of non-compliance and its causes.

Current policy is being debated in an environment where solid evidence on the magnitude of the compliance problems is scanty. Without credible numbers on EITC non-compliance and evidence about the source of non-compliance, it is difficult to design policy to address the problem.

The Internal Revenue Service has, perhaps belatedly, taken the second step needed to address non-compliance. This involves giving greater scrutiny to verifiable items on tax returns. The IRS created an uproar this year due to refund delays as they give Social Secu- rity numbers greater scrutiny.

Nevertheless, this and matching employee and employer W-2 re- ports seems to be the best way of combating the non-compliance that jeopardizes the program. As long as the credit is based on the items that the IRS is, in principle, able to verify, there is nothing inherent in the credit that would lead to unusually high levels of non-compliance.

Moreover, other areas of the tax code — for example, Schedule C — also have large amounts of non-compliance associated with them. As a matter of sound policy, the costs of reducing non-compliance in every aspect of the tax code should be compared to its benefits so that the IRS uses taxpayer resources in as efficient a way as possible.

In my invitation to testify, the Committee raised a number of proposals that could possibly enhance the targeting of the credit. I have two brief comments that I would be happy to discuss in greater detail, if asked.

First, the EITC is well targeted to low wage workers. I did some calculations for the hearing and found that more than 75 percent of EITC benefits go to taxpayers with wages that would place them in the bottom quarter of the wage distribution of all workers with children. That is, 75 percent of EITC benefits go to workers with wages below $7.30 an hour. More than 95 percent of all EITC bene- fits are paid to workers with wages below the median of $11.11 an hour.

Second, the administration has introduced a proposal to deny the Earned Income Tax Credit to taxpayers with assets, based on divi- dend and interest income above some threshold. The proposal would be relatively straightforward to administer and consequently could, in an effective manner, enhance the targeting of the EITC. I would not oppose such a proposal though there are two things that should be kept in mind when debating it.

First, it would be relatively easy to manipulate portfolios so as to avoid asset income thresholds without altering the value of the portfolio. Hence, the asset test will, in all likelihood, exclude fewer families than expected. Second, it may be counterproductive to im- plement a very restrictive asset test. While asset tests of any kind

120

enhance targeting in a static sense they also can make it nearly impossible for recipients to legally accumulate the assets necessary to take a wide-range of choices leading to greater independence, such as helping a child go to college, getting additional training or moving away from a dangerous neighborhood.

Just as with high tax rates on labor earnings, asset tests can dis- tort the economic decisions of low-income households and hence, may reduce the efficiency of anti-poverty programs.

To conclude the erosion at the bottom of the country's labor mar- ket imposes a pressing policy problem. The EITC is a sensible, well-targeted policy to address this problem. The credit is threat- ened, however, by noncompliance. I now sense, largely through the efforts of you, Mr. Chairman, that the IRS is taking strong steps to attack this problem. Their efforts should be given time to work. The reason for this is that in other respects the credit is working well. A high fraction of eligible taxpayers receive the credit; its labor market effects are probably, in aggregate, beneficial, and without a doubt they are less pernicious and alternative ways of assisting the working poor; the credit is well targeted toward poor and near-poor families; and finally, families receiving the EITC are working, an action that, in my opinion, should be encouraged.

Thank you.

Chairman ROTH. Thank you.

Mr. Greenstein.

TESTIMONY OF ROBERT GREENSTEIN,1 EXECUTIVE DIRECTOR, CENTER ON BUDGET AND POLICY PRIORITIES

Mr. GREENSTEIN. Thank you, Mr. Chairman. I appreciate the op- portunity to be here today and I certainly share the view that the principal issue facing the Earned Income Tax Credit at this point are the error rates. And I would like to return to that in just a minute. It is an area that holds particular interest for me. Some years ago I was in charge of the Food Stamp program down at the Agriculture Department and, of course, one of our principal focuses was taking steps to bring down the error rate and we did have some significant accomplishments there.

Also in the work our center does with regard to the earned in- come credit we uncovered, in 1992, some abuse relating to the health insurance credit which we brought to the attention of the Committees of Congress and the IRS and perhaps had some role in, I think, the wise decision in 1993 to repeal the health insurance credit.

But before getting to that, I did want to talk a little bit about the growth in the EITC and some of the reasons for it. As Dr. Scholz just mentioned the growth in the EITC has come during a period when there has been a significant erosion in the wages paid for low-paid work, low-skilled work. From 1977 to 1993, the poverty rate for families with children, in which a family member works, grew by nearly half. Today, over 60 percent of all poor families with children contain a worker. The proportion of workers paid a wage too low, full-time, paid a wage too low to lift them to the pov- erty line is up significantly.

1 The prepared statement of Mr. Greenstein appears on page 289.

121

And this led, starting with President Reagan in 1986, to a bipar- tisan policy emphasis, certainly in 1986 and 1990, as you know until 1993 it was on the Democratic side, to increase remuneration from low-wage work with the EITC as the principal policy instru- ment. These three expansions, 1986, 1990, and 1993, signaled an increased reliance on the EITC and a sharply decreased reliance on the minimum wage.

The purchasing power of the minimum wage is now at its second lowest level since 1955. It also signaled policy decisions to rely much less on AFDC as a means of supplementing the wages of poor, single parents with children and to shift that to the EITC. In 1972, before the EITC was created, 49 States provided AFDC as a wage supplement, where a mother with two children whose earn- ings equaled three-fourths of the poverty line. Today, just three States do.

And during this period, of course, payroll taxes increased as well. These factors had a lot to do with the specific policy decisions that were made and virtually all of the growth in the EITC that you have talked about is due to the decisions made in those three laws, 1986, 1990 arid 1993.

Each of those laws was phased in over a number of years and we had significant growth rates during that period. I did look at the CBO data and find that after 1997 when the 1993 law if fully phased in, the EITC growth rate would drop to less than 4.5 per- cent per year with most of that due to inflation. So at least we don't have a Medicaid or Medicare type of situation where there is out of control growth forever. This is specifically because of policy decisions the Congress and three Presidents made and decisions made, in large part, to move away from minimum wage and AFDC and to the EITC.

I have a table here which kind of graphically shows this. And what the table shows that is quite interesting is that for a mother with two children who works 30 or 40 hours a week, year round at the minimum wage, that even when the 1993 EITC expansion is fully phased in her disposable income will be $2,000-$3,000 lower than it was in the early 1970's before the EITC was created.

This is because even with the big expansion of the EITC that ex- pansion less than offsets the combination of the erosion of the mini- mum wage and the withdrawal of AFDC from such families over the past two decades and the increases in Social Security payroll taxes.

Now, turning to the error rate issue, I guess I have a couple of thoughts here. There is no question that much more needs to be done to reduce these error rates. I always think it is mindful to look at what we have done that has worked or hasn't worked and where we have to go.

The 1990 changes that restructured eligibility for the credit and introduced scheduled EITC for the first time, I think were signifi- cant and a GAO report in 1993 noted that they addressed the sin- gle largest source of taxpayer errors in the EITC in the pre- 1990 period.

Now, have they had any effect? In the figures you cited yesterday and that are in the GAO report, they note that from January 1994 their sample of about 24 percent of the EITC benefits are overpay-

122

ments. I do think, here I go to my experience with Food Stamps, we have to be careful not to confuse the proportion of cases that are in error and the proportion of dollars paid out in error. And AFDC, Food Stamps and Medicaid we really focus on the dollar error rate not on the case error rate. You could have lots of cases that have $10 in error and yet don't lose much money. You could have a smaller number of cases where the family is totally ineli- gible and big bucks are involved.

You referred to sanctions, I think, in AFDC and Food Stamps earlier, those are entirely based on dollar error rates.

Chairman Roth. I think that was the case with GAO, too.

Mr. Greenstein. That is right.

Chairman ROTH. I mean, they were on this program.

Mr. Greenstein. Yes. We really want to look at the dollar error rate.

Chairman Roth. So, it is consistent, is what I am saying.

Mr. Greenstein. Yes. And my point simply is that the 1988 audit found 34 percent of the EITC overpaid and the latest GAO figures from 1994 are 24 percent. Now these are not definitive but they do suggest that the 1990 changes may have gotten rid of about one-third of the problem but a big problem still remains. What do we do about it?

Senator Glenn. That is number of cases, still, right?

Mr. Greenstein. No, this is dollars.

Senator Glenn. Those are dollars.

Mr. Greenstein. Yes, 34 percent of the dollars in there — 24 per- cent in January of 1994, and that is before IRS changes this year. Here is the key point I want to make here. In the Food Stamp pro- gram, where the error rates can go lower but they are half of what they were some number of years ago — the AFDC error rates are well below what they were 10 or 15 years ago as well. You talk to people in the field, they will tell you that the single biggest dif- ference is checking, is verification. When you go from a system where you award payments without checking the information first, to one where you scrutinize it first, you get sharp reductions in error. That is the experience.

In the Earned Income Tax Credit, prior to 1990 there was not even a schedule for EITC. You didn't even have to apply for the credit. If you looked eligible, from the face of the 1040, you were awarded it. Starting in 1991, there was a schedule but IRS didn't use the information on the schedule to check anything before pay- out. Starting January of 1995, for the first time, IRS does extensive checking before payout. They check all the Social Security num- bers.

As you know, if you fall into any one of this number of categories you get this additional questionnaire. We don't have any data here, though it is my belief, based on experience with things like Food Stamps and AFDC, that the procedures instituted ought to make a significant difference. There has got to be a big difference when you verify for work payout in a number of cases, it would be my belief and I hope, but also my belief, that the error rate is now probably significantly below that 24 percent figure although we don't know how much.

123

Two or three final comments and I will be finished. I just want to say I do not mean to say that a lot more doesn't need to be done, it does. I always think we need to look at is what we have done that is working or not working, what are the additional steps that are needed. I have a couple of suggestions in my testimony for ad- ditional things I think that IRS should do.

A couple of just final points and I will close. In the work incen- tive issues which both of the other panelists have discussed, I would simply note that in a 1990 panel of the American Enterprise Institute, Bob Reischauer, Director of CBO at the time, made the important point that when we talk about the work incentive as- pects of the EITC we need to be very careful to distinguish which types of families are affected.

The data clearly show that the group among whom work hours are reduced, where the work incentive factor is most significant, is the second parent in two-parent working families who make over $11,000 a year. Reischauer's point was simply that may not be all there. If the Earned Income Tax Credit allows a mother in a two- parent family, where both parents work, the ability to choose to work a little less and spend more time raising her child, that is not all bad. The positive part, about the controls that were mentioned, is that we want to get these welfare families, who are not working at all, to start working. That is where his research and others sug- gest a positive effect.

We have the same kind of mixed effect with the marriage pen- alty, no question, there are significant marriage penalties at cer- tain parts of the EITC spectrum, as both mine and Dr. Scholz testi- mony mentions. The point that sometimes gets missed is that the EITC is also a marriage bonus for some. The issue is simply the following. You have an AFDC mother who doesn't work at all. And there is a guy who makes $10 or $15,000 a year. Does she marry him or not. If she marries him, she loses her AFDC, her kids can lose their Medicaid. The effect, these are the well-known work pen- alties built into the welfare system that can be substantial — the EITC offsets that. Right now they don't get EITC, but if they marry they can get several thousand dollar EITC that offsets the loss of welfare and eases the work penalties in the welfare system.

So, again, marriage penalty is a marriage incentive at the bottom and a marriage penalty

Chairman Roth. If you can sum up, because the hour is getting late.

Mr. Greenstein. I have one final point.

That is that I know there is a proposal around to cease the in- dexing of the EITC. I would urge you to look very carefully at that. My particular concern is that because the EITC declines 21 cents with each additional hour of earnings above $11,000 a year, that if one doesn't index it then a family whose earnings simply rise with inflation will have its payroll tax go up and its EITC go down. And the result would be a significant tax increase. I think we esti- mate it here with a couple of sample families, and over a 5 or 10 year period you could have a $400-$700-a-year tax increases. These are families at $15-$20,000 a year that are working, that are not on welfare, and I presume no one really wants to increase

124

their taxes. I think there is a problem with that proposal and let me stop there.

Chairman Roth. Thank you, Mr. Greenstein.

Chairman ROTH. Let me ask you this question, what purpose does the EITC have for those without children? Should this part of the program be kept?

Mr. Kosters. I would actually favor keeping it, Mr. Chairman. It seems to me that there are different ways of looking at equity, and that there is a case for a small credit for people who are work- ing and have low-income, low wages, and who do not have children to support. It seems unlikely that this is going to be the cause of much in the way of compliance problems or other personal incen- tive problems because the credit is so small.

Mr. Greenstein. If I could add, I would agree with Dr. Kosters on this. When the issue was first raised in early 1993, I originally had some questions about it. Well, what turned me around on it was when I looked at the historical data on tax burdens and it showed the following.

Prior to the extension of the Earned Income Tax Credit to these families, the effective tax rate, the proportion of income paid in Federal Income and Payroll tax combined, for the poorest fifth, this is CBO data, the poorest fifth of non-elderly households without children was 15 percent, that is 15 percent of their income in Fed- eral taxes.

That was double the percentage for the poorest fifth of families with children and more than five times the percentage for the poor- est fifth of the elderly.

We also found that from 1980 to the present the average tax bur- den of the poorest fifth of non-elderly households without children rose 38 percent, the Federal tax burden rose 38 percent. The rea- son being primarily that when payroll taxes went up, the one group we never offset them on were the poorest fifth of families without children.

So I think, as you know, and Dr. Kosters said, this is a very small credit for the households without children and it phases out at $9,000 a year, and I tend to think that it should be retained.

Chairman Roth. Mr. Greenstein, I think it was you, that in your opening statement, said in part, we have a much bigger EITC that partly replaces or supplements the minimum wage. Are you saying that this is a substitute for a minimum wage, that we don't need that or

Mr. Greenstein. No. I guess I am saying two things here. What I am saying here and then I would like to comment is that in the debates, particularly in 1990, much of the debate among Members of Congress, the Bush administration and a number of Members of Congress — particularly any number of conservative Democratic Members of Congress as well, Congressman Stenholm I remember having a conversation on this, very explicit discussion — that it made more sense to put more emphasis on the Earned Income Tax Credit than on the minimum wage because of the belief, both be- cause the Earned Income Tax Credit is formally targeted and be- cause of the belief that the minimum wage may have some adverse employment effects.

125'

Now my own sense is that you need a balance between the Earned Income Tax Credit and the minimum wage and one of the reasons the Earned Income Tax Credit has been made as big as it has, is leading to some of the issues discussed in this hearing, is because the minimum wage has eroded as much as it has.

What we had said several years ago suggested a substantial in- crease in the Earned Income Tax Credit but not quite as big as was enacted in 1993, coupled with a more modest increment in the min- imum wage. But it does seem to me that that was just not headed towards restoring some of the lost ground in the minimum wage. And so long as Congress does not restore some of the lost ground, then I think the problem with cutting the Earned Income Tax Credit will be the adverse affects on the living standards of the working poor and their benefits relative to welfare.

So I guess where it leaves me is one could either scale the EITC back somewhat and do a modest increase in the minimum wage if one is not going to do that, then I would try to do everything that one could to reduce the error and fraud rates in the EITC but with- out reducing the Earned Income Tax Credit because I think mil- lions of honest working families, whose wages have eroded, some of whom no longer have health insurance, and who pay higher pay- roll taxes, they need it.

Chairman Roth. Now, the EITC will be phased out roughly at $28,000, I think next year, in 1996 which is fairly high. After all in this article by you, Dr. Kosters, you noted that the President's salary when he was Governor of Arkansas was only $35,000. And my question to you, is the proposed upper income limit too high?

Mr. KOSTERS. In my opinion, yes. And let me explain very briefly why. As you move the income limit up you get to a part of the in- come distribution where there are more and more families, so that you begin influencing the work incentives of larger and larger num- bers of people and influencing them in a way that leads them to reduce their work effort. So it seems to me that the higher you go up in the income distribution the more likely you are to have an unfavorable balance — more people discouraged and working less compared with people who are encouraged to work and who might work more.

Moreover, I think that to the extent that we think of this as a program for the working poor, it has gone well beyond that range in my opinion and consequently I would favor cutting back the upper limits substantially even though a counterpart of that would be some combination of a quicker entrance into the phase out range, a higher implicit tax rate, or lower benefit payments. But if we are going to make the same magnitude of transfer payments it seems to me there are better ways of doing it than through the IRS.

Chairman Roth. Are you saying that it would be better, admit- ting that this is essentially a transfer program, to take it out of the IRS and put it either into other agencies that have welfare pro- grams or possibly move it to the State level?

Mr. Kosters. Yes, Senator Roth, I certainly would agree with that. I think that there may be a role for a small Earned Income Tax Credit for working people with dependents, probably one that is somewhat larger than the one for those without dependents. But

126

I believe that it would be preferable to make these transfer pay- ments through other agencies that are involved in Food Stamps, in AFDC, in housing subsidies, and the like. I don't really see any point in having people pay tax preparers to qualify for this pro- gram, and to have the IRS enlarge a bureaucracy to ensure compli- ance here, when we already have many social workers and auditors in these other programs. I don't argue that these programs should not be reformed. I think they should be reformed and I think we would probably get farther in terms of encouraging work if we pro- vided help, counseling, and encouragement for people to work in connection with administering those programs then we get through administering the Earned Income Tax Credit.

Chairman Roth. One more question for each of you, would you summarize as to what effect you think the current EITC program has on work incentives?

Mr. ROSTERS. In summary I would say that some people are dis- couraged from working somewhat. Those in the upper income ranges, between $10,000 and $25,000 — are particularly discour- aged. I would say two to three times as many are discouraged from working as are encouraged. And as I already mentioned, it is only among the 10 percent or so of families with dependents who are in the relevant income range — and who have no one working — where work is encouraged.

Now, what we need to recognize, at least according to experts who have been working on welfare reform type issues, that in many instances where no one is working people are unable to work for one reason or another — health or disability and the like.

We need to recognize that we are looking to the EITC to encour- age really a very small fraction of families to consider working where there is no one working and work is feasible. Consequently, I think the favorable work incentive effects that we can realisti- cally expect from the EITC are likely to be extremely small. There- fore, I think we would do better by changing the way we approach work in our other transfer programs, and using these other pro- grams to administer the transfer payments.

Chairman Roth. Mr. Greenstein.

Mr. GREENSTEIN. I would give a different answer. I think that Congress and the States, both parties, virtually every State over 20-some years, have made clear that they are moving away from using welfare as a way to provide wage supplements to working poor families. As I said, there are only three States left that pro- vide AFDC to a mother of two children who works and is at three- quarters of the poverty line. And as you know, the welfare system largely leaves out the two-parent families. So as in many areas of policy, it is a "compared to what" question.

I would not favor taking working families, two-parent families, single-parent families that are not in or have left the welfare sys- tem and putting them back into the AFDC and the welfare system. So I think that would be a mistaken way to go.

There would appear to be some reduction in work from the peo- ple in the area where the credit goes down as your earnings go up. As Dr. Scholz's testimony States, we don't know if that is sizable because we don't know if those people are really aware of their marginal tax rate. They may be aware that they get an EIC at the

127

end of the year that offsets other taxes and their tax burden is lower, without being aware of the marginal tax rate effect; and if so, there is a new study that I have just read in the past week done at the Kennedy School at Harvard that suggests labor market data indicate that at least among — they only looked at single mothers — that among single mothers there may not be a work disincentive effect from the EIC on that down slope.

The big issue, I think, is the point that Dr. Scholz raised of the EIC's effect in bringing people who otherwise weren't working at all into the labor market. And I hope you will look at this in the context of welfare reform. If we have welfare reform that has time limits, puts significant limits on cash assistance and takes millions of mothers with young poor children and says get into the workforce, there isn't cash assistance there for you, and there is a block grant with a limited amount of money, the only way that is going to work and not cause some real tragedies and some prob- lems for States is if mothers on AFDC do move into the workforce more quickly. I think an EIC, reformed to eliminate the errors, is an essential building block of that.

If we are going to say you have got to go from welfare to work, we have got to build on the positive effects of the EIC. I think many mothers on welfare don't

Chairman Roth. But as I understand what you are saying, there should be a close relationship in the reform between the two.

Mr. Greenstein. I would like to see more education of mothers on AFDC. Here is this earned income credit. You don't work; you don't get it. You go to work; here is what you are going to get.

Most of them don't know that if they leave welfare for work, they will get this credit. If more of them knew that, I think we could have a more positive effect.

But I would be very nervous, if we, as Dr. Kosters said, reduced the upper

Chairman Roth. I want to keep these to summaries now. Yes, sir?

Mr. Scholz. In the aggregate, the EITC, in my belief, has a small but positive effect on work. Within that aggregate effect, there are compositional effects. As Marv pointed out, more wealthy EITC recipients probably are induced to work less. Lower-income EITC recipients are probably working more. And on that, I think the effect is likely to be positive.

Relative to using the $27 billion that we spend on the EITC and putting it into the AFDC or food stamp programs, the work incen- tives of the EITC are much better. After some disregards for AFDC, a household starts losing 66% cents for every dollar they earn. In food stamps, a household starts losing 24 cents for every dollar they earn. The EITC subsidizes wages and, consequently, is a better designed program.

Chairman Roth. Thank you.

Senator Glenn.

Senator Glenn. Thank you, and I appreciate your consideration, Mr. Chairman. I had to go off to another hearing, and I am glad I could get back for this final panel.

Do any of you feel it is sort of a chicken-and-egg approach here? Are employers paying lower wages because they know that, in ef-

128

feet, Uncle Sam will subsidize their payroll? Mr. Greenstein, any indication of that?

Mr. Greenstein. I am not aware of any evidence one way or the other on that. The other two witnesses may know more.

I would say that since the bulk of the EITC is provided at the end of the year in a lump-sum payment rather than through em- ployers, I don't think employers are that conscious of it.

Also, single workers who don't have children don't get the EIC above $9,000 a year, and a worker making $20,000 a year married to a spouse making $20,000 a year doesn't get it either. So an em- ployer can't really know which of its employees are and aren't get- ting it. I suspect the impact is not large on employers.

Senator Glenn. Mr. Scholz, any indication?

Mr. SCHOLZ. I know of no empirical evidence on the topic. I would be very surprised if workers did not get the full benefit of the EITC, however, given what I know about the structure of low- wage labor markets.

Senator Glenn. I was thinking, though, do employers tend to pay lower because they know that people will get EITC?

Mr. SCHOLZ. That would be the idea that part of the benefit of the EITC is shifted to employers in the form of being able to pay lower wages. I think that is very unlikely that that happens. Con- sequently, the employee gets the full benefit of the credit.

Senator Glenn. I would think it would, too.

Mr. Kosters, what do you think?

Mr. Kosters. I agree with the previous statements. I know of no evidence that is the case, and I think it is implausible.

Senator Glenn. When you are talking about the impact on indi- vidual families and their own income status, I was thinking a while ago it is amazing we have gone through 2 full days of hearings here and nobody has given us a bell-shaped curve. We usually have a bell-shaped curve that shows where most of the money goes. Do any of you have a bell-shaped curve that shows where it goes which would show what the median is for people receiving this — in other words, how much they need it? It would seem to me that is a rath- er key factor here.

In other words, you have talked about people phasing out here at the $28,000. If they were the biggest recipients of this whole bundle of $29 billion, it would be different than if the people who are getting — are at poverty level, say, in the two-thirds center of the bell-shaped curve. Do you have any figures like that?

Mr. SCHOLZ. I don't have your bell curve, but I do know that roughly half the EITC benefits that you are talking about, roughly half the $27 billion goes to families with incomes below the poverty line, and about half go into the "near poor," on up to the $27,000, $28,000 income.

Senator Glenn. Very good.

Mr. Greenstein. Let me add that the 1993 changes, I think this was one of the purposes behind their design, improved the targeting. They shifted a larger proportion of the benefits to the people closer to the bottom.

We also should note there is a $28,000 income limit for families with two kids. It is about $24,000 for a family with one child, and it is $9,000 for workers without children.

129

Mr. SCHOLZ. As a point of reference, the median family income in 1993 was around $36,000. So those upper figures are still quite a bit below the median family income.

Senator GLENN. Poverty level is at?

Mr. GREENSTEIN. For a family of four, I think it is about $16,000.

Senator Glenn. Someone earlier this morning said they thought it was $16,000 in earlier testimony.

Whether this encourages people to work or not, I haven't heard of any — maybe you have — anecdotal or even statistical evidence of people refusing a promotion, refusing an advancement, refusing a bonus because it was going to cut their EITC. Have you ever heard of cases like that?

Mr. SCHOLZ. Not the specific cases, but the effect that you might expect to see is with the secondary worker, a second earner in a particular household not taking a job that he or she might other- wise take, or a primary earner not taking a second job. That is probably where you would be likely to see that disincentive effect.

Mr. KOSTERS. About the only thing I have heard about this ques- tion is that I have heard tax preparers say it is amazing to them how many of the people who come to them have income in the range where they get the maximum credit.

Mr. GREENSTEIN. I would simply note there was a GAO study in 1993, and the GAO — of course, this is based on theory, but the GAO's analysis of the research suggested the significant effect in terms of those who would work a little less were wives in two-par- ent working families. That is where I raise that question of it is both a negative, less work, but it is a positive, if it enables them, if they wanted to spend more time raising their children and fiscal pressures wouldn't allow them to.

So that is a little different than when you have a single parent who sits at home and collects welfare and doesn't work at all.

Senator Glenn. Mr. Greenstein brought up this matter of index- ing, and I didn't hear comments from the other two on this yet. I would like to get your comments. Dr. Rosters, do you think that we should keep the EITC indexed?

Mr. KOSTERS. I think there is a good case for indexing the EITC.

Mr. SCHOLZ. I strongly think we should index the tax system.

Senator GLENN. You what? I am sorry.

Mr. SCHOLZ. I strongly think that we should index the tax sys- tem, including the EITC.

Senator GLENN. Okay. It has been a little incongruous to me be- cause we have some proposals to do away with indexing here, and yet over in the House — and I know we are not having a hearing on House legislation, but over in the House they are talking about indexing capital gains, and we have proposals in the Senate — I don't know whether it has been introduced yet or not, but we have at least one Senator who is going to say we should not index EITC and thinks that should be cut out as a way of saving money. It is a little incongruous that we are talking about indexing capital gains over there where half the benefits go to the richest 3 percent of American households, and yet we would eliminate it if affects the poorest two-thirds.

There was one thing in the Plain Dealer that somebody gave me in my office this morning, the Cleveland Plain Dealer, about GOP

130

tax plan means huge windfall for big business, and it goes into all this thing of how, if they go through with that over in the House, we have got about a $100 billion windfall for industry which will go to — three-quarters of it will go to the largest 1 percent of the corporations.

Now, I know this isn't the hearing this morning that we are hav- ing here, Mr. Chairman, but we really have some dichotomy of in- terests here when we are talking about indexing things like that and cutting out indexing for EITC. So we may want to submit some other questions, Mr. Chairman. I know this has been a long hear- ing this morning. We appreciate your forbearance in sticking with us this long, gentlemen, and we will submit additional questions we may have.

Chairman Roth. Yes, the record will be kept open for 2 addi- tional days.

Again, gentlemen, I appreciate your being here today. We un- doubtedly will want to consult with you further as progress is made in re-examining this program. Your testimony has been very help- ful.

Thank you.

[Whereupon, at 1:30 p.m., the Committee was adjourned.]

APPENDIX

PREPARED STATEMENT OF MARGARET MILNER RICHARDSON

Mr. Chairman and Members of the Committee: With me today are Mike Dolan, Deputy Commissioner, Phil Brand Chief Compliance Officer, Judy Van Alfen, Chief, Taxpayer Service and Ted Brown, Refund Fraud Executive.

We appreciate the opportunity to be here today to discuss the IRS' efforts to stop refund fraud in general and Earned Income Tax Credit (EITC) fraud in particular. The IRS has addressed tax refund fraud through its Questionable Refund Program since 1977. Teams of trained personnel in each of the 10 Service Centers have used both manual techniques and computer criteria to select suspicious returns for fur- ther review. However, technology has significantly improved the capabilities of both government agencies and financial institutions to deliver money faster. Because the risk of fraud is greater with shorter payment cycles, both public and private institu- tions must be more vigilant than ever in guarding against fraud.

Shortly after becoming Commissioner a little less than 2 years ago, I recognized the need to step up fraud detection efforts, and the IRS has taken many steps since then to ensure that fraud detection receives the highest priority. The IRS is and will remain committed to detecting and preventing attempts to undermine our tax sys- tem by those who are unwilling to comply with the tax laws. Our fraud prevention efforts are and will continue to be balanced with the need to safeguard taxpayers' rights and privacy.

Fraud is a dynamic, constantly changing and adapting phenomenon that is not unique to the government. The IRS, like other financial businesses such as credit card and insurance companies, is challenged on a daily basis by individuals who at- tempt to cheat the system. The IRS has a significant additional challenge, however, we cannot screen all of our customers based on credit history or other information — we must take all comers.

Mr Chairman, and distinguished Members of the Committee, this morning, I would like to share with you what the IRS has learned about refund fraud, describe for you the IRS' actions during the current filing season to stop this fraud, and dis- cuss with you our future fraud prevention plans.

Filing Fraud Studies

Understanding the fraud schemes confronting the IRS is essential to planning the most effective methods to detect fraud and prevent its reoccurrence. This is a dif- ficult task, especially when fraud perpetrators think creatively, relish devising com- plex schemes, and adapt continuously to new fraud controls.

Before this Committee last July, I reported on three filing fraud studies that the IRS planned to provide a more comprehensive analysis of the characteristics and ex- tent of refund fraud. The first study involved a small, statistically valid sample of 1,000 returns filed electronically during January 1994 which claimed the EITC. The EITC claimed on these returns was verified by personal contact with taxpayers, re- turn preparers, and employers. The preliminary results of this study, as reported to this Committee last July, showed that roughly 35% to 45% of the 1.3 million re- turns with EITC filed electronically through January 28, 1994, contained errors that required adjustments, both up and down, in the amount of EITC claimed. Approxi- mately 50% of the EITC claims with errors were believed to result from uninten- tional errors; the errors in the remaining 50% of the returns appeared to be the re- sult of intentional misrepresentations to qualify for EITC. Taxpayer characteristics gleaned from this study aided in the development of additional fraud controls we have implemented for the 1995 filing season. The final analysis of this study should be available in the next month, and we will share the results with this Committee.

The second study, which was conducted in February 1994, involved 2,200 tax- payers whose returns had been filed electronically by Electronic Return Originators (EROs). The purpose of this study was to determine whether refunds from the elec- tronic filing system (ELF) were being received by taxpayers as issued. This study

(131)

132

has also been completed and in only a handful of the cases did there appear to be evidence that EROs were keeping a part of a taxpayer's refund.

The third study is currency underway. It involves a statistically valid, random sample of approximately 2,000 refund returns filed electronically and on paper claiming EITC filed throughout the 1995 filing season. Results from the third study will be used to expand our understanding of issues identified during the first study. The field work and analysis of results will be completed in the fall, and we will share the results with this Committee.

Maintaining effective fraud prevention demands continuous assessment of emerg- ing trends and constant revision of prevention mechanisms. Strategies which are perfectly satisfactory today, may be of no use tomorrow. Thus, the study of refund returns filed during this filing season that is currently underway will provide valu- able information that will be used to plan our strategy for next year's filing season and beyond.

Fiscal Year 1995 Initiatives and Results

As part of its continuing efforts to prevent and detect fraud, the IRS has devel- oped and implemented numerous systemic verifications and enhancements for the 1995 filing season. In addition, significant resources are being directed to preventing and detecting questionable and fraudulent refund claims before the refunds are paid. These initiatives include increased verification of taxpayers' Social Security numbers, additional checks of returns claiming certain credits, ERO suitability checks and increased ERO monitoring, and additional compliance resources devoted to fraud detection and prevention. An important part of our strategy includes delay- ing all refunds that, as a result of computer analysis and fraud identification pro- files, appear erroneous or fraudulent. This additional time for review, in most cases up to 8 weeks, has helped and will help us detect fraud schemes, including duplicate uses of Social Security numbers.

Verification of Social Security Numbers

Internal studies and the report of an outside expert, have confirmed that fraud was being perpetrated through the use of incorrect and invalid Social Security num- bers. As a result, during this filing season the IRS is devoting substantial resources to ensuring that taxpayers claiming refunds use the proper taxpayer identification number — generally a taxpayer's Social Security Number (SSN). If electronically filed returns have no SSN, have an invalid SSN, or more than one taxpayer uses the same SSN (duplicate SSNs), the returns are not accepted into the system. A correct, valid SSN must be provided for the taxpayer, spouse, and dependents before an elec- tronically filed return will be accepted. So far this filing season, over 3.9 million oc- currences (not returns) of missing, invalid, or duplicate SSNs have been identified on electronically filed returns resulting in the affected returns being rejected.

The additional checks of SSNs are not limited to ELF returns, however. The IRS is also checking paper returns for missing, invalid, or duplicate SSNs. Failure to provide a valid SSN results in a delay of the refund until the matter is resolved.

In one Service Center, we have identified over 400 uses of the same invalid SSN. One preparer prepared returns using this phony SSN over 400 times — 64 times as the SSN for the primary taxpayer; 113 times as the SSN for children being claimed for EITC; and 261 times as the SSN for children being claimed as dependents. Be- cause of our duplicate SSN report, in this one situation, we have already delayed refunding over $380,000 pending further review.

We have spent a lot of time both before and during this filing season urging tax- payers to use correct Social Security numbers (SSNs) on tax returns for themselves and their dependents. We emphasized the importance of accurate SSNs this filing season by including a message to that effect on the cover of all tax packages and through many public service announcements. In December 1994, over 180,000 tax- payers who filed in 1994 with incorrect or invalid SSNs received letters from the IRS alerting them to be more careful on their 1995 tax returns. If the taxpayers did not have a SSN for themselves or their dependents, the letter advised them to contact the Social Security Administration before filing their tax return. While our increased scrutiny of SSNs may cause delays for legitimate taxpayers this year, once the SSN problems are corrected, these taxpayers should not experience delays in fu- ture years because of SSN problems.

Additional Scrutiny of Returns Claiming Certain Credits

Our studies of fraud also revealed, as detailed above, that a large amount of fraud is related to the EITC and the motor fuel excise tax credits As a result of these findings, during this filing season, we are performing additional checks on returns that claim these credits to ensure that only those taxpayers who are entitled to such credits receive them. Refunds are being delayed on some returns to allow us addi-

133

tional time to verify claims prior to issuing the refunds. Because of the additional time needed to complete this review, in some cases taxpayers who claim credits may initially receive their refunds of withheld income taxes followed by a separate re- fund check for the credits. If a taxpayer's refund is delayed, a notice will be sent explaining the reason for the delay. Generally, the refund will be sent within 8 weeks from the date of the notice.

Although our fraud screens are designed to detect suspicious returns, some tax- payers who have filed complete and accurate returns will have their refunds de- layed. We regret any inconvenience; I would like to note that taxpayers who have legitimate hardships as a result of the refund delay have been helped through our Problem Resolution Program.

At the beginning of this filing season, the IRS estimated that 82 million individ- ual refund returns would be filed in 1995 and up to eight percent of these refunds could be delayed with the new screens and filters put into place. Through March 24, 1995, 40.9 million refunds have been issued — 38.3 million were issued for the full amount of the refund; 2.6 million were partial refunds. Only 2.6 million refunds have been delayed in their entirety. These numbers are consistent with our esti- mates, and we continue to project that approximately eight percent of total refunds claimed this filing season will be delayed.

Some examples of potentially fraudulent refund schemes we have detected this fil- ing season are:

(a) 73 paper returns prepared by a preparer in Virginia were filed at our Philadelphia and Austin Service Centers, each of which claimed large Schedule A deductions. Because of our new automated detection systems, an additional 200 ELF returns filed by the same preparer were identified, again with large Schedule A deductions.

(b) Numerous returns with Schedule C (self-employed) net income in the $8,000 to $10,000 range claiming head of household filing status were iden- tified by one Service Center. These returns also claimed the full EITC. Most of the taxpayers have similar or identical surnames and live at or around the same address. To date, 112 returns have been identified with over $200,000 in refunds claimed.

(c) Another Service Center identified 23 suspicious returns that were pre- pared by the same preparer. These returns claimed EITC and all listed a child under the age of one; thus no SSN was required for the child. Total refunds claimed were in excess of $48,000.

Working with the Department of Justice and U.S. Attorneys, the IRS continues to actively pursue prosecution of criminal violations where appropriate. For exam- ple, in FY 94, 51 return preparers, convicted of fraudulent refund schemes, were sentenced to an average prison term of 20 months. Thus far in FY 95, 14 return preparers, convicted of fraudulent refund schemes, were sentenced to an average prison term of 21.5 months.

The IRS has received many positive responses to our fraud prevention efforts this filing season. I believe that most taxpayers understand that the IRS needs the addi- tional time to verify the accuracy of refunds claimed to maintain the integrity of the tax system.

Screening and Monitoring of Electronic Return Originators (EROs)

We have also learned that some EROs have been responsible for initiating or aid- ing a large amount of refund fraud. Thus, as part of our fraud prevention efforts this filing season, new policies and procedures were implemented for screening EROs before permitting them to access the IRS electronic filing system (ELF). Fin- gerprint and credit' checks were conducted on new ERO applicants to better ensure that only appropriate and responsible individuals participate in electronic filing. The IRS received approximately 38,000 applications for admission to ELF in 1995. Of the 33,000 applications that had to undergo suitability checks, 1,500 applicants were rejected because of failure to meet our more stringent admission requirements. Another effort we have undertaken throughout this filing season in our district offices is enhanced monitoring of EROs. So far, these monitoring visits have been extremely successful. For example, while on a monitoring visit to an ERO believed to be in non-compliance with program requirements, the ERO told us: "If you think I am bad, you should look into what another ERO is doing." That contact led to an- other ERO who was not complying with the program requirements and identifica- tion of a potentially abusive scheme involving discounting of refunds. Through March 4, 1995, we have conducted over 3,600 monitoring visits, resulting in the sus-

134

pension of 103 EROs from the program and the issuance of warnings to an addi- tional 303 EROs.

Another fraud prevention step taken this filing season was the elimination of the Direct Deposit Indicator (DDI). In the past, this indicator signaled an ERO that a taxpayer's refund would not be reduced to satisfy another government debt. Al- though the DDI has been used by lenders to determine whether to issue refund an- ticipation loans, the IRS has no involvement in these loans. Our experience over the last few years with electronic filing and the DDI showed that refund fraud schemes were assisted by the availability of refund anticipation loans. Thus, the IRS is no longer providing the indicator. Lenders are still free to make refund anticipation loans based on their usual lending criteria.

For the '995 filing season, 232 IRS offices and over 1,000 volunteer sites around the country are offering free electronic filing. As of March 24, 1995, over 122,000 individual tax returns were electronically filed from these sites. These taxpayers will have the advantages of free electronic filing, including its accuracy (almost 99.5%), the acknowledgement of receipt of the return, and faster notification to tax- payers in the event IRS questions arise.

We continue to build on our partnership with practitioners and EROs. The vast majority of practitioners and EROs are interested in maintaining the integrity of our tax system; they recognize their responsibility to prepare, file, or transmit cor- rect information to the IRS. However, when we identify those few who abuse the authority of their position by committing fraud or who fail to adhere to our program guidelines, we will take action to remove them from the program and pursue crimi- nal enforcement to the full extent where appropriate. Stopping fraud requires the combined efforts of all our partners in tax administration — tax return preparers, EROs, tax practitioners, and Congress.

Enforcement Activities

As I stated earlier, this filing season, the IRS has in place new systemic screens to detect questionable and fraudulent refund claims. In addition to this effort, our Criminal Investigation Division (CID) through its Questionable Refund Detection Teams (QRDT) is using new technology to aid in the detection of refund fraud schemes. For example, the Electronic Fraud Detection System (EFDS), an auto- mated fraud detection system, was installed in all five electronic filing centers this filing season, after a successful pilot of the program last filing season in Cincinnati. Before EFDS, the results of fraud screening were printed on paper forms to be re- viewed. EFDS converts this paper system so that it is an on-line research tool which can be used to validate claims and identify multi-return fraud schemes.

While EFDS is used in the electronic filing centers, another automated detection system, called AUTO-WIF has been installed in all 10 Service Centers. Its primary purpose is to provide IRS Service Centers processing paper returns with some of the same capability of EFDS for scheme identification and quicker access to electroni- cally filed data.

The most sophisticated fraud schemes are devised by those skilled in computer programs and techniques. They assume the existence of systemic filters and design their fraud schemes to circumvent these filters and pass through the system unchal- lenged. The sophisticated fraudsters test the system from time to time to make sure they understand the parameters being used. With this information, they increas- ingly generate multiple transactions and attempt to incorporate sufficient random- ness or variation to minimize the risk of detection.

To identify these sophisticated fraud schemes, we are using the Los Alamos Na- tional Laboratory to design software to detect anomalies and match patterns in large data sets. New anomaly detection/pattern recognition tools were developed and are being tested at the Cincinnati Service Center this filing season. Returns with these patterns can be identified and removed from normal processing for further scrutiny. As we continue to identify the items on returns that are predictive of fraud, we will add them to our systemic filters.

In addition to enhancing our systemic filters to detect more questionable refund claims this filing season, we have substantially increased the enforcement resources dedicated to identifying fraudulent schemes, as well as examining questionable claims. Criminal Investigation resources in our Questionable Refund Detection Teams were increased by 11 percent and Examination resources were increased by 277 percent — over 1,700 enforcement staff years are being devoted to curbing the abuses and fraud. As of March 24, 1995, we have identified 623 ELF schemes. In connection with these schemes we have delayed $15.3 million claimed on 7,095 re- turns. We have also identified 632 paper return schemes involving 4,312 returns and have delayed refunds claimed of $8.3 million.

135

In past filing seasons, refunds were not delayed until after a potentially fraudu- lent scheme had been identified. As a result, the refunds on many returns involved in a scheme would be issued before the scheme was detected. The delay of question- able refunds this filing season has provided additional time to identify fraudulent claims and to select questionable claims for examination before the refunds are paid.

Earned Income Tax Credit

Although detection and prosecution of refund fraud are important, it is virtually impossible to prosecute every instance of fraud. Thus, the IRS must continue to build barriers to fraud. The IRS' goal is to prevent fraudulent returns from entering the system, and one of the biggest challenges in meeting this goal is to install the "up-front" fraud controls that will effectively detect and prevent fraudulent refund claims from entering the system.

A number of the initiatives and systemic enhancements installed in the 1994 fil- ing season and the current filing season were designed to stop fraud involving the EITC. This filing season, we estimated that about 20 million taxpayers would claim their EITC, an increase of about 5 million over 1993. This increase results from ex- panding the EITC to taxpayers without children. We have just received our first re- port of EITC claimed for this filing season. As of March 1995, 10,940,000 returns have been filed claiming the EITC. This is up slightly from the 10,429,000 returns filed claiming EITC at the same time last year — an increase of only 511,000 returns. The total EITC claimed so far is $13.8 billion— an increase of $2.6 billion over the $11.2 billion claimed at this time last year.

These results may be an early indication that the fraud control initiatives put in place this filing season are reducing the number of fraudulent claims involving EITC, thus making the EITC unattractive to the fraudsters and preserving the cred- it for those who have earned it. However, when we complete the filing season study that I mentioned earlier, we will be able to measure EITC compliance more pre- cisely.

Through these same systemic filters and detection efforts that are addressing EITC fraud, the IRS is making strides in stopping fraud in other areas, such as motor fuel excise tax credits. For example, on one return filed this season, a self- employed beautician claimed the tax-free use of over 42,000 gallons of gasoline, gen- erating a refund of $6,000. On this return and many like it, the refunds were stopped.

It has been estimated that between 80% and 86% of all eligible families actually claimed the EITC in 1990. Through our education and publicity efforts, the IRS is making a concerted effort to reach an even larger percentage of eligible families. For example, last year we sent 14.7 million EITC recipients information about the ad- vanced earned income tax credit (AEITC). The Treasury Department contacted cor- porate CEOs to gain their support and solicit other CEOs and national organiza- tions to provide information about AEITC. In addition, we are using our taxpayer education programs to promote the AEITC to those who are eligible by encouraging employers and community or social service organizations to conduct seminars for employees and clients. The IRS and the Small Business Administration are also co- sponsoring employer seminars in strategic locations throughout the filing season to train employers how to compute the AEITC for their employees.

Workers who qualify for the AEITC can get up to $105 per month in their pay- checks— whether they get paid weekly or bi-weekly — by filling out a very simple Form W-5, Earned Income Tax Credit Advance Payment Certificate and providing it to their employers. By claiming the earned income credit on an advanced basis, taxpayers who are eligible for EITC can avoid potential refund delays and use these funds during the year to pay for expenses. They do not have to wait until they file their returns to get the credit. So far this filing season, preliminary results show that more taxpayers have opted for AEITC than in all of 1993.

Future Fraud Prevention Plans

Mr. Chairman, I assure you that the IRS is committed to stopping all fraud, in- cluding EITC fraud, and that we will continue our efforts to ensure that only those hard working Americans who are eligible for the EITC receive it. Although we are still in the midst of our current filing season, at the direction of Mr. Brown, the Filing Fraud Executive, we have already begun planning our fraud prevention strat- egy for next filing season.

We are gaining valuable information this filing season through studies, ERO mon- itoring, and enforcement activities, on which to base the modification and refine- ment of our current strategies. Over the next few months, we will be reviewing this information. Based on the results of our review, we will revise the standards used

136

to screen EROs, and adapt the systemic screens used to detect fraud during this filing season, and, if necessary, put in place new filters.

While we will continue to enhance our detection and prevention efforts, the key to improving our ability to detect fraudulent refundable credit schemes is our Tax Systems Modernization program. Without modern equipment and software, applying expert systems analysis to large data bases is virtually impossible. Tax Systems Modernization will not only provide the computing power and capacity needed to apply sophisticated fraud detection systems, but it will also provide us with more timely access to information.

Mr. Chairman, now more than ever, we need to find a way to assure a stable funding vehicle for the completion of Tax Systems Modernization. Fiscal Year 1996 is a pivotal year for the IRS as we continue our plans to acquire and implement major new systems. What happens to our FY 1996 budget will impact the tax ad- ministration system of the future, shaping our ability to effectively administer the tax law and collect all the revenue that is due.

Conclusion

As I stated earlier, fraud is a dynamic, constantly changing phenomenon. Preven- tion and deterrence are clearly the keys to controlling it. We will continue our pro- grams to prevent, detect, investigate, and prosecute all types of refund fraud. Mr. Chairman, even if the IRS is successful in our current efforts to eliminate all EITC fraud, our job will nut be done. In our experience, when one avenue of fraud is shut down, fraudsters merely migrate to other more accessible avenues. As I have stated before, in some instances it may be necessary to delay questionable claims for re- funds while they are carefully scrutinized and pay interest, rather than risk allow- ing fraudulent claims. The IRS will remain vigilant in its fight against fraud to en- sure that those who choose not to comply with the law are caught.

Mr. Chairman, this concludes my prepared remarks. My colleagues and would be happy to answer any question you or other Committee Members may have.

PREPARED STATEMENT BY LYNDA D. WILLIS

SUMMARY

The Earned Income Credit (EIC), which is expected to provide about $22 billion in tax credits in 1995, is a major federal effort to assist the working poor. The EIC is intended to (1) offset the impact of Social Security taxes on low-income workers and (2) encourage low-income individuals to seek employment rather than welfare. GAO's statement makes the following points:

— A reliable overall measurement of noncompliance with EIC provisions has not been made since 1988. But noncompliance appears to be a problem. The Internal Revenue Service (IRS) studied electronically filed EIC claims during 2 weeks of January 1994. IRS estimated that 29 percent of the returns claimed too much EIC, and that about 13 percent of them may have done so intentionally. Judging by problems spotted by IRS per- sonnel, noncompliance on EIC paper returns is also a concern. In 1994, IRS withheld refunds totaling about $500 million from about 400,000 paper return filers due to insufficient proof that they qualified.

This year IRS has taken several steps to detect and prevent erroneous payments to EIC claimants. If implemented effectively, these steps could help improve the overall level of EIC compliance.

— Although the EIC is intended to assist the working poor, EIC eligibility criteria do not consider all of the resources recipients may have to sup- port themselves and their families. At your request, the Joint Committee on Taxation (JCT) has estimated that changing eligibility criteria to test for taxpayers' wealth could yield between $318 million and $971 million in revenue savings in fiscal year 1997, depending on the design of the test. Another approach taking into account taxpayers' resources would add certain income to taxpayers' adjusted gross income when determining EIC awards. According to JCT estimates, up to $2.1 billion could be saved in fiscal year 1997 by recognizing this income. Either attempt to better measure resources available to taxpayers would make the EIC more com- plex and add to the burden on taxpayers and IRS. Also, income informa- tion reported on tax returns can only roughly reflect taxpayers' actual wealth, and using such data to determine EIC eligibility could raise fair- ness concerns.

137

-No one knows how many illegal aliens receive the EIC. Illegal aliens may receive the EIC if they meet the credit's eligibility rules. Awarding the EIC to illegal aliens, however, works at cross-purposes with federal poli- cies that prohibit illegal aliens from legally working in the United States. If the EIC criteria were revised so that all EIC recipients needed valid Social Security numbers for work purposes, illegal aliens would no longer qualify.

Mr. Chairman and Members of the Committee: We are pleased to be here today to assist in your effort to better ensure that only the working poor receive the Earned Income Credit (EIC). As you requested, we are providing a report on the EIC along with our testimony today. I ask your permission to have the report in- serted in the hearing record. Our testimony and the accompanying report are part of our ongoing work for you.1

Background

Originally authorized in 1975, the EIC provides assistance to low-income working taxpayers to offset the impact of Social Security taxes and to encourage them to work. At various times Congress has broadened EIC coverage and increased the credit amount to ensure that the EIC amounts would not fall in purchasing power, to increase or maintain the progressivity of the tax system, and to better ensure that working individuals will have incomes above the poverty line. As figure 1 illus- trates, with these changes the overall cost of the EIC is expected to increase more than five-fold in real terms between 1988 and 1996, when the EIC costs are esti- mated to total $24.5 billion.

The most recent changes to the EIC, in the Omnibus Budget Reconciliation Act (OBRA) of 1993, increased the maximum credit available and the income level at which individuals can qualify for the credit. For the first time, it granted eligibility to certain low-income taxpayers without children. As figure 2 illustrates, the credit gradually phases in, plateaus at a maximum amount of $3,370 for a taxpayer with two qualifying children in 1996, and then phases out until it reaches zero.

1 Tax Administration: Earned Income Credit — Data on Noncompliance and Illegal Alien Re- cipients (GAO/GGD-95-27, Oct. 25, 1994) focused on EIC noncompliance.

138

FIGURE 1: GROWTH IN EIC PROGRAM COSTS (1988 - 2000)

GA0 Growth In EIC Program Costs

(1988-2000)

Millions of dollars (1994)

30000 27P4i r^a^a

11,127

1990 1992 1994 1996 1998 2000

Projected

Source: Fiscal year estimates from the Presidents' 1990, 1992, 1994, 1996 budgets.

FIGURE 2: RANGE OF EIC FOR RECIPIENTS WITH TWO QUALIFYING CHILDREN (1996)

GA0 Range of EIC for Recipients With Two Qualifying Children (1996)

Earned Income credit (1994 dollars)

$10,999 $11,000

$27,000

$1 ' $8,424 $8,425

Range of recipient's Income □ Phase in: recipient receives PI Maximum range: recipient R Phase out: recipient $.40 for each $1 dollar of receives fixed $3,370 EIC receives $.21 less for

income each $1 of income

Source: Congressional Research Service.

139

Noncompliance

Noncompliance has been and continues to be a problem for the EIC.2 For instance, compliance measurements done by the Internal Revenue Service (IRS) in 1988 esti- mated that about 42 percent of EIC recipients received too large a credit and about 34 percent of the total EIC paid out may have been awarded erroneously.

Although a current, statistically-valid measure of overall EIC compliance does not exist, the results of limited studies and of IRS efforts to enforce the EIC suggest that a significant compliance problem remains. An IRS study of electronically-filed EIC returns during a 2-week period in January 1994 found an estimated 29 percent of returns claimed too much EIC; 13 percent of the returns v/ere judged to have in- tentional errors — a surrogate measure of possible fraud.

This filing season, IRS has expanded its efforts to ensure EIC compliance. In doing so, IRS is using lessons learned from its 1994 study and enforcement experi- ence to improve its systems to identify possible noncompliant returns. Verifying So- cial Security number accuracy is key to IRS' efforts. IRS checks the accuracy of So- cial Security numbers — expanding their efforts this year to emphasize those used for dependents and EIC qualifying children. As of March 17, 1995, IRS' verification procedures had identified nearly 4.1 million instances of problems with returns. These primarily involved returns that did not appear to contain valid SSNs for de- pendents or qualifying children.

In addition, as of March 17, 1995, IRS had delayed refunds to at least 2.9 million EIC claimants for up to 8 weeks. This delay is intended to allow IRS staff time to identify any SSNs that have been used on more than one tax return. IRS identified duplicate SSNs as a problem during the 1994 filing season. For the delayed returns, IRS generally sends out the portion of any refund that was due to overpayment of taxes but withholds the EIC portion of the refund claimed.

Not surprisingly with a large new initiative, IRS experienced some problems as it began checking for duplicate SSNs. These problems included difficulties in con- structing the data base to use in identifying the duplicate SSNs, poorly-organized computer listings that enforcement personnel found difficult to use, and cum- bersome procedures for coordinating among IRS service centers. IRS national office officials told us that initial problems with the duplicate SSN system had been over- come early in the filing season. But compliance personnel continue to report prob- lems using duplicate SSN data. We intend to continue monitoring this effort.

Although it is too early to assess the success of IRS' new or expanded enforcement initiatives, the steps taken seem to be focusing appropriately on current indicators of problematic returns. Despite IRS' efforts to better verify EIC claimants' eligibility before processing refunds, IRS cannot currently verify all eligibility criteria before sending refunds to taxpayers. In the long run, sound enforcement of the EIC may require even better verification of recipients' eligibility before refunds are made.

We have made several recommendations in the past that could help to make the EIC less of a problem for IRS and taxpayers. As discussed more fully in Appendix I, those recommendations called for eliminating differences between the definition of a qualifying child for EIC purposes and the definition of a dependent for purposes of claiming a dependency exemption; encouraging the advance payment option, whereby persons eligible for the EIC can choose to receive it in advance as part of their paychecks; and moving toward timely computer matching of employer wage in- formation with tax return data.

Better Measuring EIC Filers' Resources to Determine Eligibility

Although the EIC is intended to provide assistance to the working poor, unlike certain welfare programs, taxpayer wealth is not taken directly into account in de- termining EIC eligibility or the amount of the credit received. EIC criteria also do not consider all types of income taxpayers may receive.

At your request, we assessed the potential changes in overall EIC costs that might result from including a wealth test and a more comprehensive adjusted gross in- come test in determining eligibility. We also evaluated the administrative implica- tions of expanding the eligibility criteria. Generally, to facilitate administration of the expanded eligibility criteria, we initially looked at items that are currently re- ported in some form on the individual's income tax return.

For the wealth test, we analyzed asset-derived income such as taxable interest and dividends, tax-exempt interest, estate and trust income, rental income, and cap- ital gains. For the expanded adjusted gross income test, we first analyzed the im- pacts of including nontaxed Social Security income, tax-exempt interest, and nontaxed pension distributions in the taxpayer's adjusted gross income. At your re-

2 Noncompliance includes erroneous EIC claims caused by negligence, mistakes, confusion, and fraud.

140

quest, we subsequently added child support payments — which do not currently ap- pear on any IRS form — to the income items.

Based on our work, you requested that JCT provide revenue estimates for various eligibility options we had reviewed. According to JCT estimates, denying the EIC to taxpayers whose income from wealth exceeds a certain threshold could reduce program costs $318 to $971 million in fiscal year 1997, depending on the design of the test.

Expanding taxpayers' adjusted gross income to include nontaxed Social Security income, tax-exempt interest, and nontaxed pension distributions could yield $1.45 billion in the same period, according to JCT estimates. Also adding child support payments to the expanded adjusted gross income would increase 1997 revenue sav- ings by another $686 million.

However, adding an indirect wealth test or an expanded adjusted gross income definition to the EIC eligibility criteria would add to the EIC's complexity and ad- ministrative burden Complexity has been a continuing EIC issue because it can lead to increased errors and dissuade deserving taxpayers from claiming the credit. Of the potential changes to EIC criteria, adding child support to taxpayers' adjusted gross income likely would cause the greatest complexity, because information on such income is not collected by IRS, and systems may not exist to generate the in- formation.

There are significant limitations in measuring potential EIC recipients' actual wealth through income reported on tax returns. For instance, such a test would not measure the value of taxpayer assets such as capital stock funds that yield little, if any, annual income. These limitations could raise concerns that taxpayers with similar wealth could be treated differently for the EIC.

Illegal Alien Recipients

The Internal Revenue Code does not prohibit illegal aliens from receiving the EIC, if they meet the prescribed eligibility requirements. However, illegal aliens cannot be employed lawfully in the United States. Because the EIC is intended in part to encourage employment, it works at cross purposes with the prohibition on employ- ment of illegal aliens

Although no one knows how many illegal aliens may be claiming and receiving the EIC, IRS estimated that a minimum of 160,000 taxpayers, out of about 8.7 mil- lion who filed paper returns claiming the EIC in 1994, were likely to be illegal aliens.3 IRS expected most of these refunds to be denied, because taxpayers would not support their claims by verifying that the dependent met the age, relationship, and residency requirement.

Some unknown portion of returns may also be filed by illegal aliens who use SSNs belonging to other individuals. IRS' new efforts to detect duplicate uses of SSNs, if successfully implemented, should reduce the number of illegal aliens as well as U.S. citizens incorrectly receiving the EIC.

A Senate bill you introduced in 1994 and the administration's Tax Compliance Act of 1995 (H.R. 981 and S. 453) would deny the EIC to illegal aliens. The administra- tion's proposal would require that all EIC recipients provide SSNs that are valid for employment in the United States for themselves, for their spouses, if applicable, and for qualifying children. Because illegal aliens cannot qualify for SSNs that are valid for employment in the United States, they would not be able to receive the EIC. The administration's proposal would permit IRS to use streamlined procedures to enforce the requirement that EIC claimants have valid work-related SSNs.

The administration estimates that requiring all EIC recipients to provide valid work-related SSNs and using streamlined procedures to enforce this requirement would yield about $400 million in revenue savings in fiscal year 1997.

Mr. Chairman, this concludes my prepared statement. My colleagues and I would welcome any questions that you may have.

APPENDIX I

What Could Be Done To Make the EIC Less of a Problem?

Refundable credits, like the EIC, pose a challenge for tax administrators. In addi- tion to the concerns about fraud, there are equally important concerns that not all

3 IRS officials made this estimate based on their enforcement experience and the number of taxpayers entering a code "205(c)" instead of an SSN for their qualifying child. EIC claimants are required to provide an SSN or taxpayer identification number for themselves and their qualifying children. The designation 205(c) is often used by taxpayers to indicate they are not eligible to receive an SSN

141

those eligible for the EIC are receiving it. We have made several recommendations in the past that could help to make the EIC less of a problem for IRS and taxpayers.

The definitions of a qualifying child for purposes of claiming the EIC and of a de- pendent for purposes of claiming a dependency exemption are not the same. A key difference in the two definitions is the requirement, for purposes of claiming a de- pendency exemption, that the taxpayer provide over 50 percent of a dependent's support (referred to as the "support test"). There is no support test in the definition of a qualifying child for EIC purposes. We addressed this problem in a March 1993 report, in which we analyzed four alternatives to simplify the laws on dependent exemptions, including two that would change the support test.4 On the basis of our analysis, we recommended that Congress consider enacting legislation that would substitute a residency test similar to that used in the EIC program for the depend- ent support test when the dependent lives with the taxpayer.

Persons eligible to receive the EIC can choose to receive it in a lump sum pay- ment after filing a tax return or in advance as part of their paycheck. In February 1992, we reported that less than 1 percent of EIC recipients in 1989 took advantage of that second option.5 Although use of the advance payment option would help tax- payers benefit from the credit sooner, it could also create problems for IRS if per- sons receiving the advance payment later filed a tax return but did not report that they had received the credit in advance. Under IRS' returns processing procedures in place at the time we did our review, those persons could receive the credit again as a lump sum payment. We recommended that IRS take various steps to (1) better ensure that eligible taxpayers are aware of the advance payment option and (2) pre- vent those who take advantage of that option from receiving the credit a second time. When last we checked, IRS had taken steps to better publicize the availability of the advance payment option but had not revised its procedures to protect against duplicate payment of the EIC.

With respect to fraud on electronically filed returns, we recommended in Decem- ber 1992 that IRS work toward electronically matching employer wage information with electronic return data.6 That kind of match is currently beyond IRS' computer capabilities. Currently, employer wage information other than that provided by tax- payers is not available to IRS until after it has processed taxpayers' returns. This is because of the time it takes to verify the information and correct any errors.7 IRS has begun to test the possibility of getting partial year's wage information from the States and using that to verify that the taxpayer is employed and to have some in- formation on the taxpayer's amount of earned income.

PREPARED STATEMENT OF MR. HERSCH

Good morning, Mr. Chairman and Members of the Committee. I was invited here to discuss the ease with which fraud can be perpetrated on the IRS' electronic filing system. I am grateful for the opportunity to appear before you today regarding these problems and, specifically, the easy target that the Earned Income Tax Credit makes for people like me who are tempted by its invitation for abuse.

I have some qualifications to speak on this issue. In 1993, I was indicated for fil- ing false tax returns and for money laundering that I committed as a tax preparer for the tax years 1991 and 1992. I pled guilty to all of these charges in January 1995, and am currently awaiting sentencing in the United States District Court in Boston. I would like to describe briefly for you my background, my tax preparation business, and the nature of the schemes I used to defraud the government.

Background

I got into the tax preparation business because in 1990, I loaned someone else money to start a tax business, he took off, leaving me with computers and equip- ment that he purchased and 1,000 tax returns to file. I had no license or other cer- tification to prepare tax returns myself, and no experience; but I quickly discovered

4 Tax Administration: Erroneous Dependent and Filing Status Claims (GAO/GGD-93-60, Mar. 19, 1993)

5 Earned Income Tax Credit: Advance Payment Option Is Not Widely Known or Understood by the Public (GAO/GGD-92-26, Feb. 19, 1992).

GTax Administration: IRS Can Improve Controls Over Electronic Filing Fraud (GAO/GGD-93- 27, Dec. 30, 1992).

'Under the Electronic Management System — one of many planned components of Tax Sys- tems Modernization — IRS expects to electronically receive tax returns, tax information docu- ments (like W-2s), and correspondence. Electronic transmission of W-2s would enable IRS to more quickly verify and correct the information, thus offering the possibility of having that in- formation available to match with data being reported on electronic returns.

142

that none of this was needed, not even to prepare and file the 1,000 returns that my erstwhile business associate had left for me. Those returns had been sent to him by check cashing businesses in the Philadelphia area.

Before continuing with my story, I should explain briefly the nature of the "check cashing" business and their customers. The check cashing businesses that I dealt with the first year and in subsequent years were located in depressed inner city neighborhoods; their customers were usually poor. The customers did not have bank accounts, and they used check cashing businesses to cash their paychecks and wel- fare payments, to pay their bills, and to conduct their other basic day-to-day trans- actions. These businesses charge high rates — often a high flat fee or a percentage of the amount of the check or transaction — in order to handle their customers' busi- ness. As I learned, the check cashing businesses also advertised and provided tax preparation services. Those services were usually provided by an outside and unli- censed preparer, like me, who would receive "customers" for the check cashing busi- nesses.

Returning to my story, I prepared and filed in 1990 all of the 1,000 returns that had been left by my business associate, and I did so honestly, relying on the infor- mation supplied by the customers to the tax preparation businesses. I received only my legitimate filing fee. In the course of reviewing the customers' tax information, however, I discovered clear evidence that many of them were lying to the IRS on their returns in order to take advantage of the EITC. And they were getting away with it. No problem. The customers were falsely overstating the number of their de- pendents, lying about their status as "head of household," and claiming to have earned wages for work that they never did by falsifying the W-2 that they submit- ted to me. Maybe because there were so many people doing it and maybe because of the small amount of money involved for each taxpayer, the IRS never seemed to care.

EITC could stand for "Easy Income for Tax Cheats." As I discovered, it was a par- ticularly popular vehicle for fraud for several reasons. First, the EITC was a refund- able credit that a taxpayer gets depending on how little they earned within a range that qualified them as low-income taxpayers. Second, the EITC benefits low-income taxpayer only to the extent that they show dependents on their returns. The tax- payer claims some wages — but not too much — adds up two dependents, and thereby maximizes his refund, without adequate verification. Third, the EITC was so well known in the tax preparation business that they promoted it heavily and based much of their short term lending business — in the form of "refund anticipation loans," or "RAL's" — on it.

My Business and Fraudulent EITC Refunds

My business was coming from check cashers and tax preparation companies. I as- sembled a network of check cashing companies in 25 States that contracted to have me prepare and file tax returns for their customers for the coming year. My busi- ness grew dramatically. My company filed a total of 9,000 returns in 1992, for tax year 1991, which netted my customers approximately $18 million in total refunds. Of that total, I would guess that roughly half of the returns contained false informa- tion about dependents, wages, or filing status that allowed the customers to receive more money through the EITC than they were entitled to.

That year, in 1992, I recognized how easy it was. All I needed were fake names and Social Security numbers that I made up by using any combination of nine num- bers. I never worried about the IRS cross checking any of this information. I just made up fictitious returns for fictitious filers. I personally filed 200 fictitious returns that year, which netted me roughly $200,000 in refunds. In addition to the money I made from my fictitious filers, I also made money processing the false returns given to me by real customers. For these customers, I, and often the tax preparation company who had sent me the customers, would keep a portion of the taxpayer's overstated EITC refund.

In 1993, business continued to boom. I formed a new company, Quik Tax Dollars, which was a joint venture between myself and a company called Monetary Manage- ment Corporation, which owned 130 check cashing businesses. I ran the day-to-day operations of this tax preparation and filing company. That year, we filed 29,000 tax returns, resulting in roughly $50 million in refunds. I estimate that 40 percent of those refunds were based on falsely obtained EITC's. As with the year before, I made money by pocketing a portion of the overstated refunds received by real cus- tomers. I also profited by turning non-qualifying taxpayers into qualifying taxpayers entitled to a refund and pocketed the whole amount. I filed approximately 400 of those returns that year, yielding roughly $500,000 in false refunds.

That same year, the rules of the game changed a little, although nothing I could not overcome. The IRS started cross checking Social Security numbers with the first

143

four letters of a taxpayer's last name. Now, I could no longer just fabricate tax- payers' names and Social Security numbers. Instead, I simply took the first four let- ters of a customer's name, changed the last several letters. I used the customer's actual Social Security number, but inflated the wages and number of dependents, and, often, changed the filing status, all to maximize the refund. In most cases, I would file the return, and pocket the inflated amount after giving the customer his refund amount.

Refund Anticipation Loans

I have explained the substance of the tax fraud I was engaged in. I want to make it clear, however, that none of this would have been possible without the refund an- ticipation loan (RAL) practices of the tax preparation business. The RAL's brought the customers in the door, and the RAL practices made it possible for me to take a cut of a real taxpayer's fraudulently inflated EITC or to bilk the Government for a phony return. The RAL business works very simply. A customer looking to file a tax return and obtain the money quickly walks into a check cashing or tax prepar- ing company. The customer hands over whatever tax information he has and at the same time applies for a RAL. The tax information comes to me, and my company prepares and files the tax return electronically with the IRS. When the IRS receives the electronic return, it usually acknowledges the return, which means only that the IRS has quickly scanned the return for very limited purposes. This acknowledge- ment does not mean that the return is accurate. But this acknowledgement is suffi- cient to permit the check to be cut and the RAL to be issued. I then had the author- ity to cut the loan check directly to the customer, netting out my fees, as well as the fees of the bank and the check cashing outlet. The amount of interest charged can be staggering in States that do not regulate these practices. I would sometimes receive a portion of the tax preparation company's take as a kickback Only a few large banks are really significant players in the RAL business.

Getting Caught

Despite how straightforward my schemes were, I was caught only because several of my employees of my company became informants and went to the authorities. I am confident that if the employees had not turned me in, the IRS would never have caught on and that I would still be in business today.

Conclusion

This is really just a brief summary of the innumerable schemes opened by the IRS' electronic filing system. For example, last year I discovered a widespread tax fraud scheme in Southern California involving fake W-2's which I reported to=-the IRS. In all modesty, it would take several hours for me to share with you the vir- tually endless possibilities. Thank you for your time today, Mr. Chairman and Mem- bers of the Committee. If you have any questions, I will be very happy to answer them.

PREPARED STATEMENT OF MR. STEIN

Mr. Chairman, and Members of the Committee, my name is Dan Stein, and I am the Executive Director for the Federation for American Immigration Reform, or FAIR. FAIR is a national public interest organization working to end illegal immi- gration. We also support a general moratorium on most legal immigration (such as that suggested in S. 160, introduced by Senator Richard Shelby). We support an im- migration policy that serves the American people and our interests as a Nation. With 70,000 members in all 50 states, FAIR has become the leading organization in America working for tighter immigration laws. Today, I am presenting FAIR's views on the growing use of the Earned Income Credit by alien unlawfully present in the United States, and suggesting ways that the Congress can slow this alarming trend.

Mr. Chairman, at the outset, I would like to acknowledge your excellent past lead- ership on a range of very important immigration law enforcement initiatives. The work pursued under your guidance by the Permanent Subcommittee on Investiga- tions in years past has been in the very forefront of congressional investigations and analysis of the relationship between international organized crime and the increase in illegal alien criminals operating inside the United States. An excellent expose on ABC News "Day One,"1 on the current problems of the Immigration and Naturaliza- tion Service in detecting and deporting criminal aliens was a direct by-product of

See Transcript, air date January 12, 1995.

144

the work undertaken under your supervision by the Joint Committee on Investiga- tions.

The subject of my statement here today represents another bi-product of that con- tinuing interest in improved law enforcement, and we wish to recognize your initia- tive in requesting an analysis by the General Accounting Office (GAO) examining the use of the Federal Earned Income Credit (EIC) by illegal aliens and aliens with- out authorization to work in the United States. The EIC is a part of the tax code's effort to provide a direct tax credit for persons with earned income in the United States. It was begun in 1975 to offset the impact of Social Security taxes on low- income families, and to encourage low-income families to seek employment rather than welfare.

Mr. Chairman, in my statement, I will be frequently referencing the GAO report you requested on the topic of this testimony: Tax Administration: Earned Income Credit — Data on Noncompliance and Illegal Alien Recipients; GAO/GGD-95-27 (Octo- ber 25, 1994).

EARNED INCOME CREDIT FOR ILLEGAL ALIENS: A PRODUCT OF BALKANIZED ENFORCEMENT POLICIES

Background

Current Internal Revenue Service (IRS) operating practices are producing the re- sult that anyone who earns taxable income in the United States — even someone who works illegally in the U.S. — and meets certain non-immigration-related IRS quali- fications may receive the EIC. Tax Analysts, 94 STN 190-129, 190-130 (July 29, 1992). (Field Service Advice: IRS Field Office, Des Moines, Iowa, District asks, "Can an illegal alien claim the EIC?" IRS Office of the Chief Counsel, IRS Washington, D.C.: "I don't see why not.")

The IRS makes no attempt to verify the alienage of a claimant with the Immigra- tion and Naturalization Service (INS), nor does it report suspected illegal aliens to the INS as a part of its routine operations.

The IRS estimates that in 1994 at least 160,000 illegal aliens claimed the tax credit. This low estimate is based on speculation that those filers who were engag- ing in apparent fraud or who were unable to produce valid Social Security numbers were probably illegal aliens. Alienage was never verified, and no "illegal alien" was ever actually identified as such. These 160,00 EIC filers mentioned in the GAO re- port as probable illegal aliens were never actually identified as illegal aliens; they were only surmised to be so by virtue of not reporting any Social Security numbers or reporting only an invalid one. However because the IRS looked only at claims above a certain dollar threshold, and because many illegal aliens have fraudulently obtained "valid" Social Security numbers, 160,000 is certainly not the total number of illegal alien resident filers. The number is probably far higher.

Failure to exclude illegal alien workers — most of whom work at low wage jobs — is costing the federal government and the American taxpayer hundreds of million of dollars each year. Data from the 1986 amnesty furnished by the Immigration and Naturalization Service show that almost 70 percent of the 3.2 million aliens who received amnesty lived in families with incomes so low the filers would qualify for the EIC.2 Under the current practices followed by the IRS, illegal aliens are eligible for up to $2,528 in EIC cash payments in 1994.3 In 1990, illegal immigrants are informally estimated to have received $250 million in EIC benefits.4 In 1994, if only half of the 160,000 received the $2,528, the cost would be approximately $200 mil- lion.5 The actual number is probably much higher. Because of the current dismal state of birth and immigration documentation in the United States, many aliens can obtain legal U.S. birth certificates and other documentation that would enable them to obtain legitimate Social Security numbers through fraud.6 Only through a mecha- nism that would insure only authorized aliens obtain Social Security numbers will illegal aliens be denied the EIC. One such proposal can be found in the recent report of the Commission on Immigration Reform (chaired by Barbara Jordan) that would

2 Immigration Reform and Control Act: Report on Legalized Alien Population," INS, Depart- ment of Justice, March 1992. A total of 69 percent of amnesty recipients have family incomes of less than $25,000.

3The amount will rise to $3,370 in tax year 1996. 26 U.S.C. §32 (1994).

4 See, Beck, Roy, "The Checks are in the Mail," National Review (April 17, 1995).

5 The INS estimates that many of the probable 160,000 illegal aliens without valid Social Se- curity numbers will not receive the credit because their claims were also fraudulent on other, independent grounds.

6 Powell, Joy, Omaha World-Herald, "Workers Get Valid Papers by Illegal Means." March 12, 1995, at Al (attached).

145

recommend a model state-by-state electronic birth record system with on-line inter- active capacity.

Mr. Chairman, the EIC program as administered by the IRS is symbolic of what is creating rising resentment in the electorate at large. The failure of Federal agen- cies to share basic information in the enforcement of immigration laws demonstrates how one major Federal interest, control over the nation's borders, can be subordi- nated to another Federal interest, tax collection (or, in this case, disbursement). The three agencies involved, the IRS, the INS and the Social Security Administration, rarely cooperate in such a way that the INS gets key information needed to enforce immigration laws. When the INS does get, it rarely has the means, manpower or political will to use the information effectively. The result? Illegal aliens quickly dis- cover that they can file for the EIC without any fear of detection and removal for immigration law violations. Immigration law enforcement has been completely sub- ordinated to other areas as an important priority in the field of Federal program administration.

In the past 20 years, the ability of Federal agencies to rely on the INS for prompt, effective verification of alien status has disappeared. No longer does the INS have quick, ready and reliable access to case files to verify the Social Security number presented by an alien. The Social Security Administration has also had a notable lack of interest in ensuring the integrity of the numbers it has issued. Until re- cently, proof of citizenship was not required in the issuance of Social Security num- bers. Until recently, the SSA provided permanent Social Security numbers to aliens who were only authorized to work temporarily. Although now the SSA provides "temporary work authorized" Social Security numbers, there is no evidence the IRS takes the expiration dates into account when processing the EIC. In short, there is virtually no interaction between the INS, the IRS and the SSA to insure that illegal aliens do not obtain the EIC. This situation is peculiar in light of the vast range of areas listed in 26 U.S.C. §6103 where Congress has authorized the IRS to share information in tax returns to a variety of agencies for a variety of purposes.

Eligibility

Illegal aliens are eligible for EIC cash payments due to a series of legal loopholes and mistaken presumptions.

• The presumption by the IRS that all government programs and benefits are extended to illegal aliens unless specifically prohibited. According to the IRS, since neither congressional legislation nor the Internal Revenue Code specifi- cally prohibit illegal aliens from EIC eligibility, they may receive the benefit. Congress has expressly prohibited non-resident aliens from receiving the EIC even if they earn the money legally, 26 U.S.C. §32(c)(2)(B)(iii), so, by exten- sion, the IRS has concluded that because aliens unlawfully resident and ille- gally working are not excluded from EIC eligibility by statute, they must be included. It is this logic that has enabled the EIC to become what is effec-_ tively another welfare magnet drawing persons illegally to the United States.

• According to David Simcox, a fellow at the Center for Immigration Studies the Federal Government provides funds to religious and immigrant-aid groups to persuade immigrants (no distinction is made concerning legal sta- tus) to encourage and assist immigrants in applying for the EIC. This prob- ably explains why EIC claims are filed electronically in numbers that far ex- ceed their overall percentage of tax forms filed.7 In fact, private publications for immigration lawyers and activists advise that immigration status is irrele- vant in filing for the benefit,8 and the private promotion of EIC filing is big business.

o As currently administered by the IRS, the EIC program does not require that an illegal alien obtain a valid Social Security number to file his/her tax return and receive the EIC. If a return has an invalid number, a blank space or the code 205(c), the IRS simply assigns a temporary Taxpayer Identification Number (TIN) allowing the return to be processed and the checks to be mailed. This is consistent with the current IRS position regarding illegal alien EIC eligibility: since they are eligible, there is no need to verify information with the INS. This must change, ultimately.

7 In 1993, 8.7 million EIC filers filed paper returns, and 6 million filed electronically. GAO report at 3.

8 See, footnote 4.

146

Recommendations

As a general matter, Congress needs to restate the presumption that illegal aliens are ineligible for all Federal benefits unless specifically authorized by legislation that includes illegal aliens for receiving any Federal benefit, including the EIC.

• Congress must encourage the establishment of a national computer verifica- tion system to coordinate birth and death records of all 50 States to stop ille- gal aliens from fraudulently obtaining valid Social Security numbers and con- tinuing to receive the EIC. As I mentioned, this solution is supported by Bar- bara Jordan and her Commission on Immigration Reform. Currently, the IRS has a pilot program to verify electronically the SSN's of payees furnished to banks and other payors that pay interest or dividend. And the IRS denies the EIC to all filers who have an excessive amount of interest or dividend pay- ments in the relevant tax year. If the IRS can verify information of this type in the normal course of events, it can certainly obtain verification of U.S. citi- zenship, alienage status and Social Security numbers.

• Mr. Chairman, we prefer the legislation you introduced last year in the 103rd Congress (S. 2552) to the Administration's proposed plan. Your bill would have statutorily defined illegal aliens, many nonimmigrant and temporary aliens, as well as "non-work authorized aliens" as ineligible to receive the EIC. The Administration bills (H.R. 891 and S. 453) tie eligibility to "work authorization," a concept not defined in law until the Immigration Reform and Control Act of 1986. 9 The mechanism chosen by the Administration relies upon the ineligible alien being unable to obtain a Social Security number. Further, the Administration does not exclude those cases when an alien may have accumulated the earned income while in an "ineligible" work status, even though he/she may have a Social Security number — and work authoriza- tion— at the time the tax return is filed. The main flaw in the Administra- tion's proposal is that it fails to consider the vast number of ways in which an alien ineligible to work can still obtain a Social Security number, or the ways in which an alien may present a valid Social Security number on a tax

"" return that was only issued by SSA for work purposes for a temporary period of time. Without mandating cooperation between the INS, the IRS and the SSA in making eligibility determinations, the use of fraudulently-obtained valid Social Security numbers to obtain the EIC will mushroom. This means tfiat the Administration's bills would only deny the EIC to aliens who had not attained valid Social Security numbers, regardless of the means employed by the alien in obtaining those numbers, and regardless of the alien's actual im- migration status in fact at the time of the tax filing. Further, the Administra- tion bill would still permit a wage earner to claim a dependent (or more) who does not reside with the principal alien in the United States, nor does the Ad- ministration exclude wages withheld when the alien was working "out of sta- tus" from the determination that a now-work-authorized alien may qualify for the EIC. We believe all of the above areas need to be addressed to insure that illegal aliens do not look to the EIC as another welfare rip-off program fi- nanced on the backs of U.S. taxpayers.

• The IRS should be required to notify the INS of the reported addresses of EIC claimants who are either unable to provide valid Social Security numbers or — after verification with the SSA — cannot provide evidence of lawful residence. Unless aliens unlawfully present are convinced that tax filing will raise the possibility of inquiry by the INS the problem will continue. In that regard, we are disappointed that nothing the Administration's plan would improve co- operation between the INS and the IRS in the enforcement of Federal immi- gration laws. We hope you will consider mandating more institutional co- operation and data sharing between INS, IRS and SSA.

Mr. Chairman, thank you very much for this opportunity to appear before you to discuss this important public issue. I would be very happy to answer any questions you may have at this time.

100 Stat. 3359 (1986), Pub. L. No. 99-

147

The Checks Are in the Mail

ROY BECK

Washington, D.C.

ONCE again it is the season when tha Internal Revenue Service takes money sent in by Americans and mails some of it back to illegal ah'*"** as a land of end- of-the-year Donus through the Earned Income Credit program.

When IKS officials first confirmed this long-standing practice to ma last spring, they said they had no other option because Congress had never prohibited it. Senator William Roth re- quested a study by the General Ac- counting Office, which reported last fail that illegal aliens can receive di- rect cash payments of up to $2,528. Foreign narinnnls who are working il- legally in this country can jet the checks if they have a dependent child and ™"» less tbw 523,755 a year.

The GAO report was ignored or over- looked by the news media. But it caught the attention of then Treasury Secretary Lloyd Bentsen. who said the oracricB should cease. Tucked mside the Clinton Administration's latest ree- ^irmwiHarjnns on tax. policy is a provt- sian to stop the subsidy far illegal low- wage workers — but not until nest year. To assume that these bonuses will soon be ended is to npfiCT-^grirnar^ tha resilience of a pervasive system of in- centives and loopholes that the United States provides far citizens of other na-

It was pablic cynicism about the gov- ernment's good faith in ending this system that fueled the overwhelming passage of Califormas Proposition 187 last fall and spurs imitators today.

The Earned Tncr.mo Credit program was set up in 1375 as a work incentive for Americans with low-paying jobs. The credit sometimes works like most tax credits, reducing the tax owed. But the program often requires the IRS to send recipients a check for consider- ably more money than was withheld by their employers (if indeed any was withheld) the previous year. "We re- gard EIC as a form of welfare," says Mark Mullett, an side to Senator Roth.

The Fedei

make sure that illegal aliens don t miss out an these payments. David Simcox, a fellow at the Center for Immigration Studies, says federal funds are pro- vided to religious *nri immigrant-aid croups to persuade an^ assist immi- grants (whether legal or illegal) to file for the EIC checks. Publications for im- migration lawyers advise them that immigration acatus is irrelevant in £Q- ing for the benefit.

The IRS does not know for certain which applicants for SIC checks are il- legal aliens, but it haa a fairly good idea that they account for most of the applications ^ftcinjf valid Social Se- curity numbers. (Virtually every legal resident over the age of one year has a

both laws without getting caught, the IRS will 3end him a cash bonus.

Even if this practice is halted in a quick display of bipartisanship, at least three troubling issues remain:

L Resourceful illegal aliens can con- tinue to get the annual EIC bonus if they obtain valid Social Security cards by using frauaulent birth cercincaces. Dan Stem of the Federation for Amer- ican Immigration Reform suggests that the only lairing solution is to adopt a font of the proposal of Barbara Jordan «n^ her Immigration Reform Corn-

er. BtcA i» Wasmngmrt tdiior if m , taiy Social Contract and author of Charon? Americas Coutm. -

Social

Security spokeswoman Lynn Shiiler.i Forms without valid numbers are sent to the -TJnposiable Unit" at one of the ten IBS service cencers, where a bu- reaucrat assigns them temporary Tax- payer Identification Numbers, which look like Social Security numbers. That enables the IRS to keep its paper- work organized so thac it can proceed to send checks to filers who are proba- bly illegal aliens.

Federal law forbids anyone who is not a VS. citizen to enter the country without government approval and to scay longer than his visa allows. If a foreigner succeeds in violating the law, 1386 legislation makes it a crime for be hired Nonetheless, if a foreign worker succeeds in violating

R». that person !

verification system to coorttinate birth and death records of all fifty states.

2. Congress should consider chang- ing the presumption that ail govern- ment programs and benefits are in- tended to trrrpnr) to illegal aiiens unless otherwise specified. Congress might pass legislation that prohibits il- legal aliens from parririparing in any federal program or benefit unless spe- cifically included.

3. As usual, the comulauve costs

those from illegal. Simcox says his studies of IBS records intrirna some $250 million in ETC subsidies for ille- gal aliens in 1S90. But he found five times that am™mt going to legal immi- grants. 'EIC has become another case study in the baffling dilemma of oper- ating and funding complex income- transfer programs for poor residents, while the number of these residents is continuously being expanded by mfts>; illegal and legal immigration and refugee policies which import about half a million additional needy people each year." Simcox says.

If Congress does focus on he tax- payer-provided bonus checks for illegal aiiens. it should also consider a larger question: Why should the government continue to allow legal entry of hun- dreds of r^nnyanac 0f low-skilled for- eign workers whan U.S. taxpayers end up subsidizing them (and their employ- ers; because they cannot command high enough wages to pay the taxes that ■would cover their share of iaira- sSTicture and social services? Elim- inating future importanons of low- wage worxers would not only reduce EIC payments that otherwise would go to them but might also reduce EIC ex- penditures tot American laborers who. without compeanon from immigrants, would be more likely to earn non- poverty wages.

148

SUNDAY W0RLD-HE3ALD .MarraiZiSSS

Workers Get Valid Papers

By Illegal Means

BY JOY POWELL

When federal agents faded i Ne- braska meaipacking plant Mara 4, they found what ouibais sua b the biggest indication yet of an mown; proolem: illegal workers nans docu- ments; to daim (hey were bam in the United Stales.

Agents with the Imrrderanuc and Nararalizaoon Service ay the raid snowed a trend in which naey mdo- anneoted workers no longer are asm?

"steed, therare fnndulently oooonng ! birth certificates in other people's j mug ^nH using theiD to gd goron- ;

driver's licenses and Scdal Secsrity jams. -

Immigration azents said theatrest of" J33 woocers at (He Excel Cora, bee- "proeessing plant near Scaoyier. Neh_ was the most striking examoieai winch the birth-certificate scheme has been found. For the first ttae erermsnch a raid, agents found ao phony aon documents, only valid < thai were believed to hare . tamed fraudulendy.' - "We went from INS do

r for us to identify that is our own daotoe to birth certificates to get valid documents." said Paai Oimtm- jxa. a supervisory agent with ute INS Tin Oman*. "Thev're cresting t traii •which is more cifjicuit to follow. * '.'- Otriaals say ine scnene -wrc. for -the most part, because stales do not P.eamnrttoP^iCoLS

Valid Papers Obtained; Through Illegal Means:

CanrmuMd from Pa^t 1

automatically share information » one another on sucz thinas as birth :

"We're the first ones that hat seen it in a large-scale operacea like

Schuvlcr."

death ceruficates. There is no national When ucats raided the ifanfort reeisiry that would help states check Inc. plant in Grand Island a Sewea- informauon presented by workers bcr 1992. he said, they fooai (hat, when seeicmg acoimeats. ' nearly ail the 307 workers »o» were

The buih ceniCcaie fraud makes it arrested were using phony INS doca? virtually inrposnhle for employers to meats. The same has been larger one identify workers who do not have the for other raids ""ril the Seanvjer legal documents to work in the United operation, he said. ^_" ° I

States, a meatpacking comcany ' "As (hey discover how we tod oat • spokesman said. It greatly comoliotes about them, (hey change tberatshod immigration investigations, ah INS official said.

"The bottom line is that the raid illustrates in bold prat that we have an

Cansirnsea sad- illegai immigrants.

Inumgrahon arcats in Nebraska are. trying to track down vendos a the.

emerging security crisis based on the brrti-cerrifjcaie Macx mart* »nich,

appears to be based prm.inlvmTccs.

"TheV re rather easv to get." Joseph.

Lopes- Wilson, an Omaha mamey

law,.

who specializes in said of the birth ■—

arcane ano antiquated way we keep binh records." said Dan Stem, execu- tive director of the Federation for American Immigration Reform, a Washington. D.C-based group that

bills itself as working to get "lucsjiI Lots Barrios. 24, an

imnrigrancn under control." Merita who now lives in

"It has the capacity to sink all the Neb. said pnonv irimtifiranno carts-

other INS-initiated document reform and birth certificates commcofr are

plans." Stcn said. "Everything else hawked is (he streets and pso -ax

they are trying to curtail document bigger ones. such, as Los Anises.

fraud caa be undermined bvthis." Chicago and Dallas.

Birth ccruficaus are stolea or cous- Bamos. formary an Excel enoiev-

terrors can be bought at an average ee, said he believes- the cnxnpiny's.

cost of 5700. he said. The certificates management knew that wockets, was.

also can be obtained easily by people could bareiy speak pirlish had ass.

who impersonate others when writing been bom in the Umied States: £s

or visiting government viul statistics Excel a subsidiary of Cjre3 .in

offices, he said. Minneapolis, was not a ted tor. my.

Irnmieranon officers know that violations in the March 4 rati "He- some names are obtained from grave- cornpany cooperated with INS off: stones, aewsoaper obiuianes or county oils. -~ death reeoros. Chhstensen saut They of the 133 peopli suspect that other names stay belong to Q.S. ouzos who are unaware that their identities are bang used or to people who rent out tear binh certtri- caies.be said.

The fraudulent use of birth ccrzu.

cates began surfacing in resent years hired iDejsJ woreers.

when nacxing ptant officials noticed Under federal ana

that differrat anpiicants were using the laws, he said,

same idenctY. question documents mat appai

INS Wants Standardization :f

document itself or in the meshed by wnjch you obtain these documeats." he said Friday from Washington. — "To us. it's a major proolcn **— "p* it's i breeder document, is i-e-all il : . . . You can get legitunaie docasx by frauaulenuy obtauun: i eira (

Hi left the country -wan" laniv, and five annealed. Ton «cnr juvexules and were released. " •'-••'

Mark Klein, an Excel

)-l

The U.S. Imnugnuon and Natnral- izauon Serwce is pusnmg for lessla- tion that would standardize birth certificates and :her ^suance aanon- wide.

James Puleo. exesutive associate cornmisaonir of the LNS. said the fraudulent ise of birta ceruficaies by illegal intzrjgraats is eseaiaung.

"Tcere is :o standaraization on ie

Under such a svstern. would call aocticaefs Socai won eugTouicy.

149

PREPARED STATEMENT OF DANIEL B. GRUNBERG

Mr. Chairman and Members of the Committee, my name is Dan Grunberg and I am Vice President of Technology and a Director of Jackson Hewitt Tax Service. We appreciate the opportunity to present our experiences during the current tax season. This tax season Jackson Hewitt Tax Service operated 1,232 offices, both franchised and company owned, in 44 States and the District of Columbia. We will have prepared over 640.000 tax returns this year by April 30. We have been staunch supporters and promoters of the IRS' Electronic Filing program since the beginning. In fact, we have been awarded several IRS contracts to develop the test package that all Electronic Filing companies must pass to participate in the Electronic Filing program. Over 89 percent of the returns we prepare are electronically filed. We have offered free Electronic Filing in our offices since 1990.

The IRS has made tremendous strides in fighting fraud over the last four years. But this year, they humiliated and injured several million hard-working taxpayers who can least afford it. Let me tell you what happened in our offices.

On February 3, the IRS payments for refunds, made via electronic deposits to cus- tomer's bank accounts, were not made for the full refund. Twenty-eight percent of our customers only received the non-EIC portion of their refund. These partial pay- ments were made on returns that were already accepted by the IRS as having prop- er Social Security numbers and matching names.

With no warning, customers who received refunds quickly in years past and were counting on the money did not receive it. The following scenario was replayed ten thousand times in our offices; the numbers at all tax preparation offices must be in the millions. With no advance warning nor public information from the IRS, pre- parers were saying anything just to stay alive:

Yes, we did tell you your refund was $3,000, and yes, we did electroni- cally file it for you, and yes, the IRS only sent you $105, and no, we do not know why. No, I don't have your $2,895. Yes, I understand the IRS Tele-tax number says your refund was direct deposited on Friday. I don't know why it doesn't mention that it was only $105. Did you try calling the IRS? You didn't get an answer? They said your preparer is lying to you? You want proof that we don't have your money? They said we knew all about this?

During the next 3 days, we (and every other tax preparer) had a virtual revolt on our hands. Twenty-eight percent of the checks, or 45 percent of those with EIC, were for amounts far less than people were expecting, and the only proof we coui'd offer that we weren't stealing their money was a computer printout that we had hastily assembled for them. In numerous cases, police had to be called to our offices to calm taxpayers and even forestall riots.

The Associated Press reported that the Government Accounting Office did a spot check of IRS taxpayer assistance calls, and found that only 13 percent of the calls were answered during the period Jan. 30 to Feb. 10. When our customers did man- age to get through, they were told such things as (1) they did not receive their EIC because of errors their tax preparers made, or (2) that their tax preparers had their money and they should go get it from them.

We have over 500 franchisees and everyone of them has horror stories to tell. The following are just a small sample of the cases reported in our 1,232 offices:

1. A couple from our Hattiesburg, Mississippi office had their power turned off and were out of money for food. Our franchisee filled out the IRS hardship form for them and the local IRS office told them it wouldn't help them.

2. A customer at our Vallejo, California office needed her money to pay for repairs to her car. When she couldn't pay, the repair shop took her car.

3. Numerous customers have been evicted because they were behind on their rent and they had told their landlords that the money was coming to pay them. It did not come.

Keep in mind that our large stack of stories represents only 1,232 electronic re- turn originators, or EROs. This is less than 5 percent of the total EROs in the coun- try, who all have similar stories to tell.

Our customers who electronically file and claim the Earned Income Credit have an average income of $11,600; most of them have children. This is below the poverty line. Many of these are women and minorities who have jobs and are using the EIC to stay afloat during a particularly bad time in their lives. To someone who has a weekly take home pay of less than $200, the average EIC payment of $1,500 is al- most 8 weeks wages.

150

I assure you we did not know about this. If the industry had known that this would happen, the banks would not have lent money on EIC. It is our understand- ing that the participating banks are owed over 190 million dollars because of this failure to pay by the IRS. Our company prides itself on "Winning Customers for Life." If we had the opportunity, we would have told every customer applying for EIC what was going to happen.

The IRS did make an announcement in February, in a mailing to tax preparers, after many thousands of customers and tax preparers had already been taken by surprise. We are now being told that, if the IRS pays it at all, it will be 8 weeks before a check will be mailed to the taxpayer's home. It has been more than 8 weeks and we have heard from very few people who have gotten their EIC checks yet.

Even if we as tax preparers knew about the delays, should we be the ones to break the bad news to 7 million Americans? The IRS did not adequately inform tax preparers and the public about the massive changes that were about to take place in February.

Many people choose Electronic Filing and have their money direct deposited be- cause receiving checks in the mail is unsafe in their neighborhood. Paper checks also have a much higher incidence of fraud, both by fraudulent endorsement and by fraudulent claims of stolen checks. The IRS is now delivering the EIC checks to taxpayers' homes.

If a taxpayer has been evicted from their apartment, how are they going to receive a questionnaire or check in the mail? Treasury check envelopes say "Do Not For- ward."

Most taxpayers getting the EIC have been getting it year after year. They expect and count on this money to survive. The IRS could have mailed questionnaires, done name/Social Security matches and cross references, or performed audits during the summer of 1994. This would have not disrupted so many lives as the methods they choose.

The IRS 1040 Package, mailed to the public in late December, and early January says:

If you file a complete and accurate return, your refund will be issued within 21 days*. You can also get the convenience and safety of direct de- posit.

*Some refunds may be temporarily delayed as a result of compliance re- views to ensure that the returns are accurate.

Perhaps seven million taxpayers deserve more than a footnote. Fighting Fraud

We have worked closely with the IRS to prevent fraud. Previous to this tax year, we were catching hundreds in our offices after the IRS accepted the returns (NBC Dateline aired a spot about how one of our franchisees stopped a fraud ring at Hampton University). We utilized a number of pattern detection algorithms to help us in this effort. Now the IRS has fraud under control and we are seeing far fewer cases. With the name and SSN matches (see Appendix B) implemented by the IRS, it is exceedingly difficult to commit any type of organized widespread fraud. The IRS catches dependents being claimed on multiple returns up front; we, as the preparer, then assist the IRS and taxpayer in any way that we can in straightening out the discrepancy.

The IRS says "they stopped fraud" due to delaying EIC payments. They did not. All they did was not pay it out quickly. There has been talk about the IRS mailing questionnaires (Form 9598) to taxpayers before paying the EIC. This questionnaire is yet another new form from the IRS. How many people will be able to fill it out accurately — no instruction books have been written for this and tax preparers have never seen them before. We predict that significant numbers will not be returned due to fear or ignorance, not fraud.

We understand the need for some secrecy in the battle against fraud. However, this was not a case of a few percent of people being audited — this was a massive campaign to punish 45 percent of the people because of a few fraudulent filers.

The Promotion of Electronic Filing

The IRS, Jackson Hewitt and all the other preparers who helped the IRS promote EF over the years have sold it on the basis of:

1. It is painless, easy and faster than paper filing, and

2. You will be told promptly when to expect your refund, how much will be paid, and how it will be paid.

151

This year, this was taken away from the EF program. We fear that taxpayers will be disgusted with it and have no incentive to electronically file. We and other tax preparers have done everything that the IRS has asked of us. We promoted Elec- tronic Filing by making it free. We checked IDs. We did extra paperwork. We sub- mitted to fingerprinting and credit checks.

Many taxpayers have told us that they will just file a paper return next year. They say "Why file electronically if it is going to cause this much trouble?" People are very reluctant to do anything to bring them under the scrutiny of the IRS. Even when we tell them that paper returns will have the same screens placed on them, they do not believe it. We beiieve that there will be another drop in the number of electronically filed tax returns next year if strong, immediate action is not taken.

Recommendations

We believe that the following action should be taken:

1. Restore the DDL This will increase confidence in the Electronic Filing system for millions of taxpayers. The Treasury's own testimony 1 has been that electroni- cally filed returns provide more information and make it easier to detect fraud. Peo-

Ele will know if, when, and how their refunds will be delivered. The IRS should then onor the DDI once it has been given. Direct Deposited refunds are far less likely to be lost or stolen than paper checks.

2. Additional information required to document the EIC should be asked for at the time the original return is filed. There is already an EIC form that must be filled out. Consider the huge savings in time and effort on the part of the IRS, the tax preparers and the public in determining up front the amount for which tax- payers qualify, rather than after an 8-week delay.

Summary

The change in IRS procedures may have been well warranted but the methods used brought hardship to many hard-working taxpayers. We believe the IRS needs Electronic Filing and that it can reduce costs and combat fraud at the same time. It should be a partnership between the IRS and private industry, not an antagonis- tic relationship. We would be happy to work with the IRS in any way possible to find a solution to the current situation.

Appendix A: Name and SSN Mismatches

As the season started the IRS rejected about 20 percent of the Electronic Returns that we were transmitting to them. In past years, this number was less than 2 per- cent. The main reasons for the rejected returns was that a Social Security number on the return did not match a last name on file with the IRS (from the Social Secu- rity Administration's files), or that a birth year of a child qualifying the taxpayer for the Earned Income Credit did not match the child's Social Security number.

We believe that this new SSN/name checking was a necessary step for the IRS to take to eliminate fraud. We wish to make a clear distinction between these re- jected tax returns issue and the EIC non-payment issue. Once the return was ac- cepted by the IRS, there is no more question as to the validity of SSNs or the exist- ence of dependents. Some news stories have incorrectly reported that SSN problems were delaying EIC refunds.

Appendix B: IRS Advancements In Fighting Fraud

The IRS has made numerous improvements in the Electronic Filing program since its inception. The following is a partial list of changes that have reduced the inci- dence of fraud:

1. Requiring the SSN and last name to match IRS Master file information for primary taxpayers.

2. Requiring the SSN and last name to match IRS Master file information for secondary taxpayers.

3. Requiring the SSN and last name to match IRS Master file for all de- pendents.

4. Requiring matching SSN and year of birth for EIC qualifying children.

5. Requiring all SSNs on a return be filled in (no "APPLIED FOR", etc.)

6. Allowing only a single return to claim any dependent.

7. Requiring Employer ID numbers to match the IRS Master file.

8. Allowing preparers to report suspicious returns for investigation prior to refund payment.

1 Statement of Ronald K. Noble before the U.S. House of Representatives, Committee on Ways and Means, October 6, 1994, p30.

152

PREPARED STATEMENT OF FINN N.W. CASPERSEN

Mr. Chairman and Members of the Committee, I am Finn N.W. Caspersen, Chair- man and Chief Executive Officer of Beneficial Corporation. I appreciate the Commit- tee's invitation to testify today on the Earned Income Tax Credit (EITC) and fraud in the electronic tax return program.

My testimony will focus on electronic filing fraud, the Earned Income Tax Credit (EITC), the impact of Internal Revenue Service (IRS) changes on taxpayers, particu- larly those eligible for the EITC, and the electronic filing industry.

Each of these topics is of great importance to Beneficial Corporation, since one of our subsidiaries, Beneficial National Bank, is the leading issuer of Refund Antici- pation Loans (RALs). RALs are loans made available to taxpayers who file electroni- cally and are eligible for a tax refund. Beneficial National Bank, as a RAL lender, is just one company in the electronic filing industry, which includes other lending institutions, tax preparers, electronic return originators (EROs), electronic return transmitters, computer software developers and distributors, accountants, and oth- ers..

Allow me to explain briefly how the RAL program works. A participating bank, regulated by State or Federal banking authorities, makes a loan to the taxpayer in the amount of the taxpayer's anticipated tax refund, minus a service fee. This serv- ice fee is, by law, disclosed in terms of an annual percentage rate (APR) and each customer is required to complete a formal bank loan application and receives all ap- propriate disclosures. The loan is repaid to the lender when the tax refund is trans- mitted by the government to an account established for the taxpayer at the lending bank.

In the 8 years since its inception, the RAL program has proven to be extremely popular with taxpayers. I believe we are clearly filling an important consumer need by offering this product. In a recent survey by the Roper Organization of 500 ran- domly selected RAL customers, 84 percent indicated that they would use the service again, and 88 percent of repeat users and 80 percent of first-time users felt the RAL program was a "good value." It is important to note that, until this year, the RAL lenders consistently, over the last 7 years, reduced the RAL service fee. Prior to this tax season's developments, Beneficial National Bank had expected to continue with its reduced price for RALs.

In 1994, approximately 9.2 million taxpayers received RALs. Of those, Beneficial alone made 2.7 million RALs — the largest single RAL provider in the industry. The average family income of a RAL customer is $25,000-$30,000. Many hardworking taxpayers have come to rely on early availability of their tax refund. Many use their refund to pay off outstanding, often overdue, bills or meet their day-to-day ex- penses— rent, health care costs, food, and so on. And, as one would expect, a signifi- cant percentage of RAL users claim the EITC in their tax filings. Without a RAL, many could not even afford to have their tax return prepared.

Our long and successful experience with the RAL program and its millions of cus- tomers has given Beneficial important insight into the issues that are before this Committee today.

The Electronic Filing Process and Fraud

The IRS has made great strides in its efforts to deter fraud. This tax season, the IRS has taken a number of actions in its effort to reduce fraud, some of which have had dramatic effects on taxpayers. Today I will focus my attention on those actions taken which affected taxpayers who applied for RALs and/or who filed for the EITC.

As the Committee members may know, Beneficial and the electronic filing indus- try have not always agreed with the IRS with respect to these actions. Communica- tion between the IRS and the industry has been a very serious problem. Earlier this year, in fact, Beneficial National Bank and Beneficial Tax Masters, Inc., another subsidiary of Beneficial Corporation, initiated a lawsuit asking that the IRS be forced to honor the instruction of taxpayers to direct deposit their refunds to their designated bank. The lawsuit was terminated, however, when it became apparent through discovery that no solution seemed viable within the time available to rectify the situation for the current tax season. The IRS, moreover, has reassured Bene- ficial that it will make appropriate and necessary programming changes for the next tax season to avoid a recurrence of the difficulties experienced this year, and we have every expectation that the IRS will proceed in good faith. Likewise, we are also hopeful that the communication problems that contributed to the filing of the law- suit are now behind us.

Proceeding in good faith to forge a meaningful public-private partnership is, in our view, the key to effective fraud detection in electronic filing. We in the industry have advocated a public-private partnership and substantive cooperation for some

153

years now, and firmly believe that such a partnership will provide benefits for the Service, the industry, and taxpayers.

Fraud is a legitimate concern to — and a high priority for — both the IRS and the electronic filing industry. The elimination of fraud in the electronic filing industry is in the interest of the IRS, the companies that comprise the electronic filing indus- try, including Beneficial, as well as millions of legitimate taxpayers. After all, it is the legitimate taxpayers who are penalized for fraud through delays in their re- funds, higher prices and reduced availability of RALs, and, ultimately, increased taxes.

The electronic filing industry has a strong, vested interest in reducing fraud. The industry, similar to government, is exposed to losses resulting from fraudulent re- turns, and it has worked since its inception to prevent such fraud. Clearly on the issue of fraud prevention, the interests of the IRS and the electronic filing industry are closely aligned. The industry is in a unique position to provide real-time, upfront filters to prevent fraud and thus complements the IRS' proven audit expertise.

This alignment of interests is furthered by the role that Refund Anticipation Loans play in encouraging electronic filing. For the IRS, the electronic filing pro- gram is a key component of its Tax System Modernization (TSM) program. Elec- tronic filing of tax returns, which began in 1986, is the centerpiece of the IRS' multi- billion dollar TSM program to modernize the processing of tax returns by eliminat- ing paper and substituting electronic information in its place. Last year, some 14 million tax returns were filed electronically, a key contributor to the IRS' goal to save money by eliminating the costs associated with manually processing paper re- turns. In the future, more sophisticated computer systems will enable the IRS to identify and avoid fraudulent tax returns more efficiently and expediently.

RALS, in turn, are a major force behind electronic filing. In 1994, approximately 70% of the taxpayers who filed a return electronically also applied for a RAL from the four major RAL lenders, which, at the time, included BancOne, Beneficial, Greenwood Trust, and Mellon Bank. However, due to events of the past year, two of these four lenders, Greenwood and Mellon Bank, have already discontinued their RAL programs.

For its part, the electronic filing industry has undertaken major initiatives to identify and combat fraud in electronic filing up-front — before a refund is issued. The fraud detection and prevention efforts of the electronic filing industry contribute needed assistance to the IRS by providing additional screening mechanisms to de- tect fraud before returns are processed and refunds paid.

Members of the electronic filing industry have designed or implemented several upfront checks and balances to prevent fraudulent returns from ever reaching the IRS. These procedures include, but are not limited to, detecting and rejecting re- turns with W-2s that appear suspicious, checking each taxpayer's documentation upon application to file electronically, and verifying that withholding claims match the number of dependents claimed.

To give the Committee a clear understanding of the sophisticated fraud preven- tion measures that the industry has inserted at various points in the electronic fil- ing process, allow me to explain briefly how that process works. The basic electronic filing process starts when a taxpayer visits an ERO. The tax return is either pre- pared or entered into an electronic filing format by the ERO, who then completes IRS Form 8453, the taxpayer's authorization s.nd request to have the tax return filed electronically, which the taxpayer signs.

The screening process begins at this stage. While the IRS can only recommend that the ERO obtain one form of identification that has a photo and current address of the taxpayer, the RAL lender requires this identification to be viewed and re- corded on the RAL application when a taxpayer requests a RAL. The EROs are also on the lookout for suspicious electronic filing transactions, which they report to the local IRS criminal investigation unit. In addition, if the return contains a substitute W-2, the return cannot be filed electronically until after February 15 of that year. Otherwise, if everything appears to be valid, the return is sent either directly to the IRS by the ERO or to an approved transmitter, where the return is bundled with those of other EROs and sent to the IRS via modem. An additional check occurs as the software used by the EROs and transmitters contains error checking diagnostics which will highlight returns that are incorrect.

Simultaneously with the filing of the return to the IRS, if a RAL or refund antici- pation check (RAC) is requested, the tax return is sent by the ERO or transmitter to the industry's Fraud Service Bureau (FSB). The electronic filing industry has spent over $6 million in just three years to establish the FSB, which is a sophisti- cated computer analysis system that uses artificial intelligence to help identify and reduce fraud. The FSB screening process includes checks for duplicate Social Secu- rity numbers and duplicate filings as well as complex models based on known fraud

154

cases from the past seasons' tax filings. RAL applications and tax returns that are deemed potentially fraudulent by the FSB are forwarded to the appropriate IRS service center and are denied a RAL. It should be noted that the FSB has processed 27 million returns to date.

Upon IRS receipt of the return, it is processed through the tax system. If it is accepted for filing, an acknowledgement is provided electronically to the ERO trans- mitter. If the return contains math or formatting errors, or the names and Social Security numbers cannot be matched against IRS and Social Security Administra- tion files, a rejection notice is sent electronically to the ERO or transmitter. If the return is not Questionable and proceeds in a normal electronic format, the taxpayer would generally receive a tax refund in four weeks by paper check or two to three weeks by direct deposit.

Clearly, the industry has played a significant part in putting in place a series of steps to reduce fraud in connection with the RAL program.

EITC and Fraud in Tax Filing

The EITC program presents the IRS and the electronic filing industry with nu- merous challenges, and fraud is a significant problem. While the past three adminis- trations and Congress have agreed that the EITC is an efficient means of transfer- ring benefits, the administration of this benefit through the tax system has unfortu- nately opened avenues for fraud for unscrupulous taxpayers. Apparently, the EITC program, which was designed to assist low-income working citizens, is currently used by some taxpayers to defraud the IRS. Fraudulent EITC returns are filed both on paper as well as electronically. Experience suggests that if the EITC fraud were eliminated in electronically filed returns, those seeking to use EITC with fraudulent intent would simply move to paper filing. Recent changes have increased both the scope of the EITC program and the amount of credit that can be claimed, leading to even higher levels of fraud in the tax system.

There are several ways in which taxpayers have defrauded the government through the EITC program. Some methods are blatantly fraudulent, such as inten- tionally over-claiming the credit that is actually due, understating income earned, claiming dependents that do not qualify the taxpayer for the EITC, or claiming in- come which was never received.

However, there are also some compliance issues that lead to erroneous EITCs. In March 1994, the Treasury commissioned a Task Force to review the problem of tax refund fraud in electronic filing. The Task Force identified EITC non-compliance as a problem area. The EITC forms are difficult to follow and understand, and may therefore lead to unintentional claims for too much EITC. For example, one section of the Tax Code requires a dependent to live with the claimant more than six months to qualify for dependent status, while in another, the minimum length of residency is twelve months. It is also possible for a taxpayer to have a legitimate dependent deduction, but the dependent may not necessarily qualify the taxpayer for an EITC.

These compliance questions do not necessarily lead to intentional fraud on the part of the taxpayer. In fact, a 1994 study by the IRS estimated that almost 90 per- cent of the people who claim the EITC intend to comply with the program. Nonethe- less, these erroneous claims have led to the Treasury paying refunds for which the taxpayer is not legitimately entitled. This, along with 10 percent of EITC claims which are intentionally fraudulent, must be detected and avoided by the electronic filing industry and the IRS.

Impact of the IRS Decisions on Taxpayers and the Electronic Filing Industry

As I noted at the outset, the IRS has taken a number of constructive steps over the past year aimed at combating fraud. A number of these steps were suggested by the electronic filing industry. These include requiring taxpayers to provide tax- payer identification numbers for all children claimed for EITC purposes, and deny- ing the EITC to undocumented workers. In addition, the IRS recently imposed rules for EROs which require new EROs to submit to credit checks and fingerprinting. These new rules were put in place to address the IRS' valid concern about its past problems with some EROs who had fraudulent intent. The industry fully supports the IRS' decision to have stricter verification procedures for new EROs and goes even further to suggest that the IRS fingerprint all EROs and tax preparers.

Effective this filing season, the IRS made an additional two major changes in the tax return process which have had a significant effect on millions of taxpayers — namely, eliminating the Direct Deposit Indicator (DDI)1 and, in the case of hun-

^he DDI was an acknowledgement to the taxpayer, based on information in existing IRS files (Debtor Master File), that the taxpayer had no outstanding government debts in such areas

155

dreds of thousands of taxpayers, delaying the payment of the EITC portion of a tax- payer's refund. These actions, well-intentioned though they may have been, did cre- ate serious disruptions for millions of taxpayers. I believe such disruptions can be avoided in the future if the IRS works in concert with the electronic filing industry in its efforts to combat fraud.

Changes implemented by the IRS have also affected RAL lenders like Beneficial. Due to the IRS changes, the banks credit approval criteria for RALs drastically changed this tax season. This resulted in fewer customers being approved for RALs and created serious problems for millions of taxpayers who expected; but were de- nied, full RALs. In addition, we were forced to raise the price for RALs to cover in- creased risk and costs resulting from the IRS' decision to no longer provide lenders with critical information regarding existing federal liens which the refund might be used to repay. This information is almost exclusively within the government's knowledge. Therefore, the banks must assume a greater level of risk. In effect, hon- est taxpayers were forced to pay for the liens of defaulting taxpayers. Almost ten million taxpayers were satisfied with the RAL product last year and were perplexed, disappointed, and angry when it became less available and more expensive this year.

Taxpayers and their designated banks should have the right to information about any federal tax liability that may be levied against a claimed refund. Timoly provi- sion of this information by the IRS would allow taxpayers to borrow against their refunds at reasonable prices as in the past, or simply to be assured these refunds have been deposited in accounts at the designated financial institution as promised in their agreements with the IRS.

In addition to eliminating the DDI, the IRS has also delayed, by as much as 8 weeks or more, the EITC portion of countless taxpayers' refunds. These delays were in addition to the ones caused by problems in the IRS-Social Security Administra- tion computer matching program.

The IRS attributed its decision not to send the EITC portion of a refund directly to the RAL bank designated by the taxpayer to a computer or data processing prob- lem. As a result, lending institutions have been forced to stop granting RALs to many taxpayers that receive the EITC — those people most in need of RALs.

This reversal by the IRS has delayed by as much as two months a sizable portion of low income taxpayers' refunds. Large numbers of EITC recipients file electroni- cally and apply for RALs in order to receive their refund money in two to three days. They do so because they need or want their refunds quickly to meet urgent financial obligations, such as paying overdue rent or utility bills. Because of the new delay policy of the IRS, many hard-working, low-income taxpayers, who legitimately qualify for an EITC, now find themselves unable to receive their money in the time they expected. Additionally, RAL banks were financially penalized by this action. At Beneficial over $300 million is now at risk because of the IRS decision.

Since the magnitude of withheld refunds was not anticipated by taxpayers or the electronic filing industry, tax practitioners were unable to adequately warn cus- tomers of the probability that their refund money would be delayed. This IRS change caught both taxpayers and the electronic filing industry by surprise and caused confusion for the millions of legitimate taxpayers and EITC claimants who can least afford an eight week or more delay in receiving their refund.

Cooperation Between the IRS and the Electronic Filing Industry is the Answer

The problems the IRS, taxpayers and the electronic filing industry faced this tax season, and the subsequent Beneficial lawsuit against the IRS, could have been averted had the IRS and the electronic filing industry worked cooperatively and shared information within the limits of confidentiality laws.

Apart from a general commitment on the part of the IRS and the industry to work together going forward, there are a number of specific cooperative initiatives that we would like to see undertaken to address fraud concerns without harming hard- working, law-abiding taxpayers. For example, by instituting more stringent controls at both the industry and IRS levels, fraud can be more effectively detected and avoided up-front, without hurting taxpayers whose returns are in no way suspicious, but simply fall into categories targeted for fraud checks. In addition, going forward, if the IRS decides for any reason that a tax refund claim within a specific profile should be delayed for review, the taxpayer should be notified of this delay promptly. In the case of electronic filing, this should be done as part of the IRS' acknowledge- ment process. With the up-front knowledge of this potential delay, the taxpayer will be able to make informed decisions and will not make financial commitments that

as unpaid income taxes, arrearage in child support payments, and defaults in government guar- anteed loans.

156

cannot be kept. In the case of paper returns, a notice should be mailed in a timely manner. By following this procedure, the taxpayer and the electronic filing industry would immediately be aware of any delay, and some of the problems encountered this year could be substantially alleviated.

We believe there are far more effective ways to deal with electronic filing in rela- tion to EITC fraud and non-compliance. However, in order for a total fraud preven- tion plan — or revenue protection program as the IRS now calls it — to be effective, the responsibility must be shared by the IRS and the electronic filing industry. The industry can effectively act as a gateway to the tax system and initial checkpoint for fraud, at least for the approximately 45 million returns that are filed through tax preparers and the 13 million returns that were electronically filed in 1994. The EROs, electronic return transmitters, lending banks, and, above all, the industry- funded FSB can and must act as the up-front screening system for fraud prevention in electronic filing.

The IRS recognizes that tax fraud is much easier to detect in electronic filing than in paper returns, and, therefore, the growth of electronic filing is an important goal. In addition, IRS officials estimate a cost savings of $1.52 per return in the process- ing of electronic returns versus paper returns. Continued growth of electronic filing can only be advanced with greater cooperation between the Service and the indus- try. Failing that, electronic filing will continue to decline in popularity. This tax sea- son is a case in point. Given preliminary numbers to date, we estimate that this year there will be at least a 25 percent reduction in the number of returns filed elec- tronically versus last year, which will surely be a blow to the IRS' plans for a viable TSM program.

But the real impact will be felt next year when many disillusioned taxpayers will no longer choose to utilize electronic filing. In addition, we know of many tax pre- parers and electronic return originators who will not offer electronic filing services next year due to the problems of this past tax filing season.

A Public-Private Partnership to Control and Combat Fraud

In attempting to better detect, deter and combat fraud in electronic filing, it makes the most sense to form a public and private partnership between the IRS and the electronic filing industry; such a partnership is in everyone's best interest. Both parties must be willing to accept the responsibility associated with such a partnership and be held accountable. In the past, the IRS did work with the indus- try, based on its belief that the industry could help increase the use of electronic filing. If the immediate goal is now fraud prevention and detection, cooperation on this front can be as effective as that employed to promote electronic filing.

As I have repeatedly stated today, the electronic filing industry shares the goal of eliminating fraud; we have demonstrated that we can be an effective force in the fight against electronic filing fraud; and, we are willing to do more to press this fight against fraud. Increased electronic filing and fraud control are not, and should not be treated as, mutually exclusive. As the private partner, we share this goal with the IRS because our financial risk decreases directly with any reduction in fraud.

I suggest that the sharing of taxpayer information — within IRS regulations — with the industry will better enable it to work with the Service to prevent fraud. Unfortu- nately, in the past, the IRS has often used Tax Code laws and IRS regulations as the basis for not sharing information with the industry. However, there is currently sufficient flexibility in those laws and regulations to allow the IRS to share informa- tion with the industry if doing so enhances the IRS' ability to combat fraud. All that is needed is a strong directive from the Congress to the IRS that such cooperation advances the public interest. Thus, the IRS can and should provide the industry sponsored FSB with a data file of characteristics associated with returns the IRS has determined to be fraudulent. With this data, the FSB could build more effective detection models for use during the tax season.

Since this file would consist of raw characteristics with no taxpayer identification and would be able to assist, not impede, the enforcement of IRS laws, there should be no conflict with existing privacy laws and regulations. For example, although sec- tions 6103 (a) and (b) of the 1986 Tax Code require the IRS to keep all tax return information confidential, such confidentiality need not be maintained to the extent any such information does not identify a particular taxpayer. The one exception is that such non identifying information may be kept confidential if it is used to de- velop standards for selecting returns for examination. However, even in those cases the Secretary of the Treasury may disclose such information if such disclosure would not seriously impair the enforcement of Internal Revenue laws. Clearly, no such impairment can be found where the electronic filing industry is aligned with the IRS in a tightly controlled effort to assist the IRS in reducing tax fraud. More-

157

■-"■.'■ e 6103 (c) also permits the disclosure of return information to a designated third

;-i!r' if, e.g. RAL lenders, if such disclosure is specifically authorized by the taxpayer,

:"e- ii is currently the case in the RAL program by means of express language in

etAL application. The only limitation on such disclosures occurs upon a finding

e Secretary that such disclosures would impair a civil or criminal investigation.

) nt such a determination, taxpayers may demand that IRS disclose such infor-

33n to designated third parties. However, missing from current law is a mandate

t the IRS provide such information electronically to the designated third party

-'■ ■'-'■ .1in an expedited time-frame, e.g., 24 to 48 hours, as was previously done with

:- (DDI information until the DDI was discontinued. Therefore, the 1986 Tax Code

- K Id be amended to require IRS to provide DDI-type information or other tax re-

'■••-■■ .] information as directed by the taxpayer within the previously achievable time-

:: e.

'. April 1994, an industry grade association, the Electronic Filing Coalition of ■•'•-: :"i.". erica2 (EFCOA), formed a task force comprised of industry experts, tax special- '-'■'■ l'& tand enforcement professionals. The task force explored comprehensive measures ;';--; i irther detect and prevent fraud in the electronic filing program and shared its ■■'■'■-'-i ><mmendations with the IRS and appropriate committees of Congress in March ' w& g'l. The task force's analyses resulted in the publication of A Public-Private Part- iisa- ihip to Prevent Fraud in Electronic Filing, (attached) in which several rec- ::i; tte mendations are made to both the electronic filing industry and the IRS to enable '-•Melee- jjn to work together jointly to combat fraud in electronic filing. Several of these e mmendations have already been adopted by both the IRS and the industry. (See ^endix II, for the complete Task Force study.)

a the study, the recommendations were divided into initiatives for the electronic i ig industry and for the IRS. or the industry, the Task Force recommended:

: • expansion of its existing fraud prevention program to include informing cus- tomers, upon initial contact, of the criminal and civil penalties of tax fraud;

• several potential improvements to the existing FSB, including: — coordinating efforts with the IRS;

I — improving data flow,

— expanding base-line information such as fraud profiles and demographics.

'\>r the IRS, the Task Force recommended:

i • requiring more stringent qualifications for the assignment of electronic filing identification numbers;

• aggressively enforcing penalties for violations of electronic filing rules; 1 • imposing more comprehensive controls on the procedures it follows in process- ing returns;

• utilizing detailed, current information which employers already provide about their employees when filing their quarterly employment tax returns; requiring document verification for EITC claims on tax returns prior to the processing of these returns.

EFCOA's Task Force stated its belief that the improvement in data flow and ver- :ation that would result with the implementation of its recommendations would i able both the industry and IRS to combat fraud in electronic filing to their mutual nefit.

.All that Since completion of the Task Force's study, we have continued to evaluate the cooperation lud detection processes already in place and subsequently have developed addi- - industry j>nal suggestions to detect and deter EITC-related and other potential fraud: itheltiS • effective

•action in

should

• .:.->ec-

ae extent

• Issuance of an authorization form to the taxpayer in their forms booklet with a code that indicates the taxpayer met all the requirements of EITC eligibility in the prior year and will be eligible again if the income is within the qualify- ing range and the children remain the same. The form would be filed with the tax return.

• Placement of an EITC eligibility indicator on the mailing label for tax re- turns.

• IRS utilization of the enormous infrastructure the electronic filing industry already has in place. The "private" electronic filing industry owns and oper- ates every process and procedure guiding electronic filing up until the IRS

! EFCOA includes the major institutions that offer refund anticipation

■m^ua inciuo.es me major institutions tnat oner reruna anticipation loans, tax preparers, T«j'wth !'ectroruc return originators, and electronic return transmitters. The coalition was formed in 992 to address areas of shared concern among which is tax fraud prevention.

90-040 0-96-6

.

156

cannot be kept. In the case of paper returns, a notice should be mailed in a tir ly manner. By following this procedure, the taxpayer and the electronic filing indu ry would immediately be aware of any delay, and some of the problems encount* this year could be substantially alleviated.

We believe there are far more effective ways to deal with electronic filing in i tion to EITC fraud and non-compliance. However, in order for a total fraud pre tion plan — or revenue protection program as the IRS now calls it — to be effec the responsibility must be shared by the IRS and the electronic filing industry, industry can effectively act as a gateway to the tax system and initial checkp for fraud, at least for the approximately 45 million returns that are filed thrc tax preparers and the 13 million returns that were electronically filed in 1994. EROs, electronic return transmitters, lending banks, and, above all, the indu: funded FSB can and must act as the up-front screening system for fraud prever in electronic filing.

The IRS recognizes that tax fraud is much easier to detect in electronic filing in paper returns, and, therefore, the growth of electronic filing is an important j In addition, IRS officials estimate a cost savings of $1.52 per return in the pro< ing of electronic returns versus paper returns. Continued growth of electronic f can only be advanced with greater cooperation between the Service and the in try. Failing that, electronic filing will continue to decline in popularity. This tax son is a case in point. Given preliminary numbers to date, we estimate that year there will be at least a 25 percent reduction in the number of returns filed tronically versus last year, which will surely be a blow to the IRS' plans for a v TSM program.

But the real impact will be felt next year when many disillusioned taxpayers no longer choose to utilize electronic filing. In addition, we know of many tax be parers and electronic return originators who will not offer electronic filing ser es next year due to the problems of this past tax filing season.

A Public-Private Partnership to Control and Combat Fraud

In attempting to better detect, deter and combat fraud in electronic filir lit makes the most sense to form a public and private partnership between the IS and the electronic filing industry; such a partnership is in everyone's best int< Bt. Both parties must be willing to accept the responsibility associated with sc partnership and be held accountable. In the past, the IRS did work with the ii try, based on its belief that the industry could help increase the use of elect hie filing. If the immediate goal is now fraud prevention and detection, cooperate on this front can be as effective as that employed to promote electronic filing.

As I have repeatedly stated today, the electronic filing industry shares the >al of eliminating fraud; we have demonstrated that we can be an effective force i he fight against electronic filing fraud; and, we are willing to do more to press lis fight against fraud. Increased electronic filing and fraud control are not, and si lid not be treated as, mutually exclusive. As the private partner, we share this )ai with the IRS because our financial risk decreases directly with any reductii in fraud.

I suggest that the sharing of taxpayer information — within IRS regulations— ith the industry will better enable it to work with the Service to prevent fraud. Un; tu- nately, in the past, the IRS has often used Tax Code laws and IRS regulatio as the basis for not sharing information with the industry. However, there is curr tly sufficient flexibility in those laws and regulations to allow the IRS to share inft na- tion with the industry if doing so enhances the IRS' ability to combat fraud. Al! lat is needed is a strong directive from the Congress to the IRS that such cooper ion advances the public interest. Thus, the IRS can and should provide the ind try sponsored FSB with a data file of characteristics associated with returns tht RS has determined to be fraudulent. With this data, the FSB could build more effc ive detection models for use during the tax season.

Since this file would consist of raw characteristics with no taxpayer identific ion and would be able to assist, not impede, the enforcement of IRS laws, there si ulri be no conflict with existing privacy laws and regulations. For example, althougl lec- tions 6103 (a) and (b) of the 1986 Tax Code require the IRS to keep all tax r irn information confidential, such confidentiality need not be maintained to the e ent any such information does not identify a particular taxpayer. The one excepti i is that such non identifying information may be kept confidential if it is used t de- velop standards for selecting returns for examination. However, even in those ses the Secretary of the Treasury may disclose such information if such disci ure would not seriously impair the enforcement of Internal Revenue laws. Clear! no such impairment can be found where the electronic filing industry is aligned ith the IRS in a tightly controlled effort to assist the IRS in reducing tax fraud. 1 >re-

157

clns M also permits the disclosure of return information to a designated third ,er> 6103 (c) also per dlscl0Sure is specifically authorized by the taxpayer,

»rty. eg- ^Tntrv the 'case in the RAL program by means of express language in hich is aiTienUyi net limitation on such disclosures occurs upon a finding

,e ^ a5,W Sat such disclosures would impair a civil or criminal investigation. ■•theSeCrrf determination, taxpayers may demand that IRS disclose such infor- bsent f ipslnated ftrSd parties However, missing from current law is a mandate atI°n, TRS orovide such information electronically to the designated third party l3t the IRS Pr?£°e. f"p frame e g 24 to 48 hours, as was previously done with ^m^fSaUot ^til the DDI was discontinued.' Therefore, the 1986 Tax Code ie DDI information unui DDI-type information or other tax re-

SmforSn'as direTd by the taxpayer within the previously achievable time-

a,raeA .;i 1QQ4 an industry grade association, the Electronic Filing Coalition of In Apr21fF?r6A) formed a task force comprised of industry experts, tax special- menCfpn?orcement professionals. The task force explored comprehensive measures Vi^AetecT^d prevent fraud in the electronic filing program and shared its ' turthjlfinn, with the IRS and appropriate committees of Congress in March

m,ltn PrevenT Fraud in Electronic Filing, (attached) in which several rec- tT '^Sions are made to both the electronic filing industry and the IRS to enable mmendations are maae ^^ ^ electromc m Several of these

^n,endaLnsgehav8er already been adopted by both the IRS and the industry. (See Xt^Ser^^™*^* -to initiatives for the electronic lime industry and for the IRS. For the industry, the Task Force recommended: . exoansion of its existing fraud prevention program to include informing .cus- tomers upon initial contact, of the criminal and civil penalties of tax fraud, . several potential improvements to the existing FSB, including: —coordinating efforts with the IRS; —improving data flow;

—expanding base-line information such as fraud profiles and demographics. For the IRS, the Task Force recommended: . requiring more stringent qualifications for the assignment of electronic filing

. aSl^^e^eSdng penalties for violations of electronic filing rules;

! frnposfn" ?m^re comprehensive controls on the procedures it follows m process-

. Kr^detailed, current information which employers already provide about

their employees when filing their quarterly employment tax returns

. requiring document verification for EITC claims on tax returns prior to the

processing of these returns.

EFCOA's Task Force stated its belief that the improvement in data flow andjer-

ification that would result with the implementation of its ^^^^^^3

enable both the industry and IRS to combat fraud in electronic filing to their mutual

^Smce completion of the Task Force's study we have continued tc .evaluate .the fraud detection processes already in place and subsequently ha% >e developed addi tional suggestions to detect and deter EITC-related and other potential fraud.

. Issuance of an authorization form to the taxpayer in their f0™8.^0^^ a code that indicates the taxpayer met all the requirements of ErTC ! elig , b ity in the prior year and will be eligible again if the income is w1*11?^,^^ ing range and the children remain the same. The form would be filed with

. PlacementTan EITC eligibility indicator on the mailing label for tax re-

. ^utilization of the enormous infrastructure the f^^^ne^oS already has in place. The "private" electronic filing industry °™ fnd °J« ates every process and procedure guiding electronic filing up until the 1Kb

2EFCOA uicludes the major institutions that offer refund [anticipation loans, tax preparers electronic return originators, and electronic return transmitters. The coalition was termed in 1992 to address areas of shared concern among which is tax fraud prevention.

158

center receives the return. By working together with the industry, the IRS could provide enhanced manual and software checks at the tax return origi- nator sites, utilize enhanced software processing at the electronic transmitter locations, and make better use of the FSB. The IRS cannot effectively control fraud by only looking at the back-end of the tax return process. By utilizing the real-time fraud detection capabilities of the RAL banks, the IRS could leapfrog two decades of computer development at no cost to the taxpayer.

The electronic filing industry understands and shares the IRS' concerns about fraud. We believe fraud can be most effectively prevented through increased and continued cooperation and communication between government and private indus- try.

Conclusion

Mr. Chairman and Members of the Committee, I trust that I have clearly con- veyed to you today Beneficial Corporation's perspective on the important issues be- fore you. We, as a company and as part of the electronic filing industry, are commit- ted to taking concrete action to reduce fraud in electronic filing. And, our actions over the past several years have demonstrated that commitment. We are confident that effective fraud prevention programs can be implemented that will not result in the wholesale hardship inflicted upon taxpayers — particularly low income tax- payers— this tax season. Such disruption can be avoided — if, and only if, the IRS and the electronic filing industry work together going forward. And if we are suc- cessful in forging an effective public-private partnership, we will certainly make strong advances to establishing a widely utilized, safe, and affordable electronic fil- ing tax system for this country.

Thank you very much.

ELECTRONIC FILING COALITION A Public-Private Partnership to Prevent Fraud In Electronic Filing Introduction

In March 1994, the Electronic Filing Coalition of America (EFCOA) formed a Task Force of industry experts, tax specialists, and enforcement professionals to explore comprehensive measures to prevent fraud in electronically filed income tax returns. The Task Force's extensive deliberations and analyses have resulted in the publica- tion of this study: A Public-Private Partnership To Prevent Fraud in Electronic Fil- ing.

The elimination of fraud in electronic filing is in the interests of both the Internal Revenue Service and the companies that comprise the electronic filing industry. Nei- ther the private sector nor the Government will benefit if the future development and expansion of electronic filing is short-circuited by real or perceived threats of fraud.

Of course, fraud in either paper or electronically filed returns costs the U.S. Treasury money, but there is another, equally costly consequence of unchecked fraud in electronic filing: The IRS has made the elimination of paper returns and the substitution of electronic information the centerpiece of its Tax Systems Mod- ernization (TSM). Without careful and prudent planning, well-intentioned actions — that, in the name of fraud reduction, might significantly reduce the number of elec- tronically filed returns — would put the entire modernization plan at risk. Such an unintended consequence would surely negatively impact the ability of the IRS to continue on its path to efficiency and effectiveness. At the same time, as a result of tax return fraud, the companies that provide refund anticipation loan (RAL)1 pro- grams also risk considerable financial losses.

After completing a thorough and detailed review, EFCOA's Task Force has ar- rived at a series of recommendations and action steps. Since effectively combating fraud requires a public and private partnership, the Task Force's recommendations are aimed at assisting the industry members and the IRS in the elimination of the windows in the electronic filing process through which fraud currently enters the tax system undetected. EFCOA offers these recommendations, and those that will undoubtedly be made in the future, as part of an on-going industry effort to attack and reduce fraud in electronic filing.

1 A refund anticipation loan (RAL) is a rapid turnaround advance made by a financial institu- tion in the full or partial amount of a taxpayer's requested refund.

159

Executive Summary

The recommendations are divided into initiatives for the electronic filing industry and for the IRS. The Task Force believes that the improvement in data flow and verification that would result from the recommendations in this study would enable both the industry and IRS to better combat fraud in electronic filing to their mutual benefit.

For the industry, the Task Force strongly recommends that it expand its existing fraud prevention program to include informing customers, upon initial contact, of the criminal and civil penalties of tax fraud. In addition, the Task Force has identi- fied several potential improvements to the industry's existing fraud detection proc- ess (The Fraud Services Bureau.2), including coordinating efforts with the IRS, im- proving data flow and expanding base-line information such as fraud profiles and demographics.

For the IRS, the Task Force's recommended initiatives include requiring more stringent qualifications for the assignment of electronic filing identification numbers (EFINs), aggressively enforcing penalties for violations of electronic filing rules, im- posing more comprehensive controls (such as requiring the verification of additional taxpayer information) on the procedures it follows in processing returns, utilizing detailed, current information which employers already provide on their employees when filing their quarterly returns and requiring documentation verification for Earned Income Tax Credit (EITC)3 claims on tax returns prior to the processing of these returns.

RECOMMENDATIONS

Apart from the mechanics of the actual processing of electronically filed tax re- turns, the Task Force considered a number of issues that have been associated with the perpetration of fraud: the IRS' reduced time frame for detecting fraud, the role of the EITC, and the role of RALs. These issues have been carefully reviewed in the course of the development of the Task Force's recommendations.

• Concern about the time frame of electronic return filing. The Criminal Inves- tigation Division (CID) of the IRS has expressed concern that the time frame between the electronic filing of a return and payment of a refund may be too short for it to adequately screen for fraud. The Task Force firmly believes that adoption of its recommendations will better enable the IRS, with indus- try support, to detect fraud within the existing time constraints.

• Importance of RALs in the electronic filing program. The vast majority of elec- tronically filed returns were filed in conjunction with applications for RALs. The Task Force has determined that RALs do not cause fraud in electronic filing and further are an advantage to the development and expansion of the overall electronic filing program.

• Administration of the government's EITC program through the tax system. The Task Force recognizes that improved controls in this program are crucial as EITC is currently the area most widely used to defraud the IRS. For exam- ple, ineligible dependents, unmatched wage statements (fraudulent W-2s) and Schedule C's 4 are all being used in efforts to fraudulently claim this gov- ernment entitlement. Recent tax law changes have expanded the EITC and may have increased the opportunity for tax fraud. Further, fraudulent EITC- related returns are filed both on paper as well as electronically. Tests suggest that if RALs were eliminated in electronically filed returns, the fraudulent EITC schemes would simply move to paper. Consequently, the Task Force has concluded that EITCs which are claimed on the tax returns — not RALs — are the basis for most fraud found in electronically filed returns. The Task Force suggests that the screening processes relating to the validity of EITC claims should be expanded and enhanced.

2 The Fraud Service Bureau has been developed by the major refund anticipation loan lenders. It is a sophisticated computer system that uses statistically based scorecards and neural net- work technologies to help identify and reduce fraud.

3 Earned Income Tax Credit, provided by Congressional mandate through the tax filing proc- ess, is a tax credit that is given to qualifying, low-income taxpayers who have an eligible de- pendent child. The maximum potential EITC is $2,528.

4 Form Schedule C shows the profit or loss from business. Schedule C is used to adjust the earned income amount in order to qualify for the maximum EITC.

160'

FOR THE ELECTRONIC FILING INDUSTRY

RECOMMENDATION 1: INCREASE TAXPAYER AWARENESS OF THE PEN- ALTIES ASSOCIATED WITH FILING OF FRAUDULENT TAX RETURNS

Although the filing of a fraudulent tax return is a federal offense, there has been little sustained effort to emphasize the seriousness of the offense at the initiation of contact between the taxpayer and the electronic return originator (ERO). Such an early emphasis would underscore the following facts to customers: that both the ERO and the IRS are working to prevent fraud, that electronic filing offers no "easy way" to defraud the Government, and that both the industry and the IRS have a "zero tolerance" for fraud in electronic filing. This would be a powerful deterrent to any taxpayer who might contemplate committing crimes.

The Task Force suggests the industry members take the following actions to im- plement this recommendation:

• All EROs should display prominently posters which clearly delineate the po- tential penalties for tax return fraud. Further, these posters should be re- quired by IRS rule.

• The industry should inform all of its customers of the penalties associated with filing fraudulent tax returns. This should be done in writing as part of the application forms and/or marketing materials. In addition, the IRS should delineate the penalties as part of the text on the Form 1040 and Form 8453. 5

RECOMMENDATION 2: ESTABLISH A CLOSER WORKING RELATIONSHIP WITH THE IRS

Within the confines of confidentiality laws, the industry should attempt to share more information from its Fraud Service Bureau (FSB) with the CID. By providing information and data to CID quickly, the industry will assist the IRS in its on-going efforts to identify those returns with a greater probability of fraud before refunds are actually paid.

The Task Force suggests the following as some of the actions which the industry's FSB take to implement this recommendation:

• FSB should attempt to enter into a confidentiality agreement under which the CID and the FSB would be able to share statistical and characteristic infor- mation of fraudulent tax returns.

• The additional information gained through this agreement should be used to further improve the models that both organizations use in screening the tax returns.

• FSB should explore with CID ways that it could access the FSB database, ei- ther in-house or via telecommunications, to give CID the opportunity to re- view questionable returns as soon as they are received at the FSB.

RECOMMENDATION 3: PERFORM INITIAL FRAUD DETECTING METHODS PRIOR TO FORWARDING RETURNS TO THE IRS

Current industry practices already have the capability of detecting potentially fraudulent returns before they are forwarded to the IRS. In such cases, the informa- tion is turned over to CID. An expansion of the industry's fraud detection methods would reduce the number of fraudulent returns actually being filed. With CID's prompt involvement, it may lead to the apprehension of the criminal(s) and provide a greater deterrence to others who may be inclined to commit fraud through elec- tronic filing.

The Task Force suggests that the following actions be taken to implement this recommendation:

• Industry participants should make a greater effort to flag for the IRS those returns they suspect to be fraudulent.

RECOMMENDATION 4: INCREASE THE CONTROLS ON THE FSB'S EXISTING FRAUD DETECTION PROCEDURES

The Task Force noted that the four major RAL banks, consisting of Bank One, Beneficial National Bank, Greenwood Trust and Mellon Bank, have spent approxi- mately $6 million during the past two years to establish the FSB to screen elec- tronic returns associated with RALs for fraud. The system currently employs the use of statistical and neural scorecards to weight tax characteristics and to return scores to the RAL lenders for their loan decision.

5 Form 8453 is the U.S. Individual Income Tax Declaration for Electronic Filing. This form provides the taxpayer's authorization to have their return filed electronically, authorizes the IRS to notify the ERO whether a direct deposit request will be honored and captures the taxpayer's signature.

161

The Task Force believes that increased controls will further improve the FSB's fraud detection capabilities and thus prevent the IRS and the RAL lenders from making refund payments and loans on fraudulent returns.

The Task Force suggests the following actions be taken to implement this rec- ommendation:

• FSB should obtain from the IRS and maintain a listing of electronic filing identification numbers (EFINs) which the IRS has shut down and their asso- ciated principals for ERO application processing. The IRS should be notified of any EFINs closed by the RAL banks for procedure violation, subject to the resolution of any issues of confidentiality.

• FSB should build a valid employer identification number (EIN) list from past years' data to assist in determination of valid W-2 EINs.

• FSB should incorporate additional pattern recognition systems within the FSB process and include this output in the scoring systems.

• FSB should design a method of determining variances in returns processed from year to year and setting tolerance levels which would provide the nec- essary alert if exceeded. Each RAL lender must set its own tolerance levels within the common FSB methodology.

• FSB members should encourage the IRS to have representatives visit on-site at the bank to work with the bank fraud specialists when appropriate.

FOR THE INTERNAL REVENUE SERVICE

RECOMMENDATION 1: ISSUE AN EFIN, THROUGH THE FORM 8633 6 APPLI- CATION PROCESS, TO EACH AND EVERY INDIVIDUAL OR BUSINESS IN- VOLVED WITH EACH AND EVERY ASPECT OF THE ELECTRONIC FILING

This should include all levels of participation regardless as to whether the entity has "direct contact and/or interaction" with the taxpayer — consisting of service bu- reaus, collection points, centralized processing points, tax preparers, stores, drop off collection points, etc. This will enable the IRS to screen all industry participants for past crimes and fraudulent activities.

RECOMMENDATION 2: ENFORCE MORE STRINGENT QUALIFYING STAND- ARDS FOR BUSINESSES WHICH REQUEST AN EFIN

Obtaining an EFIN should be viewed as a privilege and not a right. By better monitoring whom it allows to participate in the electronic filing program, the IRS will be able to cut off unsuitable businesses and individuals before they aid in the filing of fraudulent returns.

The Task Force recommends the IRS impose the following strict controls in deter- mining the suitability of an individual or business engaged in any aspect of process- ing electronic returns:

• Administer and assign EFINs at the IRS district level, which is the Service's level closest to the tax system users and thus most likely to be able to detect unsuitable EFIN applications.

• Implement a required training program that EROs must complete prior to the IRS activating an EFIN. The program should include obtaining a certain level of continuing professional education credit or other suitable training regard- ing Publication 1345 7 procedures prior to being approved for an EFIN by the Service Center.

• Require the EFIN applicants to provide the names and Social Security num- bers of all principals involved with the business. These names and Social Se- curity numbers should be checked for any prior association with an entity for which the EFIN has been suspended by the IRS.

• Check the Employer Identification Number (EIN) on each Form 8633. Recent issuance of an EIN should be a flag for further verification of the tax return. The IRS should also be more stringent in the controls it imposes upon the EIN application process, i.e. EINs should not be issued by telephone.

» Move the deadline for filing a Form 8633 from December 1 to November 1, thus providing an additional month to perform the necessary suitability screening before the start of the filing season.

• Maintain a database of all suspended or revoked EFINs which should include the names and Social Security numbers of all the principals involved. This list should be made available to industry members so they are aware that they should no longer deal with these businesses.

6 Form 8633 is the application to participate in the electronic filing program.

7 Publication 1345 is the handbook for electronic filers of individual income tax returns.

162

RECOMMENDATION 3: POLICE THE BUSINESSES TO WHICH IT GRANTS EFINs

By proving that the penalties for violation of electronic filing rules and procedures will be enforced, the IRS will deter the filing of fraudulent returns by EROs — as long as the potential discovery and the penalties outweigh the benefits of doing so.

The Task Force suggests that the IRS take the following steps to implement this recommendation:

• Establish standardized administrative penalties for violation of electronic fil- ing rules and procedures by businesses who have been granted EFINs. Cur- rently none exist beyond the suspension of the EFIN.

• Impose monetary fines and/or suspension, revocation or criminal processing beyond what is listed in Publication 1345. Currently, the EFIN will be sus- pended or revoked, but there are no standard monetary penalties specifically targeted at electronic filing fraud.

• Strictly enforce the penalties on violators in the industry.

RECOMMENDATION 4: IMPOSE MORE STRINGENT CONTROLS, BOTH IN- TERNAL AND IN CONJUNCTION WITH EROs, ON THE EXISTING TAX RE- TURN PROCESSING PROCEDURES

There is some concern about the time period that CID has to detect fraud in the case of electronically filed returns. If both the electronic filing industry and the IRS, working together, institute sufficient controls on the electronic filing process, the time frame between the submission of an electronically filed return and payment of the refund should be adequate to detect the majority of attempted fraudulent tax refund claims filed electronically.

The Task Force suggests the following internal IRS controls to implement this recommendation:

• Verify and check all Social Security numbers for duplication prior to provid- ing an acknowledgement to the ERO. Check across all Service Centers to de- termine if a Social Security number has already been processed for a refund, used as a qualifying dependent for EITC or used as a personal exemption.

• Perform a full-match validation of the EIN (beyond the first two numbers) prior to sending an acknowledgement to the ERO. Maintain an up-to-date database of valid EINs to use as a master list for validation. Reject any re- turn with an invalid EIN.

• Establish clearer guidelines regarding the acceptability of Substitute W-2 forms (Form 4852). Include an indicator on the electronic W-2 which identi- fies a substitute W-2 rather than simply processing it as a non-standard W- 2. Refunds for returns containing a substitute W-2 should not be processed via direct deposit.

In addition, the Task Force suggests the following controls for the IRS and EROs to follow jointly:

• Provide an indicator in the record layouts which the ERO could use to flag suspect returns for the Service. Acknowledgement of the returns deemed sus- pect by the ERO should be delayed for 1-2 days while the IRS completes ad- ditional checks. EROs would need some type of legal protection against subse- quent action by the taxpayer if they are incorrect.

• Standardize procedures for the EROs, lending banks and electronic return transmitters to follow when reporting questionable refunds to the IRS/CID. Provide some form of feedback to the ERO regarding action being taken by IRS/CID.

RECOMMENDATION 5: DEVELOP A DATABASE OF VALID EMPLOYER/EM- PLOYEE RELATIONSHIPS

The Task Force believes this would enable the IRS to discover many of the fraud- ulent Form W-2s that are currently escaping undetected. No additional costs to the taxpayer need be imposed as employers are already required to file Form 94 18 and Form 941 A on which the employer must list all employees, their earnings and their respective Social Security numbers. Currently, Form 941A is detached and for- warded directly to the Social Security Administration.

Furthermore, the Task Force believes that this recommendation makes the pro- posed acceleration of W-2 due date unnecessary. If the IRS took the information from the third-quarter Form 941 which is due October 31, the IRS would have two

8 Form 941 is the Employer Federal Quarterly Tax Return.

163

months to prepare the database before the first individual tax returns for the year are filed.

The Task Force suggests the following actions be taken to implement this rec- ommendation:

• Create a multiple copy Form 941 A, and keep one copy for IRS use while for- warding the other to the Social Security Administration. IRS would therefore retain access to the additional information needed for the database with mini- mal or no additional cost to employer taxpayers.

• Cross-check all W-2 forms included with tax returns against this database.

• If feasible, verify that the employers actually deposit the required payroll taxes.

RECOMMENDATION 6: STRENGTHEN THE ELIGIBILITY CRITERIA AT BOTH THE ERO AND IRS LEVEL IN THE VALIDATION OF EARNED INCOME TAX CREDITS

Taxpayers requesting an EITC should provide adequate documentation to estab- lish their eligibility for the credit.

The Task Force suggests that the following actions be taken to implement this recommendation:

• The IRS should mandate that copies of qualifying documents be either at- tached to Form 8453 or retained with the ERO copy of Form 8453. Suggested documentation includes school records, birth certificate and Social Security cards. These documents would prove the existence of the qualifying child and validate that the residence requirement is met.

• ERO penalties should be created for non-compliance.

• The industry and the IRS should meet for the purpose of generating further ideas to implement this recommendation.

Conclusion

EFCOA is committed to continuing its on-going efforts to reduce fraud in elec- tronic filing. It firmly believes that its Task Force's recommendations will go a long way toward the prevention, if not the elimination, of fraud in electronic filing in all the areas of current concern in electronic filing — including the time frame for proc- essing, the role of RALs and the administration of the EITC through the tax system. In fact, if these recommendations are adopted by both the electronic filing industry and the IRS — and fraud is subsequently reduced — the electronic filing program will continue to be the centerpiece of the Tax System Modernization program. Coopera- tion between the electronic filing industry and the IRS is essential to achieve the mutually beneficial goal of eliminating fraud in electronic filing.

164

United States General Accounting Office

ri AjT) Briefing Report to the Chairman,

y-xjn^J Committee on Governmental Affairs,

U.S. Senate

EARNED INCOME CREDIT

Targeting to the Working Poor

GAO/GGD-95 122BR

165

GAO

United States

General Accounting Office

Washington, D.C. 20548

General Government Division

The Honorable William V. Roth, Jr.

Chairman, Committee on Governmental Affairs

United States Senate

Dear Mr. Chairman:

This is our second report responding to your interest in the Earned Income Credit (eic).1 In this report, which follows our briefing for you, we present information on eic noncompliance and assess changes and administrative issues that might result from potential changes to eic eligibility criteria These criteria would take into account (1) measures of taxpayer wealth and (2) more sources of income when determining who qualifies for the credit.2 We also provide information on illegal alien recipients of the eic and describe the administration's proposal, which is similar to your 1994 proposal, to exclude such aliens from eligibility.

^"™ ™T^^ eic noncompliance has been a continuing concern of the Internal Revenue KeSUltS IT! tJnel Service (ms). Current, reliable noncompliance measures do not exist for

the entire eic program. An irs study of noncompliance for returns filed electronically during two weeks in January 1994 found that an estimated 29 percent of these recipients received too much eic, and 13 percent intentionally claimed foo much. The extent of such noncompliance for paper returns is unknown but also of concern to the ms.

Concerned about eic noncompliance and refund fraud generally, ms has taken steps to detect and prevent erroneous refund payments. These include developing and applying improved criteria for detecting noncompliant returns and checking for the use of the same Social Security number (ssn) on multiple tax returns. These steps have resulted in many more taxpayers being asked to provide evidence of eic eligibility and in delaying refunds to at least 2.9 million eic claimants as of March 17. 1995. In addition, as of March 17, irs had sent out almost 4. 1 million notices primarily when returns did not appear to contain valid ssns for dependents or, in the case of eic, for qualifying children. Although these steps may

'Tax Administration Earned Income Credit— Data on Noncompliance and Illegal Alien Recipients (GAO/GGD-95-i;T < .i on EIC noncompliance.

2 As you requested, we initially assessed the magnitude of change likely to result from taking wealth and additional sources of income into account when awarding the EIC On the basis of this work, you requested that the Joint Committee on Taxation (JCD provide official revenue estimates. We present those estimates in this report.

GAO/GGD-95I22BR Earned Income Credit

166

inconvenience or burden taxpayers, if implemented effectively, they could help IBS improve eic compliance.

Taxpayers' earned income, and in some cases their adjusted gross income (aoi), as well as whether they have children meeting certain age and residency tests, determine eic eligibility and credit amounts. Unlike certain federal welfare programs, taxpayers' wealth (e.g., the value of property or other investments they own) does not affect eic eligibility. In addition, the eic does not consider certain forms of income in determining how much, if any, credit taxpayers will receive, eic eligibility criteria could be changed to take into account wealth and additional forms of income.

The jct estimates that denying the eic to taxpayers who have some wealth, as indirectly measured by their asset-derived income, could yield $318 to $971 million in revenue savings in fiscal year 1997, depending on the wealth test design. These revenue savings represent potential reductions in eic program costs resulting from changing eic eligibility criteria. In addition, taking nontaxed Social Security income, tax-exempt interest, and nontaxed pension distributions into account in taxpayers' agi for credit calculations could yield $1,449 billion in revenue savings in fiscal year 1997, according to jct estimates. Also, taking child support payments into account would increase revenues in fiscal year 1997 by $686 million.

However, adding an indirect wealth test or an expanded agi definition to the eic eligibility criteria would add to the eic's complexity. Complexity has been a continuing eic issue because it can lead to increased errors and dissuade deserving taxpayers from claiming the credit. Of the potential changes to eic criteria, adding child support payments to taxpayers' agi likely would cause the greatest complexity because information on such income is not collected by IKS and systems may not exist to comprehensively generate the information.

Although an indirect wealth test for the eic that uses tax return data might be more practical than a more comprehensive test, it would have significant limitations in measuring potential eic recipients' actual wealth. For instance, such a test would not measure the value of taxpayer assets like capital stock funds that yield little, if any, annual income. These limitations could raise concerns that taxpayers with similar wealth would be treated differently for the eic.

GAO/GGD 95122BR Earned Income Credit

167

Background

The eic is a refundable tax credit available to low-income working taxpayers with children and, beginning in tax year 1994, certain taxpayers without children. Congress established the Eic in 1975 to achieve two long-term objectives: (1) to offset the impact of Social Security taxes on low-income workers with families and (2) to encourage low-income individuals with families to seek employment rather than welfare.

For tax year 1993, about 14.7 million taxpayers claimed about $15 billion in eic benefits. To be eligible for any eic in tax year 1993, a taxpayer must have had earned income of less than $23,050 and had one or more qualifying children who met the age, relationship, and residency tests. The Omnibus Budget Reconciliation Act (obra) of 1993 increased the number of taxpayers eligible for the eic and the credit amount These changes began in tax year 1994 and will be fully effective in tax year 1996. The maximum income qualifying for the Eic will rise to $27,000 in tax year 1996, the maximum credit will rise to $3,370 for tax year 1996,3 and total Eic cost in fiscal year 1996 is expected to reach nearly $25 billion (in 1994 dollars).

Objectives, Scope, and Methodology

Our objectives were to (1) present information about eic noncompliance and what steps irs is taking to control such noncompliance and (2) review the impact on the amount of Eic paid that might result from potential changes to the eic eligibility criteria that would reflect taxpayer wealth and additional sources of income and administrative issues which could arise due to these changes. In addition, we were asked to provide information about illegal aliens receiving the eic and to describe the administration's proposal to exclude illegal aliens from eligibility.

To review the effects of possible changes to eic eligibility criteria, we obtained and analyzed data from the irs' Statistics of Income Division (soi) and from the Bureau of the Census' Current Population Survey (cps). For our various objectives, we also met with Treasury, irs, jct, and Congressional Budget Office officials; visited iks Service Centers in Cincinnati, OH, and Fresno, CA; and reviewed relevant literature on the eic. See appendix I for more details on our methodology.

We did our work from August 1994 through February 1995 in accordance with generally accepted government auditing standards. On March 21, 22, and 23, 1995, we discussed our draft report with Department of the

than shown here t

GAO/GGD 9J.122BR Earned Income Credit

168

Treasury and ms officials who are responsible for administering the ac, ensuring compliance, and analyzing potential policy changes. The officials generally agreed with the material in the report but offered updated data and suggestions for improving the clarity of presentation. We made appropriate changes to the report to reflect their comments.

As agreed with your office, unless you publicly announce the contents of this report earlier, we will not distribute this report until April 4, 1995. At that time, we will send copies of this report to various interested congressional committees, the Secretary of the Treasury, the Commissioner of Internal Revenue, and other interested parties. We will also make copies available to others on request

The major contributors to this report are listed in appendix II. Please contact me on (202) 512-8633 if you have any questions about this report.

Sincerely yours,

Lynda D.Willis

Associate Director, Tax Policy and Administration Issues

GAO/GGD-95122BR Eirned Inco

169

Contents

Letter \

Briefing Section I !

r» i j Targeting the EIC 8

rJaCKgrOUna Ensuring That the Working Poor Receive the EIC 9

Range of EIC Recipients With Two Qualifying Children ( 1996) 10

How the EIC Is Awarded 1 1

Growth in EIC Program Costs (1988-2000) 12

Broader Coverage and Larger Credit Amounts Increase EIC 13

Program Costs

Briefing Section II

14

IRS Studies 14

IN OnCOmplianCe 1RS Is studying EIC Noncompliance 14

1994 Filing Season Enforcement Results 16 IRS' 1994 Enforcement Efforts Address Noncompliance 16

1995 Interim Enforcement Results 18 IRS' 1995 Countermeasures for Addressing Noncompliance 18

Briefing Section III

20

Better Measuring EIC e Options for Measuring Wealth 20

Filers' Resources Wealth Thresholds 22

Wealth Thresholds for Qualifying for the EIC 22

Wealth Test Results 24

Lower Wealth Thresholds and Broader Measures Reduce EIC 25 Program Costs

Income 26

AGI Criteria Could Be Expanded 27

Expanded AGI Results 28

Results of Expanding AGI to Include Nontaxed Social Security 29 Income, Tax-Exempt Interest, and Nontaxed Pension Distributions

Additional Considerations 30

Complexity Increases With Wealth and Income Tests 30

Child Support Data Not Reported 32

Including Child Support Payments in AGI Reduces Costs but 32 Imposes Greater Administrative Difficulty

GAO/GGD95 122BR Earned I

170

Briefing Section IV EIC Eligibility

Appendix I Objectives, Scope, and Methodology

Appendix II

Major Contributors to

This Report

Glossary

	34
Illegal Alien Recipients	34
Some Illegal Aliens Receive EIC	34
Excluding Illegal Aliens	36
Proposals to Exclude Illegal Aliens From EIC Eligibility	36

Abbreviations

Aid to Families With Dependent Children

Adjusted Gross Income

Current Population Survey

Earned Income Credit

Internal Revenue Service

Joint Committee on Taxation

Omnibus Budget Reconciliation Act

Statistics of Income

Social Security Administration

Social Security number

taxpayer identification number

UAOAJUI> 96 I 22BR Elrned I

171

Briefing Section 1

Background

G^0 Targeting the EIC

How can we better ensure that only the working poor receive the ESC?

Reduce noncompliance

Revise EIC eligibility

• Wealth test

• Income test

• Illegal aliens

GAO'GGD 95 122BR Earned Income Credit

172

Ensuring That the Working As requested by the Chairman, Senate Committee on Governmental Poor Receive the EIC Affairs, this report addresses how the federal government can better

ensure that only the working poor receive the eic. Specifically, the report

• presents information about noncompliance with the eic and what the irs is doing to increase compliance and thus exclude ineligible taxpayers from receiving the credit;

• assesses changes that may result from potential changes to the criteria used in determining eic eligibility. These changes would take into account more of the resources that taxpayers could use to support themselves and their families (resources not taken into account when determining eic eligibility include taxpayers' wealth and certain forms of income);4 and

• presents information about how many illegal aliens receive the eic and discusses the administration's proposal, which is similar to Senator Roth's 1994 proposal, to exclude illegal aliens from eligibility.

GAO/GGD-95122BR E*m«i In

173

Briefing Section I

GPD Range of EIC for Recipients With Two Qualifying Children (1996)

Earned income credit (1994 dollars)

$10,999 $11,000

$1 $8,424 $8,425

Range of recipient's income I I Phase in: recipient receives 23 Maximum range: recipient $.40 for each $1 dollar of receives fixed $3,370 EIC

$27,000

Phase out: recipient receives $.21 less for each $1 of income

Source Congiessional Research Service

GAO/GGD -9S 122BR Earned Income Credit

174

Briefing Section I

How the EIC Is Awarded The Eic is a refundable tax credit available to low-income working

taxpayers with children and, beginning in 1994, certain taxpayers without children. Congress established the eic in 1975 to achieve two long-term objectives: (1) to offset the impact of Social Security taxes on low-income workers with families and (2) to encourage low-income individuals with families to seek employment rather than welfare.

ER eligibility and credit amounts generally are determined according to the taxpayers' earned income and whether they have qualifying children who meet age, relationship, and residency tests. The credit gradually phases in, plateaus at a maximum amount, and then phases out until it reaches zero. If the taxpayers' earned income or adjusted gross income (agi) exceeds the maximum qualifying income level, they are not eligible for the credit When the taxpayers' agi falls in the credit's phase-out range, they receive the lesser amount resulting from using their earned income or agi in calculating the credit.

As the figure illustrates, when changes made in the 1993 Omnibus Budget Reconciliation Act (obra) are fully in effect in tax year 1996, taxpayers with two children and whose earned income ranges from $1 to $8,424 will receive $.40 for each dollar earned. For taxpayers with incomes between $8,425 to $10,999, the amount of eic received will remain stable at $3,370. Taxpayers whose income falls between $11,000 and $27,000 will receive a declining amount of eic, with the credit falling $.21 for each additional dollar of income.

QAO/GGD 95 -122BR Earned Income Credit

175

GPD Growth in EIC Program Costs (1988-2000)

Millions of dollars (1994)

30.000

24,550

15,010

1996 1998 2000

Projected

Source Fiscal year estimates tram the Presidents' 1990. 1992. 1994. 1996 budgets

GAO/GGD-96122BR Earned Income Credit

176

Briefing Section I

Broader Coverage and Larger Credit Amounts Increase EIC Program Costs

Total program costs5 for the eic have increased dramatically as Congress has broadened its coverage and increased the amount of credit available. The figure shows that between 1988 and 1996 total eic program costs are estimated to increase over five fold in real terms, from $4.4 billion in 1988 to an estimated $24.6 billion in 1996. Congress has increased the coverage and amount of the credit for reasons such as to ( 1) ensure that Eic amounts would not fall in terms of purchasing power, (2) increase or maintain the progressivity of the tax system, and (3) better ensure that working individuals will have incomes above the poverty line.

The most recent change to the eic, in the obra of 1993, increased the maximum credit available and the income level at which individuals can qualify for the credit, and made certain low-income taxpayers without children eligible. The maximum credit amount for a family with two children is rising from $ 1 ,5 1 1 for tax year 1993 to $3,370 in tax year 1996. The maximum income qualifying for the eic is rising from $23,050 in tax year 1993 to $27,000 in tax year 1996. Finally, beginning in tax year 1994, individuals without a qualifying child are eligible for the credit if they (1) are at least 25 but less tnan 65 years old, (2) are not a dependent of another taxpayer, and (3) have earned income and agi of $9,000 or less. These taxpayers will be eligible for a maximum credit of $306, adjusted for inflation.

i refundable tax credit As such, the portion of the credit t

the federal tax law The lefundable pottil I the Ell i- considered a federal outlay. We totaled tl

tax expenditure estimale and the outlay estimate from appropriate versions of the President's Bin to arrive at "total EIC program costs "

GAO/GGD95122BR Earned I

177

Briefing Section n

Noncompliance

G*> IRS Studies

No comprehensive, reliable EIC data since 1988

IRS1 1994 2-week study

• Electronic returns only

• 29 percent received too much EIC ($358 million)

• 13 percent were judged to have "intentional" errors ($183 million)

IRS1 ongoing study

IRS Is Studying EIC Currently, no reliable data exist on the extent of noncompliance among all

Noncompliance E!C claimants. The most recent Taxpayer Compliance Measurement

Program, from 1988, showed that about 42 percent of EIC recipients received too large a credit and about 32 percent were not able to show they were entitled to any credit About $1.9 billion (34 percent of the total eic paid out) was awarded erroneously. However, the impact of the significant changes to the eic since 1988 suggest that a new compliance

GAO/GGO-95 122BH E«jn«i Income Credit

178

measurement is needed. For instance, in the intervening years, Congress has broadened the eic coverage, increased the credit amount, and revised filing status and qualifying children criteria

In response to concerns about Eic-related fraud, irs studied a sample of those returns filed during a 2-week period in January 1994. Study results are only generalizable to electronic returns filed during this period, irs' preliminary analysis of these returns showed that an estimated 29 percent of the 1.3 million eic returns filed electronically during the period had claimed too large a refund and about 13 percent of the returns filed were estimated by irs as having intentionally claimed too much eic. Of the $1.5 billion of Eic claimed in this period, an estimated $358 million was erroneously claimed — about $183 million, or 12 percent, was classified as intentional error. This intentional error category comes closest of any ms category in the study to measuring eic fraud.6 About 3 percent of taxpayers claimed a total of about $7 million less Eic than they were entitled to receive.

In the fall of 1994, irs began reanalyzing the electronic returns using additional income data that were not available earlier. The results of the additional analyses are not yet available, irs officials expect the analyses to lead to a higher estimated error rate.

irs is doing a 1995 study that will yield a noncompliance estimate for the entire eic program. The study will include a random sample of Eic returns filed electronically and on paper throughout the 1995 filing season. Preliminary results may be available in June 1995.

"Determining whether a refund is fraudulent requires determining the taxpayer's intent, which is

GAO/GGD-95 122BR Eimed, Income (

179

Briefing Section II

GAD 1994 Filing Season Enforcement Results

Paper returns: Extent of EIC noncompliance unknown

• 400,000 returns stopped in 1994 because they lacked SSN for child

• $500 million in refunds delayed, most indefinitely

Paper and electronic returns

• 77,781 fraudulent returns in 1994

• $43 million in refunds incorrectly paid out; $117 million in refunds stopped

• 91 percent involved the EIC

Source IRS data

IRS' 1994 Enforcement In addition to the 2-week electronic EIC return study, res detected

Efforts Address noncompliance in its noimal efforts to detect inaccurate or improper

Noncompliance returns as they are processed. These data document noncompliance that

IRS discovers while processing eic returns but do not measure the universe

of noncompliance.

GAO/GGD-95 122BR Earned Income Credit

180

In 1994, during initial manual reviews of tax year 1993 eic paper returns. irs personnel identified about 150,000 taxpayers who claimed the eic although the return information indicated they were not entitled to it. Most problems resulting in disqualification related to qualifying children, such as a child exceeding the age limit irs' initial computerized reviews of about 6 million electronically filed eic returns resulted in about 610,000 rejection notices being sent out. The rejections occurred when the qualifying child's Social Security number (ssn) did not match the Social Security Administration's (ssa) records.

Beginning in January 1994, irs personnel also stopped the processing of returns that lacked an ssn for the qualifying child and had a tax refund exceeding a threshold. Following statutorily required notice procedures, irs suspended the eic refund and asked the taxpayer to submit proof of their eic qualification. If proof was provided, the refund was released. If the taxpayer submitted insufficient proof or failed to respond, irs' policy was to permanently deny the refund. As of September 30, 1994, irs had delayed about $500 million in potentially erroneous Eic refunds claimed on about 400,000 of about 8.7 million paper returns, irs officials expected most of these refunds to be permanently denied because many taxpayers did not respond to requests for information or could not support their claims.

After electronic returns pass initial computer checks and paper returns pass manual checks, data is entered into irs computers, which then identify returns that are potentially fraudulent. These potentially fraudulent returns are reviewed by fraud detection teams. The number of fraudulent returns detected has grown steadily over recent years. As of December 31, 1994, of the total number of returns reviewed, irs had identified 77,781 as fraudulent — 44,137 on paper and 33,644 electronic returns. About $43 million in fraudulent refunds was not detected soon enough to stop the refund to taxpayers, but irs stopped about $117 million in refunds from being released. About 91 percent of the fraudulent returns claimed the eic

GAOA5GD-95-122BR Earned In

181

GAD 1995 Interim Enforcement Results

Delayed and rejected returns

• About 355,000 paper returns delayed

• About 3.7 million electronic returns rejected

• 2.9 million EIC returns delayed up to 8 weeks to check for duplicate SSN use

About 1 percent of new electronic return originator applicants rejected

IRS' 1995 Verifying ssn accuracy is key to irs' 1995 qc enforcement efforts. For

Countermeasures for paper returns, as enters into computers the taxpayer's ssn and, starring

Addressing this year" dePenuent and E,c qualifying childrens' ssns. When returns have

N y missing or invalid ssns (i.e., do not match ssa records), ms delays the

Noncompliance return and contacts taxpayers to resolve the problem. As of March 17,

1995, irs had delayed the refunds for about 355,000 paper returns that lacked a valid dependent or qualifying child's ssn.

GAO/GGD95 122BRS

182

irs has added controls to prevent returns with missing or invalid ssns, or ssns already used by another taxpayer, from being filed electronically. All returns with these problems are to be rejected and returned for correction. As of March 17, irs had sent out 3.7 million rejection notices7 principally for electronic returns with ssn problems related to a questionable refund. About 1. 1 million notices were primarily due to the eic qualifying child's ssn or year of birth not matching ssa records. The remaining 2.6 million notices were primarily due to dependent ssn problems.

In addition, irs is working to identify uses of the same ssn on more than one tax return. Through a new tracking system, ms intends to identify potentially problematic returns and trigger enforcement Because of past Eir fraud problems, irs is concentrating on Eic returns. IRS delays refunds up to 8 weeks from the time a notice is sent to taxpayers to allow staff time to identify duplicate ssn uses and fraud schemes — about 7 million Eic returns could be delayed. As of March 17, about 2.9 million eic refunds had been delayed. As of March 17, irs national office officials told us that initial problems with the duplicate ssn system were overcome early in the year. However, compliance personnel we spoke with said that problems with duplicate ssn data continued to impede their effective use of the system in mid-March.

irs has begun checking criminal and credit histories of new return originator applicants who wish to file taxpayers' returns electronically. Due to these checks, irs had rejected about 1 percent of applicants.

As of mid-March, statistics were not available on fraudulent returns detected this year.

number of returns

GAO/OGD-95 122BR Earned Income Credit

183

Briefing Section III

Better Measuring EIC Filers' Resources

^> Wealth

Comprehensive wealth test may be impractical

A narrower test, measuring income that is derived from taxpayers' assets, may be more practical

• Imperfect measurement

• Fairness concerns

Options for Measuring Congress requires income and wealth tests for certain welfare programs

Wealth like Aid to Families With Dependent Children (afdc). States use

questionnaires to determine the welfare applicant's degree of need. A similar wealth test for the eic likely would require additional irs resources, or a diversion of current resources, to obtain and verify the data. Although states might administer a wealth test for irs, such an arrangement likely would take time to perfect

GAO/GGD-95-12*BR E*ned Income Credit

184

Alternatively, a test that uses income earned from assets as an indirect indicator of wealth is perhaps more immediately practical.8 Such a test could measure income reported on tax returns that is derived from taxpayers' assets and compare that income to an income threshold. This is the general approach proposed by the administration and incorporated in House and Senate versions of H.R. 831, a bill to permanently extend the health care deduction for self-employed individuals.

In evaluating an indirect wealth test, Congress might wish to consider several options. Asset-derived income is in several income categories These include taxable interest and dividends, tax-exempt interest, estate and trust income, rental income, and capital gains. A wealth test that includes a broad array of asset-derived income might better measure taxpayers' wealth than a less inclusive test

However, no wealth test relying on tax-return information can completely measure a taxpayer's wealth. For example, the value of a home,9 valuable collections, and stocks that appreciate but pay few, if any, dividends is not reflected on tax forms except when the assets are sold. Within the constraints of using data reported to irs, broader income measures might come closest to measuring a taxpayer's overall wealth, but the measures nevertheless could incorrectly represent some taxpayers' wealth while more accurately measuring others' wealth. These limitations raise fairness concerns since taxpayers with similar wealth could be treated differently for purposes of the eic.

Taxpayers' wealth is somewhat taken into account through the f some asset-derived income The wealth t disqualifying t

GAO/GGD-95122BRjP«med I

185

GL40	Wealth Thresholds
	Income realized Presumed value of from assets underlying assets8
$1,000 $16,700
1,500 25,000
2,500 41,700

i the assumption c

Wealth Thresholds for Congress will need to set a threshold amount above which taxpayers

Qualifying for the EIC would be disqualified from receiving the eic if it wishes to adopt an

indirect wealth test for the credit The House version of H.R. 831 proposes a $2,500 threshold, indexed for inflation. The credit would phase out as asset-based income rose above $2,500 and disappear when such income was at least $3,150. The Senate version of the bill proposes a $2,450

GAO/GGD-95-122BR Earned Inco

186

unindexed threshold with different income items and provides no phase out. One consideration in selecting a threshold amount could be the level of assets, or wealth, that a given level of income may represent Possible thresholds could include, for example, amounts of $1,000, $1,500, and $2,500. Assuming a 6-percent annual realized rate of return, about $16,700 of assets would generate $1,000 of income. About $25,000 and $41,700 of assets would generate $1,500 and $2,500 of income, respectively.

However, the relationship between asset-derived income and the underlying asset value may vary widely. For example, if a taxpayer reports $1,000 of bank account interest, the average annual account balance likely would have been between $10,000 (at a 10-percent interest rate) and $30,000 (a 3.3-percent rate). If $1,000 of interest was earned on a tradeable bond, the value of the underlying bond could he outside those bounds because a bond's value rises or falls as interest rates change.

Associating income reported on tax forms with an underlying asset value is most problematic for capital gains income. For example, $1,000 of reported gain could come from a successful $1,000 investment in stocks that doubled in value. But, a $1,000 capital gain also could come from a $100,000 investment in stock that performed very poorly. The association between reported capital gains and underlying asset values also is complicated because the return on the assets could have been accumulated over many years.

It is difficult to say whether a goal of treating taxpayers of similar means similarly is better served by implementing a broad wealth test that combines income from assets with widely varying rates of return, or implementing a narrower test that ignores some assets completely. In considering a threshold amount for an indirect wealth test, lower thresholds may be more appropriate if the relationship between income and the value of underlying assets is less likely to vary widely among taxpayers.

UAO/GGD-95-122BR Earned 1

187

GAD	Wealth Test Results
	Income derived from taxpayers' assets (wealth) Option 1 Option 2 Option 3
Taxable interest3	•
Taxable dividends3	•
Tax-exempt interest	•
Net estate and trust income
Net rentalb income
Net capital gains0
Income threshold	Dollars in millions
Estimated revenue in FY'97 from reductions in EIC
$1,000	$685	$833	$971
1,500	505	637	766
2,500	318	405	524

■Less investment interest paid

"Net rental real estate income, net income from Real Estate Mortgage Investment Conduits,

'Net capital gains (Schedule D) and other gains (Schedule 4797) Source JCT estimates tor Senator William V Roth. Jr

GA(VGGB-95.122BR I

188

Lower Wealth Thresholds As more sources of income derived from taxpayers' assets are included in

and Broader Measures an Elc wealth test and as the test threshold is lowered, eic program costs

Reduce EIC Program Costs would be reduced further. Joint Committee on Taxation (jct) revenue estimates illustrate this point. 10 For example, option 3, including the broadest base of income derived from taxpayers' assets, coupled with a $1,000 threshold, is estimated by jct to raise $971 million in revenue in fiscal year 1997. In contrast, option 1, including only taxable interest and dividends and tax-exempt interest, coupled with the higher $2,500 threshold, is estimated to yield $318 million in revenue in fiscal year 1997. jct's latest published estimate for the eic program's total cost in fiscal year 1997 is $25.8 billion. Thus, $971 million would represent about a 3.8 percent reduction in the total eic program costs, and $318 million would represent about a 1.2 percent reduction.

'"Although figures provided by JCT are revenue estimates, these estimates can be thought of as "program cost" reductions. The EIC is a refundable tax credit As such, the portion of the credit thai offsets taxes owed by EIC recipients is considered a lax expenditure. i.e. , a reduction in taxes due t preferential provision in the federal tax law The refundable portion of the EIC is considered a fedei outlay. Since most EIC recipients owe no taxes, most of total EIC "costs" come from the refundable outlay, portion of the credit

GAO/GGD-95 I22BR Earned 1

189

G^0 Income

Income not included in AGI test

• Nontaxed social security income

• Tax-exempt interest

• Nontaxed pension distributions

GAO/GGD 95 122DR Earned Income Credit

190

Bettt-r Measuring EIC Filer*' Resources

AGI Criteria Could Be Expanded

Although the amount of eic a taxpayer receives is based largely on earned income, the amount, if any, also depends in part on other sources of income. Taxpayers' agi can limit their eic payments. In addition to earned income, agi includes income from other sources, such as investments, alimony received, and unemployment compensation. When taxpayers' agis fall within the eic phase-out range, eic payments are the lower of those resulting from using taxpayers' agi or earned income. When agis exceed the top of the eic phase-out range, taxpayers are ineligible for the credit regardless of their earned income level. Adding income elements to calculations of the agi for the eic, thus, would be an incremental change that would enable Congress, if it so desired, to take into account a fuller range of taxpayers' incomes in determining the amount of credit taxpayers would receive."

One alternative for expanding agi could be to include nontaxed Social Security income, tax-exempt interest income, and nontaxed pension income. These income sources are excluded from agi for purposes of calculating income tax liabilities but are sources of support available to individuals. 12 Of the three income items, Social Security is the largest income source to eic recipients. Although most taxpayers eligible for the eic do not receive Social Security income, several hundred thousand do.

"AGI would be expanded to include those other income items only for purposes of the EIC and not for income tax liability

lsThe taxable portions of Social Security income and pensions are included in taxpayers' AGIs.

GA4VGGD 9S 1 ;.'»« Karned i

191

G^0 Expanded AGI Results

Expanding AGI would yield $1,449 billion in FY'97.

Combining an expanded AGI and a wealth test would yield somewhat less revenue than the sum of the two.

Source JCT estimates lor Senatoi William V Roth, Jr

GAO/GG1) 95 1 22 BR Earned Income Credit

192

Results of Expanding AGI As requested by Senator Roth, jct estimated the revenues that would

to Include Nontaxed Social result from including nontaxed Social Security income, tax-exempt

Security Income interest income, and nontaxed pension income in taxpayers' agi for

X, „ .. ' , purposes of eic eligibility. It estimated that $1,449 billion in revenue would

1 ax-exempt interest, ana bg realized m fiscaJ year 1997 from q^ cnarige. This would represent

Nontaxed Pension about a 5 6 percent reduction in the estimated $25.8 billion cost of the eic

Distributions program for fiscal year 1997.

If both a wealth test and an expanded definition of taxpayers' agis were adopted simultaneously for the eic, the net result in revenues would be somewhat less than the sum of savings from each test independently. This would occur because some of the taxpayers disqualified by one test would also be disqualified by the other, but these reductions resulting from the disqualified taxpayers should not be counted twice when estimating net revenue savings achieved by implementing both proposed changes.

GAO/GGD-95-122BR Earned Income Credit

193

Better Measuring EIC Filers' Resources

GAD Additional Considerations

Wealth and income tests

• Increase complexity and burden

• May discourage EIC claimants Wealth test-Broader measures may

• Capture more wealth

• Reduce incentives to shift investments

• Include some unverifiable income Income test

• Most social security income is not reported on Form 1040

Complexity Increases With A weaJth test and an expanded agi definition for credit determination Wealth and Income Tests purposes would have additional consequences besides reducing qc

program costs. Both changes would increase the complexity of the eic and impose burdens on taxpayers in determining their eligibility and on nts in ensuring compliance. Complexity has long been a concern of the ms. Complexity contributes to taxpayer errors and the eic's high noncompliance rate, ms officials also expressed concern to us that a wealth test might discourage some legitimately qualified taxpayers from applying for the QC — a longstanding concern. Some research suggests that

GAO/GGD-96122BR Earned Income Credit

194

between 14 percent and 25 percent of eligible taxpayers do not claim the

EIC.13

Complexity concerns might be alleviated, in part, because a small portion of EIC recipients have income that would be taken into account in either the wealth test or the expanded agi. For example, in 1992 about 82 percent of Eic recipients had no income from any of the sources included in a broad indicator of wealth. Furthermore, about 50 percent of eic recipients use paid preparers.

Specifically for a wealth test, broader measures may capture more sources of wealth and might reduce taxpayers' incentives to shift investments to maintain their EIC eligibility. However.IRS would be unable to verify tax-exempt interest income because it receives no third-party information reports to use in checking the accuracy of taxpayers' returns. Short of an audit, irs also may be unable to verify the cost basis used in determining certain capital gains that could be included in a wealth test

For the expanded agi definition, tax-exempt income would, of course, be unverifiable. In addition, only taxpayers whose Social Security income is taxed report their Social Security income to IRS. KS would have to collect Social Security income data for an expanded agi test The ssa provides all Social Security recipients with an annual Form 1099 that records their Social Security income and a computer tape containing this information is provided to IRS. However, under existing systems, according to the IRS, more than a year likely would elapse before irs would be able to match taxpayers' claimed Social Security income to the ssa tape.

luipmviiit! Hi.' I>ehv.T> ,,[ hVm-fiLs t.> rii,> W..rlnn* i\i..r Prnj-isals '.> li.-f. urn the Dnrni

1 Tin el al , Improving the I leliverv nf Benefits to the Working PoorPr IncMi,:. I jv 1 reilii I'Mitfni;" Ainen.dj, T.ix P'.liry [nslitul. FVh I >*4

(iAO/GGD-9S-122BK Earned Income Credit

195

G^0 Child Support Data Not Reported

Including child support in the expanded AGI may yield $686 million in FY'97.

Administration would be difficult.

• Data on child support not reported to IRS

• Verification of child support payments, if reported, may be difficult

lor Senator William V Roth. Jr

Including Child Support The agi definition for eic purposes could be expanded to include child

Payments in AGI Reduces support payments. Including child support payments would recognize that

Costs but Imposes Greater such PaJ™ents "« p3* °f a family's ^p011 ** estimated by jct, adding

a^ • ot f TYffi H-t child support to agi along with the other items discussed earlier would

Administrative uuncuity result m a savin^ of $2 135 bUlion in fi^ year 1997 (about a 8.3 percent

reduction in total eic program costs for that year). This is $686 million more than if the agi definition did not include child support payments.

GAO/GGD-96122BR Earned Income Credit

196

However, the administrative issues associated with incorporating child support payments in eic eligibility criteria appear to be more formidable than for the other income items, such as Social Security payments, that could be added to agi for determining the Eic. Although child support payments may be a factor in determining whether a divorced or separated parent may claim a child as a dependent for income tax purposes, child support income itself is not required on any irs forms. Therefore, if child support were to be considered in determining eic eligibility, irs would need to begin collecting this information.

irs' ability to ensure compliance is impeded when it cannot verify the accuracy of information reported on tax returns. Independent verification of child support payments could be difficult. In cases where child support agreements are overseen by a court or a state or local social services agency, the overseeing agency may be able to report to the irs the amounts of child support paid, irs could use such a report to verify the accuracy of child support data used by taxpayers claiming the eic. However, systems would need to be developed to routinely report this data to ms.

When courts or social service agencies do not oversee child support payments, third party verification of payments may be unavailable. Furthermore, the parents could have incentives to misreport the amounts paid. That is, a custodial parent might claim that payments were not made because such payments could affect their ability to claim the child as a dependent for tax purposes. Noncustodial parents who had not properly paid child support would have an incentive to claim they had paid it if for no other reason than to avoid child support enforcement procedures.

197

Briefing Section IV

EIC Eligibility

G^0 Illegal Alien Recipients

No reliable data on number of illegal aliens who received EIC

• IRS suspects that more than 160,000 received EIC in 1994

Many more may have received it

• Duplicate uses of SSNs

• Temporary identification numbers

Some Illegal Aliens The Internal Revenue Code does not prohibit illegal aliens from receiving

Receive EIC tne nc if they meet the prescribed eligibility requirements, irs forms do not

require illegal aliens to identify themselves as such; therefore, res does not know how many illegal aliens may be claiming and receiving the EIC.

IRS needs an identification number, generally the taxpayer's ssn, to process a tax return, irs assigns a temporary Taxpayer Identification Number (tin) when any taxpayer files a return with an invalid ssn, a blank space, or the

GAO/GGD-9S-422BR Earned income Credit

198

code "205(c)."'4 The designation 205(c) is often used by taxpayers to indicate they are not eligible to receive an ssn. Thus, ms officials said taxpayers who enter this code are likely to be illegal aliens.

Limited data from manual reviews under the 1994 eic Compliance Initiative show that a minimum of 160,000 taxpayers,15 out of about 8.7 million who filed paper returns claiming the eic, entered 205(c) instead of an ssn for a qualifying child.16 Given use of the 205(c) code for qualifying children and their enforcement experience, ms officials believe these returns likely were filed by illegal aliens, irs expects most of these refunds to be denied because taxpayers will not be able to support their claims by verifying that the dependent met the age, relationship, and residency requirement

Some unknown portion of returns filed with ssns may also be filed by illegal aliens. For example, if illegal aliens use ssns of other individuals when filing a return and ms does not detect the ssn duplication, they may receive an Eic refund, ms' new efforts to detect duplicate uses of ssns, if successfully implemented, should reduce the number of illegal aliens as well as U.S. citizens incorrectly receiving tax refunds.

'The designation 205(c) refers to section 205(c) of the Social Security Act as amended whi the rules for issuing SSNs Most legal aliens can obtain an SSN if they meet the applicable requirements. Section 20G(c | di «a not refer specifically to illegal aliens.

I6EIC claimants are required to provide a TIN for qualifying children

i.Aii ..(,Ii-95-I22BR Earned Income Credit

199

GPD Excluding Illegal Aliens

Administration's compliance proposal

• Exclude illegal aliens from EIC eligibility

• Require recipient to have an SSN

• Allow IRS to use streamlined enforcement procedures

Administration estimates $400 million in reduced EICs in fiscal year 1997

Proposals to Exclude Illegal Aliens From EIC Eligibility

A Senate bill introduced in 1994 by Senator Roth and the administration's Tax Compliance Act of 1995 (H.R. 981 and S. 453) introduced in 1995 would deny the Eic to illegal aliens. Illegal aliens cannot be employed lawfully in the United States. On the other hand, the eic, which is intended in part to encourage employment, under current law, can be paid to illegal aliens. Thus, the eic works at cross purposes with the prohibition on employment of illegal aliens.

200

Briefing Section IV

The administration's proposal would require that all eic recipients provide ssns that are valid for employment in the United States for themselves, for their spouses, if applicable, and for qualifying children. Because illegal aliens cannot qualify for ssns that are valid for employment in the United States, they would not be able to receive the eic.

The ssa provides the irs with a computer tape with names and ssns of those individuals having ssns. The data include codes indicating whether the ssn is for employment purposes or other purposes, such as qualifying for Social Security benefits. This data provides a tool for irs to use in enforcing the administration's proposal. We have not assessed the adequacy of the information in the tape for enforcement purposes or the potential for illegal aliens to fraudulently receive work-related ssns from the ssa.

Finally, the administration's proposal would permit irs to use streamlined procedures to enforce the requirement that eic claimants have valid work-related ssns. The administration proposes that irs be permitted to notify taxpayers who do not provide valid ssns that they are not eligible for the eic. Within 60 days, taxpayers would either have to provide valid ssns or request that irs follow deficiency procedures. Deficiency procedures protect taxpayers' rights through notices to the taxpayer and opportunities for rebuttal of irs' concerns and petition to the Tax Court Taxpayers who fail to respond within 60 days to irs' proposed notice regarding lack of valid ssns would be required to refile an amended return with correct ssns to obtain the eic.

The administration estimates that requiring all eic recipients to provide valid work-related ssns and using streamlined procedures to enforce this requirement would yield about $400 million in additional revenue in fiscal year 1997.

i. vuic.Ji sir. !._'iK turned Income Credit

201

Appendix I

Objectives, Scope, and Methodology

As requested, our objectives were to (1) present information about eic noncompliance and what steps res is taking to control noncompliance and (2) review the impact on the amount of eic paid that might result from potential changes to the eic eligibility criteria that would reflect taxpayer wealth and additional sources of income and administrative issues that could arise due to these changes. In addition, we were asked to provide information about illegal aliens receiving the Eic and to describe the administration's proposal, which is similar to Senator Roth's 1994 proposal, to exclude illegal aliens from eligibility.

In responding to all of these objectives, we met with and obtained reports and data from officials with the Department of the Treasury and res' national office. Primarily in response to the objectives concerning noncompliance and illegal aliens receiving the Eic, we also met with officials and reviewed relevant procedures and data in ms' Cincinnati, OH, and Fresno, CA, Service Centers.

Specifically for our objective of assessing the effects of certain revisions to the criteria used in determining the amount, if any, of eic that is awarded,

reviewed relevant literature on the eic to understand its requirements and to determine the results of others' analyses of eic eligibility modifications; met with staff from the jct to discuss advantages and disadvantages of using possible proxy measures of wealth to determine eic eligibility; interviewed Congressional Budget Office analysts regarding their consideration of modifications to eic eligibility requirements; met with Department of the Treasury Office of Tax Analysis staff to discuss policy and administrative issues associated with eic eligibility modifications; and

met with res national office officials, including the National Director of Submissions Planning; the Chief of Service Center Compliance; the Director of the Service's Tax Forms and Publications Division; and the National Director, Applications Design and Development Management, to discuss administrative issues associated with eic eligibility modifications.

In addition, we performed statistical analyses to make preliminary estimates of the potential effects on the overall amount of eic program costs and the number of eic recipients of using measures of taxpayers' wealth and additional sources of their income when determining how much, if any, eic would be awarded. After we determined the general magnitude of change that might result, we discussed our preliminary

202

Objectives, Scope, and Methodology

results with Senator Roth and he requested revenue estimates, shown in our study, from the jct. "

To do our analysis, we first determined how much eic would have been awarded and to how many recipients had the eic rules that will be in effect in 1996 actually been in effect in tax year 1992. Applying the 1996 rules, which increase the credit amount and the number of individuals who are eligible for the Eic, provided us with a more realistic indicator of the potential effects of changing the eic eligibility rules.

Using this as a base measure, we computed the amount of eic that would have been awarded and the number of recipients if (1) wealth measures of varying breadth coupled with varying cutoff thresholds had been used in determining eic eligibility and (2) nontaxed Social Security income, nontaxed pension distributions, and tax-exempt income had been included in taxpayers' acis in determining how much, if any, eic they would have received.

To assess the ttkely effects of changing the eic eligibility criteria, we obtained and analyzed data from the Internal Revenue Service's (irs) Statistics of Income Division (soi) and from the Bureau of the Census' Current Population Survey (cps).18 The cps data was needed because soi only has data that can be obtained from tax returns. Critically absent from soi was data related to nontaxable Social Security income. (Social Security income is reported on Forms 1040 and 1040A, but only to the extent that some portion of the income is subject to tax.) We used tax year 1992 soi data and March 1993 cps data, which incorporates economic data for the calendar year 1992, because it conformed to the same period as the most recent soi data available.

Using these data, we simulated the effect of broadening the definition of agi to include nontaxed Social Security income as well as nontaxed private pensions and tax-exempt interest income. (The latter two items are reported on tax forms.) We used Census' simulation of hc recipients on its cps data set to estimate the Social Security income received by the actual soi population. Overall, we believe that this simulation procedure yielded a

170ur methodology produces an estimate given the taxpayer income c 1992 but assuming that the 1996 EIC rules had then appued The n which provides an indication of future changes in revenues due to changes n takes into account forecasted changes in interest rates and other e

GAO/GGD-96 122BK Earned Income Credit

Objective!, Scope, and Methodology

conservative estimate of the reductions in nc program < number of affected eic recipients.

GAO/GGD-85-12ZBR Earned Income Credit

204

Appendix II

Major Contributors to This Report

r, i /-< .*„>.,.. Michael Brostek, Assistant Director, Tax PoUcy and

General Government Administralion i^

Division, Washington, James A. Wozny, Chief Tax Economist, Tax Policy and

rj Q Administration Issues

Mary Phillips, Assignment Manager Clifford Tuck, Economist-in-Charge

TinHnnati Rao-innal William Bricking, Evaluator-in-Charge

Cincinnati Kegionai E|len Soltow EvaiuatOT Office

GAO/GGD-St-IZIBK Earned Income Credit

Glossary

Income

205

Taxable Earned Income

For the eic taxable earned income includes (1) wages, salaries, and tips, (2) union strike benefits, (3) long-term disability benefits received prior to i retirement age, and (4) net earnings from self -employment.

Nontaxable Earned Income

Among the earned income items that are nontaxable are (1) voluntary salary deferrals, such as 401(k) plans or the federal thrift savings plan, (2) pay earned in a combat zone, (3) basic quarter and subsistence allowances from the U.S. military, (4) housing allowance or rental value of a parsonage for the clergy, and (5) excludable dependent care benefits.

Unearned Income

Items that are not earned income include (1) interest and dividends, (2) Social Security and railroad retirement benefits, (3) welfare benefits (including afdc payments), (4) pensions or annuities, (5) veterans' benefits, (6) workers' compensation benefits, (7) alimony, (8) child support, (9) unemployment compensation (insurance), (10) taxable scholarship or fellowship grants (not reported on Form W-2), and (11) variable housing allowance for the military.

Adjusted Gross Income

In addition to taxpayers' earned income, AG! includes their income from other sources, such as investments, alimony received, and unemployment compensation.

Qualifying Child

A qualifying child (1) is an eic claimant's son, daughter, adopted child, grandchild, stepchild, or foster child, (2) is under age 19 or under age 24 and a full-time student or any age and permanently and totally disabled, and (3) lives in the claimant's home in the United States for more than half of the year (or all of the year if a foster child).

Working Poor

The term "working poor," while used in reference to the intended beneficiaries of the eic, is not defined in statute. Generally, for purposes of this report, we use the term to refer to those individuals who meet the current eic income criteria, or the revised criteria that we analyze. The revised criteria do not alter the basic eic income criteria, but rather include a fuller range of potential eic recipients' resources in determining whether basic eligibility criteria are met

206

TESTIMONY OF

U.S. SENATOR DON NICKLES

SENATE COMMITTEE ON GOVERNMENT AFFAIRS

The Earned Income Tax Credit April 5, 1995

MR. CHAIRMAN, I want to thank you and the ranking member for inviting me to address the Committee on Government Affairs about an out-of-control federal spending program whose faults have been too long overlooked.

Mr. Chairman, the earned income tax credit, or EITC, was enacted in 1975 to offset payroll taxes for low-income families with children and provide an incentive for work. The credit was first expanded in 1990, and it was again expanded dramatically in 1993 as part of President Clinton's tax bill. Unfortunately, what began as small work "bonus" has ballooned into a massive wealth redistribution program.

During his state of the union address in January, President Clinton called the 1993 EITC expansion a "working family tax cut".

"We took the first step in 1993 with a working family tax cut for 15 million families with incomes under $27,000, a tax cut that this year will average about $1,000 a family." (Bill Clinton, State of the Union, January 24, 1995)

In fact, Mr. Chairman, the EITC is not a tax cut. It is the federal government's fastest growing and most fraud-prone welfare program.

As the chart I have brought with me illustrates, the EITC will cost $17 billion this year, more than twice as much as it cost in 1992. EITC growth rates for the last four years are 55%, 18%, 22%, and 55% respectively, and the program's cost continues to grow out of control into the near future, reaching $26 billion in fiscal year 2000. Amazingly enough, the EITC will in fiscal year 1996 eclipse the federal cost of Aid to Families with Dependent Children.

With regard to the EITC's original purpose of reducing the tax burden on working families, consider the following facts. Ninety-percent of the cost of the EITC is direct handout, or federal outlays paid directly to individuals who have zero income tax liability. Only ten percent of the cost of the EITC is a tax refund. Consider further that, although the EITC is supposed to encourage work, the Government Accounting Office found that the average EITC recipient worked only 1 ,300 hours a year, compared with a normal work year of 2,000 hours. The maximum EITC is equally available to both a fry-cook who works 2,000 hours per year at $5.00 per hour and a part-time lawyer who works 100 hours per year at $100 per hour. Finally, the President's expansion of the EITC opened the program up to taxpayers without children, a dramatic departure from the program's original purpose.

The most unsettling part of the EITC story, Mr. Chairman, brings us to the purpose of your hearings yesterday and today. Numerous studies of the EITC by the IRS, GAO, and others

207

have revealed massive program losses due to fraud and error. It is my understanding that you, Mr. Chairman, have estimated that between 30 percent and 40 percent of all EITC benefits are lost to fraud and error, which is an amazing statistic.

The primary sources of EITC fraud are people who falsely claim they have children or understate the age of their children, people who fabricate jobs, and married couples who claim to be divorced. Studies and hearings have also revealed growing tax return preparation fraud, whereby tax preparers recruit low-income people, set-up fictitious companies to pay the recruits, and then file electronic returns claiming refunds which are split with the recruits.

Why does the EITC attract such abuse? Because the EITC offers big checks. The maximum credit for a multiple-child family is $3,1 14 in 1995, and although recipients can elect to collect the credit in equal paycheck installments rather than in lump sum, less than one-half of one percent choose to do so according to the GAO.

Mr. Chairman, the IRS has come under fire this year for their initiative to reduce fraud by cross-checking Social Security numbers on all returns and by running returns seeking refundable credits through computerized screening systems and holding refundable credits until the return is verified. While the delays these activities may cause law-abiding taxpayers are regrettable, I applaud the IRS for this initiative.

However in order to curb this program's massive growth, we must do more than reduce fraud and error. We must overhaul the entire program and reduce its cost.

As a member of the Senate leadership's working group on entitlements, I was asked to examine options for reducing the cost of the EITC. What I discovered was that the massive expansion of the program in 1 993 must be reversed. Congress took the first step with the recent passage of the self-employed health deduction bill, which included a provision to'deny EITC benefits to taxpayers with significant sources of passive income. I think we couid have done even more with that provision, Mr. Chairman, to ensure that taxpayers with substantial assets do not qualify for the EITC.

I believe the EITC credit levels which Congress doubled in 1993 must be scaled back, and the top income eligibility levels of $24,394 to $26,691 must be reduced to more closely reflect need. Further, the expansion of the EITC to taxpayers without children should be repealed. Finally, we should examine other options, including proposals to fold this program into a welfare block grant to the states.

Mr. Chairman, I commend you for holding these hearings and I look forward to working with you as these issues are addressed by both the Government Affairs and Finance Committees.

EARNED INCOME TAX CREDIT

	Credit	Maximum	Min Income	Phaseout	Federal Cost	%
YEAR	Percent	Credit	for Max Credit	Income	(in billions)	Growth
1976	10.0%	$400	$4,000	$8,000	$0.8
1977	10.0%	$400	$4,000	$8,000	$0.9	12%
1978	10.0%	$400	$4,000	$8,000	$0.9	0%
1979	10.0%	$500	$5,000	$10,000	$0.8	-11%
1980	10.0%	$500	$5,000	$10,000	$1.3	63%
1981	10.0%	$500	$5,000	$10,000	$1.3	0%
1982	10.0%	$500	$5,000	$10,000	$1.2	-8%
1983	10.0%	$500	$5,000	$10,000	$1.2	0%
1984	10.0%	$500	$5,000	$10,000	$1.2	0%
1985	11.0%	$550	$5,000	$11,000	$1.1	-8%
1986	11.0%	$550	$5,000	$11,000	$1.4	27%
1987	14.0%	$851	$6,080	$15,432	$1.4	0%
-1988	14.0%	$874	$6,240	$18,576	$2.7	93%
1989	14.0%	$910	$6,500	$19,340	$4.0	48%
1990	14.0%	$953	$6,810	$20,264	$4.4	10%
1991	17.3%	$1,235	$7,140	$21,250	$4.9	11%
1992	18.4%	$1,384	$7,520	$22,370	$7.6	55%
1993	19.5%	$1,434	$7,750	$23,050	$9.0	18%
1994	30.0%	$2,528	$8,425	$25,299	$11.0	22%
1995	36.0%	$3,114	$8,650	$26,691	$17.0	55%
1996	40.0%	$3,560	$8,900	$28,524	$20.0	18%
1997	40.0%	$3,668	$9,170	$29,387	$23.0	15%
1998	40.0%	$3,780	$9,450	$30,289	$24.0	4%
1999	40.0%	$3,900	$9,750	$31,249	$25.0	4%

Tata for FY9I through FY99 for multiple-child beneficiaries

Provided by the Office of Senator Don Nickles, 04/04/95

209

210

FOR RELEASE AT 10:00 A.M. April 4, 1995

STATEMENT OF

LESLIE B. SAMUELS

ASSISTANT SECRETARY (TAX POLICY)

DEPARTMENT OF THE TREASURY

BEFORE THE

COMMITTEE ON GOVERNMENTAL AFFAIRS

UNITED STATES SENATE

Chairman Roth and Members of the Committee:

I am pleased to have the opportunity to discuss the goals, design, and effectiveness of the Earned Income Tax Credit (EITC). The Administration is strongly committed to the goals of the EITC which are to make work pay and to lift workers out of poverty in the most efficient and administrable manner possible. The Administration's commitment to the EITC program for low-income working families has been demonstrated through a number of legislative and administrative actions since January, 1993. In February 1993, we proposed a significant expansion of the EITC in order to improve its effectiveness both as a work incentive and an income support program. With certain modifications, Congress enacted the Administration's proposals as part of the Omnibus Budget Reconciliation Act of 1993 (OBRA 1993).

Since the passage of OBRA 1993, we have proposed further steps to improve the administration and targeting of the EITC, while reducing the costs of the program. Four of these proposals were included in the Uruguay Round Agreement Act of 1994 (URAA). As a consequence of that legislation, the EITC is denied to nonresident aliens and prisoners; taxpayers are reqtrired to provide a taxpayer identification number for each EITC qualifying child regardless of age; and the Department of Defense is required to report to both the IRS and military personnel the non-taxable earned income used in computing the EITC.

We have also made several proposals which are still pending final legislative action. In this year's budget, we proposed that the EITC be denied to taxpayers with $2,500 or more of interest and dividend income. A similar provision is included in the conference agreement on H.R. 83 1 , a bill to extend and expand the 25 percent deduction for health insurance costs incurred by self- employed individuals. The budget also includes proposals to deny the EITC to undocumented workers and provide the IRS with the authority to use simpler and more efficient procedures

211

-2-

when taxpayers fail to supply a valid social security number. In addition, last year's Administration welfare reform bill proposed demonstration projects to test alternative methods of administering advance payments of the EITC. We hope that Congress will act on these outstanding proposals.

As Commissioner Richardson will testify, the Administration has taken other significant actions to strengthen the integrity of the EITC program. We have expanded our outreach efforts to ensure that eligible low-income individuals are aware of the EITC and the advance payment option. We have also conducted studies of EITC compliance and the broader issue of refund fraud. Last spring, Secretary Bentsen appointed a Task Force to conduct an independent investigation of the refund fraud, and Under Secretary Noble presented their interim findings and call for aggressive action to the Ways and Means Oversight Subcommittee last October. This year, we have intensified our scrutiny of returns claiming the EITC in order to prevent erroneous refunds from being paid to ineligible individuals.

Members of your Committee are considering ways in which the EITC could be redesigned to reduce programmatic costs and non-compliance while increasing the work incentives associated with the program. Before considering significant changes to the program that makes work pay, we would urge Congress to wait until we have had time to observe the effects of both recent legislation and our enhanced compliance efforts. Moreover, we are concerned that some alternative proposals to redesign the EITC would actually cause both non-compliance and work disincentives to increase. In particular, we would strongly object to proposals which would add complexity to the EITC eligibility criteria and which would thereby be difficult for the IRS to administer.

In the remainder of my testimony, I will discuss in some detail the goals of the EITC and the actions taken by the Administration to strengthen the effectiveness of the EITC, as well as our views regarding suggestions of possible modifications to the EITC.

Description of Earned Income Tax Credit for Low-Income Workers

The EITC is a refundable tax credit available to a low-income worker who has earned income and meets certain adjusted gross income (AGI) thresholds. Because the credit is refundable, individuals can receive the full amount to which they are entitled, even if the amount exceeds their income tax liability. The amount of the credit increases significantly if an individual has one or two qualifying children.

To be eligible for the EITC, a taxpayer must reside in the United States for over six months. Nonresident aliens are not entitled to the EITC beginning in 1995.

A child qualifies a filer for a larger EITC by meeting relationship, residency, and age tests. To meet the relationship test, the individual must be a child, stepchild, descendent of a child, or foster child of the taxpayer. The child must generally reside with the taxpayer in the United States for over half the year. For foster children, the residency test is extended to the full year.

212

A qualifying child must be under the age of 19 (24 if a full-time student) or be permanently and totally disabled.

By tax year 1997, a taxpayer must provide a taxpayer identification number (TIN) for each qualifying child. Social security numbers generally serve as TINs. Some taxpayers are unable to obtain social security numbers. Under section 205(c) of the Social Security Act, social security numbers are generally issued only to individuals who are citizens or who are authorized to work in the United States.

Structure of the Credit. The credit is determined by multiplying an individual's earned income by a credit percentage. For a family with only one qualifying child, the credit percentage for 1995 is 34 percent. The credit amount increases as income increases, up to a maximum income threshold. For 1995, the income threshold is $6,160. Therefore, if there is only one qualifying child, the maximum basic credit for 1995 is $2,094 (34 percent of $6,160).

The credit is reduced and eventually phased out once AGI (or, if greater, earned income) exceeds a certain phase-out threshold. For 1995, the phase-out threshold is $11,290. The phase-out is accomplished by reducing the credit by a phase-out percentage. In 1995, for a family with only one qualifying child, the credit is reduced by an amount equal to 15.98 percent of the excess of AGI (or, if greater, earned income) over $11,290. The credit is completely phased out and is no longer available to taxpayers with incomes above the end of the phase-out range. In 1995, this income level is $24,396. The income thresholds for both the phase-in and phase-out ranges are adjusted for changes in the cost of living.

If there are two or more qualifying children, the credit percentage, income thresholds, and phase-out percentage are higher. For 1995, the credit percentage for families with two or more children is 36 percent of the first $8,640 of earned income. Filers with earnings between $8,640 and $11,290 are entitled to the maximum credit of $3,110 (36 percent of $8,640).

The phase-out percentage for these families is 20.22 percent. As in the case of the credit for families with one child, the credit is phased out starting at $1 1,290. However, the phase-out range for families with two or more children extends to $26,673.

In 1996, the credit percentage for families with two or more children will increase to 40 percent of the first $8,900. Filers with earnings between $8,900 and $1 1,620 will be entitled to the maximum credit of $3,560 (40 percent of $8,900). The phase-out percentage will also increase to 21.06 percent, and the phase-out range would extend to $28,524. Thereafter, the income thresholds for both the phase-in and phase-out ranges would be adjusted for changes in the cost of living. (The dollar amounts shown for 1996 are estimates.)

Workers without children may claim the EITC if they are between 25 and 64 years of age and are not claimed as a dependent on another taxpayer's return. For these workers, the basic credit is 7.65 percent of the first $4,100 of earned income. In 1995, the phase-out range for these workers is between $5, 150 and $9,230 of AGI (or, if greater, earned income). The phase-

213

out percentage is also 7.65 percent. The income thresholds for both the phase-in and phase-out ranges are adjusted for changes in the cost of living.

Figures 1 and 2 show the EITC credit structure for 1995 and 1996, respectively.

Advance Payments of the EITC. There are two ways to receive the EITC. Individuals can claim the credit by completing a Schedule EIC when filing their tax return at the end of the year. Alternatively, individuals with qualifying children may elect to receive a portion of their EITC in advance by filing a Form W-5 with their employer. These individuals are entitled to receive on an advance basis up to 60 percent of the credit allowable for a family with one qualifying child. The employer is not required to verify a person's eligibility for the credit.

At the end of the year, the employer notifies both the IRS and workers of the actual amounts of advance credits paid to individual workers on the Form W-2. When filing tax returns at the end of the year, these workers reduce the amount of EITC claimed by the amount of advance payments received.

Questionable Claims: The IRS must follow normal deficiency procedures when investigating questionable EITC claims. First, contact letters requesting additional information are sent to the taxpayer. If the necessary information is not provided by the taxpayer, a statutory notice of deficiency is sent by certified mail, notifying the taxpayer that the adjustment will be assessed unless the taxpayer files a petition in Tax Court within 90 days. If a petition is not filed within that time and there is no other response to the statutory notice, an assessment is made in which the EITC is denied.

Goals of the EITC

In developing the Administration's agenda for the EITC, we have been guided by the three basic principles of tax policy: efficiency, fairness, and simplicity. Specifically, we have sought expansions and modifications to the EITC in order to achieve the following four goals:

(1) to make work pay for those who might otherwise be on welfare;

(2) to ensure that an individual who works full time throughout the year will not live in poverty;

(3) to target benefits to those with the greatest needs while minimizing distortions; and

(4) to make it easier for eligible individuals to claim the credit and for the IRS to verify their eligibility.

I would like to address each of these four goals in more detail.

For low-income families, the EITC makes work pay in two ways. Unlike many other assistance programs for low-income families, the EITC is limited to working families. Moreover, the credit amount initially increases - rather than decreases - for each additional dollar of earnings. As a consequence, the EITC is different from other low-income assistance

214

-5-

programs that are characterized by a reduction in benefits for each additional dollar of earnings. The EITC significantly increases the marginal return from working for both those who do not work at all and those who work less than full-time at minimum-wage jobs throughout the year.

The positive link between the EITC and work can also offset the work disincentives created by other tax and transfer programs. Workers are taxed at the combined employer and employee rates of 15.3 percent on the first dollar of earnings for the old-age, survivors, disability and health insurance (OASDHI) programs. Beyond a relatively low income threshold, food stamp benefits are reduced by 24 cents for each additional dollar of earnings. The EITC. with its positive credit rate on low earnings, is the only program designed to help offset the marginal tax rates imposed by these other programs.

A person who works at a full-time job for the entire year should not live in poverty. The Federal government assists low-income families in a number of ways. The Federal government requires employers to pay workers at least the minimum wage, and provides direct assistance to families through food stamp benefits and the EITC. For most families dependent on a worker who earns the minimum wage, it takes both food stamp benefits and the EITC to lift them out of poverty.

Last month, Secretary Rubin visited a VITA site here in the District of Columbia. At the site, he met Rhonda Clark, a mother from Maryland. Talking of her experiences with the EITC, Ms. Clark said, "I enjoy working and I want to continue. The EIC gives me some of the help I need - to keep working, to stay independent, and to support my family. It's a help I can not do without." Ms. Clark's experience provide a vivid example of how the EITC makes a difference in people's lives by encouraging them to work and providing them with additional assistance.

The benefits of the EITC should be targeted to families with the greatest needs and to those who can be best served by the positive incentives associated with the EITC. As a consequence, the credit rate is highest at very low earning levels, thus reaching individuals who are often making the critical choice between work and welfare. Because larger families have greater needs than smaller families, taxpayers with two or more children are entitled to a larger EITC than taxpayers with one or no children.

Families with incomes slightly above the poverty level also require assistance. Wages have stagnated for many workers and declined markedly for low- wage workers. Between 1973 and 1993, real hourly wages of full-time male workers at the tenth percentile (that is, those whose wages are just above those of the lowest-paid 10 percent of workers) declined 16 percent, while real hourly wages at the median fell 12 percent. By providing the EITC to families with incomes of up to $28,524 in 1996, the program provides a cushion to protect them from the effects of wage stagnation.

We recognize that the targeting of the EITC to the neediest workers can also create work disincentives. First, the ETTC increases the income of all recipients, allowing them to maintain

215

their standard of living with less work effort. For very low-wage workers, these negative effects are largely offset by the fact that the credit also increases their after-tax wage rate and thus the pay-off to work. As incomes increase above $11,290, EITC benefits begin to phase-out. As a consequence, the marginal tax rates for families of modest means increase. Among recipients in the phase-out range, the EITC could cause some individuals, primarily the spouses of other workers, to reduce the numbers of hours worked in response to higher marginal tax rates.

In this regard, the EITC is similar to any benefit program which targets assistance to the very neediest families. We cannot target assistance to low-income families without causing marginal tax rates to increase for families with slightly higher income. However, we can seek to minimize such distortions.

The Committee asked what changes would reduce marginal tax rates for taxpayers in the phase-out range. The positive EITC rate for very low-wage workers could be lowered, or the earnings threshold for the positive rate could be lowered. Both of these approaches would reduce the maximum credit. Alternatively, the maximum credit might not be changed, but would begin to phase-out at lower levels of income. These three approaches reduce the marginal tax rates for most families in the phase-out range, but only by imposing significant costs on families with much lower incomes. Reducing the EITC rate would, for example, make work far less attractive for those currently outside the workforce. For these reasons, we strongly oppose these three approaches that reduce the incentives to work for low-wage workers.

Alternatively, the phase-out range could be extended to higher levels of income. This approach lowers the marginal tax rates for those workers currently in the phase-out range, but extends eligibility for the EITC to families with higher levels of income and would increase their marginal tax rates.

In designing a program to make work pay, it is impossible to reduce the marginal tax rates in the phase-out range without either increasing the marginal tax rates on other families or reducing the credit amount for most families. There is a trade-off between the appropriate level of the EITC and the length of the phase-out range. A balance must be struck between the benefits of a higher credit rate for very low-wage workers and the costs of either higher marginal tax rates for families in the phase-out range or a longer phase-out range. There is, in short, no simple solution. Changes that reduce marginal tax rates for one group of workers require increases in marginal tax rates for other workers or reductions in benefits for working families. Proponents of alternatives should demonstrate how their proposals result in a better balance of costs and benefits than current law.

The fourth goal of the EITC is simplicity and verification. If eligibility rules are simple, taxpayers can more easily claim the EITC and avoid costly errors. With simple and verifiable eligibility rules, the IRS can also better ensure that the EITC is paid only to taxpayers who are eligible for the credit.

Simplicity is particularly important, because eligible individuals can claim the EITC directly

216

-7-

when they file their tax return. It is likely that this simple application process has contributed to high participation rates in the program. It has been estimated that between 80 and 86 percent of eligible persons claimed the EITC in 1990.

From the IRS's perspective, it is easier to verify eligibility for the EITC if the rules are simple. Moreover, because the IRS does not ordinarily interview EITC claimants, it is important that eligibility be based on criteria which can be verified as quickly as possible through independent reporting sources. Simplicity and verification prior to the payment of the EITC are key to the successful operation of the program.

Legislative and Administrative Actions in 1993 and 1994

As I outlined in the beginning of my testimony, the Administration and Congress have taken a number of legislative and administrative actions during the past two years in order to improve the effectiveness and administration of the EITC. I would like to review with you our accomplishments during this period.

OBRA 1993. OBRA 1993 expands the EITC and makes the program more effective in achieving its policy objectives.

First, OBRA 1993 increased the returns from working for those outside the workforce and for other very low-wage workers. (See Figure 3.) For very low-wage workers without qualifying children, the EITC offsets the employee portion of the OASDHI tax. For a family with one child, the credit rate for those with low earnings was increased by 1 1 percentage points from 23 percent to 34 percent. For a family with two or more children, the credit rate for those with earnings "below $8,900 in 1996 was increased by 15 percentage points from 25 percent to 40 percent. For low-wage workers with two or more children, the EITC will fully offset the combined employee and employer portions of the OASDHI taxes and the food stamp benefit reduction formula.

The OBRA 1993 expansion was also a critical step forward toward achieving the goal that a full-time worker should not live in poverty if he or she works throughout the year. In combination, a minimum wage job, food stamp benefits, and the EITC can lift a single parent with one or two children out of poverty. But, the income (including the EITC and food sumps and subtracting the employee portion of OASDHI taxes) of a family of four with only one full- time, minimum wage worker falls below the official poverty threshold. Prior to the passage of OBRA 1993, the poverty gap for a family of four would have been $2,435 in 1996. The OBRA 1993 expansion significantly closes that gap. Since the minimum wage has not kept pace with inflation, the job is not completed yet. This is why the President has proposed that the minimum wage be increased over two years by 90 cents.

OBRA 1993 reduced the poverty gap for minimum wage workers by increasing the maximum benefits by nearly $1,500 in 1996 for a family with two or more children. For these families, this increase in the maximum credit, without a change in the phase-out range, would

217

have resulted in a phase-out rate of 30 percent. In OBRA 1993, we tried to find a balance between the goals of providing low-income families with sufficient income support, while minimizing the marginal tax rates placed on families with higher, but still modest, levels of income.

Thus, the increases in the maximum credit were accompanied by changes in the income thresholds. For all families with children, the beginning of the phase-out range was lowered by about $1,600. As a consequence, the phase-out rate actually fell slightly for a family with one child since the end of the phase-out range was left unchanged. To reduce the disincentive effect of high marginal tax rates, eligibility for the EITC was extended to families with two or more children that have incomes of up to $28,524 (or about $3,000 above the prior level). The combination of these factors increased the phase-out rate from 17.86 percent to 21.06 percent, rather than 30 percent.

While the definitive effect of OBRA 1993 can not be measured yet, we believe that the legislation will, on net, increase work effort. While some workers with larger families will face slightly higher marginal tax rates, they are unlikely to change their behavior much in response. These are individuals who are already very attached to the work force. They cannot easily adjust their hours of work in response to a small change in tax rates; they need both their jobs and the EITC to meet their day-to-day needs, and most employers will not allow them the discretion to work fewer hours. The effect of the higher marginal tax rates on some workers in the phase-out range will likely be far outweighed by 'he effect of the increase in the credit rate. By making work pay, the OBRA 1993 increase in the credit rate will encourage non- workers to enter the workforce and other low-income part-time workers to increase their hours of work.

Finally, OBRA 1993 simplified the eligibility criteria for the EITC beginning in 1994 by eliminating the two supplemental credits for health insurance coverage and for taxpayers with children under 1 year of age. These two supplemental provisions added several paragraphs to the instructions, 10 additional lines on the Schedule EIC, and two additional look-up tables. The IRS could not easily verify eligibility for the supplemental credits because it did not receive independent verification of taxpayers' eligibility for them. These changes should improve compliance by reducing errors and improving verification.

URAA. URAA contains several provisions to improve the targeting of the EITC to those with the greatest need. Under this legislation, nonresident aliens are denied the EITC beginning in 1995. Under prior law, nonresident aliens could receive the EITC based on their earnings in the United States, even though they were not required to report their world-wide income to the IRS. Thus, it was possible for a wealthy foreign student to obtain the EITC based on his or her earnings as a teaching assistant sft an American university.

In addition, prisoners will not be eligible for the EITC based on their earnings while incarcerated. In the past, prisoners generally would not have been able to claim the EITC because they did not reside with a qualifying child for over half the year. When the EITC was

218

made available to workers without children in 1994, it became possible for prisoners to receive the EITC based on their earnings at prison jobs. Because this provision was made effective for tax year 1994, the EITC will not be paid to these individuals.

URAA also contained two provisions to improve the administration of the EITC. By 1997, taxpayers will be required to provide TINs for all dependents and EITC qualifying children, regardless of their age. By requiring EITC claimants to provide the TIN of all children, regardless of age, URAA improves the ability of the IRS to verify the eligibility of a taxpayer for the EITC.

Under the legislation, the Department of Defense is required to provide military personnel and the IRS with information regarding housing and subsistence allowances. These allowances are not taxable, but are includable in earned income for determining eligibility for the EITC. The savings from this provision are somewhat offset by another provision which extends EITC eligibility to military personnel stationed abroad. The IRS will be better able to verify eligibility of military personnel because of the requirement to report non-taxable earnings to both the taxpayer and the IRS.

Administrative Actions. The Administration has taken a number of steps to ensure that eligible individuals know about the EITC and the advance payment option. While many eligible persons receive the EITC, fewer than 1 percent of EITC claimants receive the credit through advance payments. The reasons for the low utilization rate are not fully known. One possible explanation is that workers simply do not know that they have the option of claiming the credit in advance. A General Accounting Office study in 1992 provided some support for this theory when investigators found widespread ignorance about the advance payment option among low- income workers.1

The Administration has intensified its efforts to alert taxpayers of their eligibility for advanced payments. As one of the first steps, President Clinton announced a Federal campaign in 1994 to enroll eligible government workers in the advanced payment system. The Treasury Department and a group of business executives have also joined forces to encourage private- sector employers to notify their workers about the advanced payment option. As required by OBRA 1993, the IRS sends out notices to EITC claimants after the filing season, informing them about the advance payment option and (although not required by the 1993 legislation) aiso supplying a Form W-5 for their use.

The Administration has also taken steps to ensure that those who are not eligible for the EITC do not receive it. As Commissioner Richardson will explain, the IRS has taken a number of responsible and needed steps to limit the EITC to those who are entitled to the credit. Beginning this year, the IRS is validating the social security numbers on tax returns claiming the

1 U.S. General Accounting Office. Earned Income Tax Credit: Advance Payment Option is Not Widely Known or Understood bv the Public. (GAO/GGD-92-26, February 19, 1992).

219

-10-

EITC. Refunds on returns with incorrect or missing numbers will be delayed while the IRS checks the accuracy of the refunds claimed. The IRS is also increasing its screening and review of all returns to ensure that only those taxpayers entitled to refunds receive them. As a consequence, refunds may be delayed on other questionable returns.

Finally, the IRS stopped providing Direct Deposit Indicators in the 1995 filing season to lenders who were providing refund anticipation loans. This action is also expected to reduce compbance problems that were associated with refund anticipation loans. The IRS's actions this filing season have been applauded as both responsible and necessary by Ways and Means Oversight Subcommittee Chairman Johnson and Ranking Member Matsui in a recent "Dear Colleague" letter to House members.

FY 1996 Budget Proposals

The Administration included several proposals to improve the targeting and administration of the EITC in this year's budget submission. We are ready to work with Congress on these proposals and hope that Congress promptly enacts them.

Deny EITC to taxpayers having more than $2.500 of taxable interest and dividends. Under this proposal, the EITC would be denied to taxpayers having more than $2,500 of taxable interest and dividends beginning in 1996. This threshold would be indexed for inflation thereafter.

This proposal would improve the targeting of the EITC to the families with the greatest need. Under current law, a taxpayer may have relatively low earned income and be eligible for the EITC, even though he or she has significant interest and dividend income. Most EITC recipients do not have significant resources and must rely on their earnings in order to meet their day-to-day expenses, but taxpayers with significant interest and dividend income can draw upon the resources that produce this income to meet family needs.

This proposal, with some modification, is included in the conference agreement for H.R. 831, a bill to extend and expand the 25 percent health insurance deduction for self-employed individuals. The conference agreement lowers the asset income threshold to $2,350 and expands the categories of income subject to the threshold to include tax-exempt interest and net positive rents and royalties. The asset income threshold would not be indexed.

In developing the Administration's proposal, we considered a broader list of asset income subject to the cap. We recognized that a broader list might increase equity, by treating the recipients of certain other types of asset income in the same manner as those who receive interest and dividend income. An expanded list would also reduce the incentive to choose a particular type of investment based on its tax or refund consequences. However, we were also concerned because the inclusion of net positive rents and royalties would also add some complexity to the determination of the EITC. These items are not reported separately on the Form 1040. We rejected the broader list of asset items because we were also concerned that

220

low-income taxpayers could not convert real estate holdings and other types of assets into cash as easily as savings accounts and stocks in a time of need.

While we do not oppose the inclusion of tax-exempt interest and net rents and royalties in the conference agreement on H.R. 831, we are very concerned about the asset income threshold not being indexed. We believe that the asset income threshold should be indexed in the same manner as all other income parameters for the EITC. Without indexation, the number of persons affected by this provision will increase over time. By 2000, the threshold would be equal to about $2,075 in 1996 dollars and would increase the number of affected taxpayers from about 550,000 to 650,000.

EITC Compliance Proposals. Under this budget proposal, only individuals who are authorized to work in the United States would be eligible for the EITC beginning in 1996. Taxpayers claiming the EITC would be required to provide a valid social security number for themselves, their spouses, and their qualifying children. Social security numbers would have to be valid for employment purposes in the United States. Thus, eligible individuals would include U.S. citizens and lawful permanent residents. Taxpayers residing in the United States illegally would not be eligible for the credit.

In addition, the IRS would be authorized to use simplified procedures to resolve questions about the validity of a social security number. Under this approach, taxpayers would have 60 days in which they could either provide a correct social security number or request that the IRS follow the current-law deficiency procedures. If a taxpayer failed to respond within this period, he or she would be required to refile with correct social security numbers in order to obtain the EITC.

In combination, these provisions would strengthen the IRS's ability to detect and prevent erroneous refunds from being paid out. In addition, the proposals would improve the targeting of the EITC by providing the credit only to individuals who were authorized to work in the United States.

Taxpayer Systems Modernization. The budget submission for the IRS contains funding for the continuation of its taxpayer systems modernization (TSM). We urge the Congress to continue to fund TSM. TSM is vital to the long-run efficiency of the IRS's collection functions. TSM will also enhance the IRS's ability to detect erroneous EITC claims.

Welfare Reform

In June 1994, the Administration introduced the Work and Responsibility Act (H.R. 4605). Since then, the Administration has indicated its desire to work together with the Congress to develop a proposal to reform our welfare system. One of the provisions in H.R. 4605 provided additional flexibility to States with respect to the EITC and should be considered as part of a comprehensive welfare reform package.

221

-12-

The proposal would allow four demonstration projects to determine the effects of alternative methods of delivering advance payments of the EITC. States would apply to the Department of the Treasury to provide advance payments of the EITC directly to eligible residents through a State agency. Such agencies could include food stamp offices, Employment Services, and State revenue departments. State plans would be required to specify how payment of the EITC would be administered. To finance these payments, States would reduce payments of withholding taxes (for both income and payroll taxes) from their own employees by the amount of the advance payments made during the prior quarter. The four selected projects could operate for three years beginning in 1996.

This pilot program is designed to determine whether another approach would be more effective for delivering advance payments than the current employer-based system. For example, a State could choose to allow all eligible EITC recipients to apply for advance payments. By receiving the credit as they earn wages, workers would observe the direct link between work effort and the EITC. Through a State program, individuals could have a choice of receiving the credit from a neutral third-party, without fear of the consequences of notifying their employers of their eligibility for the EITC. Moreover, they could receive assistance in determining the appropriate amount of the EITC to claim in advance.

A State could instead choose to target the advance payments of the EITC to welfare recipients - as a way of driving home the message that "work pays." These individuals may not know about the EITC, and how it can "make work pay," because they do not have to file a tax return if their adjusted gross incomes are below the tax thresholds (which are generally less than the poverty thresholds).

If the legislation passes, we will be evaluating these demonstration projects in order to understand better how individuals respond to receiving the advance payments of the EITC. We will be paying careful attention to whether the use of State agencies can increase both utilization of the advance payment system and labor force participation by non-workers.

States also have the resources to verify many of the eligibility criteria for the credit better than employers, reducing the risk of erroneous payments being made to ineligible persons. This option would also allow for an evaluation of alternative delivery systems on compliance.

Other Suggestions

The Administration will evaluate other proposals to modify the EITC program design by the same criteria we apply to our own proposals:

(1) Does the proposal make work more attractive to those outside the workforce and to others with minimal ties to the workforce?

(2) Does the proposal reduce the poverty gap for full-time workers?

(3) Does the proposal improve the targeting of the EITC to the neediest individuals and families in the least distortionary manner? and

-13-

(4) Does the proposal make it easier for eligible taxpayers to claim the EITC and for the IRS to verify their eligibility before refunds are paid out?

We are concerned that many of the options being discussed before the Committee do not meet these criteria.

In anticipation of this hearing, the Chairman requested a report by the General Accounting Office (GAO) on alternative measures of determining need for the EITC. Among the options considered by the GAO are proposals to expand the definition of income for purposes of determining eligibility for the EITC to include such items as non-taxable social security benefits and child support. We have reservations concerning these options and will provide our views when more information is available regarding the specific nature of the proposals.

Families receiving both the EITC and social security benefits would include some older partially retired workers, as well as full-time workers married to retirees. In both of these cases, they may still qualify for the EITC because they have minor children at home, students away at college, or permanently disabled sons and daughters who require their care. If a proposal also extends to families receiving disability insurance benefits, it would affect workers married to disabled persons who cannot work. To a small extent, a portion of workers' social security benefits represent the return of their own contributions from previously taxed income. Reducing their EITC benefits could also compound the work disincentives already present in the social security programs.

The tax system does not count child support as income to the custodial parent because child support payments are a continuation of the other parent's obligation to support his or her child. We are also generally concerned about proposals which would add complexity to the determination of EITC eligibility and would be difficult to verify. In particular, the IRS does not currently receive information about child support payments.

We also oppose other modifications outlined in the Chairman's invitation to testify today. Requiring a minimum annual number of hours of work to qualify for the EITC would fail to achieve its stated objective of targeting the credit to low-wage workers instead of highly paid part-time or part-year workers. Instead, it could deny the credit to many low-wage workers who may not be employed throughout the year, or who cannot find a full-time job. It could also reduce the incentive for low-wage part-time workers to increase their attachment to the workforce. In addition, this requirement would impose a significant reporting burden on taxpayers and their employers. It would be especially difficult to monitor the hours of work by self-employed individuals.

We would object to provisions that would base the EITC on both custodial and noncustodial parents' income. Strengthening our child support enforcement rules is a legislative priority of the Administration. However, we do not believe that this proposal would encourage the collection of child support payments. A more likely effect is to further penalize many single parents who cannot collect support payments from the absent parent. Even if they receive child

-14-

support payments, many low-income single parents need the assistance of the EITC to compensate for the additional costs they face raising a family on their own, such as the costs of child care while they work.

Finally, we would object to a restoration of pre-1991 law, which based eligibility for the EITC on the dependent and head of household rules. Under these provisions, married taxpayers were required to demonstrate that they provided over one-half of the support of an EITC qualifying child, while a head of household was required to show that he or she provided over one-half the costs of maintaining a home in which a qualifying child resides. These rules were very difficult for taxpayers to understand because they did not conform well to ordinary people's notions of caring for a child. Moreover, the IRS could not enforce these rules because information was not readily available on taxpayers' expenditures in support of their children or household. In response to concern about high error rates in the EITC program, the Treasury Department worked together with Congress to modify and simplify these rules in 1990. We believe that the current residency and relationship tests are simpler to administer and verify than the pre-1991 rules. Restoring the old rules would increase complexity and non-compliance.

Senate Republican Task Force on Entitlement Reform: Last month, the Senate Republican Task Force on Entitlement Reform released a draft report which included a recommendation to reduce expenditures on the EITC by $27 billion. A key component of the Task Force plan does not meet the criteria listed above for analyzing proposed changes to the EITC.

Under the Task Force plan, the income thresholds for determining EITC benefits would no longer be indexed. To meet the revenue target in the report, the benefit threshold levels in 1996 could not be adjusted for inflation since 1994. Under such a plan, EITC recipients would be entitled to a maximum benefit of $3,370 in 1996, a reduction of $190 relative to current law. The maximum benefit amount would not change after 1996. In 2000, about 17.8 million taxpayers (80 percent of total recipients) would have benefits reduced or eliminated by this proposal.

The Task Force report states that "a further expansion through eligibility indexation is unaffordable and unwarranted." But, indexation is not an expansion of the existing EITC program - it is necessary to ensure that taxpayers do not lose eligibility for the EITC. Under current law, an estimated 21.1 million taxpayers will claim the EITC in 1996. If benefit thresholds are not adjusted for inflation since 1994, participation would shrink to 18.2 million by 2000.

The Task Force justifies these reductions by stating that the "program contains rampant fraud and abuse, which the IRS has awkwardly tried to restrain this year." Apart from the proposals contained in the FY 1996 budget, all of the proposed savings would result from the elimination of indexation of the EITC benefit thresholds. Eliminating indexation does not address the issue of fraud and abuse at all. Instead, it denies eligibility for the EITC to millions of law-abiding working taxpayers and reduces the benefits of millions of others who are playing by the rules. The tax code contains a number of provisions which are indexed for inflation each year. These

224

-15-

include the personal exemption, standard deduction amount, and the width of the income tax brackets. It is inappropriate to suspend indexation on the one provision which is solely targeted to low-income taxpayers.

The Administration is committed to improving compliance with the EITC rules. Its actions in the last two years are clear evidence of this commitment. The compliance problems which the Administration is addressing should not be used as an excuse to eliminate or reduce the EITC benefits to all low-income working people. Consequently, the Administration strongly opposes this ill-conceived proposal.

This concludes my remarks. Thank you once again for providing me with the opportunity to testify. I would be pleased to answer any questions that the Committee may have.

225

		c
		£
		2
		£ .	2	c
		o :		2: :
If) s T-		£ i o : <5 '	O co 6	2 j 2 ! o : o i Z	/ -
*f		I			/
•o				/
d>
2:
O			/
a
H			/
0)			.''' /
E			/
o			/
o			y /
c			/
■o			/
0) c		O
(0 UJ		5"	I
0)			cm" ***** **
1-			**«»^
■ ■			*^s^ ■* /
^			**^N^ ▼" :
£			X. 8\
3			^S*.
CD			^w^
U.	1		1 1 1 1 1 1 ^^

8§

5-g

CO CO CM CM i-

s;unouJV UP&iO

226

8

T5

2 o

*

E

8

c

(0 UJ

a>

CM

a>

v. 3

	c
	ffl
	E>
	2	»'
	S -a « .
	O =	k- :
	_ : n	2 1	,'
	More neC	O j	j
	&"• ° £:
	i	y /
	H	/ /
		/
/
3,560 56
<* '.. *1
CM
*»
*».%
*»». ^^
\^S^ */
••«./\. CM = -
-:^>\ S\
'*Sjv
*^*s^\
1 1 1 1 1 1 1 ^

8

8

o a>

CO CO

CM CM i-

sjunowv JIPQJQ

227

		i	ll
		i	&
fc.		CO	m ;
%£		o	O	/ ~7
5*.
— w £
redi Pha Dolla
ieTaxC 3, Fully dren, 1996
o o> g			/
ned Inc 0BRA1 o or More
8(		||
J5tȣ III cc	si		5|
• « *
•C 9 a
2k			^N.
3 ffl
IL		1	1 1 1 1 1 ^S

8

1

5 J

o

E

Q 8

x —

I

§

8 8 8 8

IT) O lO O

CM* CM* t-* t-"

s;unouuv 1!P©JQ

FOR DELIVERY AT 10:00 A.M. April 5, 1995

ORAL STATEMENT OF

LESLIE B. SAMUELS

ASSISTANT SECRETARY (TAX POLICY)

DEPARTMENT OF THE TREASURY

BEFORE THE

COMMITTEE ON GOVERNMENTAL AFFAIRS

UNITED STATES SENATE

Chairman Roth and Members of the Committee:

I am pleased to have the opportunity today to discuss the goals, design, and effectiveness of the Earned Income Tax Credit (EITC). The Administration is strongly committed to the goals of the EITC, which are to make work pay and to lift workers out of poverty in the most efficient and administrable manner possible. The Administration's commitment to the EITC program for low-income working families has been demonstrated through more than a dozen legislative and administrative actions since early 1993.

Goals of the EITC

Since the creation of the EITC in 1975, bipartisan support for the program and its goals has been growing. Republicans and Democrats alike have viewed the EITC as the best viable alternative to current welfare programs. With its message of "work pays," the EITC helps reduce dependency on welfare and increase reliance on jobs. Congress has voted to significantly expand the EITC in three major pieces of legislation: first, with the Tax Reform Act of 1986 during President Reagan's administration; then with the Omnibus Budget Reconciliation Act of 1990 during President Bush's administration; and finally, with the Omnibus Budget Reconciliation Act of 1993 (OBRA 1993) during President Clinton's administration.

The EITC program, as expanded by 3 Presidents, has made a difference, as Secretary Rubin was recently reminded. Visiting a VITA site, the Secretary met Rhonda Clark, a mother from Maryland. Talking of her experiences, Ms. Clark said, "I enjoy working and I want to continue. The EIC gives me some of the help I need -- to keep working, to stay independent, and to support my family. It's a help I can not do without." Ms. Clark's experience provides a vivid example of how the EITC makes a real difference in people's lives by encouraging them to work and providing them with additional assistance.

goals:

-2-

The Administration is committed to the EITC and advancing the following four key

(1) to make work pay for those who might otherwise be on welfare;

(2) to ensure that an individual who works full time throughout the year will not live in poverty;

(3) to target benefits to those with the greatest needs while minimizing distortions; and

(4) to make it easier for eligible individuals to claim the credit and for the IRS to verify their eligibility.

I would like to address each of these four goals.

First, for low-income families, the EITC makes work pay in two ways. Unlike many other assistance programs for low-income families, the EITC is limited to working families. Moreover, the credit amount initially increases - rather than decreases - for each additional dollar of earnings.

The positive link between the FITC and work can help offset the work disincentives created by other tax and transfer programs, such as social security taxes and food stamp benefits. The EITC. with its positive credit rate on low earnings, is the only program designed to help offset the marginal tax rates imposed by these other programs.

A second goal is to ensure that a person who works at a full-time job for the entire year will not live in poverty. For most families dependent on a worker who earns the minimum wage, it takes both food stamp benefits and the EITC to lift them out of poverty.

Third, the benefits of the EITC should be targeted to families with the greatest needs and to those who can be best served by the positive incentives associated with the EITC. As a consequence, the credit rate is highest at very low earning levels where individuals are often making the critical choice between work and welfare. Because larger families have greater needs than smaller families, taxpayers with two or more children are entitled to a larger EITC than taxpayers with one or no children.

Families with incomes slightly above the poverty level also require assistance. Wages have stagnated for many workers and have declined markedly for low-wage workers. By providing the EITC to families with incomes of up to $28,524 in 1996, the program provides protection from the effects of wage stagnation.

The fourth goal of the EITC is simplicity and verification. If eligibility rules are simple, taxpayers can more easily claim the EITC and avoid costly errors. With simple and

verifiable eligibility rules, the IRS can also better ensure that the EITC is paid only to taxpayers who are eligible for the credit.

Simplicity is particularly important, because eligible individuals can claim the EITC directly when they file their tax return. It has been estimated that between 80 and 86 percent of eligible persons claimed the EITC in 1990.

Description of Earned Income Tax Credit for Low-Income Workers

The design of the EITC under current law reflects a balance among these four goals. The EITC program design is a direct result of actions taken by Congress and both prior and current Administrations. Let me review quickly the basic structure of the credit.

The EITC is a refundable tax credit available to a low-income worker who has earned income and meets certain adjusted gross income (AGI) thresholds. Because the credit is refundable, individuals can receive the full amount to which they are entitled, even if the amount exceeds their income tax liability. The amount of the credit increases significantly if an individual has one or two qualifying children.

To be eligible for the EITC. a taxpayer must reside in the United States for over six months. A child qualifies a filer for a larger EITC by meeting relationship, residency, and age tests. These criteria are described on the board.

The credit is determined by multiplying an individual's earned income by a credit percentage. For a family with only one qualifying child, the credit percentage for 1996 is 34 percent. The credit amount increases as income increases, up to a maximum threshold. For 1996, the threshold is estimated to be $6,340. Therefore, if there is only one qualifying child, the maximum basic credit for 1996 is S2.156 (34 percent of $6,340).

The credit is reduced and eventually fully phased out once AGI (or. if greater, earned income) exceeds a certain threshold. For 1996. the phase-out range is estimated to begin at $11,620. The phase-out is accomplished by reducing the credit by 15.98 percent of the excess of AGI (or, if greater, earned income) over this $1 1,620 threshold. The credit is completely phased out and is no longer available to taxpayers with incomes above the end of the phase-out range. In 1996. this income level is estimated to be $25,109.

If there are two or more qualifying children, the credit percentage, income thresholds, and phase-out percentage are higher. In 1996, the credit percentage for families with two or more children will increase to 40 percent of the first $8,900, with a maximum credit of $3,560 (40 percent of $8,900). The phase-out percentage will be 21.06 percent, and the phase-out range will be extended to $28,524.

Workers without children may claim the EITC if they are between 25 and 64 years of age and are not claimed as a dependent on another taxpayer's return. For these workers, the

231

basic credit is 7.65 percent of the first $4,230 of earned income. In 1996, the phase-out range for these workers is between $5,280 and $9,510 of AGI (or, if greater, earned income). The phase-out percentage is also 7.65 percent.

For these three credits, the income thresholds for both the phase-in and phase-out ranges are adjusted for changes in the cost of living.

Advance Payments of the EITC. There are two ways to receive the EITC. Individuals can claim the credit by completing a Schedule EIC when filing their tax return. Alternatively, individuals with qualifying children may elect to receive a portion of their EITC in advance by filing a Form W-5 with their employer. These individuals are entitled to receive on an advance basis up to 60 percent of the credit allowable for a family with one qualifying child.

Legislative and Administrative Actions in 1993 and 1994

As I mentioned at the outset, the Administration and Congress have taken a number of legislative and administrative actions during the past two years to improve the effectiveness and administration of the EITC.

OBRA 1993. OBRA 1993 expands the EITC and makes the program more effective in achieving its policy objectives.

First, OBRA 1993 increased the returns from working for those outside the workforce and for other very low-wage workers. For low-wage workers with two or more children, the EITC will fully offset the combined employee and employer portions of the OASDHI taxes and the food stamp benefit reduction formula.

The OBRA 1993 expansion was also a critical step forward toward achieving the goal that a full-time worker should not live in poverty if he or she works throughout the year. In combination, a minimum wage job. food stamp benefits, and the EITC can lift a single parent with one or two children out of poverty. The OBRA 1993 expansion significantly closes the poverty gap. However, since the minimum wage has not kept pace with inflation, the job is not completed yet. This is why the President has proposed that the minimum wage be increased over two years by 90 cents.

In OBRA 1993, we also balanced the goals of providing the neediest families with sufficient income support, while minimizing the marginal tax rates placed on families with higher, but still modest, levels of income.

While the definitive effect of OBRA 1993 can not be measured yet, we believe that the legislation will, on net, increase work effort. While some workers with larger families will face slightly higher marginal tax rates, they are unlikely to change their behavior much in response. These are individuals who are already very attached to the work force. They

232

cannot easily adjust their hours of work in response to a small change in tax rates; they need both their jobs and the EITC to meet their day-to-day needs, and most employers will not allow them the discretion to work fewer hours. The effect of the higher marginal tax rates on some workers in the phase-out range will be outweighed by the effect of the increase in the credit rate. By making work pay, the OBRA 1993 increase in the credit rate will encourage non-workers to enter the workforce and other low-income part-time workers to increase their hours of work.

Finally, OBRA 1993 simplified the eligibility criteria for the EITC beginning in 1994 by eliminating the two supplemental credits for health insurance coverage and for taxpayers with children under 1 year of age. Largely as a consequence of the repeal of these provisions, we have been able to greatly simplify the Schedule EIC.

The Uruguay Round Agreements Act (URAA). The Uruguay Round legislation contains four provisions to improve compliance as well as the targeting of the EITC to those with the greatest need. First, non-resident aliens will not be entitled to the credit. In addition, effective for 1994, prisoners will not be eligible for the EITC based on their earnings while incarcerated. The legislation also improves the administration of the EITC in two ways. By 1997, taxpayers will be required to provide TINs for all dependents and EITC qualifying children. The Department of Defense is also required to provide military personnel and the IRS with information regarding housing and subsistence allowances. The IRS will be better able to verify eligibility of EITC claimants because of these two requirements.

Administrative Actions. As Commissioner Richardson testified yesterday, the Administration has taken a number of steps to ensure that eligible individuals know about the EITC and the advance payment option. The Administration has also taken significant steps to ensure that those who are not eligible for the EITC do not receive it.

FY 1996 Budget Proposals

The Administration included several proposals to improve the targeting and administration of the EITC in this year's budget submission. We are ready to work with Congress on these proposals and hope that Congress promptly enacts them.

First, the EITC would be denied to taxpayers having more than $2,500 of taxable interest and dividends beginning in 1996. This cap would be indexed for inflation thereafter. This proposal would improve the targeting of the EITC to the families with the greatest need, as taxpayers with significant interest and dividend income can draw upon the resources that produce this income to meet family needs.

The Administration's proposal, with some modification, is included in the conference agreement for H.R. 831, a bill to extend and expand the 25 percent health insurance deduction for self-employed individuals. The conference agreement lowers the investment

-6-

income cap to $2,350 and expands the categories of income subject to that cap to include tax- exempt interest and net positive rents and royalties. The investment income cap would not be indexed. At the outset, approximately 550,000 taxpayers would lose benefits under this proposal.

We are very concerned that the investment income cap is not being indexed. We believe that this income cap should be indexed in the same manner as all other income parameters for the EITC, and just as personal exemptions, standard deductions, and tax-rate brackets are indexed. Without indexation, savings will be discouraged and the number of persons affected by this provision will increase over time. By 2000, the number of affected taxpayers would increase from about 550,000 to 650,000.

EITC Compliance Proposals. Our second budget proposal would improve the targeting of the EITC. Only individuals who are authorized to work in the United States would be eligible for the EITC beginning in 1996. Taxpayers claiming the EITC would be required to provide a valid social security number for themselves, their spouses, and their qualifying children. Social security numbers would have to be valid for employment purposes in the United States.

In addition, the IRS would be authorized to use simplified procedures to resolve questions about the validity of a social security number. Under this approach, taxpayers would have 60 days in which they could either provide a correct social security number or request that the IRS follow the current-law deficiency procedures. If a taxpayer failed to respond within this period, he or she would be required to refile with correct social security numbers in order to obtain the EITC. In combination, these provisions would strengthen the IRS's ability to detect and prevent erroneous refunds from being paid out.

Welfare Reform

In the Administration's 1994 welfare reform proposal (HR 4605), States would be given additional flexibility with respect to the EITC by allowing four demonstration projects to determine the effects of alternative methods of delivering advance payments of the EITC.

Other Suggestions

With respect to other suggestions to modify the design of the EITC program, the Administration will evaluate those proposals using the same criteria we apply to our own proposals:

(1) Does the proposal make work more attractive to those outside the workforce and to others with minimal ties to the workforce?

(2) Does the proposal reduce the poverty gap for families with full-time workers?

234

(3) Does the proposal improve the targeting of the EITC to the neediest, in the least distortionary manner? and

(4) Does the proposal make it easier for eligible taxpayers to claim the EITC and for the IRS to verify their eligibility before refunds are paid out?

We are concerned that many of the suggestions under discussion do not meet these criteria. For example, we would object to a restoration of pre- 1991 law, which based eligibility for the EITC on the dependent and head of household rules. These rules were very difficult for taxpayers to understand because they did not conform well to ordinary people's notions of caring for a child. The Bush Administration worked together with Congress to modify and simplify these rules in 1990. Restoring the old rules would increase complexity and non-compliance.

We would also have concerns about lowering the investment income cap below the $2,350 figure in H.R. 831. While we do not want taxpayers with substantial investment assets to receive the EITC, we also do not want to discourage hard-working individuals from being able to accumulate sufficient capital for important investments, such as a down payment on a home or business.

Targeting the EITC

We recognize that the targeting of the EITC to the neediest workers can also create work disincentives. As incomes increase above the beginning of the phase-out threshold, EITC benefits begin to decrease. As a consequence, the marginal tax rates for families of modest means increase. Among recipients in the phase-out range, the EITC could cause some individuals, primarily the spouses of other workers, to reduce the numbers of hours worked in response to higher marginal tax rates.

The EITC is similar to any benefit program which targets assistance to the very neediest families. We cannot target assistance to low-income families without causing marginal tax rates to increase for families with slightly higher income. However, we can seek to minimize such distortions.

In designing a program to make work pay, it is impossible to reduce the marginal tax rates in the phase-out range without either increasing the marginal tax rates on other families or reducing the credit amount for most families. There is a trade-off between the appropriate level of the EITC and the length of the phase-out range. A balance must be struck between the benefits of a higher credit rate for very low-wage workers and the costs of either higher marginal tax rates for families in the phase-out range or a longer phase-out range. There is, in short, no simple solution. Proponents of alternatives should demonstrate how their proposals result in a better balance of costs and benefits than current law.

235

Senate Republican Task Force on Entitlement Reform

Last month, the Senate Republican Task Force on Entitlement Reform released a draft report which included a recommendation to reduce expenditures on the EITC by $27 billion. Under the Task Force suggestion, the income thresholds for determining EITC benefits would no longer be indexed.

This represents a reversal of one of the key provisions in the Tax Reform Act of 1986. As a result of legislative action in the eighties, the tax code contains a number of provisions which are indexed for inflation each year. These include the personal exemption, standard deduction amount, and the width of the income tax brackets. It is unfair to suspend indexation on the one provision which is solely targeted to low-income taxpayers.

Under current law, an estimated 21.1 million taxpayers will claim the EITC in 1996. If benefit thresholds are not adjusted for inflation since 1994, participation would shrink to 18.2 million by 2000.

The Task Force justifies these reductions on compliance grounds. However, there isn't a single compliance issue that is addressed by eliminating indexation. Instead, the suggestion denies eligibility for the EITC to millions of law-abiding working taxpayers and reduces the benefits of millions of others who are playing by the rules.

The Administration is committed to improving compliance with the EITC rules. Its actions in the last two years are clear evidence of this commitment. Consequently, the Administration strongly opposes the Task Force's proposal which would reverse the long- standing, bipartisan support for a program that makes work pay.

This concludes my remarks. Thank you once again for providing me with the opportunity to testify. I would be pleased to answer any questions that the Committee may have.

236

United States Senate

Government Affairs Committee

Hearings on Effectiveness and Design of the

Earned Income Tax Credit

Washington, D.C.

April 5, 1995

Summary of statement by

George K. Yin

Professor of Law

University of Virginia School of Law

In my testimony, I describe the following options that Congress should consider to curb noncompliance in the EITC program:

1. Continue to evaluate whether the EITC benefit should be provided through the tax system rather than as a direct government expenditure. Most analysts believe that although the tax system promotes higher participation in a welfare-type program like the EITC, the potential for noncompliance is also greater.

2. Replace much of the program with an exemption from the payment of payroll taxes on the first $5,000 or $10,000 of wages. The exemption could be easily administered through an adjustment to the payroll tax withholding tables, and compliance could be expected to be very high.

3. Provide some of the EITC benefit to low- income workers through a tax credit awarded to their employers.

4. Reduce the size of the program and the amount of the benefit provided.

5. Calculate the EITC benefit based on income amounts that can be easily verified by the IRS, such as earnings reported on a W-2 form. Delay the EITC award until the IRS has a matching employer copy of the W-2 form.

6. Require a taxpayer to reside with a qualifying child for the entire taxable year in order for the taxpayer to be entitled to the EITC.

7. Bar the claiming of the EITC on electronically filed tax returns.

237

United States Senate

Government Affairs Committee

Hearings on Effectiveness and Design of the

Earned Income Tax Credit

Washington, D.C.

April 5, 1995

Testimony of:

George K. Yin

Professor of Law

University of Virginia School of Law

Mr. Chairman, Members of the Committee:

Thank you for inviting me to testify before your committee on the effectiveness and design of the earned income tax credit (EITC). I am a consultant to the Treasury Department's Task Force on Tax Refund Fraud, and have performed research on the EITC under the sponsorship of the American Tax Policy Institute, a nonpartisan organization interested in promoting sound tax policy, but I appear before you today in my individual capacity and not as representative of any group.

The focus of my testimony is to respond to the following question: how can Congress achieve the general goals of the EITC program in a more efficient manner? I leave to others to address the threshold question of whether the program creates desirable incentives and ought to be continued at all.

A. Scope of the problem

First, a brief word about the scope of the problem. The following table compares the growth in total federal expenditures for the major means-tested income-support programs between 1986 and 1996:

Growth in federal expenditures for means-tested income-support programs, 1986-96

Total Federal Expenditures ($ billions) and Growth Rates

	1986	1993	1986-93	1996	1986-96
Program	spending	spending	increase	spending (proj.)	increase (proj.)
EITC	2.0	13.2	560%	25.1	1155%
SSI	9.5	20.3	114%	27.0	184%
food stamps	12.5	24.8	98%	n/a	n/a
AFDC	9.2	13.8	50%	14.8	61%

Source: U.S. House Comm. on Ways & Means, Overview of Entitlement Programs: 1994 Green Book: Background matertial and Data on Programs within the Jurisdiction of the Committee on Ways and Means (Comm. Print 1994) at 262 (Table 6-25), 389 (Table 10-21), 704 (Table 16-13), 782 (Table 18-11).

The table indicates that since 1986, the EITC program has far outstripped the growth in all of the other major means-tested income-support programs including the Supplemental Security Income (SSI) program, Aid to Families with Dependent Children (AFDC) and food stamps. Indeed, the nominal growth in the EITC program over the ten-year period between 1986 and 1996 is projected to be over 1000%. According to these figures, by 1996, federal spending for the EITC program will be over one and one-half times as much as the federal share of the AFDC program. In other words, virtually

-2-

239

overnight, the Congress has created the proverbial 800-pound gorilla.

The consequences of this phenomenal growth are entirely predictable. There is not yet any reliable data on noncompliance rates following the major changes to the program enacted in 1993 and first effective in 1994 and subsequent years. But last fall, the GAO reported to the Chairman a preliminary IRS study for tax year 1993 - when the EITC program was still "only" about a 500-pound gorilla - which revealed that about 29% of the EITC claims were excessive and that just under one-half of those errors, or about 13% of the EITC claims sampled, involved intentional taxpayer error. In short, the program has grown very large very quickly, probably too quickly for any agency to administer effectively, particularly an agency such as the IRS whose mission has not traditionally included the delivery of welfare-type benefits like the EITC.

An obvious way to improve the efficiency of the program is to address the issue of noncompliance. If Congress can insure that only those entitled to benefits from the program in fact receive them, the goals of the program can be achieved at a significantly reduced cost. Hence, most of my comments will deal with methods of reducing noncompliance in the EITC program. But Congress needs to* act boldly if it wants to gain control over the problem. Nibbling around the edges, such as the proposals put forth by the Administration to date, may take a few pounds off of that gorilla, but will not address the core issue.

3-

240

B. Should the EITC program be administered through the tax system?

If improving compliance in the EITC program is a major concern, then Congress needs to continue to evaluate whether the EITC benefit should be provided through the tax system rather than as a direct government expenditure. Most analysts believe that although the tax system promotes higher participation in a welfare-type program like the EITC, the potential for noncompliance is also greater. In other words, the tax system transfers benefits pretty effectively, both to those entitled to the benefit and, unfortunately, to those not so entitled. Indeed, in some research I recently completed with several others, we tentatively arrived at exactly that conclusion. Compared to programs like AFDC and food stamps, the EITC program has a higher participation rate but also a higher noncompliance rate. Thus, if Congress is willing to sacrifice some participation in exchange for improved compliance, it ought to reconsider whether the tax system is the appropriate vehicle to deliver the EITC benefit.

I realize that the notion of shifting to a direct expenditure program is probably the last thing on Congress's mind these days. But Congress should not be deceived into thinking that a program implemented through the tax system is somehow without administrative cost. Recent events highlight all too clearly how the EITC program taxes the IRS's resources, and potentially jeopardizes its much more important mission of collecting revenue in a fair and efficient manner.

-4

241

Moreover, as you know, Congress right now is struggling with ways to control the federal budget as well as to reform programs providing benefits to low- income households. One effort to accomplish each of those objectives is to convert various entitlement programs to block grant programs to be administered by the states. Yet the EITC program, as currently designed, is an open-ended entitlement program administered by the federal government. Anyone meeting the conditions of the tax statute and, indeed, far too many who don't meet those conditions, is "entitled" to the EITC cash benefit. It may well be that Congress wants to confer "preferred" status on the EITC program and maintain its entitlement nature. But that judgment should be made affirmatively, and not by default due to the fact that the program is part of the tax system. Congress should evaluate the EITC program in the same manner as direct expenditure programs benefitting low- income households.

C. Redesigning the EITC program within the tax system

Assuming the EITC benefit will continue to be provided through the tax system, what other design changes might result in a significant improvement in compliance? Here are two ideas. First, Congress should replace much of the program with an exemption from the payment of payroll taxes on the first $5,000 or $10,000 of wages. The exemption might apply to the employee or the employer's portion of those taxes, or both. As this

5-

242

committee knows, the EITC originated in part as an effort to rebate to low- income workers the payroll taxes collected from them. Instead of collecting payroll taxes and then trying to return those amounts to workers in the form of the EITC, it would make much more sense simply to refrain from collecting the payroll taxes in the first place.

The beauty of this idea is that an exemption could be easily administered by employers through an adjustment to the payroll tax withholding tables. Taxpayers would not need to file returns to get the benefit. Further, they would get their benefit in each paycheck rather than as a lump sum at the end of the year. Finally, compliance could be expected to be very high because there would be no net cash benefit being transferred by the government back to taxpayers. Hence, the incentive to commit fraud to obtain the benefit would not be nearly as strong as under the current program.

To be sure, many in Congress might be fearful of tampering with the Social Security system. They might object to a proposal that decouples the link between Social Security taxes and benefits. The reality, however, is that for low-income workers, the EITC program has already decoupled the link between taxes and benefits. Such workers ostensibly pay Social Security taxes, and thereby become entitled to Social Security benefits, even though the EITC payment completely reimburses them for their Social Security contributions. They, in effect, pay no Social Security taxes yet are entitled to receive Social Security benefits. My proposal is simply to

-6

243

accomplish the exact same result but in a direct fashion, by not collecting the Social Security taxes in the first instance.

Another reform idea is to provide some of the EITC benefit to low-income workers through their employers. For example, a tax credit could be awarded to the employer of certain qualifying workers. The theory is that the same general transaction - the hiring and compensation of a qualifying worker -- can be subsidized by providing a direct benefit to either the employer or the worker in the transaction if the benefit is capitalized in the compensation arrangement.

Once again, the advantage of this idea would be to simplify administration of the program. It would be easier to administer because of the far fewer numbers of employers than workers. Further, employers are more used to dealing with the IRS than are low-income workers so that at least noncompliance due to unintentional errors should be reduced. Finally, the greater dollar amounts involved per-employer rather than per-worker would make IRS enforcement efforts more cost-effective. In a sense, this idea would convert part of the EITC program into a mini-block grant program to be administered by the business community rather than by the federal government.

True, the experience with a similar employer tax credit, the targeted jobs tax credit (TJTC), has not been very positive. But a number of features unique to the TJTC program ~ its start and stop history, the limited

244

duration of the subsidy to a portion of first year wages, the highly targeted nature of the subsidy directed towards individuals like ex-convicts who are undoubtedly the subject of negative stereotyping, to name a few - may help to explain that program's ineffectiveness. In contrast, a broadly applicable, employer-based subsidy program which is permanent has the potential for avoiding many of the TJTC's pitfalls.

D. , . Smaller changes to the design of the existing program

Finally, let me offer the committee four additional suggestions to improve compliance without a major redesign of the EITC program.

1 . Probably the single best step Congress can take to curb noncompliance without a major change is to reduce the size of the program and the amount of the benefit provided. Under current law, the size of the benefit available from the program no longer bears any relationship to taxes owed by the person making the claim. Accordingly, given our self-assessment tax system, it is just too easy to file a fraudulent claim that is virtually undetectable by the IRS.

For example, by 1996, an individual with two or more qualifying children who reports $8,000 in self-employment income would be entitled to an EITC benefit of 40% of that amount, or $3,200. If that amount were reported, the individual would owe a self-employment tax of about $1,200, but no income taxes. By reporting that income, therefore, the

-8

245

individual would receive a net cash benefit from the government of about $2,000 ($3,200 - $1,200) plus Social Security retirement credit. There is therefore a strong incentive in certain cases to falsify the existence of income. Could the IRS easily establish that the individual in fact earned less than the amount of income reported? No. The tax laws and administrative procedures are generally designed to ferret out income understatement cases, not the reverse situation of possible overstatements of income. Information filing, for example, permits the IRS to verify that taxpayers do not omit items of income on their tax returns. But if taxpayers voluntarily report more income than their paper trail might suggest, there is little the IRS can do to detect an error in the absence of a full-scale audit.

2. A second suggestion is to change the program so that the size of the EITC benefit is determined based on income amounts that can be easily verified by the IRS. For example, the benefit could be calculated, as under current law, as a percentage of earned income, but only earnings reported on a W-2 form would qualify. Further, the EITC award to a taxpayer might be delayed until the IRS has a matching employer copy of the W-2. This rule would preclude a taxpayer from getting any EITC based on self-employment income and various other forms of earned income, all items difficult for the IRS to verify, as well as earnings listed on a W-2 form which is fabricated. This change would also simplify the process of claiming the credit for all taxpayers.

9-

246

3. A third suggestion is to require as a condition of receiving the EITC that a taxpayer reside with a qualifying child for the entire taxable year rather than for just more than half of the year. Under current law, the qualifying child requirement is the source of much confusion and potential fraud. More than one taxpayer, or the wrong taxpayer, may try to claim a particular child, and some taxpayers even claim fictitious children. This change would do nothing to curb the claiming of fictitious children, but it would provide a more easily verifiable test, and would clear up some of the confusion arising in situations where a child lives in several different homes in the course of a year. Unless a child resided with the taxpayer for the entire taxable year, no credit would be available to the taxpayer.

4. Finally, Congress ought to consider barring the claiming of the EITC on electronically filed tax returns. Just as a burglar is going to burglarize the most vulnerable house on the street, those of a criminal mind are constantly casing the soft spots in our federal tax and transfer system. Surely one of the most vulnerable points is the receipt of EITC money through an electronically filed return. The speed with which a fraudulent claim may be processed and awarded makes that transaction a particularly appealing one to criminals.

To be sure, the IRS has taken important steps during this filing season to curb fraud effected through electronic filing, and Congress should probably await an evaluation of the IRS's latest efforts before barring

10

247

electronic EITC claims altogether. But there is no strong policy reason to expedite EITC claims, particularly if doing so makes the system more susceptible to fraud.

I recognize that the IRS, tax preparation services, and certain financial institutions may all object to a complete bar of electronically filed EITC claims. The short answer to those in the private sector is that they are not the intended beneficiaries of the EITC program or of electronic filing. If their incidental benefits are limited by this change, so be it. As to the IRS, I think the burden is on them to establish sufficient control over the problem to

justify the continued availability of electronic filing in this area.

* * *

Mr. Chairman, this concludes my prepared statement. I am happy to entertain any questions of the committee and would be pleased to work with your staff to develop appropriate legislative changes in this area.

-11-

248

STATEMENT BY

DONALD R. HUSTON

DOVER, DELAWARE

TO THE UNITED STATES SENATE

COMMITTEE ON GOVERNMENTAL AFFAIRS

WASHINGTON, D.C.

APRIL 5, 1995

THANK YOU FOR ALLOWING ME TO TESTIFY TO PROVIDE PRACTICAL INFORMATION REGARDING THE ISSUES AND PROBLEMS TAX PREPARERS FACE BECAUSE OF THE EITC.

I HAVE BEEN PREPARING PERSONAL INCOME TAX RETURNS FOR THIRTEEN YEARS. THE OPPORTUNITY FOR FRAUD AND MISINTERPRETATION INVOLVING EITC HAS ALWAYS EXISTED. WITH THE ADVENT OF ELECTRONIC FILING AND THE RAPID REFUND OR REFUND ANTICIPATION LOAN (RAL) A NEW OPPORTUNITY FOR FRAUD WAS AVAILABLE. SINGLE MEN WERE CLAIMING THEIR GIRL FRIENDS' CHILDREN AS WAS THE GIRL FRIEND. MARRIED COUPLES WERE SPLITTING TO FILE INDEPENDENTLY AS HEAD OF HOUSEHOLD, WITH EACH CLAIMING A CHILD. GRANDPARENTS WERE CLAIMING GRANDCHILDREN BECAUSE THE PARENTS WERE ON WELFARE. RESIDENT AND NONRESIDENT ALIENS WERE COMING ON THE SCENE AS PART OF ORGANIZED GROUPS. THE STATE DEPT OF HEALTH AND SOCIAL SERVICES WAS COACHING THEIR WELFARE RECIPIENTS ON HOW TO GET AS MUCH EITC AS POSSIBLE BY TELLING THE PREPARER THE COUPLE LIVED APART SO THAT BOTH HUSBAND AND WIFE COULD CLAIM EITC. NON-EXISTENT DEPENDENTS WERE APPEARING. BOYFRIENDS WOULD COME TO HAVE THEIR TAX RETURN PREPARED WITH A LIST OF SOCIAL SECURITY NUMBERS AND NAMES HE COULD NOT PRONOUNCE MUCH LESS SPELL.

AS A TAX PREPARER I DO NOT FUNCTION AS AN AUDITOR OR POLICEMAN. I HAVE AN OBLIGATION TO REPORT OBVIOUS CASES OF ATTEMPTED FRAUD AND MUST REFUSE SERVICE TO ANYONE WHO WOULD ATTEMPT OR INSIST ON CIRCUMVENTING THE LAW.

I AM ALSO NOT SUPPOSED TO DECIDE WHETHER AN INDIVIDUAL CAN SUPPORT HIM/HER S'ELF ON SUCH MEAGER INCOME AND CLAIM TWO DEPENDENT CHILDREN WITHOUT OUTSIDE ASSISTANCE.

DURING THE 1993 TAX SEASON FOR TAX YEAR 1992 I ESTIMATED THAT ABOUT 85 PERCENT OF THE RETURNS I PREPARED INVOLVING EITC INVOLVED SOME MANNER OF FRAUD. TAX YEAR 1992 APPEARED TO BRING EVERYONE OUT OF THE WOODWORK BECAUSE IT WAS FAST MONEY. THE WORD HAD SPREAD FROM THE PREVIOUS YEAR. THE IRS WAS ILL-PREPARED FOR THIS SITUATION. THE IRS WAS IN FACT HELPING TO PERPETUATE THE PROBLEM BY GIVING EITC EVEN THOUGH IT WAS NOT DESERVED. THE IRS WAS NOT CHECKING THE RETURNS TO SEE IF PEOPLE WERE USING THE SAME ADDRESSES, SAME CHILDREN OR EVEN IF THE SSN FOR OTHER THAN THE PRIMARY TAX PAYER WAS LEGITIMATE.

MOST OF THE RETURNS INVOLVING EITC REFUNDS WERE PROCESSED THROUGH

249

ELECTRONIC FILING AND THE RAL. THE FILERS DID NOT CARE ABOUT THE COST OF THIS TYPE OF FILING BECAUSE IT WASN'T THEIR MONEY. EVEN THE PEOPLE THE IRS CAUGHT BECAUSE THEY OWED BACK TAXES, BACK CHILD SUPPORT OR FEDERALLY SPONSORED LOANS RECEIVED CHECKS FOR THE REMAINDER OF THEIR REFUND DIRECTLY FROM THE IRS. THE IRS WAS HELPING TO PERPETUATE THE FRAUD. THE PREPARERS HAD TO SEEK OUT THE PERPETRATORS TO ASK THEM TO PAY THEIR PREPARATION AND PROCESSING FEES. SOME PAID AND SOME DID NOT.

WHEN PRESIDENT CLINTON STATED HE WAS INCREASING THE EITC, I KNEW SOMETHING HAD TO BE DONE. IT WAS AT THIS TIME THAT I CONTACTED SENATOR ROTH'S OFFICE IN DOVER DELAWARE TO EXPLAIN THE SITUATION AND MY ESTIMATE OF THE AMOUNT OF FRAUD INVOLVED IN RETURNS INVOLVING THE EITC. THE FIRST FEEDBACK I RECEIVED WAS A LETTER FROM THE SENATOR EXPLAINING HIS COURSE OF ACTION AND A COPY OF A LETTER HE HAD ADDRESSED TO THE TREASURY DEPARTMENT. I ALSO RECEIVED ADDITIONAL LETTERS AND COPIES OF RESPONSES FROM THE TREASURY DEPARTMENT AND FURTHER INVESTIGATIONS REVEALING A MORE REALISTIC ESTIMATE OF THE AMOUNT OF FRAUD. THIS FEEDBACK WAS VERY ENCOURAGING.

THE FRAUD HOWEVER CONTINUED IN TAX SEASON 1994. RAPID REFUNDS CONTINUED WITHOUT ABATEMENT. HOWEVER, THE TAX PREPARERS IN MY OFFICE WERE TRYING TO STEM THE FRAUD AS MUCH AS POSSIBLE WITHOUT PLACING THEMSELVES IN AN ADVERSARIAL ROLE. THE NEWS MEDIA WERE ALSO EXPOSING THE GROUPS WHO WERE EXPLOITING THE EITC AND THE IRS" INABILITY TO CURTAIL THE FRAUD. THIS TYPE OF ACTIVITY WAS ALSO PRESENT IN DELAWARE.

AT THE BEGINNING OF THIS TAX SEASON THEIR WAS A GLIMMER OF HOPE AS PROCEDURES AND RULES WERE TIGHTENING. MORE STRINGENT PROCEDURES AND CHECKLISTS EVOLVED TO HELP TAX PREPARERS WEED OUT THE FRAUDULENT FILERS. ONLY ONE FINANCIAL INSTITUTION WOULD BE INVOLVED IN RAL'S IN OUR AREA. IN THE PAST THE COST OF ELECTRONIC FILING WAS THIRTY FIVE DOLLARS AND WITH A BANK FEE OF ANOTHER THIRTY. IN 1995 THE BANK FEE WOULD VARY FORM THIRTY FIVE TO NINETY FIVE DEPENDING ON THE SIZE OF THE REFUND. THE BIGGEST DETERRENT WAS THE ACTIVITY OF THE IRS AND ITS NEW-FOUND ABILITY TO CROSS CHECK SOCIAL SECURITY NUMBERS WITH NAMES AND DATES OF ISSUE. THE WITHHOLDING OF EITC FUNDS FROM THE FINANCIAL INSTITUTION WAS PROBABLY THE REAL CLINCHER. THE WORD SPREAD FAST. AND RAPID REFUNDS APPEARED TO COME TO A SCREECHING HALT.

I FOR ONE AM THANKFUL FOR THE PROCEDURES THE IRS HAS INSTITUTED THIS YEAR. THEY WERE LONG OVERDUE. WE PREPARERS NEED BETTER GUIDELINES FOR DEALING WITH EITC FILERS AS A WHOLE. I DO NOT RELISH TELLING A FILER HE/SHE CANNOT CLAIM THEIR CHILDREN AS DEPENDENTS BECAUSE EITHER THE SSA OR THE DEPT OF HEALTH AND SOCIAL SERVICES IS PROVIDING MORE THAN THEY ARE. I BELIEVE THAT IF THE DEPT OF HEALTH AND SOCIAL SERVICE HAD TO REPORT TO THE IRS THE AMOUNTS PROVIDED TO EACH WELFARE RECIPIENT IT WOULD ALSO HELP STOP ANY FURTHER FRAUD AND PROVIDE AN INCENTIVE FOR PEOPLE TO GET OFF THE WELFARE ROLLS. I ALSO BELIEVE THAT MINOR CHILDREN RECIPIENTS OF SOCIAL SECURITY SHOULD BE INCLUDED IN THE EITC EQUATION AND OVERALL EQUATION OF DEPENDENCY .

250

THE IRS, THROUGH THE EFFORTS OF SENATOR ROTH AND HIS STAFF, HAVE MADE A SIGNIFICANT DIFFERENCE THIS YEAR IN THE ABATEMENT OF- FRAUDULENT TAX RETURNS INVOLVING EITC.

ELECTRONIC FILING IS ESTIMATED TO BE DOWN ABOUT EIGHTEEN PERCENT THIS YEAR AS WELL AS THE TAX PEPARATION BUSINESS. THERE WILL BE PRESSURE BROUGHT TO RESTORE EITC FUNDING FOR FINANCIAL INSTITUTIONS INVOLVED IN RAL'S. WE HAVE AN OBLIGATION TO TEACH FINANCIAL RESPONSIBILITY AND THAT CANNOT BE DONE IF FILERS ARE ALLOWED TO SQUANDER MONEY IN SO FRIVOLOUS A MANNER.

THANK YOU FOR ALLOWING ME TO TESTIFY BEFORE THIS COMMITTEE. THERE ARE NUMEROUS INCIDENTS AND ANECDOTES INVOLVING ALL MANNER OF FILERS FOR EITC. UNFORTUNATELY, TIME DOES NOT ALLOW ME TO RELATE ALL OF THESE TO YOU. I WILL BE HAPPY TO ENTERTAIN ANY QUESTIONS RELATING TO MY EXPERIENCES AND CONCERNS.

RESPECTFULLY SUBMITTED,

251

TESTIMONY OF

DEBORAH WALKER

CHAIR, TAX EXECUTIVE COMMITTEE

TAX DIVISION OF THE AMERICAN INSTITUTE OF CERTIFIED PUBLIC ACCOUNTANTS

BEFORE THE COMMITTEE ON GOVERNMENTAL AFFAIRS OF THE UNITED STATES SENATE

HEARINGS ON THE EARNED INCOME TAX CREDIT

APRIL S, 1995

AMERICAN INSTITUTE OF CERTIFIED PUBLIC ACCOUNTANTS

1455 Pennsylvania Avenue, N.W.

Washington, DC 20004-1081

(202) 737-6600

252

INTRODUCTION

Good morning, Mr. Chairman, and members of this distinguished Committee. We appreciate this opportunity to testify today on the earned income tax credit (EITC) for individual taxpayers. I am Deborah Walker, Chair of the Tax Executive Committee of the American Institute of Certified Public Accountants (AICPA). The AICPA is the national professional organization of CPAs, with more than 320,000 members. Many of our members are tax practitioners who, collectively, prepare income tax returns for millions of Americans.

The AICPA urges that simplification of the tax system be made a legislative priority. In particular, the EITC is an area in critical need of simplification.

We strongly urge the Committee and Congress to re-write the EITC rules to be understandable and usable by the taxpayers that this provision is intended to benefit - low-income wage earners. This group of taxpayers generally lacks the ability to deal with complex tax laws and is unable to pay for tax preparation assistance. The AICPA welcomes proposed changes to make the credit more effective and offers several suggestions.

BACKGROUND ON THE EITC

The refundable EITC was enacted in 1975 with the policy goals of providing relief to low-income families from the regressive effect of Social Security taxes, and improving work incentives among this group. According to the IRS, EITC rules affect almost IS million individual taxpayers.

Over the last few years, the number one individual tax return error discovered by the IRS during return processing has been the EITC, including the failure of eligible taxpayers to claim the EITC, and the use of the wrong income figures when computing the EITC The frequent changes made over the past twenty years contribute greatly to the credit's high error and noncompliance rates.

In fact, the credit has been changed 10times(1976, 1977, 1978,1979, 1984,1986, 1988,1990, 1993 and 1994). The credit now is a nightmare of eligibility tests, requiring a maze of worksheets. Computation of the credit currently requires the taxpayer to consider:

■ 9 eligibility requirements;

■ the number of qualifying children — taking into account relationship, residency, and age tests;

■ the taxpayer's earned income - taxable and non-taxable;

■ the taxpayer's AGI;

■ threshold amounts;

■ phase out rates; and

■ varying credit rates.

As part of RR 831, the House passed last week and the Senate is currently considering a new factor in determining eligibility — the amount of interest and dividends, tax-exempt interest, and net rental and royalty income received by a taxpayer, even if total income is low enough to otherwise warrant eligibility for the EITC

253

To claim the credit, the taxpayer may need to complete:

■ a checklist (containing 8 complicated questions),

■ a worksheet (which has 9 steps),

■ another worksheet (if there is self-employment income),

■ a schedule with 6 lines and 2 columns (if qualifying children are claimed); and

■ Usually, the normal Form 1 040 (rather than Form 1 040A or Form 1 040EZ).

For guidance, the taxpayer may refer to 7 pages of instructions (and 39 pages of IRS Publication 596). The credit is determined by multiplying the relevant credit rate by the taxpayer's earned income up to an earned income threshold. The credit is reduced by a phaseout rate multiplied by the amount of earned income (or AGI, if less) in excess of the phaseout threshold.

While Congress and the IRS may expect that the AICPA and its members can comprehend the many pages of instructions and worksheets, it is unreasonable to expect those individuals entitled to the credit (who will almost certainly NOT be expert in tax matters) to deal with this complexity. Even our members, who tend to calculate the credit for taxpayers as part of their volunteer work, find this area to be extremely challenging. In fact, we have found that the EITC process can be a lot more demanding than completing the Schedule A - Itemized Deductions, which many of our members complete on a regular basis for their clients.

Our analysis suggests that most of the EITC complexity arises from the definitional distinctions in this area. While each departure from definitions used elsewhere in the Code can be understood in a context of accomplishing a specific legislative purpose, the sum of all the definitional variances causes this Code section to be unmanageable by taxpayers and even the IRS. We recognize that many of the additions and restrictions to the credit over the years were for laudable purposes. However, the rules are so complex that the group of taxpayers to be benefited find them incomprehensible and are not effectively able to claim the credit to which they are entitled.

SUMMARY OF OUR LEGISLATIVE RECOMMENDATIONS

We recommend that Congress adopt the following changes to the EITC:

1. Simplify definitions and the calculation. (See Appendix for specific administrative proposals that we intend to pursue with the IRS).

2. Define "earned income" as taxable wages (Form 1040, line 7) and self employment income (Form 1040, line 12).

3. Modify the "qualifying child" rules.

A. Replace the "qualifying child" definition with the already existing dependent child definition.

B. Increase the incremental amount of credit provided for two children versus one child

254

C. Use the dependency exemption rather than the ETTC to provide benefits for children.

4. Combine and expand the denial provisions.

A. Deny the credit for taxpayers with: foreign earned income, alternative minimum tax liability, andAGI that exceeds earned income by $2,350 or more.

5. Modify the EIC Table or provide a percentage rate instead of the table. SPECIFIC LEGISLATIVE RECOMMENDATIONS

1. Simplify definitions and the calculation.

The current rules for the ETTC, as previously noted, provide different rules depending upon the number of qualifying children the taxpayer claims. The many rates, thresholds, limitations, and classifications regarding this credit are confusing. For the 1994 tax year, the parameters are as follows:

	1 Child	2 or more Children	No Children
Credit rate (%)	26.30	30.00	7.65
Phaseout rate (%)	15.98	17.68	7.65
Earned income threshold	$7,750	$8,425	$4,000
Phaseout threshold	$11,000	$11,000	$5,000
Phaseout limit	$23,755	$25,296	$9,000
Maximum credit	$2,038	$2,528	$306

RECOMMENDATION: Congress should simplify the definitions and the calculation of the credit. Specifically, as detailed below, we suggest the definitions of qualifying child and earned income be modified. The many rates, thresholds, limitations, and classifications regarding this credit should be referenced to other similar thresholds and classifications throughout the Code. These changes, and the changes listed below, would reduce the number of pages needed for the worksheets. Schedule EIC, EIC Table, and instructions. (See Appendix for specific administrative proposals that we intend to pursue with the IRS to simplify the definitions and calculation.)

Define "earned income" at tftTflH* ""?* (Form 1040. line 71 and setf-emplovmcnt income (Form 1040. line 12).

The current ETTC definition of "earned income" needs to be simplified. Currently, to calculate the credit, the taxpayer must take srao account all earned income, including amounts not otherwise reported on the tax return or not taxed. This is one area where numerous errors are made. As GAO states in its September 1993 report GAO/GGD-93-145, Tax Policy: Earned Income Tax Credit: Design and Administration Could Be Improved, "determining the amount of income that should be included in calculating the credit poses a problem for taxpayers and IRS."

Currently, potentially eligible recipients must take into account:

■ Taxable earned income (wages, salaries, and tips; union strike benefits; long- term disability benefits received prior to minimum retirement age; and net earnings from self-employment), PLUS

■ Nontaxable earned income (defined in the instructions as: contributions to a 401 (k) plan and military housing and subsistence, excludable dependent care benefits, pay earned in a combat zone, the value of meals or lodging provided by an employer for the convenience of the employer, housing allowance or rental value of a parsonage for clergy, voluntary salary reductions such as under a cafeteria plan, and "anything of value that is not taxable which you received from your employer for your work").

Furthermore, because taxable scholarships and fellowship grants are reported on Form 1040 line 7, taxpayers are instructed to subtract taxable scholarships or fellowship grants not reported on the Form W-2. This one exception complicates the calculation and is not verifiable, as it is not on the Form W-2.

In addition, as discussed in our administrative recommendations in the Appendix, the earned income calculation does NOT include various other forms of income not on line 7 of the Form 1040 (i.e., welfare benefits, workers' compensation benefits, alimony, child support, unemployment compensation, social security and railroad retirement benefits, pension and annuities, interest and dividends, and variable housing allowances for the military). The exclusion of these items is mentioned in IRS Publication 596, but is not mentioned in the worksheet or instructions. Since these items are taxable, but are not wage income (line 7 of the Form 1040), taxpayers may inadvertently include these items as "earned income".

Currently, the calculation of "earned income" involves a detailed knowledge of tax terminology, such as: "excludable", "taxable", "for the convenience of the employer", and "voluntary salary reductions". The definition of taxable income includes many items not commonly thought of as earnings. In addition, the definition of "nontaxable earnings" is unique to the EITC and is defined in different ways in the instructions and IRS Publication S96 (as addressed in our administrative recommendations in the Appendix). Most people think that "earned income" is wages. Omissions are likely to happen when uncommon terms are used to cover many items that normally are not treated as earnings.

RECOMMENDATION: Congress should define earned income as wages appearing on line 7 of Form 1040, plus self-employment income from line 12 of the Form 1040.

Earned income should only include taxable income, as the statute originally provided when it was created in 1975. As the GAO points out, much of "this (nontaxable) income is not reported to recipients or to IRS, " and IRS has no way right now of verifying the nontaxable amounts. GAO slates, "we do not see a need to provide space on the tax return for nontaxable eamedincome since less than 3 percent of eligible taxpayers claim (report) this type of income. " Therefore, nontaxable income should be removed from the EITC definition

256

of "earned income " to make the process simpler for the majority of taxpayers who need to complete this worksheet.

We also believe that there should be no exceptions to this taxable earned income definition. If Congress wants to treat taxable scholarships and fellowships different from taxable wages, taxable scholarships and wages should not be reported on in line 7 of the Form 1040. Alternatively, if taxable scholarships are to be treated as taxable wages (line 7 of Form 1040), the current EITC subtraction for scholarships and fellowship grants should not be allowed

Modify the "qualifying child" rules.

According to a GAO analysis, most EITC errors have been linked to issues involving filing status and qualifying children. The qualifying child test is complex. Taxpayers are confused by the "qualifying child" definition and the different definition for a dependent. The definition of "eligible child" complicates the ETTC instructions for determining eligibility The IRS attempts to communicate that a "qualifying child" usually does not have to be a dependent. However, there are a few exceptions that confuse taxpayers. For example, if one divorced parent has custody of the child, but the other parent claims the child as a dependent, the parent with custody can claim the child as a "qualifying child" for the EITC, but can not claim the child as a dependent. Also, if a child is married, the child must be a dependent (i.e., over half of the child's support is provided by the taxpayer) to claim the child as a "qualifying child" for the EITC. This married child exception confuses taxpayers.

Additionally, the different EITC treatment for different taxpayers - depending on the number of children - seems unnecessarily complex, especially for the minor additional benefit derived. The maximum additional credit for more than one child is only $490. The minor additional benefit is illustrated by an eligible taxpayer with $4,000 of earned income receiving a credit for one child of $1,059, while for two or more children, the taxpayer's credit is $1,208, a difference of only $149. What is tins differential meant to reflect? Clearly, the difference cannot be cost. In addition, there is no EITC difference between taxpayers with two children and taxpayers with three or more children.

Also, taxpayers with and without children are treated differently with regard to their eligibility for the advance EITC. The advance EITC is available only to taxpayers with qualifying children There does not appear to be any reason for this difference The EITC should focus on one goal - earned income.

RECOMMENDATION: The rules throughout the Code, and especially in this area, could be simplified if just one definition was used consistently. Congress should eliminate the distinction between "qualifying child" and "dependent child". Section 32(c)(l)(A)(I). which currently allows the EITC to certain taxpayers with non-dependent children, should be changed If the term "eligible children" is restricted to dependent children, section 32(c)(l)(A)(I) could be cross referenced to section 151.

257

This definition would provide an easy reference to information already on the Form 1040, line 6, and would eliminate the need for the additional information currently required on the Schedule EIC. We also suggest that the married dependent child test, which is rarely applied, be removed

Alternatively, if Congress deems that the "qualified child" is a better definition than "dependent child", then the "qualified child" test should be used for the dependency exemption as well. Either way, there should be just one definition of child in the Code.

RECOMMENDATION: The spread m the amount of credit for one child and two children should be made more significant than under the current EIC Table. The difference between one and two children in the current table is so small that it could not possibly reflect a cost differential and it is too incomprehensible for it to be a motivating factor in individual conduct.

RECOMMENDATION: Even greater simplification would result if there was no EITC differential based on the number of children. The current three classes of EITC recipients and three considerations at each point in the process are cumbersome. If just one class of EITC recipient existed the "qualifying child" versus dependent child confusion would be eliminated, making the credit process much easier. In addition, if this recommendation is adopted all EITC recipients would be able to claim the advance EITC.

As stated previously, an objective of the credit is to remove the regressivity of the Social Security tax for lower-income individuals. This objective applies to all lower income taxpayers, regardless of the number of children in the home. Thus, eliminating the incremental amount of the credit based on the number of children would not detract from the staled objectives of this provision. The calculation and the EIC Table would be simplified and the additional information on age and social security numbers of children (currently required on a separate Schedule EIC) would not be needed for the EITC.

RECOMMENDATION: Congress should coordinate all of . the Code s tax provisions related to children These child-based tax provisions include: the incremental child EITC, the child tax credit, the dependency exemption deduction, and the proposed family tax credit m H.R. 1215. All of these child tax benefits should be provided through one mechanism — the dependency exemption. The dependency exemption takes into account the total number of children in the household versus the EITC, which only accounts for up to two children in a household

However, since the current dependency exemption is a deduction rather than a credit, the result is regressive (that is, the higher the tax bracket, the greater the tax benefit) at the income levels that the EITC can be claimed Therefore, if the dependency exemption is to take part of the place of the EITC, one point Congress might consider would be replacing the dependency exemption with a refundable credit, not a deduction. The credit could be refundable and set at a fixed dollar amount per dependent child This credit could be available in advance from the taxpayer's employer, as is the advance EITC. The per child credit amount could be a round number that is easy to multiply.

258

The proposed child credit could be phased-out above some threshold AGI that is simple and consistent with other phase-out rules We suggest that the phase-outs for itemized deductions, personal exemptions, and this proposed child credit all start at the same threshold and that threshold should be a number that is easy to apply — e.g. , $100, 000 of AGI. The phase-out mechanism for all tax provisions in the Code should be the same.

Combine and expand the denial provisions.

IRS and GAO have stated that many people receiving the credit are not considered "low- income" individuals. As these individuals are identified, greater restrictions are placed on eligibility for the credit, and the computation is made more complex for all EITC recipients. As mentioned earlier, Congress recently agreed to deny the credit to individuals with interest and dividends, tax-exempt interest, and net rental and royalty income in excess of $2,350. Additionally, the credit currently is not available if the taxpayer: excludes from gross income any income earned in foreign countries, or claims a tax benefit for foreign housing amounts. An individual who owes alternative minimum tax (Form 1040, line 48) is allowed a credit, but the EITC is reduced by any alternative minimum tax. The taxpayer's AGI is used as a limitation for the credit and greatly complicates the computation for most taxpayers.

RECOMMENDATION: We support Congress limiting the credit to those taxpayers originally intended to benefit from the EITC. However, we suggest that this limitation only occur through the denial rules, not the computation rules. We recommend that all the denial provisions be included in one place. The eligibility/denial rules should include the current denial for taxpayers with foreign earned income. We suggest adding to the denial rule all taxpayers subject to AMT. This would delete the computational exception for AMT taxpayers.

Congress also should include in the denial provision taxpayers with AGI that exceeds earned income by $2,350 or more. H.R. 831 "s denial for taxpayers with $2,350 of interest or dividends should be replaced (and essentially combined) with this denial for taxpayers with AGI that exceeds earned income by $2, 350 or more. This exclusion based on AGI would deny the credit to taxpayers with all forms of unearned income (i.e., capital gains, income from partnerships and S corporations, etc.), not just taxpayers with interest and dividends of $2,350 or more. As the H.R. 83 1 proposed EITC restriction stands now, these "wealthy" individuals affected by the legislation (i.e., with this type and amount of 'non-earned income) could change their investments to earn partnership investment income, capital gains, or pay off their home mortgage and still claim an EITC. If H.R. 831 is enacted as currently drafted, when additional abuses occur. Congress will have to add more restrictions to the EITC statute. However, if our proposed broader AGI denial is used, fewer, or no, additional limitations will be needed. This denial of credit for individuals with higher incomes seems to be the intent of the unearned income restrictions.

An important result of moving the AGI calculation restriction to the eligibility rules is that the calculation would be much simpler. The calculation would no longer require a

259

Specifically, in order to implement this change, we suggest that section 32(h), which currently requires a reduction of the EITCfor taxpayers subject to the alternative minimum tax, be modified On the basis of simplicity, this provision should be combined with section 32(c)(1)(D), denying the EITC to anyone claiming a foreign earned income exclusion. Section 32(c)(1)(D) should also include taxpayers with AGI that exceeds earned income by $2,350. Section 32(a)(2) should also be modified to remove AGI from the computation. All the restrictions based on income should be contained in one paragraph or subsection, , rather than throughout this Code provision.

5. Modify the EIC Table or provide a percentage rate instead of the table

Although the IRS EIC Table takes into account all the phaseouts, the table can still be a mystery to many taxpayers. Many taxpayers are confused between the EIC Table and the Tax Table.

RECOMMENDATION: Section 32(f) currently requires that the EIC Table have income brackets not greater than $50 each. Form 1040 instructions currently include two pages of the EIC Table with $50 brackets, resulting in earned income credit intervals of$3-$8. Congress should amend section 32(f) to allow wider brackets which result in greater than $3 earned income credit intervals. This will reduce the EJC Table to half a page and will minimize the overwhelming nature of the table, and, hopefully, facilitate ease of use.

RECOMMENDATION: We suggest an even bolder alternative - eliminate the EIC Table completely, and instead provide a credit equal to a certain percent (i.e., 10 percent) of earned income. This option could be modified to provide for a few percentage levels (i. e. , 30% if earning less than $8, 000, 25% if earning between $8, 000 and $16, 000, and 20% if earning between $16,000 and $24,000). This would approximate the average credit currently allowed — $900 if earning $3,000 (with 2 qualifying children), $2,500 if earning $10,000 (with 2 qualifying children), and $2,000 if earning $20,000 (with 2 qualifying children), and would be much easier to calculate. This would be much simpler and would save space in the instruction booklets and ease much of the confusion The rates could be written directly onto the EITC line of the Form 1040. This, combined with a changes in the "earned income" definition and AGI limitation should make the worksheets, checklists, and tables a thing of the past.

OTHER REFORMS TO THE SYSTEM

Lastly, in reviewing comprehensive reform of benefits and tax reform, in general, Congress should consider the problems and complexities for low-earning Americans illustrated above. Some of the reforms listed below have been suggested as a potential solution to the EITC problems.

■ Use the EITC to eliminate the regressivity of the social security tax, by setting the refundable credit at the current social security tax rate (7.6S percent). The PICA tax regressivity results because the first dollar earned is taxed for FICA purposes, while income (for 1994) generally up to: $11,250 - married filing jointly, $8,050 - head of household, $6,250 - single, and $ 2,450 - married filing separately is exempt from the progressive income tax. This option

would permit all taxpayers with "earned income" to claim this credit regardless of their age, filing status or dependency status.

■ Limit the EITC benefit to no more than 15.3 percent (the current self-employment rate) of any self employment income reported. This would address the fraud and overreporting problems involved with self-employment income.

■ Exempt the first S£ of taxable earned income from the employee's share of social security tax and from V4 of the self-employment tax. (This tax could be administered by adjusting social security withholding and by amending Schedule SE). Exempting low-income individuals from FICA (social security and Medicare) taxes would directly address (with no paperwork) what the EITC was intended to do - mitigate the regressivity of the FICA taxes.

CQNCLVglQN

In conclusion, we have identified quite a few areas that need simplification and proposed various means to achieve it. Wesupport measures to efinmate the current EITC problems so that those who legally qualify for the EITC receive it and can claim the benefits in a simplified and easy process. The AICPA wants to again thank you for the opportunity to present our comments and recommendations for simplifying the EITC.

261

APPENDIX - Suggested Administrative QRS) Changes to the EITC Claiming Process

1 ETTC Line of Form 1040 Should be in the Credits Section of the Form Even if the

taxpayer is not allowed to claim the credit, the taxpayer must go through many procedures to find out if he/she is ineligible, and then write "NO" on line 56. Line 56 is not even in the section of the Form 1040 dealing with credits; it is in the section of the Form 1040 dealing with payments. The current placement of the ETTC line on the tax return could be confusing to taxpayers (although it is in bold). Therefore, we suggest the EITC line be moved to the credit section of the Form 1040.

2. The IRS Calculation Option Should be Presented at the Beginning of the Checklist and Instructions. IRS currently offers to calculate the EITC for taxpayers, but it mentions the option rather late in the process and requires the taxpayer to supply additional information on a separate schedule. The IRS calculation option is mentioned on the last line of the checklist, which reads "If you want the IRS to figure it for you, enter "EIC" on the appropriate line on the appropriate type of Form 1040. IRS should encourage more taxpayers to use this option and should consider mentioning this option at the top of the checklist and instructions.

3 An Information Required for the Credit Should be on the Form 1040 -Schedule EIC

and the Dependency Exemption Information Should be Combined on the Form 1040.

The Form 1040 should provide sufficient information for the IRS to determine if a taxpayer legitimately qualifies for the EITC. The IRS should inform legitimate eligible taxpayers of the correct credit amount. The taxpayer should not have to take the currently required additional steps of reading the instructions and completing the checklist, worksheet, and Schedule EIC.

We agree with GAO's Sept. 1993 report, Tax Policy: Earned Income Tax Credit: Design and Administration Could Be Improved, that stated,"most of the necessary information could be included on the tax return itself With minor modifications to the dependency and filing status sections of the Form 1040 or 104QA, all the requisite information (the already required child's name, social security number, relationship to taxpayer, and number of months lived with taxpayer, as well as the age and student/disability status of the child) would be available to determine whether a child qualified... We believe taxpayer simplification can be better achieved by the elimination of the separate EIC Schedule; the separate two-page schedule is an additional obstacle for very low-income tax filers." We, therefore, support elimination of the current separate Schedule EIC that merely covers repetitive information, and suggest the necessary information be combined into the existing Form 1040 exemption section, as shown on page 64 of the GAO report. This issue would disappear if a legislative change is made (as we proposed) so that the distinction in number of children is pursued through the dependency exemption.

We also suggest an even easier modification to the Form 1040. The only additional pieces of information (not currently required on the Form 1040) that are requested on the Schedule EIC are: (1) if the child was older than 18 - whether the child was a student under age 24 or permanently/totally disabled, and (2) the child's year of birth. The year of birth could easily take the place of column 2 on line 6c, where the IRS currently asks the taxpayer to check if under age 1. The information in (1) could also be included and coded on line 6c (i.e., next

APPENDK

to the age, put an S if a student and/or D if disabled). If the legislative change we proposed concerning the definition of "eligible child" and dependent is not made, the taxpayer also could put an "E" on line 6c to indicate that the child is an "eligible child" for the EITC The Form 1040 would then include all the information currentiy requested on the Schedule EIC.

AD Responses on the Checklist Should Consistently Direct the Taxpayer. The current locations and responses are confusing to taxpayers and should be switched. Checklist question number 5 should be worded in such a way that a YES response is positive and a NO response results in the taxpayer not qualifying for the credit (similar to all the other seven questions on the checklist). Accordingly, the YES and NO box locations to question number S should be switched too.

The Worksheet Should be Incorporated in the Schedule EIC. If the credit remains as complex as it is right now, instead of a worksheet calculation, the EITC should be calculated on an IRS designed schedule which is attached to the tax return. The Schedule EIC could be modified for this purpose to include the actual computations rather than mere taxpayer identification information. The IRS also could better monitor the credit amounts and if fraud or abuse is involved. It does not make sense for the taxpayer to first complete a checklist, then be directed to the worksheet, then complete the informational Schedule EIC, and then enter the credit from the worksheet onto the tax return. The IRS never sees the worksheet and, therefore, cannot see where the taxpayer made a mistake in the calculation and if it was intentional or not.

The Description of Items Subtracted from "Earned Income" Should be Stated Similarly in All HIS Publications. All IRS publications should clearly state the same definition and explanation of earned income. Specifically, IRS Publication 596 currently includes a detailed list of items to subtract from earned income, while the worksheet and instructions do not contain this list. The worksheet and instructions should include this list. Taxpayers may inadvertently include these hems as "earned income" . Specifically, according to Publication S96, the various forms of income that are not included in the earned income calculation (and are not subtracted on the worksheet) are not included in line 7 of the Form 1040 (i.e., welfare benefits, workers' compensation benefits, alimony, child support, unemployment compensation, social security and railroad retirement benefits, pension and annuities, interest and dividends, and variable housing allowances for the military). Our legislative recommendation to define "earned income" as taxable wages (line 7 of the Form 1040) and self-employment income (line 12 of Form 1040) would greatly simplify this problem.

The Taxpayer Should be Directed to the EIC Table Only Once. Rather than sending the taxpayer to the EIC Table twice (once for earned income and another time for AGI), the worksheet should direct the taxpayer to enter the smaller of the net earned income or the taxpayer's AGI, and then look up that smaller number in the table. The repetitive reference procedure is not necessary if the AGI is less than the beginning of the phase-out threshold. However, if that is the case, the taxpayer should be told to stop once the first credit amount is found in the table, before entering AGI and completing the rest of the worksheet nglessly.

A-2

ISMAEL GONZALEZ

ATTORNEY AT LAW

04 EAST 77TH STREET

NEW YORK, NEW YORK 10021

TEL: (212) FAX: (212)

April 4, 1995

Hon. William V. Roth, Chairman United States Senator Committee on Governmental Affairs United States Senate Washington, D.C. 20520

Re: Fraud in the Earned Income Credit Program

Dear Mr. Chairman and Members of the Committee:

I am writing to you as a private practitioner specializing in tax fraud defense in New York City and as a former attorney with the Internal Revenue Service. I also write to you as a concerned citizen living in the greatest country of the world!

Based on my past experience as a Special Assistant United States Attorney for the IRS prosecuting fraud cases and my current defense work, I am of the firm belief that Congress should abolish the Earned Income Credit program (EIC) in its entirety. The EIC is a vehicle for fraud, waste and abuse.

In the hispanic areas of New York City, particularly in our Puerto Rican and Dominican neighborhoods, there is a great misconception of the purpose and use of the EIC. Unfortunately, a small minority of these people have seen the EIC as an opportunity to rip-off our federal treasury. In 1992 and 1993, an estimated 13.6 millon dollars in false claims for EIC were filed by women on welfare with Spanish surnames in the New York City area. A great majority of them were from Upper Manhattan and the Bronx. Being that a majority of my clients are Hispanic, I am only aware of the effect of the EIC in this community. By no means am I stating that Hispanics are the sole cause for EIC fraud or they are to blame for fraud in the system. Most Hispanics are hard working and honest taxpayers.

I represented two women who were charged with preparing false EIC schedules for welfare mothers — Ana Ortiz and Miriam Peralta.

264

ISMAEL GONZALEZ

ATTORNEY AT LAW

64 EAST 77TH STREET NEW YORK. NEW YORK 10021

TEL: (212) 628-0505 FAX: (212) 828-0S40

Page 2

The enclosed copies of articles from the New York Times and New York Daily News paints my clients as somehow conning welfare mothers into believing they would get refunds of up to $1,4 00.00 if they filed EIC returns and pay my clients $25.00 to prepare these fictitious returns. Both of my clients told every welfare mother that there was no guarantee that they would receive a refund through the Earned Income Credit because they were collecting public assistance. Nevertheless, many of these mothers would ask my clients to prepare the EIC returns.

Many of these welfare mothers would come to my clients stating they heard about a $1 million dollar "trust fund for poor people" that they could apply for.

The EIC fraud was so widespread in 1993, one of my clients told me and the United States Attorney's Office for the Southern District of New York that there was one woman preparing so many EIC false returns, she was operating out of 10 apartments in Manhattan.

Of course, what facilitated this "EIC feeding frenzy" was the availability of electronic filing and rapid refunds.

Other clients have told me of situations where individuals without children buy legitimate Social Security numbers from non- filing individuals with children, so they can claim the EIC. Once they receive their refund checks, they split the proceeds of the tax refund with the person who gave them the Social Security number to use. I understand the EIC goes up to $3,000.00 in 1996.

In my opinion, the Internal Revenue Service is not doing enough to spread the word in the inner-city Hispanic communities about the criminal ramifications of filing false returns, and specifically how the tough and rigid United States Sentencing Guidelines can act as a bludgeon on their heads. By spreading the word, I mean by having Spanish-speaking employees of the Internal Revenue Service serve as speakers in churches, town-meetings and at community functions. The only time members of the Hispanic community ever see, hear or read about the Internal Revenue Service in their neighborhoods is when they are involved in a large cash seizure, narcotics or money launder ing-related enforcement in conjunction with the Unites States Drug Enforcement Administration. In my opinion, this is how they conceptualize the Internal Revenue Service. Service employees must spread the word in the Hispanic

265

ISMAEL GONZALEZ

ATTORNEY AT LAW

64 EAST 7TTH STREET NEW YORK, NEW YORK 10021

TEL (212)628-0505 FAX: (212)628-0540

Page 3

community that filing false tax returns, and in particular filing a false claim for the Earned Income Credit can land the offender in jail. If this cannot be done through community group meetings, use of the mass media, such as Spanish radio and Newspapers, then Congress should altogether abolish the Earned Income Credit. I have given talks about the EIC fraud problem on a Spanish radio station in New York City and have received numerous positive telephonic inguiries. However, I find that the Spanish press in New York is not very interested in this subject matter. I have submitted two large articles in Spanish to a well-known Spanish newspaper for free publication, but never received a response from the editors.

I believe that it would cost more tax dollars in stopping fraud in the EIC than it would in fulfilling the purpose of the legislation. Therefore, it should be eliminated.

I want to thank Senator Roth, members of his staff and the members of the Committee for giving me the opportunity to submit this statement and being able to testify before the Committee.

IG/cv

266

4 Charged In Tax Plot For Refunds

Fraudulent Returns Were Done for a Fee

By RONALD SULLIVAN

At first, there were just a few. Then a lot more, and then thousands and thousands — all of them low-income tax returns from Hispanic women and all of them seeking refunds under the Federal Government's tax-credit refund program for the working poor.

By the time they reached more than 11,000 a year ago, with claims for refunds of about $13 million, Inter- nal Revenue Service officials knew something was wrong, and yesterday they said that the returns were part of a novel conspiracy to file false income tax returns. Only about $80,000 was refunded before the I.R.S. detected the pattern of similar claims.

Working, apparently independent- ly, out of two bffices In predominantly Hispanic neighborhoods, four store- front tax return specialists filled out thousands of tax returns for women whose only income was their welfare benefits. But the tax return special- ists Improperly listed the benefits as Income on many of the returns, mak- ing the women eligible for refunds of $800 to $1,400.

The Government said the women paid fees ranging up to $35 for each return, which the Government con- tended was Illegal Income for the defendants.

In turn, the women kept all of the refunds, that Is, until the I.R.S. discov- ered the many refunds of the tame kind and stopped Issuing them, Fed- eral prosecutors said. A Federal offi- cial who declined to be Identified said It was highly unlikely that any of the women who filed for the refunds would be prosecuted, because prose- cutors believe the four defendants were at the center of the conspiracy and most of the women might not have been fully unaware that they were committing a Federal crime.

But the official said the defendants would most likely be held liable for the $80,000 that was fraudulently re- funded before the conspiracy was un- covered.

And unlike most criminal conspira- cies that Involve a good deal of scheming and detailed planning, Fed- eral officials said, this one was large-

VV till 1 A , \JUU ClUtlUS

seeking $13 million, the I.R.S. knew something was up. ^J

ly unplanned and depended in large pan on the "word (if mouth" pninci pal Kin of Hie women involved

One official who declined lo be Identified said ihere was no nvidencc that one pan ul defendants knowingly conspired with the other.

Mary Jo While, the United States Attorney for the Southern Dislncl of New York, yesterday announced the arrest of four people, Miriam Peralta and Roberto Nunez of Pearl Enter- prises of 651 West I71st Street in the Washington Heights section of Man- hattan, and Ana Ortiz and Freddy Mejla of the Wjy Travel Agency, 1530 University Avenue In the Morris Heights section of the Bronx The four were charged with conspiracy to de- fraud the United States and with help- ing file fraudulent tax returns Agents as Welfare Mothers

The conspiracy was broken, Ms White said when I.R.S. agents posing as mothers on public assistance went to the two tax preparation services in Manhattan and the Bronx and had fraudulent returns prepared by the defendants Under the Federal earned-lncome credit program, quali- fied low-Income workers with chil- dren are entitled to refunds ranging from $800 to $1,400 a year. But public assistance payments are not income and cannot be counted as taxable wages or Income In the program.

Ms. White said the returns were first noticed in September 1892, at the I.R.S. Service Center In Brookhaven, LI. At first, It was Just a few returns. But they continued to Increase and L&& employees trained to spot cer- tain patterns noticed that all of the returns were filed by single female taxpayers with Hispanic surnames from addresses In the New York City area. All of the returns, they noticed, sought refunds under the Federal' Earned Income credit program for the working poor.

By this lime, about $80,000 had been refunded before the I.R.S. stopped making any more payments and called in the Criminal Investiga- tion Division, which organized the sting operation that led to yesterday's

AH told, Ms. White said, 11,262 re- turns sought about $13 million in re- funds. Ms. White said an investigation was continuing lo determine exactly how many of the returns were pre- pared by the four defendants, and to determine If any more people might be involved

Company News:

Tuesday through Saturday

Business Day

267

it\ \o-.vs StaN \*"tc

Thousands of welfare mothers-have been duped into paying Tees of up -to S30 apiece- to'-corrupt tax preparers for bogus re- fund claims, authorities i harged yesterday.

Manhattan U.S. Attorney Man' Jo White said tax probers got wise to- the scum and refused to pay most of the S13 million "re- quested over the last two years under a tax^credit ^esigne^^oJfelp-'poo>?peb- »ple withcliitdren. rv"

But the mothers s^Lgot stuck with the fe^sj -:j|£^.

;«rei5ettUafth|T«oney;!fro^i the..ii*pa>'ers"iwhether $r not the\a*xpa*yers a're.gjaj- fing the refunds." Whfte said. "They.'re counting on the ignorance, of .the' tax- payers. %jgjsr Isi&Sg #

She*ak| Jhe >■&&& cen- tered nia^uyt n tRJJBiqica n communitie^of Manhattan and the Bronx. •.,-.

The alleged scheme came to light with the .ar- rest of four tax preparers on charges off pfepfcrtbg >fflT~reim\!sl6t Undercover Internal Revenue Service investigators posing is welfare mothers. '*

IRS official Peter F£r- rell said a 1992 tax change provided-an earned in- come credit for Mow-in- come single parents who wqrk But he said tax pre- parers were itling claims ^{^ly^UsiaiffWeltiare pay- ' mentsas wages5-- .

A totaFof il#62 alleged- ly fraudulent returns were filed -since September 1992. mainly on behalf of welfare mothers, accord- ing to Assistant U.S. Attor- ney Cathy Seibel. She said they sought $800 to $1,400 each in refunds, totaling $13.6 million, but on\y $77,000 was paid before the JRS.d-.etected the fraud. '\y. -\\ .'.-.• • AuthaMtifs .'said they would not "prosecute the recipients>-*eoaus«- ■ there was no indication they knew they were violating tho-larw,, ^,. - c;v.

the-four tfrrestod were Miriam Peralta and Ro- berto Nunez, who operat- ed at Miriam Pearl Kn tor- prises, 651 W. 171st St.. Manhattan; and Ana Ortiz and Freddy Mejia. em ployed at The Wa> Travel Ageney. 1530 University Ave.. Bronx.

If convicted." they each face up to five years in prison and a $250,000 line.

/*t^)LiAfa*r

2.-1

Testimony of

Marvin H. Kosters American Enterprise Institute

Before the Senate Committee on Governmental Affairs April 5, 1995

Improving the Design of the Earned Income Tax Credit

The earned income tax credit has become a major income transfer program. It redistributes income to low-income people who work, with almost all of its benefits going to workers with dependents. Its potential for encouraging work is most frequently cited in support of the program. Several problems with the program have been identified, however, and these problems have become increasingly serious as the program has expanded.

I will discuss three problem areas. First, although incentives created by the program encourage some who would otherwise not work to choose work instead, many more who are already working face incentives to work less. Second, the program imposes a substantial marriage penalty in many two-earner situations. Third, its high rates and high credit payments create incentives for uneconomic market activities and fraudulent claims. All of these problems are exacerbated by the growing size of the earned income tax credit (EITC) program.

The EITC Program

Since it began in 1976, the program has been periodically expanded, most recently in 1993. Provisions under the most recent expansion will be fully phased in next year. The federal budget projects total costs of the EITC in fiscal year 1996 as $26 billion in budget oudays and offsets to personal income tax payments.

The projected costs of the EITC in 1996 are roughly comparable to those for food stamps ($27 billion) and housing assistance ($27 billion). The cost of medicaid is much larger ($96 billion). Although the federal share of family support payments (AFDC) is considerably smaller ($18 billion), when state and local expenditures are included the total is somewhat larger (about

269

$33 billion). The EITC has become a major component of total federal expenditures to redistribute income to families with low incomes.

A brief description of how the EITC works is necessary to discuss its effects. I will describe its application to a family with two dependents when current law is fully phased in next year. At incomes of up to about $8,900, the credit provides a subsidy of an additional 40 cents for each dollar earned, up to a maximum credit of about $3,560. The credit is reduced by 21 cents for each additional dollar earned where the credit is phased out in the income range between about $11,600 and $28,500.' Families with one dependent receive a smaller maximum credit and are subject to correspondingly lower subsidy and phase-out rates, and low-income workers without dependents are eligible for a very small credit.

Effects on Work

To examine the effects of the EITC on work incentives, it is necessary to distinguish between people who are not working and those who are already working. Those who are working and receive incomes above the level at which they become ineligible for a credit are not directly affected by the existence of the program, although they pay some additional taxes to support payments under the program to others with lower incomes. Families with incomes above the eligibility cutoff account for about 60 percent of all families with children, as shown in Table 1.

Low-income families with no one working are unambiguously encouraged by the EITC to consider working. For them, the EITC amounts to a bonus for working as long as their income remains below the eligibility cutoff, but they receive credit payments only if they are induced to work and generate earnings. These families below the income cutoff with no one working account for almost 10 percent of all families with children. Among these families the EITC definitely has a positive effect on work incentives.

Families with someone already working but with income below the level at which they become ineligible for the EITC account for the remaining 30 percent of families with children.

270

These families are eligible to receive credit payments without any change in their work behavior. The program may induce them to modify their work choices, however, and the work incentives they face depend on their income levels.

For families with the lowest incomes, the amount of the credit increases as they work more and earn more, and they are accordingly encouraged to work more through this "substitution effect" However, the credit payment they receive would permit them to maintain or even increase their income without working as much, and this "income effect" discourages work. The total effect of these incentives is uncertain for working families in the lowest income range where the credit is being phased in. Whether they will work more or less depends on how work behavior responds to the subsidy rate -- the amount by which the credit payment increases as earnings increase -- and the amount of the credit payment. These working families with the lowest incomes account for about 7.4 percent of all families with children as shown in the table.

Families with somewhat higher incomes receive the maximum credit in an income range that includes an additional 3.8 percent of families. They receive no additional encouragement to work through the substitution effect, while the additional income they receive from the credit produces an incentive to work less.

For families with still higher incomes, where the credit is being phased out, the reduction in the amount of the credit payment as they earn more is equivalent to an additional tax on earnings. For them, work is discouraged both by the reduction in net earnings from additional work, and by the additional income from the credit payment they still receive; the (negative) substitution effect is reinforced by the income effect. Families in this income range account for about 19.5 percent of all families with children. These data indicate that the work incentive effects of the EITC are negative or neutral for about three times as many families as those with positive work incentives, and more than twice as many face work incentives that are definitely negative.

271

There is little controversy among analysts about the appropriateness of this analytical framework or about the direction of incentive effects that are at work. Studies that have examined the implications of the EITC for working families generally conclude that their work effort is likely to be reduced compared with work behavior in the absence of the program.2 Less work in response to negative work incentives among working families could be offset by greater incentives to work among families with no one previously working. One study that examines this question concludes that by encouraging some who would otherwise not work to work, the EITC may on balance encourage work.3

Can anything be done to improve the balance? In my view, the EITC discourages work for too many families, too far up the income distribution. Among all those who worked year- round and full-time in 1992, half earned less than $26,000 compared with EITC eligibility up to incomes of $28,500 in 1996. For all workers 18 years old and over, median earnings were only $18,400.

Reducing the maximum amount of the credit, reducing the income limits for eligibility for the credit, and reducing the subsidy and implicit tax rates would reduce the number of working families for whom work is discouraged, while the number of low-income families with no one working would be unchanged. Downsizing the EITC would reduce the strength of positive work incentives for families with no one working, but negative work incentives for the smaller number affected would be correspondingly reduced. In addition, of course, less income would be redistributed, and the transfers that remained would be shifted toward lower income families with children.

Another approach would require more careful integration of the EITC with the individual income tax. For example, the tax credit for children under consideration in the House of Representatives could be redesigned and integrated with the EITC to reduce marginal tax rates. A refundable child credit could be conditioned on earnings and used as a partial replacement for the current EITC for low-income families, while for families with higher incomes it could serve as a credit against tax liabilities. Work would be encouraged in low-income families because the

272

refundable credits would be contingent on earnings, while the high marginal tax rate on earnings under the current EITC could be reduced for families with higher earnings.

Work incentive effects for families with relatively low incomes could be improved by considering how such families are affected by the tax system as a whole instead of focusing on only one program at a time, such as the EITC. Examining the work incentive effects of tax and transfer programs taken together would be an even more ambitious approach. But broader consideration of how provisions in the tax code, taken together, affect work incentives would be an important step forward.

The Marriage Penalty

It has long been recognized that tax liability under the individual income tax is often higher for two individuals who are married than it would be if they were single instead. Marriage neutrality cannot be achieved under an individual income tax system with progressive marginal tax rates and equal taxation of married couples with the same incomes. Until recently, however, marriage penalties have not been a problem for families with relatively low incomes because they either paid no taxes or their tax liabilities were small and rates were very low. The EITC brought the marriage penalty to low-income families, and the major expansion of the credit in 1993 made it a significant problem.4

The marriage penalty is particularly severe for people who — if they were not married -- would each be eligible for the maximum credit. Marriage would eliminate one maximum credit payment and bring the couple well into the income range where the remaining credit is being phased out Under the most adverse circumstances the marriage penalty could amount to over $5,000, which for the family affected would be about 25 percent of the combined income of the married couple. Such a tax penalty creates an economic disincentive for marriage, and it reduces economic rewards from employment for married secondary workers compared with their unmarried counterparts.

273

One way to remedy this problem would be to introduce a two-earner credit, based on the same logic as the two-earner deduction in the individual income tax law from 1981 to 1986. This would reduce but usually not eliminate the penalty. Another way to ameliorate the problem would be to reduce the size of the EITC. A significant reduction in subsidy and phase-out rates and in the maximum credit amounts would greatly reduce the importance of the marriage penalty.

Incentives for Waste and Fraud

The high subsidy rate for earnings at low family incomes produces incentives to engage in both legal and illegal activities to take advantage of payments under the EITC program. Because the earnings subsidy rate very substantially exceeds payroll tax liabilities, the EITC makes attractive market work that would not otherwise be undertaken. It encourages activities that could be characterized as taking in one another's laundry when doing one's own would make more sense. Activities of this sort could be perfectly legal, but they would contribute little to economic well-being and mainly produce transfers for EITC recipients.

The rewards produced by credit payments and subsidy rates that are high in proportion to income have also apparently stimulated illegal activities. Audits by the IRS have shown a high incidence of credits claimed that are in excess of amounts for which families were eligible, and many credit payments for which families were ineligible. Many of these problems may represent legitimate errors, and efforts have been made to make the program more easily administrable. Nevertheless, the high incidence of problems suggests that fraud has been important.3

Assuring compliance with EITC provisions poses a particularly awkward problem for the IRS. The forms it has developed and the training of its auditors have been oriented primarily to identifying hidden or under-reported income, not to looking for ways in which income from earnings might be exaggerated. Earlier tax reforms removed many low-income families from the tax rolls, but the EITC has swept many low-income families back in. Auditing many returns, each with a relatively small dollar amount at stake, is quite expensive in relation to the revenue impact of errors and noncompliance that are identified. The IRS may also be reluctant to press

274

too hard to recoup overpayments, and it may find them very difficult to collect. Yet, unless reasonable compliance can be assured, the IRS is likely to be seriously embarrassed and the EITC program discredited.

In considering the auditing and compliance role of the IRS for the EITC, it should be recognized that a large bureaucracy to administer welfare and other social services is already in the field. This bureaucracy is often criticized for devoting too much attention to documentary paperwork and auditing, and too little to counseling and helping families in other ways to improve their circumstances. It makes little sense in this context to establish a separate, parallel bureaucracy in the IRS to administer another program to supplement incomes of low-income families, many of whom also receive benefits under other programs.

As in the case of work incentives and the marriage penalty, the compliance problem is greatly exacerbated by the size of the EITC. Smaller subsidy rates and credit payments would reduce incentives to claim credits to which the family is not entitled. If the size of the EITC were significandy reduced, whether the beneficiaries received partial payment of their estimated credit contemporaneously with their paycheck, or in a lump sum at the end of a tax year (as most now choose), would also be a matter of less consequence.

Summary

The three problems with the EITC that I have described have become more significant as the program was expanded. Although intended to encourage work by helping to "make work pay," the EITC discourages work for too many, too far up the income distribution. As it is currently structured, the EITC imposes a significant marriage penalty on low-income working people. And serious compliance problems have become evident in administering the EITC. All of these problems would be greatly ameliorated by making the EITC less generous.

Since payroll taxes absorb a significant portion of earnings for workers with low incomes, a strong case can be made for relieving those with the lowest incomes from the burden of these taxes while maintaining comprehensive program coverage. An EITC that is significantly more

275

generous, however, introduces more disincentives to work, builds up a significant marriage penalty, and invites fraud. In my view, the EITC should not be abolished. But I believe that we have expanded the program too much and that we are now placing too much reliance on the EITC under the guise of improving incentives to work. Although I believe that the EITC has become too generous, I do not argue that income transfers to low-income families with children should necessarily be reduced. I think the goals of providing income support to families with the lowest incomes and putting to work those who can and should work would be better served by scaling back the EITC and placing more reliance on other income support programs. These other programs should also be reformed to place more emphasis on making work a condition for eligibility.

Endnotes

1. Income levels for the EITC are indexed for inflation. The estimates reported are from Reducing Entitlement Spending. Congressional Budget Office, September 1994.

2. For example. Earned Income Credit: Design and Administration Could be Improved. U.S. General Accounting Office, September 1993; "The Earned Income Tax Credit: Participation, Compliance, and Antipoverty Effectiveness," John Karl Scholz, National Tax Journal. Vol. 47, No. 1, 1994; "Promoting Work through the EITC," Janet Holtzblatt, Janet McCubbin, and Robert Gillette, National Tax Journal. Vol. 47, No. 3, 1994; and "The EITC and the Working Poor," The American Enterprise, Marvin H. Kosters, May/June, 1993.

3. "The Earned Income Tax Credit and Transfer Programs: A Study of Labor Market and Program Participation," Stacy Dickert, Scott Henser, and John Karl Scholz, paper prepared for NBER Tax Policy and the Economy conference, November 1994.

4. See "Alleviating Marriage Penalties in the Income Tax and the Earned Income Tax Credit," Anne L. Alstott, Tax Notes, February 27, 1995.

5. Earned Income Credit: Targeting to the Working Poor, U.S. General Accounting Office, March 1995.

276

Effects on Work Incentives and Proportions of Families Affected by the Fully Phased in Earned Income Tax Credit

Families with children	Percent of Families with Children	Effects on Work Incentives
by Income Range and EITC Eligibility	Income Substitution Total Effect Effect Effect
All Families with Children 38,895 thousand	100%

Families with Incomes Too High for Eligibility for the EITC

23,163 thousand 59.1

• Families in Income Range Where Credit is Phased Out (Credit Payment Declines as Income Increases)

7,594 thousand 19.:

Families in Income Range Eligible for the Maximum Credit

1,489 thousand 3.!

• Families in Income Range Where Credit is Phased In (Credit Payment Increases as Income Increases)

2,871 thousand 1 .<■

Families with No Earnings Who Would Be Eligible for Credit if Working

3,774 thousand 9.:

Estimates are based on (he March. 1992 current population survey data reported in The Earned Income Tax Credit EITC. Current Law and the Clinton Proposal, Thomas Gabe, Congressional Research Service, May 25. 1993. Family incomes will be higher than levels reported in 1992 when current law is fully phased in next year because of inflation, but indexation of EITC income thresholds and of the maximum credit amounts to an inflation adjustment that offsets this potential distortion.

277

Testimony before the Senate Committee on Governmental Affairs

John Karl Scholz

Economics Department, La Follette Institute of Public Affairs ,

and the Institute for Research on Poverty

University of Wisconsin-Madison

1180 Observatory Drive

Madison, Wisconsin 53706

(608)262-5380

Wednesday, April 5, 1995

Mr Chairman and Members of the Committee:

Thank you for inviting me to testify on the effectiveness and design of the earned income tax credit (EITC). The EITC plays a central role in policy as it affects the working poor. Over its history, the EITC has received strong bipartisan support, as reflected by the fact that it was increased sharply in 1990, under President George Bush, and again in 1993, by President Bill Clinton. The real value (in 1993 dollars) of the maximum EITC varied between $739 and $1,068 between 1975 - the year the ETTC was adopted - and 1990. By 1996 the real value of the EITC will be roughly three times its 1990 level as a consequence of the 1990 and 1993 legislative changes.

The effectiveness of the ETTC depends on (i) whether or not people eligible for the credit actually receive it, (ii) the labor market effects of the policy, and (iii) ensuring that those not eligible for the credit do not receive benefits. The first part of my testimony focusses on these aspects of policy design. The final section discusses a variety of EITC-related policy options.

Labor markets for low-skilled workers in the U.S.

To understand the role of the EITC, and in particular, it's charmed existence over the past 20 years, a couple of facts about the labor market for low-skilled workers in the U.S. are useful. As Table 1 shows (next page), there has been a striking change in the bottom of the nation's earnings distribution, beginning in the 1970s, but accelerating over the 1980s. In 1973, the median male without a high school diploma earned $24,079 (in 1989 dollars), by 1989 the median worker with the same level of education earned $14,439. The trend is nearly as dramatic for males with only a high school degree. In 1973, the median male had an income of $30,252, by 1989 the median male with a high school degree had an income of $21,650. Average hourly earnings in private nonagriculture Industries peaked in 1973 at $8.55 per hours (in 1982 dollars) and, in 1994 was $7.40 per hour. The erosion of labor market opportunities for people with low levels of education has placed enormous strain on the nation's antipoverty programs. Against this backdrop, the EITC has provided an important supplement to the earnings of low-skill workers.

The participation rate of the credit

A family receives the EITC by filing a tax return, but many low-income families are not legally required to file returns. A married couple with two children, for example, was required to file a tax return in 1994 only if they had income above $16,150, though with an income considerably less than

278

that, the couple would be entitled to an EITC of over $3,000. If the EITC is to be successful at meeting the objective of "making work pay," families who are eligible for the credit should receive it.

Table 1: Median Income of PerMns 25 and Over, by Educational Attainment and Gender, Selected years, 1989 Dollars

		High	School		College
Year	1-3 Years			4 Years	4+ Years
1967	$22,858			$26,894	$39,186
1970	23,442			28,034	40,527
1973	24.079			30,252	41,065
1979	18,697			26,416	36,626
1983	15,133			21,932	35,188
1989	14,439			21,650	37,553
lource:	Robert Haveman and John Karl Scholz, "Transfers, Taxes, and Welfare Reform," National Tax

It is surprisingly difficult to estimate the percentage of EITC-eligible taxpayers who receive the credit - the EITC participation rate. Household surveys generally collect the information needed to determine eligibility but do not provide information on EITC recipiency. Tax data are best for estimating EITC recipiency, but not all households file tax returns so they are unsuited for estimating EITC eligibility. In an earlier paper I used unique data that allowed me to determine EITC eligibility and EITC recipiency in the same data set: specifically. 1 use data from the 1990 Survey of Income and Program Participation (SIPP) matched by social security number to selected items from individual income tax returns.' The paper describes, in great detail, the steps needed to calculate participation rates. Of the roughly 10 million taxpayers eligible for the credit in 1990, 1 estimate that between 75 to 90 percent received the credit. For reasons given in the paper, my preferred estimates place the participation rate at 80 to 86 percent.

The high participation (or take-up) rate of the credit is striking when compared to the AFDC participation rate of 62 to 72 percent and the food stamp participation rate of 54 to 66 percent2 The

'For a more detailed discussion of participation rates, see John Karl Scholr, "The Earned Income Tax Credit: Participation, Compliance, and Anti-poverty Effectiveness," National Tax Journal. 47(1) (March, 1994): 59-81.

'These figured are calculated from the 1986 and 1987 panels of the Survey of Income and Program Participation (SIPP) by Rebecca M. Blank and Patricia Ruggles, in their paper "When Do Women Use AFDC and Food Stamps? The Dynamics of Eligibility Versus Participation," mimeo, Northwestern University and the Urban Institute, Washington, D.C.. June, 1993.

279

United Kingdom has an EITC-like program called the Family Credit. It is administered through the transfer system and directed toward families with children. Official estimates place the participation rate of the Family Credit at around SO percent Thus, both compared to other in-kind transfers in the U.S. and comparable work-related benefits in the United Kingdom, the EITC gets high marks for reaching those who are eligible for the benefit.

The EITC's labor market incentives

The EITC has different labor supply incentives depending on the taxpayer's income relative to the subsidy, flat, or phase-out range of the credit The subsidy range of the credit increases the worker's marginal return to labor. For taxpayers with incomes in the subsidy range, the wage subsidy is thought to encourage work. At the same time, the income supplement provided by the EITC is thought to decrease recipients' labor supply because more money in hand means that they will choose to work less. The net effect is ambiguous. Households in the flat range of the credit receive the maximum EITC payment and no marginal subsidy for increased work, so these households have no marginal incentive to increase their hours of work, and the EITC supplement provides incentive to work less. In the phase-out range, the EITC is reduced as additional income is earned, which is akin to an additional tax on earnings. Thus the additional tax and the additional income both encourage workers to decrease their hours of work For households that are not working, it is hoped that the wage subsidy provided by the EITC will encourage participation in the labor market.

The aggregate labor market effects of the credit depend on the distribution of taxpayers within the credit's ranges and the degree to which they respond to incentives. In my National Tax Journal paper 1 estimate that 77 percent of EITC recipients will have incomes that fall in the flat or phase-out range of the credit in 1996, which raises the concern that the EITC may lead to a net reduction in the labor supplied by low-income workers. The fact that an incentive exists, however, does not necessarily mean that people act on the incentive in an economically significant manner. In the following sections I describe work performed with Stacy Dickert and Scon Houser on the likely effect of the EITC on labor market behavior.1

The effect of the EITC on hours of workers.

A large body of work in economics examines the effect of taxes and transfers on the hours worked by those in the labor market. Dickert, Houser and I survey this literature and use the empirical results from these studies to simulate the effects of the OBRA93 EITC expansions from 1993 to 1996.

Table 2, which is drawn from our paper, shows the effect of the 1993 EITC expansion, using 3 sets of labor supply estimates from studies using kinked budget sets because (i) they provide a consistent treatment of taxes and (ii) between the estimates of MaCurdy, Green, and Paarsch on the

'The following discussion of the tax treatment of low-income households and the effects of the EITC on labor market participation and hours of work is described in more detail in Stacy Dickert, Scott Houser, and John Karl Scholz, "The Earned Income Tax Credit and Transfer Programs: A Study of Labor Market and Program Participation," Tax Policy and the Economy. James M. Poterba (ed.), National Bureau of Economic Research and the MIT Press, 1995.

280

low end, and Hausman on the high end, the estimates appear to span the range of estimates found in the literature. As expected, we find that the EITC will reduce hours for every group except those in the subsidy range of the credit. Studies of labor supply find that secondary earners are more responsive to changes in taxes and wages than are primary earners. Accordingly, our estimated labor supply responses are larger for secondary-earners in two-parent families than they are for primary earners or single-parent families.

SimaUtod Labor Supply Responses to Changes in EITC Law from 1993 to 1996

Estimated Percent Change in Annual Hours Worked Kinked budget set simulations3 MaCurdy et al. Triest

All Recipients (160) -0.09 -0.54 -4.04

By credit range

Subsidy (80)

Flat (140)

Phaseout (160) By marital status

Husbands (180)

Wives (140)

Single female heads (160)

Single male heads (160) By sex

Male (180)

Female (160)

Note: See Dickert, House:, and Scholz (199S) (the complete citation is in footnote 3) for further detail. Median monthly hours are in parentheses.

Our reading of the literature suggests Triest's estimates most closely reflect consensus estimates from various branches of the literature. Triest's estimates imply that taxpayers in the subsidy range of the credit will increase hours by roughly 38 hours per year, while taxpayers in the flat and phaseout range will reduce hours by 3 and 21 hours per year, respectively. When aggregated, EITC recipients are estimated to reduce hours by an average of 11 per year. Hausman's estimates are roughly 8 times as large, while the MaCurdy et al. estimates show virtually no response. Our estimates are consistent with, but update earlier estimates from a 1993 study by the General Accounting Office.

While the credit, in aggregate, is likely to reduce the hours worked by workers, the economic significance of these effects is small. Moreover, to the extent that taxpayers are unaware of the effects

188	3.92	13.46
■0.09	-0.19	-1.79
•0.53	-1.11	-4.73
0.00	-0J4	-3.17
-1.47	-3.03	-11.36
-0.53	-1.1 1	-4.02
0.00	-0.18	-1-56
0.00	-0.34	-3.15
-0.57	-1.17	-4.33

281

of the credit on after-tax wages, both the positive effects of the credit in the subsidy range, and the negative effects of the credit on taxpayers in the phase-out range, is overstated. Thus, the best available empirical evidence suggests that the EITC has a small, but detrimental, effect on the hours of workers.

The effect of the EITC on labor market participation

What's missing from the literature and our calculations from Table 2 are estimates of the ETTC's effect on labor market participation. The credit has unambiguously positive labor market incentives on the decision of whether or not to work. Few papers provide guidance for thinking about the size of the effects of the EITC on labor market participation. This is somewhat surprising, as there is widespread agreement among economists that the strongest empirical labor market effects of wages and nonlabor income are on participation, rather than hours of work.

We identify the effects of the EITC on labor market participation by making use of the striking variation across states in the after-tax return to work.4 The typical low-income family in Texas, for example, faces relatively low cumulative tax rates - this calculation reflects both explicit tax rates from the federal income and payroll taxes, and implicit tax rates from benefit-reduction rules of food stamps and AFDC. Low tax rates are primarily a consequence of the low levels of AFDC benefits available to families in Texas. A household in similar circumstances in New York faces considerably higher rates, primarily because of the higher level of AFDC benefits. We find, for example, that average tax rates can exceed 85 percent for New York families receiving AFDC that enter the labor market and work anywhere from 8 to 35 hours per week. This implies that this family, when making labor market participation decisions, will receive no more than 15 cents for every dollar earned in the labor market over a broad range of hours. If tax rates affect labor market participation, we expect labor market participation to be higher in low-benefit states than in high-benefit states, holding other observable characteristics constant.

Our empirical work shows that net wages positively and significantly affect labor market participation and negatively affect transfer program participation, particularly for single parents. Moreover, this result is not an artifact of our particular empirical specification, but clearly emerges in the underlying data even when we do not control statistically for other factors. If we classify one- parent families by their predicted wage rate and then plot their probability of working, it appears, looking across wage groups, that the probability of working generally rises with wage rates. Even after crudely controlling for human capital by looking within wage categories, tax rates appear to exert a strong negative effect on the probability of labor force participation.

The empirical evidence on the effects of the OBRA93 EITC expansion on labor market behavior is summarized in Table 3. We model the effect the EITC increase on net-of-tax wages, and calculate the implied change in the probability that individuals work. At the mean, the ETTC increases the net wage of single-parent families by 15 percent. This leads to a 6.2 percent increase in their probability of working. If each new labor market entrant works an average of 20 hours per week for 20 weeks a

*We characterize this variation using a detailed, microsimulation model we have developed that uses monthly data for the 1990 calendar year drawn from the Survey of Income and Program Participation (SIPP). The model is coded in the computer language C, runs on a personal computer, and contains detailed modules for SSI, AFDC, food stamps, the federal income tax, state income taxes, and payroll taxes. See Dickert, Houser, and Scholz (1995) for further details.

year, the hours of single parent families would increase by roughly 72 million hours per year. Similar calculations show that the annual hours of primary earners in two-parent families would increase by roughly 12 million hours.

Table 3: Labor Market and Transfer Program Effects of the OBRA93 EITC Expansion, 1993 to 1996

Labor Market Effects
	New Labor Force Participants	Families in the Labor Market
	Percent Change in Net Wage	Annual Hours Change Due to Labor Force Participation (million)'	Annual Hours Reductions of Workers (million)2	Average Annual Reduction in Hours
Single Parents	15.0	72.8	26.4	10.1
Primary Wage Earners	19.6	12.1	136	7.7
Secondary Wage Earners	-5.0	-10.4	14.5	30.3
Total		74.4	54.5	11.2
Transfer Program Participation Effects
	Number Leaving Program	Mean Annual Benefit	Mean EITC Payment
Single-Parent Families	398,384	S6.844	S2.040
Two-Parent Families	117,757	S4.702	S2.842

'The estimation of the change in hours from new labor force participation assumes that, persons work 20 hours per week for 20 weeks per year.

:We use the estimated net wage and virtual income elasticities from Tries t (1990).

on average, these

The EITC will probably discourage labor force participation of secondary workers in rwo-parent families, as entering the labor market is likely to place the family's income in the flat or phase-out range of the credit. Specifically, on average we calculate that the after-tax return to work for secondary earners will fall by nearly 5 percent as a consequence of OBRA93. We expect that to

283

decrease labor market participation of secondary earners, leading to a reduction of roughly 10 million hours when using the same assumptions as above. Thus, we estimate that the aggregate increase in hours generated by increased labor force participation would be roughly 74 million hours.

The increased hours resulting from higher rates of labor force participation can be compared to the reduction in hours caused by the credit shown in Table 2. We use Triest's (1990) labor supply parameters for the simulation. We estimate that the annual hours of work for single-parent families in our sample wiil fall by 26.4 million hours, for primary earners they will fall by 13.6 million hours, and for secondary earners they will fall by 14.5 million, leading to an aggregate reduction of 54.5 million hours.

Taken together, the simulations suggest that the aggregate reduction in hours supplied by working households, 54.5 million, would be more than fully offset if new labor force participants work an average of 20 hours per week, 20 weeks per year. If either hours, or weeks worked are lower the offset will be smaller, while if new labor market entrants work more than 400 hours a year, the aggregate effect will be a larger positive number. The important point, however, is that one cannot forget about the participation margin when thinking about the labor market effects of the EITC.

Our evidence, like other papers before us, show that labor market participation appears to be sensitive to changes in wages and taxes. Hence, a policy like the OBRA93 EITC expansions that substantially increases the after-tax return to working for those not in the labor market are likely to elicit greater labor market participation. This increase in participation is likely to offset (or more than offset) the reduction in hours among those who work. The modest positive labor market effects of the EITC should be contrasted to the detrimental effects on both labor market participation and hours of work associated with other income transfer programs.

Compliance Issues

In past years a large number of ineligible taxpayers claimed the EITC, according to unpublished data from the IRS's Taxpayer Compliance Measurement Program (TCMP). In 1988 10.4 million taxpayers claimed the EITC, whereas the TCMP for that year estimates that only 7.1 million were entitled to the credit, indicating that over 30 percent of EITC claimants were ineligible. Of the S5.6 billion in EITC claims, the 1988 TCMP estimates that nearly $2 billion (33.6 percent) were claimed inappropriately. A General Accounting Office official testified in 1993 that "the credit has been the source of more taxpayer mistakes than any other individual income tax provision."

EITC rules were changed in 1990 and a 2-page schedule was added to the tax forms because of the widespread perception of compliance problems, but there is little publicly-available evidence that the 1990 changes significantly reduced noncompliance. The IRS did a special study during last year's filing season, examining returns filed electronically during the first two weeks of the filing season, and found that roughly 30 percent of EITC claims were inappropriate. It is not clear that this number is representative of the universe of filers, and getting information on this whole general topic is very difficult, but the limited available evidence indicates that noncompliance is still a major problem.

Noncompliance takes many forms, though one of the frustrations with thinking about this topic is in assessing the relative importance of different circumstances. Parents that are separated (or even living together) may file separate returns and both claim the same child as a dependent. Taxpayers may make up children. Taxpayers may misreport earnings or AGI. Some of these cases may not reflect the intent to commit fraud, others clearly do.

284

The future viability of the EITC depends on the IRS being able to reduce EITC noncompliance. Doing so requires two distinct tasks. First, programs must be put in place to detect accurately and report the degree of noncompliance and its causes. Current policy is being debated in an environment where solid evidence on the magnitude of compliance problems is scanty. The IRS study for last year reported that 29 percent of EITC claimants requested too large a credit, and 13 percent appear to be intentionally inflating their claims. The press release for this hearing cited a 40 percent fraud rate. Without credible numbers of EITC noncompliance and evidence about the source of noncompliance, it is difficult to design policy to address the problem.

The Internal Revenue Service has, perhaps belatedly, taken the second step needed to address noncompliance. This involves giving greater scrutiny to verifiable items on tax returns The IRS has created an uproar this year due to refund delays as they give social security numbers greater scrutiny. Nevertheless, this and matching employee and employer W-2 reports, seem to be the best ways of combatting the noncompliance that jeopardizes the program. As long as the credit is based on items the IRS is, in principle, able to verify, there is nothing inherent in the credit that would lead to unusually high levels of noncompliance. Moreover, other areas of the tax code, for example, Schedule C also have large amounts of noncompliance associated with them. As a matter of sound policy, the costs of reducing noncompliance in every aspect of the tax code should be compared to its benefits, so the IRS uses taxpayer resources in as efficient a way as possible.

Even if the IRS is able to perfectly verify social security numbers and wage and salary income, future compliance problems may arise with self-employment income. Between 1975 and 1990 the subsidy rate of the EITC was roughly the same as the combined employee and employer share of the payroll tax. In this period, as long as the payroll tax and EITC subsidy are about the same, taxpayers will not overstate self-employment income in order to increase their EITC Doing so would increase the taxpayer's EITC but would obligate the taxpayer to pay social security taxes on the self- employment income, eliminating any advantage to falsely reporting income.

With the sharp increase in the EITC, there are now stronger incentives to manipulate income. A taxpayer who does not work and has two children could receive a payment from the IRS of $3,370 in 1996 (in 1994 dollars) by reporting self-employment income of $8,425 ($8,425x.40). Doing so would require the taxpayer to pay $1,289 ($8,425x.l53) in social security taxes, leaving a net benefit to the transaction of over $2,000. The IRS is not well-equipped to uncover overreporting of incomes, and the payoffs to monitoring compliance in this area are certainly small relative to other areas of compliance. Of course, the taxpayer's claims need not be illegal. Two families could care for each other's children or watch each other's houses. They could exchange payments of $8,425 for doing so and both receive a net benefit of more than $2,000 if neither had any other sources of taxable income.

It is, of course, not yet clear how people will respond to these incentives to manipulate income, as there is no comparable situation in the tax code. My guess is that over time taxpayers or paid tax preparers will begin to take advantage of the incentive to overstate income in the subsidy range of the credit. The IRS will surely monitor closely the amount of income reported by low-income taxpayers that occurs in forms not subject to information-matching procedures (i.e., income from self- employment). An increase in the proportion of income occurring in these forms will be an early signal that a problem may be developing.

A solution to the potential problems associated with self-employment income would be to restrict the expanded EITC to income reported on W-2s (and only allow an EITC equivalent to the employer and employee share of payroll taxes for other sources of income), though that would create an inequity

285

between low-income wage earners and self-employed households.

Additional policy considerations before the Committee

In my National Tax Journal papeT, I estimate that roughly 50 percent of ETTC payments by 1996 will go to taxpayers with incomes below the poverty line and that roughly 36 percent of total EITC payment will directly reduce the "poverty gap" - the difference between total cash income and the poverty line. Because of the design of the credit, however, all benefits (other than those obtained fraudulently) go to taxpayers with incomes that are no more than 73 percent of the median family income in 1993. Put differently, the highest income an EITC-eligible taxpayer can have is $27,000 while the median family income in 1993 was $36,959. In this sense, the EITC is fairly well targeted to working poor or near-poor families. Nevertheless, the EITC could be more tightly targeted to lower income taxpayers by increasing the phase-out rate of the credit. Given the previous evidence on labor market incentives, doing so would probably not have particularly large deleterious labor market effects.

Place an hours requirement on the EITC

There are additional ways mat EITC benefits could be more tightly targeted. For example, Congress could establish a minimum annual number of hours of work to quality for the EITC This would require employers to report hours of work to the IRS. The purpose of the proposal would be to target die credit to low paid workers as opposed to highly paid part-time or part-year workers. My sense is that this proposal would be problematic. It would significantly increase compliance burdens of the income tax to require employers to report the hours of work of their employees. Moreover, there is little evidence that a significant number of EITC recipients are high-wage, part-time employees. Based on tabulations from the Survey of Income and Program Participation, I calculate that more than 75 percent of EITC benefits go to taxpayers with wages that would place them in the bottom 25 percent of all workers with children (below $7.30 per hour). More than 95 percent of all EITC benefits are paid to workers with wages below the median of SI 1.1 1 per hour. Given the extremely small gain in targeting and the very large increase in administrative burden from this idea, it should not be pursued.

Require the identification of both parents on tax forms claiming the EITC

A second proposal would require taxpayers to identify both parents of children claimed for the EITC on tax forms in order to use this information to ensure that both parents income is taken into account when determining the EITC amount. The proposal appears to be motivated by a concern that the EITC provides a disincentive for parents to be married.

In fact, the ETTC provides very strong incentives for some taxpayers to marry and others to separate. Consider, for example, a single man with two children and a single woman with two children. Both have incomes of $11,000. By 1996, each will be eligible for an EITC of $3,370. If they marry, their joint income will be $22,000 and they will be eligible for a credit of $1,054. By marrying, their combined EITC falls by almost $5,700, or more than 25 percent of their combined earned income. Similarly, a two-eamer couple with four children and with both the husband and wife making $1 1,000 would increase their combined after-tax incomes by more than $5,700 by separating

and maintaining separate households.3 Thus, it is clear that the EITC creates very large financial incentives for some taxpayers not to many and for others to separate.

At the same time, the credit increases the incentive for some households to marry. Consider, for example a single man earning $1 1,000 and a mother with two children with no earned income. If this pair marries, they will be eligible for an EITC of $3,370. In general, positive incentives to many are provided to low- or zero-earning taxpayers with children; and positive incentives for separation (or negative incentives for marriage) are provided to couples with children when each has modest earned income. My suspicion is that more taxpayers have an incentive to many than to separate because of the EITC. but I know of no empirical evidence on the topic, nor any evidence that suggests people manipulate their legal living arrangements to respond to these incentives.

As the IRS scrutinizes social security numbers, it is important to ensure that separated parents do not both claim the same child in order to receive the EITC. The IRS presumably already has the information needed to enforce this. Beyond that, it does not make sense to me to reduce the EITC a taxpayer could receive because a separated spouse's income or assets, as long as those resources are not available (in the form of a child support order) to the child.

Change the tax treatment of some payments for the purpose of the EITC

The third proposal asks whether the tax law should treat some monetary and non-monetary payments from other sources as nontaxable income to be added to income in determining the EITC. Revising the earned income definition would be intended to encourage work and ensure that total federal benefits provided to individuals are restrained and directed toward those who provide the support of children.

A current serious problem with the EITC is that several unverifiable sources of income are cunently included in the EITC income base. These include housing allowances or the rental value of a parsonage for the clergy, excludable employer-provided dependent care benefits, nontaxable military quarters and subsistence benefits, voluntary salary reduction amounts (e.g., deductions to 401(k) plans), and anything else of value (money, goods, or services) received from someone for services performed even if it is not taxable (IRS Publication 596). With perfect compliance the inclusion of these items in the EITC income base may enhance the targeting of the credit, but reporting these income items is essentially voluntary. Given the difficulty of enforcing these provisions, I would support a proposal to restrict EITC eligibility and benefits to verifiable items.

Receipt of the EITC does not affect one's eligibility for, or level of benefits available from AFDC, food stamps, or other federal means-tested programs for the poor. Changing this aspect of the EITC would be ill-advised, given the already extremely high tax rates many low-income households face. If the EITC resulted in additional benefit reductions, tax rates on many low-income households could easily exceed 100 percent, which would undoubtedly severely inhibit labor force participation.

^Separating would also reduce their (small) federal tax liability. If the family owns their home, it Is not even clear that they could not continue to live in the same house. One partner would receive the house in the separation agreement The homeowner could then "rent" a portion of the home to the separated spouse and children.

287

Incorporate an asset test on the ETTC

The fourth proposal would deny the EITC to taxpayers with assets (based on dividend, interest, and tax-exempt bond income) above some threshold. This proposal would be relatively straightforward to administer, and consequently could, in an effective manner, enhance the targeting of the EITC. I would not oppose such a proposal, though two things shouid be kept in mind when debating it. First, it will be relatively easy to manipulate portfolios so as to avoid the asset income thresholds without altering the value of the portfolio. Hence, the asset test will in all likelihood exclude fewer families than expected. Second, it may be counterproductive to implement a very restrictive asset test. While asset tests of any kind enhance targeting in a static sense, they also can make it nearly impossible for recipients to legally accumulate the assets necessary to take a wide range of choices leading to independence - helping a child attend college, acquiring additional training, or moving away from a dangerous neighborhood.* While asset tests might help target benefits to the poorest members of society, they also make it extremely difficult for program recipients to acquire skills that may help them or their children avoid poverty in the future. As with high implicit tax rates on labor earnings, asset tests distort the economic decisions of low-income households and hence reduce the efficiency of antipoverty policy.

Conclusions

Over the past 20 years the EITC has been a favored policy tool for assisting low-income families with children. Between 1975 and 1990 the EITC remained roughly constant in real terms. By 1996 the real value of the ETTC will nearly triple from its 1990 level. No other major program directed toward low-income families has grown at a comparable rate in recent years.

The effectiveness of the EITC will depend, in part, on its effect on labor market behavior. Most workers that will receive the credit have incomes that place them in the flat or phaseout range of the credit, where the credit provides an unambiguous incentive for people to work fewer hours. Using recent estimates from the empirical literature on taxes and labor supply, we find that the change in incentives caused by the 1993 expansion of the credit are expected to lead to a reduction in hours of work by those already in the labor market. Our central estimate predicts an overall reduction of roughly 54 million hours a year for families in our sample.

One of the attractions of the EITC is that it provides an unambiguously positive incentive to people not working to get a job. We find, both in descriptive tabulations and empirical models, that the after-tax wage has an economically and statistically positive effect on labor market participation and negative effect on transfer program participation. Thus, when fully phased in, the 1993 EITC expansion will increase labor force participation. Our preferred estimates suggest the magnitude of the

'For example, a family receiving food stamps cannot have financial assets exceeding $2,000 (53,000 if the family unit has a member older than 59), and cannot own assets and a car whose combined market value exceeds 56,500. A family receiving AFDC cannot have more than 51,000 of financial assets. In a well publicized case, Cecelia Mercado was ordered by the Connecticut Supreme Court to repay 59,342.75 in AFDC payments because her daughter, without her knowledge, had taken a part-time job and, in a year and a half, had saved almost 55,000 toward attending college. To become recertified to receive AFDC, the daughter and her brother (who had saved nearly $1,000) had to spend their saving until the household's assets fell below the 51,000 asset limit (William Raspberry. Atlanta Journal and 8/21/92. p. A 10).

11

288

participation effect will be large enough to offset the decrease in hours worked by workers.

While economists are predisposed to concentrate on the labor market responses to the EITC, the credit gets mixed reviews on two additional, important, policy dimensions. The participation rate - the fraction of eligible taxpayers that actually receive the credit - is quite high. The credit is reaching its intended beneficiaries. At the same time, many that are not eligible for the credit receive it. With the recent expansion of the EITC, the amount of overpayment could reach $10 billion annually, an amount that, if it persists, will surely cause the credit to be scaled back, if not repealed altogether. I am hopeful that recent IRS attention to this problem will reduce the error rate and the credit will remain in place. My support for the credit results from the fact that (i) a high fraction of eligible taxpayers receive the credit, (ii) its labor market effects are probably, in aggregate, beneficial, and without a doubt, they are less pernicious than alternative ways of assisting the working poor, (iii) the credit is well-targeted toward poor and near-poor families, and (iv) families receiving the EITC are working, that is, they are "doing the right thing." Given the performance of labor markets for low-skill workers over the last 20 years in the U.S., it sound social policy to try to supplement the earnings of working poor families.

TESTIMONY OF ROBERT GREENSTEIN, EXECUTIVE DIRECTOR CENTER ON BUDGET AND POLICY PRIORITIES

Senate Committee on Government Affairs April 5, 1995

I appreciate the opportunity to testify before the Committee today. I am executive director of the Center on Budget and Policy Priorities, a non-profit public policy organization that examines federal and state fiscal policies with an emphasis on policies affecting low- and moderate-income families. The Center is funded by foundations. Last year, I also served as a member of the Bipartisan Commission on Entitlement and Tax Reform.

The Center has been engaged in analyzing issues related to the earned income credit for a number of years. In a related aspect of our work, we also work closely with the IRS, the National Governors Association, the United Way, and a number of businesses, charities, and other non-profit organizations in distributing information about EIC eligibility rules and filing procedures to state and local agencies and organizations that work with low-income working families, so that eligible families may be apprised of their eligibility for the credit and so that ineligible families will not seek to receive it. Through this work, we also occasionally learn of abuses related to the credit; when this occurs, we seek to bring these abuses to light so action may be taken to correct them.

In 1992, for example, we learned of abuses involving the EIC health insurance credit. We notified the IRS and the congressional committees of jurisdiction of our findings, and the oversight subcommittee of the House Ways and Means Committee mounted an investigation that confirmed these findings. The health insurance credit was repealed later in 1993, a course that we, among others, had recommended.

There is now a growing focus on the EIC. In my testimony today, I would emphasize several points.

The EIC's biggest problem involves error rates. While significant progress in reducing error was made in the 1990 reconciliation bill, more needs to be done. The EIC provision in last year's GATT bill should help. More important, in the last several months, the IRS has taken some badly needed and overdue steps that hold strong promise for making significant reductions in error rates. Legislation the Treasury submitted with the budget also should help reduce errors.

While seeking to reduce EIC error rate problems, we should also be mindful that the EIC has considerable strengths. It addresses one of the key trends that has caused a substantial increase in child poverty in recent

290

years, the erosion of wages for low-paid work. It also helps substantially in making work more remunerative than welfare. It is an important building block for welfare reform that places some limitations on cash assistance and seeks to move families to work.

Certain reforms in the EIC eligibility structure, including both the provisions of S.831 (with some modification) and proposals to bar EIC receipt for tax filers in which either the parent or the qualifying child is residing here illegally, should be made.

• Some EIC proposals need to be regarded with considerable caution. A

proposal to end the indexing of the EIC, for example, would cause large increases over time in the tax burdens of millions of low-income working families whose wages are simply keeping pace with inflation and would conflict with the goals of work-based welfare reform. Also, some proposals to lessen marriage penalties in the EIC could have perverse effects, lessening the attraction of work over welfare and transferring large sums from working poor families to middle-class families. Finally, some proposals would make the EIC significantly more complicated and be likely to increase errors as a result.

In short, we need to be mindful of the strengths of the EIC as well as of the areas where it has weaknesses that need strengthening. We need to improve the integrity of the EIC, and we need to do so in ways that do not lessen the credit's positive effects.

The Basis for the EIC

The need for and growth of the EIC in recent years is closely related to the erosion of wages for low-paid work. Between 1977 and 1993, the poverty rate for families with children in which a family member (usually a parent) works grew by nearly half. More than 60 percent of all poor families with children contain a worker.

A study by Northwestern University economist Rebecca Blank, a former staff member of the Council of Economic Advisers during the Bush Administration, found that wage erosion exceeded all other factors in importance in explaining why poverty rates did not decline more during the economic recovery of the 1980s. In addition, Census data show that the proportion of full-time year-round workers paid a wage too low to lift a family of four to the poverty line rose by one-third between 1979 and 1993 (from 12 percent of full-time year-round workers in 1979 to 16 percent in 1993).

Eroding wages for low-skilled work reduce living standards and increase child poverty. They also lessen incentives to work and decrease the gains from working relative to receiving welfare. Starting in the mid-1980s, these trends led to a bipartisan

291

policy emphasis on increasing the remuneration from low-wage work, with the EIC as the principal policy instrument.

President Reagan proposed a major EIC expansion in 1985, which was included in the 1986 Tax Reform Act. President Bush and Congress passed a further expansion in 1990. President Clinton proposed a final expansion that Congress passed in 1993.

These expansions signaled an increased reliance on the EIC and a decreased reliance on the minimum wage as an instrument of wage policy. The purchasing power of the minimum wage is now at its second lowest level since 1955. By next year, it will be at its lowest level since 1955.1

The expansions also reflected sharply deceased reliance on AFDC as a means of supplementing the wages of poor single parents with children and increased reliance on the EIC. In 1972, before the EIC was created, 49 states provided AFDC as a wage supplement to a mother with two children whose earnings equaled 75 percent of the poverty line. Today, just three states do.

The expansions also reflected, in part, increases in payroll taxes, which the EIC is designed to offset.

Finally, the EIC expansions were the principal means to attain the bipartisan goal that if a parent worked full-time year-round, the parent and his or her children should not live in poverty. The goal has been that wages (net of payroll taxes) from full-time year-round minimum wage work should, when combined with the EIC and food stamps, lift a family of four to the poverty line. When the 1993 EIC expansions are phased in fully, we will be close to attaining this goal. (The goal would be reached with a modest increase in the minimum wage.)

I would make one final observation about these expansions. A few policymakers have recently commented that the EIC is one of the fastest growing federal benefit programs and have suggested it is "out of control."

The growth rate, however, is not due to uncontrollable or unanticipated factors but rather to the explicit policy decisions made by Congress and three Presidents. As noted, the EIC was expanded on a bipartisan basis at the request of President Reagan in 1986, with strong support from President Bush in 1990, and at the request of President Clinton in 1993. Each expansion was phased in over several years. The high growth rates are a direct result of these expansions.

1 Over time, too much pressure will be placed on the EIC if it has to carry all of this load and the value of the minimum wage continues to erode. A modest strengthening of the minimum wage would be desirable.

292

Once the phase-in of the 1993 expansion is complete, the high rate ofEIC growth will end. CBO data show that after 1997, when the phase-in of the 1993 expansion is completed, the EIC growth rate will be less than 4.5 percent per year. Most of this growth will be due to inflation.

As a percentage of the Gross Domestic Product, EIC costs will decline after 1997. After that year, the EIC does not contribute to the projected growth in the federal deficit as a percentage of GDP.

This is a far cry from the situation in programs such as Medicare and Medicaid. In those programs, rapid rates of growth persist indefinitely and do not primarily stem from specific federal decisions to expand the programs.

It also should be noted that in the context of minimum wage erosion and the near elimination of AFDC as a wage supplement to poor single parents working half- time or more, the EIC expansions do not look so large. When the 1993 EIC expansions are phased in fully, single parents working at the minimum wage will still have less disposable income, after adjusting for inflation, than they did in the early 1970s before the EIC was created.

An analysis conducted last year by the Department of Health and Human Services found that with the fully phased-in EIC, the income for a mother with two children who works half-time year-round at the minimum wage will be nearly $3,000 lower — or 28 percent lower — than it was in 1972, after adjustment for inflation. (Disposable income, as used here, includes wages, AFDC, food stamps, and the EIC, minus federal income and payroll taxes.2) The loss of AFDC in many states, the erosion in the minimum wage, and the increase in payroll taxes more than outweigh the EIC increases of the past decade.

The HHS analysis found similar results for single parents with two children who work 30 hours a week throughout the year as well as for those who work 40 hours a week throughout the year. In both cases, the family's disposable income is at least $2,000 — or at least 24 percent — below 1972 levels in purchasing power.

In a nutshell, the workings of the private economy in pushing down wages for low-skilled work, combined with policy decisions in the minimum wage and AFDC areas, confronted policymakers with a need to increase the EIC substantially to avoid

2 These figures reflect average weighted AFDC benefits across the 50 states. Figures for 1972 are adjusted for inflation using the CPI-U-X1 rather than the more rapidly rising CPI. All figures are in 1993 dollars.

Average Disposable Income For a Mother and Two Children

From Wages, AFDC, Food Stamps, EIC, and Federal Taxes

(in 1993 dollars)

Number of Hours Worked Per Week At Minimum Wage Throughout the Year
Yjear 20 Hours	30 Hours	40 Hours
1972 $13,482	$14,602	$15,656
1980 11,479	12,870	13,792
1990 9,830	10,467	11,509
1993 (with EIC at fully phased-in 10,612 1996 levels)	11,956	13,653
Percentage Change in Average Disposable for a Mother and Two Children	Increase
1972-1993 -28%	-27.4%	-24.0%

Source: Department of Health and Human Services

large decreases in the living standards of working poor and near-poor families with children.

Beneficial Aspects of the EIC

The 1986, 1990, and 1993 EIC expansions reflected decisions to help cushion the erosion of wages for low-paid work, while placing less emphasis on the minimum wage and AFDC as mechanisms to accomplish this goal, as well as a desire on the part of policymakers to make work more remunerative than welfare and bring families with a full-time working parent closer to the poverty line. These expansions also reflected a recognition that with its faults, the EIC remains one of the soundest of the low-income transfer programs.

In designing income transfer programs, policymakers face a conundrum. If the benefit is means-tested and phases out as income rises, the program will inevitably create some work disincentive effects and probably also some marriage penalties. This can be avoided if the benefit is not means-tested and is made available at all income levels, without reductions in the benefit as income rises. But that would make the

294

benefit prohibitively expensive. Thus, policymakers face trade-offs in designing these programs.

The EIC differs from most other means-tested transfer programs in several critical respects. First, it is limited to those who work. Non-workers may not receive it. It helps make work more remunerative than non-work rather than the other way around.

Second, the major EIC benefit — the EIC for families with children — is limited to parents that live with their children. Absent parents may not receive it.

Third, unlike welfare benefits that decline as soon as income starts to rise, the EIC increases as earnings rise up to about $8,500 a year (for a family with two or more children). The EIC benefit does not begin to decline until income surpasses $11,000. (These figures are in 1994 dollars.) It then declines at a more moderate rate than public assistance benefits do.

The result is that the EIC rewards work and encourages those not working to enter the labor market. It distinguishes the working poor from the non-working poor and custodial parents from absent parents. As will be discussed later, its marriage incentive and work effects are mixed, with some adverse side-effects and some beneficial effects. As Gary Burtless of the Brooking Institution has noted, the EIC probably has stronger beneficial effects and more modest adverse side-effects than most or all other programs to boost the incomes of low-income families.

Current EIC Issues

I would like to turn to several current issues relating to the EIC. These include questions relating to error and fraud in the credit, work incentives, marriage penalties, and the adjustment of the EIC for inflation. I also will offer some comments on several proposals to reduce EIC costs.

EIC Error Rates

EIC error rates are too high and need to be brought down. In discussing efforts to reduce error, I would offer a caveat. Congress needs to be careful not to institute reforms that make the EIC significantly more complicated for taxpayers and the IRS. Congress also should avoid changes that the IRS cannot administer or enforce. Such steps would almost be guaranteed to increase the error rate further.

We also need to be mindful of progress that has been made in reducing error rates as well as the substantial task that remains. In 1990, staff of the Senate Finance Committee, the House Ways and Means Committee, the Treasury, and IRS worked on a

295

bipartisan basis to craft reforms to reduce the error rate. These reforms were enacted into law in that year's reconciliation bill.

The changes made in 1990 included a major overhaul and simplification of EIC eligibility rules, including the elimination of complex rules that were responsible for a significant share of the errors prior to 1990. The 1990 reforms also included the introduction of a new form — Schedule EIC — that taxpayers must file to receive the credit and that provides the IRS with key information relevant to EIC eligibility it had not previously obtained. Today, the IRS denies the credit to families that do not file Schedule EIC; in the past, it awarded the credit to families that appeared eligible whether the families indicated they were applying for the EIC or not.

These changes were significant. The IRS audits in 1985 and 1988 that found the EIC error rate to be high discovered that one of the principal reasons for the high error rate lay in a series of arcane IRS rules on household filing status that few iow-income families could understand and the IRS could not enforce. Under these rules, a single parent with children could legally file as a head-of-household only if she provided more than half of her household's support. In determining whether she provided more than half of the household's support, the parent was supposed to calculate the household's total support and include in it any AFDC payments, child support payments, and the like the family received. The parent was then supposed to calculate the portion of overall household support that she herself provided, and in so doing, the parent was not supposed to count any income from AFDC or other public benefit payments. Similarly, the parent was not supposed to count, as support she provided, any child support payments she received. If, after completing these calculations, a single parent found that more than 50 percent of the household's support came from sources other than herself, she was supposed to submit her tax return as a single fiier rather than as a head-of-household.

Most parents assume a single filer is someone who lives on his or her own, not a parent who heads a family with children. As a result, large numbers of single parents who were supposed to file as single filers submitted their returns as heads-of- households instead.

This contributed heavily to high EIC error rates since heads-of-households were eligible for the EIC but single filers were not. Adding further to the confusion — and to the EIC error rates — were complex rules on when single parents in such circumstances could claim children who lived with them as dependents.

The EIC law was rewritten in 1990 to address these problems. The new rules dropped the complex test involving AFDC, child support payments and certain other income sources and requiring a measurement of what share of household support the parent provided.

296

In September 1993, the GAO commented favorably on these changes, noting that "OBRA 1990 resolved the major administrative problems associated with complicated filing status and dependency determinations...." The GAO added that "The OBRA changes made it easier for IRS to administer the basic credit because there are no longer different eligibility rules for different filing statuses.. ..Because this issue was the largest source of taxpayer errors in the pre-OBRA period, the change should substantially reduce the number of erroneous EIC claims."3

Also important was the requirement to file Schedule EIC. Prior to 1991, families needed only to file a tax return (the 1040 or 1040A form) to get the EIC. No specific EIC questions had to be answered, and no EIC -specific information provided. The EIC worksheet the IRS had designed was printed in the instructions section of the 1040 and 1040A tax booklets and was not submitted to the IRS. The result was that IRS never received some basic information needed to make accurate EIC eligibility determinations.

Examination of IRS audit findings identified these procedures as a significant contributor to the error rate. Accordingly, the IRS and Congress changed the procedures. Schedule EIC was created. Information needed to determine EIC eligibility, including information the IRS had never previously received, is provided on the schedule. This gives the IRS more of what it needs to make accurate eligibility determinations.

The GAO report took note of this change. "In the past," the GAO observed, "IRS returns processing procedures could not detect erroneous eligibility claims.. ..in part because the tax return did not contain sufficient information. IRS also could not determine whether taxpayers who claimed the credit were eligible for it if the taxpayers failed to provide such information as the child's relation to the taxpayer or the length of time the child resided with the taxpayer."4

Exactly how much these changes may have reduced the EIC error rate is not yet known. The IRS does not yet have data comparable to its 1988 error rate data for a year since these changes were instituted. (The 1993 GAO report also noted that the 1990 law created some new complexities by adding two supplemental EIC credits. One, the health insurance credit, was particularly subject to abuse. Both of these supplemental credits were eliminated in 1993.)

General Accounting Office, Earned Income Tax Credit: Design and Administration Could be Improved, September 1993, pp. 58-59."

4 GAO, p. 6.

297

The GAO on Problems Before the 1990 Law Changes

In its 1993 report, the GAO elaborated on how some of the problems in the EIC before the 1990 law i

"...before ODRA, unmarried taxpayers with children had to file as head of household or qualifying widow(er) with dependent child to get the credit. To qualify as a head of household, a taxpayer had to provide over one-half the costs of maintaining a household; a qualifying widow(er) had to provide over half the support of the dependent child. If taxpayers did not meet the support requirement but still claimed the head of household filing status, IRS could not detect these errors when processing tax returns. Similarly, erroneous EIC claims that were based on ineligible dependents could not be detected when returns were processed. The only way IRS could detect EIC payments that were based on inaccurate return information was to audit the taxpayer's return.

"IRS also could not determine whether taxpayers who claimed the credit were eligible for it when they did not provide complete EIC eligibility information on their returns, such as the relationship of the child or length of time the child resided with the taxpayer. These situations posed a dilemma for IRS. IRS could either assume that the taxpayers were entitled to the credit or it could deny the credit and correspond with the taxpayers for the missing information. If IRS granted the credit on the basis of incomplete information, it had no assurance that the taxpayers were entitled to it. On the other hand, if IRS denied the credit and corresponded with the taxpayers for the missing information, refunds would have been delayed or some eligible taxpayers may not have responded to IRS and would not have received the credit.

"Faced with this dilemma, IRS adopted returns processing procedures that allowed most taxpayers who claimed the credit to get it even though they did not provide all the necessary eligibility information on their returns. ...if a taxpayer failed to provide information on the number of months the child resided with the taxpayer or the relationship of the child to the taxpayer, IRS would still grant the credit. We estimated that in about 21 percent of the EIC claims for tax year 1989 represented by our sample cases, the credit was granted even though the returns had incomplete EIC eligibility information.

"IRS was faced with a similar dilemma when taxpayers did not claim the credit but appeared to qualify for it on the basis of the income, filing status, and dependency information on their returns. IRS adopted returns processing procedures to give these taxpayers the credit, instead of just informing them that they might be eligible for the credit. For tax year 1990, IRS gave the credit to about 564,000 taxpayers who did not claim it. IRS does not have data on how many of these taxpayers were actually entitled to the credit."5

These problems were addressed by the changes in eligibility rules made in the 1990 law and the introduction of Schedule EIC.

GAO, pp. 56-57.

The changes in law passed in 1990, however, addressed only part of the problem. More remains to be done. I believe a large part of what is needed now entails major changes in how the credit is administered. I am encouraged that this year, the IRS has radically altered its processing procedures. The new procedures are somewhat controversial, and it would have been preferable if the IRS had better prepared commercial tax preparers and taxpayers for them. But the new procedures hold strong promise for significant further reductions in the error rate.

Prior to 1991, the IRS did not even get the basic EIC eligibility information now contained on Schedule EIC. Up until this year, the IRS made limited use, before making EIC payments, of the information it did receive. Now this is changing.

Until this year, the IRS processed the EIC almost solely based on information on the tax return, without verifying any of this information before making payments. The validity of social security numbers for parents and children was not checked before EIC checks were mailed. No social security numbers were required for very young children. (This was due to statutory limitations.) EIC refunds were not held up while questionable information was verified.

Now, social security numbers are verified before payment is made. EIC returns are scrutinized, and those subject to question on any of a number of grounds are held up, and further information gathered on them, before the EIC is awarded. In addition, legislation enacted last fall extends the requirement for the provision of social security numbers to very young children as well.

These processing changes, along with other changes aimed at commercial preparers that use electronic filing, are the most sweeping changes in processing procedures since the EIC was instituted. They are likely to have a substantial effect.

In this vein, I believe at least one other procedural change may be needed. As recently as 1993, the IRS paid EIC refunds without checking the W-2 form to see if the taxpayer had already received an EIC advance payment. This led to double payment in some cases (although not in a great number of cases since use of advance payments is very small). I believe, but am not certain, that the IRS has resolved this problem as well. If not, the IRS should institute procedures to check W-2 forms to determine if advance payments are made before issuing an EIC refund. I believe the GAO has made a similar recommendation.

The EIC and Work Incentives

Important issues are frequently raised about the EIC and work incentives. Based on economic theory, it is generally assumed that the EIC encourages work among those

10

299

working little or not at all while acting as a modest work disincentive for families whose income exceeds $11,000 and whose EIC benefit is reduced as their earnings rise.

These issues are sometimes oversimplified or misunderstood. We do not know whether the work disincentive effort is a significant one. For it to be, affected families would need to understand how the EIC affects their marginal tax rate, and it isn't clear that many do. If they know that they receive an EIC which lowers their overall tax burden but are not aware of its effect on their marginal tax rate — which is likely to be the case for a large number of families — the EIC could encourage rather than discourage them to work more. One recent analysis of labor market data finds that the EIC succeeds in increasing work among single mothers who have previously been out of the labor force and accomplishes this without diminishing work effort among single mothers whose EIC benefits decline as their earnings rise.6

Even if the EIC does have some disincentive effects upon the group of families whose EIC falls as their earnings increase, it is important to identify which types of families are affected. This point was emphasized by Robert Reischauer, former director of the Congressional Budget Office, and Henry Aaron, Director of Economic Studies at the Brookings Institution, at an American Enterprise Institute symposium on the EIC in 1990. Reischauer and Aaron noted that the families encouraged to work more by the EIC are likely to be quite different from the families that may be encouraged to work less. The families encouraged to work more, they observed, are heavily made up of single parents working little if at all — precisely the group whose work effort we most want to increase. By contrast, the families whose EIC benefits decline as their earnings rise — families with incomes over $11,000 — include a large number of two-parent families. Reischauer and Aaron emphasized that a major part of the EIC's effect in inducing modestly less work among families in this income range is likely to result from the credit's effect in enabling one parent in married families in which both parents are employed to work a little less so she can spend more time with her children.

Reischauer stressed that this should not be regarded as an adverse outcome and that it may be positive for a parent in two-parent working families to spend more time with her children. He cautioned that lumping increased work effort among single mothers who worked little with reduced work effort among married mothers in two- parent working families — and producing a negative number on the EIC's net effect on hours worked — could confuse rather than illuminate this issue.

Reischauer's point is supported by data provided in a GAO analysis. The GAO estimated that the percentage reduction in hours worked as a result of the EIC is four

6 Nada Eissa (University of California, Berkeley) and Jeffrey B. Liebman (Harvard University), "Labor Supply Response to the Earned Income Tax Credit," December 5, 1994.

times greater among wives in two-parent families than among husbands in such families. In addition, the GAO's estimates showed no significant percentage reduction in work effort among single parents.

It also is important to place the estimates of the EIC's effects on work effort in perspective. The conclusion the GAO drew in Congressional testimony in 1993 is noteworthy in this regard. The GAO testified: "[The earned income tax credit] works. It offsets payroll taxes, increases progressivity of the tax system, and provides a positive work incentive to the lowest income group with only a slight disincentive to other recipients."7

The EIC and Marriage Penalties

The issue of the EIC and marriage penalties is another complex area. The EIC penalizes marriage for some and rewards it for others. It does not represent an unambiguous marriage penalty. In addition, some proposals to reduce the marriage penalty are unwise because they would substantially lessen the EIC's work incentive efforts among poor single-parent families, diminish the attractiveness of work as compared to welfare, and shift large sums from working poor families to middle-class families.

Sometimes, hypothetical cases are cited in which the EIC imposes a very large marriage penalty. The example most commonly cited involves two potential marriage partners who each are custodial parents with at least two children living with them and each earn about $11,000. This hypothetical case does not provide the best basis for policymaking. It is the hypothetical case in which the marriage penalty is greatest. This case rarely exists in the real world.

There are few cases in which two people who wish to marry are both single custodial parents who each live with at least two children and each have incomes in this range. Most male single parents are not custodial parents and are not eligible for the EIC — and thus do not lose any EIC benefits if they marry.

Census data indicate that of all marriages that occurred in 1990, fewer titan two- tenths of one percent involved a man with two or more children marrying a woman with two or more children. The likelihood of such a marriage in cases in which both parties are custodial parents and also have incomes around $11,000 appears to be close to zero.

GAO testimony before the Subcommittees on Select Revenue Measures and Human Resources, House Ways and Means Committee, March 30, 1993.

301

This does not mean the marriage penalty issue is not significant but that more reasonable examples should be used in discussing the issue. The most common example in which the EIC can penalize marriage involves a single working mother with one child who is considering marrying a man who earns modest wages but does not live with a child or receive the EIC. The EIC that such a couple would receive if they marry could be lower than the EIC the mother currently gets. If a single mother with one child who earns $5,000 working part-time marries a male earning $10,000, the EIC benefit in 1996 would drop $82.8 If she marries a man earning $15,000, the EIC would decline $881.

If the mother earning $5,000 has tivo children, marrying a man earning $15,000 would cause a smaller EIC loss — about $200. And if such a mother marries a man earning $10,000, the EIC benefit would increase about $850.

On the other hand, if the mother earns $10,000 herself and has two children, the EIC benefit loss becomes larger. For example, if she marries a man earning $10,000, the loss is $1,760.

The EIC thus can penalize marriage. What is less well known is that it also can reward marriage substantially. The EIC offers a sizeable marriage incentive to a single mother who has no earnings and receives AFDC. This is significant, since marriage is one of the principal routes out of welfare.

In the absence of the EIC, a mother on AFDC who does not work and is contemplating marrying a man with low earnings risks losing a great deal. If she marries, she and her children will become ineligible for AFDC and also lose some of their food stamps. In addition, she likely will become ineligible for Medicaid. Depending on her children's ages and her husband's earnings, some or all of her children may lose Medicaid coverage as well. The marriage penalties embedded in the welfare system are strong.

The EIC helps to offset these losses and ease these penalties. By marrying, the couple will gain an EIC benefit of up to $2,157 if they have one child and up to $3,564 if they have two or more children. This will partially — and in some states, wholly — make up for the mother's loss of AFDC benefits when she marries.

For poor single mothers with little or no earnings, the EIC thus significantly lessens the marriage penalties in the welfare system. It provides these women an incentive to marry and leave welfare. This point is sometimes overlooked when the EIC and its effects on marriage are discussed. It is not accurate to speak of the EIC as

8 All figures in this analysis are for 1996, when the EIC expansions enacted in 1993 are phased in fully.

13

302

simply penalizing marriage. It penalizes it for some and rewards it for others, including the group for which there is the greatest concern for encouraging marriage, single mothers on welfare.

CHANGE IN EITC BENEFITS IF MARRIAGE OCCURS Mother Has One Child

Mother's Earnings	Male's Earnings are $5,000 $10,000 $15,000
0	+1,376	+2,157	+1,618
5,000	+133	-82	-881
10,000	-863	-1,338	-2,137

Mother Has Two Children

Mother's Earnings	Male's Earnings are $5,000 $10,000 $15,000
0	+1,676	+3,564	+2,854
5,000	+1,240	+854	-199
10,000	-1,034	-1 ,763	-2,816

Reducing the Marriage Penalty

One option recently suggested to ease the marriage penalty is to restructure the EIC in a budget-neutral manner so EIC benefits for two-parent families are essentially double those of single-parent families. The credit for a married couple with children would be twice as high as the credit for a single-parent family with the same income and the same number of children.

This would eliminate the marriage penalty for some of those for whom the EIC now creates such a penalty and lessen the penalty for others among this group. But in so doing, this change would have other, undesirable effects. If the credit for two-parent families is to be made twice as large as the credit for single-parent families and this is to be done without increasing its cost, the credit provided to single-parent working families — most of whom are low-income families working at low wages and not

14

collecting AFDC — must be cut sharply. The result would be to lessen incentives to leave welfare for work. The problems that would result from restructuring the EIC in this manner are significant.

Such an approach would necessitate making two-parent families with incomes well above $30,000 eligible for the EIC. To provide larger credits to two- parent families, the maximum credit for these families would have to be raised substantially. As a result, the income level at which the credit phased out entirely would have to be increased as well; otherwise, benefits would be reduced too rapidly as earnings climbed, pushing marginal tax rates too high. Under current law, the EIC will be available to families with two or more children that have incomes up to about $28,500 in tax year 1996. If the EIC is restructured so the credit for two- parent families is double that for single-parent families, the income limit for two-parent families would have to be raised substantially, probably to somewhere in the $30,000 to $40,000 range.

As the EIC income limit is increased for two-parent families, the credit will reach into a "dense" part of the income distribution; a large proportion of families with children have incomes between $30,000 and $40,000. As the credit's income limit is raised, the credit becomes much more costly since a rapidly increasing number of families gain eligibility for it. This means that as the EIC income limit for two-parent families rises, the credit for single-parent families must be cut still more sharply to pay for it.

Incentives for families to leave welfare for work would be reduced significantly. With EIC benefit levels for single-parent working families being cut sharply, the incentives for parents to leave welfare for work would be lessened. Poor single-parent families who work their way off welfare would receive substantially smaller EIC benefits and have less after-tax income than under current law. Bipartisan efforts of recent years to make work pay more than welfare would be set back, and part of the underpinning for work-based welfare reform would weaken.

Billions in EIC resources would be shifted from working poor families and their children to middle-income families. Single-parent families on EIC constitute a much poorer group than do the two-parent families receiving this benefit. Data from the 1993 Green Book show that single-parent families account for more than two- thirds of the EIC families with incomes below $10,000, while two-parent families constitute a majority of the EIC families with incomes above $20,000. The effect of reducing the EIC for single-parent families, increasing it for two-parent families, and extending it to two-

15

304

parent families at higher income levels would be to shift large amounts of income from poor or near-poor single-parent working families to two- parent families in the middle-income range.

Both the effect of the EIC on marriage and efforts to ease the marriage penalties are thus more complex than they may initially seem. This is a thorny area in which policy should move with caution.

>4ssefs

Another question is whether assets limits should be included in the EIC eligibility criteria. On the one hand, taxpayers with very substantial assets do not need the EIC. On the other hand, assets tests of the sort used in some other means-tested programs are cumbersome and costly to administer and cannot be administered through the tax code.

The approach reflected in H.R. 831 addresses these issues in a reasonable way. I would, however, recommend one modification to H.R. 831.

The $2,350 limit on interest, dividends, rents and royalties should be indexed. Otherwise, this limit will erode in real terms over time and compel families with a legitimate need for the EIC to stop saving and consume enough of their liquid assets each year to remain eligible for the EIC. Also, as the $2,350 limit erodes each year, it will disqualify families with steadily lower amounts of assets. I would suggest reducing the $2,350 limit to a level that, when indexed, yields the same level of savings over five or 10 years as the provision in H.R. 831.

I would not recommend reducing this limit below a $2,000 indexed limit. A lower limit would disqualify some poor and near-poor working families that are saving for such legitimate reasons as sending a child to college, purchasing a home, or meeting a medical emergency, an increasing concern as the proportion of low-income working families lacking health insurance rises. We should not force such families to choose between losing their EIC (if their modest assets simply keep pace with inflation) and "dissaving" so they remain below the EIC limit. (Also, homeowners are not disqualified for the EIC, and it would pose equity problems to disqualify those who are saving so they can purchase a modest home while those who already own such a home can receive benefits.)

The EIC and Inflation Adjustments

The final proposal I would like to examine is a proposal recently suggested by Senator Gregg to halt indexation of the EIC. I commend the effort to determine if the Consumer Price Index is overstating inflation and, if so, to correct for this matter. Such

16

305

a correction would affect the EIC, along with a range of other provisions of the tax code (such as the personal exemption, the standard deduction, and tax brackets) and many entitlement benefits, including Social Security.

But ending the indexation of the EIC would be a unwise course to follow. It would diminish the effectiveness of the EIC and, over time, result in substantial tax increases for large numbers of low-income working families with children. Millions of working poor and near-poor families would find that if their earnings simply kept pace with inflation, their payroll taxes would rise each year while their EIC — which is intended, in part, to offset payroll taxes — declined each year. The result would be steadily increasing tax burdens for working families that had experienced no increase in purchasing power. Moreover, because other key parts of the tax code are indexed, higher-income families would continue to be shielded from the effects of inflation on their tax burdens while working poor and near-poor families faced substantial tax increases.

This proposal also would be inconsistent with welfare reform goals because over time, it would reduce the advantages of working rather than receiving welfare. It also would sharply raise marginal tax rates on large numbers of low-income working families.

Indexing and the Federal Tax Code

Ending the indexation of the EIC would violate a principle advanced by President Reagan and enacted into law in the 1980s — that the basic features of the income tax code should be indexed so taxes do not creep up for working families whose incomes are rising only at the pace of inflation. It was President Reagan who proposed the EIC be indexed for inflation.

In some ways, indexing is even more important in the EIC than in the personal exemption and standard deduction. Indexing those two features of the tax code keeps a family's tax burden constant as its income rises with inflation. For certain EIC recipients, this is the effect of indexing as well. But for millions of other EIC families, indexing the credit is necessary to ensure that the EIC these families receive is not cut when their income rises at a rate equal to or less than inflation.

Indexing avoids cuts in the EIC benefits of millions of low-income working families because of basic features of the EIC's design. The EIC is phased out above a certain income threshold. For a family with two or more children, the threshold will be about $11,600 in 1996. For every dollar a family earns above $11,600, the EIC is reduced 21 cents. This income threshold is indexed for inflation.

17

306

The Structure of the EIC and EIC Indexing

The structure of the EIC includes an "upslope," a "plateau," and a "downslope." For each dollar a family earns up to a certain level, the value of the EIC is increased. In a family with two or more children in 1996, the EIC equals 40 percent of earnings for each dollar this family earns up to $8,910. Thus, a family with two children and earnings of $5,000 would receive a credit of $2,000. A family earning $8,910 would receive a credit of $3,564.

As the family's income rises above $8,910, the EIC remains constant (at its maximum level of $3,564) until income reaches $11,630. This is the plateau.

Once a family's income passes $11,630, the EIC drops by 21 .06 percent — or about 21 cents — for each additional dollar of income. This is the downslope. When family income reaches $28,550 the value of the credit falls to zero.

Both the dollar level at which the EIC stops increasing as earnings rise ($8,910 for a family with two or more children in 1996) and the dollar level at which the EIC starts falling as income rises (about $11,630 in 1996) are indexed for inflation. This indexation feature of the EIC was proposed by President Reagan and enacted in 1986 as part of that year's Tax Reform Act.

	Federal EIC Benefit Stnjctur With Two or More Children	c For Families Tax Year 1996
54.300 J.V00O	cwu.	>c
SiOOO
S 1,000
	V.91G M1.H0 V inliHi		tssso

18

307

If the indexation feature is repealed, a family with income at or above $11,600 whose income rises at the rate of inflation will find that its EIC falls in absolute terms (i.e., not just in inflation-adjusted terms) even though the family's purchasing power has not increased. This family also will owe more in payroll taxes. For each dollar its earnings rise, its payroll tax will increase 7.65 cents while its EIC drops 21 cents.

These effects are illustrated by the following examples, which use CBO's inflation forecast.

• Take the case of a family of four with two children and earnings of $12,000 in 1996. This family's earnings will leave it more than $4,000 below the poverty line next year; the family's income will equal a little less than 75 percent of the poverty line. The EIC the family receives bridges part of this gap. After the family's EIC and payroll taxes are figured in, the family's income will equal about $14,570, or about 90 percent of the poverty line.

The poverty line rises with inflation. Under current law, if the family's wages keep pace with inflation, its payroll taxes and its EIC rise with inflation as well — and the family's income, after taxes, remains at 90 percent of the poverty Line.

But if the EIC is not indexed and the family's income keeps pace with inflation, its EIC will decline while its payroll taxes are rising and the poverty line is increasing. After five years, if the family's income simply keeps pace with inflation, its EIC will have fallen $460 below what the family received in 1996, while its payroll taxes have climbed $167.

• Near-poor families would be affected as well. Under current law, a family with two children earning $20,000 will receive an EIC of about $1,800 in 1996. (The family's credit will offset most but not all of the family's payroll taxes of $1,530 and its federal income tax liability of $430.) Under current law, if the family's earnings rise to $20,680 in 1997 — an increase equal to the projected inflation rate — its EIC also will rise in tandem with inflation, to $1,860.

If the EIC is not indexed, however, the family's EIC will fall from $1,800 in 1996 to $1,660 in 1997, a $140 decline. The EIC will fall because the family loses 21 cents of its EIC for each additional dollar of income.

At the same time that the family's EIC would be falling, its payroll taxes would be rising. Although this family's income would not have risen faster than inflation, its net tax bill would increase.

19

308

• This effect would grow larger with each passing year. If a family's

income simply kept pace with inflation and the EIC was not indexed, the family with income equal to $20,000 in 1996 would receive an EIC five years later that was $767 smaller than the EIC it received in 1996. Its EIC would be $1,094 lower than it would have been had the EIC been adjusted for inflation. In real terms, the purchasing power of its EIC would decline by 51 percent in only a five-year period even though its real income had not risen at all.

Working poor families at still-lower earnings levels, such as those working full time year round at little more than the minimum wage, also would be adversely affected. Consider a family that earns $4.50 an hour in 1996, or $9,000 a year, and is far below the poverty line. If the family's earnings keep pace with inflation, its EIC will rise with inflation as well, under current law. If indexing is eliminated, the purchasing power of its EIC will erode. After five years, its EIC will have lost 15 percent of its purchasing power. After 10 years, the purchasing power of the family's EIC would have fallen by 32 percent. Meanwhile, the family's payroll taxes would have increased every year, and the poverty line would have risen. The family would have fallen steadily deeper into poverty.

Ending the indexation of the EIC consequently would make millions of working poor families with children poorer over time. It also would turn a steadily increasing number of near-poor working families into poor families, by pushing them below the poverty line, and reduce the EIC for millions more who work hard and are modestly above the poverty line.

Ending the indexation of the EIC would be inconsistent with the goals of "making work pay" and promoting work over welfare. If the EIC is not adjusted for inflation, as its real value declines, the gain from working at a low-wage job rather than relying on welfare will erode.

Failure to index the EIC would undermine the EIC's effectiveness in one other way as well, by weakening the work incentive features of the credit. Currently, families earning $10,000 or $11,000 a year do not face an increase in their marginal tax rates when their earnings rise with inflation. But if EIC indexing ends, such families would be pushed above the point where their EIC begins to drop 21 cents for each additional dollar earned. The marginal tax rate these families face would rise 21 percentage points.

Some of these families already face substantial marginal tax rates because they receive other benefits such as food stamps that decline as earnings rise. To raise their marginal rates another 21 percentage points due to failure to index the EIC could raise their marginal tax rates to disturbingly high levels.

20

309

Finally, ending EIC indexation would be inconsistent with other Congressional action. There is no discussion of ending the indexation of other features of the tax code such as the personal exemption, the standard deduction, or the tax brackets. In addition, the House of Representatives is likely to pass legislation to index capital gains tax benefits. Some policymakers also are proposing various tax cuts. It is difficult to discern how ending EIC indexing fits in with these other policies unless the principle is that investors and middle- and upper-income families need protection against the effects of inflation on their taxes but low-income working families do not. (Furthermore, if EIC indexation is ended while various tax cuts also are approved, low- income working families would face tax increases while families at higher income levels received tax cuts.)

Other Proposals

Finally, I would offer brief comments on several other proposals.

• There are arguments both for and against including Social Security in adjusted gross income (AGI) for purposes of determining eligibility and benefit levels for the EIC. EIC savings are desired, this option probably warrants investigation.

• It would be a mistake to attempt to include child support in AGI for EIC purposes. Attempting to count child support payments as part of AGI would pose serious problems for the IRS. Such a rule would not be enforceable to any substantial degree. The result would be higher error rates and further damage to the EIC's integrity and reputation. Furthermore, attempting to count child support payments in this manner would cause double-counting of income, since non-custodial parents pay income tax on the income from which child support payments are made.

• I would strongly recommend against reinstating the type of complex, error-prone rules that contributed to high error rates in the 1980s and were repealed in the 1990 reconciliation act. This includes rules such as the "support test," which is largely unenforceable.

• To further reduce error, I would urge consideration of a proposal made several years ago by Senator Packwood, and praised by the GAO, to modify the rules for claiming the personal exemption for a child so they match the rules for determining when a child is a "qualifying child" for EIC purposes. This would simplify tax rules and should reduce errors in both the EIC and the personal exemption areas.

21

310

Conclusion

While steps have been taken in recent years to reduce EIC error rates, more needs to be done. Some other EIC changes also warrant consideration.

We should not, however, lose sight of the EIC's virtues. The EIC is boosting the incomes of millions of poor and near-poor families with children that are working and staying off welfare. With the steady erosion of wages for low-paid work over the past 20 years and the likely continuation of this trend in the future, the EIC is a critically important element of the tax code. The EIC also provides an important underpinning for welfare reforms to move families from welfare to work. It is part of the welfare reform strategies of policymakers such as Governor Engler of Michigan.

In addition, the EIC has helped change the tax system into a system that helps lift working families out of poverty instead of pushing more working families into poverty. The Department of Health and Human Services testified that in 1984, the tax system pushed into poverty 1.8 million people who lived in families with children. When the recent EIC expansion is fully phased in, HHS estimates the tax system will lift more than two million such people out of poverty.

The EIC needs significant improvement to reduce error and fraud. But the EIC's mission remains as important as it ever. Deep reductions in the EIC benefits of honest low-income working families would not be a desirable part of the EIC reform agenda.

22

311

GAO RESPONSES TO QUESTIONS FROM SENATOR GLENN

1(a). What can be done to address these [TSM] deficiencies'? Is it realistic to as- sume they will be addressed? How long will it take?

Our work on IRS' modernization efforts has identified many management and technical issues that need to be addressed to mitigate critical risks and better posi- tion IRS to achieve success. We have concluded that, to enhance its chances of suc- cessfully delivering TSM, IRS needs to:

(1) Refocus its business strategy (including marketing and education) to maximize electronic submissions, aggressively targeting those sectors of the taxpaying population that can file electronically most cost beneficially.

(2) Define and commit to a plan and schedule for completing and institu- tionalizing process improvements for

strategic information management; systems development capability; cost/benefit analyses; and data, security, interoperability architectures.

(3) Improve organizational authority and commitment including strength- ening decision making and follow through to

• ensure strategic planning documents are complete and consistent;

• develop a formal plan and schedule for linking reengineering efforts to sys- tems development projects;

• exercise consolidated control over all information systems investments agen- cywide, including all new

• systems in research and development and operational systems being up- graded and replaced; and

• ensure that defined systems development standards and architectures are en- forced agencywide.

1(b). If these problems were corrected sufficiently and in a timely nature, would TSM be successful? What is its potential, from a cost-benefit perspective?.

The successful implementation of a major systems modernization cannot ever be fully guaranteed. However, if these steps are adequately imple- mented, IRS will have mitigated critical risks associated with its ongoing systems modernization. Following these steps should also provide IRS with a more cost effective modernization.

2. When do you think IRS books and records will be auditable? Where would they be now if they were a taxpayer or a small business?

IRS' books and records could be auditable in fiscal year 1995. To achieve auditable financial statements will require renewed emphasis and priority by IRS' senior management, especially the CFO and his staff, to ensure that corrective actions are implemented to fix the rudimentary problems that have persisted. Most of these problems relate to ensuring that all amounts reported in financial statements are supported by itemized details and appropriate evidence that transactions were valid — meaning author- ized, occurred during the period, and for the correct amount. One principal problem is that IRS does not perform reconciliations from its detailed books and records to reported financial statement amounts.

A small business or taxpayer in IRS' position would have a difficult time calculating total income, as well as supporting the costs of their operations. They would likely not meet the record-keeping requirements imposed on taxpayers for supporting a tax return — which in some respects are financial statements for some taxpayers.

3. Why hasn't the IRS, in your opinion, been able to do the rudimentary book- keeping procedures necessary to fully account for its appropriated funds — for exam- ple, reconciling its records with Treasury's records?

IRS has not done the rudimentary bookkeeping procedures necessary to fully account for appropriated funds because staff responsible for perform- ing these tasks are not properly supervised and trained, and managers have not been held accountable for ensuring that these tasks were properly performed. Further, IRS needs to ensure that it employs staff with the proper mix of skills necessary to carry out these tasks.

312

4(a). How accurate and reliable is the IRS' accounts receivable information? What are the. primary obstacles to overcome?

As we have highlighted in our financial audits and in our 1992 and 1995 High Risk reports on IRS' Receivables, the lack of accurate and reliable in- formation is the foremost problem facing IRS. This serious weakness has hindered IRS' efforts to effectively collect delinquent taxes. The extent of in- accuracies is unknown, but IRS' own sample review indicates that only about one-half of the recorded tax debts are valid and less than one-half of the valid debts are collectible. Sampling, however, only gives an estimate of the situation. Without an indication of the validity and collectibility of each account in the inventory and the sources of the invalid accounts, little can be done to design programs to eliminate unproductive cases and collect more revenue. In this regard, as we have reported IRS abated about $37 billion of its receivables in fiscal year 1994 while only collecting $23.5 bil- lion.

The basic problem emanates from IRS' antiquated financial management systems and computer and manual processes which were not designed to provide meaningful and reliable financial data. These systems and proc- esses were put in place during the 1960s and have not kept pace with the significant increase in the volume of information received or with advanced technology. In this regard, much of the tax information is still batch proc- essed and relies on computer programming designed decades ago.

IRS believes that its Tax Systems Modernization efforts will address the problem of inaccurate and unreliable information. However, those efforts are still being developed and their impact will not be known for several years.

Another problem lies with IRS' organizational structure and the need for an institutional focus to address accounts receivable issues which involve most functions within the agency. For example, currently no single organi- zational function has the responsibility or accountability for both adminis- trative and revenue accounting activities. While we have recommended that this responsibility and accountability be combined under the Chief Finan- cial Officer to provide the focus needed to correct the problems, the activi- ties remain separate.

4(b). What should we look for in the outcomes and performance of IRS' new initia- tives to improve this information? What standards should we measure it by?

IRS' new initiatives must be able to produce accurate and reliable informa- tion on the validity and collectibility of all accounts in the receivable inven- tory. Currently, the lack of such information results in a significant number of unproductive accounts which cause inefficiencies in other IRS program areas.

With more accurate and reliable information, IRS could effectively direct its efforts to the best targets of opportunity thus increasing the amount of rev- enue collected. As a minimum, performance measures should include in- creased collections and reduced abatements. Also, there should be measures that differentiate valid accounts receivable from assessments made as a-re- sult of IRS enforcement and compliance efforts.

IRS RESPONSES TO ADDITIONAL QUESTIONS FROM SENATOR GLENN

1. Question: Two years ago, when we held our first oversight hearing on browsing, you told the Committee that even one violation of a taxpayer's privacy was too many. I couldn't agree more strongly. Nonetheless, I am concerned that there have been many violations — and potential violations — in which IRS employees gained access to taxpayer files for non-business reasons. What actions have been taken by the IRS to stop browsingf

Answer: The IRS has worked with the Treasury Department in providing lan- guage to the Department of Justice's computer crime bill so it would address unau- thorized access to tax return information through use of a computer — including amendments to 18 U.S.C. 1030. In addition, the Service requested that Treasury propose amending section 7213 of the Internal Revenue Code since this statute spe- cifically addresses the confidentiality of tax information. As modified, it would ad- dress all improper review of tax information, not just computer accesses. The Serv-

313

ice favors legislation that will make the unauthorized review and accessing of re- turns and return information a misdemeanor.

While the imposition of appropriate disciplinary sanctions is an important part of ensuring that taxpayer privacy is maintained, the Service also has taken a number of preventive measures to ensure that all of IDRS users know and understand their responsibility for protecting taxpayer privacy. The Commissioner has sent a memo- randum to all Service employees emphasizing that security of taxpayer information is one of the most important issues facing the IRS today, and that any access of taxpayer information without legitimate business reason violates taxpayers' privacy and will result in sanctions — up to, and including, removal from the Service.

In addition, the Commissioner and the Chief Inspector have issued an "Inspection Integrity Alert" to 17,000 managers on privacy and security issues. The IRS also recently completed the development of training for IDRS users that emphasizes safeguarding taxpayer privacy. All managers and their employees will have com- pleted that training within the next six months. The Commissioner also approved an "IRS Information Security Policy" that lays out the foundation for functional se- curity requirements for Tax Systems Modernization.

The IRS has revised its orientation program for new employees to include ex- panded coverage of security and the protection of taxpayers' information. The mes- sages contained in the orientation session are reinforced by managers in the work place.

Screen warning software was completed in February 1995 and will be imple- mented on all IDRS terminals following the completion of labor/management nego- tiations. Employees will be required to acknowledge a privacy warning every time they sign on to "the IDRS system. Users will access IDRS using their password and then see a privacy warning screen; to further access the system, users will have to reenter their password again.

Significant progress has been made in addressing the security issues raised in the August 1993 and July 1994 Governmental Affairs Committee hearings. The Com- missioner has indicated that even one employee breaching the public trust will not be tolerated; IRS data indicates that employee misuse of IDRS is limited to less than one percent of the 56,000 IDRS users who process 1.2 billion transactions a year.

After the 1993 hearing, the IRS developed an action plan to improve the security and privacy of tax information in the IDRS database. As of February 1995, the agency has completed 30 of 35 action items. Of the remaining five, two actions are ongoing/on schedule, while 3 action items have been rescheduled. The Senior Coun- cil for Management Control, chaired by the Deputy Commissioner, oversees the ac- tion plan, the major accomplishments of which include:

• A policy statement on privacy rights was issued to all employees emphasizing the need to protect taxpayers from unnecessary intrusion into their tax records.

• Ten basic privacy principles were enumerated, establishing a public trust for protecting taxpayer privacy and safeguarding the confidentiality of taxpayer information. These principles were distributed to all employees in May, 1994 and are now being discussed in employee group meetings throughout the Service.

• Enhanced procedures for tracking account adjustment activity in IDRS were developed. The IRS now reviews transcripts and listings of IDRS cases to identify high risk account activity.

• A review of the IRS Safeguard Review Program was conducted to ensure it met statutory requirements.

• "The Guide for Penalty Determinations" mentioned earlier has now been is- sued to all employees.

• The Electronic Audit Research Log (EARL) was installed in all service centers in March 1994, allowing security staff and management to perform ad hoc, or "custom", IDRS audit-trail searches. As of mid-February of 1995, all service centers had been equipped with enhanced computer systems and EARL soft- ware to provide audit trail reviews for detecting "browsing" of taxpayer records by IRS employees.

2. Question: What specifically has been the impact of the IRS' revised penalty guide and the Electronic Audit Research Log (EARL), the computerized search for browsing patterns?

314

Answer: The Service believes that EARL is becoming an effective deterrent. In No- vember 1994, all IDRS users received a memorandum explaining EARL, which indi- cated that the system would be used to detect unauthorized access to IDRS in order to protect taxpayer privacy. Custom search scenarios have been implemented in all service centers since March 1994. Development of additional customized search ca- pability is ongoing.

As mentioned, in February 1995, upgraded hardware and enhanced name-search capability was installed at all service centers. Additional custom search scenarios are now being implemented at the service centers, greatly enhancing the ability of the EARL system to detect unauthorized abuse.

Using just the name search scenario, 41 cases or authorized access to taxpayer accounts were recently identified. Those cases were referred to management and are being adjudicated.

3. Question: For the record, where do we stand with browsing cases? I am con- cerned that if new cases continue to be found — after all the publicity and warnings — then we need to do more to prevent IRS employees from violating taxpayers' privacy. What is the status of browsing cases? How many new cases of snooping have been found? How were those cases handled? How many people have been fired from the IRS for violating taxpayers' privacy?

Answer: Since the August 1993 Senate hearing, the IRS Office of Ethics and Busi- ness Conduct comprehensively reviewed the Service wide computer security and tax- payer privacy cases in our Automated Labor and Employee Relations Tracking Sys- tem (ALERTS is used to document the status of potential disciplinary and adverse action cases throughout the Service.) This review, covering a period of two full cal- endar years, identified 1,494 potential IDRS security violations. Of these 1,142 have been fully investigated and 352 remain open. Of the 1,142 completed investigations, management action was taken in 685 cases (65 separations/resignations, 93 suspen- sions, 90 reprimands, 85 admonishments, and 352 counseling/caution sessions). The remaining 457 cases did not warrant any management action.

4. Question: You have mentioned that actions taken by the IRS are having a posi- tive impact on browsing — and yet this problem seems to persist. Are there more cases now because of the actions you have taken or despite those actions? Is there any way to know?

Answer: For the purposes of determining trends, we compared calendar years 1993 and 1994 data related to potential computer security and taxpayer privacy vio- lations. Increases were experienced in a number of categories. The data suggests that the increase in receipts for 1994 reflects the Service's efforts to implement bet- ter systems to detect potential taxpayer privacy violations. The IRS is early in the process of adjudicating these cases and, therefore, is unable to provide any addi- tional specificity at this time.

5. Question: It is my understanding that TSM will help to eliminate browsing and other violations of taxpayers privacy. If Congress does not fund TSM adequately, will it be more difficult for the IRS to ensure that Americans' privacy will be protected? With TSM, IRS employees will have greater access to tax data than they have now — How will you prevent that information from being abused?

Answer: Yes, as pointed out by the Commissioner's Advisory Group in December 1994, continuous full-level funding for security is imperative if the IRS is to success- fully develop and field sophisticated the necessary TSM infrastructure security and related information systems capabilities. These critical security components will fa- cilitate the prevention of browsing and other security violations. While no tech- nology-based system or security solution is totally fool-proof, well designed security and auditing systems will significantly reduce the chances for abuse. IRS' TSM plans include a system security architecture which will easily facilitate insertion of evolving security technology to assist in countering threats as they surface.

6. Question: GAO recently has raised some concerns about the progress and designs of the TSM system. I want this to work as promised — as I know you do, too — because it is so very important. If there are issues outstanding, now is the time to address them. When will you have validated costs for TSM's design, development and imple- mentation ?

Answer: The Office of Economics Analysis is developing the long-term economic analysis model that will be used for forecasting TSM costs and benefits. During and after TSM's implementation, the Service will conduct reviews to determine whether benefits are being achieved as expected. A more complete report will be provided

315

by September 1, 1995, as required by the Treasury Postal and General Government Appropriations Act of 1994.

7. Question: Do you have — on paper — a technical design for a comprehensive sys- tem to provide the capabilities for making this vision a reality? If so, please provide me with a copy.

Answer. Yes, the IRS has a technical design reflected in our Business Master Plan, Integrated Transition Plan and Schedule, and in the Concepts of Operation plan. These documents provide the program and project level priorities for TSM and clearly identify the milestones for achieving the Business Vision objectives. They also form the baseline set of plans and schedules against which performance is mon- itored and the success of TSM may be determined. A detailed extract of agency plans and schedules, based the actual FY 1995 TSM budget and the President's budget request for FY 1996, was completed last March. The Associate Commissioner Designee is working closely with the Chief Information Officer to complete delivery of the remaining architectural and design components of TSM.

8. Question: Know that the accounts receivable numbers continue to increase, but I was also quite pleased that, with the help of GAO, in the last two years you have gotten a more realistic sense of the real amount of money that is collectable. Although accounts receivable is increasing, isn't some of this increase good news because it means that IRS has been able to put together installment agreements with taxpayers that ultimately will bring in more revenue? Can you update us on the accounts re- ceivable situation and give us some sense of what is — and what is not collectable?

Answer: In assessing IRS accounts receivable inventory, it is important to not just focus on the dollar amount of the inventory. Like private businesses, the IRS has accounts receivable. Receivables are not inherently good or bad. Because IRS gross receipts are higher than most businesses, almost $1.3 trillion in 1994, the accounts receivable are also higher than most businesses. Furthermore, a growth in gross re- ceipts usually results in a growth in accounts receivable. Thus, most private busi- nesses rely on the ratio of receivables to gross receipts as an indicator of perform- ance. The ratio of IRS active accounts receivable inventory to tax receipts has re- mained constant at 6% to 7% over the last 5 years.

The constant ratio of receivables to tax receipts, as well as the growth due to the extension of the statute to 10 years, and the accrued penalties and interest, are in- dicators that the problem is not worse than 5 years ago. Still, the IRS' goal is to improve performance in collecting tax revenues, which includes effective manage- ment of the accounts receivable inventory.

IRS research indicates that 86.5% of the taxes that are due each year are col- lected. There are plans in place to increase this to 90% by the end of 2001 — an addi- tional $40-$60 billion without any change in the tax rates. Addressing the portion of noncompliance reflected in the accounts receivable inventory is a very important part of these plans.

As early as 1988, the IRS reported accounts receivable as a material weakness under the Federal Managers' Financial Integrity Act (FMFIA). In 1989, GAO and OMB identified IRS accounts receivable inventory as a high risk item. For the past two years, the IRS has listened to the concerns expressed by GAO and others and have taken steps to improve the management of the receivables inventory. These actions were taken as a part of a four-step strategy: to determine the correct com- position of the accounts receivable inventory; ensure the accuracy of the assess- ments that are included in the inventory; improve the currency of the inventory; and increase the collection of accounts in the inventory.

Through the FMFIA process, high risk reviews and other management tools, the IRS is giving priority to management of the accounts receivable inventory. In FY 1994 as a result of our collection efforts, the IRS collected $23.5 billion. This was a 3% increase from the prior fiscal year and was the first increase in collection yield in three years. It was accomplished with a 9.5% decrease in collection staffing. These promising results have continued into FY 1995, and result from a focus on improving the use of existing collection tools coupled with increased productivity of IRS field operations.

During January 1995, using 770 staff years of the FY 1995 Compliance Initiative in Automated Collection System (ACS) sites, the IRS implemented "early interven- tion" by telephone nationwide. Early intervention adopts private sector and state collection "best practices" techniques. Based on experience during the prototypes, the IRS projects that early intervention will result in additional revenues of $3 bil- lion over 5 years. In addition, plans are underway to consolidate the Taxpayer Serv- ice and ACS telephone sites— 70 sites will be consolidated into 23 sites. These sites

316

will all handle collection and accounts calls which will further enhance the use of the telephone contact to collect delinquent accounts.

All delinquent accounts, with a few exceptions, flow through ACS for telephone contact. When ACS does not resolve an account either through full payment, install- ment agreement, lien or levy, personal contact is required before other enforcement techniques, such as offers in compromise and seizures can be instituted. We have emphasized the importance of aggressively using these tools in appropriate cases, such as repeat delinquents. Thus, a field collection process will always be needed to resolve complex financial transactions, larger dollar cases, and repeat delinquents. The IRS' focus is to improve the efficiency and effectiveness of field op- erations, as well as ACS and other telephone operations in collecting delinquent fed- eral tax debt.

A significant improvement in the IRS' ability to collect accounts receivable will be realized through the Accounts Receivable Dollar Inventory (ARDI) expert system and the Inventory Delivery System (IDS). The ARDI expert system will allow the IRS to predict collectibility based on case characteristics. IDS, which will also com- mence operation in FY 1996, ensures cases are routed to the most effective point in the collection processing stream. These systems prioritize work so that it is as- signed to the point of most likely resolution as early in the process as possible.

The Integrated Collection System (ICS) prototype resulted in a 26% increase in revenue officer dollars collected per staff year and a 20-day decrease in the time it takes a revenue officer to close an account. Unfortunately, because of the reduction in funding for Tax Systems Modernization (TSM) in fiscal year 1995, the Service has had to delay the roll-out of ICS from 9 districts to 6 districts. In FY 1996, the IRS will continue to roll-out ICS to another 15-20 districts if funding is available. ICS will significantly enhance the productivity of our field operations, as well as de- crease the time to close accounts. All actions to date have been within the context of trying to obtain better performance from our outmoded technology. The agency expects even greater benefits in the management of the receivables inventory through the TSM program.

9. Question: As you know, I remain very concerned about the financial manage- ment condition of Federal agencies. I was a principal sponsor of the Chief Financial Officers (CFO) Act and the Government Management Reform Act (GMRA), laws de- signed to improve how government handles (and audits) its use of tax dollars. When do you expect the IRS to receive a clean opinion on its audited financial statements? What obstacles remain in the way of a clean audit opinion, and do your auditors — GAO and the Treasury Department's Office of Inspector General — agree with your upbeat assessment?

Answer: The IRS was one of the first three pilots under the CFO Act to prepare financial statements and have them audited by the GAO. The Service knew the process would be incremental, but is well positioned to obtain a clean opinion by the date the Department committed for the IRS' FY 1996 financial statement.

The IRS recently submitted to its auditors the FY 1994 financial statements, which reflect a series of accomplishments that will result in the IRS' meeting its commitment. Some examples include:

• The IRS provided assurance again that the over $1 trillion in revenue col- lected from American taxpayers was properly deposited in the Treasury.

• The IRS was able to extract and balance over 10 billion financial transactions from its master files to the overall general ledger balance so they could suc- cessfully be tested by GAO.

• The IRS implemented on time a single nationwide financial system which en- abled GAO last year to audit 100% of operations — an enormous improvement over the first audit in FY 1992 that was dependent upon old financial sys- tems.

• GAO audited over $5 billion in IRS payroll and found no errors or inappropri- ate payments made.

Furthermore, building on the results of the first two audits, the IRS has con- ducted a detailed analysis to quantify the financial effect of reporting excise taxes based on assessment rather than collection, the difference of which is approximately one tenth of one percent. The IRS has programmed the capability to collect informa- tion on transactions in process at year end, which allows the Service to analyze the effect of these transactions and adjust our financial statements appropriately. And the Service implemented a seized asset inventory system this year for collection- sized assets, which will enable the auditors to perform testing next year on the data.

317

Finally, the IRS will be positioned next year to further detail our master file rec- onciliations to specific general ledger accounts.

GAO has indicated that there are several remaining audit issues needing resolu- tion in the near future —

On the appropriated funds side: (1) IRS' accounts payable information from years

Rrior required reconciliation. IRS has completed this work and GAO has been ana- yzing the information; (2) IRS did not have an adequate seized asset inventory sys- tem in place. IRS has since installed a system that will enable auditors to test in- ventory balances.

On the revenue side: (1) GAO wants the IRS to properly post transactions-in-proc- ess across fiscal years to the proper year. The IRS has completed programming to identify the amounts that are in process and their affect on the General Ledger. The agency may need two full fiscal years of testing (FY 1995 and 1996) to assure audi- tors of the information's reliability; (2) Excise Taxes: the IRS reports amounts based on assessments rather than actual collections as required by law. Based on initial work done in FY 1993, the agency has since analyzed the impact in detail and the discrepancies are immaterial to IRS financial records. In fact, based on a four-year analysis, 99.9 percent of the assessments as reported were collected; (3) Reconcili- ations to General Ledger Accounts: While IRS has been able to balance the masterfile overall through the analysis of more than 10 billion transactions, the agency's goal is still to reconcile individual general ledger accounts. IRS has done considerable work on this issue, but additional testing is still necessary.

10. Question: I understand that there are programs in some states and in Mexico that link the award of contracts by those entities to the awardee's fulfillment of their tax obligations. As I understand it, Federal agencies now have limited quarterly in- formation reporting requirements, but have no mechanism to assess whether the con- tractor has filed or paid his or her taxes. Taking into consideration privacy concerns and the ongoing streamlining of the procurement system, how would you evaluate the merits of a legislative change requiring Federal contractors to demonstrate they are in compliance with their tax obligations before Federal contracts are awarded?

Answer: The concept has merit and deserves attention, but the IRS would reserve judgment until the details of any such legislation are proposed and reviewed.

11. Question: In previous financial audits of IRS, GAO has brought up a variety of issues needing resolution, including problems involving how excise taxes are dis- tributed to their appropriate funds, as well as problems with tracking inventory. Have these issues been settled? How?

Answer: On excise taxes, IRS has historically reported them on the basis of as- sessments rather than actual collections as required by law. In the aftermath of the FY 1993 audit, GAO and the agency were concerned over the magnitude of the pos- sible difference between assessed amounts and actual collections.

For the FY 1994 audit, IRS has done detailed analysis for the past four years and found, as mentioned in response to a previous questions, that 99.9% of assessments as reported were collected (the Highway Trust Fund, for example, in the four years

[receding the audit, more than $63 billion was assessed and 99.9% of it was col- ected). In effect, about one tenth of one percent was not actually collected. The agency believes that for financial statements purposes such a difference will not be judged a material issue.

A similar issue regards Social Security taxes: as required by law, the IRS reports FICA taxes to the SSA based on liabilities reported on the quarterly employer tax returns. SSA is responsible for certifying to the Department of the Treasury that the correct amount to be transferred to the Social Security trust funds.

Again, for the four years preceding the audit, the agency analyzed any remaining unpaid balances for these tax years. The total FICA assessed was over $1,457 tril- lion and the amount of it actually paid was 99.75 percent. Again for the purposes of the integrity of financial statements, the IRS believes such a difference will not be judged a material issue.

o

BOSTON PUBLIC LIBRARY

) 05984 368 8

ISBN 0-16-052749-X

78016CT5274941

90000

■