UNIVERSITY OF

ILLINOIS LIBRARY

AT URBANA-CHAMPAIGN

BOOKSTACKS

CENTRAL CIRCULATION AND BOOKSTACKS

The person borrowing this material is re- sponsible for its renewal or return before the Latest Dote stamped below. You may be charged a minimum fee of $75.00 for each non-returned or lost item.

Theft, mutilation, or defacement of library materials can be causes for student disciplinary action. All materials owned by the University of Illinois Library are the property of the State of Illinois and are protected by Article 16B of Illinois, Criminal Law and Procedure.

TO RENEW, CALL (217) 333-8400. University of Illinois Library at Urbana-Champaign

When renewing by phone, write new due date below previous due date. L162

330

No. cop. 3L

BEBR

FACULTY WORKING PAPER NO. 1288

Value Line Investment Survey Rank Changes and Beta Coefficients

Cheng F. Lee HunY Park

College of Commerce and Business Administration Bureau of Economic and Business Research University of Illinois, Urbana-Champaign

BEBR

FACULTY WORKING PAPER NO. 1288 College of Commerce and Business Administration University of Illinois at Urbana-Champaign September 1986

Value Line Investment Survey Rank Changes and Beta Coefficients

Cheng F. Lee, IBE Distinguished Professor Department of Finance

Hun Y. Park, Assistant Professor Department of Finance

Digitized by the Internet Archive

in 2011 with funding from

University of Illinois Urbana-Champaign

http://www.archive.org/details/valuelineinvestm1288leec

Value Line Investment Survey Rank Changes and Beta Coefficients

Abstract

We use a Value Line rank varying market model to test the existence of a possible systematic association of Value Line ranks with the beta coefficients of securities. The results indicate that about 57 percent of the companies' betas in the sample are associated with Value Line ranks and that these firms are in general small. It is also found that the mean and the volatility of Value Line ranks per se are negatively and positively related to the beta coefficient, respectively.

Value Line Investment Survey Rank Changes and Beta Coefficients

The information content of Value Line Investment Survey rank changes has attracted considerable attention of financial academicians as well as security traders. A number of studies have analyzed the performance of Value Line ranking system. The main conclusion of pre- vious studies, with few exceptions, is that Value Line rank changes have better ability to predict stock price movements than asset pricing models, i.e., several versions of the CAPM. In other words, an investor can generate excess returns even net of transaction costs by following the Value Line rank changes. This, being called the "Value Line enigma," has been used as a typical example against the semi- strong form of the market efficiency hypothesis. If the market is efficient, stock prices instantaneously adjust to reflect all publicly available information including Value Line rank changes and that knowledge of such information cannot lead to excess returns.

The purpose of this paper is to investigate the association of Value Line rank changes with security beta changes, in an attempt to explain how and why the Value Line enigma has been observed. To test the systematic relation between security beta changes and Value Line rank changes, a specification analysis technique is used. The next section describes the model for testing the association of Value Line ranks and the security beta. In the third section, we describe the data and present empirical results. The last section contains a brief conclusion.

■2-

Methodology

The empirical version of the market model to estimate the beta

2 coefficients of securities can be written as

R.to - o. + B.Jl . + e4t_, (1)

jt J jt mt jt'

where R. = the rate of return on security j in period t,

R = the rate of return on market portfolio m in period t, mt

8. = the beta of security j in period t, and

e . = the disturbance term for security j , which is assumed to have mean zero and constant variance.

If the beta of security j is related to Value Line rank changes, we may

specify the beta as a functional form of Value Line rank as:

8. = 8. + Y.V. , (2)

Jt J J Jt

where V. represents Value Line rank of security j in period t.

Then, substituting equation (2) into equation (1), we have a Value Line rank varying market model as

R.„ = a. + 6 .R . + y.(V. *R „) + t. . (3)

Jt j J mt 'j jt mt jt

The variable, V. R , in equation (3) "can be interpreted as an inter- J t mt

action variable reflecting the association of Value Line rank with the time varying beta. If the coefficient, y., is not equal to zero, we may interpret that the market reacts to Value Line ranks and thus the Value Line rank has extra explanatory power for forecasting the beta coefficient and the rate of return of the security. In addition, to obtain the average relation between the Value Line rank and the beta, the following two cross-sectional regressions will be run as:

-3-

M . = a + b6 . (4)

a , = a' + b'B . , (4)'

vj i'

where M . and a . represent the mean value and the standard derivation Vj vj

of Value Line ranks of security j, respectively.

Data and Empirical Results

Weekly ranks of all securities of Value Line were secured for the period July 1978-February 1983. Five ranks are provided by Value Line depending on the expected price performance over the next 12 months. Ranks 1 and 5 represent the best and the worst securities, respectively. Excluding the firms not included in the CRSP monthly files, we obtained 1331 companies. Monthly rate of returns on the individual securities and the value weighted NYSE index are used to estimate the coefficients of equation (3). For V. in equation (3), monthly average of weekly Value Line ranks are used.

Through examination of t-statistics of the coefficients in eq. (3), we find that 189 firms have y. significantly different from zero at the 5 percent level. The names of these companies are listed in Appendix A. To save space, the empirical results of only the first 32

companies in alphabetical order are listed in Table 1 for an exhibition

3 purpose.

For example, the American International Company's beta can be

decomposed into two components — the constant component, 3.477, and the

Value Line rank related component, -.583. In other words, the re-

sonsive coefficient of beta to Value Line rank is -.583 for American

-4-

International Company, and thus one percent increase in Value Line rank causes .583 percent decrease in the company's beta.

Insert Table 1 about here

In addition, we also find that 567 firms have t-statistics for Y. coefficients larger than one. In statistical sense, this number of companies, 756 (567 plus 189), certainly implies that the market per- ceives Value Line ranks as an important source of information in pricing securities. It is interesting also to note that most of these 756 companies are small in terms of size. Therefore, it appears that the smaller the size of the firm, the greater the impact of Value Line ranks on the determination of the beta. More importantly, most of y. coefficients are negative (even in other companies not reported here) , suggesting that the Value Line rank is negatively related to the rate of return. The lower the rank, the better the projected performance of the security and thus the higher return (note that rank 1 represents the security which is projected to perform best). This result is con- sistent with the findings in previous studies on the performance of Value Line. However, this paper shows that the result may be through the association of Value Line ranks with the beta. This is confirmed by examining the coefficients of equation (4).

The results on the cross-sectional regressions in (4) are shown in Table 2. The beta in Table 2 was estimated using monthly rate of returns on individual securities and the NYSE index, based on equation

(1). M . and a . were calculated using weekly Value Line ranks. As

vj vj * '

expected from negative coefficients of y . , in general, in Table 1, the b

-5-

coefficient in Table 2 is significantly negative. More interestingly, the results in Table 2 also suggest that the volatility of Value Line ranks per se is positively related to the beta. The b' coefficient in equation (4)' is .1490, which is significant at the 1 percent level. Since the beta is a measure of the volatility of a security relative to the market and the Value Line rank is a relative measure of projected performance of individual security, it is not surprising that the beta is positively associated with the volatility of the Value Line rank.

Conclusion

A number of previous studies have shown outstanding performance of Value Line ranking system. We use a Value Line rank varying market model to test the existence of a possible systematic association of Value Line ranks with the beta coefficients of securities. Using weekly ranks of 1331 companies for July 1978-February 1983, we find that about 57 percent of the companies' betas are associated with Value Line ranks and that these firms are in general small. This finding provides an insight into how Value Line rank changes affect the indivi- dual firm's stock price. It is also found that the mean and the vola- tility of Value Line ranks per se are negatively and positively related to the beta coefficient, respectively.

-6-

Footnotes

See, for references, F. Black, "Yes, Here is Hope: Tests of Value Line Ranking System," Financial Analyst Journal 29 (1973), pp. 10-14; T. E. Copeland and D. Mayers, "The Value Line Enigma (1965- 1978): A case Study of Performance Evaluation Issues," Journal of Financial Economics 10 (1982), pp. 289-321; C. Holloway, "A Note on Testing an Aggressive Investment Strategy Using Value Line Ranks," Journal of Finance 36 (1981), pp. 711-719; S. Stickel, "The Effect of Value Line Investment Survey Rank Changes on Common Stock Prices," Journal of Financial Economics 14 (1985), pp. 121-143.

2 See E. Fama, Foundations of Finance," Basic Books, New York

(1976).

3 The results of all other firms are available from the authors

upon request.

D/415

Table 1 Value Line Rank Varying Market Model

R. =a.+8.R +Y.(V R )+e jt j .1 m,t jjCmt jt

Corporation 8 v _1 J.

1. Am Int'l 3.477 -.583

(5.320) (-3.144)

2. Amr Corp 2.668 -.385

(5.526) (-2.176)

3. Aetna Life & Casualty 1.846 -.250

(7.124) (-2.907)

4. Albertson's, Inc. 1.501 -.319

(5.610) (-3.000)

5. Alcan Aluminum 1.964 -.290

(5.563) (-2.398)

6. Allegheny Int'l 2.832 -.457

(6.106) (-3.257)

7. Amerace Corp. 2.562 -.563

(3.722) (-2.779)

8. Amer. Broadcasting 2.025 -.396

(5.256) (-3.164)

9. Amer. Hoist Derrick 2.438 -.364

(3.611) (-1.935)

10. Amfac Inc. . -.418 .502

(-.757) ( 2.895)

11. Amrep Corp. -1.735 1.100

(-1.092) ( 2.188)

12. Anchor Hocking Corp. -.745 .371

(-1.386) ( 2.426)

13. Avon Products 1.535 -.235

(5.743) (-2.551)

14. Ball Corp. 1.632 -.347

(3.506) (-2.163)

Table 1 (cont'd.)

Corporation

B. _1

15. Bandag, Inc.

16. Bk America

17. Baxter Travenol Labs

18. Best Products

19. Boeing Company

20. Braniff Int'l Corp.

21. British Petroleum

22. Brooklyn Union Gas

23. Burroughs Corp.

24. CCI Corp.

25. Caesors World

26. Campbell Red Lake

27. Central Soye Co.

28. Champion Int'l

29. Chasebrough-Ponds

-.339 .546

(-.594) ( 2.547)

1.563 -.243

(5.225) (-2.190)

.390 .334

(1.441) ( 2.296)

2.293 -.338

(6.199) (-2.569)

2.216 -.305

(6.652) (-2.406)

2.508 -.369

(5.805) (-3.227)

-.163 .293

(-.436) ( 2.512)

1.513 -.412

(2.653) (-2.259)

1.729 -.236

(6.360) (-2.795)

.236 .493

( .346) ( 2.009)

4.354 -1.124

(2.753) (-2.217)

1.451 -.433

(3.199) (-2.520)

3.234 -.627

(3.956) (-2.765)

2.034 -.312

(4.168) (-2.016)

.221 .217

( .870) ( 2.079)

Table 1 (cont'd.)

Corporation 6 . y .

30. Cities Service 2.107 -.273

(5.010) (-2.035)

31. City Investing 2.111 -.348

(3.796) (-1.936)

32. Coleman Co. Inc. 2.032 -.431

(4.502) (-2.811)

The numbers in parentheses represent t-statistics.

Table 2

Cross-Sectional Regressions Between Value Line Rankings and the Betas of 1331 Companies*

M = a + bS . vj J

vj

a' + b'S . J

a	b	R2	D-W	a'	b1	_2 R~	D-W
3.251 (109.21)	-.2391 (-8.72)	.054	2.02	.6296 (43.07)	.1490 (11.06)	.084	2.01

*The numbers in parentheses represent t-statistics.

D-W stands for Durbin-Watson statistics,

-2 2

R represents the adjusted R .

Appendix A The List of 189 Firms with Significant y Coefficient in Equation (3)

Am Int'l Albertson's Inc. Amerace Corp. Amfac, Inc. Avon Products Bankamerica Corp. Boeing Company Brooklyn Union Gas Caesars World Champion Int'l Corp. City Investing Co. Cominco Ltd. Cox Communications Dana Corporation Deltona Corp. Disney (Walt) Prod. El Paso Co. Equitable Life Mortg. Federal-Mogul First Mississippi Fuqua Ind. General Cinema Giant Portland Cement Gulton Ind Hewlett-Packard Co. Hoover Universal Inc. Ingersoll-Rand Co. Int. Reetifier Corp. Jamesway Corp. Kdt Ints

Kyocera Corp. (ADR) Leverage Fund Boston MCA Inc.

Manhattan Industries Masco Corp. Miller-Wohl National Gypsum Niagra Shave Corp. Nortek Inc. Olin Corp.

Overseas Shipholding Pennzoil Co. Petroleum & Res. Corp. Pioneer Corp Polaroid Corp.

Amr Corp. Alcan Aluminum Amer. Broadcasting Amrep Corp Ball Corp.

Baxter Travenol Labs Braniff Int'l. Corp. Burroughs Corp. Campbell Red Lake Chesebrough Ponds Coleman Co. Inc. Computer Sciences Crane Co.

Data General Corp. Dennison Mfg. Diversified Ind. Electronic Assoc Evans Products Corp Figgie Int'l First Natl St Bancor Gatx Corp Gerber Products Giddings & Lewis Handleman Hillenbrand Inds. Hunt (Phil A) Chem Interco, Inc. Interstate Baker Jewel Companies Kansas City Southern La Quinta Motor Inns Libby Owens Ford Macmillan Inc. Map Co. Inc. Mclntyre Mines Ltd. Monarch Machine Tool New York Times Nicor, Inc. Northwest Airlines Oneok Inc.

Pacific Gas & Electric Peoples Drug Store Philip Morris Pitney-Bowes Ponderosa, Inc.

Aetna Life & Casualty Allegheny Int'l Amer. Hoist & Derrick Anchor Hocking Corp. Bandag, Inc. Best Products British Petroleum ECI Corp. Central Soya Co. Cities Service Colonial Penn Group Copperweld Corp. Cross (A.T.) Deere & Co. Dial Corp.

A G Edwards and Sons Empire Distric Elec Far West Financial Fin'l Santa Barbara Fruehauf Corp Gemini Fund Getty Oil Gleason Works Harsco Corp. Hilton Hotels Imperial Oil Ltd. "A" Int'l Flav & Frag Interstate Power Johnson & Johnson Kennametal Inc. Lennar Corp. Lockheed Corp. Madison Fund Maryland & Cup Corp. Metromedia, Inc. NVF Co.

Newmont Mining Norlin Corp Northwest Energy Opelika Mfg. Parkers Pen Co. Peoples Energy Corp Piedmont Nat. Gas Pneumo Corp. Presly, Cos.

Appendix A (cont'd.)

Onanex Corp. Revere Copper & Brass Rockwell Int'l SPS Technologies Scoa Ind. Sears, Roebuck Smith (A.O. ) Corp Southern Union Co. Suave Shoe Corp. Swank Inc. Talley Ind. Texfi Industries Travelers Corp. United Brands U.S. Shoe Corp. Vista Resources Inc. Washington Gas Light Western Pacific Ind.

RCA Corp. Revlon, Inc. Ronson Corp. Sabine Corp Scott & Fetzer Co. Shell Transport South Jersey Ind. Standard Oil (Ind. ) Sunbeam Corp Teco Energy Inc. Tesoro Petroleum Tiger Int'l Inc. Tri-South Invert Inc. U.S. Gypsum Co. Univar Wainoco Oil Washington Nat'l. Cp. Wheeling-Pitts burgh

Reeding & Bates Rexham Corp. Ryan Homes Santa Fe Industries Scottys Inc. Simmonds Prec. Prod's Southern Pacific Stop & Shop Cos Superior Oil Taft Broadcasting Texaco, Inc. Tokheim Corp. Union Oil Co. Calif. U.S. Industries Valley Nat'l Corp Warnaco Inc. Westwart Transm'n

HECKMAN I— I

BINDERY INC. |§|

JUN95

„,,,„, T„ ,,, '; N. MANCHESTER.

Bound .T»-Plum |ND|ANA46g62 V ■ '