R T I S T S
in the Work Force:
Employment and Earnings,
Research Report #37
Neil O. Alper and Gregory H. Wassail
Joan Jeffri and Robert Greenblatt
Ann O. Kay and Stephyn G. W. Butcher
Harry Hillman Chartrand
FOR THE ARTS
Digitized by the Internet Archive
in 2012 with funding from
Boston Library Consortium Member Libraries
Artists in the Work Force:
Employment and Earnings,
1970 to 1990
Artists in the Work Force:
Employment and Earnings,
1970 to 1990
Neil O. Alper and Gregory H. Wassail
Joan Jeffri and Robert Greenblatt
Ann O. Kay and Stephyn G. W. Butcher
Harry Hillman Chartrand
Research Division Report #37
National Endowment for the Arts
Seven Locks Press
Santa Ana, California
Artists in the Work Force: Employment and Earnings, 1970 to 1990 is Report #37 in
a series on matters of interest to the arts community commissioned by the Research
Division of the National Endowment for the Arts.
Cover photos: (L to R) Novelist Alice Walker (photo by Jean Weisinger); participants
in the Mayors' Institute on City Design; Victoria Finlayson and Alan Good of the
Merce Cunningham Dance Company (photo by Michael O'Neill); Glassblowing at
the Rhode Island School of Design (photo by David O'Connor).
First Printed 1996
Library of Congress Cataloging-in-Publication Data
Artists in the work force: employment and earnings, 1970 to 1990/
Neil O. Alper...[et al.].
p. cm. — (Research Division report: #37)
1. Artists — Employment — United States. 2. Artists — Salaries, etc., — United
States. I. Alper, Neil, 1949- . II. Series: Research Division report (National
Endowment for the Arts. Research Division); 37.
Manufactured in the United States of America
Seven Locks Press
Santa Ana, California
Table of Contents
Artists Who Work with Their Hands 3
Performing Artists 4
Architecture and Design Arts Occupations 6
I. The Write Stuff: Employment and
Earnings of Authors, 1 970-1 990 13
Profile of American Authors 14
Evidence on Author's Multiple Job Holding 1 6
Evidence on Authors from Non-Census Sources 17
Authors in the U.S. Census 18
Census Data Used to Analyze Authors'
Work Experiences 19
Issues in Using Census Data to Analyze
Authors' Labor Force 19
Whom Does the Census Define as an Author? 20
Growth in the Author Profession and
Where Writers Work 22
For Whom Authors Work: Distribution by Industry 24
Demographic Characteristics of Authors 25
Authors in the Labor Market 28
Earnings of Authors 32
Authors in the U.S. and Elsewhere 40
Current Population Survey 41
Non-Census Surveys 43
Alper- Wassail Survey 44
Columbia University's RCAC Survey 48
Special Artist Surveys 49
Administrative Records 52
Writers in Other Countries 55
About the Authors 58
vi I Artists in the Workforce
II. Artists Who Work with Their Hands
A Trend Report, 1 970 to 1 990 59
Data from the United States Census 60
Discrete Surveys 61
Other United States Data 65
United States Census Data, 1970-1990 65
Geographic Trends 66
Employment and Earnings 69
Discrete Surveys 72
The Artist's Work-Related, Human and Social Services 72
Information on Artists Survey (1988) 73
Information on Artists: Boston and New York 78
Artists Training and Career Project (ATC) 1990-91 79
Artists and Jobs Questionnaire 1980 81
About the Authors 84
III. Employment and Earnings of Performing
Artists, 1970 to 1990 85
Data Sources and Conceptual Issues 86
Performing Artist Surveys 87
Union Pension Records 87
Employment and Earnings Analysis 88
Labor Force Status 88
Class of Worker 92
Industry of Employment 93
Weeks of Work and Usual Weekly Hours 94
Geographic Distribution 106
About the Authors 111
Table of Contents I vii
IV. Architecture and Design Arts Occupations,
1970 to 1990 112
Comparative Occupational Groups 114
Landscape Architects 115
Ethnicity and Race 117
Employment 1 23
Ethnicity and Race 130
Professionalization and Competition 139
Design Deficit 140
Design Rights 140
Aesthetic Utopians 14 1
Forecasts 1 4 1
About the Author 142
The focus of this report commissioned by the National Endowment for the Arts
is on four distinct groups of artists: authors, architects and designers, performing
artists and artists who work with their hands. The report attempts to answer certain
basic questions about where artists live and work and what they earn. Do they
support their art or does it support them? How well did the growth and changes in
art occupations over the two decades covered by the report keep pace with those of
professionals in other fields with comparable education and training? The report also
examines trends regarding the artists' economic conditions, age, gender, race,
ethnicity, education, area of residence, employment sector, earnings and multiple job
holdings. An underlying assumption throughout the report is that the intrinsic
rewards for artists in doing the work they most want to do cannot be quantified.
The authors of the four sections of the report represent different disciplines in
art, education, and economics, but they share a general concern for the art world and
its career participants. They worked independently, focusing on their specialized
fields and studying existing data bases, including those of the U. S. Census, various
surveys of artists and industry records. Although they were often researching similar
information on artist occupations, the authors approached their studies from
different perspectives, using various analytical techniques. The result of their efforts
is a comprehensive, revealing look at issues affecting the entire set of artist
occupations covered by the Census from 1970 to 1990.
The authors generally agree on the advantages and limitations of using data from
the broad-based Census and the more limited, but closely targeted, surveys aimed at
specific artist occupations. They also concur that continuing research will lead to a
better understanding of the opportunities and challenges today's artists experience.
Because these authors can best highlight the findings of their individual studies,
their summaries of these findings appear here. It is hoped that by combining these
findings in one area, both the similarities and differences of employment experiences
in the four art fields will be apparent.
Employment and Earnings of Authors
Neil O. Alper and Gregory H. Wassail researched and wrote the section of this
report on authors. They found that writing occupations have grown rapidly the last
two decades, but that writers still constitute a very small fraction of the nation's labor
force — in 1990 American authors numbered about 107,000 or just nine in 10,000
members of the labor force. Alper and Wassail used the broadest definition of
authors: all persons who indicated occupational membership, regardless of being in
the labor force at Census time. The researchers compared the employment and
earnings of authors to all artists, editors and reporters, technical writers, and all
professional and technical workers other than artists.
Briefly, here are some of their major findings from U.S. Census data:
Artists in the Work Force
• Education. Authors were found to be well educated, averaging 15.8 years of
education in 1990, more than other artists and professional/technical workers.
• Age. Authors averaged 44 years of age, which was four to five years older than
reference groups used (other professional workers and all artists).
• Residence. California and New York are home to more authors than other states,
with the Los Angeles and New York City metropolitan areas claiming the highest
percentages of writers in their labor forces.
• Gender and ethnicity. Half of all authors were men in 1990 and 95 percent were
white. Over the 1970-1990 period the author profession had a lower percentage
of minorities than artists or other professional/technical workers. It had a higher
percentage of women than did the artist profession, but a lower percentage than
other professional/technical occupations.
• Unemployment. In 1990 only 40 percent of authors worked full time year-
round, compared to 46 percent of artists and 56 percent of other professional
and technical workers who worked full-time.
• Employment sector. Compared to the average artist and professional or
technical workers, authors were far more likely to be self-employed and less likely
to work for private sector employers.
• Earnings. In 1989 authors' total earnings averaged $23,335, less than other
professional workers, editors and reporters and technical writers, but more than
other artists. Authors' total personal income was $30,089 and their average total
household income, $62,083 (leading all reference groups in household income).
Between 1969 and 1989, the earnings of authors grew by 175 percent, which
lagged behind the reference groups and behind inflation. From 1969 through
1989, a higher percentage of authors had zero or negative earnings, and a higher
percentage earned over $90,000 than any of the reference groups.
In 1989 women authors earned only 52 percent of male writers' earnings.
Black authors earned 91 percent of white; Hispanics earned 87 percent.
Several studies of artists from 1979 to 1989 found that 70 to 90 percent of
authors worked at other jobs during the year, often simultaneously with their writing.
The majority of second jobs for authors were professional in their nature. Their
writing-related jobs were primarily teaching. Without income from their second or
multiple jobs, a majority of authors would be classified as poor.
Non-Census sources, such as artist surveys and union records, confirm many of
the Census findings, including the facts that authors are primarily white with an
average age around 40 and are well-educated compared to the general population. As
to gender, some of the studies showed an even distribution of males and females,
some 60 percent female, and union records accounted for only 22 percent female.
Regarding income, non-Census surveys showed female authors consistently
earned less than males, with different studies showing a range from about 40 percent
of male earnings to about 75 percent by 1991. The income of other family members
was found to be important in aiding writers to work at their craft and to stay well
above poverty thresholds.
The Columbia Authors' Guild study and the Writers' Guild studies found
earnings to vary greatly related to the type of writing and the employer. Generally,
poets earned the least from their craft, authors of academically oriented non-fiction
somewhat more, writers of children's books and general adult fiction were in the
middle, and writers of "genre" fiction earned the most. Writers in television earned
about 3 percent more than film writers in 1991. Those who worked for major TV or
film production companies earned more than those working for smaller production
companies, with those working for the major film production companies in 1991
having median earnings 115 percent higher than writers working for independent
Artists Who Work with Their Hands
Joan Jeffri and Robert Greenblatt wrote the section of this report which examines
trends in the visual arts occupations of painters, sculptors, craft artists and artist
printmakers. Major areas they examined, using Census data and artist population
surveys, are geographic distribution, age, education, employment and earnings.
Highlights of their findings appear here.
Between 1970 and 1990 the total artist population more than doubled, from
720,000 to 1,671,000. By 1990, painters and craft artists totaled 191,160, or 13
percent of all artists, the second largest of all artist occupations. Female painters/craft
artists by 1990 numbered 107,920, or 56 percent. For male painters/craft artists, the
rate of growth declined from 64 percent in 1970 to 44 percent in 1990.
While about three-fourths of these artists are urban residents, these percentages
have declined from 1980. Male painters and craft artists had their highest
proportions in the West in 1990, while their female counterparts moved South and
substantially decreased in the Northwest.
Median ages for male painters/craftspeople were higher at 40 than for all male
artists at 37 and higher than the general labor force. Female median ages of 39 years
were higher than for all female artists at 37, but similar to the general labor force.
Both male and female painters/craftspeople had median ages of 41 in a Research
Center for Arts and Culture study done at the time.
The area of education seems to prove the most difficult when comparing Census
figures to discrete survey figures. According to the Census, education for both male
and female painters and craft artists is just holding steady at the 4-plus years of
college level and above, and both years of graduate education and degrees are suspect
due to changes in the coding procedures of the Census. According to the findings of
the discrete surveys used here, over 40 percent of the painters/craft artists have
Self-employment rose for both male and female painters and craft artists from 32
percent of males and 34 percent of females in 1970 to 47 percent of males and 49
percent of females in 1990. For females, as self-employment increased,
unemployment declined, but the definition of self-employment for visual artists is a
highly complex one.
Females continued to earn less than males in all sectors between 1970 and 1990,
even though between 1980 and 1990, the female median income for professionals
more than doubled to $23,113, and the median income for female painters/craft
Artists in the Work Force
artists more than tripled to $22,041. For female professionals who worked 50 to 52
weeks, median income was $29,181. The median income for full-time female
painters/craft artists of $18,762 was lower than that of their artist counterparts,
whose median income was $20,825 in 1990.
In 1990 authors Jeffri and Greenblatt found that Census data showed a larger
proportion of women as painters, sculptors, craft artists and artist printmakers, with
fewer living in urban areas. They had a higher median income than all artists and the
general labor force, but were closer in median age to professionals. For both males
and females, the level of education, according to the Census, seemed to be just
holding steady or rising slightly at the higher levels. More of these artists were self-
employed, with percentages much higher than other kinds of artists. The median
income for male painters and craft artists grew more slowly than for the total work
force, male professionals and female painters/craft artists, whose median income
tripled since 1980. Finally, it appeared to the authors that part-year female
painters/craft artists earned more than their full-year counterparts.
The discrete surveys Jeffri and Greenblatt studied offered another view, one
which targets the artist population more narrowly than the Census, and which
suggests additional ways of looking at how artists view their occupations. These
surveys also identify other areas of inquiry that broaden the picture of the artist in
society. Neither the Census nor the discrete surveys purport to provide a longitudinal
data base, a problem which the authors felt the National Endowment for the Arts
should perhaps try to address for artists in the future. Finally, the authors concluded
that research indicates the need for a regular survey of artists, if possible, by the NEA,
which combines the more relevant aspects of the Census with other areas of inquiry,
some of which they have identified in their report.
Ann O. Kay and Stephyn G. W. Butcher found that the performing arts have
become an increasingly important part of American life in recent decades, at least as
measured by their growing presence. Between 1965 and 1992, the number of
professional dance companies in this country leaped from 37 to more than 250, and
their audience from 1 million to over 16 million. Professional nonprofit theatres
expanded dramatically, with 1992's count of more than 400 theatres representing
over seven times the number of theatres in 1965. Although Broadway and road shows
employ proportionately fewer professional actors, resident theatres and other forms
of professional theatres are expanding. As a percentage of total gross national
product, the motion picture industry grew by one-third between 1969 and 1989,
amusement and recreation services grew by half, and radio and television
broadcasting increased by half. Meanwhile, the number of professional orchestras
doubled, and opera companies grew fourfold since the mid-1960s.
From 1970 to 1990, performing artists as a group grew 50 percent faster than
the national labor force and slightly faster than other professional occupations. The
dancers and actors-directors occupations were among the fastest growing of all arts
occupations. By contrast, numbers for musicians-composers leveled off during the
Union survey data have shown musicians are more likely than other performers
to rely on income from jobs outside their arts profession. In 1980 only 26 percent of
musicians received all of their income from performing arts work, as against 32
percent of actors and 62 percent of dancers.
Throughout the last two decades many performing artists could not find work
in their profession. In 1990 when the national unemployment rate was 5.3 percent,
the rate for actors and directors was 13 percent; for dancers, 7 percent, and for
musicians and composers, 6 percent. During the 1970-90 period, performers'
unemployment rates consistently exceeded the national average, but each performing
arts occupation experienced its own unemployment trends. The rate for musicians
and composers rose and fell over the 1970-90 period, mirroring the national
unemployment trends. For actors and directors, unemployment held at
approximately twice the national rate in 1980 and 1990. For dancers, however,
unemployment moved down from nearly three times the national rate in 1970 to
within two points of the national rate in 1990.
Though musicians and composers had somewhat lower rates of unemployment
than other performers, those who were unemployed during 1990 were out of work
longer than other performing artists. About one out of eight unemployed musicians
and composers had been without a job for two to five years in 1990, about three
times the rate for actors and directors and twice the rate for dancers. For all
performing artists, though, long term unemployment continued to hamper the full
utilization of this segment of the work force.
Survey findings from 1980 suggest that the labor reserve is further increased by
the discouraged worker phenomenon; that is, by performing artists who leave the
labor force because they believe no jobs are available. At least 5 percent of actors and
musicians fit this category that year, with fewer dancers doing so.
Performers cycle more often between employment and unemployment than
other workers. More than three out of five actors who had any unemployment in
1980 were out of work three or more periods, as were more than one out of two
musicians and nearly one in three dancers.
Dancers were the performers least likely to work a full year. For those who
worked anytime in 1989, their median number of weeks worked was 39, compared
to 50 weeks for actors and directors and 48 for musicians and composers. The latter
were most likely to report self-employment.
The principal industry employing performing artists continued to be the
"theatres and motion pictures" industry, although during the 1980s the number of
musicians in this industry declined. The industry in which actors showed the greatest
growth (a 50-fold jump) during these two decades was in "radio and television
broadcasting and cable."
Performers as a group increased their income levels over the 1970-90 period, but
the gains were not equally shared. Actors and directors' median earnings in 1989 were
$22,000 according to Census data, up from about $12,600 ten years earlier. Census
Artists in the Work Force
data also showed one-fourth of actors-directors earning less than $10,000 in 1989. In
contrast, union pension records, which look only at earnings in the acting profession,
showed nearly nine out often actors receiving less than $10,000 in 1990. This suggests
the large extent to which actors must supplement their arts income with outside jobs.
Dancers' earnings tended to be the lowest in the performing arts. Although their
median earnings grew from about $5,400 to $8,500 between 1979 and 1989, nearly
a third of dancers earned less than $5,000 in 1989. Their earnings did not keep up
with inflation during the 1980s. Meanwhile, musicians and composers saw their
median earnings rise from $5,600 to $9,900 between 1979 and 1989, and the
proportion earning less than $10,000 dropped by nearly a third.
Personal earnings levels from acting were extremely volatile, as union pension
data show. One out of five actors who had paid work in their profession in 1990 did
not have any acting income two years later. Half of those who had earnings received
less than they did two years before. A few went from zero to $60,000 or more in two
Earnings levels depended in part on the type of work performers did. Television
work paid better for the members of one actors union, while theatrical engagements
were most rewarding for union musicians.
Performing artists tended to cluster in the West and Northeast regions of the U.
S., particularly in the New York and Los Angeles areas. Although they lived there to
be close to the historic centers of performing arts employment, these areas also had
the highest rates of unemployment for performers. As the number of performers grew
during the last two decades, so did their geographic distribution. The relative increase
in the performing artist work force in the South has been a significant geographic
trend in these 20 years.
Architecture and Design Arts Occupations
Harry Hillman Chartrand examined trends in architecture and design
occupations for 1970-1990. His findings are summarized here.
First, architecture and design occupations grew significantly over the two decades
covered by this report, representing 0.6 percent of the experienced labor force, 4.6
percent of all professional specialty workers and 45.2 percent of the arts labor force
Second, architects, decorators and designers 25 to 44 years of age dominate their
professions, accounting for more than 60 percent of those in the fields.
Third, both architecture and design were predominantly white professions (90.5
percent for both) in 1990. However, Hispanics grew from 1.8 percent of all architects
in 1970 to 5.1 percent in 1990. During the same period, Afro-American architects
increased from 2.4 percent of all architects to 2.8 percent and non-white and non-
black architects, largely Asians, grew from 3 percent to 6.7 percent between 1970
Among decorators and designers, Hispanics grew from 2.9 percent of the
profession In 1970 to 5.4 percent in 1990. Afro-Americans grew from 1.9 percent
in 1970 to 3.6 percent in 1990. Non-white and non-black decorators and designers
grew from 2.1 percent in 1970 to 6.2 percent in 1990.
Fourth, by 1990 over half of all architects, designers and decorators lived in the
South or West, with each of these regions home to approximately one-quarter of the
groups. The Northeast and Midwest have lost their historic dominance of these
Fifth, while architecture remained a predominantly male profession in 1990,
women architects represented 17.7 percent of all architects then. By contrast, women
decorators represented 73.3 percent and designers 43.5 percent in their fields in
Sixth, among architects, those with four or more years of college were 80.2
percent of the profession in 1990, increasing from 73.3 percent in 1970. Among all
decorators and designers, 35.8 percent had four or more years of college in 1990,
increasing from 21.6 percent in 1970.
Seventh, in 1970 self-employed architects represented 36.7 percent of the
profession. Self-employment decreased to 32.8 percent by 1990. Meanwhile, self-
employment among all decorators and designers increased from 15.9 percent in 1970
to 24.6 percent in 1990. Among architects, 25 percent worked part-time in 1990,
compared to 42.8 percent of decorators and designers. The high level of part-timers
among decorators and designers reflects a majority of women in these professions.
Over 75 percent of architects worked in professional services industries, i.e., in
architectural firms. Over 75 percent of decorators and designers worked in
manufacturing, retail trade or business and repairs service industries.
Eighth, in 1987 there were nearly 18,000 architectural establishments with
receipts of almost $10 billion and 137,000 paid employees. Over 70 percent of
receipts came from projects involving commercial buildings or public and
institutional facilities. Nearly 78 percent of receipts were for architectural services
excluding landscape architecture and 75 percent of receipts were from industrial,
business, and commercial companies, government and private institutions.
In 1987 there were 7,202 graphic arts establishments with receipts of $3.2
billion. The Northeast accounted for 34 percent of receipts; the South, 17.5 percent;
the Midwest, 25.5 percent and the West, 22.7 percent.
Ninth, the median annual household income of architects in 1989 was $56,773,
which was 140 percent of the labor force average, 109 percent of the typical
professional specialty worker and 128 percent of the typical artist. The median
annual income of a decorator and designer was $45,873, or 1 13 percent of the labor
force average, 88.2 percent of the typical professional specialty worker and 103
percent of the typical artist.
To understand how artists fared in the work force from 1970 to 1990 it is useful
to contrast their employment experiences with those of other professionals with
comparable education and training. This introduction will attempt to do this,
identifying the labor force terms which appear throughout the four sections of this
report and the areas of common ground covered by the various authors. It synthesizes
to some extent key points from the sections which pertain to the overall report —
points which individual authors explored thoroughly and which will be extracted to
Any study of persons in the labor force must naturally define exactly who is to
be included. For purposes of their analyses, the authors generally refer to the labor
groups defined below. They also discuss in their individual sections various aspects of
these groups as they relate most directly to the specific art fields being discussed.
• The civilian labor force includes all persons 16 years and older who are currently
working (the employed), plus those not working but looking for work (the
unemployed). Persons currently not working and not looking for work are
described as out of the labor force. Working persons in the military are obviously
• The experienced civilian labor force (ECLF) consists of all employed and
unemployed persons with recent civilian work experience (again excluding those
in the military). This more narrowly defined term has been used in recent studies
of artists commissioned by the NEA and will often appear in this report.
• Professional specialty workers (PSW) includes artists, athletes, astronomers,
dentists, engineers, lawyers, miners, nurses, physicians, physicists, optometrists,
reporters, social workers, among others.
• All artists include actors and directors, announcers, architects, authors, dancers,
designers, musicians and composers, painters, sculptors, craft-artists and artist
printmakers, photographers, teachers of art, drama and music in higher
education, and also artists and related workers not elsewhere classified.
Growth Rates in the Labor Force, 1970-1990
The experienced civilian labor force numbered 79,801,605 according to the
1970 U.S. Census of Population and Housing. About one in 100 of those, or
736,960, were artists. By 1980, the experienced labor force had increased by 30.4
percent to 104,057,985. Authors Ann Kay and Stephyn Butcher attribute this high
growth rate to a continuing move into the labor force by the baby boom generation
and an increased participation of women in the labor force. The number of artists
jumped also, growing an average of 4.0 percent per year from 1970 to 1980, or 47.3
percent over the decade. This boom in artists' numbers occurred for the same reasons
as the national labor force, plus a trend toward the increasing importance of
professional specialty occupations and the service sector in the U.S. economy — a
trend that continued in the 1980s.
By 1990, the experienced labor force had grown to 122,473,499 and the artist
labor force to 1,671,277 (see Tables 1 and 2 in the Performing Artists section).
Professional specialty occupations also grew dramatically over the 1970-90 period,
increasing from 8,800,210 workers in 1970 to 12,275,140 in 1980 and to
16,647,688 in 1990. The growth rate among artists during the 1980s exceeded that
The U.S. labor force during the 1970 to 1990 period became better educated,
more ethnically diverse and comprised of a higher percentage of women. Other
general demographic trends among all the occupational groups discussed in this
report include lower rates of marriage and household formation and a decline in the
percentage of veterans.
How the Census Classifies Artists
The authors of the report explained that all of the employment and earnings
information available from the Census flows from the occupation in which the
worker was classified. The work history (labor force status, weekly hours, weeks
worked, income) becomes attached to that occupation.
The Census asks six basic questions about current or most recent job activity
with instructions to describe the person's "chief job activity or business last week."
Respondents who had more than one job were instructed to describe the one at
which they worked the most hours. If they had no job last week, they should refer to
their last job or business since 1985.
The six questions center on:
• Industry or Employer — employer, kind of business
• Occupation — kind of work, most important activities
• Organizational Sector — private nonprofit, government, self-employed, working
The basic way artists (and others) get fitted into a category is by answering the
requisite six questions, and Census bureau employees deciding into which category
they fit based on a classification listing.
Several authors pointed out concerns over using Census data to analyze labor
market outcomes of artists revolve around the manner in which occupations are
defined. A person's choice of occupation is determined by his or her response to a
request for information on work activity "last week." Further, occupation is self-
defined; i.e., the respondent names the occupation, rather than picks the best
available choice from a list. The Census then assigns the respondent's self-defined
occupation to an existing classification (or creates a new one if warranted). Thus an
occupational classification consists of a number of individual but related
10 I Artists in the Work Force
The Census procedure forces the respondent to select one occupation only.
Those who worked at more than one job during the reference week are required to
choose only one — that at which the most time was spent during the week.
Authors Neil Alper and Gregory Wassail state that the use of Census data to
analyze labor supply, incomes and other labor market characteristics of artists has
been challenged by some social scientists with two principal criticisms: (1) that the
Census defines the term "artist" too broadly, and (2) that the Census mis-classifies or
ignores some artists because of frequent multiple job-holding in most artistic
The Census does not actually define the term "artist." Rather, the 1980 and 1990
Census creates six broad occupational classifications including one called
"Managerial and Professional Specialty Occupations" that contains all 1 1 Census
artist occupations. The 1 1 categories are the product of the NEA, and are based on
which occupations have traditionally comprised its artistic constituency. They are: (1)
actors and directors, (2) announcers, (3) authors, (4) dancers, (5) designers, (6)
musicians and composers, (7) painters, sculptors, craft-artists and artist printmakers,
(8) photographers, (9) artists, performers, and related workers, not elsewhere
classified, (10) architects (found under the category "Engineers, Architects, and
Surveyors"), and (11) college and university art, drama, and music teachers (found
under "Teachers, Postsecondary.")
While some of the 1 1 categories may indeed be too broad (the artists not
elsewhere classified category and dancers), others such as architects and radio and
television announcer categories are sufficiently narrow.
A second limitation in using Census data discussed by several authors lies in its
treatment of multiple job-holding. The Census questions force a respondent to
choose only one occupation and the respondent's labor market experiences for the
previous calendar year are then attributed to that occupation. While this procedure
is reasonable for most workers, who largely hold only one job, for artists it does not
produce an accurate picture of work hours, earnings and other aspects of labor
market behavior due to their extensive multiple job-holding.
Alper and Wassail present a case for using a more expansive definition of the
work force in analyzing the labor market experiences of authors, and artists in
general. Compared to members of most other professional occupations, artists are
more often marginalized in the labor market. They often moonlight to make ends
meet. In addition, career paths in the arts are rarely well defined. Young persons with
artistic skills and training often find it hard to start an artistic career, either in terms
of obtaining employment or in finding buyers for their work. Such artists would be
excluded from the experienced civilian labor force, since they would be classified as
being unemployed with no prior work experience, or out of the labor force. Other
artists may be working on their art but unable to sell it, while being supported by a
spouse or others. It is not obvious how they would describe themselves in the Census
questionnaire, but they may also wind up classified as out of the labor force.
Such considerations are particularly relevant to artists struggling to join the artist
labor force initially, to stay employed in it and to support themselves, with or without
Introduction I 1 1
holding multiple jobs. Compared to the general labor force where only 6.2 percent
worked in rwo or more occupations in 1989, a large percentage of artists have to
moonlight to survive.
Strengths and Limitations of Census Data
Several of the authors of this report summarized the strengths of Census and
other government research data on artist occupations:
• Data are consistent with and directly compatible with other occupations.
Their use permits the drawing of comparisons among occupations.
• Data are collected at regular intervals permitting trend analysis and the drawing
of comparisons over time.
• Large sample sizes lead to a comprehensive data base permitting detailed analysis.
• A broad picture displays central tendencies of the artist population.
• The data provide an answer to the question, "How many artists?' — a question
often used by policy makers, funders and arts groups, particularly in times of
Census data report the overall labor market activities of artists. However, Census
data sources have some limitations:
• The system of classification. Data are about primary occupation, i.e., where one
works the most hours. Therefore, artists who work more time in jobs other than their
art occupation are not classified as artists. Also, the Census classification system will
make some artists appear more successful in their craft than they really are.
• Data are of limited usefulness in addressing questions specific to specialized artist
• Data do not reflect certain labor market conditions for artists, i.e., artists who
work at a different jobs simultaneously, or in more than one art form, or those
for whom art making is only one of several careers. Thus data are unable to
account for employment and earnings in secondary/multiple jobs which artists
often have and which affect their economic welfare to the extent at times of
keeping them above the poverty level.
Strengths and Limitations of Non-Census Data
Each section of this report supplements Census data with information obtained
from direct surveys of general and specific artist groups, and from professional
organizations and industry records. While the Census is more comprehensive, these
surveys are more detailed. Authors Joan Jeffri and Robert Greenblatt point out that
some discrete surveys are also more attitudinally based than the Census, which some
economists are beginning to acknowledge as important when studying economic
1 2 I Artists in the Work Force
data on artists. Thus these other data have the indirect result of providing a forum
for artists to advocate for themselves, plus indicating areas for further research.
Surveys analyzed in this report include studies based on general surveys of artists
where a broad range of artistic occupations is targeted and the information collected
is relevant to all artists. Also included are studies based on specialized surveys of a
single artistic occupation. A third source of data is information provided by artists to
a professional organization or union.
Harry Chartrand and other authors of this report summarized some of the
advantages of these non-Census data sources:
• Data are on those generally accepted as "professional" in their field.
• Data are useful in addressing questions specific to such professionals.
• They provide valuable reference information since they offer comparable data for
• Questions geared to particular occupational groups yield answers that are more
illustrative of the labor market experiences of that group. Chartrand and others
also point out these disadvantages of non-Census data:
• Data are not necessarily compatible with other occupations.
• Data are not necessarily collected on a regular basis, limiting trend analysis.
• Surveys aimed at general artist groups may not truly represent the population of
artists and are not likely to be as statistically reliable.
• Regarding use of union records, it is difficult to generalize the results to the artist
labor force in general.
Comparing data from Census and non-Census sources cannot help but present
a more complete portrait of artists in the labor force. Figures and percentages from
such different data bases will not always agree, but despite these differences, each of
the various data sources provides a unique view or insight into the conditions
affecting the employment and earnings of artists.
The labor market for artists toward the close of the 20th century is challenging
for those risk-loving individuals who balance "making it big" in the art world against
perhaps not making it at all. The employment answer for many artists appears to be
supplementing their volatile earnings in art endeavors with more predictable earnings
from secondary jobs such as teaching.
Continuing research by the National Endowment for the Arts and others will
further illuminate market conditions for those artists who — more than any other
workers — shape today's culture.
NOTE: The following reports have been condensed from full-length statistical
studies. The authors' original comprehensive documents are available in the
Educational Research Information Center (ERIC) system. They may be accessed by
use of these ERIC numbers for Chapters I, II, & III in the order of the chapters:
ERIC #ED385856, ERIC # ED390728 and ERIC # ED390727. Chapter IV may
be accessed by use of its full original title: Architecture and Design Arts Occupations
1940 to 1990.
I. The Write Stuff:
Employment and Earnings of Authors 1 970
to 1 990
by Neil O. Alper and Gregory H. Wassail
Successful authors are frequently in the news. Winners of the major prizes for
writing receive extensive press coverage. Best selling authors make public
appearances on television, at book signings and readings in book stores. Those
authors, including television and film scriptwriters, fortunate enough to command
top money, are especially newsworthy. One example reported in the Los Angeles
Times is the $18.9 million paid author Dean Koontz by publisher Alfred A. Knopf
for a three-book contract. Two film scripts that sold for $4 million apiece also raised
some eyebrows. Given the media interest and public attention that prominent writers
enjoy, surprisingly little statistical information about them reaches social scientists,
policy analysts, policy makers or writers themselves.
The National Endowment for the Arts (NEA) commissioned this report as part
of its research on employment and earnings of artists in the labor force. The NEA
supports research on the artist work force in general and also shows a special interest
in writers through its Literature Program which has aided writers with fellowships
and other means.
This report surveys and synthesizes current knowledge on the employment and
earnings of authors between 1970 and 1990, compiling and comparing statistics on
how they go about making a living, including second occupations, and how
successful their efforts are. The information comes largely from the U.S. Census and
other government censuses, surveys of authors and from records of writers' unions
and professional organizations. It is hoped that by consolidating information from
diverse sources some sense may be made of the writing profession. In comparing
statistics among surveys and censuses, the diversity in definitions, sampling
procedures and sample sizes should be noted.
The U.S. Census, listing about 107,000 American authors, is the most
comprehensive source of information about them in the country. As mentioned in
the general introduction to this report, it affords several advantages in studying the
labor market experiences of artists, and also has some shortcomings. Because the
Census requires every worker to declare only one occupation, all work experiences
and earnings in a given year are attributed to that occupation. For authors and many
other artists, this leads to an overstatement of earnings from their craft, because direct
surveys have turned up significant percentages working in other occupations in a
14 I Artists in the Work Force
Sources of information about authors which deal with multiple-job holding
among occupations include the annual Cutrent Population Survey. Surveys
specifically aimed at writers give more detail about personal characteristics, working
conditions and earnings. Records of authors' unions and professional organizations
offer little personal information, and often provide data only on sources of income
through union recognized activity.
The author profession as treated in this work embraces poetry, fiction and
creative nonfiction, in English or in translation. Census statistics used here, unless
otherwise noted, are drawn from the broadest definition of authors and related
professionals: all persons who indicated occupational membership, regardless of
whether they were in the labor force at the time of the Census. In 1990, 14 percent
or authors indicated occupational affiliation, despite not being members of the labor
force (neither working nor actively seeking work).
Profile of American Authors
The author occupation is growing rapidly. Between 1970 and 1990 their
numbers increased by 285 percent, with half of this phenomenal growth occurring
between 1985 and 1990. Nevertheless, authors still constitute a very small fraction
of the nation's labor force, or nine in 10,000 in 1990.
Relative to the entire labor force, authors are concentrated in states along the
Atlantic and Pacific coasts, although this concentration lessened somewhat from
1970 to 1990. California and New York are home to more authors than other states
and have the highest percentage of authors in their labor force. The Los Ajigeles and
New York City metropolitan areas have the highest percentages of authors in their
labor forces, and the greatest number of authors, of any metropolitan areas in the
Members of the author profession are well educated. In 1990 they averaged 15.8
years of education, more than other artists and other professional and technical
workers. They are also older; their average age of 44 years exceeded that of artists by
five years and that of other professional and technical workers by four years. In 1990
half of all authors were men and 95 percent were white. These percentages were
higher than those of other professional and technical workers. Over the 1970 to 1990
period the author profession had a lower percentage of minorities than artists or other
professional and technical occupations. It contained more women than did the artist
profession, but fewer than did the other reference groups.
Authors had higher rates of unemployment than other professional and technical
workers, but lower rates than other artists. In 1990 only 40 percent of authors
worked full-time year-round (working at least 35 hours per week and 50 weeks per
year). By comparison, 46 percent of artists and 56 percent of other professional and
technical workers worked full-time.
Compared to the average artist and professional/technical worker, authors were
far more likely to be self-employed, and less likely to work for private sector
employers. Therefore, authors received a higher percentage of their earnings from
The Write Stuff: Employment and Earnings of Authors 1 970 to 1 990 I 1 5
self-employment, although these earnings still were typically less than half of authors'
total earnings. In 1989 authors' total earnings were $23,335. (Earnings and income
data are annual averages, and refer to the year prior to the Census.) In both 1979 and
1989 authors earned less than other professional and technical workers, and less than
editors, reporters and technical writers. In 1969 and 1989 they earned more than
Because authors worked fewer hours per year than members of the other
reference occupations, their computed hourly wage is higher than that of the other
reference groups in all three Census years. Census data also show that authors' total
personal incomes are higher than those of all artists, but lower (except in 1969) than
those of other professional and technical workers. The average total personal income
of authors in 1989 was $30,089, while their average total household income was
Between 1969 and 1989 the earnings of authors grew by 175 percent. However,
this earnings growth lagged behind that of the other reference occupations, and
behind changes in the Consumer Price Index. In constant 1969 dollars, the real
earnings of authors fell by $1,567. The earnings of other reference groups, except for
the professional and technical workers, also fell over this period.
Similar trends were found in total personal income. Between 1969 and 1989 real
personal income of authors fell by $924. Real personal incomes of the reference
groups fell also, except for other professional/technical workers. Authors' household
income, however, outpaced inflation, rising by $2,370. All reference occupations
showed increases in real household income, resulting primarily from the rising labor
force participation of women over this period.
One interesting characteristic of authors' earnings is their relative inequality. In
1989 more authors had zero or negative earnings, and more earned over $90,000
than any of the reference groups. Essentially the same earnings distribution, relative
to the other occupations, was found in 1979 and 1969.
The same array of statistics was examined for authors in the experienced civilian
labor force, and for authors who worked full-time year round. Authors in the
experienced civilian labor force earned $25,800 in 1989, or $2,465 more than all
Census authors (i.e., including authors out of the labor force in 1990). Other
occupations, when limited to labor force members, experienced comparable earnings
increases. Personal and household income rankings did not change either. The
average personal income of authors in the experienced civilian labor force in 1989
was $31,788; their average household income was $63,019.
Comparing only authors and members of other professions who work full-time
year-round minimizes earnings differences caused solely by different amounts of time
worked per year. Authors, as mentioned, worked fewer hours per year. As a result, the
annual earnings of full-time authors compare more favorably to their counterparts in
other occupations. For example, in 1989 full-time authors' earnings of $35,896
placed them above artists, editors, reporters and technical writers, and below only
other professional and technical workers.
1 6 I Artists in the Work Force
Differences in earnings and income by gender, race and ethnicity were also
examined. Given the small size of the author profession and the relatively small
percentage of minorities working as authors, racial and ethnic comparisons were
made only among white, black and Hispanic authors. In all three Census years,
significant disparities between the average earnings of men and women authors were
found. The earnings disparity narrowed between 1969 and 1989, but still remained
significant. In 1989 women's earnings were only 52 percent of men's. Differences in
median earnings and wages were smaller, indicating more men were concentrated in
the high earnings and hourly wage region. Gender differences in personal and house-
hold incomes were smaller. In fact, women authors had higher household income.
Differences in earnings among white, black and Hispanic authors were smaller.
In 1989 black authors' earnings were 91 percent of white authors earnings, and
Hispanic authors' earnings were 87 percent. Greater disparities appeared in personal
income and household income between white and minority authors. The ratio of
black to white authors' personal income was 85 percent, and the ratio of Hispanic to
white authors' personal income was 80 percent.
Evidence on Authors' Multiple Job Holding
The Census Bureau's Current Population Survey, even with its small sample size,
consistently found high multiple job holding rates for authors and artists. From the
mid-1980s to the early 1990s moonlighting rates for authors exceeded all other
professional workers. In 1991 the rate for authors was double that of all artists,
almost three times that of all other professionals and almost four times the rate of the
total work force.
Evidence from non-census surveys of authors, or artists including authors,
confirms that writers' labor market behavior in a given year involves working at more
than one job. Alper- Wassail's survey of New England artists, undertaken in 1981-82,
found that only one in five authors worked full time at writing, and the majority
worked in a job related to their writing some time during the year, primarily
teaching. Additionally, 45 percent held jobs unrelated to their writing. The evidence
shows several of these work activities were done simultaneously.
Columbia University's Research Center for Arts and Culture survey undertaken
in 1989 found that 90 percent of the writers needed to work at some non-writing job
to support their writing, and that almost half were multiple job holders at the time
they were surveyed. Similar behavior was identified in the Columbia Survey of
American Authors done for the Authors' Guild in 1979. It found that 70 percent of
authors had earnings from non-writing work, and that almost half held regular
salaried positions not as writers. Like the other studies, it found that the majority (90
percent) of those who had second jobs held professional jobs, not the traditional
service or clerical jobs of the stereotypical "starving artist."
Surveys of authors in other countries suggest that multiple job holding for
writers is not a uniquely U.S. phenomenon. A study of the members of the British
Society of Authors and the Writers' Guild found that only 1 7 percent of authors
The Write Stuff: Employment and Earnings of Authors 1 970 to 1 990 I 1 7
worked only as writers, and that 67 percent of those surveyed identified writing as a
secondary occupation. A French study found that 30 percent of the authors worked
at another trade during the survey year, and that 70 percent had done so at some time
during their writing career. In Finland, only 22 percent of the authors indicated that
they were full-time writers, and fully half indicated a non-writing occupation for tax
The importance of authors' moonlighting activity is best determined by
examining its impact on their economic well-being. Alper- Wassail found that two-
thirds of the authors' labor market earnings and almost 60 percent of their total
income comes from working in a non-writing job. The Columbia Research Center's
study data suggests that non-writing income, labor and non-labor income combined
account for almost 80 percent of total writers' income. Columbia's study for the
Authors' Guild reported that median writing income comprised 33 percent of
median total personal income, and for the group most dedicated to writing — the
"committed full-timers" — writing income was only 77 percent of total income.
Without their other jobs that are often held concurrently with their writing work,
clearly the economic well-being of authors would be much worse, and most writers
would be classified as poor.
Evidence on Authors from Non-Census Sources
The value of non-census surveys goes beyond their ability to examine the rather
unique labor market behavior of authors and other artists. Following is a summary
of additional findings from the three surveys already discussed and three studies
utilizing the administrative records of the Writers' Guild of America, west, a union
for writers in the television, radio and film industries.
Demographic characteristics of authors follow a generally consistent pattern
across all the studies. Authors are primarily white with an average age around 40
years. The proportion of non-white seems to be increasing. The two earliest studies
(Columbia's Authors' Guild and Alper- Wassail) found approximately 3 percent non-
white authors. The later survey (Columbia's Research Center) found almost 10
percent non-white, and the union's administrative records showed almost double the
number of minority members from 1986 to 1991.
The gender composition of the profession appears to be changing as well. The
earliest study found 60 percent of the authors were female. AJper- Wassail found an
even distribution, and the latest survey (Columbia's Research Center) found 60
percent female writers. The union's administrative records suggest an increase in
female writers employed from 1982, but they accounted for only 22 percent of
employed writers in 1991.
Writers, like their artistic colleagues, are very well-educated compared with the
general population — a consistent finding for all the surveys. Alper- Wassail found
they were better educated than both performers and visual artists. They were also
found, by both Alper- Wassail and Columbia's Research Center, to have started their
training to be writers about age 16, which is considerably older than performers, but
about the same as visual artists.
1 8 I Artists in the Work Force
Two of the surveys, Columbia's Authors' Guild study and Alper- Wassail, found
the income of other family members to be important in explaining the ability of
writers to work at their craft, and that family income for the writers was well above
the poverty thresholds at the time of the surveys.
Aji additional impact on writers' economic well-being is the cost they incur to
write. Alper- Wassail found that writing earnings net of costs averaged half of the
writing earnings, and that median writing earnings net of costs were negative.
Columbia's Research Center's findings were similar, showing that only about 43
percent of the authors had writing income that exceeded their costs.
Female authors were found to consistently earn less than male authors.
Columbia's Authors' Guild study showed female authors' median earnings from
writing to be 77 percent of the male median. This was a considerably smaller
difference than the Alper- Wassail finding where female authors' average earnings
from their writing were only 20 percent of their male colleagues' earnings. When
adjustments were made for the fact that women spent much less time writing during
the year, Alper- Wassail's estimated differential for hourly wage was that women
writers earned about 40 percent of male earnings. The three Writers' Guild studies
identified differentials more in line with the Authors' Guild study, and found the
earnings to fluctuate over time. In 1982 median female earnings were 73 percent of
male earnings. In 1986 they were about 60 percent of male earnings and in 1991 they
were about 75 percent of male earnings.
Considerable differences exist in earnings related to the type of writing the
author does and for whom they work, according to findings by the Columbia's
Authors' Guild study and the Writers' Guild studies. The Authors' Guild study found
that poets earn the least from their writing, authors of academically oriented
nonfiction were slightly better off, children's books writers and writers of general
adult fiction were in the middle, and writers of "genre" fiction earned the most. The
Writers' Guild studies found that writers in television tended to earn more than
writers in film, but the difference narrowed over the decade of the studies to a
difference of only 3 percent in 1991. Greater disparities were found in earnings based
on the type of firm the writer worked for, regardless of whether it was in television
or film. Writers who worked for the major television or film production companies
earned considerably more. In 1991, for example, writers working for the major film
producers had median earnings 1 1 5 percent higher than their colleagues working for
the smaller independent film producers.
Authors in the U. S. Census, 1970 to 1990
This section presents and analyzes evidence on the employment, earnings, labor
market and other related characteristics of authors using information in the 1970,
1980 and 1990 U. S. Census. The evidence is presented separately for men, women,
white, black and Hispanic authors. It compares authors to all artists, using the
definition of artist employed by the National Endowment for the Arts. Included in
the Census definition are: (1) actors and directors, (2) announcers, (3) authors, (4)
The Write Stuff: Employment and Earnings of Authors 1 970 to 1 990 I 1 9
dancers, (5) designers, (6) musicians and composers, (7) painters, sculptors,
craft-artists and artist printmakers, (8) photographers, and (9) artists, performers and
related workers, not elsewhere classified. The following Census occupations have
been added by the NEA: (10) architects, and (11) college and university art, drama
and music teachers. Secondly, authors are compared to all professional and technical
workers, excluding artists, and members of several other occupations, including two
with similar characteristics to authors: editors and reporters, and technical writers.
Census Data Used to Analyze Authors' Work Experiences
Most of the information reported on in this section is derived from the 1970,
1980 and 1990 Census Public Use Microdata Samples (often simply called PUMS).
Each contains a sample of the responses from households who completed the Census
long form questionnaire. A minority of households are asked to complete the long
form, approximately 16 percent of all households in the 1990 Census. The 1990 long
form contained 26 housing questions and 33 personal questions.
The PUMS is unique in that the unit of observation is the household and person
record. Because information about individuals is revealed, several steps are taken to
insure anonymity, including less detail about geographic location than is available in
other Census releases which use aggregated data, typically over geographical areas.
Since the inception of PUMS in 1940, the Census Bureau has extracted different
sized samples for researchers to work with, determined somewhat by the ability of
existing computer systems to store and process the data. In 1980 and 1990, a basic
five percent sample and a supplemental one percent sample were drawn. The
supplemental samples are focused on different geographic information (states, county
groups and neighborhoods), or different segments, such as the elderly population. Or
in 1990, for example, the supplemental one percent sample, like the five percent
sample, was based on a random selection of respondents who completed the long
The principal advantages of using Census Public Use Microdata are noted here.
First, the Census sampling and information collection and tabulation procedures are
very sophisticated; information gained in the PUMS can be generalized to the entire
population with relatively small margins of error. Second, samples are large,
permitting analyses of small occupations such as artists, and within that category,
authors, for example. Third, vast information is obtained from the Census due to the
wide variety of questions asked on the long form. Fourth, the Census provides
consistent observations on a large sample of the population every ten years, enabling
many types of comparisons over time.
Issues in Using Census Data to Analyze Authors' Labor Force
Controversies over the use of Census data to analyze labor market outcomes of
artists are discussed in the general introduction to this report, so only those
particularly relevant to authors will be treated here.
20 I Artists in the Work Force
Respondents to the 1990 long form questionnaire were asked to describe their
chief job activity or business the prior week. If this person had more than one job,
the job at which the most hours were worked was to be listed.
Suppose that people who call themselves authors hold non-writing jobs as well
in a calendar year. By attributing all time spent working and all earnings for an entire
year to the author occupation, the Census overestimates the importance of that
occupation in a person's total work effort and earnings.
The Census tally of the number of authors would seem to be accurate, since only
those for whom writing is their principal work declare themselves as authors. Those
devoting more time and effort to other occupations declare themselves as members
of those occupations. Of course, many persons with little or no earnings from their
art nevertheless consider themselves to be artists. If authors moonlight, then the time
spent working, the earnings and other attributes of labor market behavior are
overstated for the author profession. The evidence suggests that multiple job-holding
is common among authors and several other artist occupations to a greater extent
than in most non-artist occupations. Direct studies of authors and other artists
discussed later verify that a significant percentage of moonlighting authors' (artists')
work time and earnings are from jobs outside their artistic profession. In defense of
the Census methodology, it's worth noting that such direct surveys elicit responses
from persons who may devote only a small fraction of their work time to their art
and would appear in the Census Public Use Sample as members of other occupations.
To summarize: The Census classification system will make authors and several
other types of artists appear to be more successful in their chosen occupation than they
actually are. Despite this, it reports their overall labor market activities accurately.
Whom Does the Census Define as an Author?
Since the Census procedure permits individuals to describe their occupations,
every type of author or writer may be found in the Census PUMS. However, not every
author or writer is found in the authors occupational category. All occupations which
were included in the authors category between 1970 and 1990 are listed below:
Radio script writer
Short story writer
Television script writer
Verse writer-greeting cards
The Write Stuff: Employment and Earnings of Authors 1970 to 1990 I 21
The authors category includes all writers of fiction and many other types of
writer as well. The Census does not provide the relative distribution of persons across
Writers are also found in other Census occupation categories. For example, a
sampling of listings within the editors and reporters category reveals (in 1990)
advertising copy writer, columnist, art critic, book critic, copy writer, editorial writer,
feature writer, literary writer, news writer and sports writer. Similarly, the technical
writers category contains writers in the fields of engineering, health and science, plus
specialists in documentation and technical writing. Technical writers became an
occupational category in 1980; prior to 1980, they were included in the artists,
performers and related workers not elsewhere classified. In 1990, the latter category
contained among its unique potpourri of occupations a few at the very fringes of the
writing profession, such as crossword puzzle maker and language translator. One
might conclude that the author occupational category, if viewed as encompassing all
"creative" writing, could be construed as too broad. On the other hand, if viewed as
encompassing all those who earn their living from writing, it could be construed as
As discussed in the general introduction to this report, most studies of persons
in the labor force use the civilian labor force as their baseline definition. A variant of
this definition — the experienced civilian labor force — includes all persons not in the
military who are working, but only those unemployed persons who have had recent
prior work experience. 1 In the section immediately following, the size and geographic
and industrial distribution of authors in the experienced civilian labor force are
However, compelling arguments also favor analyzing the labor market
experiences of authors and other artists, using a more expansive definition of the
work force. There are numerous anecdotal examples of struggling writers whose
status would officially place them either out of the labor force or as members of other
occupations for extended periods in their careers. Fortunately, the Census permits the
use of a more expansive definition. It includes all persons who call themselves authors
in the Census PUMS, regardless of labor market status or work experience. These
persons will be called all Census authors.
In light of these considerations, the rest of this part of the report uses three
different definitions of the author profession: authors in the experienced civilian labor
force, all Census authors and full-time year-round authors (those who worked at least 35
hours per week and 50 weeks per year). In the next section (see Tables 1 through 4),
information on the size and growth of the profession and on industry and locational
preferences of authors is presented for those authors who are members of the civilian
labor force. In the remainder of the section (including Tables 5 through 16), where
characteristics of the representative author are presented and compared to counter-
parts in the reference occupational groups, the all Census authors definition is used.
Looking at all Census authors makes writers, as well as all artists, appear slightly
less successful relative to other professional and technical workers for three reasons:
(1) a slightly higher percentage of all Census authors is found in the out of the labor
force category; (2) a higher percentage of authors in the experienced civilian labor
22 I Artists in the Work Force
force is unemployed; and (3) a lower percentage of working authors holds full-time
employment. Looking at only those authors in the experienced civilian labor force
has a similar effect, since reasons (2) and (3) still apply. If one's objective is to
compare authors to members of other occupations when they share the same job
status, perhaps the best choice is to compare the labor market experiences of only
full-time, year round workers in each profession (those who worked at least 35 hours
per week and 50 weeks per year).
Growth in the Author Profession and Where Writers Work
The growth in the author profession over the 1970-90 period was exceptional
and erratic. Authors in the experienced civilian labor force grew more rapidly than
any other artist class, at an astonishing rate of 285 percent over this period,
numbering 106,730 by 1990. The most dramatic growth was between 1985 and
1990 when the number of authors roughly doubled (Table 1). Table 2 shows the
regional distribution of authors in the 1970-90 Census years. The nine-region
breakdown in this table is commonly used in geographic displays of Census data.
NUMBERS OF AUTHORS IN THE EXPERIENCED CIVILIAN LABOR
FORCE, COMPARED TO (1) ARTISTS, (2) PROFESSIONALS, AND
(3) ALL WORKERS, 1970 TO 1990
Source: Diane C. Ellis and John C. Beresford, Trends in Artist Occupations,
1970-1990 (Washington: National Endowment for the Arts, 1994).
In 1980 and 1990 the Pacific region, dominated by California, contained the
most authors. In 1970 this region was second in number of authors to the Mid-
Atlantic (including New York), which in turn came in second in 1980 and 1990. The
Pacific region, however, ranked third in the rate of growth of authors over the 20-year
period. The Mountain region had the highest rate of growth, closely followed by the
West South Central region. The Mid-Atlantic had the slowest growth in authors,
closely followed by New England. Larger regions naturally tend to be home to more
authors, but differences exist between authors and all members of the labor force,
both in terms of where they are located and of relative regional growth.
Location quotients are used to show regional concentration, usually of industries
or of workers. Here a location quotient is used to show occupational concentration,
the percentage of authors who reside in a region divided by the percentage of the
The Write Stuff: Employment and Earnings of Authors 1970 to 1990 I 23
entire labor force which resides in the region. For example, a location quotient greater
than one will be found in a region in which a larger percentage of the author labor
force resides than of the overall labor force. A location quotient of exactly one implies
identical shares of the author and the total labor force in a region. For any
occupation, the entire country must have a location quotient of one.
How this plays out among the nine Census regions can be seen in Table 2. The
two regions with the greatest endowment of authors relative to the entire labor force
have rather consistently been the Pacific and the Mid-Atlantic. However, there has
been a gradual trend toward a geographic distribution of authors more closely aligned
with the distribution of the entire labor force. This is shown by the convergence of
the regional location quotients toward one. Most notable in this convergence process
is New England, which in 1970 had the highest concentration of authors. Its location
quotient is now almost one.
NUMBERS AND RELATIVE DISTRIBUTION OF AUTHORS IN THE
EXPERIENCED CIVILIAN LABOR FORCE, BY CENSUS REGIONAL
NUMBER OF AUTHORS
AUTHOR LOCATION QUOTIENT
4713 6068 16804
685 654 2767
1054 2514 6490
5790 13044 28119
27752 45748 106730
DIVISIONAL BREAKDOWN: New England: CT, ME, MA, NH, Rl, VT. Mid-Atlantic: NJ,
NY, PA. East North Central: IL, IN, Ml, OH, Wl. West North Central: IA, KN, MN, MO,
NE, ND, SD. South Atlantic: DE, DC, FL, GA, MD, NC, SC, VA, WV. East South Central:
AL, KY, MS, TN. West South Central: AR, LA, OK, TX. Mountain: AZ, CO, ID, MT, NV,
NM, UT, WY. Pacific: AK, CA, HI, OR, WA.
SOURCE: Ellis and Beresford, Trends in Artist Occupations, 1970-1990 (Washington:
National Endowment for the Arts, 1994), Table 5; U.S. Department of Labor, Bureau
of Labor Statistics, Geographic Profile of Employment and Unemployment (Washington,
24 I Artists in the Work Force
More detail on author residence can be seen at the state level. Table 3 shows the
ten states in which the most authors resided in the three Census years. These top ten
states have been home to roughly two-thirds of all authors in the labor force in each
Census year, regardless of their identity and ranking. California ranked first in author
residence in 1980 and 1990, but second to New York in 1970. Washington joined
the top ten ranking in 1990. Other states ascending the rankings are Texas, rising
from ninth in 1970 to third in 1990, and Florida, rising from eleventh in 1970 to
fifth in 1990.
THE TEN STATES WITH THE MOST AUTHORS IN THE EXPERIENCED
CIVILIAN LABOR FORCE IN 1990
Amount in 1990
Amount in 1980
Amount in 1970
Top Ten in Each Year
as % of Total Author
SOURCE: Ellis and Beresford, Trends in Artist Occupations, 1970-1990
(Washington: National Endowment for the Arts, 1994), Table 5.
Another way to look at the distribution of authors among states is to examine
their percentages in each state's experienced civilian labor force, where the ten states
with the highest percentage of author to total employment are ranked. The top ten
states in this ranking are New York, California, Vermont, New Mexico, Virginia,
Massachusetts, Colorado, Connecticut, Washington and Alaska. New York and
California top both state rankings, and may well be special cases, with their
concentrations of radio, television, film, publishing and advertising industries. Many,
if not all of the ten states in this ranking are high "quality of life" states where people
with no constraints on where they can live would be likely to locate. Prime examples
are Vermont and New Mexico.
For Whom Authors Work: Distribution by Industry
When authors in the experienced civilian labor force are split into the ten broad
industry groups derived from the Census, 65.9 percent of authors are found in
The Write Stuff: Employment and Earnings of Authors 1 970 to 1 990 I 25
Professional and Related Services. In 1990 only Construction and Manufacturing
also contained more than 10 percent of the total. Because most industry groups
contain a large number of industries, it is more enlightening to examine which
specific industries employ the most authors. Table 4 shows that Miscellaneous
Professional Services employed 52.2 percent of all authors in 1990. All other
industries held no more than 7 percent of author employment in each Census year.
The top ten industries combined employed 80.8 percent of all authors in 1990. Of
the top ten industries in 1990, nine were also in the top ten in 1980, but only four
were among the ten largest in 1970.
THE TEN INDUSTRIES WHICH EMPLOYED THE MOST AUTHORS IN
1990 AND THEIR PERCENTAGE SHARE OF THE AUTHOR
EXPERIENCED CIVILIAN LABOR FORCE
% in 1990
% in 1980
% in 1970
Miscellaneous Professional Services
Printing and Publishing,
Theaters and Motion Pictures
Colleges and Universities
Management and Public
Radio and Television Broadcasting
Newspaper Printing and Publishing
Business Services, n.e.c.
and Testing Services
Top Ten in Each Year as %
of Total Author Labor Force
* Included in Miscellaneous Professional Services in 1980 and 1970.
**lncluded in Miscellaneous Professional Services in 1970.
SOURCE: Authors' tabulation 1970, 1980, 1990 U.S. Census PUMS.
Demographic Characteristics of Authors
The information that follows incorporates everyone who identified himself or
herself as an author in the Census, regardless of being in the experienced civilian
labor force at Census time. Therefore it includes (1) those who were unemployed
with no prior work experience, (2) those claiming an occupation but who are out of
the labor force, and (3) those employed by the military. By changing to the more
inclusive all Census authors description from the experienced civilian labor force
description, the number of authors covered increased significantly.
26 I Artists in the Work Force
Basic demographic characteristics of all Census authors in 1970-90 are shown in
Tables 5 through 8. Tables 5 and 6 show detailed characteristics of all authors, and
then of selected reference groups: all artists (including authors), all professional and
technical workers other than artists, and the two closely related job categories: editors
and reporters and technical writers. Tables 7 and 8 reproduce the same characteristics
of all authors, and then of male, female, white, black and Hispanic authors. 2
DEMOGRAPHIC CHARACTERISTICS OF ALL CENSUS AUTHORS VS.
SELECTED REFERENCE GROUPS, 1970
% WHO ARE:
rs' tabulations from the 1970 Census PUMS.
In 1990, the average Census author was 44 years old, and had completed just
under 16 years of education. Thus it is not surprising to discover that 43 percent of
all authors held a bachelor's degree. An additional 21 percent had master's degrees.
Of the authors in the 1990 Census, 57 percent were married and heads of their
households, 50 percent were women, 95 percent were white, 3 percent were black
and one percent was Hispanic.
Most demographic trends found in the overall population and labor force have
parallels in the author labor force from 1970 to 1990. The U. S. labor force has
become better educated, more ethnically diverse and comprised of a higher
percentage of women. Other general demographic trends observed in all the
occupational groups profiled in Table 5 and 6 include lower rates of marriage and
household formation and a decline in the percentage of veterans.
The conventional wisdom in the United States would probably regard authors as
being very similar to other artists, but authors' demographics differ in many ways
from those of artists. With respect to education, authors more closely resembled
The Write Stuff: Employment and Earnings of Authors 1 970 to 1 990 I 27
DEMOGRAPHIC CHARACTERISTICS OF ALL CENSUS AUTHORS VS.
SELECTED REFERENCE GROUPS, 1990
% WITH DEGREES:
% WHO ARE:
Drs' tabulations from the 1990 Census PUMS.
other professional and technical workers from 1970 to 1990. However, their level of
education exceeded that of all artists by about one-and-a-half years. In 1990 authors
held more of every type of college degree than all artists, and held more bachelor's,
master's and doctor's degrees than other professional and technical workers. Authors
had consistently higher rates of marriage and heading households than artists,
explained perhaps by their averaging four to seven years older than artists. This age
differential reflects both later entry and greater longevity in the occupation. It
narrowed somewhat in 1990, due probably to many new, younger entrants to the
In other respects, authors differ from both artists and other professional and
technical workers. As noted, authors in every period are older than both groups.
Persons with work-affecting disabilities 3 consistently form a higher percentage of the
author labor force than of the other groups, due to the relatively minimal physical
demands required by writing. The typical author is more likely to be a woman than
is the typical artist, but less likely to be a woman than the typical
professional/technical worker. The typical author is less likely to be a member of a
28 I Artists in the Work Force
racial minority or to be a Hispanic than either reference group members. Thus no
numerical evidence in the Census reflects the increasing popularity of African-
American and Hispanic writers that has occurred in the United States in the 1980s
and 1990s. The percentage of these two groups in the author labor force actually
declined in those decades.
The two groups, editors and reporters and technical writers, appear similar to
authors, although the age in both has been consistently younger. Educational
attainment is similar, but some race and gender differences exist. Compared to
authors, in each Census year editors and reporters have had a higher percentage of
women and technical writers a lower percentage of women. Both reference groups
have had a higher percentage of racial and ethnic minorities in their ranks in each
Data on authors who are also members of the experienced civilian labor force
show essentially the same picture and will not be treated in detail here. Generally,
authors in the civilian labor force are younger, better educated, more likely to be
married and heads of households and less likely to be women. In 1970 and 1980 they
were more likely to be white. In 1990 the percent of white and black were essentially
the same. Comparable differences exist between all Census artists and artists in the
civilian labor force, and apply to professional and technical workers as well.
Authors who work full-time year-round are a much smaller group, constituting
39.7 percent of all Census authors and 46.4 percent of authors in the experienced
civilian labor force. Full-time authors had somewhat more education, were 41
percent women, and a larger percentage belonged to racial and ethnic minorities.
The demographic characteristics of authors over the 1970-90 period, broken
into gender, race and ethnicity are shown in Tables 7 and 8. Over all three Census
years, white authors were older and had completed more formal education than
minority authors. Male authors in 1980 and 1990 were older and possessed more
formal education with a higher percentage of men writers possessing professional and
doctoral degrees. More women writers possessed bachelor's and master's degrees. In
general, a higher percentage of white writers possessed bachelor's and higher degrees
than minority writers, with the exception of Hispanic writers.
Authors in the Labor Market
In this section a variety of characteristics relating to the labor market status of
authors and their reference groups is examined. Tables 9-10 present information on
authors versus the selected reference groups for 1970-90. Tables 1 1 and 12 focus on
labor market characteristics over the 20-year period by gender, race and ethnicity
rather than by reference groups.
Several important points should be kept in mind relating to how this
information was collected. First, questions on labor market experiences may refer to
the Census year or to the prior year, depending on the reported activity in which the
respondent was engaged in during the reference week. Other statistics provided in
these tables, such as full-time, year-round status and average hours and weeks
worked, refer to work experience in the prior year, since these statistics are based on
The Write Stuff: Employment and Earnings of Authors 1 970 to 1 990 I 29
DEMOGRAPHIC CHARACTERISTICS OF ALL CENSUS AUTHORS,
RACE, AND ETHNICITY,
% WHO ARE:
the 1970 Census
DEMOGRAPHIC CHARACTERISTICS OF ALL CENSUS AUTHORS,
BY GENDER, RACE, AND ETHNICITY,
% WITH DEGREES:
% WHO ARE:
tabulations from the
30 I Artists in the Work Force
LABOR MARKET CHARACTERISTICS OF ALL CENSUS AUTHORS AND
SELECTED REFERENCE GROUPS, 1970 (Med
lians in Parentheses)
% WHO ARE:
Not in Work Force
from the 1 970 Cens
LABOR MARKET CHARACTERISTICS OF ALL CENSUS AUTHORS AND
SELECTED REFERENCE GROUPS, 1990 (Medians in Parentheses)
% WHO ARE:
Not in Work Force
tabulations from the 1990 Census PUMS.
The Write Stuff: Employment and Earnings of Authors 1 970 to 1 990 I 31
a calendar year's labor market experience. Also, for persons such as authors for whom
moonlighting is common, the labor market data — especially the annual data — is
likely to reflect time spent and experience in non-author jobs as well.
A labor market snapshot of authors in 1990 shows that 2.7 percent experienced
unemployment during the reference week and 14. 1 percent were neither working nor
unemployed (out of the labor force). From 1970-90 authors had consistently lower
rates of unemployment than all artists, but consistently higher rates (1.5 times
greater) than other professional and technical workers. However, author
unemployment rates over the 20-year period were roughly comparable to those of
editors and reporters and technical writers, and in 1990 authors had lower rates. 4
Since 1980 authors have been more likely to be found out of the labor force than
either of these groups, and than other professional and technical workers in general.
Regarding the frequency and duration of employment, in 1990, roughly the
same percentage of authors as other professional and technical workers worked at
some time during the Census year. A greater difference (3 percentage points) exists
in the percentage actually working during the reference week, suggesting that
relatively more authors drift in and out of employment in a given year than other
In 1980 and 1990 authors worked fewer weeks and fewer hours per year than all
artists, and fewer authors worked full-time year-round, due most likely to the much
higher rates of self-employment in the author profession. However, this higher rate
of self-employment for authors does not account for why other professional and
technical workers continually experience less unemployment than authors. The
difference in unemployment is more likely explained by these characteristics shared
by other artists: Few or no entry barriers to the profession, less stability in existing
jobs, frequent new jobs and assignments, changes in jobs and extended periods of no
Among all Census authors in 1990, besides the 58 percent who reported self-
employment, 8 percent reported working for government and 32 percent reported
working for private sector employers. Authors differ from other professional and
technical workers in their low percent of government employment.
Authors in the experienced civilian labor force were found to have higher rates
of both employment and unemployment rates than all Census authors. Members of
the experienced civilian labor force by definition can only be working or
unemployed; they cannot be out of the labor force. By contrast, all Census authors
can be out of the labor force. In comparing all Census authors to full-time year-
round authors, rates of unemployment and of absence from the labor force in the
Census year are naturally lower among persons who worked full-time the entire
previous year. However, full-time authors were less likely to be self-employed (46.1
percent vs. 58.0 percent for all Census authors in 1990). The comparisons among
authors by gender, race and ethnicity in Tables 1 1 and 12 generally reflect differences
found in the labor force as a whole. Men were more likely to be employed than
women, and less likely to be out of the labor force. A significantly higher percentage
of men worked full-time in all three years (57.6 percent men versus 49.4 percent for
32 I Artists in the Work Force
LABOR MARKET CHARACTERISTICS OF ALL CENSUS AUTHORS,
1970 (Medians in Parentheses)
% WHO ARE:
Not in Work Force
tabulations from the 1970 Census PUMS.
women in 1990). However, in 1980 and 1990 both black and Hispanic authors were
more likely to be full-time than white authors. Also, in both years the percentages of
blacks and Hispanic authors working for any level of government are two to three
times those of whites. Most of this differential is made up by lower rates of self-
employment among minority authors, as it is hard to imagine many full-time jobs in
government as authors. This choice in turn may reflect greater difficulties among
minorities in "making it" as full-time independent writers.
Earnings of Authors
This section examines how authors have fared economically over the 1970-90
period. Earnings and other sources of income to authors are reported and compared
to earnings of reference occupations. The term earnings refers to income from work
effort, either in the form of wages and salaries or earnings from self-employment.
Census data on earnings, as noted above, are attributed to one occupation — that
which the respondent identifies as the "chief job activity or business last week." Thus
this information does not distinguish earnings of authors received from writing or
from earnings received in other lines of work. Surveys of authors discussed in the next
section of this report consistently report that a low percentage of total earnings is
derived from writing.
The Write Stuff: Employment and Earnings of Authors 1970 to 1990 I 33
LABOR MARKET CHARACTERISTICS OF ALL CENSUS AUTHORS,
RACE, AND ETHNICITY, 1990 (Medians in Parentheses)
% WHO ARE:
Not in Work Force
tabulations from the 1990 Census PUMS.
The Census provides separate categories for wages and salaries and self-
employment. The latter category is further broken into non-farm and farm
components. Only the non-farm component is reported here in self-employment
earnings. For authors (and the reference groups discussed) farm earnings are minimal,
and clearly cannot be from writing. Nevertheless, for the sake of completeness, total
earnings reported include all wage, salary and self-employment earnings, including
Successful authors may command earnings not only from writing, but may
receive royalties from published work. Such royalties are included here in a category
called asset income which also includes interest, dividend and net rental income.
Asset income can only roughly measure actual royalty income received by authors. By
comparison, little or no royalty income is earned by non-authors, so all else being
equal, authors should receive greater amounts of asset income. 6
A wage rate is also calculated and reported here. This is defined as total earnings
from work divided by total hours worked. Several of the income entries from the
1970 and 1980 Census that are reported in the tables below were calculated by the
authors of this report. One of these — that of total earnings — was done out of
necessity, because it was not a separate category until the 1990 Census. Two other
calculations — of total personal income and total household income — were made to
minimize the adverse effects of the Census practice of capping income categories to
34 I Artists in the Work Force
preserve anonymity. In 1990, for example, the Census capped each income category
at different levels, which were determined by the earnings distribution in each
category. Total personal income was capped at $284,000, but asset income was
capped at $40,000.
In Tables 13 and 14 the earnings and incomes of all Census authors are
compared to all Census artists and other professional and technical workers over the
1970-1990 period. The actual year in which earnings and incomes were recorded was
the year prior to the Census, since all earnings and income data are annual averages.
In Tables 13 through 16 mean (average) values are reported first. Median
(midpoint) values are reported in parentheses below the mean values. Means, rather
than medians, are discussed below because many of the categories reported on have
more than half the sample reporting zero earnings. In such cases, the mean provides
more useful information than the median, which is zero. A comparison of means and
medians is helpful in interpreting some of the differences in earnings found among
occupations. As one example, the ratio of the mean to the median is later used as a
measure of earnings variability.
In 1989, the average total earnings of all Census authors were $23,335. These
earnings were about $2,100 more than the average earnings of all artists, but almost
$4,800 less than the average earnings of all other professional and technical workers.
Editors and reporters and technical writers also earned more than authors, by $1,600
and $4,700 respectively. A higher percentage of authors' earnings was derived from
self-employment than from the reference groups.
Comparisons among occupational groups over the Census years reveal that only
in 1969 did authors earn more than other professional and technical workers, and
more than editors and reporters. Since 1979, the first year for which this information
is available, technical writers have also earned more than authors. In all three years all
artists' earnings were less than those of other professional and technical workers. In
1979 only, artists' earnings were greater than authors' by about $200. 8
Authors' earnings grew at a relatively slow rate (175 percent) between 1969 and
1979, lagging behind those of the reference groups (205 percent for artists, 226
percent for editors and reporters, and 261 percent for other professional and
technical workers). Between 1979 and 1989, authors' earnings grew faster than those
of the reference groups.
These earnings increases may initially seem dramatic. However, most have not
even kept pace with rising prices. After deflating earnings to allow for changes in the
Consumer Price Index over this period, it is apparent that in constant 1969 dollars,
authors' earnings fell from $8,743 in 1969 to $5,524 in 1979, and then rose to
$6,906 in 1989. Thus in constant 1969 dollars authors' earnings fell by $1,567 over
the 1969-1989 period. By comparison, all artists combined lost $684, editors and
reporters lost $265, and other professional and technical workers gained $527 in real
earnings during this time.
Earnings data for authors in the experienced civilian labor force tell essentially
the same story, ranking the same relative to those in the other occupations: first in
1969, last in 1979, and lower than all but artists in 1989.
The Write Stuff: Employment and Earnings of Authors 1 970 to 1 990 I 35
MEAN INCOME OF ALL CENSUS AUTHORS AND SELECTED
1969 (Medians in Parentheses)
Wage & Salary
from the 1
970 Census PUMS.
MEAN INCOME OF ALL CENSUS AUTHORS AND SELECTED
Wage & Salary
>' tabulations from the 1990 Census PUMS
36 I Artists in the Work Force
The earnings of full-time year-round authors in 1989 were $35,896, or about
$12,500 more than the earnings of all Census authors, ranking first among the
reference occupations in 1969 and last in 1979. In 1989, authors earned more than
all artists, and than editors and reporters and technical writers. In real 1970 dollars,
earnings of full-time authors fell by $1,433, artists' earnings fell by $1,604, editors
and reporters' earnings fell by $1,121, and those of other professional and technical
workers fell by $525.
The hourly wages reported in Tables 13 through 16 are constructed by dividing
total earnings by hours worked. A comparison of these wages across the occupational
categories for all Census members shows that authors had the highest hourly wage in
all three Census years. In 1989, the author average hourly wage was roughly $10
more than that of other professional workers. This is mainly because authors spent
considerably fewer hours working than did the other professionals. In addition, a
number of authors worked relatively few hours per year, but were very well
compensated for their work. In two of three Census years, the wages of all Census
authors were higher than the wages of experienced civilian labor force authors. Also,
in general the wages of all Census or experienced labor force authors were higher than
those of full-time year-round authors.
Variability in Authors' Earnings
In every Census year, authors had a higher percentage of persons with zero (or
negative) earnings than people in other comparable occupations. On the other end
of the spectrum, Census evidence from 1970-90 shows that a higher percentage of
authors had earnings at the maximum level (the level at which the Census assigns
persons the same value regardless of actual income or earnings). Also, a higher
percentage of authors had asset income (and income from all sources) at the
maximum than any other occupational group.
An examination of earnings variability data for authors and reference groups in
the experienced civilian labor force shows the same relative outcomes. Naturally,
fewer workers are found with zero earnings and more are found at the earnings
maximum in each year. A higher percentage of authors again appears at both the low
and high ends of the earnings spectrum compared to the reference groups.
With full-time, year-round workers, the percentages of persons with zero
earnings shrink considerably, while high percentages of persons have earnings at the
maximum. There are minimal changes in the rankings by occupational group.
Another statistic that highlights the concentration of persons at the low end of
the income spectrum is the percentage of members in an occupation with household
or individual incomes below the poverty line. 9 The poverty status of all Census
authors and members of occupational reference groups is found in Tables 13 and 14.
Income counted toward poverty status is from all sources, including other family
members, and thus includes more than earnings. This statistic also measures the
concentration of poverty in different groups.
Poverty rates of all Census authors, and of artists in general, are higher than those
The Write Stuff: Employment and Earnings of Authors 1 970 to 1 990 I 37
of editors and reporters, technical writers and all other professional and technical
workers combined. Poverty rates fall for experienced labor force workers, and fall
further for full-time, year-round workers. What does not change is the relative
ranking of occupations within each category. Authors and all artists have the highest
incidence of poverty, regardless of the labor force definition employed. This finding
is particularly notable given that authors had the highest household incomes of all
the occupational groups.
Why are the earnings of authors subject to such variability? As mentioned earlier,
authors, among other artists, work fewer hours in a year and experience greater
variability in those hours worked. When one considers the career of authors and of
many other artists, one sees persons who move among projects or employers.
Compensation may differ greatly from move to move. A best-selling book may be
followed by one which does not sell. Rarely is there a long-term contractual
relationship with a steady employer with compensation rising gradually every year, as
there is in many other occupations. Alternative explanations exist, as put forth in the
introduction to this report. Artists, including authors, are often risk-takers and
gamble on the chance of becoming famous and wealthy through their occupation.
Then also, "psychic income" may figure in, where persons will accept lower
compensation because of the satisfaction their occupation provides in itself. Reality
may lie in a mixture of these explanations.
Other Sources of Income
Total personal income — the sum of all income sources, including earnings —
shows an improved status for authors when ranked against reference groups. In 1969,
authors' personal income ranked first, as did authors' earnings. In 1979 and 1989,
authors' personal income ranked third. The main reason for their improved status in
the personal income rankings is the greater amounts of asset income authors receive,
including royalties and income from property. Royalties are hypothesized to be the
main difference in boosting authors' personal income; their asset income is from two
to four times larger than the reference groups. Nevertheless, asset income averages
only about one-fifth of total earnings for authors.
The total personal incomes of authors in the experienced civilian labor force rank
the same in each year relative to the reference groups. Among full-time, year-round
workers, authors' relative ranking in personal incomes is similar to those noted above,
except that their personal incomes placed them first in 1989 as well as in 1969.
The personal incomes of authors and most of the reference groups failed to keep
pace with inflation from 1969 to 1989. All Census authors, for example, suffered a
loss of $924 in real total personal income over this period. Authors in the experienced
civilian labor force and full-time year-round authors also lost purchasing power over
this period, by roughly the same amount.
The financial status of authors shows improvement when one moves from
earnings to total personal income to total household income. Tables 13 and 14 show
that authors rank first in total household income each Census year.
38 I Artists in the Work Force
The primary contributor to household incomes other than the professionals
described here clearly is the spouse or partner, when one is present. Why do authors
have spouses who contribute more to household income than do members of other
professions? One possibility is that authors, with their higher educational levels,
might attract better educated spouses with higher earnings.
Authors in the experienced civilian labor force continue to rank first in total
household income. The picture changes only slightly when full-time year-round
authors are examined. Their household income places them first in 1969 and 1989,
and second in 1979.
Regarding the effects of inflation, household income increases experienced by all
occupational groups surveyed kept them ahead of inflation. All Census authors saw
real household incomes rise by $2,370 over the 1969-89 period, with reference
groups seeing comparable gains. This general conclusion held true for experienced
civilian force and full-time year-round counterparts in each occupational group. The
basic reason household incomes kept ahead of inflation is the increasing labor force
participation of women over the same period, so more family members were in the
Earnings and Income by Gender, Race and Ethnicity
Tables 15 and 16 summarize information about authors' earnings and income by
gender and race. Men dominated all personal earnings, income and wage categories
in all three Census years. Women's annual earnings as a percentage of men's increased
from 42.8 percent in 1969 to 47.1 percent in 1979 to 52.5 percent in 1989. Women's
calculated hourly wage, measured as a percentage of men's, increased from 58.2
percent in 1969 to 82.0 percent in 1979, but fell dramatically to 40.4 percent in
1989. The wage disparity in 1989 is especially noteworthy. The high hourly wage
earners referred to above are mostly men, whose mean wage in 1989 of $40 is $24
higher than the women's. However the men's median wage of $12 is less than $4
higher than the women's median.
In absolute dollar terms, differences in total personal incomes between men and
women authors are larger than differences in earnings. In 1979 and 1989 men earned
more asset income. However, women authors had greater household income — less
than $1 ,000 more than men — in all three Census years. While the household income
of men authors is about one-third greater than their personal income, the household
income of women authors is about two-thirds greater. Authors in general marry well,
but women authors particularly find successful spouses.
Less overall disparity exists among earnings of members of different races and
ethnic backgrounds. In general, white earnings are higher than those of blacks and
Hispanics, but percentage earnings differentials are narrower than those between the
sexes, and show no consistent trends. Black authors' earnings were between 80 and
90 percent of white authors' earnings, with the highest ratio in 1969. Hispanic
authors' earnings were between 64 and 115 percent of white authors' earnings, with
the highest ratio in 1979. Hourly wages showed similar small gaps in 1969 and 1979.
The Write Stuff: Employment and Earnings of Authors 1 970 to 1 990 I 39
MEAN INCOMES OF ALL CENSUS AUTHORS,
BY GENDER, RACE
, AND ETHNICITY, 1969 (Medi
ans in Parentheses)
Wage & Salary
SOURCE: Authors' tabulations from the
! 1970 Census
MEAN INCOMES OF ALL CENSUS AUTHORS,
BY GENDER, RACE, AND ETHNICITY, 1989 (Med
ans in Parentheses)
Wage & Salary
1 1 .3%
SOURCE: Authors' tabulations from the
! 1990 Census
40 I Artists in the Work Force
In 1989, however, the gap widened dramatically. Typically, larger disparities existed
between the personal and household incomes of whites versus blacks and Hispanics,
than between men and women authors.
The calculated mean wages of men authors and white authors more than
doubled between 1979 and 1989, and significantly outpaced the growth in women
and all minority wages. There was no comparable explosion in median wages for men
and whites. Thus the growth in mean wages was driven by the success of a limited
number of white men in securing very high hourly compensation for their output.
In a similar vein, there was no explosion in the earnings of whites and men in 1989.
The high mean wages reflect high hourly compensation, not high total
As discussed above, poverty levels are based on household incomes, and not just
earnings. In 1979 and 1989, the poverty rates of men authors were higher than those
of women. Similarly, in the same two Census years, the poverty rates of white authors
was lower than those of black and Hispanic authors, by wide margins. In 1969,
women authors had a higher incidence of poverty than men authors, despite having
higher household incomes. Also, in 1969 white authors had a higher incidence of
poverty than black and Hispanic authors, despite having higher household incomes.
The unusual distribution of poverty incidence among ethnic groups in 1969 may
partly be attributable to small samples of black and Hispanic authors in that Census.
Why Authors' Earnings Differ from Those of Other Artists
There is little evidence that being an author per se increases one's earnings above
those of other artists. A better explanation of authors' earnings lies in the importance
of the human capital variables. Recall that authors were consistently older and better
educated, and were more likely to be white, married and the heads of households. All
of these traits are positively correlated with earnings. Although some traits of authors
(such as more likely to be a woman and to be self-employed) predict lower earnings,
on balance, a human capital interpretation of differences in artists' endowments of
human capital would predict higher earnings for authors than for most other artist
One might also expect authors to have the strongest verbal and writing skills of
any artist occupation. Though not measured by the Census, these traits are correlated
with job success as well. There is considerable evidence that the majority of the
earnings of authors is not from writing. Thus it may be true that authors' verbal
abilities enhanced their earnings in other occupations more than any general human
capital skills possessed by other artists.
Authors in the U. S. and Elsewhere
This section continues to explore the economic condition of writers over the
period 1970 to 1990. It differs from the previous chapter because it does not use data
from the U.S. decennial censuses. The Census, while complete in such important
The Write Stuff: Employment and Earnings of Authors 1 970 to 1 990 I 41
dimensions as coverage of the population, is weak in describing the complexities of
the labor market experiences and income generating opportunities utilized by writers
and other artists.
Multiple job holding among writers will first be examined, using the Census'
close cousin, the Current Population Survey. Then information will be provided on
a number of issues relevant to the writers' socioeconomic condition, utilizing data
from a variety of surveys and studies. Last will come a brief comparison of the U.S.
writers' socioeconomic condition to writers in several other countries.
Current Population Survey
Multiple job holding is a characteristic of writers' and artists' labor market
behavior that somewhat sets them apart from most other workers. Generally the
term, multiple job holding, suggests that more than one job is held concurrently, but
the data does not always confirm this.
The Current Population Survey (CPS) interviews approximately 60,000
households annually. 10 Since artists are such a small proportion of the overall labor
force, just over one percent, and with writers being only six percent of all artists, the
accuracy of the CPS estimates as they relate to artists and writers is of concern. What
could appear as considerable change in the behavior of writers nationwide might
result from the change in the actions of a single writer in the sample. 11 However, the
CPS does provide an important supplement to the census information.
Overall the multiple job holding rate in the U.S., calculated from the CPS, has
shown a general upward trend, though it fluctuates with changes in economic
conditions. In 1989 it was 6.2 percent. 12 The CPS indicates that multiple job
holding is more common among artists and writers than other comparably trained
workers. Eliminating those artists whose employment behavior is more like that of
other professional workers than it is like an artist's (architects, designers and
photographers), then the multiple job holding rate for artists was 13.7 in 1991. In
both 1989 and 1991 the rates for writers were approximately 20 percent, the highest
rates among the artist occupations, except for that of those artists whose primary job
was teaching art at the post-secondary level. (Table 17)
Table 18 shows secondary occupations for individuals whose primary occupation
was that of author. According to the CPS, the majority of these secondary jobs were
in other professional occupations. In both 1989 and 1991 teaching something other
than writing at a college or university was the most common second job for writers.
Table 19 shows primary occupations for those whose secondary job was that of
author. Not surprisingly, these primary jobs were in professional occupations, with a
significant proportion of them being managerial.
Several characteristics of writers seem to be associated with multiple job holding.
In 1985 and 1989, writers with second jobs were more likely to be male than writers
who held only one job. The opposite was true in 1991. Writers who held second jobs
averaged 10 years younger than those who did not, suggesting that multiple job
holding is more common among the new entrants into the writing occupation.
42 I Artists in the Work Force
CURRENT POPULATION SURVEY - MULTIPLE JOB HOLDING RATES:
1985, 1989, 1991 (Percent)
Art Teachers (post-secondary) 29.3
Photographers 1 1 .3
Artists NEC 14.5
All Artists 9.8
Other Professionals 6.9
Source: Authors' tabulations and calculations from Current Population St
for May 1985, 1989 and 1991.
CURRENT POPULATION SURVEY
- AUTHOR'S SECONDARY
1985, 1989, 1991 (Percent)
Art Teacher (College)
Source: Authors' tabu
lations and calculations from Current Popi
Surveys for May 1985,
1989 and 1991.
The Write Stuff: Employment and Earnings of Authors 1970 to 1990 I 43
It is interesting to note that about one- third of the artistic occupations to be held
as a second job was that of musician. Writers, on the other hand, accounted for only
about seven percent of the secondary artistic employment in 1985 and 1991, and
almost thirteen percent in 1989.
CURRENT POPULATION SURVEY - AUTHOR'S PRIMARY
1985, 1989, 1991 (Percent)
Source: Authors' tabulations and calculations from Current Population Su
for May 1985, 1989 and 1991.
Sources of information about authors other than national censuses are special
studies which fall into three categories. One category includes studies based on
general surveys of artists where a broad range of artistic occupations is targeted. A
second category includes studies based on specialized surveys of a single artistic
occupation such as authors. A third category is information provided by artists to a
professional organization or union and retrieved from an organizations
First of the two general artist surveys to be examined is the Alper- Wassail Survey
(AW) of artists in New England. 13 The second will be the Joan Jeffri survey
undertaken by Columbia University's Research Center for Arts and Culture 14 of ten
U.S. locations, incorporating eight cities and two non-urban areas (RCAC). Both
studies were undertaken in the 1980s.
Next to be discussed is a survey explicitly designed for authors, the Kingston-
Cole study, 15 done by Columbia University's Center for the Social Sciences for the
Authors Guild Foundation. This study draws heavily on the definition of an author
44 I Artists in the Work Force
established by the Guild. This means that those included in the study were book
authors whose work had been published. This study, while undertaken in the early
1980s, refers to the situation of writers in the late 1970s.
Last to be examined is information from three Writers Guild of America, west
(WGA) studies of writers in Hollywood. 16 Utilizing administrative records, the WGA
reports provide a detailed, but limited, description of writers in the motion picture,
television and radio industries. They are necessarily limited to the information the
WGA maintains on its members and to what the members are required to report to
the WGA. One of its strengths is that with readily available administrative data,
several reports covering 1982 through 1991 could be prepared, providing a
longitudinal study not available from any other source.
The AJper- Wassail New England Survey studied artists in the six New England
states in 1981 and 1982. Authors comprised 12.1 percent of the sample. In
comparison, authors comprised 4 percent of artists in the 1980 census. Based on a
system of self-identification, 62.4 percent of the authors were "writers," 32.0 percent
were "poets" and 5.5 percent were "playwrights" and other creative writers.
Table 20 highlights the demographic characteristics of the New England authors
and compares them with other regional artists. Authors average about 43.5 years in
SOCIO-ECONOMIC CHARACTERISTICS OF NEW ENGLAND
AUTHORS, PERFORMERS, AND VISUAL ARTISTS
% Who Are
calculations from Alper-Wassall's study of New
The Write Stuff: Employment and Earnings of Authors 1 970 to 1 990 I 45
age, contrasted to 40 years average age for other artists. Authors are better educated,
with three times more holding doctorates than all artists. Writers were not likely to
attend specialized arts schools, and relatively unlikely even to major in their field
through their undergraduate education.
Authors do not dramatically differ from other New England artists regarding
marital status and racial distribution, but slightly more than half the authors are
female, unlike performing artists, but more like visual artists.
Alper- Wassail identified three separate labor markets that authors, and all artists,
were likely to participate in. One was the market for their art work. A second was
working in a job related to the arts, including teaching their art and arts
administration. The third was working in a job completely unrelated to the
production of their art work — the proverbial taxi driver or waiter, for example.
Every author in the study spent some time during the year writing, but only
about one in five were full-time writers, (see Table 21). Authors averaged about 33
weeks of the year working as authors, about the same time spent by the other artists
in the production of their art. Like their artistic peers, most authors with arts related
jobs (almost 75 percent) were involved in teaching their art at some level. Alper-
Wassall found that one in five authors were unemployed at some time during the
year, averaging 13 weeks out of work.
Almost 45 percent of the authors in this survey held a job unrelated to their
writing with about one-fourth of these jobs in non-teaching professional
LABOR MARKET CHARACTERISTICS OF NEW ENGLAND AUTHORS,
AND VISUAL ARTISTS 1981
Arts-Related Job (%)
Full-Time Artist (%)
Worked per Week
Weeks Worked in
Weeks Worked in
Non-Arts Related Job
Source: Authors' tab
study of New
46 I Artists in the Work Force
occupations. Approximately 1 5 percent held teaching jobs not related to their writing
and another 10 percent were employed in food service type jobs. Often working at
two or more jobs at a time, authors spent approximately the same number of weeks
working as writers as in work related to writing and unrelated work combined. They
appeared to earn more per hour from their non-writing work.
Table 22 details the authors' employment experiences as reflected in their
earnings and total income. Authors' total income, which included earnings from
their various jobs and such non-labor income as rent, interest and dividends, was
$17,126 in 1981 dollars, about 10 percent higher than that of all New England
artists. It is not clear from this study where authors would have included any royalty
income they might have received.
MEAN ANNUAL INCOMES OF NEW ENGLAND AUTHORS,
PERFORMERS AND VISUAL ARTISTS, 1981 (Medians in
Net Arts Earnings
Source: Authors' tabulations and
calculations from Alpe
r-Wassall's study of New England
Authors, like most artists, incur significant costs in order to produce their
writing. New England authors incurred costs of about $2,000 in 1981. The impact
on their earnings from writing was that authors' earnings net of costs were slightly
more than half their gross earnings from writing.
The hourly wage rate for the time authors spent writing was estimated to be
$2.62, less than 80 percent of the federally legislated minimum wage in 1981 which
was $3.35. The authors' estimated wage rate from the 1980 Census was almost $13,
but it reflected earnings from all the jobs held and not simply from working as a
The Write Stuff: Employment and Earnings of Authors 1970 to 1990 I 47
writer. Weekly earnings from all jobs worked, which clearly do not correct for
differences in the hours worked per week, showed authors were above average
compared to other artists.
However the author works — as an employee, self-employed or free-lance — he or
she must learn about employment and writing opportunities, or must know how to
market his/her work. More than half the authors studied by Alper- Wassail found
writing jobs through friends and relatives. Networking through former business
associates was a method utilized by almost half the authors in searching for a job. Of
course, multiple methods are used in job searches at various times. For authors in
New England these included advertisements, booking agents, private employment
agencies, public employment offices, student placement offices, and "other."
The marketing methods most commonly used by New England authors were
agents, consignment in showroom or shop, shows and fairs, advertisements, own
showroom or shop and "other."
Table 23 highlights some demographic differences between male and female
authors and their artistic colleagues in New England, while Table 24 examines the
labor market characteristics of these authors and their fellow artists. These findings
are summarized briefly here:
SOCIO-ECONOMIC CHARACTERISTICS OF NEW ENGLAND
i, PERFORMERS, AND VISUAL ARTISTS: BY GENDER
% Who Are
ations from Alper
New England artists.
Male and female writers were found to be essentially the same age. Female writers
had less formal education than males, with male writers more than twice as likely to
have a doctoral degree. The proportion of male and female authors married or living
with someone who shared income and expenses was much the same as was the
48 I Artists in the Work Force
proportion of each gender in minority groups. In general women writers did not do
as well in the labor market and were more likely to be unemployed during the year.
Female writers were half as likely to be full-time writers as their male colleagues, and
were found to work less time per year than male writers.
Differences in earnings for authors in this study are available, but not detailed
here. Generally, women authors earned about 18 percent of what male writers earned
from writing on an annual basis, but fared better with regard to other earnings.
Overall, though, the female writers' annual earnings from all sources were half their
male colleagues, while their total weekly earnings were 63 percent of what male
writers earned. Female writers' household income was only slight lower, 3 percent,
than that of male writers. Most authors who held writing-related jobs taught at some
level, with more men teaching at the college level. Male writers also were seven times
more likely to work in managerial and executive jobs. Female writers were about
seven times more likely to hold clerical jobs, while male writers were about five times
more likely to hold operative and laborer jobs, a pattern consistent with that observed
in the general population.
LABOR MARKET CHARACTERISTICS OF NEW ENGLAND AUTHORS,
, AND VISUAL ARTISTS BY GENDER 1981
Related Job (%)
Worked per Week
Source: Authors' tabi
jlations and calculations from Alper- Wassail's stu
dy of New E
Columbia University's Research Center Arts and Culture Survey
Another general survey of artists undertaken by Columbia University's Research
Center for Arts and Culture (RCAC) r focussed on examining the work- related
The Write Stuff: Employment and Earnings of Authors 1 970 to 1 990 I 49
human and social service needs of artists. It was a mail survey in 1989 of artists, not
simply authors. It covered Boston, Chicago, Dallas, Los Angeles, Minneapolis/St.
Paul, New York City, Philadelphia and San Francisco, plus Western Massachusetts
and Cape Cod.
The RCAC study provides some valuable insights into the economic condition
of authors, and with its similar findings to other studies it becomes part of a useful
body of studies that confirms the uniqueness of artists in general, and authors in
Authors in this study were very highly educated, especially those who stated their
major area of concentration as "writing/literature." More than one-third had
bachelor's degrees and more than one-half had graduate degrees. Over 63 percent had
formal degrees in arts. RCAC authors are somewhat younger, more likely to be
women, and more racially and ethnically diverse than both the Alper- Wassail and
Census authors. They started their training at 16.4 years, which is relatively old
compared to other artists. Most interesting is that the vast majority, 80 percent,
indicated that they were, at least in part, "self-taught" writers. More than 40 percent
of RCAC authors prepared for their writing careers though the use of private teachers
and/or mentors. 18
Table 25 details the labor market characteristics of RCAC authors. Their
unemployment rate was almost double the national rate. Almost 90 percent of the
writers needed to work at some non-writing job to support their art, and nearly half
of them held multiple jobs at the time of the survey. A significant proportion
indicated they worked more than a standard 40-hour week.
The authors' income sources were varied, including working as a writer or other
artistic endeavor. Only a minority had income from grants and/or awards, royalties,
and unemployment insurance. (Table 26) Barely half the authors earned more than
$500 (1988 dollars) from writing, writing-related or arts-related activity. An estimate
of the average authors' earnings is over $4,600, enough to cover the expenses
associated with producing their artistic work for just 43 percent of the writers.
Earnings from work unrelated to writing played an important role in
determining the authors' economic well-being. Almost half the RCAC writers had
total income greater than $20,000, with the average income estimated at almost
$21,700. This suggests that, on average, almost $17,000 of their income came from
work not related to their writing or other activity and from non-labor income
Special Artist Surveys
The number of studies based on surveys of a single artistic occupation are
relatively few, which makes the Columbia Survey of American Authors (CSAA)
sponsored by the Authors Guild Foundation so unique. 19 In addition to describing
authors' experiences and economic condition, it obtained information that is
idiosyncratic to the authors. It dealt with writing genre, separating information for
writers of children's book, adult fiction and adult nonfiction, for example. It also
made distinctions between different types of royalties.
50 I Artists in the Work Force
LABOR MARKET CHARACTERISTICS
OF RCAC AUTHORS,
Other Jobs to Support Writing or Other Arts Work 88.0
Multiple Job Holder
Hours Writing or Other Arts Work:
Hours Worked on Non-arts Job:
Source: Authors' tabulations and calculations
from Jeffri's study of artists.
INCOMES OF RCAC AUTHORS, 1988 (percent)
$ - 500
501 - 3,000
3,001 - 7,000
7,001 - 12,000
20,001 - 40,000
$ 0- 5,000
5,001 - 10,000
20,001 - 30,000
30,001 - 40,000
Grants - Awards Income
Royalties, etc. (median)
Writing income exceeds costs (percent)
Source: Authors' tabulations and calculations
from Jeffri's study of artists.
The Write Stuff: Employment and Earnings of Authors 1 97C to 1 990 I 51
The CSAA surveyed almost five thousand writers in 1980. Approximately 60
percent of those sampled were from the Authors Guild's membership list. The
remaining 40 percent were from a list of authors who had been invited to join the
Guild, but had not done so. Therefore, this survey included only writers as defined
by the Authors Guild's membership criteria: Writers with at least one book published
in the last seven years, three works of fiction or non-fiction published by a magazine
in the last 18 months, or a writer whose professional reputation entitles him/her to
membership according to the Membership Committee. The study excluded
academic writers, screen writers and new entrants to the writing market.
Findings from the Columbia survey are generally consistent with those from the
other two studies discussed. The survey identified 40 percent of authors as being
women. CSAA authors were well educated, predominantly white (97 percent) and
about two-thirds were married.
Fully 70 percent of the authors had labor earnings from work other than their
writing. Almost half the authors (46 percent) held regular salaried positions other
than as a writer, and 40 percent of those without regular non-writing positions
worked irregularly at things related to their writing such as editing, translating and
lecturing. Approximately 40 percent held non-teaching professional jobs, and an
equal proportion held teaching jobs, mostly at colleges and universities (90 percent).
If the CSAA authors had to live only on earnings from writing, the majority
would be in poverty. It was the non-writing activity and the income of the writers'
spouse/spouse equivalent that provided total family income comparable to, if not
better than, most professional workers.
The typical (median) free-lance writer earned $4,775 in 1979 dollars from
writing. About 10 percent of the free-lancers actually earned no income from
writing-related activities and one-fourth earned less than $1,000. The top 10 percent
of writers earned $45,000 or more, and the top 5 percent earned more than $80,000
from writing. The hourly wage for writing-related work, after controlling for the
amount of time spent, was $4.90 per hour.
In general, the CSAA authors' writing-related income was somewhat less than
half their personal income. Family income for the typical CSAA author was $38,000,
reflecting considerable work effort on the part of each author's spouse. The
contribution of the typical writer's husband to the household income was about
$26,000, and that of the typical writer's wife about $4,000.
One factor correlated with writing-related income was the author's writing genre.
Poets earned the least from books, with approximately 60 percent earning less than
$2,500, and almost 75 percent earning less than $5,000. Authors of "academically
oriented nonfiction" fared only slightly better with almost 70 percent having earned
less than $5,000. Writers of genre fiction (westerns, thrillers, science fiction, etc.) did
better. Almost one-fourth of them earned in excess of $50,000 from their writing,
while about the same proportion earned less than $5,000. By comparison, only 15
percent of writers of general adult fiction and 7 percent of children's books writers
earned more than $50,000.
The author's gender made a difference. Median earnings for female authors was
77 percent of those for male writers ($4,000 vs. $5,200). This difference did not exist
52 I Artists in the Work Force
at the extremes of the earnings distribution, as approximately the same proportion of
men and women earned less than $2,500 and more than $50,000 from writing. For
"committed full-time" writers, writing genre helped explain the gender wage
differential. Women were almost three times as likely as men to write children's
books, which usually don't generate as much income as other books.
Because the CSAA study collected information on more than one year's income,
a limited analysis could be made of the stability in authors' writing income. At the
extremes of the income distribution a great deal of stability existed. More than 80
percent of those earning less than $2,500 in 1978 from their writing did so in 1979.
Half those who did improve earned between $2,500 and $4,999, only a slight
improvement. At the other extreme, almost 90 percent of the writers who earned
$100,000 or more in 1978 did so in 1979, and of those who did change, half were
still earning between $50,000 and $99,000. Between these two extremes, fewer than
half the writers with 1978 writing incomes between $2,500 and $19,999 remained
in the same income group in 1979. Approximately 60 percent of writers with
incomes in the range of $20,000 to $99,999 were in the same income category in
both 1978 and 1979.
Writers belong to a variety of professional organizations, including PEN and the
Authors Guild, but very few of these organizations have any need to regularly collect
information on the employment or earnings of its members. Very irregularly they
may survey their members, as with the Authors Guild's study discussed above.
Writers' unions, however, do need to regularly obtain employment and earnings
information from their members, many of whom are independent contractors, rather
than more traditional employees. The only way their unions can ensure that they are
being compensated according to the negotiated rates is for the members to report
their employment activity. It is also important in establishing employers'
contributions to the unions' health and welfare funds.
Two of the largest unions for writers are the Writers Guild of America and the
Dramatists Guild. The Writers Guild of America (WGA) is the only true writers'
union because it is recognized by the National Labor Relations Board. This is because
it represents writers who are employees, rather than writers who are independent
contractors or who lease their copyrighted material to others. The WGA organizes
writers, bargains with the production companies, networks, etc., and administers
agreements in the television, radio and movie industries. Two affiliated unions make
up the WGA: the WGA, East and the WGA, west. The Mississippi River divides the
The WGA, west, the larger of the two, has been very active since the mid-1980s
in studying the employment experiences of its members, producing three reports
which are derived from their administrative records. 20 The use of these records
provides reliability that is not available from surveys of writers which rely on their
memories or records. Also, there is no problem of response rate as with a survey.
The Write Stuff: Employment and Earnings of Authors 1 970 to 1 990 I 53
On the other hand, the information in the union's administrative records
probably does not provide a complete picture of the labor market experiences of its
members. This is especially true for those writers who are not currently employed or
who were employed part-time during the year. Other sources of labor earnings,
whether from some other form of writing or not, are not included in the WGA
records. Nor is any income from non-labor sources, such as unemployment benefits,
welfare, rent, or other family members.
New entrants into the occupation are not likely to be included. One reason is the
initiation fee, $1,000 for WGA, East, for example. Newly employed writers who have
not worked enough to be admitted to the union clearly are not included either.
The union's records do not include much background information on the
writers' education, training, socioeconomic background, and other elements of their
lives that are relevant to a complete understanding of their economic condition.
Even with these shortcomings, the three reports provide an interesting picture of
the earnings and employment of writers in the television, film and radio industries —
a picture consistent with what has already been discussed.
Like the three reports written by Bielby and Bielby, the primary focus here will be on
the employment and earnings experiences of women, minorities and "older" writers.
It is, though, important to look first at the broader changes affecting WGA, west
Employment of writers in the television, film and radio industries grew by more
than 30 percent from 1982 to 1991 to almost 3,700 writers. The supply of writers
outpaced the growth in jobs over this period. In 1985, 54 percent of the WGA, west
members were employed in jobs covered by the unions' Minimum Basic Agreements
(MBA) for at least one quarter during the year. In 1991, this percentage had
decreased to 48 percent.
Earnings of WGA, west writers grew considerably from 1982 to 1991. Median
earnings (nominal) for the employed writers grew almost 117 percent, from $26,100
to $56,619. Earnings growth was not equally shared by all writers. Those in the top
5 percent of the earnings distribution saw their earnings grow by 96 percent over the
period, picking up momentum from 1987 to 1991. Bielby and Bielby concluded that
the gap between the highest paid and lowest paid writers grew in the late 1980s.
Overall WGA membership grew by 40 percent from 1986 to 1991. Female
membership grew by 60 percent and minority membership almost doubled.
However, the occupation is still dominated by white males, who represented 75
percent of the employed writers in 1991. The proportion of females was 22 percent,
and minorities, almost 3.5 percent. Older writers, defined by Bielby and Bielby as
more than 40 years old, had a 48 percent share of employment in 1991.
The differences between WGA, west's earnings by gender are similar to the
Census and other studies discussed. Throughout the ten-year period the median
earnings of employed female writers were considerably less than male earnings. By
the end of the decade women writers' median earnings were $45,995, or 75 percent
of male writers. These figures include only earnings from writing jobs covered by the
MBA. Also, the studies do not provide information on the number of hours worked
54 I Artists in the Work Force
which would enable the estimation of a wage rate.
With respect to minority writers' relative earnings, the decade from 1982 to
1991 was a period of considerable gain. In 1982 the median earnings of minority
writers was $11,780, or 40 percent of those for non-minority writers. By 1991 the
median earnings for minority writers were $48,061, or 80 percent of the earnings of
WGA, west data suggests that significant earnings differences exist among
members depending on the genre in which they are working and by whom they are
employed. In 1987 the median earnings for television writers were 22 percent higher
than those of the median film writer, but this gap closed to 3 percent by 1991 . 21
Throughout the entire period of the three WGA studies, the writers who worked
for the major film production companies earned more than the writers in any other
sector of the film or television industries. 22 Relative to the writers who worked for
independent film producers (e.g., Orion and Gaunt Films Ltd.) those who worked
for the major producers (MGM, Paramount, etc.) earned 72 percent more in 1982
and 115 percent more in 1991.
In television the pattern was essentially the same. Writers working for the major
producers of television shows (e.g., Columbia and Fox) earned 62 percent more in
1982 and double in 1991 what the writers employed by independent production
companies earned. Writers employed by the three television networks also earned
considerably less than those who worked for the major production companies.
With regard to earnings, the relative difference between female and male writers
varied within the genres. Women writers' relative earnings showed greater
improvement among film writers than television writers. In 1987 the median
earnings of female film writers was two-thirds that of white-male film writers. In
1991 the gap had narrowed to 86 percent of male earnings. In the television industry
the 1987 differential ratio was 68 percent, but improved to 77 percent by 1991.
Within the film industry earnings of female writers fluctuated considerably
relative to male writers, but from 1982 to 1991 showed improvement in major film
production companies from 75 percent to 85 percent. Among the writers in the
smaller film production companies, female earnings in 1982 were 80 percent of male
earnings, and in 1991 were 78 percent of their male colleagues.
The television industry saw female writers working for the major networks make
significant gains over the 10-year period. In 1982 female writers earned 57 percent
of male writers' earnings, but by 1990 and 1991, female writers' earnings were on par
with male writers' earnings. Women writers working for the major television
production companies in 1982 earned 61 percent of what male writers earned, but
by 1991 they were earning 90 percent of what their male colleagues earned.
Minor gains were also made in the numbers of women writers employed in
television and film industries over the ten-year period. In the film industry female
writers were 17 percent of the writers employed at the major studios in 1982 and 18
percent in 1991. At the independent studios female writers were about 14 percent of
the writers in 1982 and 18 percent in 1991. In television production female writers
for the networks moved from 24 percent of writers employed in 1982 to about 27
The Write Stuff: Employment and Earnings of Authors 1 9/C to 1 990 I 55
percent in 1991. One of the largest gains in female employment was among the
major television production companies where women comprised almost 19 percent
of writers employed in 1982 and 23 percent in 1991.
Identifying trends in earnings or employment for minority writers is made more
difficult by the relatively small numbers having been employed at all. For example,
in 1987 the 13 minority writers working at major film production companies earned
40 percent of their white male colleagues' earnings. One year later, minority
writers — only five in number — apparently earned 135 percent more than their white
male colleagues. Small absolute changes, but significant changes in relative terms,
thus impact aggregate data even when utilizing medians.
Considerable employment gains have been made by minority writers in both the
television and film industries. Over the ten-year period minority employment in the
major television production companies increased from 2.3 percent of total
employment to 4.4 percent. At the networks minority representation among the
writers increased from 1.0 percent to 4.6 percent. Absolute numbers were small,
with just 16 minority writers at the networks in 1991, up from three in 1982.
Minority employment at the major film studies increased from 1.0 percent to 3.2
percent, while at the independent studios employment for minority writers went
from 1 .0 percent to 2.2 percent. The total number of minority writers working in the
movie industry in 1991 was only 50, up from 11 in 1982.
Earnings of minority writers in the film and television industries relative to their
white-male counterparts were volatile, but generally trended upward. In 1982
minority writers employed by the networks earned 55 percent of what the white-male
writers earned. By 1991 they were earning almost 70 percent of what their white-
male colleagues earned. At the major television producers median minority earnings
moved from 57 percent of median white-male earnings in 1982 to 93 percent by
1991. In the film industry minority writers' median earnings at the major film
studios went from 25 percent of the white-maie median in 1984 to two-thirds in
1991. Gains for minority writers were greater among the independent film
producers where in 1982 their median earnings were one-third of their white-male
colleagues, but in 1991 were 18 percent larger.
Writers in Other Countries
It is interesting to note, even briefly, that general patterns found for authors in
the U.S. tend to hold for authors in other parts of the world. International
comparisons of authors are made difficult because few other countries except
Australia and Canada obtain detailed information on the earnings and labor market
experiences of the population comparable to that collected in the U.S. However,
bearing that in mind, it is possible to observe some commonalities.
The number of authors in other countries is also very small compared to the
overall work force. In Australia and Canada, where census data are available, authors
accounted for less than one-third of a percent. In the U.S. they were less than one-
tenth of a percent. As in the U.S., the majority of writers in Canada, Finland and
56 I Artists in the Work Force
Australia were men, about 55 percent of the total. In France the proportion of male
writers is much larger, about 80 percent. In several countries, as in the U.S., the
proportion of female writers has been growing.
Countries with data available on schooling showed authors to be very well
educated. In Australia 44 percent of the authors had earned the highest degree
reported in the census, compared to 8 percent for the total work force. In Canada in
1981, 42 percent had earned bachelor's degrees, while only 10 percent of the general
work force had done so. This pattern was also found for authors in Finland and
In all the countries studied female authors' income was lower than that of their
male colleagues. The differential some places was considerably less than in the U.S.
where the median income for female writers was 52 percent of male writers' incomes.
In Australia in 1986 female writers earned about 80 percent of what male writers
earned. In Canada in 1980 female writers' income was about 65 percent of male
writers' income. Even in Finland in 1984, where extensive public programs support
all artists including writers, female writers had incomes that were 68 percent of male
Several surveys of writers in Great Britain, France and Finland confirm that
writers there are not very different from writers in the U.S. In these countries the
majority of writers had held multiple jobs at some point in their careers. In England
in the early 1980s, 67 percent of writers surveyed identified writing as a secondary
occupation. Finnish writers, too, are likely to be multiple job holders and not likely
to be able to earn a living from their writing alone.
During the decades covered by this report, 1970 to 1990, there was a dramatic
growth in the number of artists in the United States, with authors standing out as the
fastest growing of the artistic occupations. According to the Census, the number of
authors almost quadrupled, while the number of artists barely doubled. This period
of growth saw considerable change in the composition of the occupation. At the end
of the period women comprised half the authors when they were only one-third at
the beginning. The proportion of non-white increased somewhat as did the
education level, which was the highest among the artist occupations and on par with
workers in other professions.
The authors' labor market experiences also underwent some significant changes.
In 1990 they were more likely to be self employed than they were in 1970, following
a similar trend in the labor force as a whole. They worked slightly more hours during
the week when they were working, but worked fewer weeks during the year.
The economic condition of authors showed some signs of improvement and
some signs of deterioration over the period. This is true relative to all other artists and
relative to other professional workers. While the growth in authors' wage and salary
earnings was considerable, it was not as large as it was for all artists and for all
professional workers, perhaps reflecting the shift toward self employment. On the
The Write Stuff: Employment and Earnings of Authors 1970 to 1990 I 57
other hand, authors' total personal income was still higher than for all artists at the
end of the period as it was at the beginning, as was total household income. The
largest gain made by authors was in their hourly wage rate. Its increase was greater
than the increase for all the artists, professional workers and other writing
occupations. As a result of some of these changes, the proportion of authors in
poverty increased, but not as much as for all artists.
While these changes were occurring in the U.S., some of them were also
occurring in other countries. Australia, Canada and Finland saw growth in the
proportion of female authors. One common element that characterizes authors, and
artists, throughout the world is the prevalence of multiple job holding. While this is
not identifiable in the Census data, every survey of artists and authors, whether in
the U.S. or elsewhere, finds this to be a common trait. In most countries the majority
of writers work at more than one job during the year. For many this involves a job
related to their writing, such as teaching, but for others it is the proverbial food
service job. Another common trait is that female authors, on average, earn less than
their male colleagues. There are many explanations for this pattern, including that
women tend to spend less time working as writers, and that they write in genres that
are not as well rewarded.
The authors occupation world-wide is undergoing considerable change. Future
research targeted to the distinctiveness of writers will be needed to determine these
patterns and the impact of the changing economic and technological environment
within which they are working. Change is inevitable, but what direction it takes and
how it impacts authors is yet to be determined.
1. In practice the two definitions are virtually equivalent. There is often no
difference between the two with respect to the occupations examined here.
2. The Hispanic question in the 1970 Census was worded differently than in 1980
and 1990, thus the information reported on Hispanics may not be comparable.
3. The Census has several questions about disability status. Tables 5 and 6 refer to
those who report a disability that either limits or prevents work.
4. Unemployment rates for the years between Censuses can be calculated from the
Current Population Survey. Sample sizes are much smaller, however, so they are not
reliable for small occupational groups such as authors.
5. Because of this practice, total earnings reported in the tables may slightly
exceed the sum of the wage and salary and self-employment earnings reported
6. One can only speculate where other types of earnings authors may receive, from
fellowships and readings, for example, would be classified.
7. For example, in 1980 total personal income was estimated as the sum of the
amounts in each of the seven categories, yielding possible amounts up to
$525,000. The Census capped it at $75,000.
8. Comparisons of median earnings among these groups tell a more dismal story.
58 I Artists in the Work Force
Authors start with the highest median earnings in 1969, and then have the lowest
in 1979 and 1989.
9. The poverty line is a federally established standard constructed by multiplying
the cost of feeding a nutritionally balanced diet by three. The poverty line
increases with increasing family size, and is adjusted annually for changes in the
cost of living.
10. Beginning with January 1994, questions about multiple job-holding are now
asked on a monthly basis in the CPS.
11. In the 1991 CPS only nine individuals indicated their primary occupation was
author, and that they held a second job.
12. Stinson (1990).
13. For example see Wassail, Alper, and Davison (1983).
14. For example see Jeffri (1989).
15. Kingston and Cole (1986).
16. Bielby and Bielby (1987), (1989), and (1993).
17. Jeffri (1989)
18. Multiple responses were permitted to this question and others in the survey.
19. Kingston and Cole, op. cit.
20. Bielby and Bielby (1987), (1989), and (1993).
21. Bielby and Bielby (1993), Table 5.
22. Bielby and Bielby (1987), Appendix Table 4, and Bielby and Bielby (1993) Table 5.
About the Authors
Neil O. Alper is an associate professor of economics at Northeastern University.
His teaching and research interests are in the area of applied microeconomics.
Previously, Professor Alper had a joint position in the Department of Economics and
the Center for Business and Economics Research at the University of Tennessee in
Knoxville. He has published a number of articles, books and reports, and presented
professional papers on cultural economics, economics of crime and labor economics.
He received his baccalaureate degree in economics from the State University of New
York at Stony Brook, and his master's and doctoral degrees in economics from the
University of Pittsburgh.
Gregory H. Wassail is an associate professor of economics at Northeastern
University. He has authored a number of articles on the economics of the arts and
culture in books and journals. In addition, he has co-authored The Arts and the New
England Economy (1989) and Artwork: The Work and Pay of Artists in New England
(1983). He has served as a consultant to both regional and state arts agencies, and to
state and local governments on issues facing the arts, education and tax policy.
II. Artists Who Work with Their Hands
A Trend Report, 1 970 to 1 990
by Joan Jeffri and Robert Greenblatt
Artists are always a product of their own times. The years covered by this report
have seen resources developing for individual artists that have changed the
artists' relationship to their own growth, careers and support systems. Such resources
have taken the form of grants, arts service organizations, commissions, and
cooperative and commercial galleries, to name a few. The G.I. Bill after World War
II, for example, became a turning point in the formal higher education of visual
Since 1970, development of the nonprofit sector on a broad scale has expanded
opportunities for visual artists, providing many more venues to show their work. Pro
bono attorneys, like Volunteer Lawyers for the Arts, have helped inform visual artists
about their rights, including copyright. These signs of progress, and many more,
continue to affect the careers and living conditions of those who work with their
This report examines trends in the visual arts occupations of painters, sculptors,
craft artists and artist printmakers — all of whom currently occupy one category of the
United States Census — from 1970 to 1990 in the areas of employment, earnings and
geographic distribution. In addition to information from the U.S. Census and related
research monographs, information will be targeted from artist population surveys
conducted by the Research Center for Arts and Culture at Columbia University and
surveys by other researchers, including one commissioned by the New England
Foundation for the Arts and a longitudinal study conducted by psychologists from
the University of Chicago.
As discussed in the general introduction to this overall report on artists, there are
both limitations and advantages to Census data sets and artist survey data sets.
However, taken together, these sources provide a broader profile of visual artists in
the United States than has been possible before.
A basic limitation of all these surveys is their regard of arts occupations as
comparable to other trades and professions. Sociologist Judith Adler discussed this:
"A study of artists in a society in which occupational membership is (fortunately) not
defined or restricted by a guild, an academy or a state system of licensing can neither
comfortably ignore problems of occupational definition nor resolve them.
Accepting, then, an imperfect definition of art occupations, there are various
limitations and advantages of the data used as a basis of this analysis.
The focus here with non-census data will be the 1980s. Several surveys which
cover different geographic locations during that decade serve to complement and
60 I Artists in the Work Force
broaden census data. Also, by 1980, public funding at the federal and state levels had
become increasingly important to visual artists. And the 1980s marked the "art
boom," as journalists describe it, a time when prices for contemporary visual art
reached an economic peak in the marketplace. Also, the visual arts provided a
catalyst, through controversy over work by artists Robert Mapplethorpe and Andres
Serrano, among many others, for a broad-based public discussion of the role,
function and support of art.
Data from the United States Census
• Multi-Categories. Those who have a particular interest in visual artists will
acknowledge that the combination of artist types in one category not only groups
together very different kinds of creators, but each group interacts very differently
with the profit and nonprofit marketplaces. Their commonality lies in the fact
that they all work with their hands. Lumping together artists whose work can
appear in multiples with artists who work in singular objects distorts the
information gathered on earnings (money made from a single object, a limited
edition, or a series of multiples), hours worked (time required to produce one
painting, one sculpture, one craft work or the original for what will become a
series of prints, glass goblets, etc.) and even geographic distribution (access to
equipment from suitable studio space to foundries and kilns).
• Multi-Jobs. Another limitation of Census data regards multi-jobs. It is well
known that artists often hold two or more jobs simultaneously. Artists included
in the Census are asked to describe their chief occupation during the previous
week and are cited under a single occupation. While the "reference week" may
limit the artist's Census occupation according to the job he spent the most hours
earning money from (i.e., taxi driving), there is no provision for his artist
occupation to be cited. Finally, the broadness of Census categories tends to hide
certain subtleties. (What about multi-media artists, for example?) Agencies like
the National Endowment for the Arts seem to have understood these differences
for funding purposes, using discrete categories which represent a broad variety of
• Multi-Art Forms. A problem not addressed by the Census or most other artist
surveys is the fact that some artists work in more than one art form. In the
process of being an artist, they pursue a number of art forms until they find the
one that best suits their mode of expression. They may work for a period of time
in one art form and find their work evolving into expression through a different
art form. Or they combine art forms, sometimes being labeled as multi-media
artists and sometimes as hyphenates: painter-craftsperson, dancer-photographer,
etc. Taken together with the multi-job and multi-career situations mentioned
above, it becomes clear that accurate "artist categories" can be extremely
How artists are placed in Census categories is discussed in the general
introduction to this report. A few major discrepancies regarding artists who work
with their hands are pointed out here.
Artists Who Work with Their Hands A Trend Report, 1970 to 1990 I 61
• Artists are classified in the Census under Managerial and Professional Specialty
Occupations with a subset of Writers, Artists, Entertainers and Athletes. Under
this are finer categories of architects, designers, etc.
• Through the decades, even within this category, at least one occupation has gone
through a major revision according to the Census: before 1980 there was no
category defining "craft artists," so it is not known where craftspeople were
identified. Since 1980, the craft artist is part of the multi-category However,
within the finer Census categories (just as examples) there are no categories for
Fiber Artist, Weaver, Goldsmith, Silversmith, Leatherworker, Papermaker or
Bookbinder. Calligraphers appear in the 1990 Census category "Artists,
Performers and Related Workers, Not Elsewhere Classified," but are not counted
in the Census figures which report craft artists.
• Before 1970 many of the artists in the multi-category analyzed in this report
were classified under Artists and Art Teachers. In 1970 the category became
Painters and Sculptors. Starting in 1980 the category became Painters, Sculptors,
Craft Artists and Artist Printmakers.
• Job titles under the Census category "Painters and Sculptors" included in artist
occupations of the 1980 Census, cardpainters, music autographers and tattoo
• In the 1980 Standard Occupational Classification Manual under the larger
classification "Fabricators, Assemblers and Hand Working Occupations," the
following were listed: Hand Sewing Occupations (Embroiderers?), Hand
Painting, Coating and Decorating Occupations and Miscellaneous Hand
Working Occupations. There were also separate listings for Bookbinders and
Cabinet Makers. Clearly, not everyone who fits into one of these categories
would claim he is an artist or craftsperson, but some would. The above examples
illustrate difficulties of inclusion as well as exclusion in Census categories.
Three surveys conducted by the Research Center for Arts and Culture (RCAC)
at Columbia University were used for this report and are summarized briefly here.
1. In 1986 The Artist's Work- Related, Human and Social Services Questionnaire
surveyed 900 fellowship applicants to the New York Foundation for the Arts
(NYFA) with a focus on visual artists categories in New York City. The RCAC
administered the same questionnaire in three separate rounds of 300 artists each
in March, June and September of 1986. In October a random sample of 50
artists was sent the questionnaire as a control group to ascertain if fellowship
applicants responded differently before, during and after the fellowship
application process, since this was a universe of artists applying for money. In
fact, there was a significant decrease in the response rate from March 1986 before
fellowships were announced to the subsequent two rounds after fellowships were
announced. The 900 artists represented 14 artistic disciplines. The total response
was 561 artists or 62 percent. Of these, 163 represent artists in the painting,
62 I Artists in the Work Force
sculpture and crafts categories in New York City. Information from the three
separate rounds has been aggregated here.
2 In 1988 the RCAC conducted the Information on Artists Survey of 9,870 in 10
U.S. locations: Boston, Cape Cod, Chicago, Dallas, Los Angeles,
Minneapolis/St. Paul, New York City, Philadelphia, San Francisco and western
Massachusetts. Artists were located with the help of local, regional and national
organizations, and the overall response rate was 4,146 or 42 percent. Although
the sample was a random one, it is not certain that it was representative of the
artist population, since one of the problems of artist definition is the universe
from which the sample is drawn. Nevertheless, this information provides a
parallel portrait to that portrayed by the U.S. Census. This report will focus on
those artists who, when asked their major field of concentration, gave
painting/drawing, sculpture or crafts as their first, second or third choice.
3. The 1991 Artists Training and Career Project surveys focused on painters and
craftspeople. In the craft survey 3,942 questionnaires were mailed to a random
sample in 1990. The response rate was 33 percent, representing 1,301
craftspeople. In the painters survey, a random sample of 2,000 painters was
mailed in 1991. The response was 960 painters or 48 percent with 2 percent
arriving after the data entry period, so 920 painters or 46 percent was the
number used for reporting.
4. In addition, the survey conducted by Gregory Wassail and Neil Alper for the
New England Foundation for the Arts between 1980 and 1982 will be used. For
this report, Massachusetts is the focal point where 8,000 questionnaires were
sent, with a response rate of 1,281 or 16 percent. Of this response, there are data
for 291 painters, sculptors and craftspeople from greater metropolitan Boston.
While this sample is small, it is interesting to compare these findings with those
of the RCAC in its Information on Artists data on Boston, bearing in mind that
the surveys are seven years apart and do not necessarily cover the same universe.
5. One other study needs special mention since it is the only truly longitudinal
study done of fine artists in this country. In 1963 two psychologists from the
University of Chicago, Mihaly Csikszentmihalyi and Jacob W. Getzels,
undertook a study which tested, interviewed and observed at work almost 300
junior and senior fine arts majors at the School of the Art Institute of Chicago.
The purpose of their study of these fine arts majors "was to find out the extent
to which various cognitive abilities... perceptual abilities, values and personality
characteristics are involved in the making of art that is thought to be creative." 2
In 1980 these scholars were able to locate 250 of the original sample of 281,
and administer the first truly longitudinal study of artists. The response was 208,
or 74 percent. This second study focused on "the vicissitudes of creativity in art"
and was also "a study of young people moving into adulthood." 3 It produced an
unpublished report in 1984, Talent and Achievement, which provides important
insights into the very areas tracked in this report: income, employment and
geographical differences, as well as other areas.
Artists Who Work with Their Hands A Trend Report, 1 970 to 1 990 I 63
• Occupation vs. Career. For some artists the ideas of "occupation" and "career"
are not the way they choose to identify themselves. (Indeed, some painters
interviewed for a RCAC project said that in their early painting days in the
1950s, "career" was not part of their professional vocabulary; they simply "were
painters.") Since Census and most other surveys ask artists to "self-identify"
either their occupation, career, or both, this creates difficulties for identifying the
• Professional vs. Amateur. There is a question of who is a "professional artist,"
with disagreement in the art world itself as to what constitutes a professional
artist. The question is also raised of who are "amateur artists," who are also
producers of art. There is no formal Academy to accredit the artist, no equivalent
to the attorney's bar exam to certify him. Even the university degrees of fine arts
are not a universally accepted standard for being an artist.
• Geographic Attitudes. A "big city" bias exists for many, including many artists,
in the art world. In most of the arts the centers of artist activity against which
artists are always comparing themselves are New York and Los Angeles. For the
visual artists addressed here, the primary locus seems to be New York.
Montana painter Karen Kitchel discussed this bias: "The most widespread
assumption in the U.S. art world is that if you're not in New York, you're
nowhere... To simplify the tremendous amount of activity going on outside of
New York City... is a transparent attempt to minimize any accurate sense of the
creative depth, diversity or market in the United States... standards obsessed with
regional hierarchy betray a lack of aesthetic focus and miss the point. They're out
of date and fail to reflect this age of travel and instant information, as well as
artists' diverse living and working arrangements." 4
An application by Arts Midwest to the NEA in 1987 stated: "Visual artists
choosing to live in the Midwest have fewer opportunities for encouragement,
recognition and financial support... A study of artists published in March 1987
by the National Endowment for the Arts indicates that our region of the country
has lost more artists than any other region over the five-year period
researched... Arts Midwest believes that artists do not have to live in New York to
succeed in their profession; although that city may boast an active artistic
climate, it should not be the only place for artists to successfully create and
market their works." 5
Some interesting data supplement these opinions. In the study of fine arts
students at the School of the Art Institute of Chicago previously mentioned, 25
percent of the former students stayed in Chicago, 17 percent lived in Chicago
suburbs and another 18 percent lived in the Midwest. Only half the artists who
lived in New York in 1981 were fully involved in art and "at least one-third had
given up on the practice of art entirely. "In fact," say the study's authors, "moving
to New York is clearly an either-or proposition: those living there are either
committed or quit; very few remain only partly committed to art." 6
64 I Artists in the Work Force
In its 1988 study of 10,000 artists in ten U.S. locations called Information
on Artists, RCAC found that of 4,146 responses, 61 percent of the painters and
58 percent of the craftspeople received art-related training in the city or region
where they currently reside.
Art as a Core Activity. Howard Becker describes some of society's attitudes
toward the artist which complicate economic analyses: "Participants in the
making of artworks, and members of society generally, regard some of the
activities necessary to the production of a form of art as 'artistic,' requiring the
special gifts or sensibility of an artist. They further regard those activities as the
core activities of art, necessary to make the work art rather than (in the case of
objects), an industrial product, a craft item or a natural object... The remaining
activities seem to them a matter of craft, business acumen, or some ability less
rare..." 7 Becker goes on to describe how that core activity called art changes over
time, painting once being regarded as skilled work and then elevated to more
special status during the Renaissance, craft activities being redefined as art.
Artist Focus. Some artists are considered "Creator oriented" with a focus on the
process and creation of the work itself, and others are considered "consumer
oriented." This is not an attempt to judge either the artists or their art, but to
point out that the former state translates economically into limited audiences
and low wages, according to Mary Jean Ryan in her dissertation, In Quiet
Desperation: Professional Strategies of the Aspiring Fine Artist (UCLA, 1985).
Research in psychology has put the economic motivation of so-called
"starving artists" in perspective by suggesting that some fine artists may operate
according to intrinsic rather than extrinsic activity. Joanna Stohs defines intrinsic
motivation: the fine artist "engages in art work for reasons such as satisfaction,
intellectual growth, or emotional or psychological goals (self-fulfillment or
gratification). There are no references to things outside the self (e.g. income or
evaluations by others). The activity is sought because it increases competence or
self-determination or provides inner rewards of personal challenges." 8
Perhaps economic success is not of primary importance to the artists
themselves, or at least to certain kinds of "fine artists." If this is true, measuring
trends in income and jobs for these artists may be relevant compared to other
professions, even other artists' professions, but may not necessarily represent the
Context: Education and Resources. Enormous changes have occurred in
opportunities for education and formal training for visual artists in the country
and in public funding. BFA and MFA Programs in Fine Arts, targeted to all the
artist specialties covered here, were well underway by the end of the 1960s. By
the late 1980s, a Directory of the National Schools of Art and Design listed 164
programs. By the 1960s state arts agencies included every state and U.S. territory,
the National Endowment for the Arts was created, and community and local arts
agencies began strengthening their force. By the 1970s new programs targeting
unthought-of constituencies were added to funding agencies (like the Expansion
Arts Program at the NEA). By the 1980s agencies which funded individual artists
Artists Who Work with Their Hands A Trend Report, 1 970 to 1 990 I 65
(like the New York Foundation for the Arts) joined together to form a seven-state
consortium and received a challenge grant from the NEA. By the 1990s
arguments of freedom of expression, censorship and the relationship between the
government and artists centered around visual artists. Other developments in
resources are mentioned at the beginning of this section.
Other United States Data
Discrete surveys included in this report have a number of common
characteristics. First, they all represent findings from artists in the 1980s. Second,
they all relied on the cooperation of the arts institution community to provide lists
of artists' names. While these lists aid in constructing a profile of people who are
considered artists, they do not identify artists who do not join institutions, apply for
funding or use service organizations. Finally, the studies included here seek to give
some insight into the national as well as the local picture of artists.
For purposes of at least rough comparison with the U.S. Census, painters,
sculptors and craftspeople are analyzed together as one category in three of the four
studies examined by this report (artist printmakers were impossible to identify
according to the definitional categories of the surveys.) Regarding the fourth survey,
The Artists Training and Career Project, results were analyzed for one national study
of painters and another of craftspeople.
United States Census Data, 1970-1990
This research monograph looks at one Census occupation category of visual
artists, which combines painters, sculptors, craft artists and artist printmakers, from
1970-1990. The accompanying tables provide the actual Census trends in
employment, earnings, education and geographic trends by age and gender.
General trends in the experienced civilian labor force from 1970-1990 are
discussed in the General Introduction and include overall growth in the labor force
and trends by gender and age. This report will concentrate on visual artists.
Between 1970 and 1990 the total artist population more than doubled, from
720,000 to 1,671,000. While the male artist population almost doubled (a rise of 46
percent) from 499,000 in 1970 to 931,000 in 1990, the female artist population
tripled from 221,000 in 1970 to 675,000 in 1990.
Female painters/craft artists by 1990 numbered 107,920, accounting for 56
percent of painters/craft artists. This compares to their percentages among all artists,
up 7 percent from 1970 to 1980 and another 3 percent from 1980 to 1990. From
1970 to 1980 there was over 7 percent point rate of growth for women artists,
although the growth rate of female workers slowed after 1980. For male painters/craft
artists, rate of growth decreased from 64 percent in 1970 to 52 percent in 1980 and
to 44 percent in 1990 (Table 1).
In 1970 painters/craft artists totaled 102,600; in 1980, they totaled 151,360, 14
percent of all artists; by 1990 they totaled 191,160, or 13 percent of all artists, the
second largest of all artist occupations, representing .174 percent of the total labor force.
66 I Artists in the Work Force
GROWTH IN ARTIST OCCUPATION BY GENDER,
1970 TO 1990 (numbers in thousands)
Sources: Painters/craft artists are
All Artists are
Notes: Sums for men and women may not add to totah
i due to round
According to Deirdre Gaquin 9 between 1980 and 1985, 41 percent of the
population changed their place of residence. She warns that statistics for 1990 in this
area are unreliable due to a change in urban/rural distinctions by the Census, so the
following figures on artists and painters/craft artists are cited with caution.
In 1980, 86 percent of male artists and 85 percent of female artists were urban
residents. By 1990, 78 percent of male artists and 72 percent of female artists were
urban residents. Urban dwellers among male painters and craft artists went from 90
percent in 1970 to 86 percent in 1980 to 80 percent in 1990.
Urban dwellers among women painters and craft artists also declined, going from
86 percent in 1970 to 85 percent in 1980 to 76 percent in 1990, renewing discussion
on whether artists need to work in the same geographic area where their work is sold.
Regarding distribution of artists by Census region, male artists by 1990 had
highest proportions residing in the West; females had highest proportions residing in
the South. Male artists had increases in the West, up to 30 percent and male painters
and craft artists had up to 28 percent increases in the West, where their highest
proportions were. Their most substantial decrease was in the Northwest.
Female painters and craft artists had highest proportions residing in the South,
and their most substantial decrease was in the Northeast (Table 2).
Artists Who Work with Their Hands A Trend Report 1 970 to 1 990 I 67
REGION OF THE EXPERIENCED CIVILIAN LABOR FORCE, PROFESSIONAL
SPECIALTY WORKERS, AND ALL ARTISTS BY DETAILED ARTIST OCCUPATION & GENDER
Total ECLF age 16+
Painters, Sculptors, etc.
Total ECLF age 16+
Painters, Sculptors, etc.
Sources: All Artists and Painters,
Sculptors Etc. tor 1990 is Greenblatt.
Data tor 1970/1980 are Citro &
Gaquin. 1990 Professional Specialty Workers in J
rom Gaquin. All other data is t'rom the U.S. Statistical
68 I Artists in the Work Force
Trend data for age patterns is particularly difficult to assess with confidence for
several reasons: the definitional changes of artists categories, the relatively small
sample size for each occupation and the challenges presented in the first section of
Between 1 970 and 1 990 baby boomers entered the labor force and male workers
declined in labor force participation, possibly due to their choosing early retirement
and the longer tenure of females in the job force.
Male artists were similar (but younger on average) in 1990 to all male workers
with their median age dropping from 37-38 in 1970 to 34 in 1980 and rising to 37
in 1990. Male painters and crafts artists had higher median ages than the general
labor force in 1990, but were closer to male professionals, moving from a median age
of 39 in 1970 to 36 in 1980 and 40 in 1990.
For all male artists the largest age group was the 25-34 year olds with 31 percent.
For male painters and craft artists, the 35-44 year old group was largest, with 29
percent (Table 3).
Female artists were similar (but younger on average) in 1990 to all female
workers, with their median age dropping from 37-38 to 33 in 1980 and rising to 37
in 1990. For all female artists the largest age group was 25-34 year olds at 34 percent,
and for female painters and craft artists the 35-44 year old group was largest with 31
percent (Table 3).
AGE AND EDUCATION OF PAINTERS/CRAFT-ARTISTS,
BY GENDER 1970 TO 1990
Percent with Education
Sources: Data from 1970/1980 are from Citro & Gaq
uin. 1990 data
The post World War II availability of education in general, and for artists in
particular, raised the educational profile of many Americans. For artists, although
Artists Who Work with Their Hands A Trend Report, 1970 to 1990 I 69
their educational levels were higher than the general work force, they were below all
professionals and showed smaller percentage point increase in proportions with
In 1990, 40 percent of male artists had completed four or more years of college
compared with 76 percent of male professionals. Forty percent of female artists had
completed four or more years of college compared with 66 percent of female
professionals. For male painters and craft artists by 1990, 37 percent had four or
more years of college while 43 percent of female painters and craft artists had four or
more years of college. Statistics on education suffer from a change in the wording of
the Census questions and are to be viewed with caution. Also, the discrete studies on
painters/craft artists provide a different profile.
Employment and Earnings
Deirdre Gaquin summarized some striking trends characterizing patterns of
employment and earnings after World War II. During those years men in the labor
force trended away from self-employment and toward working for private employers
and government. Women workers were also increasingly attracted to public sector
employment and more workers, particularly among women, were employed year-
round. Median earnings for men increased 75 percent and for women, 43 percent,
from 1950 to 1970. After 1970, real earnings adjusted for inflation declined sharply,
particularly among women, whose median earnings remained less than half those of
men throughout this period. 10
The trend in employment sectors for artists differs from that of the general labor
force in the post WW II years, as artists moved toward self employment during this
time. In 1970, 67 percent of male artists and 69 percent of female artists worked in
private firms; 10 percent of male artists and 8 percent of female artists worked in
government, and 23 percent of males and 21 percent of females were self-employed.
In 1980 self-employment was on the rise, with 60 percent of male artists and 65
percent of female artists working for private firms, 7 percent of both male and female
artists working for government, and 32 percent male and 28 percent female artists
self-employed. By 1990 self-employment for artists had increased slightly with 33
percent of males and 31 percent of female artists self-employed, compared to 62
percent of male artists and 64 percent of female artists working for private firms and
5 percent of both males and females working for the government.
Painters and craft artists show still higher percentages of self employment, so that
by 1990 nearly half of males and females were self employed. In 1970, 62 percent of
male painters and craft artists and 58 percent of female painters/craft artists worked
for private firms, 7 percent of males and 8 percent of females worked for government
and 32 percent of males and 34 percent of females were self-employed. In 1980, 48
percent of both male and female painters/craft artists worked for private firms, 7
percent of both worked for government and 45 percent of males and 42 percent of
females were self-employed. By 1990, 48 percent of male painters/craft artists and 45
percent of females worked for private firms, 5 percent of males and 4 percent of
70 I Artists in the Work Force
females worked for the government and 47 percent of males and 49 percent of
females were self-employed.
In terms of unemployment according to Census definitions, unemployment
declined for both male and female painters and craft artists from 1980 to 1990 after
a rise from 1970 to 1980.
For male artists, unemployment declined by almost 2 percent from 1970 to
1990, while female artists' unemployment declined by a full 2 percent. For male
painters/craft artists, unemployment growth was less than 1 percent, but for females
unemployment declined between 1970 and 1990 by less than 1 percent.
Problems arise in measuring year-round employment for artists, and even more
so when it comes to painters and craft artists, since the nature of employment (1)
often is not attached to a single employer; (2) is not always characterized as
"employment" — the sale of a painting is not employment; (3) is not always
measurable by the standards of full-time and part-time work the Census uses (many
artists are literally always working) and (4) does not often translate into eligibility for
unemployment benefits (because the artist's work time cannot be verified by an
"employer.") The artist may be self-employed, but he also may earn most of his
income at other work such as teaching, and declare his art income as "other."
That females earn less than males in all sectors is obvious throughout the 1970-
1990 period. (Table 4). Male painters' and craft artists' median incomes fall below
the medians for the total labor force and professionals. Female painters/craft artists'
median earnings are either close to or above those of the total female work force.
The median income for male artists grew from $8,768 to $14,219 between 1970
and 1980 and to $21,600 by 1990. For male painters and craft artists the median
income grew from $8,893 in 1970 to $12,684 in 1980 and to $18,187 in 1990.
For female artists the median income was $3,637 in 1970, $6,712 in 1980 and
$11,096 in 1990. For female painters and craft artists, the median was $3,682 in
1970, $6,612 in 1980 and $22,041 in 1990 (more than triple the 1980 figure).
For those male artists who worked between 50 and 52 weeks per year, the median
income almost tripled from $9,550 in 1970 to $27,961 in 1980 and went to $31,124
in 1990. For male painters and craft artists the median income grew more slowly
from $9,672 in 1970 to $15,112 in 1980 and to $24,320 in 1990.
Female artists enjoyed a four-fold increase in median income from $4,152 in
1970 to $17,328 in 1980 and then a smaller increase to $20,825 by 1990. For female
painters and craft artists, the median went from $5,347 in 1970 to $9,344 in 1980
and to $18,762 in 1990.
Full-year male painters and craft artists' median earnings exceeded that of their
part-year colleagues, but full-year female painters and craft artists' median income
In summary, by 1990 there was a larger proportion of women as painters,
sculptors, craft artists and artist printmakers, with fewer living in urban areas. They
had a higher median income than all artists and the general labor force, but were
closer in median age to professionals. For both males and females, their level of
Artists Who Work with Their Hands A Trend Report, 1 970 to 1 990 I 71
EARNINGS OF THE EXPERIENCED CIVILIAN LABOR FORCE,
BY DETAILED ARTIST OCCUPATION AND GENDER
Total ECLFage 16 +
Professional Specialty Occup.
All Artists (3)
Painters, Sculptors, etc.
♦Total ECLFage 14+
Professional Specialty Occup.
All Artists (3)
Painters, Sculptors, etc.
*Please note change in ECLF age
for females. (1) From Beresford Table 1 . (2) From
Statistical Abstracts, 1992, Table
656, page 414.
(3) Beresford Tables 5 &6 and
Citro & Caquin.
72 I Artists in the Work Force
education, according to the Census, seemed to be holding steady or rising slightly,
and many more were self-employed, with percentages much higher than other kinds
of artists. The median income for male painters and craft artists grew more slowly
than for the total male work force, male professionals and female painters and craft
artists (whose median income more than tripled since 1980). Finally, using figures
that should be scrutinized further, it appears that part-year female painters and craft
artists earned more than their full-year counterparts.
The Research Center for Arts and Culture conducted the three studies that
follow, searching for an understanding of what artists do which is closer to their own
perception than the Census. In addition to asking which occupation provided the
artist's major income and number of hours worked, the survey asked:
• the occupation that is primary to the respondent
• the occupation that is most important to the respondent
• the major area of concentration
• If the respondent considers himself/herself to be a professional artist.
In the three RCAC surveys, 93 percent of the respondents to The Artist's Work-
Related, Human and Social Services Questionnaire consider themselves professional
artists; in Information on Artists, 89 percent consider themselves professional artists
and in the Artists Training and Career Project, 91 percent of the painters and 86
percent of the craftspeople consider themselves professional artists.
A fourth study, the Artists and Jobs Questionnaire, commissioned by the New
England Foundation on the Arts and done by Wassail, Alper and McCabe, is based
on more traditional Census-based definitions.
The Artist's Work- Related, Human and Social Services Questionnaire
The major area of concentration in this study is listed as painting, sculpture and
crafts, but since 86 percent of the 164 respondents claimed the occupation that is
primary to them is "artist," and since there are relatively small numbers to begin with,
these findings must be viewed with caution.
This was a pilot study for the Research Center for Arts and Culture. Two of its
contributions were broadening the base of investigation to a larger geographic area
and the realization that studies which isolated particular types of artists would allow
for a much more specific investigation.
• Age and Gender. The mean age for these artists was 38 (standard deviation
9.969) and the median, 36. (The mean is used as well as the median age in this
chapter because in most cases the difference of two or more years is significant.)
Half were male and half female.
Artists Who Work with Their Hands A Trend Report 1 970 to 1 990 I 73
• Education. Eleven percent of these artists had some college; 23 percent had at
least four years of college and 63 percent had some graduate education. By
gender, 29.1 percent of males and 16.1 percent of females had four years of
college; 55.6 percent of males and 69.5 percent of females had some graduate
• Income. Twenty-five percent of these artists earned $500 or less from their art in
1985 while 12 percent earned over $20,000 from their art. By gender, 28.6
percent males and 21.4 percent females earned $0-500; 26.8 percent males and
24.9 percent females earned between $501 and $3,000; 18.8 percent males and
21.5 percent females earned between $3,001 and $7,000; 3.6 percent males and
10.7 percent females earned between $7,001 and $12,000; 9 percent males and
11.6 percent females earned between $12,001 and $20,000 and 12.5 percent
males and 9.8 percent females earned over $20,000.
Information on Artists Survey (1988)
Data presented here are for all artists and for the category of painters, sculptors
and craftspeople for age, education, income and art-related costs. These same
breakdowns are applied to Boston and New York, two of 10 sites surveyed. Tables 5
through 8 show data for all artist respondents from the broad variety of art fields used
for the study. The text below will focus on the subject of this report, artists who work
with their hands.
Age. The mean age for painters, sculptors and craftspeople is 38.6, the median, 37.
Education. Table 5 displays the educational attainment of these artists. When
INFORMATION ON ARTISTS
"Painters, et. al"
All Artists: Boston
All Artists: New York
Painters: New York
74 I Artists in the Work Force
broken down by gender, of male painters, sculptors and craftspeople, 35.5
percent have college degrees and another 44.8 percent have graduate degrees. For
females, 45.7 percent have college degrees and another 41.3 percent have
Income. Table 6 shows total income as an artist and total gross income in 1988.
For male painters, sculptors and craftspeople, individual income as an artist
shows 25.4 percent earning $500 or less; 26.6 percent earning berween $501 and
$3,000; 14.3 percent between $3,001 and $7,000; 9.9 percent between $7,001
and $12,000; 7.9 percent between $12,001 and $20,000; 9.7 percent between
$20,001 and $40,000 and 6.2 percent over $40,000.
For female painters, sculptors and craftspeople, individual income as an
artist shows 26.9 percent earning $500 or less; 33.7 percent earning between
$501 and $3,000; 15.8 percent between $3,001 and $7,000; 9.7 percent
between $7,001 and $12,000; 6.9 percent between $12,001 and $20,000; 5.3
percent between $20,001 and $40,000 and 1.7 percent over $40,000. (Totals
may not equal 100 percent due to rounding.)
Total Gross Income. For male painters, sculptors and craftspeople, total gross
income is: 7.1 percent earning $5,000 or less; 12.3 percent earning between
$5,001 and $10,000; 32 percent between $10,001 and $20,000; 20.6 percent
between $20,001 and $30,000; 15 percent between $30,001 and $40,000 and
13 percent over $40,000, similar findings to those for all artists in this study.
For female painters, sculptors and craftspeople, total gross income in 1988
is: 13.8 percent earning $5,000 or less; 21.2 earning between $5,001 and
$10,000; 33.5 percent between $10,001 and $20,000; 17 percent between
$20,001 and $30,000; 8.8 between $30,001 and $40,000 and 5.7 percent over
$40,000. (Totals may not equal 100 percent due to rounding.)
Artists' Costs. Relevant to earnings are the costs for space to work and
maintenance of one's craft. About two-thirds of the respondents answered
questions on costs. In terms of monthly costs for space, over two-thirds of the
artists, including painters/craftspeople, paid under $500 per month for
workspace in 1988. (See Table 7.)
Annual art-related expenses (excluding workspace) include art supplies and
services, tools and equipment, capital improvements, training and maintaining
their craft, publicity, marketing, travel and shipping. For painters, sculptors and
craftspeople, 86.2 percent of males and 78.6 percent of females spent under
$500 on their annual art-related expenses in 1988. Another 10.3 percent of male
painters, sculptors and craftspeople and 19.5 percent of females spent between
$501 and $2,500. Thus in 1988 over 89 percent of these artists spent $2,500 or
less on their art-related expenses.
Comments and informal information indicate that, especially for visual
artists in need of expensive equipment like kilns, often their "other"employment
(i.e. teaching) fills this need. This information is reinforced by the Artists and Jobs
Questionnaire done in New England in the early 1980s.
Artists Who Work with Their Hands A Trend Report, 1 970 to 1 990 I 75
IOA TOTAL INCOME AS AN ARTIST
"Painters, et. al"
All Artists: Boston
All Artists: New York 20.95
Painters: New York
IOA TOTAL GROSS INCOME FOR 1988
"Painters, et. al"
All Artists: Boston
All Artists: New York 5.20
Painters: New York
1 1 .40
76 I Artists in the Work Force
SCULPTORS, & CRAFTSPEOPLE
Costs of Workspace
"Painters, et. al"
Painters et al
Painters et al.
Costs: Annual Cost of Training & Maintaining Artwork
Painters et al.
Painters et al.
Painters et al.
Artists Who Work with Their Hands A Trend Report, 1 970 to 1 990 I 77
• Professionalism. Eighty-nine percent of all artists in this survey consider
themselves to be professional artists. To gain a better understanding of how
artists view professionalism, a three-way division was used which included both
external and self- assessment criteria. The groups of these definitions were done
after the data were collected to identify three main areas:
1 . The Marketplace Definition.
The person makes his/her living as an artist.
The person receives some income from his/her work as an artist.
The person intends to make his/her living as an artist.
2. The Education and Affiliation Definition.
The person belongs to an artists' association (discussion group, artists' group,
artists' co-op, etc.)
The person belongs to an artists' union or guild.
The person has been formally educated in the fine, creative, literary or
3. The Self and Peer Definition
The person is recognized by his/her peers as an artist.
The person considers himself/herself to be an artist.
The person spends a substantial amount of time working at art.
The person has a special talent.
The person has an inner drive to make art.
The person receives some public recognition for his/her art.
The above criteria were used in two questions, one which asked respondents to
identify their three most important choices in rank order in considering "someone to
be a professional artist," and another, similarly ranked, in which these "reasons apply
to you." The figures below show the overwhelming first choice in the Self-Definition
IOA: IMPORTANT CRITERIA FOR PROFESSIONAL ARTIST
(When considering someone else to be a professional)
Painters Craft Artists All Respondents
Market Definition 18% 26.1% 23.1%
Peer/Educ Definition 10.3% 12.5% 12.5%
Self-Definition 71.7% 61.4% 64.4%
IOA: IMPORTANT CRITERIA FOR PROFESSIONAL ARTIST-Self
Painters Craft Artists All Respondents
Market Definition 16.1% 36.7% 22.8%
Peer/Educ Definition 7.2% 7.6% 9.3%
Self-Definition 76.8% 55.7% 68.0%
78 I Artists in the Work Force
Information on Artists: Boston and New York
Boston had an artist population of 20,839 in 1980 according to the Census. The
'80s were a time for increased public funding through the Massachusetts Council for
the Arts and Humanities and progress on many levels for artists. The number of
artists of all kinds surveyed in Boston total 350; of those 157 were
painters/sculptors/craftspeople, the focus of the data below.
• Age. The mean age for all Boston painters, sculptors and craftspeople is 36.4; the
median age, 35.
• Education. For male Boston painters, sculptors and craftspeople, 36.8 percent
have college degrees, 47.4 percent have graduate degrees; for females 52.8% have
college degrees, 40.3 percent have graduate degrees (Table 5).
• Income as Artists. For male Boston painters, sculptors and craftspeople,
individual earnings as an artist look like this for 39 respondents: 28.2 percent
earning $500 or less; 30.8 percent earning between $501 and $3,000; 17.9
percent earning between $3,001 and $7,000; 15.4 percent earning between
$7,001 and $12,000; 2.6 percent earning between $12,001 and $20,000; 5.1
percent earning between $20,001 and $40,000 and no one earning over $40,000
For female Boston painters, sculptors and craftspeople, individual earnings
look like this for 73 respondents: 26 percent $500 or less; 42.5 percent earning
between $501 and $3,000; 13.7 percent earning between $3,001 and $7,000;
6.8 percent earning between $7,001 and $12,000; 6.8 percent earning between
$12,001 and $20,000; 4.1 percent earning between $20,001 and $40,000 and
no one earning over $40,000."
• Total Gross Income. For 37 male Boston painters, sculptors and craftspeople,
total gross income in 1988: 5.4 percent earned less than $5,000; 8.1 percent
earned between $5,001 and $10,000; 45.9 percent between $10,001 and
$20,000; 21.6 percent between $20,000 and $30,000; 10.8 percent between
$30,000 and $40,000 and 8.1 percent over $40,000 (Table 6).
For 72 female Boston painters, sculptors and craftspeople, total gross income
in 1988: 8.3 percent earned less than $5,000; 23.6 percent earned between
$5,001 and $10,000; 33.3 percent between $10,001 and $20,000; 16.7 percent
between $40,000 and $30,000; 9.7 percent between $30,001 and $40,000 and
8.3 percent over $40,000.
• Artists' Costs. Of 21 male Boston painters, sculptors and craftspeople and 41
females, virtually all pay less than $2,500 in annual art-related expenses. Over
three-quarters of male and female members of these professions pay less than
$400 a month for workspace (Table 7).
The artist population of New York City, according to the 1980 U.S. Census, was
112,954. Of these 15,640 were painters, sculptors, craft artists and artist
printmakers. The number of all artists surveyed in New York totals 485, of which 290
Artists Who Work with Their Hands A Trend Report 1 970 to 1 990 I 79
were painters, sculptors and craftspeople. The 1980s saw a huge infusion of money
in the for-profit art market arena with prices soaring for visual art. There was an
explosion of East Village galleries and commercial spaces in Soho, Noho and Tribeca
as the line between profit and nonprofit spaces became thinner. Museums asked
avant-garde artists to donate one-of-a-kind objects for reproduction, coming
perilously closer to commercial activity. Non-mainstream museums collaborated on
shows featuring a range and breadth of artists new to many New Yorkers. New York
was considered by many the seat of the art market.
• Age. The mean age for all New York painters, sculptors and craftspeople is 37.6
percent; the median age is 36.
• Education. Table 5 details the spread between males and females for college
degrees for painters, sculptors and craftspeople, with 30 percent of the males and
46.8 percent of the females having college degrees. At the graduate level 53.8
percent of the males have graduate degrees, and 47.6 of the females.
• Income as Artists. Individual earnings as an artist look like this for 79 male New
York painters, sculptors and craftspeople: 25.3 percent earning $500 or less; 32.9
percent earning between $501 and $3,000; 12.7 percent earning between $3,001
and $7,000; 10.1 percent earning between $7,001 and $12,000; 8.9 percent
earning between $12,001 and $20,000; 8 percent earning between $20,001 and
$40,000 and 1.3 percent earning over $40,000.
For 126 female New York painters/craftspeople, individual earnings look like
this: 24.6 percent earning $500 or less; 33.3 percent earning between $501 and
$3,000; 17.5 percent earning between $3,001 and $7,000; 6.3 percent earning
between $7,001 and $12,000; 10.3 percent earning between $12,001 and
$20,000; 6.3 percent earning between $20,001 and $40,000 and 1.6 percent
earning over $40,000. 12
• Total Gross Income. For all 79 male New York painters/craftspeople, total gross
income in 1988: 7.6 percent earned less than $5,000; 11.4 percent earned
between $5,001 and $10,000; 34.2 percent between $10,001 and $20,000; 20.3
percent between $20,001 and $30,000; 17.7 percent between $30,001 and
$40,000 and 8.4 percent over $40,000.
For all 127 female painters/craftspeople, total gross income in 1988: 9.4
percent earned less than $5,000; 18.9 percent earned between $5,001 and
$10,000; 32.3 percent between $10,001 and $20,000; 21.8 percent between
$20,001 and $30,000; 14.2 percent between $30,001 and $40,000 and 3.9
percent over $40,000.
• Artists' Costs. Of 44 male New York painters, sculptors and craftspeople and 59
females, virtually all pay less than $2,500 in annual art-related expenses. Of 77
males, over half pay less than $400 a month for workspace, while over half of the
118 females pay less than $400 per month for workspace (Table 7).
Artists Training and Career Project (ATC)-1 990-91
• Age, Gender, Ethnic Background. The mean age for all painters and
craftspeople from this 1990-91 survey, to which 960 painters and 1,301 craft
80 I Artists in the Work Force
artists responded, is 43.1 for painters and 43.4 for craftspeople. (Standard
deviation 11.5-12.3.) The median age is 41 for both painters and craftspeople.
As to gender, 58 percent of the painters and 54 percent of the craftspeople are
female, while 42 percent of the painters and 46 percent of craftspeople are male.
Median age for male painters is 38 and for craftsmen, 42. Median age for female
painters in 40 and for craftswomen, 41.
Regarding ethnic background, for painters 86 percent are white, fewer than
2 percent are American Indian, more than 2 percent are Asian, 3 percent are
black, fewer than 2 percent are Hispanic and 5 percent are "other." For
craftspeople, 92 percent are white, fewer than 1 percent American Indian, 1
percent each Asian and Hispanic, more than 1 percent black, and 5 percent
Education. Regarding degrees, 40.6 percent of the painters have a college degree
and 42.5 percent a graduate degree; 38.3 percent of the craftspeople have a
college degree and 33.6 a graduate degree. By gender, 33.4 percent of male
painters and 45.7 percent of women painters, and 31.5 percent of male
craftspeople and 45.2 percent of female craftspeople have a college degree; 45.6
percent of male painters, 34.2 percent of male craftspeople, 40.2 percent of
female painters and 33 percent of female craftspeople have graduate degrees
Income as Artists. For painters in 1990, 56 percent of the males and 62 percent
of the females earned less than $3,000 as artists, and 65 percent of the males and
78 percent of the females earned less than $7,000 as artists. For craftspeople in
1989, 36 percent of the males and 39 percent of the females earned less than
$3,000 as artists and 47 percent of the males and 53 percent of the females
earned less than $7,000 as artists (Table 8).
Total Gross Income. For male painters in 1990, 8.6 percent earned under
$5,000; 13.3 percent earned between $5,001 and $10,000; 25.1 percent earned
between $10,001 and $20,000; 20.2 percent earned between $20,001 and
$30,000; 14.4 percent earned between $30,001 and $40,000; 16.7 percent
earned between $40,001 and $60,000 and 1.7 percent earned over $60,000.
For female painters in 1990, more than one third earned under $10,000;
26.7 percent earned between $10,001 and $20,000; 19.8 percent earned
between $20,001 and $30,000; 10.8 earned between $30,001 and $40,000; 7.5
percent earned between $40,001 and $60,000 and 1.3 percent earned over
For male craftspeople in 1989, 11.5 percent earned under $5,000; 6.1
percent earned between $5,001 and $10,000; 14.3 percent earned between
$10,001 and $20,000; 17 percent earned between $20,001 and $30,000; 19.1
percent earned between $30,001 and $40,000, 32 percent earned over $40,000.
For female craftspeople in 1989, 19.8 percent earned under $5,000; 14.3
percent earned between $5,001 and $10,000; 21.8 percent earned between
$10,001 and $20,000; 18 percent earned between $20,001 and $30,000; 11.1
percent earned between $30,001 and $40,000, 15 percent earned over $40,000.
Artists Who Work with Their Hands A Trend Report, 1970 to 1990 I 81
ARTISTS TRAINING & CAREER PROJECT: PAINTERS & CRAFTS ARTISTS
Craft Artists (total)
TOTAL INCOME AS AN ARTIST:
$0- $501 $3,001
500 3,000 7,000
27.90 29.00 15.70
22.80 27.10 15.10
31.60 30.30 16.20
Craft Artists (1989)
20.50 18.10 12.60
21,80 14.40 11.10
18.20 21.20 13.80
TOTAL GROSS INCOME FOR 1988:
Craft Artists (total)
GROSS HOUSEHOLD INCOME FOR 1988:
Craft Artists (total)
Table 9 shows total gross household income for these artists.
• Professionalism. In this survey 91.2 percent of the painters and 85.7 percent of
the craftspeople consider themselves professionals. (See Information on Artists
Survey on professionalism discussed earlier for criteria and definitions.)
Artists and Jobs Questionnaire-1980
This study commissioned by the New England Foundation for the Arts will be
briefly summarized here. Its findings are for 291 painters, sculptors and craftspeople
from the Boston area.
82 I Artists in the Work Force
ARTISTS AND JOBS QUESTIONNAIRE (1980)
Income as Artists
• Age and Gender. The mean age for 287 Boston painters/craftspeople is 37.3; the
median is 34. For males, the mean age is 36.3; the median, 33. For females, the
mean is 38.7; the median, 35.5
• Education. For male Boston painters/craftspeople 8.4 percent have a high school
education; 28 percent some college; 45.8 percent a college degree and 57.9 percent
a graduate degree. For females, 16.5 percent have some college; 47.4 percent a
college degree and 62.9 percent have a graduate degree.
• Income. Income as artist and total gross income for Boston area artists is seen in
Tables 6 and 7 and will not be detailed separately here. Analysts of the findings
on income seemed shocked at what artists earn, commenting "Find out why
these artists earn so little." Proof of such meager earnings: for half the males and
almost one- third of these female artists worked a full year to earn under $500
from their art.
Why is there a discrepancy between the findings of all these discrete surveys and
the Census in the area of education? Recalling the definitional problems of the
Census and examples of tattoo artists and cardpainters being included as painters,
one can see how the numbers become inflated with people whose data cannot be
accurate markers for a large segment of the arts, whose information seems to be out
of sync with everything else known. In fact, most studies of artists during the last 10
years, outside of the Census, have clearly established the high degree of formal
education, as well as its lack of corresponding income.
Artists Who Work with Their Hands A Trend Report, 1 970 to 1 990 I 83
Figures that emerge from non-Census data about painters and craft artists
described here are closer to the Census figures for Professional Specialty Occupations.
Information gathered from every independent U.S. study reviewed here indicates a
huge discrepancy between what researchers have identified directly from artists, and
results from the 1990 Census. This is an area warranting substantial further
Another important area for inquiry, and one that has rarely been addressed, is the
income of artists who have abandoned art as an occupation compared to those who
have stuck with it. In Talent and Achievement, the authors report that "for both men
and women, the household income of those who had abandoned fine art by mid-life
is higher than the income of those still involved." 13 In addition, the range of
individual earnings for those artists who remained involved in fine art was $500-
$80,000, "either a feast or a famine." 14 The ranges in Research Center Studies were
Many characteristics of artists emerge which have been explored by independent
researchers and which bear continued attention. The authors of Talent and
Achievement note that of the artists they studied, "at least since their early twenties,
young people interested in art show a remarkable determination to shape their own
destiny." 15 In addition, they comment on art as a profession: "Art differs from other
occupations in that artists must find their jobs within themselves... the modern artist
is expected to develop the content and the rules of his profession from within.
External signposts are few, and ambiguous..." 16 By focusing solely on measures like
earnings and education, analysis of artists is limited to "conventional goals of
affluence and status" in the "roles prescribed by society. " r
What the discrete surveys offer is another view targeting the artist population
more narrowly than the Census and suggesting additional ways of looking at how
artists view their occupations. These surveys also identify other areas of inquiry that
broaden the picture of the artist in society. Finally, they indicate the need for a regular
survey of artists, if possible, by the National Endowment for the Arts, which
combines the more relevant aspects of the Census with other areas of inquiry, some
of which have been identified in this document.
Further investigation is needed to compare other data sets and sources to the
Census findings, and to provide a broader landscape in which to think about artists
in ways which are valuable to society, the government and to the artists themselves.
1. Judith Adler, "Artists Job Market Experiences," journal of Arts Management and
Law, 13:3 (1983), pp. 177-182.
2. Mihaly Csikszentmihalyi, Jacob W. Getzels and Stephen P. Kahn, Talent and
Achievement (Chicago, 1984) an unpublished report, p. 1.
3. ibid. p. 10.
84 I Artists in the Work Force
4. Karen Kitchel, "Speakeasy." New Art Examiner, Summer 1992, pp. 13-15.
5. C. Lynn Cowan, "The Artists' Condition from the Regional Perspective," in
C.Richard Swaim (ed.), The Modern Muse: The Support and Condition of Artists
(New York: ACA Books, 1989), pp. 33-45.
6. Mihaly Csikszentmihalyi, Jocob W. Getzels and Stephen P. Kahn, Talent and
Achievement (Chicago, 1984) an unpublished report, p. 44-45.
7. Howard Becker, Artworlds. (Los Angeles: University of California Press, 1982),
8. Joanna H. Stohs, "Intrinsic Motivation and Sustained Art Activity Among Male
Fine and Applied Artists," Creativity Research Journal, 1992, Vol. 5, p. 247.
9. Gaquin, Deidre, Constance Citro. Artists in the Workforce, 1950 to 1985. Research
Division of the NEA, Washington DC. p. III-2.
10. ibid, p. V-l.
11 and 12. Numbers may not equal 100 percent due to rounding. The small
numbers in this analysis must be taken with extreme caution; they are included
here for their site-specific purposes, and because comparisons with larger studies
indicate directions for the future.
13. Mihaly Czikszentmihalyi, Jacob W. Getzels and Stephen P. Kahn, Talent and
Achievement (Chicago, 1984) an unpublished paper, p. 305.
14. Ibid, p.306.
15. 16 and 17. Ibid. 483.
About the Authors
Joan Jeffri is the founder and director of the Research Center for Arts and
Culture at the School of the Arts at Columbia University, and the coordinator of the
Graduate Program in Arts Administration at Teachers College, Columbia University.
She is the author or editor of numerous studies on individual artists, as well as texts
in the arts management field, including Artsmoney: Raising It, Saving It, and Earning
It (University of Minnesota, 1989) and The Emerging Arts: Management, Survival,
and Growth (Praeger, 1980). Her books on artists include a series of interview books:
The Actor Speaks, The Craftsperson Speaks and The Painter Speaks (Greenwood, 1992,
1993, 1994). For nine years she was an Executive Editor of The Journal of Arts
Management and Law.
Robert Greenblatt has worked as a consultant in communications, computer
resources, survey design and analysis, and systems design for a diverse group of clients
including Columbia University's Research Center for Arts and Culture, the American
ORT Federation and the Palacio de Queluz in Portugal. He has taught mathematics,
political science and quantitative research at Pace University, Brooklyn College, the
State University of New York and Cornell University. He has a Ph. D. in
Mathematics from Yale University.
III. Employment and Earnings of
Performing Artists, 1 970 to 1 990
by Ann O. Kay and Stephyn G. W. Butcher
Artists, as a group, participate in the economy in a way unparalleled by the rest
of the work force. Most workers are involved in creating the daily necessities of
life. But artists are different. They participate in the creation of the culture. From
architects to dancers, authors to actors, sculptors to musicians, no other group's
impact is more immediate and pervasive, creating work spaces, homes, music, books,
magazines, ads, artwork, television programs and movies.
The significance of artists in the work force increased dramatically from 1970 to
1990. In 1970 artists numbered 736,960 or 0.9 percent of the experienced civilian
labor force (Table 1). By 1980 artists' numbers had jumped to 1,085,693, growing
EXPERIENCED CIVILIAN, PROFESSIONAL SPECIALTY, ARTIST AND
PERFORMING ARTIST LABOR FORCES. 1970, 1980, AND 1990.
Experienced Civilian Labor Force (ECLF)
Professional Specialty Occupations
Performing Artist Occupations
Professionals as a percent of ECLF
1 1 .0%
1 1 .8%
Artist Occupations as a percent of ECLF
Artist Occupations as a percent of Professi
Performing Artists as a percent of Artists
Source: Ellis and Beresiord, 1994.
47.3 percent over the decade. Part of this increase reflected the overall labor force
growth due to baby boomers continuing to enter the labor force, .md an increased
number of women working. But the artist growth rate also reflected the increasing
importance of the professional specialty occupations and the service sector in the
U.S. economy. This trend continued during the 1 c )S0s. boosting the artist
occupation numbers by 4.4 percent a year | Table 2). By 1990 there were more artists
as a whole and as a proportion of the experienced civilian labor force And professional
specialty occupations. The mixture of artist types changed during these two decades.
Performing artists actors directors, musicians/composers .\nd dancers - fell from 23
percent of all artists in 1 970 to IS percent bv 1990. During this time, "word" artists
and visual and design artists occupations grew (aster. However, the number o(
performing artists still increased at ,\n annual average rate of 3.3 percent from L970
86 I Artists in the Work Force
GROWTH RATES OF THE EXPERIENCED CIVILIAN, PROFESSIONAL
SPECIALTY, ARTIST & PERFORMING ARTISTS LABOR FORCES.
1970 TO 1990
Dverall Growth Rate
Experienced Civilian Labor Force (ECLF)
Professional Specialty Occupations
Performing Artists Occupations
Average Annual Growth Rate
Experienced Civilian Labor Force (ECLF)
Professional Specialty Occupations
Performing Artists Occupations
Growth rates calculated from labor force data
to 1990, surpassing the growth rate of the national and professional specialty labor
Data Sources and Conceptual Issues
This report examines trends in the employment and earnings of performing
artists and their geographic distribution from 1970 to 1990. It uses three sources for
data: Census, survey and pension data. Each of these data sources has advantages and
disadvantages which are discussed in the general introduction to this book and will
not be repeated here.
The Census data used is not from the Census form sent to most households, but
from the Census long form questionnaire which is sent to about 15.9 percent of U.S.
households. The long form has detailed questions about work history: occupation,
sector, industry, last job held, current labor force status, weeks worked last year, usual
work hours last year and detailed income from last year. The Census makes this data
available to the public in the form of various sample subsets which include all the
information except that which would identify the individual or household that
Most of the analysis in this report is based, directly or indirectly, on the 5 percent
PUMS or "Sample A" containing information on the population and households
representing 5 percent of the total U.S. population (a subset of the 15.9 percent who
received the Census long form). Since the 1980s, The National Endowment for the
Arts has constructed "Artist Extract Files" from these PUMS files created by the
Census. The Artist Extract Files (AEF) are subsets of the PUMS that cover all those
persons in the PUMS who listed their occupation as one in the arts. The data for
1970 in this report are based on Muriel Cantor's Employment Status of Performing
Artists: 1970 to 1980 (Cantor, 1987) which used the AEF files. The 1980 and 1990
data are from the authors' analyses of the AEF files. In all cases the actual counts
Employment and Earnings of Performing Artists, 1970 to 1990 I 87
which are the result of samples have been "normalized" to the actual Census counts
for the occupation as a total.
The main difficulties with using Census data to study the performing arts
occupations are conceptual. The conclusions reached throughout this report are
indicative rather than definitive for several reasons. The Census classifies the
occupation of a worker according to the job he or she had the week previous to
Census Day, April 1 . (Not all respondents fill out the form on April 1 , but this report
ignores such variations and denotes the reference week as "Census Week.") If the
worker had more than one job, the one at which he or she worked the most hours
determines the job classification. If the worker is unemployed or not in the labor
force, the occupation is classified as the last one held by the person.
For performing artists, this can be problematic. With irregular work cycles and
work availability, not all performing artists will be working in performing arts
occupations the week before Census Day. If this is so, or if most of their hours that
week were not at a performing arts job, then they are not counted as performing
artists. Wriile excluding those "career" performing artists, the Census by design will
include some "hobbyists" who happened to have earnings from their performing arts
hobby during Census week - further blurring the head count.
All of the employment and earnings information available from the Census
becomes attached to the occupation in which the worker was classified. The Census
does not clarify whether all the hours worked Census week were in an artist
occupation, or if all the weeks worked or income received last year were in
performing artist occupations. Therefore, when this report speaks of "actors' or
dancers' earnings," not all the earnings will necessarily have come from acting or
Performing Artist Surveys
The second source of information on performing artists used in this report is
from national surveys conducted by Ruttenberg et al. in 1977 and 1980. Members
of the largest performing arts unions were surveyed about their careers and specific
problems they experienced in the labor markets. Questions were tailored to the
performing arts and therefore elicited answers much more illustrative of the work
experience of such artists.
The main disadvantage with the surveys is that the universe is restricted to union
members and it is impossible to determine how their results compare with Census
data where criteria for membership in a performing arts occupation differ. Because it
is difficult to generalize the survey results to the performing artist labor force overall,
the surveys are used mostly to clarify points upon which the Census is silent.
Union Pension Records
The third source of data used here is from pension and membership records for
1990 and 1992 from the three principal actors' unions: Screen Actors Guild (SAG),
88 I Artists in the Work Force
Actors' Equity Association (AEA) and American Federation of Television and Radio
Artists (AFTRA). (AFTRA excluded data on non-actor occupations represented by
the union.) The American Federation of Musicians (AFM), the primary musicians'
union, also provided some pension information on average earnings. These unions
and the AFL-CIO Department for Professional Employees cooperated in producing
this data, which contains no personal identification information on individuals.
The pension and membership data show wage earnings solely from acting work,
unlike the Census. As with the surveys, the universe is restricted to union members.
However, through judicious use, these records supply a piece of the picture that is
missing in Census data on performing artists. Using all the information available, it may
not be possible to see the entire picture, but its outlines will be more clearly discernible.
Employment and Earnings Analysis
To identify trends in the performing arts occupations, this report examines the
employment and earnings characteristics of these artists. In chief, these characteristics
are labor force status, geographic distribution, class of employment, industry of
employment, hours worked last week and usual hours last year, weeks worked last
year and earnings. This study works with two universes. The first is based on current
labor force status as of Census Week: employed, unemployed and not a participant.
This universe or subsets of it will be used for geographic distribution, class of
employment and industry of employment. The second universe is based on the work
history for the previous year, regardless of whether the individual was employed,
unemployed or not in the labor force at Census time. This universe is used for all
characteristics based on the year previous to the Census, weeks and hours of work
and earnings. (This second universe offers a richer data source than the one limited
to Census week.)
Each of these economic characteristics will be examined as of the 1990 Census and
then, using the 1970 and 1980 Censuses and other data, the trends will be described.
For the 1980 Census the Bureau of the Census created a new category of actors
and directors to replace the former actors category. Although this means comparable
data are not available over the entire 20-year span for actors-directors, some
interesting trends are nonetheless brought to light.
It bears repeating that Census counts are probably undercounts of the numbers
of performing artists. Based on the Census long form questionnaire, the respondent's
occupation is the one in which he or she had worked the most hours the week prior
to the Census. In 1980 Ruttenberg et al. found that the preponderance of all
performing artists, except dancers, held jobs outside the performing arts and worked
more in those other jobs (Ruttenberg et al., 1981; pp. 64, 81).
Labor Force Status
In 1990 the Census counted 109,573 experienced civilians in the labor force
working in the actors and directors occupation, about 39 percent of all performing
Employment and Earnings of Performing Artists, 1970 to 1990 I 89
artists in the labor force (see Table 3). This represented a 63 percent increase over
1980 and a 173 percent increase since 1970. (The estimated number of actors and
directors for 1970 is 40,201.) As with all of the performing arts occupations, the
number of actors and directors in the labor force grew less rapidly in the 1980s than
in the 1970s, although faster than the total experienced civilian labor force in general
and professionals specifically. On an average annual basis, actors and directors
increased by 5.3 percent per year during the 1970s and 5.0 percent during the 1980s.
This slowdown was the smallest decline in growth rates for any performing arts
PERFORMING ARTISTS IN THE EXPERIENCED CIVILIAN LABOR
FORCE, 1990, AND AVERAGE ANNUAL GROWTH RATES DURING
THE 1970s AND 1980s
Actors and Directors
Musicians and Composers
All Artist Occupations
Source: Ellis and Beresford,
Dancers, the smallest performing artist occupation (7.9 percent of performing
artists in 1990), nearly tripled their numbers during these 20 years. The number of
dancers grew at 5.9 percent per year in the 1970s and 5.2 percent in the 1980s, so
that by 1990 their numbers had increased by 196 percent, from 7,404 in 1970 to
13,194 in 1980 to 21,913 in 1990.
The slowest growth rate — and the most dramatic trend — occurred among
musicians and composers (the Census group which also includes singers). In 1970
the experienced civilian labor force included 99,533 musicians and composers, the
largest group (68 percent) of all performing artists. In 1980 their numbers had
increased to 140,556 for a growth rate during the 1970s of 3.5 percent per year.
During the 1980s, however, the growth rate had slowed to merely 0.5 percent per
year so that by the 1990 Census, musicians and composers had added slightly less
than 8,000 new labor force participants (148,020 musicians and composers in 1990).
This was the lowest growth rate of any artist occupation during the 1980s. By 1 990 the
proportion of performing artists who were musicians and composers fell to 53 percent.
Throughout the period of this study, performing artists' unemployment rates
exceeded those of the civilian workforce as a whole. Of the three groups, actors and
90 I Artists in the Work Force
directors consistently had the highest unemployment rates; their rate in 1990 of
about 13 percent was well over twice the national rate of 5.3 percent. Dancers were
unemployed at a rate of 7 percent in 1990, while the rate for musicians and
composers was 6 percent. For performing artists as a whole, unemployment that year
was 8.8 percent.
Each of the three performing arts groups had its own distinct unemployment
trends during this 20-year period. For musicians and composers, unemployment rose
from 6 percent in 1970 to 8 percent in 1980 and then fell to 6 percent in 1990. Their
rate showed the same trend as the national labor force's unemployment rate, which
rose from 5 percent in 1970 to 7 percent in 1980 and fell to 5 percent in 1990.
For dancers, on the other hand, unemployment rates have declined. Starting at
14 percent in 1970, more than double the national rate, dancers' unemployment
dropped to 1 1 percent in 1980 and to 7 percent in 1990. This was only two points
above the 1990 national rate. (Because of the small number of dancers in the sample,
this difference could be statistically insignificant.)
Unemployment trends of actors and directors are more difficult to pin down. In
1 970 when actors were a separate occupational group, their unemployment rate was
32 percent — extremely high compared to the other groups. When directors were
combined with actors for the 1980 Census, the rate dropped to 15 percent. For actors
alone, the rate could have gone up or down. The Current Population Survey put the
1980 unemployment rate for actors alone at about 35 percent. However, the small
CPS sample size for groups like actors and directors can distort the findings. This
information indicates that directors have steadier employment than actors.
Length of Time Since the Unemployed Last Worked
Along with a decline in unemployment rates, the length of time since
unemployed actors, directors and dancers last worked fell in the 1980s. In 1990, 65.2
percent of actors and directors who were unemployed at the time of the Census
(April 1st) had been employed at some time that year. Nearly one-third, however, had
been unemployed since sometime in 1989. Fewer than 3 percent had not worked for
two years (since 1988) and only 1.5 percent had not worked for three to five years
(since 1985 to 1987). This showed slight improvement over findings of the 1980
Census when 61.1 percent of unemployed actors and directors had been employed at
some time between January 1st and April 1st, 1980; 33.4 percent had last been
employed during 1979; 3.5 percent had been out of work for two years and 2.0
percent had been without a job for three to five years (Table 4).
Unemployed dancers saw an even stronger downward trend in the length of time
since they had last worked. As of April 1, 1990, 68.0 percent of the dancers who were
out of work had been unemployed for less than three months, compared to 57.1
percent of unemployed dancers 1 years earlier. The proportion who had not worked
since sometime in the year prior to the Census dropped from 35.7 percent in 1980
to 24.3 percent in 1990. The proportion unemployed for two to five years was
roughly unchanged: 7.1 percent in 1980 to 7.6 percent in 1990.
Employment and Earnings of Performing Artists, 1970 to 1990 I 91
AMONG UNEMPLOYED PERFORMING ARTISTS
, WHEN DID THEY
LAST WORK? AS OF APRIL 1980 AND 1990
Unemployed as of April 1,
Unemployed actors and directors who last worked...
Two years ago
Three to five years ago
Unemployed dancers who last worked...
Two years ago
Three to five years ago
Unemployed musicians and composers who last worked...
Two years ago
Three to five years ago
Source: 1980 and 1990 PUMS.
*Less than 0.05%
Although musicians and composers had consistently lower unemployment rates
than the other performing artists, those who were out of work tended to remain so
longer. In 1990, 12.9 percent of jobless musicians and composers had been
unemployed for two to five years, compared to 4.2 percent for actors and directors
and 7.6 percent for dancers. For musicians and composers, this was an increase in the
numbers experiencing long term spells of unemployment. In 1980 the proportion
with two or more years of unemployment had been only 8.3 percent. Additionally,
there was a shift from unemployed musicians and composers who had worked during
the Census year (64.6 percent in 1980; 55.7 percent in 1990) to those who had not
worked since the year before the Census (27.1 percent in 1980; 31.4 percent in
A person is counted as in the labor force only if he or she is employed,
unemployed and looking for work, or has been laid off and Is awaiting recall. Of
those classified as "not in the labor force," most are retired or working in the home.
Others may be in school, ill or not in the labor force for other reasons. One reason
is often "thinks no job is available," the discouraged worker. Unfortunatclv.
information on discouraged workers is not available from the Census data. However,
for 1980 Ruttenberg et al. found that 6 percent of all actors surveyed (including those
who might be classed by the Census as not in the labor force) were discouraged
workers. For singers, the rate was 7 percent; musicians, 5 percent and dancers. 3
percent (Ruttenberg et al., 1981; p. 161).
92 I Artists in the Work Force
Spells of Unemployment
Concentrating on the labor force status at a point in time may be misleading; it
certainly gives an incomplete picture. A performer may be employed at Census time,
but this says nothing about his or her experiences in the labor market over time.
According to the Ruttenberg survey, during 1980, 69 percent of actors, 64 percent
of singers, 38 percent of musicians and 77 percent of dancers experienced some
period of unemployment (Ruttenberg et al., 1981; p. 131). Nationally, only 18.1
percent of those in the labor force experienced or were experiencing a period of
unemployment in 1980 (BLS, 1989).
Not only did most performing artists have some period without pay during
1980, but they also tended to have multiple periods of unemployment. In 1980, 63
percent of actors, 49 percent of singers, 56 percent of musicians and 30 percent of
dancers who were not employed experienced three or more periods without any work
during the year, either as performers or in secondary jobs (Ruttenberg et al., 1981; p.
144). By contrast, only 13.1 percent of those in the total work force who had a
jobless period in 1980 experienced three or more periods without work. Additionally,
the majority of the unemployed performers (except for dancers) did not receive
Class of Worker
The Census defines the nature of a worker's employer as the worker's "class" of
employment. The classes are private employer, self-employed, government or unpaid
family work. Throughout this century there has been a trend in the U.S. away from
self-employment toward wage and salary work in private firms, and more recently, in
federal, state and local government. Performing artist occupations, however, deviate
from the national trend. In 1990 of the 255,031 employed performing artists
counted by the Census, 65.8 percent were wage and salary employees of private
firms, 4.1 percent were government employees and 29.9 percent were self-employed
(compared to only 8.6 percent of self-employed working civilians in the labor force
overall). Of self-employed performing artists, 69.9 percent were musicians and
composers, 24.8 percent were actors and directors and 5.3 percent were dancers.
For actors and directors, 20 percent were self-employed in 1990, compared to 17
percent in 1980. In 1970 only 11 percent of actors alone were self-employed; the
difference implies that directors are more likely to be self-employed than actors.
In 1990 dancers were also 20 percent self-employed, representing a five-fold
increase over 1970 when only 4 percent were self-employed. They were, however, less
likely than actors and directors to hold government positions. Dancers had the largest
increase in the numbers of self-employed workers.
Musicians and composers were the most likely to be self-employed in each
Census, with 24 percent self-employed in 1970, 36 percent in 1980 and 38 percent
in 1990. However, the unemployed among musicians and composers were more
likely to describe themselves as self-employed; 47 percent of this group who were out
Employment and Earnings of Performing Artists, 1970 to 1990 I 93
of work described themselves as self-employed, compared to 38 percent of those who
Persons in other classes of work are also classified as unemployed when laid off
because their employer cannot find buyers for the firm's goods or services. This is not
much different from the self-employed musician who cannot find a gig.
Industry of Employment
It should come as no surprise that the majority of all performing artists were
employed in the service sector of the U.S. economy. In 1990 "entertainment and
recreation services" employed 51 percent of actors and directors, 68 percent of
dancers and 48 percent of musicians and composers. Within that sector of
employment, the main industry was "theatres and motion pictures" which employed
51 percent of actors and directors, 32 percent of dancers and 45 percent of musicians
and composers in 1990. For actors and directors, "communications" was the other
principal sector of employment, specifically, "radio and television broadcasting and
cable," which employed 32 percent of actors and directors in 1990. Next largest
industry was "colleges and universities," which employed 5 percent of all working
actors and directors.
For dancers, the other significant industries of employment were "miscellaneous
entertainment and recreation services" (35 percent) and "eating and drinking places"
(22 percent). "Eating and drinking places" fall within the retail sector, which overall
employed 23 percent of all dancers, including 1 percent in "hotels and motels." The
prominence of eating and drinking places as a major employer of dancers probably
reflects the diversity of this category which includes ballet, tap dancers,
choreographers, go-go and "exotic" dancers.
Musicians and composers were the next most likely, after actors and directors, to
work in the "theatres and motion picture" industry. The next largest employers of
musicians and composers were "religious organizations" (25 percent) and "eating and
drinking places" (7 percent). That last figure appears to represent a reversal in the
growth trend for musicians in "eating and drinking places." In 1970, 10,363 or 1 1
percent of musicians and composers were employed in that industry. By 1980 "eating
and drinking places" employed 11,558 musicians and composers. But all the gains
and more disappeared by 1990 when only 9,769 musicians and composers worked
in that industry — a decline of almost 6 percent in the 20-year period. This decline
could in part reflect the difficulty the Census has identifying musicians when thev
work part-time or intermittently in their profession. But it appears to be a factor in
the overall decline in the musicians' labor force growth rate.
An even larger factor in the declining growth rate for musicians and composers
appears to be the sharp decline in the number employed in "theatres and motion
pictures" between 1980 and 1990. In 1980, 68,468 musicians and composers or S3
percent had work in this Industry. By 1990 the numbers had dropped to 62,614
individuals and 45 percent. This was a sharp reversal of the " ' percent growth
musicians and composers enjoyed in this industry in the previous decade. An
94 I Artists in the Work Force
explanation for this trend may lie in the increased availability of electronic substitutes
for live performers in the music field during the 1980s, such as sophisticated
recording and sampling technologies.
Perhaps one of the most interesting changes in industrial employment for
performing artists resulted from combining the actor-director category and the
concomitant rise in importance of cable television. In 1970, 70 percent of all actors
worked in the "theatres and motion picture" industry and only 6 percent in the
"radio and television broadcasting and cable" industry. By 1980 the proportions had
changed to 51 percent of actors- directors employed in the "theatres and motion
picture" industry and 34 percent in the "radio and television broadcasting and cable"
industry. In 1990 these proportions were little changed (51 percent and 32 percent).
Weeks of Work and Usual Weekly Hours
The Census asks respondents if they worked at any time during the previous year
for a wage. Having worked last year forms the organizing principle for employment
and earnings characteristics of performing artists in the following sections.
The number of performing artists increases significantly when the focus shifts
from the currently employed (as of Census Week) to those who worked at all the
previous year. In 1990, 255,031 performing artists were employed at the time of the
Census, but almost a quarter more (310,925) had worked sometime the previous
year. Workers in general cycle between employment, unemployment and not in the
labor force due to a host of personal and economic circumstances. Performing artists
appear more likely to experience this cycling than most other workers. When the
focus shifts from the current to the previous year, a problem is uncertainty that the
current occupation is the same as last year's.
Among actors and directors who had any work in 1989, 53.6 percent worked 50
to 52 weeks during the year. At the other end of the spectrum 23.5 percent worked
26 or fewer weeks. This was an improvement over 1979 when 48.3 percent worked
a full year and 27.7 percent worked less than half a year. In 1969, before directors
were included in this category, only 21.7 percent of actors worked a full year, while
41.2 percent worked less than a half year. In fact, in 1969, nearly a quarter of all
actors worked 13 or fewer weeks. Improvements for actors and directors from 1970
to 1990 have resulted largely from changing the definition of the category in 1980,
as directors clearly have steadier employment than actors. The median number of
weeks worked by actors and directors in 1989 was 50, up two weeks from 1979.
Dancers were the least likely in 1989 to work a full year. Only a third of all
dancers who had any work that year worked 50 to 52 weeks. This represented some
progress as only 29.5 percent of dancers worked a full year in 1969 and 28.3 percent
in 1979. Dancers were also more likely than other performers, although marginally
so, to work less than half a year in 1989. Of all dancers who worked in 1989, 27.2
percent worked 26 or fewer weeks, a slight improvement over 1979 (33.3 percent)
and 1969 (31.1 percent). The median number of weeks worked in 1989 was 39, the
lowest for performing artists.
Employment and Earnings of Performing Artists, 1970 to 1990 I 95
Nearly half of all musicians and composers who worked in 1989 (46.1 percent)
worked a full year, which was up from 39 percent in 1969 and a similar proportion
in 1979. In 1989 the median number of weeks worked by musicians and composers
was 48, up by six weeks from 1979 (data for 1969 is not available). Taken together,
the data for musicians and composers paint a picture of a labor market that is using
its employed workers more intensively (more weeks worked and longer spells of
unemployment) compared to other performers.
For most occupations, less than 35 hours a week is considered to be part-time
employment. By this measurement, of musicians and composers who worked in
1989, 43.8 percent worked full-time and 56.2 percent part-time (Table 5). This
represented a trend toward less part-time work for musicians and composers, 63.0
percent of whom worked part-time in 1979. (1969 data not available.) It is
impossible to say whether this trend may represent an increase in hours worked at
non-performing arts related jobs.
CROSSTABULATION OF PERFORMING ARTISTS BY PART/FULL
TIME AND PART/FULL YEAR WORK STATUS IN 1989.
Source 1990 PUMS.
Dancers experienced a relative decline in full-time employment as measured by
usual weekly hours from 1979 to 1989. In 1979, 56.0 percent were employed full-
time and 44.0 percent part-time. By 1989 the proportions had changed to 50.4
percent working full-time usual weekly hours and 49.6 percent part-time.
For actors and directors, those employed full-time increased slightly from 74.6
percent working full-time hours in 1979 to 77.2 percent working full-time hours in
1989. When actors and directors worked, they were more than twice as likely as other
performers to work long hours; 37.4 percent had more than 40 hours in a typical
week in 1989.
Because it will be important later when earnings are discussed, Table 5 brings
together the discussion of weeks and usual weekly hours worked by performing artists
in 1989. It is interesting to note that whether actors and directors are employed part-
or full-year, they are more likely to work full-time — 35 or more hours a week. For
dancers, musicians and composers, part-year workers are more likely to work part-
time and full-year workers to work full-time. This has a significant effect on earnings
96 I Artists in the Work Force
levels for these occupations, as will be shown.
The Census does not answer the question that is probably foremost in the mind
of those interested in the labor market experiences of performing artists: how much
of the work reported in the Census is in the performing arts and how much is in
other jobs? Most performing artists are multiple jobholders. Sixty-four percent each
of actors and singers, 73 percent of musicians and 37 percent of dancers held jobs
outside of the performing arts in 1980 (Ruttenberg et al., 1981; p. 81). In contrast,
only 5.2 percent of all workers nationwide held multiple jobs in 1970, 4.9 percent in
1980 and 6.2 percent in 1989 (Stinson, 1990; p.4). In 1991 the multiple job-holding
rate for managerial and professional specialty workers was 7.3 percent, the highest
rate among the major occupational groups, but not as high as rates for performing
artists (Census, 1992). The most frequent reasons performers gave in 1980 for
working outside performing arts were (1) there was not enough performing arts work
and (2) the outside work supplemented their income from the performing arts.
The Ruttenberg study found that in 1980 actors were the most likely to have
worked 50 days or fewer in performing arts (61 percent). (For a regular eight hour
work day and five day work week, the usual number of days with pay is 260 per year.)
Singers tended to have more performing arts work than actors; only 31 percent
worked 50 days or fewer. Thirty-four percent of musicians worked 50 or fewer days
in their field. Dancers had the most work time in a performing arts job. Only 19
percent of dancers worked 50 days or less, but 47 percent worked over 250 days,
giving dancers the highest proportion with roughly full year employment in the
performing arts. Only 5 percent of actors, 8 percent of singers and 12 percent of
musicians worked as much as dancers in performing arts jobs (Ruttenberg et al.,
1981; p. 64). By contrast, Census data showed dancers the least likely to work a full
The differences might arise from small sample sizes in both data sources, the
mixture of types of dancers in the Census and the union status of dancers in the
Ruttenberg study. They may also reflect that actors and musicians are more likely
than dancers to get supplemental work outside their performing arts profession,
thereby boosting the number of weeks they report employment relative to dancers.
The Ruttenberg study reported that of those who held a job outside the performing
arts, 47 percent of actors worked 100 or more days in that job and 1 7 percent worked
over 250 days in 1980. The proportions for singers were even greater: 59 percent
worked over 100 days. Fifty percent of musicians worked 100 or more days in their
other jobs. Dancers, the least likely to have a job outside the performing arts, also
worked less in those jobs when they did have them; only 22 percent worked more
than 100 days in their non-performing arts jobs (Ruttenberg et al., 1981; p. 1 16).
Clear-cut information on performing artists' incomes is difficult to come by, due
to multiple job holding, irregular hours and periodic joblessness, with and without
Employment and Earnings of Performing Artists, 1970 to 1990 I 97
Data on income by type is available in the Census data. In the Artist Extract
Files, which are derived from the Census, the data for each person include income
from wages and salaries, non-farm and farm self-employment, retirement, interest
and dividends, public assistance and "other" sources. Interpreting this income
information for performing artists is not easy. The breakdown of wages between
performing arts work and/or a support job cannot be known. Interest income could
be residual or royalty income from a recorded performance or interest from a
checking account. Caution is called for when interpreting the data.
Union pension records, on the other hand, show wage and residual income from
performing arts work only. However, they may not represent all of a person's earnings
in the performing arts field, and they lack data on earnings from jobs outside the
field. Neither do they include precise information on how much work time the
earnings represent. (The union pension records do include weeks worked, but
according to the standard practice, one to seven days worked in a seven day period is
recorded as a full week of work. This system makes it difficult to tell who is actually
working more than whom.) Also, information on labor force status comparable to
that found in the Census and CPS is lacking.
Despite the shortcomings of the Census and pension data, a great deal can be
said about the earnings of performing artists.
Median and Mean Earnings
The last three Censuses have recorded ever higher earnings for each group of
performing artists. In 1969 actors had median earnings of $5,936; dancers, $3,332
and musicians and composers, $2,958. By 1979 the median earnings of dancers
increased 62 percent to $5,404 and musicians and composers 88 percent to $5,561.
These increases, however, were outpaced by the 98 percent rise in average consumer
prices in the U.S. from 1969 to 1979.
Data for actors and directors are not available for 1969 because directors were
not included with actors in the 1970 Census. In 1979, however, the median earnings
of actors and directors were $12,564, an increase of 1 1 1.7 percent over actors alone
in the 1970 Census. Clearly the merging of directors with actors in the 1980 Census
boosted the earnings of the group relative to actors alone.
By 1989 the median earnings of performing artists were higher: for actors and
directors, $22,000 (up 75 percent from 1979); for dancers, $8,500 (up 57 percent)
and for musicians and composers, $9,900 (up 78 percent). Average prices increased
by 68 percent during this time, so for all performers except dancers their median
earnings outpaced inflation. (See Table 6.)
The mean earnings of performing artists also increased from 1979 to 1989, but
at a higher rate than inflation. Additionally, mean earnings were higher than median
earnings, which showed some individuals' earnings were much higher than the rest
of their groups earnings. The mean earnings of actors and directors were $16,498 in
1970, or 31 percent higher than median earnings; by 1989 they had increased to
$32,261 (up 96 percent), with the difference between the mean and median or
98 I Artists in the Work Force
MEDIAN EARNINGS OF PERFORMING ARTISTS WORKING IN
1969,1979 AND 1989.
1969 1979 1989 1969-79 1979-89
Actors & directors
Musicians & Composers
Inflation (CPI-W, 1982/84 + 100
Sources: NEA Research Division Note #10 "Artists Real Earnings Decline 37 Percent in the
1970's," Washington, DC, March 5, 1985, and Artist Extract File for 1990 Census
spread increased to 47 percent. This growing spread indicates that the earnings of the
highest paid members of the professions, perhaps "superstars," increased faster than
earnings of the profession as a whole. (See Table 7.)
The same trend held for dancers, musicians and composers. For dancers, their
mean earnings were $7,062 in 1979 and increased by 72 percent to $12,152 in 1989.
Musicians' and composers' mean earnings were $16,233 in 1989, an increase of 105
percent over their mean earnings of $7,923 in 1979. The spread also increased. In
1 979 the mean was 3 1 percent higher than the median for dancers' earnings, but 43
percent higher in 1989. For musicians and composers, the spread increased from 43
percent in 1979 to 64 percent in 1989.
MEAN EARNINGS OF PERFORMING ARTISTS AND INFLATION
1979 TO 1989.
1979 1989 1979-89
Actors and directors
Musicians and composers
Inflation (CPI-W, 1982/84=100)
Sources: Artist Extract Files for 1980 and 1990; Bureau of Labor Statistics
Distribution of Earnings
Single numbers like the mean or median do not tell the whole story of how well
or poorly performing artists have done in the labor market at each point in time. For
example, in 1989 about 1 percent of all performing artists earned zero or less (Table
8). Twenty-four percent of actors and directors, 53 percent of dancers and 49 percent
of musicians and composers earned less than $10,000 in 1989 — from all work. The
similarities in earnings distributions for performing artist occupations end there.
Actors and directors are much more evenly spread out in terms of earnings and
consistently earn more than their dancer or musician peers. About three-quarters of
Employment and Earnings of Performing Artists, 1970 to 1990 I 99
MEAN EARNINGS OF PERFORMING ARTISTS AND INFLATION
1979 TO 1989.
$1 to $999
$1,000 to $4,999
$5,000 to $9,999
$0 to $9,999
$10,000 to $19,999
$20,000 to $29,999
$30,000 to $39,999
$40,000 to $49,999
$50,000 to $59,999
$60,000 or more
Source: 1990 PU 'MS.
dancers, musicians and composers earned less than $20,000 in 1989, but less than
half of all actors and directors who worked in 1989 earned less than $20,000. At the
other end of the earnings distribution, 13 percent of actors and directors (1 in 8)
earned $60,000 or more in 1989, but only 1.3 percent of dancers (1 in 80) and 3.9
percent of musicians and composers (about 1 in 25) earned that much the same year.
Although actors and directors were the most likely to earn in the higher brackets
in 1989, the distribution of earnings improved during the 1980s for all performing
artist occupations as seen in Tables 9, 10 and 1 1 for 1979 and 1989. Data for 1969
are not available for earnings distributions.
The earnings distribution shifted markedly for actors and directors between
1979 and 1989. The proportion earning less than $10,000 during the year decreased
from 44 percent in 1979 to 24 percent in 1989. The proportion who earned at least
$10,000, but less than $20,000, also decreased, from 28 percent to 20 percent. At
the other end of the earnings distribution, every interval above $20,000 increased its
share of actors and directors earning in those ranges. The proportion of actors and
directors with the highest incomes, $60,000 or more, grew the most, increasing from
5 percent in 1979 to 13 percent in 1989.
An upward shift in the earnings distribution also occurred for dancers during the
1980s. In 1979, 75 percent of dancers earned less than $10,000, but by 1989 the
proportion of dancers in that range had fallen to roughly SO percent. More than a
quarter earned between $10,000 and $19,999 in 1989, compared to less than 20
percent earning in that range in 1979.
The distribution of earnings for musicians and composers also improved
between 1979 and 1989. The proportion earning less than $10,000 fell from 71
percent in 1979 to about one half in 1989. The proportion earning $30,000 or more
increased substantially from 3.6 percent in 1979 to IS. 3 percent in 1989.
100 I Artists in the Work Force
FREQUENCY DISTRIBUTION OF ACTORS AND DIRECTORS WHO WORKED
IN YEAR PREVIOUS TO CENSUS,
1980 AND 1990 CENSUSES.
Shift in %
$1 to $999
$1,000 10 54,999
$5,000 to $9,999
$0 to $9,999
$10,000 to $19,999
$20,000 to $29,999
$30,000 to $39,999
$40,000 to $49,999
$50,000 to $59,999
$60,000 or more
Source: 1980 and 1990 PUMS.
FREQUENCY DISTRIBUTION OF DANCERS WHO WORKED IN YEAR
PREVIOUS TO CENSUS,
BY EARNINGS. 1980 AND 1990 CENSUSES.
Shift in %
$1 to $999
$1,000 to $4,999
$5,000 to $9,999
$0 to $9,999
$10,000 to $19,999
$20,000 to $29,999
$30,000 to $39,999
$40,000 to $49,999
$50,000 to $59,999
$60,000 or more
Source: 1980 and 1990 PUMS.
FREQUENCY DISTRIBUTION OF MUSICIANS & COMPOSERS WHO WORKED
IN YEAR PREVIOUS TO CENSUS,
1980 AND 1990 CENSUSES.
Shift in %
$1 to $999
$1,000 to $4,999
$5,000 to $9,999
$0 to $9,999
$10,000 to $19,999
$20,000 to $29,999
$30,000 to $39,999
$40,000 to $49,999
$50,000 to $59,999
$60,000 or more
Source: 1980 and 1990 PUMS.
Employment and Earnings of Performing Artists, 1970 to 1990 I 101
Full-time versus Part-time Earnings
A number of variables affect earnings. The number of hours worked per week
and the number of weeks worked per year are major factors determining the level of
earnings. To simplify this analysis, performing artists are grouped according to
whether they worked part-time and part-year (less than 35 hours per week, less than
50 weeks per year) or full-time and full-year (35 or more hours per week, 50 weeks
or more per year). As Table 12 shows, these differences have a dramatic effect on
FREQUENCY DISTRIBUTION OF PERFORMING ARTISTS WHO
WORKED IN YEAR PREVIOUS TO CENSUS, BY EARNINGS AND BY
PART/FULL TIME AND PART/FULL YEAR WORK STATUS,
EARNINGS FROM 1990 CENSUS.
* & Part-year
$1 to $999
$1,000 to $4,999
$5,000 to $9,999
$0 to $9,999
$10,000 to $19,999
$20,000 to $29,999
$30,000 to $39,999
$40,000 to $49,999
1 1 .3%
$50,000 to $59,999
$60,000 or more
Source: 1980 and 1990 PUMS.
For all performing artist occupations (without regard to hours or weeks worked),
the most common earnings interval was the $0 to $9,999 range (Table 8, earlier).
When part-time and part-year artists are separated from full-time and full-year
artists, the most prevalent interval for part-time and part-year artists continues to be
the $0 to $9,999 interval, with 65 percent of actors and directors, 76 percent of
dancers and 71 percent of musicians and composers (Table 12). Interestingly; the
proportions of all performing artists earning SO to $9,999 are closer for all three
occupations when just part-time and part-year workers are examined. Clearly, some
of the earnings differences between actors and directors and the other two groups
discussed above arise from differing proportions of part-time workers (Fable 5« earlier).
For full-time and full-year work, the most prevalent range tor performing artists
overall is $10,000 to $19,999. Twenty percent of actors and directors, 42 percent of
dancers and 29 percent of musicians and composers earn in this range. Notably.
substantial proportions of all performing artists who work full-time, all year earn in
the higher ranges. For example, when all performing artists are considered, 18
102 I Artists in the Work Force
percent of actors and directors and 13 percent each of dancers, musicians and com-
posers earned in the $20,000 to $29,999 range. When only full-time, full-year workers
are considered, however, about a quarter of each group had earnings in this range.
Much has been made of the low levels of performing arts earnings and the fact
that median earnings generally have failed to keep pace with inflation. In analyzing
the meaning of available data on their earnings, though, it is important to keep that
data's limitations in mind.
One problem is the temptation to analyze Census data as if the performing artists
recorded in the 1970 Census are the same ones recorded in the 1980 and 1990
Censuses. This is not the case. As noted earlier, the performing artist labor force has
increased substantially over the last 20 years, faster than the labor force in general.
This suggests that the performing artist labor force has a greater proportion of new
and relatively inexperienced workers, and these less experienced performers might
inflate the lower end of the earnings distribution. Those just starting out will earn
little and if they don't make it will likely drop out of the performing arts work force
to pursue another career. Those who do make it will probably see their earnings rise
over time. By the next Census, the discouraged performers are accounted for in
different occupations and a new, larger group of aspiring performers are now counted
in the performing artist labor force. The Census data will reflect the successes of
performing artists who have found steady work by showing a shifting of the earnings
distribution toward higher incomes.
Generally, researchers have found that performing artists earn less than other
professionals with similar experience and education. However, Ruttenberg et al.
found in 1980 that 50 percent of actors, 42 percent of singers, 49 percent of
musicians and 23 percent of dancers received less than half of their income from
performing arts work (Ruttenberg et al., 1981; p. 182). Thus it cannot be assumed
that earnings of performing artists reported in the Census are from performing arts
work alone and that as professionals, their earnings suffer in comparison to those of
other professionals. What may suffer is the availability of steady employment as a
performing artist relative to other professionals. It cannot be demonstrated using
Census data that the artists' rates of pay are lower.
Another factor heavily influencing earnings is the artist's other occupation. The
1981 survey by Ruttenberg et al. found that of those with second jobs in 1980 over
half the actors and singers, over a fourth of musicians and one third of dancers were
in sales, clerical or service jobs — jobs with a history of low pay and benefits. Fewer
than 1 5 percent held professional jobs as their supplementary occupation and the
majority indicated that pay of their secondary job was less that their pay in
performing arts. Most of those with a second job tended to choose those jobs that
gave them the flexibility needed to pursue an arts career.
Union Earnings from Pension Records
This section examines wage and residual earnings information provided by the
three main actors' unions from their pension records. These three unions represent
most of the nation's professional actors who perform on stage, screen, television,
Employment and Earnings of Performing Artists, 1970 to 1990 I 103
radio or video. Comparable data could not be provided by the two principal unions
in the musical arts. This discussion will therefore concentrate on actors, except in the
area of average earnings by source, where certain information is also available for
Data furnished by the actors' unions cover all members of the particular union
who had worked anytime within the five years previous to 1990. The Screen Actors
Guild (SAG) had data for 1992 only, while the other two unions, Actors' Equity
Association (AEA) and Aanerican Federation of Television and Radio Artists
(AETRA), furnished data for both 1990 and 1992. (AFTRA's pension data used here
excludes categories of members who worked in non-acting fields.)
The pension data was combined into two master files. The first file covers
individuals with at least $ 1 in earnings in 1 992 who belonged to any of the three
unions. The second file covers individuals regardless of earnings who were active
members in both 1990 and 1992 of AFTRA and/or AEA. Both files contain, for each
individual, the source of income by union and whether it was wage or residual
income; however, neither file included personal identifying information. The
category of wage income in these pension data files corresponds to the wage and
salary income category of the Census. However, an actor's residual income would
have been included in the Census as interest, royalty and dividends — a category that
was not examined in the previous section on earnings information from the Census.
The first file, covering actors and extras of all three unions who earned at least
$1 in 1992, included 84,637 actors and extras (extras are included in the Census'
actor/director category). As an indication of overlapping memberships, 56 of these
actors and extras belonged to only one union, 31 percent belonged to two unions and
13 percent belonged to all three unions. The second file, which covers members of
AEA and AFTRA who were active in both 1990 and 1992, includes 93,766 actors
and extras. The differences between the two universes are (1) SAG members who are
members of SAG only, (2) any actor/extra who earned nothing, and (3) active
members in 1992 who were not active in 1990.
The Census and pension data are not strictly comparable, but their differences
help point out some unique labor market problems faced by performers. The data are
not comparable because (1) the Census has stricter inclusion criteria both for
occupational and labor force status and (2) the Census data includes directors. Of
those 84,637 union actors in the pension records, many were probably omitted from
the tally of Census actors because they did not spend the majority of their work hours
during Census Week working as actors. Similarly, there were probably other union
actors who were out of the labor force altogether during Census Week. The Census'
stricter criteria may help explain why the 1990 Census counted 109,573 actors and
directors, while the union's figures for actors alone just two years later are about 28
Based on the 1992 union data covering all 84,637 actors and extras with
earnings, the vast majority (about 69 percent) earned less than $10,000 per year from
104 I Artists in the Work Force
wages and residuals (Table 13). The most comparable Census figures are from 1989
when only 24 percent of actors and directors reported earning $10,000 or less. The
differences lend support to the hypothesis that Census data on performers' earnings
are more reflective of their non-performing arts work, and that combining actors and
directors in the Census data has masked the labor market experiences of actors alone.
This hypothesis is further supported by looking at the high end of the earnings
distribution. In the 1989 Census data 13 percent of actors and directors earned
$60,000 or more, whereas only 6 percent of actors in the pension data earned that
FREQUENCY DISTRIBUTION OF EARNINGS FOR ACTORS WITH EARNINGS
THROUGH THE ACTOR UNIONS (Wages & Residuals).
AFTRA, and SAG,
1992 Wages & Residuals
through AEA, AFTRA and/or
SAG by Actors with Earnings
$1 to $999
$1,000 to $4,999
$5,000 to $9,999
$1 to $9,999
$10,000 to $19,999
$20,000 to $29,999
$30,000 to $39,999
$40,000 to $49,999
$50,000 to $59,999
$60,000 or more
Source: Union pension
records, AEA, AFTRA and SAC.
Earnings over Time
The pension data which covers only AEA and AFTRA members has information
for the same 93,766 individuals in 1990 and 1992, providing a means to examine
how individuals fare over time in the acting profession, albeit for a relatively short
Overall, the earnings distribution is fairly stable. Eighty-nine percent in 1990
earned less than $10,000 per year and 87 percent earned less than $10,000 per year
in 1992. The distribution changed little at any earnings level.
What is masked in the distribution as a whole is the volatility of personal
earnings between these two years. Of those without any earnings in 1990, 25 percent
earned something in 1992 — some going from "zero to 60" thousand or more in two
years. Conversely, of those who had earned something in 1990, 18 percent earned
nothing in 1992. Although more than half of the 93,766 earned nothing from their
craft in both years, of the remainder about 50 percent earned more in 1992 and
about 50 percent earned less. These statistics include actors whose earnings changed
Employment and Earnings of Performing Artists, 1970 to 1990 I 105
within their range. Looking at the same earnings intervals, about one-third moved to
a higher interval, about one-third stayed the same and about one-third dropped.
Source of Earnings
Earnings levels also differ significantly by source. For SAG members alone,
information was obtained from the pension records on the source of earnings for
their members broken down as follows: actor versus extra, wages (from any and all
employers) versus residuals and by type of performance (theatrical motion picture,
television, TV commercial and industrial educational). The data are for 1992 (Table 14).
AVERAGE WAGES AND RESIDUALS FOR SAG MEMBERS,
BY TYPE OF WORK. 1992.
Type of Earnings & Work
Theatrical Motion Picture
Theatrical Motion Picture
Theatrical Motion Picture
All types of work 
 Detail aggregated to preserve
Source: SAG Pension Records.
The highest average wage was for actors with television work who earned an
average of $15,712 for the year (1992). The highest average residual was from TV
commercial work, $11,284. The lowest average wage was for extras in theatrical
motion pictures, $651, while extras earned the most in television and in TV
commercials, about $1,168. These averages probably show the same amount of
variation as do earnings as a whole, but they do provide a clue as to why earnings vary
so much. It may be that some types of work pay more, or simply that more work is
available in those categories. The data cannot disclose which is the case.
Similar information is available from the American Federation of Musicians
(AFM) which provided summary data on its members' average earnings by type of
106 I Artists in the Work Force
performance from pension records covering 43,552 members in 1992. The highest
average pay for musicians came from New York City theatricals, where average pay
was $15,164 for the year (Table 15). The lowest was for performances on network or
commercial radio at $680. Average earnings for TV firms were $3,318 and motion
These data are essentially the total of all earnings from a particular source (e.g.,
TV commercials) divided by the number of performers with that kind of work. They
are average earnings per person from that kind of work, not average earnings per
performance. While it is not possible to tell from these data which types of
performances paid better, these figures do show the extent to which each type of work
is serving as a source of income for actors and musicians.
AVERAGE WAGES FOR MUSICIAN MEMBERS OF AMERICAN
FEDERATION OF MUSICIANS, BY INDUSTRY & OVERALL. 1992.
Type of Earnings & Work
(Symphonic & non-symphonic)
Jingles and commercials (Radio and TV
Phonograph demo recordings
Network radio (commercial radio)
Fairs, rodeos and circuses
New York City theatricals
Local casual and steady engagements
All industries 
 Includes union local officials and employees. The sum
of the detail exceeds the total
number of earners because earners can
have earnings in more than one
Source: AFM Pension Records.
This section details, as much as practicable, where performing artists live and
work, based on Census data. The Census defined regions are: Northeast, Midwest,
South and West. Each region is composed of divisions of which there are nine: New
England, Middle Atlantic, East North Central, West North Central, South Atlantic,
East South Central, West South Central, Mountain and Pacific. Unless specifically
noted to the contrary, actors and directors are included in the following 1970 data,
unlike most of the preceding analysis.
Employment and Earnings of Performing Artists, 1970 to 1990 I 107
Distribution by Region
In 1990 the West region of the United States had 36 percent of all actors and
directors, more than any of the other three regions. In 1970 only 29 percent of actors
and directors lived in this region, a virtual tie with the Northeast where 3 1 percent
lived (Table 16). By 1990 the Northeast was close to the same number at 29 percent.
The Midwest's share dropped from 18 percent of actors and directors in 1970 to 14
percent in 1990. The South remained essentially unchanged at 21 percent in 1990,
up from 18 percent in 1980 but the same as in 1970.
DISTRIBUTION OF PERFORMING ARTISTS BY REGION OF
THE UNITED STATES. 1970 TO 1990.
Actors and Directors
Musicians and Composers
Source: Ellis and Beresford,
The pattern for dancers could not have been more different. The West employed
40 percent of all dancers in 1970, dropping to 32 percent in 1980 and to 28 percent
in 1990. The major gainer was the South whose share of dancers in 1990 was 32
percent, up from 19 percent in 1970 and 25 percent in 1980. With minor
fluctuations, the Northeast and Midwest had about the same proportion of dancers
in 1990 as in 1970. In 1970 the Northeast was home to 24 percent, in 1990, 23
percent. The Midwest had 17 percent in both 1970 and 1990.
Musicians and composers followed yet another trend. While the Northeast
incurred a slight loss in its share of musicians and composers and the West a slight
gain over the 20-year period, the Midwest lost a substantial share, going from 2^
percent in 1970 to 20 percent in 1990. The South was the major gainer. In l l )~0. 2"
percent of musicians and composers lived there while 32 percent lived in the South
The overall trend represented a shift in employment to the South for dancers,
musicians and composers and to the West for actors and directors. 1 [owever, no area
108 I Artists in the Work Force
had a net loss of any performing artist group over the 1970 to 1990 period. The
changes in proportion for all three groups are due solely to different rates of growth
of the performing artist work forces within the regions.
Distribution by Division
At the level of Census division, the proportion of actors and directors who live
in a particular area has been remarkably stable over the last 20 years. The notable
exceptions are the Pacific region which increased its share from 25 percent in 1970
to 32 percent in 1990 and the West South Central region (Texas, Arkansas,
Oklahoma and Louisiana) which decreased its share from about 10 percent in 1970
to about 5 percent by 1990. Again, these changes reflect solely different growth rates
of the actor and director work forces in those areas.
If the last two decades were boom years for actors and directors in the Pacific
area, they seem to have been bust years for dancers. The Pacific states' share of
dancers shifted from 32 percent in 1970 to 19 percent by 1990. The other two areas
with high concentrations of dancers in 1970 also saw their shares fall. In the Middle
Atlantic states which include New York, the share fell from 22 to 19 percent over the
20-year period. In the North East Central states which include Illinois, the
proportion of dancers fell from 15 percent in 1970 to 12 percent in 1990.
The big gainers in terms of dancer population were the South Atlantic
(Maryland to Florida on the Atlantic seaboard) and the West South Central states
(Texas, Oklahoma, Arkansas and Louisiana). From 1970 to 1990 the South Atlantic
increased its share of resident dancers from 1 2 to 18 percent; the West South Central
states, from 6 to 1 1 percent.
The geographic distribution of musicians and composers has remained fairly
stable since 1970. The major shifts have been from the Middle Atlantic states (chiefly
New York) to the Pacific and South Atlantic states. As with the other performing
artist occupations, the changes in shares for musicians and composers reflect different
Unemployment by Division, 1 990
Nationally, in 1990 the unemployment rate for actors and directors was 13
percent; for dancers, 7 percent and for musicians and composers, 6 percent.
However, in areas where the absolute numbers of performing artists are quite high,
the unemployment rate also tends to be higher than the national average. Using a
statistical test to determine the degree of trend, where percent indicates no
relationship and 100 percent indicates a full relationship, the degree to which the
trend is true for actors and directors is 95 percent; musicians and composers, 59
percent and dancers, 50 percent.
Across occupations, four divisions consistently are home to over 10 percent of
performing artists: Middle Atlantic, East North Central, South Atlantic and Pacific.
The Middle Atlantic and the Pacific have consistently higher proportions of the
Employment and Earnings of Performing Artists, 1970 to 1990 I 109
nation's performing artists and unemployment rates consistently higher than the
national average. The Middle Atlantic had 24 percent of actors and directors in 1990
with an unemployment rate of 16 percent (compared to 13 percent nationally for
performers). The Pacific had 32 percent of actors and directors with an
unemployment rate of 20 percent. The trend holds for the other performing artists
In contrast, the other two major divisions had unemployment rates lower than
the national average. The East North Central had 10 percent of all actors and
directors in 1990, but an unemployment rate of 6 percent, less than half the national
rate. The same held true for dancers (12 percent of dancers, 6 percent unemployment
rate) and musicians and composers (14 percent with 6 percent unemployment rate).
In the South Atlantic division the unemployment rate for all three groups of
performing artists was 5 percent.
Geographic Concentration of Performing Artists
Looking at the divisional level, it is clear that performing artists are concentrated
in specific areas of the country. Most of those performing artists who lived on "one
of the coasts" lived in the Pacific states (California, Oregon, Washington, Alaska and
Hawaii) if they were in the West or the Middle Atlantic states (New York, New Jersey
and Pennsylvania) if they were in the Northeast. In 1990, 56 percent of actors and
directors lived either in the Pacific states (32 percent) or the Middle Atlantic states
(24 percent). For dancers, musicians and composers, 38 percent lived on either coast.
Moving to even greater detail, the New York City and Los Angeles metropolitan areas
were the major areas of residence for performing artists. In 1990, 37 percent of
actors and directors lived in Los Angeles or New York, with more in Los Angeles. For
dancers, 16 percent lived in either city, with more in New York. For musicians, 16
percent lived in these two metro areas, with 8 percent in each.
In 1990, 60 percent of all employed actors and directors worked in just six states
(California, New York, Texas, Florida, Illinois and Massachusetts). For employed
dancers, 39 percent were concentrated in just four states (New York, California,
Florida and Texas), and the same four states (with California first) provided work for
40 percent of employed musicians. The reverse side of this concentration finds ~^
states providing only 10 percent of the work for actors And directors in 1 990. I lalf of
.ill states plus the District of ( Columbia offered employment for just 1 2 percent of the
nations dancers, including three states with no dancer employment reported. Sixteen
percent of working musicians were spread among half of the states And the District
A significant trend over the last twenty years has been the Increasing importance
of the performing artist workforce in the South, notably dancers and musicians-
composers. B) 1990 nearly one thud of the members ot those two professions were
concentrated in the South Atlantic division. The overall proportion of actors and
directors in the South, on the other band, Stayed at about one out of five over the
twenty years, despite a small Increase in then percentage in the South Atlantic states.
110 1 Artists in the Work Force
The most recent data from the Census and other sources discussed here leave
little doubt that performing artists face a cluster of labor market obstacles, ranging
from a shortage of full time jobs to a lack of steady income from their profession to
the limited geographic areas where work is likely to be found. As a result, the talents
and resources of the nation's performers are significantly underutilized. In the
interests of promoting and nurturing the arts in this country, the government,
through NEA, would seem to have a direct stake in improving the employment and
earnings opportunities for performing artists.
Aji understanding of the work force needs of performing artists is shaped by the
ability to observe and document those needs. This cannot be done with precision at
present. The incidence of multiple jobs and income sources, and the on-again, off-
again employment patterns common in this field complicate any analysis of the
performers' work experience. The primary source of data, the Census long form
questionnaire, is not designed to gather the detailed data needed to explain their
situation. The authors therefore recommend that the NEA consider how to
encourage the development of additional data sources which could include:
(1) Special surveys of performing artists. The best way to overcome the gaps left by
the Census and CPS questionnaires would be to develop a national survey of
performing artists. This will require cooperation from performing arts unions,
theatres, dance organizations and others with lists of members of employees from
which the survey samples could be drawn. Longitudinal studies of performing
artists as they pursue their careers would be of the utmost importance and
(2) Records of organizations in the field, including unions and performing arts
organizations. Public and private cooperation should be sought in developing
new data sources, building on existing resources of these organizations. As labor
organizations in the field develop their own databases and computer capacity, the
NEA should consider providing technical assistance to help them design data
systems that will also meet national informational needs through easy
accessibility and standardized record keeping that does not compromise the
required level of confidentiality.
(3) A special Current Population Survey questionnaire on performing arts
employment. This should be administered periodically to obtain regular, current
information from the general population on income and employment in
performing arts occupations. This special CPS should include questions to
identify persons who consider themselves performing artists and to differentiate
between their arts related and non-arts related work and income.
Employment opportunities in the performing arts obviously depend on a
sustained level of support for drama, dance and music. A continuing priority for
the Endowment should be to undertake activities that encourage public and
Employment and Earnings of Performing Artists, 1 970 to 1 990 1111
private investment in the performing arts, as well as to develop and expand
audiences. Special attention should be given to activities that will generate jobs
and income for professionals in the performing arts.
About the Authors
Ann Kay is a senior member of Ruttenberg, Kilgallon & Associates, a
Washington, DC, based labor consulting firm. She coordinated the firm's 1980
survey of employment and earnings among members of performing arts unions,
which resulted in the 1981 report, Working — and Not Working — in the Performing
Arts. She has also conducted a variety of research and evaluation studies on other
issues related to employment and skill development.
Steve Butcher is a doctoral candidate in economics at The American University
in Washington, DC. He is also employed by Ruttenberg, Kilgallon & Associates
where he consults for labor unions on a variety of issues including earnings and
working conditions. His only direct experience with the performing arts was a junior
high school performance in Peter Pan as Michael; it was panned by the critics.
IV. Architecture and Design
Arts Occupations, 1 970 to 1 990
by Harry Hillman Chartrand
Architects and designers are the visual ecologists of our society. They cultivate the
images and forms that shape the human environment, applying the insights
and findings of the fine arts to our daily lives. They shape, color and mold the
skylines of our cities, clothes, shopping centers, cereal packages, houses, furniture,
TV sets, offices, factories, churches and temples of today's society. Architects and
designers contribute what the ancient Greeks called kosmosr. the right ordering of the
multiple parts of the world.
Even the words and sayings of architects and designers form part of our
contemporary vocabulary. It was Louis Sullivan, architect of the first skyscraper, the
Wainwright Building in St. Louis (1890), who said, "Form follows function." Frank
Lloyd Wright coined the phrase, "Organic architecture." Mies van der Rohe,
godfather of the International Style, is credited with saying, "Less is more."
If architects and designers are concerned with the present, preservationists are
dedicated to conserving the past — Williamsburg, for example — while planners are
concerned with the shape of the future.
But who are these shapers of image and molders of form? What is their age, race,
ethnicity, sex and education? Where do they live and work and how much do they
earn? These questions will be addressed in this study.
Two principal sets of data will be presented and assessed. Of course, statistics
cannot explore quality and excellence in architecture and design, but they do provide
a means to understand the factual context from which quality and excellence emerge.
The first set of statistics is produced by the Bureau of the Census of the
Department of Commerce, the National Center for Education Studies of the
Department of Education, and the Bureau of Labor Statistics of the Department of
Labor. Their advantages and disadvantages are discussed in the general introduction
to this report. The second set of data is from membership organizations including the
American Institute of Architects, the American Society for Landscape Architects, the
Industrial Designers Society of America, the American Institute of Graphic Arts, and
the American Planning Institute. Advantages and disadvantages of such data sets are
also in the general introduction.
Architecture and Design Arts Occupations, 1970 to 1990 1113
Both data sets have been primarily collected from databases and special studies
conducted or commissioned by the Research Division of the National Endowment
for the Arts. These include the biennial Source Book of Arts Statistics. In addition, a
special data set was provided by Deirdre Gaquin of Washington (1990 Census of
Definitional difficulties occur when comparing Census of Population data with
that from the Bureau of Labor and from representative organizations. For example,
the Census of Population's Classified Index of Industries and Occupations identifies 13
job titles for architects, including landscape architects. Census of Population data is
presented for all 13 types of architects. In the case of Bureau of Labor Statistics data,
architects and landscape architects are reported separately, as are data from the
American Institute of Architects and the American Society for Landscape Architects.
In all cases, however, marine and naval architects are excluded.
For designers, the situation is worse. The Census of Population for 1980 and
1990 identifies at least 98 occupational titles under "Designer," ranging from
window trimmers to industrial designers to flower arrangers to fashion designers. By
contrast, the Occupational Handbook 1992-93 of the Department of Labor formally
defines six design occupations as follows: Industrial designers, interior designers, set
designers, fashion designers, textile designers and floral designers.
The Bureau of Labor Statistics, on the other hand, distinguishes between
designers, interior designers, and merchandising displayers and window trimmers. As
well, distinct data is available for graphic and industrial designers from the
representative associations. Educational data provide the finest degree of distinction
between the various design disciplines.
In addition, Census of Population data for different years have been derived from
different sample sizes. For example, data for the 1950 through 1980 Census (Citro
and Gaquin 1987) are derived from a 5 percent sample, while some Census data for
1970 through 1990 (Ellis, Beresford 1994) have been derived from a 16." 7 percent
sample. Such differences can cause analysis to be somewhat jerky, bouncing from one
data definition and sometimes sample size to another.
Regarding urban and regional planners and preservationists, no Census of
Population data were available for purposes of this study, so no trend analysis was
possible. Preservation is, at present, mainly a speciality practiced by architects,
designers and planners.
A "trend" refers to the general direction of a phenomenon over time. For
purposes of this study, this means the general direction of demographic
characteristics, education, employment and earnings of architects .\\m\ designers.
114 1 Artists in the Work Force
When discussing trends for a demographic or economic variable, a convenient
summary measure is the average rate of growth. Growth rates have three strengths.
First, while one cannot compare apples and oranges, one can quite properly compare
the rate of growth of apples and oranges. Second, growth rates, as a single number,
provide a succinct summary of trends. Third, theoretically, growth rates can be used
to project trends into the future, assuming the future reflects the past.
For purposes of this study, the restricted least squares method is used because it is
considered the best single indicator of the trend.
Comparative Occupational Groups
The general introduction to this NEA report on artists defines the three
occupational groups of experienced civilian labor force (ECLF), professional specialty
workers (PSW) and all artists.
Trends in architect and designer demographics, education, employment and
income will be compared to these three groups.
Between 1970 and 1990, the experienced civilian labor force grew at an average
rate of 24.3 percent each decade from 80.1 million to 123.5 million. In this same
time period professional specialty workers grew at an average rate of 21.5 percent
each decade from 1 1.7 million to 16.6 million workers. Between 1970 and 1990, all
artists grew at an average rate of 57.1 percent each decade from 1 1.7 million to 16.6
The number of architects between 1970 and 1990 grew at an average rate of 60.1
percent each decade from 56,125 to 157,759.
Until 1980, decorators and designers were recognized as distinct occupational
categories in the Census of Population. Drawing upon Citro and Gaquin (1987), the
distinction is maintained whenever possible. Between 1970 and 1990, decorators and
designers grew at an average rate of 78.9 percent each decade from 185,954 to 600,810.
Decorators increased between 1970 and 1990 at an average rate of 82.1 percent
each decade from 74,004 in 1970 to 240,800 in 1990. During this time designers
increased at an average rate of 76.9 percent each decade from 111,950 in 1970 to
360,000 in 1990.
The total number of artists was 6 percent of all professional specialty workers in
1970 and had increased to 10 percent by 1990. Architects represented 0.5 percent of
PSW in 1970 and then increased to 1 percent in 1990. Decorators and designers
advanced from 1.6 percent of all PSW in 1970 to 3.6 percent in 1990.
As a percent of all artists, architects went from 5 percent in 1970 to 9.4 percent
in 1990. During this time decorators and designers represented 27.7 percent of all
artists in 1970 and 35.8 percent in 1990.
To provide a basic understanding of architectural occupations, descriptions
derived from the Occupational Handbook 1992-93 (Bureau of Labor Statistics, 1993)
Architecture and Design Arts Occupations, 1 970 to 1 990 1115
will be used for architects and landscape architects.
The Handbook notes that architects provide a variety of services to individuals
and organizations from initial discussion with clients through constructions
requiring a variety of skills including design, engineering, managerial and
The architect and client must first discuss purposes, requirements and budget.
Based on these discussions, the architect prepares a report specifying requirements
and prepares drawings presenting ways to meet a client's needs.
After the initial proposal is accepted, the architect develops final construction
documents showing the building's appearance, including drawings of structural
systems (air-conditioning, electrical, heating, plumbing and ventilation) and
sometimes landscape plans. Architects also specify building materials and sometimes
interior furnishings. They must follow building codes, zoning bylaws, fire regulations
and such ordinances as access for the handicapped.
The architect may assist in getting construction bids, selecting a contractor and
negotiating contracts. S(he) may be engaged to ensure contractors follow the design,
use specified materials and meet quality standards. The job is not completed until all
construction is finished, required tests performed and construction costs paid.
Architects design a variety of buildings — offices, apartments, schools, churches,
factories, hospitals, houses and airports — as well as such multi-building complexes as
urban centers, college campuses, industrial parks and entire communities. In
addition to design, architects may advise on site selection, cost and land-use studies
and long-range land development.
Some specialize in one type of building, in construction management or in
managing their own firm, doing little design work. They often work with engineers,
urban planners, interior designers and landscape architects.
The Handbook notes that landscape architects design residential areas, parks,
campuses, shopping centers, golf courses, parkways and industrial parks to be
functional, beautiful, and compatible with their setting. They plan building
locations, roads and walkways, arrangements of flowers, shrubs and trees. They
redesign streets to limit car traffic and improve pedestrian access and safety. They also
work on natural resource conservation and historical preservation.
Landscape architects may be hired by organizations like real estate developers
starting new projects and municipalities constructing airports or parks. Often they
are involved from project conception and work with architects and engineers
determining the best arrangement of roads and buildings. Thev develop plans
indicating new topology, vegetation, walkways and landscape amenities.
Landscape architects discuss with clients the purpose of the project and funds
available. They analyze site elements such as climate, soil, slope, drainage and
vegetation. They observe the fall of sunlight and access to existing buildings, roads,
walkways and utilities. Then they prepare preliminary plans which are subject to
116 1 Artists in the Work Force
change. Many now use Computer Aided Design and Drafting (CADD) and video
simulators to help clients access proposals.
Working with other professionals in the design phase, once the design is
complete, landscape architects draw up detailed plans, including written reports,
sketches, models, photographs, land-use studies and cost estimates. Once the plan is
approved, they prepare working drawings showing existing and proposed features,
outline methods of construction and materials required. Many supervise installation
of their design. Some are involved in construction, but this is generally done by a
contractor or developer.
Some landscape architects work on a variety of projects, while others specialize
in residential, historical, restoration, water improvements, parks, playgrounds or
shopping centers. Others work in regional planning and resource development,
feasibility, environmental impact and cost studies, or site construction. Yet others
teach at the college or university level.
While Census data does not distinguish between types of architects, two data sets
provide a more detailed view of the profession: the American Institute of Architects
(ALA) and the American Society of Landscape Architects (AS LA).
If one accepts the 1990 Census count of 157,759 architects in the experienced
civilian labor force and, further, that membership in the ALA and ASLA is mutually ex-
clusive (which is not necessarily true), then the 56,802 ALA members represented 36.0
percent of all architects and the 10,443 ASLA members, 6.6 percent. The remaining
90,514 or 57.4 percent of Census architects were not affiliated with either organization.
Data concerning the age distribution of architects is available only from the
Census of Population. Accordingly, all architects are reported, including landscape
architects. (Exhibit 1)
Between 1970 and 1990 architects of all ages increased at an average rate of 60.1
percent each decade from 56,125 in 1970 to 157,759 in 1990.
Architects aged 16 to 24 increased at an average rate of 29.6 percent each decade
from 3,175 in 1970 to 7,245 in 1990.
Architects aged 25 to 34 years increased between 1970 and 1990 at an average
rate of 62.3 percent each decade from 15,300 in 1970 to 53,032 in 1990. As a
percent of all architects, they increased from 27.7 percent in 1970 to 33.6 percent in
Between 1970 and 1990 architects from 35 and 44 years increased at an average
rate of 92.6 percent each decade from 15,500 in 1970 to 52,256 in 1990. As a
percent of all architects they grew from 27.7 percent in 1970 to 33.1 percent in 1990.
Those aged between 45 and 54 years increased between 1970 and 1990 at an
average rate of 40.6 percent each decade from 1 1,900 in 1970 to 24,266 in 1990. As
Architecture and Design Arts Occupations, 1970 to 1990 I 117
Percentage Distribution of Census Architects
by Age Category. 1970 & 1990
45-54 Yrs \
a percent of all architects they decreased from 21.2 percent in 1970 to 15.4 percent
Architects from 55 to 64 years increased at an average rate 44.0 percent each
decade from 7,250 in 1970 to 15,438 in 1990. They decreased from 1 2.9 percent in
1970 to 9.8 percent in 1990 as a percent of all architects.
Those over 65 years increased between 1970 and 1990 an average rate of 47.0
percent each decade from 2,625 in 1970 to 5,522 in 1990. As a percent of all
architects they went from 4.7 percent in 1970 to 3.5 percent in 1990.
Ethnicity and Race
Data concerning ethnicity and race oi architects are presented from the 1970,
1980 and 1990 Census of Population. It reports all architects (Exhibit 2). Data ate
also presented for members of the .American Institute for Architecture. The two data
sets are not directly comparable.
Between 1970 and 1990 Hispanic architects increased at an average rate ol 124
percent each decade from 938 in 1970 to 8,006 in 1990. This was significantly faster
than growth of Hispanics in the general labor force (74 percent per decade), raster
than growth of Hispanics among professional specialty workers ( l) T(-> percent per
decade), and faster than growth in I [ispanics among artists in general ( 1 13.6 percent
118 1 Artists in the Work Force
Percentage Distribution of Census Architects by
Selected Racial Categories. 1970, 1980 & 1990
1,273 v - 1,601
or 2.4% \ / or 3%
' " )
3,013 v - 5,549
or 2.8% \ / or 5.2%
or 2.8% X
/ or 6.7%
per decade). As a percent of all architects, they increased from 1.8 percent in 1970 to
5.1 percent in 1990. The number of non-Hispanic architects decreased from 98.8
percent of all architects in 1990 to 94.9 percent in 1990.
Black or Afro-American architects increased at an average rate of 70.5 percent
each decade from 1,273 in 1970 to 4,429 in 1990. This was faster than growth of
whites (59.1 percent per decade), but significantly slower than growth among
architects of other races, for example Asians (118.7 percent). Black architects did,
however, increase in numbers faster than blacks in the general labor force (an average
rate of 26.6 percent per decade), faster than blacks among professional specialty
workers (at an average rate of 55.2 percent per decade), but slower than blacks among
Architecture and Design Arts Occupations, 1970 to 1990 I 119
artists in general (at an average rate of 72.3 percent per decade). As a percent of all
architects, blacks increased from 2.4 percent in 1970 to 2.8 percent in 1990.
If one compares 1990 Census data with reported 1989 members of the American
Institute of Architects. Census Hispanic architects were 5.1 percent of all architects
compared with 2.9 percent of AIA members. Census blacks represented 2.8 percent
of all architects compared to 1.7 percent of AIA members. Whites represented 90.5
percent of all Census architects compared to 89.4 percent of AIA members.
Data concerning the residence of architects is available from the Census of
Population and the two representative architectural associations. It is presented here for
the four principal Census regions: the Northeast, South, Midwest and West (Exhibit 3).
Percentage Distribution Architects by Region
1970 & 1990
Between 1970 and 1990 architects living in the Northeast increased at an average
rate of 69.6 percent each decade from 15,375 in 1970 to 41,596 In 1990. As a
percent of all architects, they decreased from 27.4 percent in 1970 to 26.4 percent in
1990. They were 26.4 percent of all architects in 1990.
Architects living in the South increased at an average rate of 60.3 percent each
decade from 14,150 in 1970 to 43,141 in 1990. They were 27.4 percent of all
architects in 1990.
1 20 I Artists in the Work Force
Midwestern architects increased at an average rate of 41 percent each decade from
12,975 in 1970 to 27,566 in 1990. They were 17.5 percent of all architects in 1990.
Western architects increased at an average rate of 66.5 percent each decade from
13,625 in 1970 to 45,456 in 1990. They represented 28.8 percent of all architects in
A comparison of the 1990 Census regional distribution with the 1990
membership in the American Institute of Architects and the 1991 membership in the
American Society for Landscape Architects shows that:
•the Northeast accounted for 26.4 percent of Census architects, 21.5 percent of
AIA members and 21.4 percent of ASLA members.
•the South accounted for 27.4 percent of Census architects, 31.8 percent of AIA
members and 34.7 percent of ASLA members.
•the Midwest accounted for 17.5 percent of Census architects, 19.9 percent of
AIA members and 17.4 percent of ASLA members, and
•the West accounted for 28.8 percent of Census architects, 26.4 percent of AIA
members and 26.5 percent of ASLA members.
Data concerning the gender of architects is available from the Census of
Population and from reporting members of the AIA.
Between 1970 and 1990 female architects increased on an average rate of 180.1
percent per decade from 2,075 in 1970 to 23,723 in 1990. This compares with an
average rate growth among women in the ECLF of 24.5 percent, of 41.1 percent
among PSW and 86.9 percent among all artists. As a percentage of all architects,
women went from 3.7 percent in 1970 to 17.7 percent in 1990.
Comparing the 1990 Census with the reported 1989 AIA members shows
women were 17.7 percent of Census architects and 9.2 percent of AIA members.
Education and Licensing Requirements
According to the Occupational Handbook, all states and the District of Columbia
require licensing before calling oneself an architect or contracting to provide
architectural services. While many architectural school graduates work in the field
without being licensed, a licensed architect is legally responsible for all work.
Licensing usually requires a professional architectural degree, a period of practical
training or internship (usually three years) and passing the Architect Registration
In most states the professional degree is from one of 96 architectural schools
accredited by the National Architectural Accrediting Board. There are several types
of degree. Over half of all architects have a five-year Bachelor of Architecture. There
is a two-year Master's with a pre-professorial degree in architecture or a selected area
and three- or four-year Master's for those with a degree in another discipline. Many
combinations and variations of these programs also exist.
Architecture and Design Arts Occupations, 1970 to 1990 I 121
A typical five-year Bachelor program includes architectural history and theory,
building design including technical and legal requirements, math, physical sciences
and liberal arts. Many schools also offer graduate programs beyond a professional
degree for research, teaching and certain specialties.
With respect to landscape architects, the Handbook notes that a Bachelor's or
Master's degree in landscape architecture is usually required. The Bachelor's program
is usually four to five years. Of the two types of Master's degrees, a three-year program
is for those with a Bachelor's degree in another discipline (most common), and a two-
year program is for those with a Bachelor's in landscape architecture.
In 1990, 51 colleges and universities offered 61 undergrad and grad programs
accredited by the American Society for Landscape Architecture. Typically, courses
include surveying, landscape design and construction, ecology, structural design, city
and regional planning, history of landscape architecture, plant and soil science,
geology, design and color theory and general management. The design studio is an
important aspect of the curriculum. Students are assigned real projects providing
hands-on experience. Prerequisites often include English, math and social science.
Some 44 states require licensing based on the Uniform National Examination
(UNE), admission to which usually requires a degree from an accredited school plus
one to four years work experience. Some states require an additional exam on law and
plants indigenous to that state. The federal government does not require a license.
In states requiring licenses, entrants are called interns until licensed. They may
do research, prepare base maps or participate in actual design work, supervised by a
licensed landscape architect who takes legal responsibility. After several years beyond
receiving their license, they may become associates or partners, or open their own offices.
Data concerning the educational attainment of architects is available from the
Census of Population, the Department of Education and the American Society for
Architects with only elementary education declined at an average rate of -56.1
percent each decade from 1 , 1 50 in 1 970 to 1 58 in 1 990. As a percent of all architects
they decreased from 2.1 percent in 1970 to 0.1 percent in 1990. (Exhibit 4)
Architects with one and three years of high school education decreased at an
average rate of -8.6 percent each decade from 1,600 in 1 1 )~0 to 1,150 in 1990.
Compared to all architects, they declined from 2.9 percent in 1970 to 0.7 in 1990.
Between 1970 and 1990 architects with four wars of high school increased at an
average rate of 13.1 percent each decade from 4, v 25 in 1970 to 6,676 in 1 990. As a per-
cent of all architects, they decreased from 8.4 percent in 1970 to 4.2 percent in 1990.
Architects with one to three years of college or university education increased
between 1970 and 1990 at an average rate of 68.8 percent each decade from 7,500
in 1970 to 23,256 in 1990. Compared to all architects, they increased from 13.4
percent in 1970 to 14.7 percent in 1990.
Architects with four years or more of college or university education increased at
122 I Artists in the Work Force
Percentage Distribution of Census
Architects by Education
1970 & 1990
Percentage Distribution of Degrees Awarded in
Architecture and Landscape Architecture
an average rate of 68.4 percent each decade from 41,150 in 1970 to 126,519 in
1990. As a percent of all architects, they increased from 73.3 percent in 1970 to 80.2
percent in 1990.
Degrees and Enrollment
Using Department of Education data, in 1988-89 there were 6,386 college or
university degrees awarded in architecture at the Bachelor level (73.3 percent of
degrees), Master's (26.2 percent) and Doctoral (0.5 percent). There were 1,164
degrees in landscape architecture awarded at the Bachelor level (75-9 percent) and at
the Master's level (24.1 percent). Data from the American Society for Landscape
Architecture indicate an increasing professionalism in the field. In 1971 there were
22 accredited programs in landscape architecture of which 18.2 percent were at the
Master's level. By 1991 there were 64 accredited programs of which 36.1 percent
were at the Master's level. Female students in accredited programs in 1990-91
represented almost 31 percent of all students, compared to 15 percent of 1990
Architecture and Design Arts Occupations, 1970 to 1990 I 123
Data concerning the employment of architects are available from the Census of
Population, Census of Service Industries and from two representative associations.
Census data do not distinguish landscape architects from architects in general.
Architects employed in the private sector increased at an average rate of 76.2
percent each decade from 28,225 in 1970 to 92,029 in 1990. As a percent of all
architects, they increased from 49.3 percent in 1970 to 59.8 percent in 1990. (See
Between 1970 and 1990 architects employed in the public sector increased at an
average rate of 25.3 percent from 6,775 in 1970 to 1 1,208 in 1990. As a percent of
architects, they decreased from 12.2 percent in 1970 to 7.3 percent in 1990.
Self-employed architects increased at an average rate of 47.1 percent each decade
from 20,375 in 1970 to 50,535 in 1990. As a percent of all architects, they decreased
from 36.7 percent in 1970 to 32.8 percent in 1990.
The unemployment rate for architects was 1.2 percent in 1970 and 2.4 percent in
1990. This compares with experienced civilian labor force unemployment of 4.1 percent
in 1970 and 5.5 percent in 1990. For professional special workers the corresponding
rates were 1.8 percent unemployed in 1970 and 2.1 percent in 1990. For all artists, the
corresponding rates were 4.5 percent in 1970 and 4.8 percent in 1990.
The 1980 and 1990 Census shows that full-time architects increased as a percent-
age of all architects from 70.7 percent in 1980 to 73.2 percent in 1990. Accordingly,
more than a quarter of all architects work only part-time. Female architects accounted
for 1 1.8 percent of full-time architects in 1990, but 25.7 percent of part-time architects.
Growth Rate of Census Architects by Class of Work
% 25 50 75
i i i
124 I Artists in the Work Force
While over 75 percent of Census architects in 1990 were employed in the
professional service industries (mainly architectural, engineering and surveying
firms), architects were also employed elsewhere. Of the other industries reporting
architects, 12 percent were in agriculture, 3.5 percent in public administration, 2.8
percent in both construction and manufacturing, 2.1 percent in transportation,
communications and public utilities, 2.1 percent in educational services, 1.4 percent
in retail trade, finance, insurance, real estate and business and repair services
industries (Exhibit 6).
Percentage Distribution of Census Architects
by Major Industries. 1990
Agriculture 17,000 or 12.0% M
Construction 4,000 or 2.8%
Manufacturing 4,000 or 2.8%
Transportation, Communications & Public Utilities
3,000 or 2.1%
Retail Trade 2,000 or 1 .4%
Finance, Insurance & Real Estate 2,000 or 1.4%
Business & Repair Services 2,000 or 1 .4%
Educational Services 3,000 or 2.1%
Public Administration 5,000 or 3.5%
"excludes 107,000 in Professional Services or 75.4%
The Census of Service industries provides a breakout of architects from
landscape architects. In 1987, within the broad category called Construction,
Finance and Service Industries, 62,520 architects were employed, representing
0.5 percent of employment in industries reporting these occupations. Of these, 85
percent were general architects working in miscellaneous service industries; 8.4
percent were landscape architects employed in miscellaneous service industries; 4
percent were general architects employed in the construction industries and 2.7
percent were general architects in business services industries.
Architectural services are provided by three types of businesses: architectural,
engineering and surveying service establishments. In 1982 engineering
establishments employed 5,218 architects (compared to 31,871 by architectural
Architecture and Design Arts Occupations, 1970 to 1990 I 125
firms), while in 1987 surveying establishments employed 158 architects (compared
to 40,583 by architectural service firms). For purposes of this analysis, no further
reference will be made to engineering or surveying service establishments.
The number of architectural services firms increased from 13,414 in 1982 to
17,777 in 1987. Total receipts increased from $5.9 to $9.9 billion. Paid employees
increased from 105,270 to 136,809, while architects as a percent of total
employment declined from 30.3 percent to 29.7 percent. The number of sole
proprietors increased from 8,039 in 1982 to 8,950 in 1987. Compared to all
establishments, sole proprietorships declined from 59.9 percent in 1982 to 50.4
percent in 1987.
Regarding the distribution of architectural establishments, staff and fees by
project type, source and client for 1982 and 1987, in-house projects generated $5.1
billion in 1982 and $8.6 billion in 1987. Of total in-house work, commercial
buildings accounted for more than 40 percent, while public and institutional
facilities accounted for more than 25 percent in each year. All other types of projects
accounted for less than 1 percent of revenues.
As to the source of receipts including work done outside of architectural firms,
on average for both years: architectural services, excluding landscape architecture,
produced more than 75 percent of revenues; work done outside, but reimbursable,
more than 10 percent; consulting and design engineering more than 6 percent and
all other activities accounted for the balance.
Regarding fees from clients, on average: industrial, business and commercial
clients paid more than 33 percent; government more than 23 percent; private
institutions more than 17 percent; private individuals more than 8 percent and all
other clients, less than 19 percent.
Regionally, the Northeast accounted for 21.9 percent of all establishments in
1987 and 24.1 percent of all receipts; the South for 32.1 percent of the firms and
29.4 percent of receipts; the Midwest for 18.1 percent of firms and receipts and the
West for 28.7 percent of firms and 28.4 percent of receipts.
By contrast, firms in the Northeast belonging to the American Institute for
Architecture accounted for 23 percent of all AIA firms; the South for 32 percent; the
Midwest for 1 5 percent and the West for 30 percent.
Data about the income of architects are available from the Census of Population
and two representative associations. Census data due to definition changes are
presented only for 1980 and 1990, so no meaningful growth rate analysis is possible.
In 1990 architects working full-time and earning $7,500 or less accounted for
1.9 percent of all architects; those earning $7,500 to $14,999 in the year before the
Census, 3.3 percent; $15,000 to $24,999, 13.6 percent; $25,000 to $3 &,999, 23.4
percent; $35,000 to $49,999, 29.4 percent; $50,000 to $69,999, 16.2 percent;
$70,000 to $99,999, 6.2 percent and those earning more than Si 00,000, 6 percent
126 Artists in the Work Force
- - ■ I -
'edian income for full-time male architects was S40,l 10 and for females,
_ 451 * "3.4 percent of males. Median full-time earnings of male architects were
141 percent of the median full time earnings of males in the experienced civilian
labor force; equal to earnings of male professional specialty workers and 129 percent
of artists median full-time earnings.
-chitects living in households with an annual income under S 1 5-000 in the year
before the Census accounted for 3.4 percent of all architects; those earning $15,000
to S24.999, 5.9 percer 34,999, 11.4 percent; $35,000 to $49,999,
20.4 p j,000 to $64,999, 19.2 percent; $65,000 to $94,999, 23.4 percent;
S9: Do SI 24,999, 8.2 percent and those earning more than $125,000 the year
before the Census, 8 percent.
The median household income for male architects was $56,952 and for females,
r ,639 or 9".~ 7 percent of males. The median household income with either a male
or female architect was 5 5 3. Median household income of all architects was 140
percent that of the experienced civilian labor force, 109 percent of professional
specialty* workers and 128 percent of art:
Exhibit 8 shows the median annual compensation of ALA members by position
Members of the .American Society* for Landscape Architects working in the
private sector in 1991 had an annual median income, from all sources, of $43,575.
Architecture and Design Arts Occupations, 1970 to 1990 I 127
Median Annual Compensation of Members of the
American Institute of Architects by Position. 1990.
4,282 Positions in
1 ,655 Positions in 686 Firms
1 ,234 Positions in 412 Firms
Architect III $37,000
1 ,262 Positions in 427 Firms
Architect II $33,800
4,282 Positions in 2,481 Firms
Architects I $30,000
4,282 Positions in 2,481 Firms
2.884 Positions in 1,224 Firms
Landscape Architect $33,000
120 Positions in 62 Firms
Those working in the public sector earned $41,475, while those in academic
positions earned $49,350.
To provide a basic understanding of design occupations, descriptions derived
from the Occupational Handbook 1992-93 Bureau of Labor Statistics, 1993) are given
Designers organize and design articles, products and materials to serve a purpose
and to be visually pleasing. Pleasant surroundings, beautiful clothes and floral
arrangements boost our spirits, while eye-catching products and packaging are more
likely to attract buyers. Designers usually specialize, for example on cars, furniture,
home appliances, industrial equipment, movie and theater sets, packaging, flower
In developing a design they first determine the needs of the client and potential
users. They consider size, shape, weight, color, materials and the way a product
functions, as well as maintenance, safety and cost. They take into account — and
often set — style and fashion trends. They usually sketch several possible designs
which are present for final selection to an art or design director, a product
1 28 I Artists in the Work Force
development team, a play, film or television director, or a client.
The designer then makes a model, sample or detailed plan drawn to scale.
Increasingly, Computer Aided Design and Drafting (CADD) is used, while industrial
designers use Computer Aided Industrial Design (CAID) to create a design and
communicate it to automated production tools.
Designers may supervise craft workers who carry out the design. Owners may
devote much time to developing business contacts and to administrative tasks like
reviewing catalogs and samples.
Design is not one but a number of fields including:
• Industrial Designers develop and design manufactured products like appliances,
cars, computers, medical, office and recreation equipment and children's toys.
They combine artistic talent with market research on product use, marketing,
materials and production methods to create the most functional and appealing
design and make products competitive in the marketplace.
• Interior Designers plan space and furnish interiors of homes, hotels, offices,
public buildings, restaurants, stores and theaters. With a client's tastes and needs
in mind they prepare working drawings and specifications for interior
construction, furnishings, lighting and finishes including crown moldings,
coordinating colors and selecting furnishings, floor and window coverings. They
also plan additions and renovations. They must design in accordance with
federal, state and local building codes.
• Set Designers study scripts, confer with directors and conduct research to
determine appropriate styles and then design sets for film, television and theater.
• Fashion Designers design wearing apparel and accessories. Some high-fashion
designers are self-employed and design for individual clients. They make fashion
by establishing the "line" and colors. Some cater to specialty stores or high-
fashion department stores. They design original garments, as well as follow
established trends. Most work for apparel manufacturers and adapt clothing to
the mass market.
• Textile Designers design fabrics for garments, upholstery, rugs and other
products using their knowledge of materials and fashion trends.
• Floral Designers cut and arrange fresh, dried or artificial flowers and foliage into
designs expressing the sentiments of the sender. Duties depend on the size of the
shop and the number of designers.
While Census data does not distinguish between types of decorators and
designers, it does identify 98 distinct occupations. Two data sets provide a more
detailed view of the profession. The first is from the Industrial Designers Society of
America (IDSA). The second is from the American Institute of Graphic Arts.
Summary findings on data from all federal sources and from the two representative
associations are given below.
Architecture and Design Arts Occupations, 1970 to 1990 I 129
Data concerning the age distribution of decorators and designers is available only
from the Census of Population and only for combined decorators and designers. The
combined group of all ages increased between 1970 and 1990 at an average rate of
78.9 percent each decade from 185,954 in 1970 to 600,810 in 1990.
The combined group aged 16 to 24 years increased at an average rate of 42.3
percent each decade from 28,765 in 1970 to 65,526 in 1990. As a percent of all
combined decorators and designers, they declined from 15.5 percent in 1970 to 10.9
percent in 1990 (Exhibit 9).
Percentage Distribution of Census Combined Decorators
Designers by Age Category. 1970 & 1990.
Decorators and designers aged 25 ro 34 years increased at an average rate of 91.9
percent each decade from 48,478 in 1970 to 200,628 in 1990. As a percent of
combined decorators and designers they increased from 26.1 percent in 1970 to 33.4
percent in 1990.
Combined decorators and designers aged 35 to 44 years increased at an average
rate of 112.2 percent each decade from 42,900 in 1970 to 169,075 in 1990. As a
percent of all the combined group they increased from 2}. 1 in 1 1 )~*0 to 28.1 in 1990.
Between 1970 and 1990 decorators and designers aged 45 to 54 increased at an
average rate of 71.4 percent each decade from 36,468 in 1970 to 97,821 in 1990. As
1 30 I Artists in the Work Force
a percent of all decorators and designers they decreased from 19.6 percent in 1970 to
16.3 percent in 1990.
The group of decorators and designers aged 55 to 64 years increased at an
average rate of 34.5 percent each decade from 29,039 in 1970 to 50,388 in 1990. As
a percent of all decorators and designers, they decreased from 15.6 percent in 1970
to 8.4 percent in 1990.
The group 65 years and older increased at an average rate of 54.6 percent each
decade from 7,354 in 1970 to 17,372 in 1990. As a percent of all decorators and
designers, they decreased from 4 percent in 1970 to 2.9 percent in 1990.
Ethnicity and Race
Data concerning ethnicity and race of combined decorators and designers is
presented from the 1970, 1980 and 1990 Census of Population.
Between 1970 and 1990 Hispanic decorators and designers grew at an average
rate of 130 percent each decade from 6,815 in 1970 to 32,296 in 1990. This was
significantly faster than growth of Hispanics in the general labor force (an average
rate of 74 percent per decade), faster than growth of Hispanics among professional
specialty workers (an average rate of 97.6 percent per decade) and faster than growth
in Hispanics among artists in general (an average rate of 113.6 per decade). As a
percent of all decorators and designers, they increased from 2.9 percent in 1970 to
5.3 percent in 1990.
Between 1970 and 1990 black or Afro-American decorators and designers
increased at an average rate of 106.5 percent each decade from 4,445 in 1970 to
21,204 in 1990. This was faster than growth of whites (an average rate of 59.7
percent per decade), but significantly slower than growth of decorators and designers
of other races (an average rate of 180.5 percent). Black decorators and designers did,
however, increase faster than blacks in the general labor force (an average rate of 26.6
percent per decade), faster than blacks among professional specialty workers (an
average rate of 55.2 percent per decade) and faster than blacks among all artists (an
average rate of 72.3 percent per decade). As a percent of all decorators and designers,
blacks went from 1.9 percent in 1970 to 3.6 percent in 1990. Whites declined from
96 percent in 1970 to 90.2 in 1990. Other races increased from 2.1 percent in 1970
to 6.2 percent in 1990 (Exhibit 10).
Data concerning the residence of decorators and designers came from the Census
of Population, the Industrial Designers Society of America and the American
Institute of Graphic Arts. It is presented for the Northeast, South, Midwest and West
Designers and decorators living in the Northeast increased at an average rate of
44.8 percent each decade from 75,766 in 1970 to 149,888 in 1990. Compared to all
designers, they decreased from 32.5 percent in 1970 to 25 percent in 1990.
Architecture and Design Arts Occupations, 1970 to 1990 I 131
Percentage Distribution of Census Combined
Decorators & Designers by Selected
Racial Category. 1970, 1980 & 1990.
Percentage Distribution of Census Combined
Decorators & Designers by Region
1970 & 1990
Those living in the South increased at an average rate of 78.5 percent each
decade from 53,886 in 1970 to 168,390 in 1990. Compared to all designers, they
grew from 23.1 percent in 1970 to 28 percent in 1990.
Midwestern designers and decorators increased at an average rate of 58.5 percent
from 58,764 in 1970 to 137,005 in 1990. As a percent of the total, they decreased
from 25.2 percent in 1970 to 22.8 percent in 1990.
Designers and decorators in the West increased at an average rate of 86.9 percent
1 32 I Artists in the Work Force
each decade from 44,474 in 1970 to 145,527 in 1990. As a percent of all designers,
they increased from 19.1 percent in 1970 to 24.2 percent in 1990.
By comparison, for 1987 members of the American Institute for Graphic Arts:
Northeast, 42 percent; South, 17 percent; Midwest, 17 percent and West
22 percent. And for members of the Industrial Designers Society of America:
Northeast, 29.7 percent; South 16 percent; Midwest 31.9 percent and West, 22.4
Data concerning gender is available from the Census of Population and reporting
members of the American Institute of Graphic Arts.
Between 1 970 and 1 990 female decorators and designers increased at an average
rate of 106.4 percent per decade from 71,262 in 1970 to 333,032 in 1990. This
compares with an average rate growth per decade of women among the experienced
civilian labor force of 24.5 percent; among professional specialty workers, 41.1
percent and among all artists, 86.9 percent. As a percentage of all decorators and
designers, women increased from 38.3 percent in 1970 to 55.4 percent in 1990.
Female decorators increased at an average rate of 100 percent per decade from
30,717 in 1970 to 176,500 in 1990. As a percentage of all decorators, women
increased from 58.5 percent in 1970 to 73.3 percent in 1990.
Female designers also increased from 1970 to 1990, growing at an average rate
of 1 16 percent per decade from 27,975 in 1970 to 156,500 in 1990. As a percentage
of all designers women went from 25 percent in 1970 to 43.5 percent in 1990.
By contrast, in 1987, 46 percent of reporting members of the American Institute
for Graphic Arts were women.
According to the Occupational Handbook (Bureau of Labor Statistics, 1993),
creativity is crucial for designers and decorators, together with a strong color sense,
an eye for detail, balance and proportion and sensitivity to beauty. Sketching is
especially important for fashion design. Some formal preparation in design is
Important in all but floral design.
Educational requirements for entry vary. Industrial design requires a Bachelor's
degree, interior design a four-year Bachelor's in fine art. Interior designers must also
be familiar with federal, state and local building codes, as well as toxicity and
flammability standards. In fashion design some formal education, such as a two-to
four-year degree is important, plus knowledge of textiles, fabrics, ornamentation and
fashion trends. Floral designers usually
need only a high school degree as most learn on the job.
Formal training in some design disciplines is available from professional schools
offering certificates or associate degrees. Four-year college and university programs
grant a Bachelor of Fine Arts. The curriculum includes art and art history, principles
Architecture and Design Arts Occupations, 1970 to 1990 I 133
of design, designing and sketching. Also, specialized programs like garment
construction, textiles, mechanical and architectural drawing, computerized design,
sculpture, architecture, marketing and basic engineering. Persons with architectural
training also qualify for some design occupations, especially interior design.
Computer aided design is taught, especially in industrial design.
In 1991, the National Association of Schools of Ait and Design accredited 166
post-secondary institutions in art and design. Most award a degree in art, some in
industrial, interior, textile, graphic or fashion design. Many allow entry into a
Bachelor's program only after a year of basic art and design courses.
The Foundation for Interior Design Education Research accredits interior design
programs and schools. There are 89 accredited programs in the U.S. and Canada
located in schools of art, architecture and home economics. Some colleges and
universities offer degrees in floriculture and floristry and provide training in flower
marketing and shop management. Floral design is also taught in private schools.
Interior design is the only design discipline subject to government regulation.
The District of Columbia licenses and 14 states regulate use of the title. Membership
in a professional association is a mark of achievement, usually requiring completion
of three or four years of post-secondary education in the field, at least two years of
practical experience and completion of the National Council for Interior Design
Data concerning the educational attainment of decorators and designers is
available from the Census of Population (for combined decorators and designers),
from the Department of Education and from the American Institute of Graphic Arts.
Decorators and designers with only elementary education declined from 10,823
in 1970 to 9,745 in 1990. Compared to all decorators and designers, they decreased
from 5.8 percent in 1970 to 1.6 percent in 1990 (Exhibit 12).
Those decorators and designers with one to three years of high school increased
at an average rate of 23.8 percent each decade from 22,869 in 1970 to 33,823 in
1990. Compared to all decorators and designers, however, they decreased from 12.3
percent in 1970 to 5.6 percent in 1990.
Those with four years of high school increased at an average rate of 29.8 percent
per decade from 62,743 in 1970 to 111,573 in 1990. They decreased, however,
compared to all in the field from }?>.7 percent in 1970 to 18.6 in 1990.
Decorators and designers with one to three years of college or university
increased at an average rate of 120.5 percent each decade from 49,375 in 1970 to
230,409 in 1990. Compared to all in the field they increased from 2d. 6 percent in
1970 to 38.4 percent in 1990.
Those with four or more years of college or universit) education increased at an
average rate of 1 17.2 percent each decade from 40,144 in 1970 to 21 S,24() in 1990.
Compared to all decorators and designers they increased from 21.6 percent in 1970
to 35.8 percent in 1990.
1 34 i Artists in the \\ ork Force
Percentage Distribution of Census Combined Decorators &
Designers by Education. 1970 & 1990
Degrees and Enrollment
Department of Education data shows that 5,054 college or university degrees
were awarded in 1988-89 in design at the Bachelor level (93.3 percent) and Master's
level (6.7 percent).
In 198^ there were 174 graduate design programs offered by American colleges
and universities of which 35.1 percent were in graphic design, 13.2 percent in
industrial design, 28.8 percent in interior design and 23 percent in textile design. At
both the Bachelor and Master level, percentages of students enrolled and graduated
were highest in graphic design. Interior design, industrial design and
communications design were the other three areas largest in enrollment and degrees
By contrast with 1990 Census figures, of reporting members of the American
Institute for Graphic Arts, 13 percent had roughly three years of college or university
compared to 38.4 percent for all decorators and designers. And 8~ percent had four
years or more of college or university' education compared to 35.8 percent for all
decorators and designers.
Architecture and Design Arts Occupations, 1970 to 1990 I 135
Data concerning the employment of decorators and designers is available from
the Census of Population, the Census of Service Industries and from representative
Decorators and designers employed in the private sector increased at an average
rate of 69.9 percent each decade from 144,554 in 1970 to 416,352 in 1990. As a
percent of all decorators and designers, they decreased from 80.4 percent in 1970 to
72.1 percent in 1990.
Those employed in the public sector increased at an average rate of 68.4 percent
each decade from 5,105 in 1970 to 15,046 in 1990. Compared to all in the field,
they declined from 2.8 percent in 1970 to 2.6 percent in 1990.
Self-employed decorators and designers increased at an average rate of 1 1 5. 1 each
decade from 28,586 in 1970 to 142,178 in 1990. As a percent of the field, they
increased from 15.9 percent in 1970 to 24.6 percent in 1990.
The unemployment rate for combined decorators and designers was 3.3 percent
in 1970 and 3.9 percent in 1990. This compares with the experienced civilian labor
force rate of unemployment of 4.1 percent in 1970 and 5.5 percent in 1990. For
professional speciality workers the corresponding rates were 1.8 percent in 1970 and
2.1 percent in 1990. For all artists the corresponding rates were 4.5 percent in 1970
and 4.8 percent in 1990.
Data for full- and part-time decorators and designers are from the 1980 and
1990 Census. Drawing upon work by Ellis and Beresford (1994), full-time
decorators and designers increased compared to all decorators and designers from
53.8 percent in 1980 to 54.4 percent in 1990. Accordingly, nearly half of all
decorators and designers worked only part-time. Female decorators and designers
accounted for 45.5 percent of full-time workers in 1990, but 72 percent of part-time.
Exhibit 13 presents the distribution of Census combined decorators and
designers by major industries in 1990. Over 80 percent were employed in retail trade
(29.4 percent), business and repair industries (26.4 percent) and manufacturing
industries (24.9 percent). Other professional services industries employed 7.7
percent. There were 2.6 percent in wholesale trade; 2.1 percent in transportation,
communications and public utilities; 1.5 percent in construction, entertainment and
recreation; 1.3 percent in educational services and less than 1 percent in finance,
insurance, real estate, public administration, agricultural industries, personal services
industries and health services. Decorators and designers represented 0.5 percent of
total employment in industries reporting these occupations.
In 1987 the broad Census category called Construction, Finance and Service
industries accounted for 61,130 employed decorators and designers, including
48,040 designers (excluding interior designers) and 15,090 interior designers. In
total, they represented 0.5 percent of total employment in industries reporting these
1 36 I Artists in the Work Force
Percentage Distribution of Census Combined Decorators &
Designers by Major Industries. 1990.
Total Decorators & Designers 531,000
I Agriculture 3,000 or 0.6%
Construction 8,000 or 1 .5%
132,000 or 24.9%
156,000 or 29.4%
Business & Repair Services
140,000 or 26.4%
Personal Services excl. Private Households 1 ,000 or 0.2%
Entertainment & Rec. Services 8,000 or 1 .5%
Health Services excl. Hospitals 1,000 or 0.1%
I Educational Services 7,0000 or 1 .3%
Other Professional Services 41 ,000 or 7.7%
Public Administration ;4,000 or 0.8%
occupations. Of these 65.6 percent worked in business services industries; 29.3
percent in miscellaneous service industries; 4.1 percent in construction; 2.1 percent
in amusement and recreation services; 1.6 percent in motion pictures and
0.6 percent in museums, botanical and zoological parks.
In the broad category called Non-Manufacturing Industries in 1988 there were
244,020 decorators and designers made up of 140,770 designers (excluding interior
designers), 43,580 interior designers and 59,670 merchandise displayers and window
trimmers representing 0.5 percent of total employment in industries reporting these
occupations. Of these 43.9 percent were in retail trade; 11.7 percent in wholesale
trade; 3.4 percent in electric, gas and sanitary services and 0.2 percent in
In the broad category called Manufacturing Industries in 1989 there were
39,890 made up of designers (excluding interior designers), representing 0.5 percent
of total employment of the industries reporting these occupations. Of these 15.2
percent were in apparel and other textile products; 12.7 percent in printing and
publishing; 12.1 percent in transportation equipment; 10.8 percent in industrial
machinery and equipment; 10.3 percent in miscellaneous industries and fewer than
10 percent in all other reporting industries.
Architecture and Design Arts Occupations, 1970 to 1990 I 137
The Census of Service Industries offers insight into the establishments providing
graphic arts services. Two types of businesses provide these services: commercial art
and graphic design firms and graphic design firms. For purposes of this analysis, no
further reference will be made to commercial art and graphic design firms.
In 1987 there were 7,202 graphic design establishments with receipts of $3.2
billion. The Northeast accounted for 30 percent of establishments and 34.3 percent
of receipts; the South for 23 percent of establishments and 17.5 percent of receipts'
the Midwest for 22.3 percent of establishments and 25.5 percent of receipts and the
West for 24.6 percent of establishments and 22.7 percent of receipts.
With respect to type of practice, billings and employees, the Industrial Design
Society of America reported that in 1989 of responding groups 56.6 percent were
consulting groups accounting for 52.3 percent of design employees; 39.6 percent
were corporate design groups accounting for 45.6 percent of design employees and
3.8 percent were other types of groups accounting for 2.2 percent of design
employees (Exhibit 14).
Percentage Distribution of Industrial Design Society of
America Groups by Practice, Billings & Desgin Employees
9 or 3.8%
33 or 56.6%
93 or 39.6%
67 or 28.5%
54 or 23.0%
114 or 32.0%
60 or 25.5%
Groups reporting billings up to $249,999 a year accounted for 23 percent of all
reporting groups and 6.1 percent of design employees; groups with billings between
$250,000 and $499,999 accounted for 25.5 percent of groups and 13 percent of
employees; groups with billings between $500,000 to $999,999 accounted for 23.0
1 38 I Artists in the Work Force
percent of groups and 18.6 percent of employees; Groups with billings of more than
$1 million accounted for 28.5 percent of all groups and 62.3 percent of design
The American Institute of Graphic Art reported that in 1987 free-lancers were 7
percent of its members; 21 percent were self-employed; 28 percent were owners or
partners of firms; 44 percent were employees and 1 percent were unemployed.
In 1987 of reporting AIGA members 46 percent were employed in a design
firm; 28 percent in a non-design firm; 13 percent in educational institutions; 8
percent in a publishing house; 5 percent in non-profit institutions; 2 percent in
governmental institutions and 2 percent in other types of organizations.
Data concerning the income of decorators and designers is available from the
Census of Population for 1980 and 1990 and from the two representative associations.
In 1990 decorators and designers working full-time and earning $7,500 or less
accounted for 5.4 percent of all decorators and designers; those earning $7,500 to
$14,999 in the year before the Census, 14.4 percent; between $15,000 and $24,999,
25 percent; between $25,000 and $34,999, 22.3percent; between $35,000 and
$49,999, 19.4 percent; between $50,000 and $69,999, 8.6 percent; between
$70,000 and $99,999, 3.1 percent and those earning more than $100,000, 1.9
percent. (Exhibit 15)
The median income for full-time earnings of male decorators and designers was
$32,549 and for females, $20,394, or 62.7 percent of what their male counterparts
Decorators and designers living in households with an annual income under
$15,000 in 1989 accounted for 7.8 percent of all decorators and designers; those
earning between $15,000 and $24,999, 1 1.2 percent; between $25,000 to $34,999,
14.8 percent; between $35,000 and $49,999, 21.7 percent, between $50,000 and
$64,999, 17.1 percent; between $65,000 and 94,999, 16.2 percent; between $95,000
and $124,999, 5.6 percent and those earning more than $125,000, 5.5 percent.
The median income for male decorators and designers was $47,688 and for
females, $44,308. Median income for females was 92.9 percent of males. The median
household income with a working decorator or designer, either a male or female, was
Median household income of all decorators and designers was 1 13 percent of the
experienced civilian labor force; 88.2 percent of professional specialty workers and
103 percent of all artists.
A summary of findings concerning architecture and design occupations between
1970 and 1990 appears in the Executive Summary at the beginning of this NEA
report on artists and will not be repeated here. Rather, this section will briefly treat
Architecture and Design Arts Occupations, 1970 to 1990 I 139
Percentage Distribution of Census Combined Decorators &
Designers by Full-time Earnings in Year before Census
1980 & 1990
$100,000 or more
2,193 or 1.1%
32,193 or 16.0%
48.302 or 26.3%
"50,736 or 25.2%
$100,000 or more
6,780 or 1.9%
19,413 or 5.4%
52,035 or 14.4%
90,522 or 25.0%
80.595 or 22.3%
four aspects of professionalization and competition which affect the employment and
earnings in architecture and design professions.
Professionalization and Competition
Since the Industrial Revolution, the engineer has been at the vanguard of
integrating scientific knowledge into the physical structures, instruments and utensils
of daily life. With the ascendancy of science, the engineer has displaced the architect
as the "master builder." It is rumored that fewer than 10 percent of all construction
projects in the U.S. engage architects, with the remaining 90 percent reputedly in the
hands of engineers. To some developers architects are too concerned with aesthetics
and not enough with function, cost and efficiency.
If architects are considered "soft" compared to engineers, then designers are
considered "soft" by architects. Thus while all states and the District of Columbia
require formal licensing of architects, only the District licenses and only 14 states
regulate use of the term "interior designer." Tension between the two professions was
evident in 1990 with passage of the LaValle-Koppell Bill in New York. The bill
140 I Artists in the Work Force
resulted from an agreement between the architectural profession and interior
designers on the scope of designers' work. In essence the bill: established a legal
definition of interior design; established a category of "certified interior designers"
and set minimum standards for education, training and testing; distinguished
interior decorators from interior designers and defined the type of interior
construction designers can perform as "not materially related to or materially
affecting the building systems."
Competition and growing demand for interior design has resulted in many
architecture firms forming interior design departments.
With respect to architecture, the U.S. is doing well on world markets. In 1989
there were 200 "design" firms competing for international contracts worth $7.4
billion. Design firms are those that develop plans for construction projects, as
opposed to construction companies. Without differentiating between engineering
and architectural firms, the U.S. accounted for 67 (33.5 percent) of competing firms
and $3.2 billion (43.5 percent) of international design work.
With respect to design of consumer products, the situation is quite different,
with European imports traditionally dominating the top end of the line in the U.S.
Consider the trade balance. In 1982 total U.S. exports were $252 billion, or 0.8
percent of GNP. Arts-related exports were $12 billion, or 5 percent of total exports.
In 1982 imports amounted to $306 billion, or almost 10 percent of GNP. Arts
imports were $37 billion, or 12 percent of all imports. Thus the U.S. had a trade
deficit with the rest of the world of $55 billion, or 1.7 percent of GNP. The arts trade
deficit was $25 billion, or 45 percent of the total trade deficit.
Growth in the number of decorators and designers during the 1980s suggests
that an effort is being made to fill this trade gap. Good design adds value to products
and makes them more competitive in the domestic and export market.
It has been estimated that the U.S. lost more than $13.5 billion to copyright
pirates around the world in 1986 (Hoffman 1989). Unfortunately, there is no
estimate of loss due to design piracy.
While European countries and Japan have long provided design protection, the
U.S. offers a design patent that requires not just that a design be different and
distinctive, but that it be new, useful and not "obvious" to others skilled in the trade.
Product designers consider their work as creative and original as that of painters,
sculptors and writers who enjoy copyright protection. Designers, however, too often
must watch their work being copied by others with little fear of being sued under
existing law (Andrews 1990). To the degree product design protection increases the
value of design, then to that degree the employment and earnings of designers will
increase and their status in corporate hierarchies would rise.
Architecture and Design Arts Occupations, 1970 to 1990 I 141
Frank Lloyd Wright, like members of the German Expressionist Movement, the
Bauhaus and the "International Style," believed that architecture and design could
change the human condition.
"It was one of those illusions of the 20s," recalls Philip Johnson. "We were
thoroughly of the opinion that if you had good architecture the lives of people would
be improved... and people (would) improve architecture... This did not prove to be the
case." (Hughes 1981:164)
For more than a half century, the International Style of rectangular glass boxes
dominated construction in downtown America. Aesthetic Utopians wanted buildings
and objects to approach an aesthetic ideal of perfection while developers and
manufacturers wanted to produce at the lowest possible price, thus fueling the
dominance of the style. But in the 1980s architects and designers began to reject this
mainstream of modern architecture and design. Not just the formal harmonies and
proportions of Gropius, Mies van der Rohe and Le Courbusier were rejected, but also
their social and ethical ideals. Functionalism, however, was not rejected.
This stylistic rejection became known as "Postmodernism." It is characterized by
an eclecticism of styles and reversion to pre-modern architecture before the
Without a dominant style, the public is confused while the architecture and
design professions search for a new guiding light. A new style fuels the employment
and earnings of architects and designers. Perhaps Wrights dream of a distinctive
American style will result from this contemporary Postmodern confusion.
Employment opportunities for architects, landscape architects and designers are
projected to rise faster than the average for the labor force as a whole through the year
2005. Most job openings, however, will result from people transferring to other fields
or leaving the profession.
Demand for architects and landscape architects depends on local construction
which, in turn, is sensitive to the economic cycle. Furthermore, architects must meet
licensing requirements in each state before they can practice there, which limits
mobility. As well, competition for the most prestigious firms will continue.
Computer-aided design and drafting is becoming more prevalent, but is not
expected to reduce demand tor architects. Rather, it should allow more options to be
developed and changes in plans made more easily, hopefully improving the quality of
1 andscape architects will no doubt find construction growing in the long term,
but mainly outside the major cities. Ivpicallv such sites have large surroundings
requiring more landscape designing than urban sites. And as the COS1 ot land
increases, good landscape design will become more desirable.
Increased development ot recreation spaces, wildlife refuges and parks will also
require landscape architects, as will growing concern about the environment and
142 I Artists in the Work Force
historical preservation. Also, local, city and regional planning is requiring increased
mixed land reclamation and refurbishing of existing sites.
Designers should also continue to be in demand. Continued competition and
emphasis on product quality and safety, on design of new business and office
products, on high-tech products in medicine and transportation will also stimulate
demand for industrial designers.
To the professional statisticians of the Federal government, the staff of
representative organizations and the Research Division of the NEA, whose long term
efforts provided the evidence presented in this report, many thanks and
encouragement are offered. In a society in which "If you're not counted, you don't
count!" their on-going efforts aid and assist materially in making the case for the arts
before the court of public opinion.
About the Author
Harry Hillman Chartrand received his M.A. in Economics (1974) from Carleton
University in Ottawa prior to operating his own consultancy, FUTURES. Clients
included the Canadian Federation of Mayors and Municipalities, Tri-Level Task
Force on Public Finance, Bureau of Intellectual Properties and Secretary of State,
Canada Council. From 1981 to 1989, he served as Research Director for the Canada
Council. From 1989 to 1994, Mr. Chartrand served as Chief Economist of Kultural
Econometrics International. In July he returned to private practice in Saskatoon and
now serves as Administrator for the Saskatchewan Arts Alliance. He has written
extensively concerning cultural economics.
Australia Council for the Arts (1991). The Arts: Some Australian Data, 4th Edition, Sydney.
Bielby, William T. and Denise D. Bielby (1987). Pay Equity and Employment Opportunities
Among Writers for Television and Feature Films. Los Angeles: Writers Guild of
America, west (WGA, w), June.
1989 Hollywood Writers' Report: Unequal Access, Unequal Pay. WGA,w, April.
(1993). The 1993 Hollywood Writers' Report: A Survey of the Employment of Writers
in the Film, Broadcast, and Cable Industries for the Period 1987-1991. WGA,w, June.
Blau, J.R., Architects and Firms: A Sociological Perspective on Architectural Practice.
Cambridge: MIT Press, 1984.
Bradshaw, Tom, "Predicting Artist Employment in the Year 2000," in Cultural Economics
88: An American Perspective, Shaw, D.V., W.S. Hendon, V.L. Owens (eds), Association
for Cultural Economics (ACE), University of Akron, 1989.
(1984) "An Examination of the Comparability of the 1970 and 1980 Census Statistics
on Artists." In William S. Hendon, et al (eds.). The Economics of Cultural Decisions.
Brighton, A. and N. Pearson. The Economic Situation of the Visual Artist. London: Calouste
Gulbenkian Foundation, 1982.
Brown, P.L., "A Legal Leg for Designers," New York Times, Aug. 2, 1990, Cl & 6.
Architecture and Design Arts Occupations, 1970 to 1990 I 143
Bureau of the Census, Classified Index of Industries and Occupations: 1980 Census of
Population, PHC 80 R4, U.S. Dept. of Commerce, Washington, D.C. Nov. 1982.
Bureau of Labor Statistics. Handbook of Labor Statistics. Washington, D.C, August 1989.
Occupational Handbook 1992-93, Bulletin 2400, Washington, D.C, 1993.
Canada Council (1984). A Canadian Dictionary & Selected Statistical Profile of Arts
Employment, 1981. Ottawa, Research and Evaluation.
Cantor, Muriel C, Employment Status of Performing Artists, 1970 to 1980. Washington,
D.C: NEA, Research Division, 1987.
Center for Education Statistics, Degrees and Other Formal Awards Conferred Survey, U.S.
Dept. of Education, Washington, D.C, annual.
Citro, Constance F., and Deirdre A. Gaquin (1987). Artists in the Workforce, 1950 to 1985.
Washington: NEA Research Division.
Clark, K., Civilization, BBC & John Murray, London, 1969.
Data User Services Division, U.S. Bureau of the Census. Census of Population and Housing,
1990: Public Use-Microdata Samples U.S. Technical Documentation. Washington,
D.C: U.S. Bureau of the Census, 1983 and 1992.
Diamonstein, B., Fashion: The Inside Story, New York: Rizzoli, 1985.
Economic Report of the President. Washington, D.C: U.S. Govt. Printing Office, 1991.
Ellis, Diane C and John C Beresford. Trends in Artist Occupations: 1970-1990.
Washington, D.C: NEA Research Division, 1994.
Filer, Randall K., "Arts and Academe: The Effects of Education on Earnings of Artists,"
Journal of Cultural Economics, Vol. 14, No. 2, Dec. 1990, 15-40.
(1989). "The Economic Condition of Artists in America." In Douglas V. Shaw, et al,
(eds.), Cultural Economics 88: An American Perspective. Akron, ACE.
(1986). "The 'Starving Artist' — Myth or Reality? Earnings of Artists in the United
States." Journal of Political Economy, 94, 56-75.
Gaquin, Deirdre. Special Tabulation of 1990 Census. Washington, D.C. 1994.
Hale, R.D., "An Economic Analysis of the Architectural Profession," Journal of Cultural
Economics, Vol. 4, No. 2, Dec. 1980, 27-38.
Hanson, A.F., "History of Art — Part Four: The Modern World — 5." Twentieth Century
Architecture, 3rd. Edition. New York: Abrams 1986, 746-767.
Harris, N. The Artist in American Society. New York: Simon & Schuster, 1970.
Heikkinen, Merja (1989a). Tilannekuva Kirjailijoista (The Social and Economic Position of
Finnish Writers). Helsinki: The Arts Council of Finland.
(1989b). "A Situational Portrait of Writers: A Study of the Position of Authors in
Finland (English Summary)," Helsinki: The Arts Council of Finland.
Horowitz, H., "The Status of the Artist in the U.S.A.," Journal of Cultural Economics, Vol.
17, No. l,June 1993,29-48.
Hughes, R., Shock of the New, Chapter 4: "Trouble in Utopia." New York: Alfred A. Knopf,
Jeffri, Joan. "The Artist in an Integrated Society." In Public Money and the Muse. Stephen
Benedict, ed. New York: Norton, 1991.
, J. Hosie and R. Greenblatt. "The Artist Alone: Work-Related and Social Service
Needs-Selected Findings." Journal of Arts Management and Law 17 (Fall 1987), 5-22.
Jeffri, Joan, ed. (1989). Information on Artists: Cape Cod. New York: Columbia
University/Research Center for Arts and Culture.
Jeffri, J. and D. Throsby. "Professionalism and the Visual Artist." European Journal of
Cultural Policy, 1.1, 1994.
Keens Company (1994). Literature Field Overview Study. Prepared for the NEA, February.
144 I Artists in the Work Force
Kingston, Paul, and Jonathan R. Cole (1986). The Wages of Writing: Per Word, Per Piece,
or Perhaps. New York: Columbia University Press.
Knox, P.L. (ed.) The Design Professions and the Built Environment. New York: Nichols
Lorenz, C, The Design Dimension: Product Strategy and the Challenge of Global
Marketing. New York and Oxford, Basil Blackwell, 1986.
Meggs, P., A History of Graphic Design. New York: Van Nostrand Reinhold, 1983.
Moore, K., E. Jacob, T. Bradshaw, et al, "Design Arts" in The Arts in America, NEA,
Washington D.C., 1988, 205-244.
Moulin, Raymonde. L'Artiste, l'institution, et le marche. Paris: Flammarion, 1992.
Research Division, A Source Book of Arts Statistics, Bi-annual, NEA, Washington, D.C.
Rowe, P.G., Design Thinking. Cambridge: MIT Press, 1984.
Ruttenberg, Friedman, Kilgallon & Associates, Inc. Working and Not Working in the
Performing Arts: A Survey of Employment, Underemployment and Unemployment
Among Performing Artists in 1980. Washington, D.C: Labor Institute for Human
Enrichment, Inc., AFL-CIO, 1981.
Ruttenberg, Friedman, Kilgallon, Gutchess & Associates, Inc., Survey of Employment,
Underemployment, and Unemployment in the Performing Arts. Washington, D.C:
Human Resources Development Institute, AFL-CIO, 1978.
Santos, F.P. (1976). "Risk, Uncertainty and the Performing Artist." in Mark Blaug (ed.) The
Economics of the Arts. Boulder: Westview Press.
Schuyler, D., The New Urban Landscape, Baltimore: John Hopkins Press, 1987.
Statistics Canada (1993). The Nation: Employment and Income by Occupation. Ottawa,
March and April.
Stinson, John F. Jr., "Multiple jobholding up sharply in the 1980s." Monthly Labor Review.
Washington, D.C: U.S. Dept. of Labor, Bureau of Labor Statistics. July 1990:3-10.
Throsby, D. and D. Mills. When Are You Going to Get a Real Job? Sydney: Australia
Towse, Ruth. "The Economic and Social Characteristics of Artists in Wales." Dept. of
Economics, University of Exeter, undated mimeo.
U.S. Bureau of the Census. Statistical Abstract of the United States, 1992 (112th Edition).
Washington, D.C: U.S. Bureau of the Census, 1992.
Waits, C. Richard and Edward M. McNertney (1989). "The Incomes of Cultural Providers:
A Review of Current Research." In Douglas V. Shaw, et al (eds.), Cultural Economics
88: An American Perspective. Akron: ACE.
Wassail, Gregory H. and Neil O. Alper (1984). "Determinants of Artists' Earnings." In
William S. Hendon, et al, The Economics of Cultural Industries. Akron: ACE.
(1985). "Occupational Characteristics of Artists: A Statistical Analysis." Journal of
Cultural Economics 9, 13-34.
1990). "When Is an Artist an Artist?" Journal of Arts Management and Law. 19, 37-50.
(1992a). "Earnings of American Artists: 1960-1980." 7th International Conference on
Cultural Economics, Fort Worth, October.
(1992b). "Toward a Unified Theory of the Determinants of the Earnings of Artists." in
R. Towse and A. Khakee (eds.) Cultural Economics. Berlin: Springer- Verlag.
_and Rebecca Davison (1983). Art Work: Artists in the New England Labor Market.
Cambridge: New England Foundation for the Arts.
Whitney, P. (ed.) Design in the Information Environment, Carbondale and Edwardsville:
Southern Illinois University Press, 1985.
for the arts
Seven Locks Press
Santa Ana, California