In his 1995 book, The End of Work, Jeremy Rifkin predicted that employment in the third, or nonprofit, sector would grow increasingly important in the U.S. What is not widely appreciated, however, is that employment in the nation’s nonprofit organizations already plays a major role in the nation’s economic life.1 As of 1996, roughly 1.6 million nonprofit organizations existed in the United States. Nearly 400,000 of these organizations are known as “member serving” organizations and include social and fraternal organizations, labor unions, and business and professional associations. The remaining 1.2 million are public-benefit organizations of the sort commonly considered to be part of the charitable segment of the nonprofit sector, including churches, social service providers, and action agencies.

These 1.2 million public-benefit organizations had expenditures of approximately $525 billion as of 1996, the equivalent of more than seven percent of the nation’s gross national product. More important for our purposes here, they employed nearly 11 million people, or approximately eight percent of the nation’s workforce. This was more than three times the number employed in agriculture and larger than the numbers employed in construction, transportation, communication, finance, insurance, and real estate. In addition to paid employment, nonprofit organizations also employed the equivalent of 6.3 million full-time volunteers, boosting their workforce to 17.2 million workers.2

Nor is an important nonprofit sector an exclusively American phenomenon. Indeed, work we have done under the Johns Hopkins Comparative Nonprofit Sector Project reveals that employment in the nonprofit sector, as a share of total employment, is higher in four out of the 22 countries studied than it is in the United States. (See Figure, page 8.) In contrast to the seven percent nonprofit employment in the United States, the comparable figure is 12.4 percent in the Netherlands, 11.5 percent in Ireland, 10.5 percent in Belgium, and 9.2 percent in Israel.3

Who are these nonprofit employees? Where do they work? What do they do? Although we have learned a great deal about these matters in the past decade or more, significant blind spots remain. The basic sources of data about nonprofit employment continue to be imperfect at best, yielding estimates that vary wildly (see Nonprofit Employment Box). What is more, we have great difficulty tracking the changing balance between nonprofit and for-profit employment in particular fields. Concern over this issue has grown in recent years because of the increasing competition from for-profit firms in fields that nonprofits once dominated, but lack of reliable data obscures the debate.

In this brief article we review some of the limitations in the available data sources on nonprofit employment, outline what we are beginning to learn about the contours of the nonprofit workforce from the available data, and then describe an effort we have under way to close some of the remaining gaps in knowledge about this crucial facet of nonprofit operations. We hope a better understanding of the vital importance of nonprofit employment as well as the need for having more accurate data on this sector will come out of this discussion.

Based on existing data sources, a number of salient features of the nonprofit workforce are becoming apparent, though the exact parameters remain somewhat blurred because of the limited quality of the data. The most important of these features are as follows:

• Quite large. Depending on the source, estimates of the number of people working for nonprofit organizations at the present time range from 6.9 million to nearly 11 million, that is, from just over five percent of the American labor force to eight percent.4
• Concentrated in three industries. Eighty percent of all nonprofit employees work in three industries—health, education, and social services.5
• Growing disproportionately. Between 1977 and 1996, employment in the nonprofit sector grew by an annual rate of 3.3 percent. During this same period, overall U.S. nonagricultural employment grew by an annual rate of only 1.9 percent.6
• Disproportionately female. Two-thirds of all nonprofit employees were female as of 1994 compared to 44 percent of all nonagricultural employees in other sectors of the economy.7
• Disproportionately minority. Fifteen percent of employees in the independent sector in 1994 were African-American, compared with ten percent of all nonagricultural employees.8
• Highly educated. About 40 percent of all employees in the nonprofit sector have college or advanced degrees compared to 17 percent in the for-profit sector. Among full-time employees, 45 percent of nonprofit employees had college or advanced degrees compared to 41 percent for the public sector and 19 percent for the for-profit sector.9
• Highly skilled. Professional specialty occupations accounted for 38 percent of all employees in the nonprofit sector as of 1990, compared to 32 percent in government and eight percent in the for-profit sector.10
• Higher wages in selected occupations. The mean salaries of several professional occupations, including physicians and psychologists, are from 10 to 25 percent lower in the nonprofit sector than in the for-profit sector. However, the mean salaries for other professions, such as registered nurses, health technologists, health aids, practical nurses, early childhood teachers’ assistants, medical managers, and kindergarten teachers are from 10 to 25 percent higher in the nonprofit sector than in the for-profit sector.11 The 1990 census also suggests that mean wages and salaries for females, African-American, Hispanics, and other minority groups were higher in the nonprofit sector than in the for-profit sector though lower than in government.12

While we are gaining a clearer picture of the nonprofit workforce, important gaps and uncertainties still remain both with respect to the overall scale of nonprofit employment and with respect to its composition and characteristics. To close these gaps, two sets of steps will be needed.

Census and Population Survey Data: In the first place, there needs to be an effort to perfect the data available from the decennial Census and Current Population Survey. Nonprofits have an important role to play in making sure their employees are aware that they work for a nonprofit organization so that census tabulators will get a more accurate count of employment in this sector.

ES-202 Data: The best hope for perfecting the aggregate picture of nonprofit employment and its distribution by industry at the present time is to tap into the data collected as part of the ES-202 program (for more on this program see Nonprofit Employment Box, next page). The ES-202 program is a partnership between the U.S. Bureau of Labor Statistics and State Employment Security Agencies. It covers 97 to 99 percent of all nonagricultural employment, including 98 percent of all nonprofit employment. Data are collected on a quarterly basis from firms in each state, and can be broken down by detailed industry code and at the county level for every county in the country. It is thus much more timely and complete than the Census of Service Industries, which has long been the major source of reliable aggregate employment data on the nonprofit sector.

Without some way to differentiate the nonprofit employers from the for-profit ones, the ES-202 data can provide only aggregate figures. Work under way at the Johns Hopkins Center for Civil Society Studies is seeking to overcome this obstacle. In collaboration with State Employment Security Agencies in a number of states, the Johns Hopkins Nonprofit Employment Data Project is developing a series of flags that states can use to identify the nonprofit employers in their ES-202 data sets.
Lester Salamon and Leslie Hems have already produced a pilot report drawing on this body of data for the state of Maryland.13 They found that:

• The Maryland nonprofit sector, exclusive of churches, employs considerably more people than had previously been understood—about eight percent of all full-time employees, or nearly one in every twelve employees. This is more people than are employed in all of manufacturing in the state.14
• The nonprofit sector employs a striking 19 percent of all workers in Baltimore City, or nearly one in every five workers in the city.
• Nonprofit organizations are significantly present in all regions of the state, including the rural counties of Western Maryland and the Eastern Shore, and not just in the urban areas around Baltimore and Washington.
• Maryland’s nonprofit sector has been a major contributor to the state’s employment growth, accounting for half of all the net new jobs that Maryland generated between 1989 and 1996.
• Despite their substantial growth, Maryland nonprofit organizations have been losing market share to for-profit firms in the larger fields of nonprofit action such as health, education, and social services between 1989 and 1996.

The Maryland Association of Nonprofit Organizations (MANO) was able to use this information to communicate the importance of the nonprofit sector to Maryland state and local policy-makers. In 2000, MANO specifically used the report to convince state policy-makers to include nonprofits in legislation providing a transportation and commuting tax credit for state businesses, highlighting the importance of nonprofit businesses to the state. Similar insights will hopefully soon be available in other states as well as nationally.

Individuals interested in finding out more about the Nonprofit Employment Data project can visit the Center for Civil Society Studies website ( Pilot studies in over ten states are under way. State reports with a detailed analysis of state nonprofit data will be published on the Center website as they are produced.

The nonprofit sector is a major economic force, and in many regions is growing in terms of employment, market share, and income. At the same time it is confronting significant changes and growing competition from for-profit firms in its traditional areas of activity. Without accurate and timely data, the ability of the nonprofit sector to cope with these changes will continue to be severely compromised. Policy-makers, the media, and the general public have little information about this sector in general and do not have an accurate picture of the economic importance of this sector. Second, without detailed, timely information, this sector lacks a degree of self-knowledge that is critical for identifying key new workforce trends. Improving the basic data available to nonprofit organizations is essential to improving management. The Johns Hopkins Nonprofit Employment Data Project is designed to aid in this process, but additional work is needed as well. Nonprofits must make sure they have the information they need to operate effectively in today’s new competitive environment.

1. Salamon, Lester. 1999. America’s Nonprofit Sector: A Primer. 2nd Edition. New York, NY: The Foundation Center, p. 6.
2. ibid., p. 22.
3. ibid., p. 39.
4. See U.S. Bureau of Labor Statistics. 1999. Current Population Survey, Table 19. Washington, D.C.: U.S. Bureau of Labor Statistics, and Salamon, 1999.
5. See Hodgkinson, Virginia Ann, Murray S. Weitzman, John Abrahams, Eric Crutchfield, and David R. Stevenson. 1996. Nonprofit Almanac: Dimensions of the Independent Sector, 1996-1997. Washington, D.C.: Independent Sector, p. 132.
6. ibid., p. 129.
7. ibid., p. 136.
8. ibid., p. 137.
9. Hodgkinson, Virginia Ann, Murray S. Weitzman, Stephan M. Noga, and Heather A. Gorski. 1993. National Summary: Not-For-Profit Employment From the 1990 Census of Population and Housing (Preliminary Findings). Washington, D.C.: Independent Sector, p. 3.
10. ibid., p. 7.
11. Hodgkinson et al., 1996, p. 133.
12. Hodgkinson et al., 1993, p. 3.
13. Salamon, Lester and Leslie Hems. 1999. “Maryland’s Nonprofit Sector: A Major Economic Force.” Baltimore, MD: Johns Hopkins University, Center for Civil Society Studies.
14. Salamon, Lester. 1997. Private Action/ Public Good: Maryland’s Nonprofit Sector in a Time of Change. Baltimore, MD: Maryland Association of Nonprofits, p. 9.

Additional References

Independent Sector. 1998. In Brief: America’s Nonprofit Sector: Facts and Figures on the Independent Sector. Washington, D.C.: Independent Sector.
Statistical Abstract of the United States. (
1997 Economic Census: Census of Service Industries. Comparative Statistics for the States. ( E97SUSI.htm)

Five major sources of data are currently available to help clarify the size and evolution of the nonprofit workforce. Unfortunately, however, each of these data sources suffers from limitations that make it difficult to determine exactly the scale and composition of the nonprofit labor force.

Census of Service Industries: The first of these sources is the Census of Service Industries, which is conducted every five years by the U.S. Census Bureau. Beginning in 1982, this source began to designate the tax status of establishments. As a consequence, it now provides highly reliable data on the number, employment, and wages of nonprofit organizations with at least one paid employee broken down by type of organization using the Standard Industrial Classification. In addition, it provides comparable data on for-profit firms. Unfortunately, however, it regularly takes the Census Bureau three to five years to process and release the data, which means that as of mid-2000 the latest data available were from the 1992 service census. This leaves the nonprofit sector largely blind to the fundamental shifts that are under way in nonprofit versus for-profit involvement in certain industries. In addition, the service census covers only service industries, which means that it excludes nonprofit credit unions and thrift stores as well as education, a major component of the nonprofit sector. Finally, while this source provides detailed information on employment by industry, it does not provide information on the demographic characteristics or occupations of the employees themselves.

ES-202 and Bureau of Labor Statistics Surveys: The Bureau of Labor Statistics in the U.S. Department of Labor produces far more timely employment data. BLS oversees the collection of monthly employment data from workplaces throughout the country under its ES-202 program linked to the unemployment insurance system. In addition, it conducts a quarterly sample survey of workplaces throughout the country. Both of these sources cover most nonprofit workplaces and generate timely data on numbers of employees and wages broken down by industry. The ES-202 source also provides these data at the county level. Unfortunately, however, neither of these sources identifies nonprofit workplaces. And, as with the Census of Service Industries, they provide only the most basic information on the demographic and other characteristics of the workers.

Population Census: Far more detailed demographic information on nonprofit workers is available from the decennial population Census. Beginning in 1990, the Population Census queried respondents about the nonprofit status of their place of employment. By linking responses to other demographic information, it is therefore possible to develop a fairly complete profile of the occupations, education, and related demographic characteristics of the nonprofit workforce. The problem, however, is that this source assumes that respondents know the status of their workplace. In practice, it appears that many do not. Thus, the size of the nonprofit workforce as reflected in the Census data is well below that estimated from other sources, as highlighted by Hodgkinson et al. in their National Summary of Not-For-Profit Employment. The Population Census reported a figure of 7.7 million compared to the 8.6 million estimated by the Census of Service Industries, which excludes education, and the 9.8 million estimated by Independent Sector using a combination of Census, Labor Department and Education Department statistics. Especially missing from the Census count are persons in lower skilled jobs and with lower education levels, particularly in the health and education sectors. Beyond that, the Census is now a decade old. (See Hodgkinson et al., 1993, pp. 1-2.)

Current Population Survey: The Current Population Survey, a monthly survey of a representative sample of households in the U.S. population, provides the basis for several Bureau of Economic Analysis publications on employment and earnings in all industries. Since the survey includes a question about whether the respondent works for a nonprofit, it can provide a breakdown of employment by for-profit and nonprofit status and by industry, occupation, and demographic background. The CPS clearly yields more timely information. However, it suffers from the same problem as the decennial census in that respondents often do not know whether their employers are nonprofit organizations. Reflecting this, estimates of nonprofit employment from the Current Population Survey are even lower than those for the decennial census (6.9 million versus 7.7 million). In addition, the CPS is based on a sample, so aggregate numbers are not available at the state and county level, only for the nation and for regions.

Nonprofit Almanac: The Independent Sector regularly issues a Nonprofit Almanac that compiles data from these other sources and also seeks to fill in some of the gaps through a variety of data manipulation efforts. For example, the Almanac seeks to overcome the undercounting of nonprofit employment in the Current Population Survey by applying such estimates as those of the nonprofit share of employment derived from earlier Census of Service Industries data. It does the same to the annual Bureau of Labor Statistics estimates. While this does not reflect any shift in the nonprofit versus for-profit balance in particular industries, it nevertheless generates a more up-to-date, if rough, estimate of nonprofit employment than the Census of Service Industries alone would permit.

Dr. Lester M. Salamon is a professor at Johns Hopkins University and director of the Johns Hopkins Center for Civil Society Studies. He has been studying the nonprofit sector for over 20 years and is the author of America’s Nonprofit Sector: A Primer. Dr. Sarah Dewees is coordinating the Nonprofit Employment Data project at the Center for Civil Society Studies at Johns Hopkins University.