HHS Data: Nonprofit Jobs Picture Mixed, Faster Job Growth than in For-Profit Sector, Small Nonprofits Hit Hard by Recession

Print Share on LinkedIn More

CC Photo: Jamie Thompson

Based on an analysis of four years of federal government data on tens of thousands of nonprofit, for-profit, and government employers, there are some important and startling findings about employment trends in the nonprofit sector that should interest most nonprofit agency leaders and staff:

  • Overall, the nonprofit sector is generating jobs at a faster rate than the private sector. The resilience of the nonprofit sector—and of charitable donors—during a recession is noteworthy, especially as nonprofits perform critical functions in providing a safety net for the poor and disadvantaged during a recession.
  • Nonprofit-sector job creation, however, has been largely concentrated in large nonprofit employers with 1,000 employees or more. We would guess that these nonprofits are likely to be hospitals and universities.
  • There is a reasonably logical correlation of the growth in nonprofit sector employment with the arrival of federal stimulus dollars starting in 2009 and extending through 2010. We know that a significant portion of the stimulus funds were directed to programs run by nonprofit organizations. It is possible, however, that with the end of stimulus spending, the employment gains of the nonprofit sector in recent years could be reversed.
  • Despite the increase in overall nonprofit employment and an increase in the number of establishments associated with large nonprofit employers, there appears to have been a contraction in the number of small nonprofit employers. Perhaps small nonprofits have adjusted to the recession by shifting from paid personnel to volunteer staffing. If the recession ends up double-dipping, more small nonprofits are likely to disappear or be taken over by larger organizations.

Is this data from the Census Bureau? The Bureau of Labor Statistics? The Internal Revenue Service? No, these potential findings are drawn from annual studies conducted by the Agency for Healthcare Research and Quality (AHRQ), a little-known agency associated with the Department of Health and Human Services (HHS).

No one collects undeniably accurate and reliable data on nonprofit employment, no one. Counts of employment by the Bureau of Labor Statistics, for example, are really estimates based on samples, often stratified samples of employers or employees to make sure that information is collected for specific types and classes of employers and for specific geographic regions.

So it is a boon to researchers of the nonprofit sector that a unit of the Department of Health and Human Services also conducts an annual survey of employers—like others, based on a stratified sample—in order to collect health-insurance-related information. Every year since 1996 (with the exception of 2007), AHRQ has surveyed employers to identify and assess the “number and types of private health insurance plans offered, benefits associated with these plans, premiums, contributions by employers and employees, eligibility requirements, and employer characteristics.”

This data series, known as the Medicare Expenditure Panel Study (MEPS), includes an “Insurance/Employer Component” that gathers health expenditure and benefit information from a national sample of tens of thousands of public- and private-sector employers—including nonprofit employers. For researchers interested in the nonprofit sector, there is a side benefit: the data include estimated counts of nonprofit employers and employees by size of employer and by full-time or part-time status of the employees. When examined over a period of years, these data reveal lots of hidden nuggets of insight into changes in nonprofit employment and nonprofit employers.

AHRQ researchers interview a large number of employers via a detailed telephone and written survey, typically sampling some 38,000 to 40,000 private sector establishments drawn from the Census Bureau’s most recent business register. The MEPS response rate for the 2006, 2008, 2009, and 2010 interview periods has typically hovered between 78 and 86 percent. That high response rate is due to several rounds of outreach, including an initial survey mailing, a second mailing if there is no response to the first, and a telephone follow-up if neither mailing secures a response.

Multiple years of data from these carefully constructed national samples reveal often-surprising trends concerning rates of nonprofit job creation, changes in full-time versus part-time employment, and the number and types of nonprofit firms and establishments. 

Job Creation

Like many other data sources, the AHRQ surveys suggest that the nonprofit sector is outpacing the private sector in job creation:

Table I: Nonprofit Employment by Firm Size


Total nonprofit employment

Less than 10 employees

10-24 employees

25-99 employees

100-999 employees

1000 or more employees





























The MEPS statistics suggest that nonprofit employment plummeted in 2009 as the national recession ravaged the sector but then increased dramatically in 2010. Why? The obvious answer is the American Recovery and Reinvestment Act of 2009—a.k.a. ARRA or “the stimulus”—which funneled money to nonprofits such as Community Action agencies, Head Start groups, health clinics, and hospitals, allowing them to retain and hire staff. Who knows how many people’s livelihoods during the darkest days of 2009 and 2010 were maintained by ARRA dollars (funding that has more or less come to an end now)?

In 2010, ARRA-funded weatherization employment increased from 10,666 full-time jobs in the first quarter to 15,415 in the fourth as weatherization programs ramped up their activity. ARRA-funded employment in the Health Center Integrated Services Development Initiative increased from 7,068 to 8,267 over the same period, ARRA-funded Community Services Block Grant jobs grew from 7,966 in Q1 to 18,431 in Q3 (the Q4 statistic for this particular program is missing), jobs in the homelessness prevention and rehousing program grew from 3,481 to 4,252, and so on. In addition to these ARRA jobs, accelerated medical reimbursements from Medicaid and Medicare created even more nonprofit employment.

Overall, total nonprofit-sector employment grew by 8.8 percent from 2009 to 2010 according to the MEPS survey. Breaking down the data in terms of nonprofit size reveals a lot of variation, however. Employment among nonprofit employers with 10 to 24 employees was up 3.6 percent. For organizations with 25 to 99 employees it was up 6.4 percent. Between 100 and 999 employees: up 1.9 percent. More than 1,000 employees: up an astounding 16.7 percent. But at nonprofits with fewer than 10 employees employment actually dropped 3.8 percent.

Part-time vs. Full-time

Another revealing way to break down the data is to look at full-time versus part-time jobs. Part-time employment in the nonprofit sector increased 5.3 percent between 2009 and 2010, as compared with a 2.2-percent increase for all private sector employers. Part-time employment in nonprofit organizations with 1,000 or more employees increased a whopping 31.1 percent.

Table II: Part-time Nonprofit Employment by Firm Size


Total (part-time nonprofit)

Less than 10 employees

10-24 employees

25-99 employees

100-999 employees

1000 or more employees



447,854 (11.7%)

322,816 (8.4%)

648,327 (17.0%)

967,568 (25.3%)

1,434,014 (37.5%)



486,015 (12.5%)

327,332 (8.4%)

559,350 (14.4%)

1,185,903 (30.5%)

1,327,970 (34.2%)



467,533  (12.7%)

298,540  (8.1%)

512,362 (13.9%)

1,316,193 (29.8%)

1,099,844 (35.6%)



474,726 (12.2%)

354,298  (9.1%)

631,212 (16.2%)

986,818 (25.4%)

1,442,947 (37.1%)

Nearly one out of four nonprofit employees were part-time workers in 2010, compared to slightly more than one out of five private-sector employees overall. (This kind of data is important to the AHRQ because part-time workers are less likely to qualify for employer-provided or employer-subsidized health insurance.)

Nonprofit Firms and Establishments

Although the data is presented by “firm” size, the survey is actually administered among “establishments.” An establishment is defined as “a specific workplace or business location,” while a “firm” is a “business entity consisting of one or more business establishments under common ownership or control.” Consequently, one firm could have several different establishments. Given that the survey sample is specifically constructed to make national estimates, the changes in the number of establishments in the survey do reveal something about the structure of both nonprofit and for-profit business entities in this recession.

Table III: Nonprofit Establishments by Firm Size


Total number of nonprofit establishments

Less than 10 employees

10-24 employees

25-99 employees

100-999 employees

1000 or more employees



251,174 (51.4%)


61,720 (12.6%)

85,498 (17.5%)

26,843 (5.5%)



277,902 (52.0%)

70,378 (13.2%)

66,779 (12.5%)

87,262 (16.3%)

32,233 (6.0%)



257,460 (50.4%)

64,592 (12.6%)

62,590 (12.3%)

90,200 (17.7%)

36,007 (7.0%)



255,727 (49.4%)

64,102 (12.4%)

69,577 (13.5%)

89,398 (17.3%)

38,441 (7.4%)

Again, it appears that the stimulus was one factor leading to an increase in the number of nonprofit establishments between 2009 and 2010, but the number of nonprofit establishments is still down 3.2 percent as compared with 2008 (a much sharper decline than the 1-percent drop among all private-sector establishments). Particularly noteworthy is the shrinkage in the number of establishments controlled by nonprofit firms of fewer than 10 employees, as compared to the increasingly large share of establishments tied to nonprofit employers with 1,000 employees or more.

The Nonprofit Economic Engine

Everyone can argue methodology and whether the AHRQ’s MEPS studies are a better or worse indicator of changes in the nonprofit sector than other data series. For one thing, the MEPS studies focus on employers, not entities. As a result, among nonprofits—especially nonprofit corporations that are combined with other tax-exempt entities such as religious institutions—the MEPS analysis is missing the larger number of nonprofits that exist without any staff.  As a result, a contraction in the number of nonprofit establishments or firms in the MEPS studies does not necessarily mean that there was a corresponding reduction in the overall number of nonprofits for those years. 

But the AHRQ’s MEPS studies should help nonprofits and policy-makers recognize the economic importance of the nonprofit sector writ large. It’s more than what the sector absorbs and spends in the form of charitable giving and governmental grants and contracts—which is a significant though not huge portion of the nation’s Gross Domestic Product. It’s also that the nonprofit sector provides employment to millions of people. When debates occur within the Washington beltway over whether employers can afford to provide health insurance or the best methods of generating jobs in the American economy, policy-makers might benefit from looking at the AHRQ data to better appreciate the nonprofit sector as a significant and resilient provider of jobs and incomes.

  • Nesly

    Here is a great article about non profits facing the recession

  • Michael Wyland

    It would be interesting to compare the MEPS data on nonprofit establishments with the changing numbers of nonprofits listed on the IRS’s Publication 78 after the recent initial purge.

  • jdzlindeed

    Can I use this data in my grant proposals and reports if I cite it properly? Our CDC could use it.

  • rick cohen

    Why not? It’s a public data source. You should check out the actual AHRQ/MEPS data on the website (the article has the link) because you might find other nuggets of particular interest to you and your colleagues. I always recommend people looking at the original data tabulations. Clearly I’m focusing in the article on what struck me and I’m making observations based on my experience and my research, but you might see more and reach new and different conclusions. Do tell the NPQ Newswire readers what you do with the numbers. Your NPQ Newswire reading peers might be very interested.

  • Kristin Coombes

    I appreciate your reminder to our policy makers that not only does our sector attempt to care for those most effected by economic turmoil, we are also trying to maintain employment and benefits for a large segment of the population. I would be interested to also know the employment numbers by age of nonprofit. It seems many new nonprofits are growing while older, traditional, programs have been hit even harder.

  • Jim Chitwood

    🙂 Great work Rick

  • rick cohen

    Great question, Kristin. But you’re actually pointing out how lousy the data really is on the nonprofit sector. The MEPS data isn’t really aimed at the nonprofit sector, so we get data based on nonprofit employers by size, but that’s it. It would be great to see employment data collected more regularly with disaggregations including age of organization and other factors. The sector’s ability to get the government to start generating this kind of data will depend on the the sector’s ability to explain its use, to show how the kind of information you’re requesting would be used to improve nonprofit sector management and policy. Thanks for reading.