The Region

The Hidden Economy of Nonprofits

We know surprisingly little about nonprofits. Does it matter? Do they?

Ronald A. Wirtz - Editor, fedgazette

Published September 1, 2006  |  September 2006 issue

If you would, draw a picture of the face of nonprofits.

Perhaps you envision a food shelf providing meals for the hungry. Or maybe the United Way or Boy Scouts. Possibly the Red Cross or Salvation Army. Many might picture a religious organization or countless other entities serving the less fortunate or children or both. Whatever your choice, it is more than likely a face drawn from the compassionate history of charities.

Now, try to describe the collective body of nonprofits. Is it big or small? Growing or shrinking? Which parts of the body are healthy, and which ones sick? And here are a few extra credit questions: How has that body changed over time? What helps its different parts grow, and what causes them to die? How does it interact with the other two members of the economic family: for-profits and the government? What does the nonprofit sector accomplish in the course of a year, and is it running faster, or more efficiently, than the year before?

What's that? You've got a cake burning in the oven? Call it the what's-his-name syndrome: We recognize the face, but we don't know much else about the economic traits and activities of the nonprofit sector.

Whoa, you might be saying—we're talking about nonprofits here. Next you'll be squeezing Mother Teresa into a supply-and-demand curve. What's the point?

One key reason to develop a clearer understanding of nonprofits is that they "are growing disproportionately" to the economy as a whole and are slowly assuming a larger share of the overall economy, said Kirsten Grønbjerg, Efroymson Chair in Philanthropy at the Center on Philanthropy at Indiana University. The sector has changed considerably, having seen explosive growth in terms of total organizations, organizational mix, employment, revenues and assets. You might say the nonprofit meek are inheriting the earth. Said Grønbjerg, "Clearly, it would seem to be useful to understand what's driving this."

Ten Years of Nonprofit Growth


Number of Nonprofits






Public charities




Private foundations




Noncharitable organizations








Source: National Center for Charitable Statistics at the Urban Institute

Ten-Year Change in Types of Nonprofits

Chart: Ten-Year Change in Types of Nonprofits

But in spite of this growth, we know comparatively little about the nonprofit sector. What we do know—or think we know—with relative certainty merely scratches the surface, and yet still comes with a lot of measurement caveats. Very few historical and trend data are available on the many smaller "activity fields" that make up the nonprofit sector. The matter gets worse as we attempt to understand more sophisticated economic benchmarks, like output or productivity.

So you might say the nonprofit sector is something of a hidden economy. We know it's widespread and growing, but we don't have good measures for its economic activity or overall contributions, and those measures that we do have likely undervalue it. We obsess over for-profit sectors and their performance, but we're largely ignorant about nonprofits.

"Nonprofits are an afterthought in the economy, so all stats are relative to business," said David La Piana, corresponding via e-mail. He is president of La Piana Associates, a consulting firm to nonprofits based in Emeryville, Calif. "Nonetheless, nonprofits are a significant economic component and employer. We are missing something important by not [understanding] their output."

In some ways, the economics and activities of nonprofits are almost akin to how Secretary of Defense Donald Rumsfeld described the difficulties in Iraq: There are known-knowns, known-unknowns and unknown-unknowns. For example, the Internal Revenue Service has collected data on nonprofit organizational growth over time. OK, good start. But these data are fraught with methodological gaps that lead to both overcounting and undercounting the nonprofit universe (more on this in a bit). Without reliable data at the most fundamental level—organizations, employment—more difficult economic measures like output and productivity are beyond our grasp, and economists are hard-pressed to offer more than a shrug regarding the sector's performance and influence in the broader economy.

The known-knowns

First, a little housekeeping. People generally lump all nonprofits together as a group—as does a lot of research on this topic—when in reality the term "nonprofit" covers many types of organizations, with widely varying missions, sizes, revenue streams and other organizational and sectoral characteristics.

At their root, nonprofits are political creations, granted a federal exemption from paying various taxes on the theory that they offer some type of quasi-public service, and because people connected with the organization don't receive private gain from it. In all, several dozen nonprofit categories are notched into the federal tax code.

Here's what we know about the nonprofit sector: It's large, and it's growing rapidly. In terms of employment and/or gross receipts, in fact, it's believed to be larger than the finance and insurance industry, construction, wholesale trade or durable goods manufacturing. As a group, the number of U.S. nonprofits registered with the IRS has jumped by 30 percent since 1996, tallying more than 1.4 million, according to the National Center for Charitable Statistics, which uses IRS tax returns to compile a comprehensive database on nonprofits. By comparison, over a similar 10-year period (1993-2003), the number of private establishments (not including sole proprietorships) increased by about 14 percent, according to U.S. Census figures.

When it comes to trend-spotting in the sector, three types or groupings—public charities, private foundations and noncharitable organizations—encompass all nonprofits and also offer useful insights into sector trends. Public charities and private foundations both fall under the tax-code umbrella of 501(c)(3)s: They are exempt from federal and other taxes, and donations to such organizations are deductible from income taxes. Noncharitable organizations—essentially, everything besides 501(c)(3)s—are also tax exempt, but contributions to them are not deductible. Overall growth among nonprofits masks a markedly different pace among these three main groups. The number of public charities, for example, has risen by almost 60 percent in the past decade and now totals 850,000; foundations grew by 70 percent, topping 100,000 this year. In contrast, noncharitable organizations saw their numbers decline by almost 7 percent across the country—to 460,000—driven in part by the decrease in fraternal and other social associations. As a result, the nonprofit sector is becoming increasingly concentrated in 501(c)(3)s; their share of the sector has typically grown about 10 to 15 percentage points in many states over the past decade, often making up 55 to 70 percent of all nonprofits in a given state.

And though most nonprofits are small, as a group they are getting bigger and richer. In the past decade, gross annual receipts have doubled to $2.6 trillion, with total assets reaching $3.3 trillion—and these are conservative figures because of who is counted (more on this below). The explosion in revenue and assets tends to be concentrated in larger nonprofit organizations, particularly health care and education.

Charitable donations are also growing, and many smaller humanitarian organizations still depend heavily on such giving. But the sector is increasingly less dependent on donations for growth: The $260 billion in charitable donations received last year made up only about 10 percent of all nonprofit gross receipts. Fees for service now provide easily the largest chunk of nonprofit revenue, much of it in health care and education. In Wisconsin, for example, these two nonprofit subsectors make up about one-fifth of nonprofits but lay claim to about three-fourths of gross receipts.

Though national data on nonprofit employment aren't particularly solid, the rise in total nonprofit organizations, along with solid data from Minnesota (one of few states to track the sector), suggests that nonprofit employment is going gangbusters. From 1993 to 2004, nonprofit employment in Minnesota grew by 48 percent to more than 250,000 workers—an annual average growth rate of 4.4 percent. That's better than twice the private sector's employment growth rate for Minnesota and the nation—both of which came in at about 20 percent-over the same period.

And now for the footnotes

That might feel like strong statistical footing, but don't let go of the proverbial railing. Even these fairly straightforward measures of nonprofit economic activity are rife with caveats.

For example, the biggest pot of data on nonprofits comes from the IRS, which requires organizations to register with it in order to receive tax-exempt status; it also requires organizations with revenue over $25,000 to file an annual 990 tax return. Though it offers easily the best historical record of nonprofit growth, "there [are] problems with the IRS tracking of nonprofits," according to Kevin Rafter, a research associate at the Institute for Nonprofit Organization Management (INOM) at the University of San Francisco. "It has precision issues."

And how. IRS data do the seemingly impossible feat of simultaneously undercounting, overcounting and double counting. The data undercount because numerous types of organizations are exempt from filing 990s, most notably religious organizations, estimated to number 330,000. The data also overcount the total number of nonprofits because defunct organizations are not regularly purged from records. Rafter and others believe the dead weight could be 20 to 30 percent—possibly higher—of all organizations. IRS data also double count organizations at times, because some organizations register two entities, a 501(c)(3) so donations can be tax deductible and a 501(c)(4) to gain the unrestricted ability to lobby politically.

GuideStar is a leading online source of nonprofit information, much of it gleaned from 990s. Last year, GuideStar CEO Robert Ottenhoff wrote in the Nonprofit Quarterly that filing exemptions and errors in the returns that are filed create "a gaping hole in providing nonprofit data."

David Renz, director of the Midwest Center for Nonprofit Leadership at the University of Missouri-Kansas City, called the 990 data "terribly inaccurate," compounded by the fact that the IRS has "no motivation to get it right." As nontaxed entities, nonprofits do not represent a revenue stream to go after in terms of possible abuse. Any IRS efforts would be an exercise in data accuracy only.

That might not matter if there were other data agencies to fall back on. But surprisingly, federal data offices like the Bureau of Labor Statistics and the Bureau of Economic Analysis offer virtually nothing on the sector—or at least nothing specific. Technically, nonprofit organizations and workers are counted in major government surveys, like the BLS's Occupational Employment Statistics survey. But the OES does not distinguish between nonprofit and for-profit workers. Given the small-minority status of nonprofits in terms of the total private market, data and trends on nonprofits simply get lost in the for-profit wash. With the assistance of some academic researchers, the BLS is developing better nonprofit measures for its Quarterly Census of Employment and Wages, but BLS officials say they won't be available for some time.

Don't you love me?

Why doesn't the government track the nonprofit sector? There are many reasons, some of them simple, others less so. Several sources noted that little attention is paid to the nonprofit sector data because most see nonprofits as economically insignificant. Rather than having a clear and purposeful role in the economy, "nonprofits are seen to fill a role left over from government and the for-profits," said Rafter, from INOM. Some likely view the sector as economically "trivial," according to Grønbjerg, from Indiana University. "There is probably a tendency to assume that whatever is important is in the for-profit sector."

That perception of nonprofits likely filters down to the priorities of data-gathering agencies. "It's a matter of administrative necessity and convenience. We're not willing to pay the costs" of getting good nonprofit data, Renz said. His organization produces a unique regional data set on the nonprofit industry in the greater Kansas City region. "I've tried to get government and foundation funding, and they don't care."

But even if time, interest and resources were all available, measuring the nonprofit sector in economic terms offers some unique challenges. In a nutshell, "some of the results are hard if not impossible to quantify or to verify," said Patrick Rooney, director of research at the Center on Philanthropy at Indiana University, via e-mail. Nonprofit motivations and outcomes are more nebulous and "sometimes completely intangible. It is a far cry from outputs and prices from the for-profit sector. But it is still important."

Employment offers a small window on the problem. Even if accurate data were gathered on workers drawing compensation from nonprofit organizations, such data would miss the contribution of volunteer labor to the nonprofit sector. Currently, only crude estimates of volunteer labor exist, and a very modest understanding of its value to organizations.

When it comes to even more sophisticated economic measures—like output and productivity—the matter quickly gets complicated. Technically speaking, total output by nonprofits is included in gross domestic product, according to Gabriel Medeiros, of the BEA's Current Industry Analysis Division, also corresponding by e-mail. Nonprofits are also included in value-added measures (equal to compensation and a measure of capital consumption) that are used for GDP figures. But these figures do not represent a genuine calculation of nonprofit output, but rather sort of a better-than-nothing measure.

"There are many reasons why nonprofit output is hard to pin down," said Medeiros. For starters, measuring the value of services, whether for-profit or nonprofit, is tougher than measuring the value of goods. "There are more imputations made when it comes to services. ... Services are a lot harder to price," he said. In addition, many nonprofit services either have no price or the price charged has been subsidized by other revenue streams.

Because of this and other difficulties, nonprofit output is calculated by the BEA on an inputs (or expense) basis, according to Medeiros. A lot gets lost in the translation, mostly because many nonprofit inputs are nonmonetary. Renz gives an example of a regional mental health clinic in the Kansas City region. It operates on an all-volunteer basis, and even the facility was donated. The only expenses are for utilities like heat, water and electricity, which means "the (expense) measure of input doesn't have any bearing to output" of this mental health facility.

The lack of an accurate output yardstick has a ripple effect on other measures that economists and others typically find useful, like productivity—a measure "we know nothing about in the nonprofit sector," said Shawn Sprague, with the BLS Division of Major Sector Productivity, via e-mail. It's a pretty simple matter, really: Productivity is an efficiency measure for how well economic inputs are converted into outputs. Without real output measures, nonprofit productivity simply cannot be computed accurately. If you want to go through the calculations—outputs divided by inputs over time—the sector's productivity rate is negligible because annual inputs and outputs are the same.

Mere academic debate?

This might all seem very ethereal—interesting to discuss, but ultimately impractical—but the push to measure has already swept much of the nonprofit sector as nonprofit organizations, their boards and funders have taken a keen interest in outcomes, efficiency and performance measurement.

Over the past few years, very public controversies have arisen over the sector's performance and efficiency. In 2003, former U.S. Senator and presidential candidate Bill Bradley created a furor when he suggested that charities were wasting $100 billion a year, mostly on inefficient fundraising and administrative practices, as well as overlapping program services. More recently, the Michael and Susan Dell Foundation, one of the nation's largest charitable foundations, has challenged nonprofits to become more efficient.

Even at the level of individual organizations, the nonprofit sector struggles with the notion of output measurement—and for many of the same basic reasons seen at the broader, sectoral level. In a recent paper on nonprofit organizational effectiveness, Robert Herman and David Renz, both of the Midwest Center for Nonprofit Leadership, wrote that nonprofits struggle to measure effectiveness at the organizational level because they lack "the simple criterion of bottom-line profit or loss. ... The reality is the [nonprofit] effectiveness is more complicated."

So in the place of identifiable outcomes, financial and other input measures often serve as proxies for effectiveness and performance outcomes. For example, much attention is paid by nonprofit watchdog groups and others (notably United Way) to overhead costs and administrative efficiency (known in nonprofit circles as financial or funding ratios). Those with small funding ratios are touted as more efficient because more of an average dollar goes directly for services, rather than administrative costs. Online firms like GuideStar and Charity Navigator have carved a niche in making such information available for hundreds of thousands of organizations, thanks to IRS 990 records.

But some are also beginning to question whether the goal of administrative efficiency might discourage nonprofits from making necessary investments in their infrastructure. Taken to the extreme, it turns into a race to the administrative bottom and assumes that all administrative costs are inefficient and wasteful, when in fact good administrative infrastructure is essential to good programs. (See "Circle, or square?")

An obsession over economic and financial measurement might also be missing the point entirely about the role of nonprofits in society, ignoring important noneconomic contributions that nonprofits make to community and individual quality of life, according to Grønbjerg, from Indiana University. What's important "is not so much what (nonprofits) do, but what difference they make in the community" in terms of quality of life. "We are particularly far away from knowing anything about noneconomic contributions."

Still, La Piana, from California, believes it's fair to ask about nonprofit outcomes. "If nonprofits are in business to combat social problems ... then asking what are they accomplishing is valid. On the other hand, it is a bit like asking if the growth of hospital emergency rooms is reducing the number of car wrecks." La Piana points out that nonprofits often deal "with the results of social problems, not the causes. Child abuse services often do more to help children and families than to prevent abuse, for example. But it is a discussion to have. I have no patience with nonprofits that say it is impossible to measure what they do. It is extremely difficult, but also essential to do so."

Don't know what I don't know

So, given the many gray areas and all the measurement difficulties, is it worth the trouble to try to get a better handle on what nonprofits contribute to the nation's economy?

"Yes, it is important," said Charles Weinberg, who does research on nonprofits as a marketing professor at the University of British Columbia. "If we don't understand a major part of the economy, how are we going to know how the economy operates?"

Better economic data on nonprofits are important "for the same reason that for-profit businesses devour economic statistics—so they can identify trends and stay ahead of them," according to Woods Bowman, a DePaul University economist, and formerly with the Chicago Fed, corresponding via e-mail.

But to many, it's still a matter of out-of-sight, out-of-mind. "To the extent you don't know, you don't know why you should care" about the fact that comparatively few macro data on the industry are available, said Renz, in Kansas City. But he said there are very good—and urgent—reasons for getting a better economic understanding of the nonprofit sector.

"Why it is more important now—to a degree that it concerns me, and I'm not alone—is that nonprofits are an agent of government service delivery," Renz said. The federal government has been abdicating more of that system to nonprofits, and given the resources at stake, not to mention social stability considering the populations served, the system's reliability and efficiency are paramount.

"There are significant policy implications when you don't understand the health of the delivery system. ... If you're going to take this imprecise information (on nonprofits) and make policy, you're really playing the blind-man-and-the-elephant game," said Renz.

So what's the holdup, you might ask? Most point back to the lack of systematic attention and data-gathering from traditional research agencies, particularly the federal government and academia. To Weinberg, the fundamental problem is a dearth of economists and academics who are interested in the nonprofit sector as a research focus. He's particularly puzzled by the fact that many areas of organizational and economic research on the for-profit sector are saturated. That means economists and academics can have a major impact on discovery within the nonprofit field. Yet there is little allure for newcomers.

Why? "It's a positioning problem," says Weinberg, the marketing expert. The sector doesn't receive much attention in the classroom or from scholars, and it is widely called the "third sector," which appears to denote its value—and thus prestige—vis-à-vis research in (or on) the private sector or government. "That's not a good marketing position to be in."

Some, like Bowman, from DePaul, argue that research on the nonprofit sector is really not so far behind. "We know less about it than the for-profit sector primarily because it is a newer field of inquiry," its genesis coming in the early 1970s after federal legislation required more reporting by organizations. He adds that the sector "is arguably more complex, so in a sense there is more to know."

And neither should one be left with the impression that no research whatever exists on the nonprofit sector. In fact, there's a growing body of research at the micro and organizational level on such topics as board governance, management, fundraising and charitable giving, performance measurement and other matters. With 27 chapters, the forthcoming second edition of Walter Powell and Richard Steinberg's The Nonprofit Sector: A Research Handbook promises to make a notable contribution to the economic, political and historical understanding of nonprofits, and will fill in "some conspicuous gaps," according to the editors. But the book—several years in the making—is a case study in the slow-moving nature of nonprofit research.

Medeiros, from the BLS, believes more information will be available in the future. With a growing service economy, "there has been greater interest in the services measures, and both Census and the BLS I'm sure have been fighting for a bigger budget so that they can try to measure that sector of the economy appropriately." But any real progress, he said, "depends largely on the budget climate."

Rafter, from INOM, hopes that the researchers will continue to produce better data on the sector, but sees something of a chicken-and-egg scenario. There are very few data on nonprofits, he says, because there is very little demand for it, and there's low demand because people know so little. Rafter has done work in producing regional snapshots of the nonprofit sector in California, and he said, "people are not knocking down our door for it."

But as more data become available, and they're tailored to geography, subsector, mission and other specific targets, the sector will start demanding more information because of its intrinsic value in understanding trends. Rafter believes much of the demand will come from foundations and other philanthropists who are beginning to require new metrics for evaluating their grant-making.

Currently, he said, "the data are very imperfect. But I have hope that in 10 to 20 years they will be much better."


Related Links