Accounting as a Tool for Legitimization of Nonprofit Organizations

This article’s purpose is to sensitize scholars and students to the powerful role accounting plays in the nonprofit sector. People give to nonprofit organizations, not for their own material benefit but for other reasons. Although some may give to enhance their prestige or to pacify an employer (Gordon and Khumawala 1999; Khumawala and Shroff 2023), many give to get what Andreoni (1989) refers to as a “warm glow.” However, nobody feels a warm glow if they worry they are being scammed. Potential donors need reassurance that charities are legitimate and will use donated funds prudently. When donors are not able to personally observe how nonprofits distribute funds, they must rely on reports.¹ This situation is an example of the “principal-agent” issue common in business.²

Accounting fills a key role in engendering legitimacy for good organizations. A combination of legal regulation and accounting disclosure has helped foster a large and growing nonprofit sector in the U.S. Over 1.6 million 501(c)(3) organizations were registered with the IRS in 2020, per Independent Sector (2021) data. In their 2022 survey, Independent Sector (2022, 13) found 59 percent of Americans trust that nonprofit organizations “will do what is right.”

Organizations that do not meet norms in reporting or controls or accounting measures risk being deemed illegitimate by donors and regulators. Among the 12 percent of respondents to the Independent Sector (2022, 22) survey with low trust in nonprofit organizations to “do what is right,” common concerns involved scandals, lack of transparency, and fear of organizations “potentially pocketing money.” Accounting and auditing play an important role in identifying scams. However, there is a real danger of mislabeling legitimate organizations as illegitimate.

I use three vignettes to illustrate the key role of accounting and controls in establishing, or impairing, organizations’ credibility. In the first, in approximately 814 B.C.E., the Bible says that King Jehoash of Judah became concerned that donations for repair of the Temple in Jerusalem were being misused and instituted a “lockbox” system of controls that helped ensure funds were used appropriately. In the second, William Booth, the first leader of The Salvation Army, used internal controls and regular audited reports in the U.K. in the 1800s to engender confidence. In the third, a U.S. company called Pallotta TeamWorks (“PTW”) used internal controls, auditing, and transparency to try to establish credibility for its fundraising efforts on behalf of AIDS, breast cancer, and other charities. It initially had tremendous success, but in 2002, PTW went out of business, in large part because of public concerns that the costs of its events were above established norms.

After discussing the vital role accounting plays in reducing agency costs and supporting public trust in the nonprofit sector, I make four other points. First, some accounting measures frequently used by donors are highly imperfect indicators of whether nonprofit managers are operating effectively or whether they are operating in good faith. Using imperfect indicators can result in two errors—erroneously accepting “bad” organizations as “good” and unfairly condemning good organizations. Second, nonprofit managers react strategically to being measured. This can lead both to (desirable) real efforts to improve controls and efficiency and to dysfunctional actions, such as cutting out necessary administrative functions and gaming the accounting system. Third, any use of accounting to separate good from bad actors will favor some types of organizations over others. Too little attention has been paid to which types of organizations have supported or opposed particular accounting measures. Finally, it is important for potential donors and regulators to use accounting reports carefully and knowledgeably. Much can be done to help educate the public about how to evaluate organizations.

The Bible provides two highly similar accounts of an incident occurring during the reign of King Jehoash of Judah.³ The king had previously ordered extensive work to be done to repair the Temple. Priests and Levites were supposed to collect money from the people and use it for repairs. During the 23^rd year of his reign, he grew concerned that the work was not getting done.⁴

The king and the priests agreed on a new system of handling collections that instilled accountability and engendered trust. Instead of priests collecting money from people in various towns (as the II Chronicles account describes) or individually from donors (as the II Kings account describes), the priests agreed to take no part in either the collection process or the supervision of repairs. Donors put their payments for the temple restoration project into a hole in the top of a locked chest, located near the door of the Temple. Whenever the chest was getting full, the high priest and the royal scribe would jointly open it, count the contents, and deliver the money to the people supervising the repairs. The money was used solely for such direct costs as wages for laborers and carpenters and for materials like wood and hewn stone.⁵

According to II Chronicles, chapter 24, verses 10 and 11, people “gladly” brought money and threw it into the chest. “Much money” was collected. There was more than enough to finish the job. Verse 13 says that the money left over was used to make gold and silver temple utensils.

Several points are noteworthy in this story. First, the temple repair project was, by its nature, highly visible to the people and the king. King Jehoash was in a position to realize the work was not happening, and he lost trust in the priests. Second, the original system, where the priests directly collected the money for repairs, was subject to abuse. The priests had the opportunity to divert the donations meant for repairs to other purposes, including their own personal expenses. There were reasonable grounds for public suspicion of the process. Third, the king and priests put into place a highly visible process, in which the counting was done by two parties who did not report to each other—the king’s scribe and the high priest. Also, the repair work was supervised by people who had no duties of collecting funds. This system, combined with visible progress of the repair work, restored trust in the project to the point where people gave “gladly.”⁶

Irvine’s (2002) description of the early history of The Salvation Army in the U.K. shows how accounting can give an organization credibility.

In 1865, when William Booth started the organization that grew into The Salvation Army, he needed to raise money. Unlike the Anglican Church, The Salvation Army would not get tithes raised by the government. Booth had no connections with the aristocracy or upper classes, so he needed to raise money from a public that was not used to fundraising appeals.

In some ways, his problem of combatting mistrust is similar to the first vignette. People worried that Booth would use their donations to enrich himself. But The Salvation Army’s task of establishing credibility was greater than the ancient priests’ for at least two reasons. First, the work of The Salvation Army in helping the needy was less visible to donors than was the repair work on the Temple. Second, The Salvation Army had numerous branches that collected and used funds rather than one central site.

From the very beginning, there was public concern that Booth would keep the contributions. Per Humphreys (2015), the 1878 founding documents of The Salvation Army allowed Booth total control of the funds, which seemed to justify skeptics’ concerns about potential self-dealing. On the other hand, the founding documents required annual audits. Booth pledged he would take no money for himself. He did not draw a salary, and he died poor. This helped to address concerns about his own conduct.

More generally, the organization established a rigorous accounting system. Each local “division” prepared monthly “balance sheets,” and each regional “corps” produced them quarterly. The central organization produced annual, audited “balance sheets.” From the examples reproduced by Irvine (2002), these were not balance sheets in the modern sense of the word. Instead, they showed inflows into the organization during a period from offerings and other sources; outflows for rent, salaries, and other expenses; and balances of funds left on hand.

Auditing was used to enhance control and credibility. The headquarters audited statements from “divisions” and “corps.” Outside auditors reported on the overall organization’s statements from the 1860s onwards. The audited balance sheets were published in the organization’s magazine, The War Cry. Irvine (2002) reproduces one of these balance sheets, from 1880, including the following promise of transparency:

No system of controls is perfect, and the occasional failures emphasized the vital importance of controls and accountability. For example, accounting irregularities led to Switzerland expelling The Salvation Army in 1883 (Humphreys 2015).

Although there were some criticisms and some control failures at particular divisions, The Salvation Army used its record of controls and audits to respond to critics. It generally grew throughout this period and is now one of the world’s largest charitable organizations (Barrett 2022).

In the early 1990s, Dan Pallotta founded a fundraising business, Pallotta TeamWorks (“PTW”), which grew and thrived through 2001 but collapsed in 2002. Its history through 2001 illustrates again the importance of accounting in establishing donor trust and enabling growth. The collapse in 2002 illustrates the critical effect of public perception that an organization has not met accounting norms of performance. In this case, as discussed later, the norms were of questionable appropriateness. One reason to focus on this case is the unusual amount of available material, including Pallotta’s books (Pallotta 2008, 2012, 2020), his TED talk (Pallotta 2013), data on the PTW website (Pallotta TeamWorks (PTW) 2002), a Harvard Business School case written before the collapse (Grossman and Kind 2002), case studies by Tinkelman (2007, 2009), and news stories. A second reason is the size of PTW’s operations. Over its nine years of operations, it attracted about 180,000 walkers or riders to its events and raised over $305 million in contributions (PTW 2002).

PTW’s main business was organizing large fundraising events for charitable sponsors. Between 1994 and 2002, it organized numerous bicycle rides for AIDS and AIDS vaccine research and three-day walks for the Avon Foundation’s breast cancer research program, as well as a few events to raise money for other causes.

A typical PTW breast cancer 3-Day Walk or AIDS Ride was a major undertaking, with many participants. The 13 3-Day events in 2002 averaged close to 3,000 participants each, and AIDS bicycle rides in 2001 and 2002 generally had over 1,000 participants (PTW 2002). Each walker and rider had to obtain a significant level of pledges from friends and relatives to participate. The average contributions raised per participant for PTW’s 3-Day Breast Cancer Walks and AIDS bicycle rides after 1998 were over $3,000.

The ability of these events to generate profits for research was heavily dependent on participant volume. They illustrate the relations between cost, volume, and profits we typically teach in managerial accounting courses. Here, I use the term “returns for the cause” in place of “profits.” The events required significant up-front funding to cover costs of planning, attracting participants, arranging routes, obtaining insurance, organizing tent cities to lodge participants, setting up sound stages for evening activities, and ensuring there was adequate food and medical care available. PTW spent $15 million on marketing its 2001 events (Grossman and Kind 2002). The incremental costs of servicing an extra participant were quite low, e.g., providing a t-shirt and some meals. Thus, we would expect to see no “returns for the cause” when events had low levels of volume and then returns increasing, as a percentage of donations, after some break-even volume level. The percentage “return to charity” for PTW’s various AIDS bicycle rides varied from 0 to 73 percent, largely correlated with volume levels (PTW 2002).

PTW’s ability to succeed depended on its credibility with both its charitable sponsors and with donors. Each of PTW’s events was sponsored by a nonprofit organization, such as an AIDS group or the Avon Foundation. These groups provided both seed money and their imprimatur, so it was vital for PTW to maintain their trust. Second, the success of the events depended on the willingness of walkers and riders to ask their friends and families for large pledges. Thus, the riders and walkers had to be so confident in PTW’s handling of funds that the participants would ask their friends to give.

Dan Pallotta and PTW had to overcome considerable public skepticism for several reasons. First, PTW was a for-profit company, and people were concerned that Pallotta could personally profit to an excessive degree. Second, such large events, with minimum pledge levels, were unconventional. Pallotta (2008) believed that, in order to attract large numbers of participants, PTW had to be willing to spend. Critics disliked the idea of “lavish” spending, whether that related to PTW paying its staff market wages, its use of “glossy” promotional materials, paying for full-page advertisements, or the quality of food at its events.

Pallotta made a number of efforts to obtain and keep public trust. PTW worked on a fixed fee for each contract, set in advance. PTW had no custody over donor contributions. All funds came into a bank “lockbox” controlled by the donor (Grossman and Kind 2002). (This use of a lockbox harks back to King Jehoash.) Because the donations were made directly to a charity’s lockbox, not to PTW, they were deductible. PTW had to submit detailed receipts to its sponsors to be reimbursed for expenses. PTW complied with applicable state filing requirements for professional fundraisers. PTW (2002) voluntarily published, on the internet, a detailed “Record of Impact” that showed, for each event, the number of participants, the average amount raised per participant, total contributions raised, the total dollars netted by the charitable organizations, and four types of expenses: participant support, marketing and “awareness” of the cause, administration, and PTW’s production fees.⁷ Over the 1994–2002 period, PTW’s production fees totaled about $23 million, or roughly 4 percent of the $556 million total charitable donations from participants. Pallotta (2008) argues that his firms’ fees were modest in proportion to the success they achieved. PTW noted that its practices were praised in a Harvard Business School case study (Grossman and Kind 2002).

For some time, PTW’s efforts succeeded, and the business thrived. From 1997 to 2001, per PTW’s (2002) Record of Impact, the AIDS bicycle rides had roughly 8,000 to 9,000 participants per year, raising between $26 and $29 million per year. Participation in the Avon Breast Cancer 3-Day Walks grew from about 2,300 in 1998 to 26,300 in 2001, and related contributions rose from $7 to $85 million.

In 2002, PTW ran its largest number of events and raised its largest amount of money for charity but collapsed when its sponsors abandoned it. The sponsors’ underlying concern was that the PTW events were getting a reputation for not returning enough money, as a percentage, to the causes. Up through 2002, the Council of Better Business Bureaus had considered returns to charity of 50 percent or more to be sufficient (Tinkelman 2007). In 2001, the average return on PTW’s AIDS Rides was only 47 percent and the returns on its AIDS Vaccine Rides were only 21 percent (PTW 2002). A 2002 The New York Times article (Harbert 2002) by a participant in an AIDS Vaccine ride was titled “Charity on Wheels: Riding your Heart Out, Then Feeling Betrayed.” The participant was upset that too much of the funds collected, in her view, was spent on event expenses. In April 2002, a class-action lawsuit was filed against PTW on behalf of participants in the 2000 and 2001 AIDS Vaccine Rides, citing the low returns to charity reported by PTW and claiming PTW had “taken too much money off the top” (Kaiser Health News 2002).

Such criticism hurt PTW’s ability to attract participants to its events and, critically, its relations with the events’ sponsors. Per Pallotta (2008), the group that had sponsored the California AIDS Ride in prior years expressed concern that the event costs were too high a percentage of the donations raised and started its own competing 2002 event. The number of AIDS Rides participants fell 54 percent, from 8,349 in 2001 to 4,507 in 2002, in large part due to the competition and the controversy.

In April 2002, the Avon Foundation, which had sponsored the 3-Day Breast Cancer Walks, expressed a similar concern and ended its contracts with PTW (Pallotta 2008), effective after the end of the 2002 fundraising season. In part, the controversy over the AIDS Vaccine Rides was affecting PTW’s reputation. In addition, Avon was aware that the standards used by monitors would become tougher in 2003 (Tinkelman 2007), and future compliance was likely to be difficult. Some monitors already looked for events to have at least a 65 percent return to charity, and the average for the nine 2001 Breast Cancer 3-Days was only 60 percent, with four events returning lower amounts. Thus, Avon was subject to criticism for its association with what were seen as high-cost events. The PTW’s 2002 Breast Cancer Walks for Avon were its last events. The breast cancer events were such a major part of PTW’s business that PTW could not survive without them.⁸

Ironically, the very efforts to be transparent that had helped PTW establish its legitimacy were used against it. Because PTW did not try to game the system by classifying any marketing expenses as program expenses, its ratios looked worse than competitors. Because it separately reported statistics on every event, critics could focus on its least successful events as evidence of its inefficiency. Thus, the class action lawsuit brought on behalf of Vaccine Ride participants cited the low 21 percent return to charity for those rides in 2001 but ignored the overall average return on all PTW events in 2001 of 52 percent. The overall average met then-existing guidelines, but the lawsuit, using data PTW disclosed, focused on events that violated the guidelines.

The principal-agent problem arises whenever some “principal” has trouble seeing what his or her “agent” is doing (Jensen and Meckling 1976). In the nonprofit world, donors are “principals” and nonprofit organizations are their “agents.” The donors want organizations to use donations to further good causes. Typically, donors cannot see directly how the money is spent. For example, American donors to UNICEF who want to help impoverished children in foreign countries are highly unlikely to go to the places where UNICEF distributes aid or to meet the children it helps.

Unfortunately, there have been many examples throughout history of donated funds being misused or stolen. Fishman (2007), who studied scandals involving nonprofit organizations from 1200 to 2000 C.E., wrote, “One thing I can say with assurance is that scandal and fraud have always been rampant among charitable organizations.”

Both large- and small-scale scams continue to arise. In the 1990s, John Bennett was convicted of defrauding donors to New Era Philanthropy of $135 million, essentially through a Ponzi scheme (Stecklow 1996). In 2021, the Federal Trade Commission, along with 46 other agencies in 39 jurisdictions, filed a complaint against a number of for-profit charity fundraisers of making “abusive, unsolicited [and] deceptive fundraising calls.”

Donors have good reason to be cautious, just as they did in the time of King Jehoash.

To survive, the nonprofit sector needs donors’ trust. From early times, the tools used to reduce the principal-agent problem for donors have included laws, internal controls, and accounting reports.

Donors rely on the law to stop abuses by nonprofit organizations in three major ways. First, the law forbids fraud. People who ask for money under false pretenses can be sent to jail. John Bennett was sentenced to ten years in prison for the New Era Philanthropy fraud (Stecklow 1996). Second, donors can restrict their donations to particular purposes. Organizations can be sued for diverting funds to other uses. Third, no one owns, and has a right to profit from, a nonprofit organization. Unlike regular businesses, charities are not legally allowed to pay out profits to anyone. Without a profit motive, managers are likely to focus more on fulfilling the organization’s mission of service. Many scholars, following Hansmann (1980), believe this “nondistribution constraint” is vital to the existence of the nonprofit sector. Donors are more likely to get a “warm glow” from giving to a nonprofit organization than from one that distributes dividends to shareholders.⁹

“Internal controls” are the various processes that organizations use to ensure that their policies are followed and to reduce the likelihood of fraud. Nonprofit organizations use a variety of control procedures, including putting the ultimate authority in the hands of a board of trustees rather than the executive director. The use of control procedures has a long history, and the three vignettes in this article all demonstrate internal controls. The procedures put in place by King Jehoash included centralized collections, separation of the building renovation donations from other donations, use of a locked box, dual people opening cash receipts, and establishing accountability over cash receipts. The Salvation Army and PTW used a variety of internal controls, audits, and reports to show they were not diverting donated cash.

For donors who ask, “What is the charity doing with my money?,” accounting can report what money was spent on.¹⁰ Accounting reports serve to legitimize the reputable organizations. Audits by well-known and trusted accounting firms enhance the reports’ credibility.

U.S. charities today rely on regular provision of financial information to reassure donors. Nonprofit organizations above a minimum size generally must file financial and governance information each year with the IRS on Form 990.¹¹ The Form 990 was first required in the 1940s and has been widely available since the 1990s (Hager and Flack 2004). Form 990 contains financial information, names and compensation of board and key officers, information about amounts paid to major contractors, natural and functional expenses, professional fundraising fees, descriptions of program service accomplishments for the three largest programs, information about lobbying activities, and loans to or from major donors or officers. By law, these filings are open to the public, and the IRS now makes them available on its website. They are readily available on the internet through sites like Candid (formerly GuideStar) and ProPublica’s Nonprofit Explorer.¹² Although the IRS does not require these reports to be audited, some states require audits of organizations above a certain size.¹³ Many large nonprofit organizations voluntarily publish annual audited financial reports, because audits add credibility.¹⁴

Foundations and government agencies typically require detailed reporting on how their grants were used. They may have the right to do special-purpose audits of grantees.

There are also several independent monitoring groups that issue ratings on charities. They include the Better Business Bureau (BBB) Wise Giving Alliance, Charity Navigator, and Charity Watch. Organizations that receive high ratings from these organizations brag and use the ratings to burnish their images.¹⁵ Organizations can pay to display the Better Business Bureau seal on their websites. Research has generally supported a link between favorable monitors’ ratings and donations (Neely, Saxton, and Maharaj 2023).

The combination of laws, internal controls, private monitoring, and accounting reports has done a tremendous amount to reassure donors. As noted earlier, there are now about 1.6 million 501(c)(3) nonprofit organizations (Independent Sector 2022). According to Giving USA (BWF 2022), estimated 2021 charitable giving was $485 billion.

Although the FASB in Statement of Accounting Concepts No. 4 (Financial Accounting Standards Board (FASB) 1980) recognized that information on both the service efforts and accomplishments of nonprofit organizations was important to donors, Paragraph 53 states

In the decades since then, the FASB has not promulgated any rules for measuring and reporting service accomplishments. In the absence of such requirements, donors have used other information, especially spending data, to judge organizations.¹⁶

Since the 1960s, charities have often been compared based on their “program ratio,” their “overhead ratio,” and the ratio of fundraising expenses to donations raised (Hager and Flack 2004). Both archival and experimental research indicates that charities with “better” program and fundraising ratios are more attractive to donors.

The program ratio is the percentage of expenses spent on program services. The overhead ratio is the converse: the percentage of expenses spent on things other than program services. For nonprofit organizations, that means fundraising and management and general (administrative) expenses. The program ratio and overhead ratio, together, add to 100 percent, so organizations with high program ratios have low overhead ratios and vice versa.

The ratio of fundraising costs to donations raised is used to gauge the efficiency of fundraising efforts in terms of generating net revenue from particular events or campaigns “for the cause,” after accounting for expenses of raising the money.

The ratios are easy to compute, using data organizations report on their Form 990s or financial statements. They seem to be a useful tool to compare organizations. Since most donors do not have the time or energy to do extensive research before donating, such simple measures are attractive (Styron 2021).

Monitors and periodicals that have reported such ratios include Forbes, U.S. News & World Report, Worth, Money, GuideStar, the Combined Federal Campaign, the Better Business Bureau’s Wise Giving Alliance, Charity Navigator, and Charity Watch (Hager and Flack 2004).

The ratios considered acceptable have changed over time. In various opinion polls from the 1980s to 2012 (cited in Garven, Hofmann, and McSwain 2016), what people thought was an appropriate minimum program expense ranged from 77 percent to 82 percent. This implies people think overhead should be, at most, 23 percent. When the BBB Wise Giving Alliance was formed by the merger of two other groups in 2002, it indicated that organizations should spend at least 65 percent of their expenses on program expenses. It also indicated that fundraising expenses should not exceed 35 percent of related donations (Tinkelman 2009). These standards were more restrictive than those of its predecessor organizations. As discussed above, most PTW events complied with the prior standards but appeared excessively costly under the 2003 standards (Tinkelman 2007, 2009). The change in standards added pressure on PTW and the charitable sponsors of its events.

An overhead ratio measure is, at best, an incomplete answer to the donor’s information needs. Assume that a donor wants to help supply vaccines to poor people in another country. She wants to give to the charity that will use her money to vaccinate the most people. So the ultimate question is how many people will be protected from disease based on her gift.

Let us follow the process between her gift and people being protected.¹⁷ She contributes money.

1. The organization either spends it this year or saves it. The money saved may be spent to immunize people in years to come. It may also be used on fundraising efforts that bring in money for additional vaccinations. It may earn investment returns that can help fund future purchases of vaccines.

2. The money spent this year is spent on some combination of administrative costs, fundraising, and programs. Fundraising efforts will bring in more money, much of which will be used to immunize people. Administrative spending may make the organization’s programs more effective. Thus, both administrative and fundraising efforts may help the organization do exactly what the donor wants—immunizing more people. However, both make the overhead ratio higher and thus “worse.”

3. Some of this year’s spending is on programs. This is the only part of this process that is captured by the program ratio.

4. Some money spent on programs will be spent effectively, and some will not. The program expense ratio does not distinguish productive from wasteful spending. A program ratio looks higher for an organization that overpays for salaries and services than for one that operates efficiently. The ratio does not measure how many people were given vaccines or track whether vaccines were carefully refrigerated so they had full effectiveness.

Running an effective program requires making choices between saving or spending funds received and between spending on programs now versus investing in fundraising efforts and administrative infrastructure. Managers must make trade-offs between the quality and quantity of current service efforts. Focusing on just the ratio of current amounts spent on programs and total expenses is myopic and distorts decision making. Program quality suffers without adequate administrative structure. Limiting fundraising stunts future growth.

A second way percentage ratios are incomplete measures is that they ignore the scale of the effort. Small events, such as local bake sales, may have very low fundraising ratios, but they don’t raise much money. PTW events were large and required professional staff as well as some volunteer help. Pallotta (2008) has consistently argued that raising major sums for research, and actually finding a cure for cancer, is more important than keeping overhead ratios low.

A third way the traditional measures are incomplete is that most volunteer effort is not recognized and measured. Volunteer-led bake sales would look far less efficient if the time of the volunteers was valued and recorded as both revenue and as fundraising cost.

Another major problem with making decisions based on a program service ratio is familiar to anyone who has taken an introductory economics or managerial accounting course. The program ratio shown to our donor reveals, on average, what percentage of last year’s expenses was spent on programs. She shouldn’t care about last year’s average effect. She should be focused on the incremental impact of her donation this year, i.e., the “marginal” impact. Unfortunately, as a general matter, the average and marginal program ratios are unrelated.¹⁸ Donors have no way of determining the marginal impact of their contributions from financial statements.

The PTW events discussed above exemplify the difference between the average “return to the cause” and the marginal returns from an additional event participant or donor. Regardless of the overall average fundraising cost of a PTW event, once the upfront costs were covered, the marginal costs associated with additional participants were very low. The vast bulk of the money raised by incremental participants went “to the cause.” The point holds more generally—for any activity with high fixed costs and low variable costs, the major factor in how much money it raises is the level of participation. The more people who come, the lower the costs per person and the better the ratios look.

There can be many reasons, other than dishonest or self-enriching management, for an organization to have “bad” overhead or fundraising ratios in a year or on a particular event. Unpopular causes, or those with widely dispersed or low-income donor bases, need to spend more heavily to raise funds than popular causes with highly concentrated groups of wealthy supporters. New organizations need to spend to develop reliable donor lists. In some cases, the expenses related to raising funds may come in an earlier year than the related donations, skewing ratios for both years. Lecy, Searing, and Li (2023) present evidence that, on average, larger organizations have higher program ratios than smaller ones. This would be consistent with there being a fixed cost component to management and general expenses.

Ratios can be heavily impacted by factors other than the managers’ good or bad faith. The PTW Record of Impact data (PTW 2002) illustrate this point. PTW oversaw 34 Breast Cancer 3-Day Walks from 1998 to 2002. Even though these were highly similar events, with the same event organizer and the same charitable sponsor, the returns to charity varied considerably. For the 13 events in 2002, the returns averaged 57 percent but ranged from lows of 26 percent and 37 percent to highs of 66 percent. Factors such as weather, donor fatigue after the 2001 attacks, competing events, and adverse publicity all affect turnout, and turnout affects the percentage return.

Nonprofit organizations react in predictable ways to being judged on their overhead and fundraising ratios. Some lobby about the measures. Some use funds more efficiently for their missions. Some take economically unsound actions to avoid criticism. Some use accounting alternatives, or lies, to appear to comply. And some revolt against being measured.

Nonprofit managers understand that donors and the public want organizations to focus on providing services efficiently. Reputable organizations try hard to operate effectively at low costs. They use significant volunteer labor and donated materials. They avoid having lavish offices. Avon’s breast cancer walks after 2002 tried to reduce costs by making the events last two days with a circular route, with only one overnight camping location, as opposed to the prior three-day events, with two campsites plus a location for a closing ceremony (Tinkelman 2007, 2009).

To avoid criticism, managers often underspend on administrative costs and pass up opportunities to raise funds. In a 2017 survey of 200 nonprofit executives, Parsons, Pryor, and Roberts (2017) found that about 70 percent of managers reported feeling pressure to report favorable ratios to appeal to donors. Over half said they were willing to make real changes in spending to report better ratios, especially by cutting administrative expenses. “Furthermore, 30 percent of respondents claim they would accelerate program expenses and 18 percent indicate they are willing to manipulate accounting information to meet a financial benchmark.”

Mitchell and Calabrese (2019) say “one of the best substantiated conclusions in nonprofit studies” is that minimizing overhead, rather than improving organizational effectiveness, impairs it. They cite studies associating overhead minimization with lower effectiveness, fundraising inefficiency, reduced organizational resilience, underinvestment in administrative capacity, and increased financial vulnerability. Harmful effects of minimizing fundraising ratios include “increased compliance and regulatory costs, misleading solicitations and misled donors, fundraising inefficiency, and inefficient charitable output provision.”

Mitchell and Calabrese (2019) conclude these harmful effects “may be acceptable if the purpose of nonprofit management is to demonstrate trustworthiness through conformity to fiscal norms but is problematic if the purpose of nonprofit management is to efficiently achieve meaningful organizational outcomes.”

The limited academic research on the link between overhead spending and organizational effectiveness suggests that there is a “sweet spot” between spending nothing on programs and spending 100 percent on programs. Below some level, organizations don’t spend enough on central administration and fundraising to be fully effective. Above the sweet spot, the additional spending on management and fundraising doesn’t have a commensurate program benefit. The research is limited and should be treated cautiously, but it suggests the sweet spot for overhead may be from around 30 percent to around 40 percent of expenses.¹⁹ The sweet spot is likely to vary across types of nonprofit organizations, depending on their cost structures and sources of funding. This is roughly consistent with the BBB Wise Giving Alliance guideline that program expenses should be at least 65 percent of total expenses.

The Urban Institute’s Nonprofit Overhead Cost Study found many examples of organizations that were starved for administrative capacity. Wing, Hager, Rooney, and Pollak (2005) report:

Their article was titled “Paying for Not Paying for Overhead.”

People under pressure to meet an accounting standard have incentives to either shade their accounting in favorable ways, or to lie. As noted above, Parsons et al. (2017) survey of 200 executives found 18 percent admitted to being willing to manipulate their accounting to show better ratios.

Study after study has found the same thing—an implausible number of organizations report spending nothing on fundraising.²⁰ In a 2004 paper published as part of the Urban Institute’s Nonprofit Overhead Cost Project, the Urban Institute (2004) looked at a database of about 220,000 filings on Form 990 and noted “One-fourth of nonprofits reporting $1 to 5 million in contributions report zero fundraising costs.”

Organizations use a variety of methods to show low fundraising.²¹ Some aggressively allocate a large share of joint costs to programs and a low amount to fundraising (see Jones and Roberts 2006). Per Hager (2003), some classify costs of paid fundraisers in program categories. Keating, Parsons, and Roberts (2008) found 74 percent of the regulatory filings in their sample misclassified some costs of using fundraising telemarketing firms. In some cases, organizations report the donations received net of fundraising costs, which improves the ratio of costs to donations. In other cases, the costs and fundraising events are run by affiliated organizations and the charitable beneficiary reports only the net proceeds, hiding the fundraising costs.

In a 2009 article, Gregory and Howard (2009) coined the term “Nonprofit Starvation Cycle” to summarize the harmful process that damages nonprofit organizations. The starting point is donors’ unrealistic expectation that organizations can thrive with minimal fundraising and administrative spending. In the next step, some organizations cut overhead spending to suboptimal levels and others under-report their actual overhead spending. In both cases, they seem to be complying with donors’ expectations. This reinforces donors’ belief that their expectations are reasonable and puts even more pressure on organizations to cut costs. Gregory and Howard (2009) expect the process to continue until organizations, starved of adequate infrastructure, go out of business.²²

The evidence is that, indeed, over time, nonprofit organizations are reporting spending less on overhead. Lecy and Searing (2015) found that over 25 years, starting in 1985, the reported average overhead costs of nonprofit organizations fell by 2.6 percentage points, from 20.9 percent of expenses to 18.3 percent.

PTW may have been naive, because it made no effort to classify any of the event expenses as program costs. Most similar organizations did allocate substantial parts of event costs to programs. Tinkelman (2007) looked at cost allocations by 16 organizations that raised over $5 million for causes related to cancer in 2005. The average amount of joint costs allocated to programs was 44 percent. Jones and Roberts (2006) studied joint cost allocations by 155 charities and found an average of 53.6 percent allocated to programs. Based on these studies, it seems PTW might have justified allocating about half the event expenses to programs. Tinkelman (2007) notes Avon allocated even more of the expenses of its successor breast cancer walks to programs. Dan Pallotta (2008) noted,

If everyone else is slanting their numbers, organizations that play it straight look bad.

When people resent an accounting measure of performance, they revolt against it. Dan Pallotta, stung by the collapse of PTW, has been a leader in raising public consciousness about the problems of using overhead ratios. He has written books (Pallotta 2008, 2012, 2020) and given two TED talks that have had over six million viewers.²³ He founded an organization, the Charity Defense Council, to be a voice speaking out against unfair criticism of nonprofit organizations and resisting restrictive regulations.

The Charity Defense Council has tried to change the narrative about both overhead costs and fundraising. For example, it points to the committed, essential workers at nonprofit organizations whose salaries are classified as “overhead.” It sells t-shirts that say, “I’m Overhead.” Pallotta (2012) has suggested running ads with a tagline of “I don’t want to be the only donor” that point to the importance of fundraising in mobilizing support.

Some influential voices in the nonprofit world have joined in trying to de-emphasize overhead ratios. The heads of GuideStar, the BBB Wise Giving Alliance, and Charity Navigator signed a letter titled “The Overhead Myth” (Berger, Harold, and Taylor 2013), saying, “The percent of charity expenses that go to administrative and fundraising costs—commonly referred to as overhead—is a poor measure of a charity’s performance.” They go on to say that, although extreme values of the overhead ratio could indicate fraud or poor management, “in most cases focusing on overhead without considering other critical dimensions of a charity’s financial and organizational performance does more damage than good.” They urge donors to pay attention to other facets of nonprofit performance, including governance and results. Berger et al. (2013) add:

However, the BBB Wise Giving Alliance and Charity Navigator continue to use ratios in their rating standards.

Accounting standards and practices are created in a process that involves lobbying by interested parties. Charitable monitors’ standards are also subject to comment and pushback by various parties. Clearly, as discussed above, many nonprofit organizations want to substantiate their legitimacy by showing they use funds responsibly. They have an interest in having credible accounting and auditing standards.

According to Seville (1987), the early standard-setting efforts by the AICPA and the Accounting Principles Board did not apply to nonprofit organizations. For years, there were no authoritative nonprofit accounting standards. However, once the AICPA issued Statement of Auditing Procedures No. 23 in 1951, which required auditors to say whether statements were in accordance with GAAP, it became necessary to define nonprofit GAAP.

Some well-established nonprofit organizations tried to act collectively to fill this gap. In 1964, the National Health Council and the National Social Welfare Assembly (1964) collaborated to publish Standards of Accounting and Financial Reporting for Voluntary Health and Welfare Organizations. Together, these groups had 54 member organizations. Per Seville (1987), these standards were influential and were widely seen as best practices. In 1975, the National Health Council, National Assembly of Social Workers, and United Way of America published a revised version of these standards. These standards remained influential for years.

The FASB began a project to define the objectives of nonprofit accounting in 1973, which resulted in a Statement of Accounting Concepts in 1980 (FASB 1980). FASB (1980, paragraph 59) showed its awareness of the link between accounting and public confidence in nonprofit organizations:

In 1993, the FASB asserted itself as the standard setter for nonprofit accounting when it issued Statements of Accounting Standards No. 116 and 117 (Financial Accounting Standards Board (FASB) 1993a, 1993b). These statements provided rules for recognizing donations, classifying expenses, categorizing net assets, formatting financial statements, and other key matters.

Any accounting guidelines used to distinguish good from bad organizations will favor some types of organizations over others. One can therefore expect different organizations would favor, or disfavor, particular rules.

As an example, the FASB codification generally forbids the recognition of volunteer efforts unless certain specified criteria are met.²⁴ Although a major reason for this rule may be related to the difficulty of tracking and valuing volunteer time, it also serves to make organizations that use large numbers of volunteers for fundraising events look more efficient than if they had to record the value of volunteer time as both a contribution and as a fundraising cost. Thus, the rule makes the fundraising ratios of volunteer-staffed events look better relative to those staffed by professionals.

The FASB (1993b) required organizations to report their expenses in the functional categories of management and general, fundraising, and program in order to help financial statement users assess how organizations were using their money. The IRS Form 990 already required this classification. Thus, both the FASB and the IRS required organizations to report the data needed to compute fundraising, overhead, and program ratios.

Monitoring organizations have set their standards for fundraising and overhead ratios at levels that are easy for established organizations, addressing popular causes, to meet. Bhattacharya and Tinkelman (2009) looked at the expected impact of the BBB’s new financial ratio standards in 2003. In their sample of 469,525 organizations’ 2001 Form 990s, about 75 percent reported spending zero on fundraising or administrative costs or both. Bhattacharya and Tinkelman found this implausible, but clearly such organizations reported low overhead. Of the remaining approximately 111,000 organizations, about 75 percent would have met the pre-2003 standards and slightly more, about 79 percent, would have met the 2003 standards. This leaves only about one out of five of the 111,000 organizations, or about 5 percent of the full sample, who would not report complying with the standards.

As Hager and Flack (2004) note, using overhead and fundraising ratios to judge organizations favors organizations that are larger and long-established, have popular causes, and willing to use sophisticated accounting methods to skew their reported results. Such organizations can be expected to support the use of ratio measures. The Supreme Court has noted that there is no necessary connection between high fundraising and fraud and that laws restricting fundraising ratios discriminate against unpopular causes.²⁵ Based on these laws’ impact on the freedom of speech of unpopular causes, the Supreme Court has consistently struck down laws requiring nonprofit organizations to meet minimum program spending ratios or to avoid high ratios of fundraising to donations.²⁶

Organizations that see the standards as unfair or inappropriate are in an awkward position. They have to argue against the standards on logical grounds and also overcome people’s suspicion that they are too badly run to meet reasonable standards. This is quite a challenge, since opinion surveys consistently indicate donors believe overhead spending levels are important. Thus, it was hard for Dan Pallotta to get allies in protesting that the BBB standards were unreasonable.

Lobbying for loose standards has been subtle and away from public attention. It has generally involved accounting rules defining the expense categories. The definition of “management and general” expenses for reporting purposes is far narrower than the popular conception of overhead, and the definition of “program expenses” is wider than most people believe. Also, if an organization is trying to both raise money and educate people about a cause during an event, some of the money spent gets classified as a public education program expense.

Under current GAAP, and under prior industry standards dating back to the 1960s, program expenses include all the direct and indirect costs directly related to that program. For example, the program expenses of a charity that supports cancer research include not just the amounts it pays to researchers but the occupancy costs of the buildings they work in and the costs of the people who supervise them. Only costs that relate to the overall functioning of the organization, rather than particular programs, are categorized as “management and general.”²⁷

The issue of allocating costs arises when organizations spend money on raising awareness of particular problems as part of their mission. For example, breast cancer charities try to educate women on steps they should take to protect themselves against this disease. PTW’s breast cancer events included educational material in their outreach advertisements as well as in materials distributed during the events. If some of these “joint costs” are booked as program costs rather than overhead, the overhead ratio looks better.

GAAP standards on when it is proper to allocate some costs between fundraising and public education have gone back and forth. In the 1960s, in their initial set of standards, the National Health Council and the National Social Welfare Assembly (1964) forbade allocating any joint costs to programs unless an organization could show its primary purpose of that event was for program purposes. Otherwise, all the costs were considered fundraising. Later, in Statements of Position issued by the AICPA, the standard was loosened. Some joint costs were allocable to programs as long as the costs met certain criteria (American Institute of Certified Public Accountants (AICPA) 1987). One reason the PTW events reported “bad” overhead ratios is because PTW did not allocate any costs to programs. The events run by successor organizations and many competing organizations did allocate such costs.

To summarize, although organizations may not have wanted the bad optics of seeming to lobby for a right to spend unlimited amounts on overhead, quiet discussions of obscure accounting definitions have resulted in standards that are looser than donors may realize. The great majority of organizations report meeting the standards.

As discussed above, a key component of the nonprofit starvation cycle is unrealistic donor expectations about appropriate overhead spending. Much of the public concern over PTW arose from people applying to particular events their views of what normal fundraising levels should be and criticizing PTW for events that fell short.

Research indicates that people’s aversion to overhead can be reduced by additional disclosures, explanations, and more favorable framing. Saxton, Neely, and Guo (2014) find that providing additional, voluntary disclosures of financial and performance data positively affects donations. Several studies indicate that, when people are given explanatory disclosures, they are less likely to choose organizations based simply on financial ratios.²⁸Qu and Daniel (2021) find when the word overhead is avoided and donors are told an organization is spending on building organizational capacity, they are more likely to give. Donors who are more committed to a cause are more likely to overlook high overhead ratios (see Newman, Shniderman, Cain, and Sevel 2019). Further research on factors affecting overhead aversion could be very helpful to nonprofit managers and regulators.

Accounting is essential to the nonprofit sector because financial reports are basic to establishing donor trust. Accounting serves as a legitimizing institution. The U.S. would not have 1.6 million nonprofit organizations without the legitimacy that regular financial reporting conveys.

The inability of accounting to measure the true effectiveness of nonprofit organizations has caused major problems. Donors use ratios that do not reliably separate efficient and well-meaning organizations from inefficient or abusive ones. Organizations react to these donor preferences by underspending on needed administrative and fundraising efforts and, in many cases, by presenting misleading financial reports. Good organizations end up starved of funding for administration and unable to invest major amounts in fundraising efforts.

When teaching students about nonprofit accounting, I believe it is essential to stress donor concerns. Since the time of King Jehoash, donors have worried about diversion or misuse of contributions. State laws forbidding distribution of nonprofit assets, IRS rules on inurement and excessive compensation, auditing practices, and FASB rules for categorizing expenses by function, classifying net assets as either restricted or unrestricted by donors, and public reporting of financial results and officer salaries are all responsive to these donor concerns. Students should understand the purpose of the FASB rules and should also understand managers’ motivations to slant reporting to look more favorable to donors. Understanding managerial motivations and the gray areas in the accounting rules is essential for judging audit risk. Pedagogical research on how to communicate these concepts could be very useful.

Donors, regulators, and the press have historically relied unduly on simple ratios to distinguish good from bad organizations. However, new types of information are now coming out. For example, Charity Navigator now seeks to rate organizations on up to four main areas.²⁹ One is the traditional “accountability and finance” area. The other three are “impact & results,” “culture and community,” and “leadership and adaptability.” Candid (formerly GuideStar) has begun allowing organizations to earn bronze, silver, gold, or platinum ratings on their transparency. Organizations are supposed to disclose information about their strategy, plans, and metrics used to assess their progress.³⁰ Research is needed on whether donors understand and use these new metrics, whether the metrics accurately measure what they purport to measure, and whether organizations are gaming the measures.

Finally, there has been relatively little research on the lobbying and pressures on regulators and the FASB that have helped shape the nonprofit accounting and auditing environment. Rules are rarely neutral in their impact, and research could help us understand who they help and who is disadvantaged.