A zero on the fundraising line is not the whole story
A 2004 study counted both fundraising and event costs. Current returns let us separate an explicit zero from a missing amount, then look for event spending elsewhere in the filing.
By the CharitySense Data team
Published Updated
01/What the original 37%
What the original 37% measured
In 2004, researchers at the Urban Institute and Indiana University found that 37% of full Form 990 filers receiving at least $50,000 in direct private contributions reported no fundraising or special-event expense. Their tax-year-2000 group held 74,360 organizations. The finding helped put a basic question on the table: what can an overhead ratio tell us if the costs behind it are reported in different ways?
The study offered several plausible explanations for a zero, including volunteer fundraising and support from a related organization. It also pointed to the difficulty of allocating staff time across functions. The separate finding that nearly 13% reported no management-and-general expense concerned a broader group of operating public charities. We do not combine the two percentages here.
Our first question was narrower. When we open current electronic returns, can we tell a reported zero from a missing amount? And can an event cost appear elsewhere on the same return?
02/current cohort with
A current cohort, with the comparison limits in view
We read the latest available full Form 990 for each of 485,109 organizations, then focused on the 277,653 whose latest full return is for tax year 2024. To approximate the study's private-contribution threshold, we selected the 138,765 reporting at least $50,000 in current-year "other contributions." This modern line excludes amounts separately identified as government grants, federated campaigns, and support from related organizations. It is a useful proxy for direct private support, but it is not an exact match to the line on the 2000 form.
The counts are 57,448 explicit zeros, 1,921 missing amounts, and 79,377 positive amounts. Nineteen returns reported a negative value, which we kept distinct rather than forcing into a zero category. A zero here describes the return's fundraising-expense column. It says nothing by itself about every way an organization raised support.
The 37% from 2000 and the 41.4% from 2024 are different measurements. The earlier study counted the absence of both fundraising and special-event costs; our headline identifies an explicit zero on one modern expense line. Form design and the contribution threshold also changed. We would mislead readers by drawing a trend line between them.
03/zeros become less common
The zeros become less common as contributions rise
The explicit-zero share falls across the four contribution ranges, from 56.6% among filers with $50,000 to $250,000 in other contributions to 15.5% among filers with at least $10 million. The largest group still contains 742 explicit zeros out of 4,786 organizations. Size changes the frequency of the reporting pattern; it does not explain any particular return.
| band | organizations | explicit Zero | missing | positive Event Expense | zero Pct |
|---|---|---|---|---|---|
| $50k to $250k | 54,919 | 31,091 | 1,133 | 5,424 | 56.6 |
| $250k to $1m | 48,298 | 18,664 | 521 | 2,468 | 38.6 |
| $1m to $10m | 30,762 | 6,951 | 243 | 823 | 22.6 |
| $10m or more | 4,786 | 742 | 24 | 49 | 15.5 |
Select a bar to see the number of returns behind it and the event-expense cross-check.
Counts behind each range
- $50k to $250k in other contributions: 31,091 of 54,919 latest full Form 990 filers for tax year 2024 (56.6%) reported an explicit zero in fundraising expense; 1,133 had no amount on that line; and 5,424 of the explicit-zero filers reported positive direct event expenses on Schedule G.
- $250k to $1m in other contributions: 18,664 of 48,298 latest full Form 990 filers for tax year 2024 (38.6%) reported an explicit zero in fundraising expense; 521 had no amount on that line; and 2,468 of the explicit-zero filers reported positive direct event expenses on Schedule G.
- $1m to $10m in other contributions: 6,951 of 30,762 latest full Form 990 filers for tax year 2024 (22.6%) reported an explicit zero in fundraising expense; 243 had no amount on that line; and 823 of the explicit-zero filers reported positive direct event expenses on Schedule G.
- $10m or more in other contributions: 742 of 4,786 latest full Form 990 filers for tax year 2024 (15.5%) reported an explicit zero in fundraising expense; 24 had no amount on that line; and 49 of the explicit-zero filers reported positive direct event expenses on Schedule G.
04/event expense can sit
An event expense can sit beside a fundraising zero
The current Form 990 keeps direct fundraising-event expenses out of its functional fundraising-expense column. Those costs belong in the event section of the revenue statement. The supplemental fundraising schedule gives more detail for organizations crossing its event-reporting threshold. We found 8,764 organizations with an explicit zero in the fundraising column and a positive direct event expense on that schedule. That is 15.3% of the explicit-zero group.
This overlap is a floor. A missing supplemental schedule does not establish that an organization had no event costs. Nor can we infer from an event expense how the rest of its fundraising work was staffed or allocated. It does show why a zero on one line should be read with the rest of the return, precisely the distinction the older study had to make.
05/Other support can change
Other support can change the reading too
Among the 57,448 explicit-zero filers, 2,380 also reported positive contributions from related organizations. Another 1,177 reported positive federated-campaign contributions. Those groups can overlap, and neither figure tells us who bore the cost of soliciting a gift. They make the research question more interesting: a filing can show substantial incoming support without describing a conventional in-house appeal.
The same caution applies to the broader "contributions and grants" total. If we used that total for the $50,000 threshold, the 2024 cohort would grow to 184,145 and the explicit-zero share would be 49.8%. That is a different population because the total includes government grants and other separately reported sources. A seemingly small choice of line changes the headline by 8.4 percentage points.
For another view of why functional expense labels can be hard to compare, see our overhead analysis. The related accounting blind-spot essay looks at what a different kind of total can leave out.
Behind the Data
We reviewed every organization's latest available full Form 990 in the current archive on September 28, 2026, and used only tax-year-2024 latest returns for the article's modern percentages. We selected filers with at least $50,000 in current-year other contributions, with the lower bound included. Each organization appears once. The four chart ranges use that same contribution amount. Counts are organizations, not filings across multiple years.
We read the return's total fundraising expense in its functional-expense statement. A numeric zero, an absent amount, a positive amount, and a negative amount remain distinct. For the event cross-check, we looked for a positive direct expense on the supplemental fundraising-event schedule among explicit-zero filers. That schedule is required only when the applicable event receipts and contributions exceed the IRS threshold, so its absence cannot be interpreted as no event cost. The related-organization and federated-campaign counts are reported contribution components, not evidence of who performed fundraising.
The 2004 study used tax-year-2000 returns, an older form design, and a direct-private-support threshold. Its 37% combined two expense categories. Our modern private-support proxy and explicit-zero measure therefore cannot reproduce the historical statistic or measure change over time. Our archive also excludes organizations that are not required to file and latest returns for other tax years from the 2024 analysis. We describe what these returns report, not what any charity actually spent outside them. For a broader account of the archive, see our methodology.
Sources
- CharitySense Data analysis of 485,109 latest available full Form 990s, including 277,653 latest tax-year-2024 returns. Read September 28, 2026.
- Urban Institute and Indiana University, What We Know about Overhead Costs in the Nonprofit Sector (2004; tax-year-2000 analysis).
- Internal Revenue Service, 2024 Form 990 instructions, especially the functional fundraising column and event-expense reporting.
- Internal Revenue Service, Schedule G instructions, including the event-reporting threshold.
Ask the research team
What would you want us to pull from the filings? Tell us, and a future piece may answer it.
Final Thoughts
The historical 37% remains a useful prompt to inspect how nonprofits report their costs. The clearest current finding is more specific: an explicit zero is common, a missing amount is much rarer, and thousands of the zero filers also report direct event expense. The form makes those facts visible in different places. Reading them together gives us a better question for any individual organization than an overhead percentage alone can provide.
About the author
The CharitySense Data team
Research team
We read public nonprofit filings and other sources to understand how the sector works. This blog shares what we find, along with the limits of the evidence. Each article explains how its figures were measured and cites the records or primary sources behind them.