Urban Institute's new survey: 20% of the smallest charities saw a funding disruption in 2025, versus 55% of the largest. Losses ran the other way.

If your organization's annual expenses are under $100,000, you were the least likely kind of nonprofit to have government funding disrupted in 2025. Among the small organizations that did lose government money, the typical loss was large enough to threaten the place.
That is the pattern in a new Urban Institute report, published October 14 alongside a companion workforce report. Both draw on the National Survey of Nonprofit Trends and Impacts, which reached 2,737 organizations in 2025 and 2,328 in early 2026. The core sample is 501(c)(3) public charities with at least $50,000 in annual expenses and revenue, and it excludes many organization types, among them hospitals, schools, day cares, houses of worship, grantmakers, and membership associations. What is left looks like the service-providing nonprofit most readers here run.
Thirty-five percent of nonprofits reported at least one government funding disruption during 2025, or 49 percent of those receiving any government funding. Split by budget, the ladder is close to straight: 20 percent of organizations with $50,000 to $99,999 in annual expenses were disrupted, rising to 29 percent at $100,000 to $499,999, 44 percent at $500,000 to $999,999, 49 percent at $1 million to $9.9 million, and 55 percent at $10 million or more.
Read that as exposure rather than fragility, which is how the report reads it: the higher rate among larger organizations "does not necessarily mean they were less resilient; but they were more exposed." Bigger nonprofits hold more government money, in more separate pieces. A disruption counts as any of three things during calendar 2025: losing a grant, contract, loan, or other government assistance, whether canceled outright or with committed funds pulled back; a delay, pause, or freeze; or a stop-work notice.
Among organizations that did lose government funding, the median loss was 9 percent of total revenue. By size, the medians run 28 percent at $50,000 to $99,999, then 14, 12, and 5 percent through the middle bands, before ticking back up to 9 percent at $10 million or more. Not a clean line: the largest organizations were not the least affected proportionally. But the smallest band stands well apart from everything above it.
The tail matters more than the median. Among the smallest organizations that lost government funding, 29 percent lost half or more of total revenue and 56 percent lost at least 20 percent. Across all sizes, 17 percent lost half or more.
Cancellation was not the whole story. Delays, pauses, and freezes were common, and long-running. Among nonprofits reporting one to six delays, 48 percent had at least one that ran 12 weeks or longer. One caveat if you run a small shop: Urban excluded the $50,000 to $99,999 band, because too few answered, and organizations with seven or more delays.
Eleven percent of nonprofits that had a delay still had an unresolved one in early 2026, about 3 percent of the sector carrying an open question into this year. Of those, 76 percent said at least one had already run 12 weeks or longer.
A 12-week gap on a reimbursement contract is a cash-flow event whether or not the grant survives. You front payroll, keep serving people, find out later.
Among disrupted nonprofits, 48 percent decreased overall expenses, 47 percent drew on reserves, 70 percent increased charitable and philanthropic fundraising, and 17 percent borrowed or increased a line of credit. Twelve percent increased fundraising from government sources, against 8 percent sector-wide.
The report is careful here, and so should we be. These are actions taken at any point in 2025, not actions caused by a disruption, and undisrupted nonprofits did much the same: 59 percent also increased charitable fundraising, 37 percent also drew on reserves. Among disrupted organizations taking each action, 63 to 65 percent attributed it to the disruption.
Urban flags the limits of that toolkit. Everyone raising emergency money at once can intensify competition for the same dollars, and organizations with strong fundraising capacity, which tend to be larger, may have better access. Borrowing creates a repayment obligation. Reserves spend down what you were keeping for the next shock, which is why how many months you hold stops being abstract.
The workforce report traces where the money came out. Among staffed nonprofits that were disrupted, 30 percent cut headcount against 20 percent of staffed nonprofits overall. Disrupted organizations reduced programs at 25 percent against a sector rate of 16 percent, and people served at 20 percent against 11 percent. Hiring plans have not recovered either, sliding from 57 percent of staffed nonprofits expecting to hire within a year in late 2024 to 38 percent by spring 2026.
Name your concentration before anyone else does. Pull revenue by source and find your largest single government grant or contract as a share of the total. If that line is 20 percent or more, you are in the territory where a loss forces real decisions, and more than half of the smallest organizations that lost government funding landed at or above it.
Then separate the two problems, because the fixes differ. Against a delay you need cash: a line of credit arranged while your finances look fine, a reserve you can reach, a funder you have already told. Against a loss you need diversification, which takes quarters rather than weeks. Our reporting on how far federal grants fell and the continuing resolution and grant timing covers the environment those plans must survive.
One caution on the numbers. These are self-reports from executive directors, weighted for national representativeness, not agency records, though the size bands come from 990 data. Urban puts the margin of error on prevalence estimates at about plus or minus 2 percentage points, which does not extend to the medians or thin subgroups, and completion rates were 6.1 and 4.9 percent. The report concedes the hardest-hit organizations may have closed before being surveyed, and the workforce frame, commissioned by Independent Sector, does include some hospitals and universities.
Size changed the shape of the risk more than the amount of it. If you are large, you were likely to get hit somewhere and able to absorb it. If you are small, you probably escaped, and among the minority who lost government money the median loss was 28 percent of total revenue. Find your largest government line as a share of revenue this week. That number tells you which problem you have.
One email when new briefings publish. No noise, unsubscribe anytime.