RECIPES / ENDOWMENT DRAW-RATE TRACKER

How much of the endowment is being spent?

5 years of federal finance filings, turned into an estimated endowment spending rate for private nonprofit colleges. This is an estimate from federal filings. A high draw rate is a reason to look closer, not proof a school is in trouble.

IPEDS Finance F2Part HFY2020–FY2024
Latest median4.4%FY2024, provisional
Above 7%16.4%211 of 1,287 eligible schools
School-year rows6,5801,357 schools across five years
Latest statusProvisionalFY2024 will be updated when the final release is published
§ Sector view

An independent estimate from public federal data

ipeds-endowment-283292ca75a4abda
Generated 2026-08-10

Each year covers schools that were private nonprofits and reported Finance Part H. The denominator includes rows with a positive beginning value and reported inputs. Rows must also pass the component identity check (the beginning value plus the reported parts equals the year-end value). Rates use the absolute reported spending amount. This sign-normalizes the mixed positive and negative reporting convention in FY2020–21.

Fig. 1 · Reported draw-rate distribution
LINE P10–P90 · BAR P25–P75 · DOT MEDIAN
0%5%7%10%15%FY2020FY2020: 10th percentile 0.0%, 90th percentile 7.0%FY2020: middle half 1.9% to 5.0%FY2020 median: 4.0%FY2021FY2021: 10th percentile 0.0%, 90th percentile 6.9%FY2021: middle half 1.8% to 5.2%FY2021 median: 4.1%FY2022FY2022: 10th percentile 0.0%, 90th percentile 6.1%FY2022: middle half 1.5% to 4.4%FY2022 median: 3.5%FY2023FY2023: 10th percentile 0.0%, 90th percentile 8.1%FY2023: middle half 2.2% to 5.5%FY2023 median: 4.3%FY2024FY2024: 10th percentile 0.0%, 90th percentile 9.6%FY2024: middle half 2.3% to 5.6%FY2024 median: 4.4%Spending distribution ÷ beginning-of-year endowment value →

Reference lines mark 5% and 7%. The scale stops at 15% to keep the main distribution readable. The table below includes every eligible school, including rates beyond the chart.

Many things put a school on these lists besides financial stress: board-approved spending of unrestricted quasi-endowment funds, drawing down a completed capital campaign, deploying a large one-time gift, or ordinary volatility in a small endowment, where a single transfer can swing the rate by whole percentage points. Appearing here is not evidence of fiscal irresponsibility. Rates come from each school's own federal IPEDS filing. Some states' UPMIFA statutes presume imprudence only above a 7% rate measured against a multi-year average value — a different measure than the single-year rate shown here.

Fiscal yearEligible / reportersMedianAbove 5%Above 7%Above 15%Release
FY20201,335 / 1,3364.0%final
FY20211,331 / 1,3334.1%final
FY20221,316 / 1,3173.5%final
FY20231,303 / 1,3044.3%final
FY20241,287 / 1,2904.4%provisional
§ Pick a school

Compare a school to the sector

1,357 schools

No school is preselected or highlighted, so the sector-wide view stays up until you choose one.

§ What the threshold means

How to read the 7% line

Typical endowment payout policies often target roughly 4–5%. Some states' versions of UPMIFA include a presumption above 7%. That test uses a multi-year average value and applies only in a minority of states. This recipe divides one year's reported spending distribution by that year's beginning value. Crossing 7% here does not establish a UPMIFA violation.

Coverage note

263 schools in this dataset have no current public school page. This can include colleges that later closed or left the current directory. Their histories still appear here.

§ Pull the fields yourself
curl 'https://api.collegedata.fyi/rest/v1/ipeds_facts?ipeds_id=eq.152080&field_key=in.(endowment_value_begin,endowment_value_end,endowment_new_gifts,endowment_investment_return,endowment_spending_distribution,endowment_other_change)&data_year=gte.2020&data_year=lte.2024&select=ipeds_id,data_year,field_key,value_numeric,quality_flag,source_table,source_variable,release_type&order=data_year.asc,field_key.asc' \
  -H 'apikey: <anon key>' \
  -H 'Authorization: Bearer <anon key>'

Then compute abs(endowment_spending_distribution) / endowment_value_beginafter retaining reported rows with a positive beginning value and verifying the Part H component identity.