The Overhead Percentage That Funds an Entire Research University
A federal research grant pays for the project and a negotiated additional percentage for the buildings, utilities, and administration behind it. That percentage is one of the most consequential numbers in higher education.
Two Kinds of Cost
A research grant funds direct costs: the salaries of the researchers, the equipment, the materials, the participant payments. Those are attributable to a specific project.
The project also consumes things that cannot be attributed to it individually. The building it occupies, the electricity and specialised ventilation, the animal care facility, the library, the research compliance office reviewing protocols, the grants administration staff, and the financial systems producing the reports the funder requires.
Those are indirect costs, formally called facilities and administrative costs, and they are recovered as a percentage applied to the direct cost base.
How the Rate Is Set
The rate is not chosen by the university. It is negotiated with a cognisant federal agency through a periodic exercise in which the institution submits a detailed proposal allocating its actual costs between the research function and everything else.
The facilities component covers depreciation on buildings and equipment, operations and maintenance, interest on debt for research facilities, and library costs. The administrative component covers departmental and central administration and compliance.
Crucially, the administrative portion has been capped at twenty six percent since the early 1990s, regardless of what an institution actually spends. Only the facilities component varies with the negotiation.
| Component | Basis | Constraint |
|---|---|---|
| Facilities | Buildings, utilities, maintenance, interest | Negotiated from actual costs |
| Administration | Departmental and central administration | Capped at 26 percent |
Because the administrative half is capped and the facilities half is not, institutions with expensive laboratory buildings negotiate high rates and institutions with heavy compliance burdens and modest buildings do not. The rate measures physical plant more than it measures overhead.
Why the Rates Vary So Much
Published rates range from under forty percent to over seventy. The variation is driven principally by the facilities component, which reflects the cost of the buildings the research occupies.
An institution running wet laboratories with intensive ventilation, containment facilities, and animal care has a genuinely higher cost per research dollar than one conducting mostly computational or social science work.
Location matters too, since construction and energy costs differ substantially by region.
The result is that comparing headline rates across institutions is close to meaningless without knowing what kind of research each conducts.
The Recurring Argument About Capping It
Proposals to cap indirect cost rates at a uniform lower percentage have been made repeatedly, and the arguments on each side are worth stating properly.
The case for a cap is that the rates are opaque, negotiated bilaterally, vary in ways that are difficult to justify, and consume a substantial share of a research budget that could otherwise fund more projects. Private foundations frequently pay ten to fifteen percent and research still happens.
The case against is that the costs are real and somebody pays them. If the federal government pays less, the institution funds the difference from tuition, endowment, state appropriation, or clinical revenue, all of which are cross subsidies from other activities. Institutions without those resources would reduce research.
The foundation comparison is weaker than it appears, because foundation funded research is generally conducted in facilities the federal government helped pay for, which means low foundation rates are subsidised by federal rates rather than demonstrating that the costs are unnecessary.
The Cross Subsidy Nobody Advertises
The uncomfortable finding in analyses of research finance is that even at current rates, most research universities do not fully recover the cost of conducting research.
Studies of the gap have found institutions contributing substantial amounts of their own funds per research dollar, sourced from tuition, endowment income, clinical operations, and state support.
That means research at most institutions is a cost centre subsidised by other activities, undertaken for reasons of mission, prestige, and faculty recruitment rather than because it pays.
Understanding that inverts the usual framing. Indirect cost recovery is not a profit margin on research, it is a partial reimbursement of costs already incurred.
What It Means Practically
For anyone working with grant budgets, several implications follow.
The rate applies to a defined base, generally modified total direct costs, which excludes equipment above a threshold, tuition remission, and the portion of subawards above a limit. That means the effective overhead on a grant is lower than the headline rate.
Different rates apply to different activities, with organised research, instruction, and off campus work each carrying separate rates.
And a funder that pays a lower rate is not saving the institution money, it is shifting the cost onto the institution, which is why some universities decline funding from sources that cap recovery too aggressively.
The Bottom Line
Indirect cost rates recover the buildings, utilities, and compliance infrastructure that make research possible and that no grant pays for directly. They vary because laboratory buildings cost more than offices, and the administrative half has been capped for three decades regardless of what compliance actually costs. Capping the rest would not eliminate the costs, it would move them onto tuition and endowments, which is the substance of an argument that recurs every few years and is usually conducted as though the money were optional.