We have built a quiet preference for the proposal with the lowest overhead. It looks like good stewardship. A leaner budget seems to send more of every dollar to the people we serve, and on the page it does. But the lean number hides a second payment we make later, and that second payment is usually larger than the first.
The costs we squeeze are not luxuries. They are the things that keep an organization upright between grants: the finance officer who can close the books cleanly, the system that tracks a payment so it does not have to be tracked twice, the training that means a control is followed rather than just written down. We call these indirect costs, and the word itself does some of the damage. It makes them sound optional, like a tax on the real work, rather than the structure that lets the real work happen at all.
What we are actually buying when we underfund them
When we pick the partner with the thinnest overhead, we are not buying efficiency. We are buying a deferral. The capacity that was not funded does not disappear. It is borrowed from somewhere. A program officer stays past midnight covering a finance gap. A reconciliation slips because no one had the hours to do it on time. A small local organization wins the grant by promising to run on almost nothing, then spends the year quietly cannibalizing the systems that would have made it auditable and durable. Two years on, we look at that same organization, see weak controls and thin records, and conclude it was not ready for direct funding. We read the symptom of our own underfunding as proof that the underfunding was wise.
We starve the function that keeps an organization standing, then point to the wobble as a reason not to feed it.
This is not a story about anyone acting in bad faith. The funder who favors low overhead is trying to be a careful guardian of public or private money, and that instinct is honest. The implementer who promises to run lean is responding rationally to what wins the bid. The incentive is doing exactly its job. The trouble is the job we set it. We designed a competition that rewards looking cheap over being sound, and a competition can be redesigned.
The overhead myth costs the frontline most
The deepest cost lands where we can least afford it. An organization with no funded slack has no room to absorb a shock, so the first hard week breaks something. It has no margin to invest in the people who carry the work, so it loses them and pays the recruiting and retraining bill again. It cannot build the back office that would let it take money directly, so it stays a permanent subcontractor, dependent on intermediaries who take their own slice. The pursuit of a low headline rate produces a sector that is fragile, dependent, and more expensive in the round. We optimized the visible number and lost the thing the number was supposed to protect.
Building true cost into the deal
The fix is not to stop caring about value. It is to change what value means on the page, and this is work funders and implementers can do at the same table, because neither benefits from a partner that quietly hollows itself out.
Fund the true cost of delivery, including the indirect functions, as a deliberate line rather than a residual that gets trimmed when the budget is tight. An organization that can keep clean books and retain its staff is a better steward of the next grant, not a worse one.
Stop treating a low overhead rate as a mark of virtue in the scoring. A rate that is implausibly low is a warning, not a win. It usually means a cost has been hidden, not removed, and it will surface later as risk we both have to manage.
Invest in the systems that make a partner auditable, rather than withholding that investment and then citing the resulting weakness as a reason to keep control upstream. Shared, recognized due diligence and funded financial systems turn a fragile partner into a durable one. That is cheaper for everyone over a few years than rebuilding capacity from scratch each time a grant ends.
None of this lowers the bar on accountability. It raises it, because an organization with funded systems can actually meet the standard we set, rather than performing compliance on hours it does not have.
The test is simple, and we can apply it to our own choices before we apply it to anyone else’s. The cheapest partner on paper is rarely the cheapest partner in the end. If a budget looks lean because the structure that holds it up was left unfunded, we have not saved the money. We have only agreed to pay for it twice, and the second bill tends to arrive at the worst possible moment, charged to the people we came to serve.