Money across borders: what to settle in the agreement before funds move

Rules on eligible costs, overhead rates, tax treatment and foreign receipts differ by institution, funder and jurisdiction, and they change. Treat this as a checklist of questions to ask your own research office and your partner's — not as guidance on any specific rule.

A collaborative grant agreement will describe the scientific work in several pages and the financial arrangements in a paragraph. Then the project starts, and almost every delay in the first year turns out to be financial or administrative rather than scientific.

The first question, asked early

Before anything else: can your partner's institution receive funds from your funder at all, and in what form?

This sounds like a formality. It is not. Institutions differ in whether they can accept foreign research funds directly, whether the receipt requires a specific type of agreement, how long approval takes internally, and what the money can then be spent on.

The answer determines the structure of everything else, and it takes one email to the partner's research office to find out. Ask it before the proposal is written, not after it is funded.

Three ways money usually moves

Direct sub-award — your institution transfers funds to theirs under a subcontract. Cleanest when it is possible, and it makes the partner a formal party with their own budget and reporting duties.

Payment in kind — you buy the equipment, consumables or services and ship or pay directly, rather than transferring cash. Common where direct transfer is difficult, and it works, but be careful: an arrangement where the partner never controls any budget line puts them in a weaker position on every subsequent decision.

Reimbursement — the partner spends and then claims back. This shifts cash-flow risk onto the side least able to carry it, which is worth saying out loud. If reimbursement is the only route, an advance for the first period solves most of the problem.

Four points where assumptions differ

Overhead. The rate and what it covers differ substantially between institutions and countries. Write the rate into the agreement in figures, not as "standard institutional rate", which means two different things to the two signatories.

Eligible costs. Funders differ on what a grant may pay for — salary top-ups, participant compensation, equipment above a threshold, travel class, publication fees. An expense that is routine for one side may be unallowable for the other, and this is best checked line by line before the budget is fixed.

Personnel costs. How researchers' time is charged differs: some systems charge a share of an existing salary, others pay supplements for project work, others cannot pay individuals from the grant at all. Neither side should assume the other's model.

Participant compensation and field expenses. Cash payments in field research often cannot be documented the way a Western finance office expects. Agree the documentation standard in advance — a locally appropriate record that your finance office has accepted in writing beforehand — rather than discovering the mismatch at the first report.

Timing, currency and fees

Timing. Write down when each tranche moves and what triggers it. The most common failure is a first tranche that arrives months after work is supposed to start, because internal approval on one side takes longer than anyone allowed for.

Currency. State which currency the budget is denominated in and who carries exchange-rate movement. Over a three-year project this is not a trivial amount, and leaving it unstated means it is carried by whoever is weaker in the negotiation, by default.

Fees. Bank charges on international transfers are taken from the transfer. If the agreement says an amount, say whether it is before or after fees — otherwise the partner receives less than the budget line and has to explain a shortfall they did not cause.

Tax. Ask both research offices whether any withholding applies. This is not a question researchers can answer, and it can change the net amount materially.

Reporting

Agree three things:

What financial reports are required, in what format, and when.

What supporting documents must be kept, and by whom — and for how long after the project ends.

Who prepares them. If the answer is "the partner", make sure the partner's budget includes time for that work. Financial reporting on an international grant is a real administrative load, and assigning it without resourcing it is a common way to build resentment into a collaboration.

Clauses worth writing down

Four, all short, all saving a difficult conversation later:

What happens to unspent funds at the end of a period, and whether they can be carried forward.

What happens if the project ends early — who keeps equipment purchased with grant funds, and what happens to committed staff costs.

Who may reallocate between budget lines, up to what amount, and with whose approval.

Who pays publication charges, particularly open-access fees, which can be a significant line and are often assumed by each side to be the other's responsibility.

One practice that prevents most of this

Put the two research offices in contact with each other early, and let them talk directly rather than relaying through the researchers.

They speak a shared professional language, they know the questions that matter, and they will surface constraints in one exchange that would take the researchers three months to discover. It is the highest-value hour in the whole setup phase, and it is routinely skipped because nobody thinks to arrange it.