Aid, Migration, and the Money That Goes Home

This page reports numbers and the disagreements between them. It does not recommend an immigration policy or an aid policy, and where the research genuinely conflicts, both sides are shown and the reader is handed the switch.

One. What America sends

Foreign aid is about one percent of the federal budget. Ask Americans and most guess many times higher.

In 2026 it also collapsed. Congress appropriated roughly $50 billion, itself 16 percent below 2025. But USAID had been shut down, nearly all ten thousand staff terminated along with as many as 280,000 contractors, and the surviving programs handed to the State Department. As of 1 July, actual spending for the year stood just under $7 billion.

Two. What migrants send

Money moving in the same direction, from the same country, to the same places, on a completely different scale.

Two of those rows measure money sent from the US specifically. Two measure everything arriving from everywhere. That is a real inconsistency in the sources rather than a rounding choice.

It barely moves the comparison, because roughly nine in ten remittance dollars reaching Guatemala and Honduras come from the United States. But it is worth knowing which row is which.

Three. What it costs, and earns, at home

Here the evidence is genuinely contested, and the argument turns on an accounting choice most summaries never mention.

The National Academies found that whether immigration helps or hurts the budget depends on how you charge people for things like national defense and interest on the debt.

If each immigrant is charged an average share of those costs, immigration looks fiscally negative in both the short and long run. If they are charged the marginal cost, which is close to nothing, because another resident does not require another aircraft carrier, the long-run effect is positive. Same data, same researchers, opposite answer.

Four. Does the aid actually reduce the migration?

This is the reason usually given for the spending, so it deserves a real answer rather than a shrug.

The answer is that for the poorest countries the evidence points the other way, and for the rest the aid is nowhere near large enough to matter.

Emigration does not fall as a poor country develops. It rises, and only begins to fall once the country crosses roughly seven to eight thousand dollars of income per head, a pattern known as the mobility transition. The mechanism is unglamorous: leaving is expensive, and people too poor to leave do not. Development relieves the constraint on going.

Honduras, at $6,586, sits below the turn. Aid that succeeded in raising Honduran incomes would, on the best available evidence, raise emigration first. Not as a paradox but as the base case. El Salvador at $11,669 and Guatemala at $12,641 sit above it, so there the direction is at least favorable.

Mexico is the confirming case and it cuts both ways. It crossed the threshold long ago, and the net flow to the United States fell to zero and then reversed, with more Mexicans returning than departing. That is exactly what the transition predicts. It also took decades and was driven by trade, industrialization and falling fertility. American aid was a rounding error in that story.

None of which says aid is unjustified. It may be defensible on humanitarian or strategic grounds, and this page takes no position on that. It says the migration-control argument is the weakest of the available arguments for it, and for the poorest countries it probably runs backwards.

Five. The lever that is actually large

If the object is more money reaching households in those countries, the aid budget is not where the leverage is. Sending remittances costs 6.36 percent globally as of the third quarter of 2025 and just under 5 percent on the United States to Mexico corridor, against a United Nations target of 3 percent. Newer payment rails already move that corridor for under one percent.

Closing that gap to the 3 percent target would put roughly $20 billion a year back into families abroad. That is about thirty times the Central America root-causes budget, and roughly three times everything America disbursed in aid this year.

And it is a question about payments rules, not about foreign assistance.

What it all adds up to

Read together, the five panels say something the separate arguments miss.

American aid is small, has just been cut hard, and is usually defended as a way of reducing migration. Migration from those same countries produces a private flow home roughly eighteen times larger, spent by families rather than administered by agencies, and it rises when the receiving country is in trouble rather than falling.

And the effect of that migration inside America is negative or positive depending on an accounting rule, a generation, and a time horizon. Which is why two honest people can read the same report and quote opposite conclusions.

None of that settles what anyone should do. It does suggest the three questions are really one question, and that answering any of them while ignoring the other two produces a number that cannot be right.

Sources and limits

Aid: congressional appropriations for 2026 and disbursement tracking through 1 July. Remittances: World Bank and KNOMAD data, plus central bank reporting for Guatemala and Honduras. Migration: the DHS Office of Homeland Security Statistics, where Mexico remains the leading country of birth. Fiscal estimates: the Congressional Budget Office, 2024, and the National Academies of Sciences, 2016.

Four limits, and they matter.

Remittance data captures formal transfers and misses cash carried by hand, so the real flow is larger than shown. The corridor and total-inflow rows are not perfectly comparable, as noted above.

Fiscal estimates rest on assumptions that advocacy groups on both sides pick to suit their conclusion, which is why only the CBO and the National Academies appear here, and why the assumption that flips the answer is a switch rather than a footnote. And aid, migration and remittances all push on each other in ways no table settles.