The world set two remittance-cost targets for 2030 — under 3%, nothing above 5%. Across the 103 economies the World Bank prices, 52 clear the goal and 24 remain above the ceiling — and what a country pays barely tracks how much it leans on the money

In 2015 the world's governments wrote two numbers into the Sustainable Development Goals. Target 10.c: by 2030, "reduce to less than 3 per cent the transaction costs of migrant remittances and eliminate remittance corridors with costs higher than 5 per cent." The 3% goal is also a standing G20 commitment. We pulled the World Bank's own scorecard for the target — the average cost of sending money *to* each country, indicator SI.RMT.COST.IB.ZS, latest value per economy (the series was last updated 2026-07-13). One hundred four economies carry a figure. One, Malawi, is recorded at −49.69% — a negative "cost" is a data artifact, so we set it aside and read the remaining 103. Ninety-two of those are 2023 values; eleven are older (2016–2021), a vintage mix we flag below. Against the 3% goal, 52 of the 103 economies clear it and 51 do not. Against the 5% corridor ceiling — the line the target said should no longer exist — 24 are still above it, close to one in four. Twenty-seven sit in the 3-to-5% band. The median receiving country is essentially at the goal, 2.92%; the unweighted mean across countries is 4.04%, pulled up by a heavy tail. That tail is where receiving money still costs most: Cuba (19.63%), Benin (17.35%, a 2020 reading), Angola (12.78%), Syria (10.71%, 2016), Sierra Leone (10.34%), Tajikistan (10.33%), Vanuatu (9.49%), Eritrea (9.16%) and Afghanistan (9.02%) head the list. Then we joined cost to dependence — remittances as a share of GDP (indicator BX.TRF.PWKR.DT.GD.ZS) — for the 102 economies present in both series. The Pearson correlation between the two is 0.05: effectively none. The most remittance-reliant economies do not, as a rule, pay the most to receive. Tajikistan leans hardest of all, remittances worth 57.7% of GDP, and pays 10.33% — but El Salvador (27.5% of GDP) receives at 2.00%, Honduras (30.1%) at 2.62%, Nepal (26.0%) at 2.53%, and Comoros (20.8%) at 1.74%. The expensive tail and the dependent tail are not the same countries.

The following is analysis, not fact. What the full 103-economy read adds, over the two or three high-cost countries a development desk usually names, is that the intuitive story is wrong. "The poorest and most remittance-dependent get gouged hardest" is the reflex; across the whole panel it barely holds — a 0.05 correlation is a scatter, not a line. Several of the cheapest corridors run into the most dependent economies, and several of the most expensive run into ones that lean on the money far less. The expensive tail clusters instead among small, fragile, isolated or sanctioned economies — Cuba, Syria, Afghanistan, Eritrea, South Sudan — and across Sub-Saharan Africa, which the World Bank separately reports as the costliest region to send to. ^mmodeled We did not test why: no market-concentration, sanctions, exchange-control or corridor-volume variable was joined to these figures, so the clustering is a pattern the data are consistent with, not a cause they isolate. ^sspeculative The ledger's honest verdict is split. On the 3% goal the world is roughly half-way there by count of countries, with four years left. On the 5% ceiling — the promise that no expensive corridor would survive — a quarter of receiving countries are still over the line on their *average* alone.

The strongest counter is that our country-average distribution flatters the record. The number a migrant actually faces is the World Bank's transaction-weighted global average of sending $200 — 6.39% in Q4 2023 and 6.49% in Q1 2025, more than double our unweighted country median of 2.92%. Because remittances move disproportionately through a few high-volume, high-cost corridors, weighting by dollars sent pushes the true cost far above the typical country. Sending to small states averaged 8.68%. By that measure the target is nowhere close. Our median describes the median *country*; it does not describe the median *dollar*. Two more caveats. The 5% target is defined at the corridor level, not the country average, so a country under 5% on average can still hide corridors above it — our count of 24 is a floor, not a ceiling. And eleven of the 103 readings predate 2023 (Syria's is 2016, Benin's 2020); the direction of the ledger is robust, but a handful of individual placements in the expensive tail could shift on a fresh reading.