The cost of sending money home varies by a factor of more than twenty depending on which two countries are involved. Sending US$200 from the UAE to India costs about 1.5%. Moving money between some African neighbours costs over 30%. Neither figure has much to do with distance, and the reason is worth understanding, because it also tells you what would fix it.
This is a companion to the guide to global remittance flows, which covers the volumes.
The short answer
Remittance costs are set by market structure, not geography. Where volumes are high and many providers compete on digital rails, prices fall below 3%. Where volumes are thin, currencies are many and compliance costs are spread over few transactions, prices reach double figures. The cheapest and most expensive corridors on earth can be similar distances apart.
The spread
The chart uses a single basis — the cost of sending US$200 — because remittance pricing is highly non-linear and comparisons across different transfer sizes are meaningless. Percentage costs almost always fall as the amount rises, since a large part of the fee is fixed.
At the cheap end sit the Gulf-to-South-Asia corridors: UAE to India at roughly 1.5%, Saudi Arabia to Pakistan at roughly 2.1%. Both already clear the UN's SDG 10.c target of below 3% by 2030, without subsidy.
At the expensive end sits Sub-Saharan Africa, averaging about 8.78% as a destination region — roughly triple the target and roughly four times what the best corridors charge.
Regional averages, though, hide the real extremes. The costliest routes are within Africa rather than into it. The Tanzania to Uganda corridor has been recorded among the world's most expensive, with a US$500 transfer averaging 33.58% — about US$167.90 to move US$500. (That figure is on a US$500 basis, not the US$200 basis used in the chart; on the smaller amount the percentage would typically be higher still.) South Africa to Botswana, Tanzania to Kenya, Angola to Namibia and South Africa to Angola complete the region's most expensive set.
For scale: in 2021, sending US$200 from Tanzania cost about 27%, and from South Africa about 14.7%. Those are older figures on a different vintage, and costs have edged down since — but they establish the order of magnitude that intra-African senders have been paying.
Why the gap exists
Four structural factors explain almost all of it, and none is distance.
Volume
Remittance provision has high fixed costs — licensing, compliance, settlement infrastructure, agent networks — and low marginal costs per transaction. A corridor moving billions of dollars a year spreads those fixed costs across an enormous base. A corridor moving a few hundred million cannot. This alone accounts for much of the gap between Gulf-to-South-Asia and intra-African routes.
Competition
Where several providers compete, margins compress. Where one or two operators hold an effective monopoly — often because exclusivity agreements tie up the agent network on the receiving side — they price accordingly. Exclusivity arrangements have historically been a significant driver of African corridor costs, and unwinding them tends to move prices faster than anything else.
Currency and settlement
A dollar corridor into a large economy settles through deep, liquid markets. An intra-African transfer may require converting into and out of two thinly traded currencies, frequently routing through a third currency to do it. Each conversion carries a spread, and those spreads are often larger than the visible fee — which is why headline fee comparisons understate what senders actually pay.
Compliance costs that fall the wrong way
Anti-money-laundering and counter-terrorist-financing obligations are broadly fixed per provider and per corridor. Applied uniformly, they weigh most heavily on small operators serving thin routes — precisely the corridors that most need new entrants. The result is a de-risking pattern where banks withdraw correspondent relationships from small markets, leaving fewer and costlier channels behind.
What high costs actually do
The obvious cost is the money taken. It is larger than it looks: at 8.78% rather than 3%, a household receiving US$3,000 a year loses roughly US$173 annually to the difference. For a family close to subsistence, that is not a rounding error.
The second effect is the one that shows up in national statistics. High formal costs push flow into informal channels. Where a licensed transfer costs 20% and an informal operator charges a fraction of that, informality is not a preference; it is arithmetic. This is why the corridors with the worst measured costs are also the corridors whose recorded remittance totals understate reality by the most — the measurement problem and the pricing problem are the same problem.
That connection matters for policy design. A government that wants better remittance data and a government that wants cheaper transfers are pursuing the same objective through the same lever.
What brings costs down
The evidence from the corridors that already beat the target is reasonably clear, and it is not exhortation.
- Interoperable digital rails. Mobile money and instant payment systems that talk to each other across borders remove agent-network economics from the equation entirely. This is where the fastest reductions have come.
- Ending exclusivity. Agreements that lock a receiving-side network to a single operator are a straightforward competition problem with a straightforward remedy.
- Proportionate compliance. Risk-based rather than blanket obligations, so that a small operator moving small sums between neighbouring countries is not carrying the same fixed burden as a global bank.
- Price transparency. Senders can only choose the cheaper option if the total cost, including the exchange-rate margin, is visible before they commit. Fee-only disclosure systematically hides the larger half of the charge.
None of this requires subsidy. The corridors already below 3% got there through volume and competition, which is the strongest evidence available that the target is achievable where the structure allows it.
What to watch
- Whether Africa's digital corridors compress prices the way South Asia's did. This is the single biggest determinant of whether the 2030 target is met globally, since the region furthest from it is also the one with the fastest mobile-money growth.
- The gap between fee and total cost. As visible fees fall, exchange-rate margins tend to absorb some of the reduction. Look at the total, always.
- Correspondent banking relationships. Continued de-risking withdrawals from small markets would push costs back up regardless of what happens to technology.
The uncomfortable arithmetic of remittance pricing is that it is regressive by construction. The poorest senders, moving the smallest amounts, along the thinnest corridors, pay the highest percentages — and the fixed-cost structure means that is the natural result rather than an aberration. Fixing it requires changing the structure, which is slower than announcing a target and considerably more effective.