Sending Money to Malawi? We Can't Tell You What It Costs

We have priced every other South African money-transfer corridor we track. This one we are not going to price, and the reason is more useful than a number would have been.

What the data actually looks like

Thirteen services were surveyed on South Africa to Malawi. Sorted by the exchange-rate margin each applies, they fall into two groups that do not touch:

Look at the second row again. Six independent providers land within 3.68 percentage points of each other, and a gap of 25.94 points separates them from everybody else — with nothing in between.

That is not what competition looks like. Competitive pricing produces a spread, not two tight clusters with a chasm between them. What it looks like is two different exchange rates being applied to the same currency pair.

The row that settles it

If the split were about the kind of provider, it would be explainable. It is not.

The higher group contains four bank-type services and two non-banks. The lower group contains one bank-type service and six non-banks. Both groups contain both kinds.

And then there is this:

One institution, one currency pair, one survey — and two views of the exchange rate 58 percentage points apart.

A single bank cannot genuinely hold both of those opinions about what a rand is worth in kwacha. The only reading that fits is that the two products transact at different rates.

Why we will not just publish "the cheap half"

The obvious move is to report the lower group and ignore the rest. We considered it and it is wrong.

We do not know which group is right.

If the lower group is quoting a rate that a recipient cannot actually get, its apparent cheapness is an illusion and the real cost of those services is much higher than the table says. If instead the higher group is the anomaly, the lower group is fine.

The data cannot distinguish between those. Publishing "the cheapest way to send money to Malawi is 14.46%" would be presenting one of two possibilities as settled fact — on a corridor used largely by people sending small amounts home, where being wrong costs them directly.

So there is no cheapest on this page, no median, and no ranking.

What we are not claiming

We are not saying any provider is doing anything wrong. A margin computed against a benchmark tells you the two numbers differ. It does not tell you which is the honest one.

We are not explaining the cause. There is an obvious candidate — an official rate diverging from the rate people actually transact at, which is a well-known phenomenon in economies under currency stress. This dataset cannot demonstrate that, and we did not research Malawi's monetary arrangements. So we are describing what the numbers do, not why.

We are not saying the survey is faulty. It reports what providers quoted. The difficulty is comparing quotes that rest on different rates.

The one thing we can say, and it is not good

Even taking the lower group entirely at face value — the most generous reading available — Malawi is expensive.

Its cheapest surveyed option, at 14.46%, is the second-worst floor of any South African outbound corridor in the survey. Only China is worse:

Every other corridor's cheapest option is under 8.3%. On Malawi, the best verifiable case is nearly double that — and the best case may not be the real case.

Why this matters more here than it would elsewhere

A pricing puzzle on a corridor between wealthy countries is an academic irritation. On this one it is not.

The South Africa to Malawi corridor carries money sent by people working in South Africa to households at home, usually in small amounts, usually monthly, and usually where the money is spent on necessities rather than saved. Three things follow.

The percentage is the whole cost. On a small transfer there is no scale to absorb a bad rate. A sender moving a few hundred rand has no ability to batch their way out of it, and no realistic alternative rail.

The person bearing the error is not the person choosing. The sender picks the provider; the recipient discovers what arrived. If a quoted rate is not the rate the recipient can access, the gap lands on someone with no say in the decision and often no way to query it.

A wrong recommendation compounds monthly. Naming a "cheapest" that turns out to be quoting an inaccessible rate would not cost a reader once. It would cost them every month until they noticed — and the design of the mistake is that it is hard to notice, because the app said the transfer went fine.

That is the calculation behind not publishing a ranking here. On a corridor where we could be confident, a table helps. On this one, the cost of being confidently wrong falls on households least able to absorb it.

What to do instead, which needs no answer to the puzzle

Here is the useful part, and it works regardless of which group turns out to be right.

Ask what arrives, in kwacha. Not the rate. Not the fee. Put the same rand amount to three providers and ask each: how many kwacha will my recipient receive? Write down three numbers and pick the biggest.

That comparison is immune to the entire problem on this page. It does not depend on any benchmark, any margin calculation, or which rate is the "real" one. It measures the only thing that matters to the person at the other end.

Then confirm with the recipient. On a corridor where quoted rates diverge this much, the number that counts is what actually lands in the account or is handed over at the counter — not what the app said before you sent. Check the first transfer with a small amount before committing to a pattern.

And do it again periodically. Whatever is causing the divergence, it is not a fixed property of the corridor. A provider that was best last year is not necessarily best now.

How current this is

The survey was field-collected between 10 February and 11 March 2025. On a corridor showing this much rate dispersion, an eighteen-month-old snapshot is even less of a guide to today than usual.

Treat everything here as a description of a structural problem rather than a price list — because that is exactly what it is.

Source: World Bank, Remittance Prices Worldwide, 2025 Q1 (CC BY 4.0). Analysis by Rateweb; the World Bank does not endorse it.

How does this affect YOUR Money OS?

On a corridor where quoted rates disagree by 26 percentage points, the single question "how much will actually arrive" is worth more than any comparison table — including one we could have published.

Check my free OS score

FAQ

What is the cheapest way to send money from South Africa to Malawi? We are not naming one. Six of the thirteen surveyed providers apply exchange rates around 58% away from the other seven, and we cannot establish which group reflects the rate a recipient can actually get. Naming a cheapest would present a guess as a fact.

Why can other corridors be priced but not this one? Because on every other corridor the providers' rates form a continuum. Here they form two clusters 26 percentage points apart, with the same bank appearing in both — which indicates two different exchange rates rather than competitive pricing.

Is one of the providers cheating? We are not saying that and the data cannot show it. A margin measured against a benchmark tells you the numbers differ, not which one is honest.

So how do I compare providers? Ask each one how many kwacha your recipient will receive for the same rand amount, and pick the biggest number. That is immune to the whole problem, because it measures what arrives rather than what rate was used.

Is sending money to Malawi expensive? On the most generous reading of the data, yes. The cheapest surveyed option is 14.46%, the second-worst floor of any South African outbound corridor — and that best case may not be the real case.

Will you price this corridor later? If a survey round shows the rates converging, or the divergence becomes explainable from a source we can stand behind, yes. Not before.

Related

• What sending money within Africa really costs — 50 corridors
• The cheapest way to send money to Zimbabwe from South Africa
• Sending money to Lesotho or Eswatini