Wise Owns Its Rails. Remitly Rents Them. Who's Winning?

Alex: One company that moves money home for migrants owns its payment rails, and keeps 52 cents of every hundred dollars. Another rents its rails, keeps about two dollars, and grew faster last year. So in remittances, does owning the pipes decide who wins? A founder from Tanzania learned his answer the hard way. He wanted to build a simple app on top of other companies payment systems.

Alex: By his own account those systems failed so often that he had no choice. He applied for licence after licence, 17 of them, and built the rails himself. Today, one of his customers is MoneyGram, a company he used to send money through. So that's one answer. Own the rails, or lose.

Alex: The numbers say it's messier than that. I'm Alex, and this is Curiosity Code. Most weeks I talk to someone building in fintech or AI. This one is just me working through a question I've already answered twice in public in two different ways. People working abroad send hundreds of billions of dollars home every year.

Alex: And for many countries, this is one of the biggest sources of foreign money they have. It is also still expensive. Sending $200 costs about 6.4% on average, against a target of three. So there is real money here and a clear reason to compete for it. There are two ways to build a remittance business.

Alex: You can own the plumbing, the licences, the bank connections, the money sitting in each country ready to pay out, or you can build the app and the brand and rent the plumbing from partners. The company keeping 52 cents owns its plumbing, the one keeping two dollars rents it. Both models are working right now and that's the puzzle. Here's the thing, this experiment has already been run somewhere else at speed and in public. Crypto spent a decade building rails that anyone could use.

Alex: Then watched where the money went. Same two layers, the one that moves the money and the one that owns the customer. Its rails are open so nobody can own them and the answer came back clean. The money went to whoever already had the customers. Remittance rails are the opposite.

Alex: They are licensed, scarce and slow to get. As the founder from the opening puts it, you cannot vibe code a licence. So crypto tells you what happens when the rails are free. And the question here is whether a licence changes that answer, which is why the guests I keep coming back to on this one built in crypto. Their episodes are linked in the show notes.

Alex: Colin Butler spent three years running institutional capital at Polygon, one of the big blockchain networks, betting that value would build up in the infrastructure. On this show last week, in his words, value didn't accrue to the infrastructure layer where I was at. My big mistake at Polygon was believing that it would. Where did it go? To whoever already had the customers.

Alex: Elena Shiltseva, talking about stablecoins, expected the technology to end up in the background of remittances. Her words: they will use blockchain on the background as an infra for international remittances. Infrastructure nobody sees. So my guests lean one way, the customer wins and the pipes become invisible. The founder from the opening would disagree, and so, it turns out, would I.

Alex: I put that claim to a founder I recorded with this week, whose episode is out in a couple of weeks. He did not dispute that a bank could do it. There is no law of physics stopping them, he said. His answer was that most people are perfectly happy with what their bank offers, while a small group want everything the market has. Both can be right, for different customers.

Alex: Here is where I land. This summer I wrote that a fintech that doesn't own its rails is a guest in someone else's house. And that distribution is worth more than margin. Remittances is where both hold at once. Owning the rails decides whether you can survive on thin margins.

Alex: Owning the customer decides whether you get to use them. And in much of Africa, neither owns the last mile. The telco does. Let me unpack that because those two posts sound like they contradict each other. Nuvei agreed to buy Payoneer.

Alex: The obvious read was consolidation. Mine was that Nuvei was buying relationships it could never win on its own. Distribution. In September, I wrote the other side. Build on someone else's core and you can furnish the house beautifully but you don't hold the keys.

Alex: The owner sets the rules and the pricing and changes both when it suits them. Remittances shows you why both are true. Rails and customers answer different questions. The rails answer: can you do this cheaply and reliably in a market where the banks' systems are 15 years old? The customer answers: will anyone pick you to do it?

Alex: You need yes to both. You can buy one of them, you mostly can't buy the other. Here is what the numbers say on both sides. Start with the case for owning the rails. Wise has connected directly into national payment systems in several countries, including a settlement account at the Philippine Central Bank.

Alex: In its last financial year, it moved $243 billion across borders. For every $100, it kept 52 cents. Three quarters of its payments arrived in under 20 seconds and banks now use Wise's system for their own customers' transfers. That is owning the rails at its best, cheap, fast, and a product other people buy. The founder from the opening tells a smaller version of the same story.

Alex: Before he built his own rails, he says, partners were failing about 25 times a month. Afterwards, over 99% of transfers arrived within an hour. Now the other side. Remitly owns very little of its plumbing. Its annual report describes a network of partners across more than 5,000 corridors without local operations in each country.

Alex: Last year, it sent almost $75 billion, up 37%. That's faster growth than Wise. It also made its first full year of profit. And Remitly keeps about $2 on every hundred. Four times what Wise keeps.

Alex: Customers pay it anyway. And the corridor that costs the most is the one this question is really about. Sending money to Sub-Saharan Africa still costs about 8.5%, well above the global average. So let me recap. Wise owns the rails and wins on cost.

Alex: Remitly rents the rails and wins on growth. Both are working. Which tells you the rails decide your cost and the customer decides your share. So what does this mean if you're building or backing a payments business in emerging markets? First, ask whether the rails can be owned at all, where they are a commodity, as in crypto, distribution wins by default, and there is no argument to have.

Alex: In remittances, the licence is the only reason this is even a question. Second, own the rails where they are broken and rent them where they work. The founder from the opening did not build infrastructure for the margin. He built it because the systems underneath kept failing. If a partner is reliable, renting is cheaper.

Alex: Third, owning rails means owning money in every country you pay out in, sitting there before the customer presses send. His company grew so fast it broke, by his own account and needed a $50 million credit line to keep up. Rails are a balance sheet decision as much as a technology one. Fourth, look at who owns the last step. In Kenya, the money usually lands in M-Pesa, the mobile money service run by the telco Safaricom, with over 43 million active customers.

Alex: Neither Wise nor Remitly owns that. Everyone pays to reach it. Which leaves the question I can't answer yet. If the last mile in so many of these countries belongs to a telco, is the real winner in remittances the company nobody thinks of as a remittance company? If you've built in one of these corridors, I'd like to hear where you landed.

Alex: Leave a comment if you are watching or find me on LinkedIn if you're listening, and follow the show wherever you are.