Stripe Paid $1.1 Billion for Stablecoins. It Did Not Buy a Single One.

The largest deal in stablecoin history bought licences and counterparties. The token was never the hard part.

Share
Stripe Paid $1.1 Billion for Stablecoins. It Did Not Buy a Single One.

The most expensive deal in stablecoin history bought plumbing, not currency. That distinction is the whole lesson.

Stripe closed its acquisition of Bridge last week for $1.1 billion, the largest deal the stablecoin sector has seen. It is worth being precise about what that money bought, because almost every account of it has been imprecise in the same direction.

Stripe did not buy a stablecoin. It did not buy a token, a reserve, a mint, or a claim on anybody's dollars. It bought a company that makes it possible for other companies to move stablecoins — issuance, conversion, and the connections at either end where digital dollars turn back into the kind a bank will recognize.

It bought the plumbing.

That distinction sounds pedantic until you have tried to run a cross-border payment through a stablecoin and discovered which part of it was hard. The token was never the hard part. The token is the easiest component in the entire chain, and it has been solved for years.

What actually changed hands

Stripe confirmed the deal in late October, and Patrick Collison marked it with a line that has been quoted ever since: stablecoins are room-temperature superconductors for financial services. The acquisition closed on 4 February. At $1.1 billion it is the largest Stripe has ever made.

Bridge itself is three years old. It was founded in 2022 by Zach Abrams and Sean Yu, who came out of Coinbase and Square, and it is based in San Antonio. It had raised around $58 million, and when it closed a $40 million Series A last year it was valued at roughly $200 million.

Hold those two numbers next to each other. Two hundred million, then eleven hundred million, inside a year. The company did not become five times better at anything in twelve months. The market changed its mind about what the asset was.

What Bridge sells is business-to-business infrastructure: an interface that lets a company issue, hold, move and convert stablecoins without building any of that itself. It is an orchestrator. It sits beside card rails and bank rails rather than replacing them, which is precisely why it fits inside Stripe rather than competing with it.

Stripe had been walking toward this for a while. It turned crypto purchases back on in Europe in July, and in October it launched a way for merchants to accept stablecoins directly. Buying Bridge is the same bet, made with conviction and a balance sheet.

Collison's framing on the day it closed is the honest version: that everyone programmatically moving money will eventually want a stablecoin strategy. Not everyone will want a stablecoin. They will want a strategy — which is a statement about operations, not about currency.

The part that was never hard

Here is where I think most of the commentary goes wrong, and where someone who runs a processor sees something different.

Take an ordinary cross-border payment and try to run it on a stablecoin. Walk the whole chain, not the exciting middle of it.

Someone has to turn real money into tokens. That is an on-ramp, and it requires an institution willing to take fiat from an identified customer, in a jurisdiction where it is licensed to do so, under rules about who that customer may be.

Something has to hold the tokens between the two ends. That is custody, with its own regulatory treatment, its own insurance question and its own operational risk, which is not the same risk as a card balance and does not sit in the same place on a balance sheet.

The tokens have to cross. This is the part everybody writes about. It costs cents and takes seconds and it is, genuinely, very good.

Then somebody has to turn the tokens back into money that a bank on the other side will accept, in the local currency, into an account that belongs to the person being paid. That is an off-ramp, and it requires another licensed institution, in another jurisdiction, willing to do the same thing in reverse, for a counterparty it has never met.

And underneath all of it: sanctions screening at both ends, transaction monitoring that understands chain analytics, a reconciliation process for a ledger that does not look like any ledger your core system was built for, tax treatment, accounting treatment, and a support team that can answer what happens when a transfer goes to the wrong address and cannot be recalled.

Count the hard parts in that list. Exactly one of them is the token, and it is the one that works.

Everything else is permission, counterparties and operations. Which is to say: everything else is the business Bridge is in, and the business Stripe just paid a billion dollars to be in.

The licenses are the asset

If the token is commodity and the code is replicable, what exactly costs a billion dollars?

Permissions. The right to touch real money at both ends, in each jurisdiction, under a regulator that has looked at you and said yes.

That is the part no engineering team can ship. You cannot write it, you cannot buy it on a marketplace, and you cannot move fast and break it, because the consequence of breaking it is not a bug report. A money transmission authorisation, a payment institution licence, a banking relationship that survives a compliance review — each takes years, each is granted locally, and each can be withdrawn. Multiply that by every country where you want the money to land.

Which is why the interesting sentence in this deal is not about stablecoins at all. Stripe bought a company whose value is that it already has the arrangements, in enough places, to make the easy part usable.

This is a familiar shape to anyone in payments, and we should recognise it rather than be surprised by it. It is the same reason a card network is valuable. Not the message format — the message format is public and boring. The value is that thousands of institutions have agreed, in writing, under a shared rulebook, to accept it. The asset is the agreement, not the technology.

Stablecoins are heading to the same place. The protocol will be free. The permissions will not.

From a Mexico City desk

In México the usual stablecoin pitch lands squarely, which is exactly why it is worth being careful about what this deal actually proves.

The pitch is: here is a way to hold and move dollars without needing a US bank account. In a country with a volatile currency and restricted access to dollar accounts, that is a genuinely powerful offer, and it is the dominant retail story here.

But there is a second proposition underneath it, and it is the one an acquisition of licences is really about. What we do not have — what no country outside the United States has — is direct access to the system that settles those dollars. Our dollars move internationally through correspondent banks, which means through somebody else's compliance department, on somebody else's schedule, at a price we are quoted rather than negotiate. That dependency is the actual constraint, and it is a constraint the token genuinely does address.

But only if the other end exists.

A stablecoin corridor is not a corridor until somebody licensed on the far side will convert tokens into local money and put them in a named account. Until that exists, what you have is a fast way to move value to a place where it cannot get out — which, operationally, is not an improvement. It is the same dependency wearing better clothes.

So the question for an issuer or processor in this region is not whether to have a stablecoin strategy. It is narrower and more useful: for each corridor that matters to your clients, who is the licensed off-ramp, what are they regulated by, and what happens to your customer's money on the day that institution loses its banking relationship?

That is a counterparty question. We know how to ask counterparty questions. We have been asking them about correspondent banks for forty years. This is the same question about a newer counterparty, and the fact that the middle of the chain is now cryptographic does not change what we should be diligencing at the ends.

My judgment

The sector is going to spend the next several years arguing about which stablecoin wins. That argument will generate enormous commentary and very little that matters operationally, because the token is converging on commodity and commodities do not have winners in the sense the argument imagines.

The question that will actually determine outcomes is duller and much harder: who holds the permissions at both ends of the corridors you care about.

Stripe just answered that question with a billion dollars. Read the deal as a valuation of licensing and counterparty arrangements, and it stops looking like a crypto bet and starts looking like something far more conventional — a payments company buying distribution into places its existing rails do not reach cheaply.

For those of us operating outside the United States, the strategic fork is real and it is close. Either you become an on-ramp and off-ramp in your own market, which is a licensing and compliance project measured in years and defended for decades, or you rent somebody else's, which is fast, cheap, and makes you a reseller of a dependency you do not control.

Both are legitimate choices. What is not legitimate is drifting into the second one while telling the board you are doing the first.

Three questions for the people who own this

1. For each cross-border corridor your clients actually use, can you name the licensed institution on the far side that converts tokens into local money? If you cannot name it, you do not have a corridor. You have a demonstration.

2. When a customer's funds sit as tokens between your on-ramp and somebody else's off-ramp — six hours, overnight, a weekend — whose balance sheet are they on, and what does your regulator call that position? Has anyone asked your regulator, or is the answer currently an assumption?

3. On-ramp of your own, or rent someone else's? One is a multi-year licensing programme. The other is a dependency. Which one has your institution actually chosen, and can you point to the meeting where that was decided?

Reply to this email with a line or two. The sharpest answers will shape the next issue.

Sources

Ledger Insights, 21 October 2024 — Stripe confirms the acquisition of Bridge; Patrick Collison's description of stablecoins as room-temperature superconductors for financial services. Link

TechCrunch, 5 February 2025 — Stripe closes the $1.1 billion Bridge acquisition; Bridge's founding, founders, funding history and prior valuation, and Stripe's earlier crypto moves. Link

Bridge's funding figures and prior valuation are reported by TechCrunch citing PitchBook. The $1.1 billion price was first reported by Michael Arrington and subsequently confirmed in Stripe's own communications; I have not seen a filing that states it.