Visa's Moat Isn't the Network. It's the Contracts the DOJ Calls a Cage.

Tap a debit card for groceries and, in the time it takes the receipt to print, a question has already been answered that you never got to ask: which network carried the transaction. There is, in principle, more than one road that money could have taken. In practice, for the large majority of U.S. debit purchases, the answer was decided long before you reached the counter — not by you, not by the merchant in the moment, but by contracts signed in rooms you'll never see. That is the part of Visa's moat almost no one talks about, and it is the part a federal antitrust suit is now built around.⁵

The official story is that Visa is protected by network effects: billions of cards in wallets, millions of merchants that accept them, each side making the other more valuable until no challenger can break in. It's a beautiful story, and it's true at the surface. But it is also the investor-friendly name for something the Department of Justice describes far less flatteringly — a contractual web designed, it alleges, to make sure the road money takes is Visa's regardless of which road would have been cheaper.

Here is the thesis, plainly: Visa's deepest protection is not the network you can see — it's the contracts you can't. The 'network effects' framing describes how Visa got dominant. The contracts describe how it stays that way even where a rival rail already exists and would cost less.

Start with what Visa is actually paid for

To see the moat, first clear away the most common myth about it. People say Visa earns 2 to 3 percent on every transaction. It doesn't. That figure is the merchant discount — the bundled price a merchant pays its own bank — and the biggest slice of it is interchange, which by Visa's own published fee schedule flows from the acquirer to the cardholder's issuing bank, not to Visa.⁴ Visa collects a separate, far thinner network assessment fee. The lender's cut and the network's cut are different animals, and conflating them overstates Visa's per-swipe take by an order of magnitude.

That thin fee, multiplied by staggering volume, is the whole machine. In fiscal 2024 Visa processed 233.8 billion transactions on $13.2 trillion of payments volume¹ and turned that into $35.9 billion of net revenue, up 10% on the year.³ Against $12.3 billion of operating expenses, that's a GAAP operating margin of roughly 66% — software-like economics on a fee measured in fractions of a percent.¹ A business this profitable invites the obvious question: why hasn't anyone competed it down? The flattering answer is the network. The contested answer is the contracts.

Where the moat stops being organic

Network effects are real, and they explain why Visa sits in the dominant seat: it captured 61.1% of all U.S. card spending in 2024, against 25.8% for Mastercard, in a market where card spending topped $10 trillion.⁸ If the story ended there, the moat would be a triumph of scale and ubiquity, the kind no one is obliged to resent. But the DOJ's complaint, filed in September 2024 in the Southern District of New York, tells a sharper story about debit specifically — the one corner of the card world where federal law was supposed to force competition.⁵

Debit is different because the law made it different: regulation requires that debit transactions be routable over more than one network, so a merchant should be able to send a transaction down a cheaper rail. That's exactly the kind of opening that should erode a moat. The DOJ alleges Visa closed it not with a better product but with contracts. Visa handles more than 60% of U.S. debit transactions and collects over $7 billion a year in debit processing fees, the government says⁵ — and the way it protects that, the complaint argues, is by paying issuers and binding merchants and acquirers into routing arrangements that keep the volume on Visa even where an alternative exists.

The numbers in the complaint are the part that should make an investor sit up. According to the DOJ, Visa's merchant and acquirer routing contracts alone foreclose at least 45% of all U.S. debit transactions — not Visa's debit, all debit — and at least 75% of Visa's own debit volume is insulated from competition through its combined web of agreements.⁶ If those figures hold up, then three-quarters of Visa's debit moat is not the network choosing Visa. It's the contracts removing the choice.

The fair objection — and why a judge didn't buy it

The honest counter to all this is that an allegation is not a verdict. The DOJ filed a complaint; Visa denies it; a thin assessment fee on a genuinely useful global network is not, by itself, a crime, and incumbents with 60-point market shares get sued whether or not they did anything wrong. A scale advantage built over decades will always look, from the outside, indistinguishable from foreclosure — because dominance and exclusion produce the same market share. That objection is real, and a careful reader should hold it.

But it lost a meaningful round. In June 2025 a federal judge declined to dismiss the case, ruling that the government's allegations were plausible enough to proceed.⁷ That is not a finding that Visa broke the law. It is something narrower and still important: a court looked at the claim that Visa's debit dominance rests on contracts rather than competition and decided it was a serious enough argument to put to trial. The 'natural network effects' narrative is no longer uncontested fact. It is now a contested theory of the case — and the contest is happening in a courtroom, not an analyst's note.

What the threat actually is — and what it isn't

If you want to know where Visa is genuinely exposed, ignore the headlines about Big Tech. Apple Pay and Google Pay are not rail competitors — they run on top of Visa's network as a wallet layer, which makes them distribution amplifiers that route more volume onto Visa, not less. The wallet is paint on the road, not a second road. The real threat is the thing the DOJ case circles: anything that lets money move without touching the card network at all — real-time account-to-account rails, and regulated debit routing that actually forces transactions onto cheaper alternatives. Where those win, Visa's thin fee simply isn't collected, and no contract can foreclose a payment that never enters the system.

So the anatomy of the moat comes apart into three layers. There is the genuine, defensible utility — instant authorization, fraud rules, acceptance in 200 countries — that earns Visa a real place in the wire. There is the network ubiquity that made it the default. And underneath both, holding the debit volume in place where the law tried to pry it loose, there is the contractual web. The first layer is a moat anyone would admire. The third is the one a court is now deciding whether to drain. The most expensive misreading of Visa is to assume those three are the same wall — and that all of it is equally safe.

Visa's slivers add up to one of the most profitable businesses on earth because it stands in the one place every card transaction has to pass. The flattering question is how it got there: a network so complete that nobody can route around it. The harder question — the one a judge let proceed in 2025 — is whether, in debit at least, it stays there because no one can route around it, or because it wrote the contracts that make sure no one is allowed to try. A moat built on a network survives a competitor. A moat built on contracts survives only as long as the contracts do.