Every time a card is tapped at a terminal, a small piece of financial infrastructure activates that most consumers never think about. Visa does not lend money, does not set interest rates, and does not decide who gets approved for a card. Yet in fiscal year 2023 it generated roughly $32.7 billion in net revenue and converted more than half of that into net income. Understanding why requires looking past the Visa logo on the plastic and into the mechanics of the network sitting behind it.
Four Party Network Mechanics
Visa operates what the industry calls a four-party model: the cardholder, the merchant, the issuing bank that provides the card, and the acquiring bank that processes the merchant's transactions. Visa itself sits in the middle as the messaging and settlement layer, routing authorization requests between issuer and acquirer in a fraction of a second and guaranteeing that the transaction data is standardized and secure. It does not extend the credit line, does not hold the deposit, and does not absorb the loss if a cardholder defaults. That risk sits with the issuing bank. Visa is, functionally, a toll operator on a road it does not own outright but has spent five decades building the on-ramps for.
Revenue Without Credit Risk
Because Visa is not a lender, its revenue model looks different from a typical bank. It earns from data processing fees tied to the volume of transactions it authorizes, from service fees calculated as a small percentage of payments volume, and from international transaction fees when a purchase crosses currencies. Combined, these produce a take rate on total payments volume that is measured in fractions of a percent, often cited in the range of 0.13% to 0.15% of volume processed. In calendar 2023, Visa's network processed north of 233 billion transactions. Multiplying a tiny fee across an enormous and growing base of transactions is the entire economic engine, and it is a structure that has historically produced operating margins above 65%, a level unusual for a company of Visa's revenue scale.
Scale as the Moat
The competitive position rests less on any patent or proprietary technology and more on the difficulty of replicating a network that already touches over 100 million merchant locations and thousands of issuing banks across more than 200 countries and territories. Every new bank that issues a Visa card makes the network marginally more valuable to merchants, and every new merchant that accepts Visa makes the card marginally more valuable to a cardholder. This two-sided network effect is why the payments industry has consolidated around a small number of global rails rather than fragmenting into dozens of regional systems. Building a comparable footprint from scratch would require persuading millions of merchants and thousands of financial institutions to adopt a new standard simultaneously, a coordination problem that has proven extraordinarily difficult for challengers to solve at global scale.
Where Competition Actually Comes From
Mastercard is the most direct structural analogue, running a similar four-party model at a somewhat smaller scale, and the two together have historically processed the large majority of global card-based payments outside China. American Express and Discover, by contrast, run closed-loop networks in which they act as both network and issuer, a different risk and revenue profile entirely. The more interesting competitive pressure in recent years has come from account-to-account payment rails, such as Pix in Brazil, UPI in India, and various real-time payment systems in Europe, which allow money to move bank-to-bank without a card network intermediary at all. These systems have grown quickly in specific domestic markets and represent a genuine long-term question about how much of future payments volume routes around, rather than through, networks like Visa's.
Regulatory and Currency Crosswinds
Interchange fees, the portion of the transaction fee that flows to issuing banks, have been the subject of regulatory intervention in multiple jurisdictions, including fee caps imposed in the European Union and ongoing litigation and legislative proposals in the United States around interchange and routing rules. Because Visa's own take is smaller than the interchange fee and separate from it, direct regulatory pressure on Visa's own pricing has been less severe than on issuing banks, but merchant lobbying against network fees generally has been a recurring feature of the business's operating environment for over a decade. Currency translation is another structural factor: because a meaningful share of volume is international, movements in the dollar against other major currencies can add or subtract several percentage points from reported revenue growth in a given year without reflecting any change in underlying transaction activity.
The rule to internalise
A payments network's economics are best understood by separating the fee it earns from the risk it does not carry. Visa's business model has historically demonstrated that transaction volume, not credit exposure, is the variable worth tracking, and that a network's durability tends to rest on the mutual dependency between merchants and issuers rather than on any single technological advantage. For a long-term investor building a mental model of the payments industry, the more useful exercise is tracking how account-to-account rails and regulatory interchange rules evolve over time, since those forces, not competition from another card network, represent the more structural test of the toll-booth model over the coming decade.
Educational content only. Not investment advice.