By the late 2010s, paying digitally in Brazil was expensive. Two card networks dominated the market and charged high fees for every transaction. Small shops often refused card payments because the fees ate into their margins, and millions of Brazilians without bank accounts could not pay digitally at all. The Banco Central do Brasil had two options. It could wait for the banks and card networks to build a cheaper system themselves, knowing they had little reason to do so because the existing system was profitable for them. Or it could build a new system itself, which had never been attempted at this scale by a central bank.
What Brazil needed was an instant payment system that worked across every bank, was free for ordinary people to use, and could not be controlled by any one company. The market was not going to produce that. The banks and card networks that ran the existing system had no reason to build a free alternative to a service that was making them money.
The Banco Central do Brasil launched PIX in November 2020. Participation was mandatory for every licensed bank above a certain size. Transfers were free for individuals and for small merchants. The central bank wrote the technical standards, ran the central settlement system, and required banks to offer PIX inside their existing apps, alongside other payment options. The reasoning was that by building and running the basic system itself, the central bank could open up competition on the services built on top of it. Banks and fintechs could still compete on apps, features, and customer experience. They just could not compete on the underlying rails.
By December 2024, more than 6 billion PIX transactions were taking place every month. On a single day in late December 2024, the system processed 252 million transactions, a record. By the end of 2024, the value of PIX transactions was around 80% higher than the combined total for credit and debit cards in Brazil. New features are being added: PIX Automático, which allows recurring payments, and Tap-to-Pay, both rolling out across 2025 and 2026.
Two limits are worth noting. By building and running the system itself, the central bank has become a single point of failure for a very large share of Brazil's payments. When PIX goes down, the country notices. Competition concerns that used to focus on the card networks have shifted toward the central bank itself.
Public infrastructure sometimes works better than private competition, but only when the regulator is willing to build it, run it, and take political responsibility when something goes wrong. PIX is a model that works when the central bank has the legal mandate, the technical capacity, and the appetite to be a long-term operator of national infrastructure. Other countries that want similar outcomes have to ask whether their central bank has all three. Where any one of them is missing, the PIX model is much harder to copy than the headline numbers suggest.