Brazil built the developing world's most competitive payments market. The instant-payment system Pix, the rise of Nubank, and a long tail of more than two thousand licensed and unlicensed payment players all flowed from a regulatory posture that let small institutions enter cheaply and scale fast. Banco Central do Brasil is now tightening that posture sharply — and its own projections suggest the squeeze will reshape the sector.
On November 3, 2025, the National Monetary Council and the central bank issued Joint Resolution CMN nº 14 and Resolution BCB nº 517, replacing flat, license-type capital floors with an activity-based methodology: minimum capital is now a function of the operations a firm actually performs and the risk they carry, phased in at 25% of the gap by December 2026, 50% by June 2027, 75% by December 2027, and 100% by January 1, 2028 (Joint Resolution 14). For payment institutions, the floor climbs from roughly R$1 million toward as much as R$9.2 million depending on activities. Reuters reported the central bank expects around 500 firms to recapitalize, lifting aggregate sector requirements from about R$5.2 billion to R$9.1 billion (Reuters/Investing.com).
The numbers the central bank itself published
The most striking figures come from Banco Central's own Financial Stability Report, detailed by Brazilian legal press on May 28, 2026. Among institutions analyzed, 679 — roughly 39% — could fall short of the new requirements. The risk is concentrated almost entirely among nonbanks: about 5% of multiple banks face a shortfall, against 63% of payment institutions, 81% of direct-credit societies (SCDs), 83% of foreign-exchange brokers, 86% of microcredit societies, and 92% of peer-to-peer lending platforms (SEPs) (Migalhas). The same report estimates the systemic capital deficiency at just 0.5% of the system's reference equity — a number worth holding onto.
This sits inside a broader enforcement package. On September 5, 2025, Resolutions BCB nº 494 through 498 tightened authorization, banned coworking and virtual-office addresses for licensees, capped Pix and TED transfers through unauthorized institutions at R$15,000, and set a hard May 1–31, 2026 window for previously unlicensed e-money issuers and acquirers to file for authorization or wind down within 30 days (Resolution 494).
The strongest case for the crackdown
The regulator is not acting on a hunch. The September rules followed Operation Carbono Oculto (Hidden Carbon), a police investigation that exposed how organized crime had threaded fintechs and IT service providers into fuel-sector money laundering, using thinly capitalized shells to obscure beneficial owners (Bloomberg Law). When a payment institution can be stood up in a shared office with a million reais and routed billions in dirty flows, the integrity of the entire rails is at stake. Activity-based capital is also, in principle, smarter regulation than the old type-of-license buckets: it scales the buffer to what a firm does rather than how it is labeled, and the "same activity, same risk, same regulation" framing is one most economists would endorse. The 18-month phase-in is a genuine accommodation, not a cliff.
Where proportionality breaks down
Grant all of that, and the design still conflates two different problems. Money laundering is an AML, KYC, and beneficial-ownership failure. Capital adequacy is a solvency instrument. The governance, physical-premises, and transaction-cap rules in Resolutions 494–498 are precisely targeted at the laundering threat — they make shell structures harder to build and easier to trace. A blanket capital floor does not. It does little to stop a well-funded bad actor and a great deal to evict small, legitimate operators whose only deficiency is being small.
Banco Central's own 0.5% figure is the tell. If the firms projected to fall out of compliance collectively represent half a percent of system reference equity, they are not a systemic solvency risk — which is the problem capital requirements exist to solve. A rule that knocks out 63% of payment institutions and 92% of P2P lenders while curing 0.5% of system fragility is not calibrated to solvency. It is, in effect, a market-cleansing tool wearing a prudential label, and regulators should say so plainly rather than borrow the credibility of capital rules for an industrial-policy outcome.
The predictable result is consolidation. Smaller fintechs will be acquired, merged, or shuttered, and the competitive fringe that disciplined Brazil's incumbents — the same fringe that produced Pix-era pricing pressure — will thin. That raises concentration in a banking market already dominated by a handful of giants, the opposite of what a pro-competition regulator should want.
A more surgical path
The better instrument is the one Banco Central already deployed in September: rigorous authorization, real-premises requirements, beneficial-ownership transparency, transaction caps on unlicensed flows, and active AML supervision of the IT providers Carbono Oculto exposed. Those tools attack the laundering channel directly. Where solvency genuinely warrants higher buffers — institutions taking real credit or settlement risk — capital should rise. But applying a uniform activity-based floor that disqualifies the overwhelming majority of nonbank lenders, when the regulator's own data shows negligible systemic exposure, trades a competitive, innovative market for a tidier one. Brazil can keep its rails clean without thinning the field that made them world-leading.