Open in interactive viewer → charts, metric popovers & call review

Nu's $1.1 Billion Quarter Isn't the Story — The Private Payroll Switch Is

Record profit, a U.S. launch, and a quiet shift from private-payroll caution to offense — while the tape votes no on Brazilian credit.
NU · Earnings Call · 2026-11-12

The quarter that changed the perimeter

On paper, Nu Holdings delivered a victory lap: 143 million customers, $6.3 billion of gross revenue, “record net income of $1.1 billion in the quarter” — David Velez Osorno, Founder, Chief Executive Officer and Chairman · 2026-11-12, a 32% return on equity and a 19.3% efficiency ratio. But the more important change is the perimeter. Mexico is now operating as a bank, Colombia is accelerating, Nu U.S. launched on September 10, and Nu Global went live on October 15. David Velez said Nu had “already reached customers across all 50 states” — David Velez Osorno, Founder, Chief Executive Officer and Chairman · 2026-11-12 in the first few weeks of the U.S. rollout. That is a different company from the one that spent years explaining why it would not rush into new markets. The stock, however, entered the print roughly a third below its high, with October's Brazilian election result weighing heavily on the shares. So the report sets up a clean tension: operating momentum is accelerating, but the market is pricing Brazil macro and political risk, not the compounding of a 143-million-customer franchise.

Private payroll: from "not yet" to "lean in"

The most consequential language change on the call was about private payroll lending. For several quarters, management treated the product as a promising but undercooked market. In August 2025, CFO Guilherme Marques do Lago warned that

first payment defaults are set to be in the 10% to 18%, which I think it is higher than what at least we would expect so far. So we are not yet fully comfortable with the quality of the collateral

Guilherme Marques do Lago, Chief Financial Officer · 2025-08-15
. In November 2025 he said Nu remained “more cautious than some of the other players in the industry with respect to its cost of risk” — Guilherme Marques do Lago, Chief Financial Officer (CFO) · 2025-11-13. Even in February 2026, when he acknowledged that the feared cannibalization and subordination risks had not materialized — “we have not yet seen any evidence that any of those 2 risks that you've laid out are materializing” — Guilherme Marques do Lago, Chief Financial Officer · 2026-02-26 — Nu still would not lean in. By August 2026, David was saying Nu was “slowly getting more comfortable with the product” — David Velez-Osomo, Founder, Chief Executive Officer and Chairman · 2026-08-13. This quarter, Rob Livingston removed the hedge: “we are getting very comfortable with private payroll at this point” — Rob Livingston, Chief Financial Officer · 2026-11-12, adding that originations are doubling quarter-over-quarter off a small base and that Nu sees no reason it should not be a leading private payroll provider within a year. That is a genuine strategic pivot, not a new talking point. The company-specific keyword private payroll space jumped into Nu's top five this quarter. But the tape disagrees. The global keyword basket for private payroll loans — which includes NU, BBD, ITUB and BBSE3 — is down roughly 25% over 30 days and about 13% over 360 days. Either the market is underestimating Nu's underwriting edge, or it is discounting a credit product that is growing just as Brazil's consumer cycle looks fragile. That debate is the heart of the quarter.

The NPL optical illusion

Headline delinquency looks worse. The 15-to-90-day ratio rose from 4.8% to 4.9%, and the 90-plus ratio rose from 6.9% to 7.3%. Management's rebuttal is structural: interest-earning balances are now 61% of the portfolio, up from 59% last quarter and 44% three years ago, and delinquency is concentrated in those balances.

As their share grows, delinquency measured over the total portfolio naturally increases even without a deterioration in underlying credit performance or an increase in exposure. ... On that basis, our 15- to 90-day ratio improved to 8.0% from 8.2% last quarter, while our 90-plus ratio increased to 12.1% from 11.6%.

Rob Livingston, Chief Financial Officer · 2026-11-12
On a recurring basis, risk-adjusted NIM expanded from 11.9% in Q2 to 12.7% in Q3. The reported 12.1% was hit by a 63-basis-point one-off charge on receivables acquired from payment institutions later placed into extrajudicial liquidation by Brazil's central bank. This is the quarter's crux. Nu is deliberately taking more risk — keyword risk expansion is now the company's top keyword — because risk-adjusted returns justify it. The vocabulary rotated from last quarter's top keyword, Cost of credit, to this quarter's offense. That is exactly what makes investors nervous: expanding unsecured personal loans and private payroll into a soft Brazilian consumer cycle is either smart counter-cyclical share-taking or late-cycle bravado. David's answer is the same one he has given for quarters: assume the future is worse than the past, underwrite with a buffer, and require every loan to be NPV-positive in a stress case. The philosophy is consistent; the difference is that Nu is now applying it at much larger scale.

Beyond Brazil — and the AI subtext

Mexico added 870,000 customers to reach 16.6 million, and bank status lifted deposit insurance coverage 16-fold, from roughly $12,000 to $195,000 per customer. Colombia reached 6.1 million customers, became the No. 1 card issuer by cardholders with 3 million, grew its credit portfolio 80% year-over-year in FX-neutral terms, and is benefiting from Bre-B, the instant-payment system that has reached 82% adult penetration in its first year. Those are the kinds of adoption curves that make the "Nu model travels" thesis credible rather than promotional. The U.S. is the wildcard. Nu's initial proposition — 3.5% APY, fee-free transfers to Brazil, Mexico and Colombia, and a no-annual-fee card with 1.5% cash back — is aimed at mass-affluent customers and diaspora communities. Rob previously said U.S. credit models would take 12 to 30 months to build confidence. The launch does not invalidate that caution; it starts the clock. Nu Global, built on stablecoin infrastructure and already in 37 countries, is even earlier. David framed AI as the eventual disruptor of U.S. banking economics: “beginning of the end for a lot of those profit pools” — David Velez Osorno, Founder, Chief Executive Officer and Chairman · 2026-11-12. That ties Nu to the global personal AI agents theme, but Nu's real AI edge is less visible: a credit foundation model trained across 140 million customers, which it hopes will shorten the learning curve in new markets. The moat remains the primary banking relationship — roughly 60% of customers use Nu as their main account, giving it positive selection in the credit stack.

What to watch

Nu trimmed full-year efficiency guidance to 19%–19.5% from around 20%, a sign that AI and disciplined hiring are producing more operating leverage than planned. The effective tax rate rose to 19.2%, and management expects 2027 in the low-to-mid 20s. On M&A, David set an extremely high bar, with organic growth still the strong preference. The December 8 Investor Day will be the next catalyst. The quarter proves Nu can compound through a bad macro backdrop. But the stock is not trading on the compounding; it is trading on Brazil. The private payroll switch and the deliberate risk expansion are the two variables that could either extend the growth runway or mark the top. The tape currently says skepticism. Nu's answer is that it is taking share, not taking excess risk. That argument now has to survive a full credit cycle.