Earnings call transcript: Nu Holdings misses Q1 2026 EPS, stock steady

Nu Holdings reported record Q1 2026 revenue of $5 billion and net income of $871 million, up 41% year over year, but EPS of $0.18 missed the $0.20 consensus by 10%. The market reaction was muted, with the stock up just 0.23% aftermarket to $12.85, as investors balanced strong growth against higher credit loss provisions and seasonal pressure on risk-adjusted NIM, which fell to 9.5%. Management reiterated a full-year 2026 efficiency ratio around 20% and described AI-driven transformation, U.S. expansion, and SME lending as key growth drivers.

Analysis

NU’s real signal isn’t the headline EPS miss; it’s that operating leverage and credit discipline are now coexisting at scale. The market is still treating credit provisioning as a beta-to-default-rate story, but management is framing it as a function of product mix, tenor, and upfront IFRS-9 mechanics — which means the earnings power is more durable than the quarter-to-quarter noise suggests. That matters because the stock’s recent de-rating has created a setup where a normalizing provision cadence could re-rate the multiple quickly if investors stop anchoring on a single-quarter margin compression.

The second-order winner is the platform itself: AI-driven productivity is not just lowering cost, it is increasing model velocity in underwriting, collections, and product launches. If the company can keep shortening decision loops while pushing into higher-income and SME segments, the addressable profit pool expands with lower CAC than competitors can replicate. The risk is that investors overestimate how linear this adoption is — any slippage in data quality, regulatory friction, or a credit event in newer segments would challenge the “AI-enabled resilience” narrative fast.

The contrarian point is that the market may be underappreciating how much of the current debate is about framework mismatch rather than fundamentals. NU is increasingly a self-funding compounding story with optionality in U.S. expansion and SMEs, but the stock will likely remain range-bound until the sell side converges on a cleaner model of managerial vs IFRS economics. If that happens over the next 1-2 quarters, the current drawdown could prove to be a transition phase rather than a broken thesis.

Key Decisions for Investors

• Go long NU over a 1-3 month horizon on any post-earnings weakness below recent trading range; risk/reward improves if the market fades the provision noise and focuses on normalization in risk-adjusted NIM.
• Pair long NU / short a mature regional bank with slower growth and lower operating leverage (e.g., JPM or MS) over 3-6 months; NU should outperform if AI-driven efficiency and customer monetization continue to compound.
• Buy NU call spreads 3-6 months out to express upside from multiple expansion without full downside exposure; best entry is after any fade following the initial earnings reaction.
• For more defensive positioning, wait for a second-quarter credit update before adding aggressively; a real credit wobble would likely show up first in provisioning cadence and late-stage delinquency, not headline revenue.
• Monitor NVDA only as a thematic read-through: NU’s AI narrative is idiosyncratic, so avoid over-trading the name as a pure AI beta proxy unless broader AI sentiment is the dominant tape driver.