Stock analysis · NYSE: NU · October 5, 2026 · Prices in U.S. dollars

Nu Holdings Stock: Underwriting the Digital Bank’s Next Expansion

A 13% election-day rally restores the share price, but durable returns depend on credit-adjusted earnings—not political enthusiasm or customer scale alone.

Executive Summary

Recommendation: Hold; selective accumulation only with a larger margin of safety. Nu Holdings closed October 5 at $15.18, up 13.03% from $13.43. Its digital banking model has moved beyond a customer-acquisition story: second-quarter net income reached approximately $1.1 billion, management reported a 33% return on equity, and deposits supported a growing credit portfolio. The investment question is whether those returns survive deliberate expansion into higher-risk lending and newer markets.

At roughly 20.7 times trailing earnings and 5.5 times book value, NU is not priced like an ordinary low-growth bank. That premium can be earned through sustained monetization, inexpensive funding and good underwriting. It can also contract quickly if credit costs rise or dollar-denominated earnings disappoint. The latest rally improves the technical picture but reduces the compensation for taking those risks today.

The operating engine

139 million customers; approximately $17 monthly revenue per active customer and an 83.5% activity rate in Q2. Deepening existing relationships matters more than merely adding accounts.

The credit constraint

Risk-adjusted net interest margin reached 12.4%, but 90-plus-day delinquency rose to 6.9%. A stronger margin and a higher late-stage delinquency ratio can coexist.

The price discipline

The close stands 6.3% above the 50-day average. A bank valued at a premium needs repeatable earnings, not simply a favorable election interpretation.

Sources: Stock Analysis, October 5 closing price and valuation; Nu Holdings Q2 results, released August 13, 2026. Company growth comparisons below are foreign-exchange neutral unless noted.

Ninety-Day Price and Trend Dashboard

Daily trading observations, July 7–October 5, 2026; no prices are interpolated on market holidays. All three series are in USD. KAMA is the 14-period Kaufman adaptive moving average; SMA uses 50 trading sessions. Source: PortfolioAI market-price history. Indicators include preceding price history.

October 5 close and trend reference points
MeasureValueInterpretation
Closing price / daily change$15.18 / +13.03%Sharp reversal from the October 2 close of $13.43.
50-session SMA$14.2742Price is 6.3% above the intermediate trend reference.
14-period KAMA$13.2392Price is 14.7% above the adaptive average after a rapid rebound.
200-session SMA / 14-session RSI$14.6965 / 61.38Above the long-term average; RSI is below the conventional 70 overbought threshold.
Closing-price range in chart$12.23–$15.68September 28 low and September 3 high; these are closes, not intraday extremes.

Company Overview: A Relationship Bank Without Branches

Nu operates the Nubank platform, centered on Brazil with expanding businesses in Mexico and Colombia. Credit cards, accounts, deposits, personal loans, investments and insurance create multiple ways to monetize a customer relationship. The attraction is not just low distribution cost. A primary banking relationship can produce deposits, transaction data and opportunities to sell additional products without rebuilding a branch network.

Brazil remains decisive: almost 118 million of the group’s 139 million customers were there at June 30, approximately 85% of the total. Mexico had 15.8 million and Colombia more than 5 million. Customer diversification is progressing, but the shares remain substantially exposed to Brazilian borrowers, funding conditions, currency and policy.

The mature-market task is deeper engagement; the expansion-market task is profitable replication. Management said Mexico’s full banking operation launched in August, creating a broader product opportunity than the earlier credit-first model. Its reported Q2 revenue per active customer of $12.3, versus $5.6 in Brazil at a comparable stage, is encouraging. It is not proof that Mexico will reproduce Brazil’s eventual profitability: customer mix, regulation, deposit pricing and credit behavior can differ.

Source: Nu Q2 operating and business highlights. Comparative market-stage figures and competitive claims are management’s assessments.

Industry and Market Analysis: Brazil First, Fintech Second

The October 5 move was a reminder that NU is not interchangeable with a U.S. fintech stock. 24/7 Wall St.’s October 5 report linked the rally in Brazil-exposed equities to Flávio Bolsonaro outperforming polls in the first presidential-election round. A runoff remained ahead. The relevant investment distinction is between a lower perceived country-risk premium and higher demonstrated company earnings. An election reaction can deliver the former immediately; it cannot establish the latter.

Nu competes on convenience, product breadth, funding and credit selection—not on an app alone. Established banks can respond with their own digital distribution and deposit offers. MercadoLibre is a useful alternative exposure, but its commerce, payments and lending mix differs materially from Nu’s bank-centered economics. Comparing their headline earnings multiples without adjusting for credit losses, funding and business mix would obscure rather than resolve the choice.

Sell-side expectations: Stock Analysis consensus last updated October 1, 2026
Consensus measureReadingImplication at $15.18
Analysts / aggregate rating22 / BuyBroadly favorable, but not unanimous.
Average / median one-year target$18.64 / $19.0022.8% / 25.2% above the close.
Low / high target$12.00 / $23.00−20.9% / +51.5%; wide uncertainty around value.
September recommendation distribution13 Strong Buy, 5 Buy, 3 Hold, 1 SellA historical monthly snapshot, not 22 new October 5 opinions.

Analyst targets provide an expectations benchmark, not downside protection. The consensus already assumes improving earnings, while the large target range acknowledges meaningful disagreement. NU should be judged against the operating assumptions behind that growth rather than the apparent percentage upside alone.

Source: Stock Analysis analyst consensus and forecasts, dated October 1; target-price comparisons calculated against the October 5 close. Forecasts are estimates, not company guidance.

Technical Analysis: A Rebound That Still Needs Confirmation

The shares rose 11.5% from the July 7 close to October 5, but that endpoint conceals a difficult path. NU fell to a $12.23 close on September 28 before recovering 24.1% in five subsequent sessions. Monday’s close reclaimed both the 50-day and 200-day averages. This is a meaningful repair of the price structure, not yet evidence of a smooth, durable uptrend.

The 50-day average increased from $14.2524 on October 2 to $14.2742 on October 5. A single upward turn after a large gap deserves less weight than several weeks of rising averages and sustained closes above them. KAMA near $13.24 remains well below the price, showing how quickly the rebound outran the preceding trend. RSI at 61.38 supports improved momentum without independently proving that the shares are inexpensive.

  • First confirmation area: the $14.27–$14.70 moving-average zone. Holding that region on a pullback would be more constructive than surrendering it immediately.
  • Overhead reference: the September 3 closing high of $15.68, about 3.3% above the latest close. Sustained trading beyond it would strengthen the recovery case.
  • Failure references: $13.43, the pre-rally close, and then the $12.23 September closing trough. A return to these levels would undermine the immediate breakout thesis.

These are historical reference areas, not guaranteed floors, ceilings or recommended stop orders. Political headlines can produce opening gaps that bypass a planned exit price. The technical argument favors patience after a large one-day move rather than treating the rally as a mandate to chase.

Fundamental Analysis: Credit-Adjusted Growth Is the Real Moat

Q2 2026 company-reported operating and managerial highlights
MetricQ2 resultInvestment reading
Gross revenueNearly $5.9 billion; +39% YoYBroad monetization growth; gross revenue is not interchangeable with net revenue.
Net interest income / margin$3.7 billion / 22.9%Greater lending and unsecured mix lift yield before credit losses.
Cost of credit$1.7 billion; −9% QoQSeasonal improvement helped the quarter; do not assume it repeats every quarter.
Risk-adjusted NIM12.4%, versus 9.5% in Q1A stronger lending spread after credit costs; the essential durability test.
Net income / ROEApproximately $1.1 billion / 33%Strong current profitability, not a through-cycle earnings guarantee.
Credit portfolio / deposits$39.4 billion / $45.3 billionFunding scale supports lending; deposits are customer liabilities, not surplus shareholder cash.
15–90 day / 90-plus-day NPL ratios4.8% / 6.9%Early delinquency improved 16 basis points; late delinquency rose 35 basis points.
Efficiency ratio19.5%, versus 17.6% in Q1Higher quarterly expense intensity, despite improvement from 21.3% a year earlier.

Source: Nu Holdings August 13 results announcement. Growth rates are FX neutral. The release includes IFRS financial results and complementary managerial measures; gross revenue, efficiency and risk-adjusted margin retain the company’s presentation.

Why the margin improvement deserves both credit and scrutiny

Nu’s Q2 credit portfolio grew 37% year over year, faster than deposits at 18%. Lending can grow faster than funding for a period without creating a liquidity problem, but the divergence requires attention. Deposits increased 6% sequentially and the consolidated cost of deposits was 88% of interbank rates. The bank’s funding advantage remains valuable only if that cost discipline survives competition and expansion.

The unsecured-lending mix raises the stakes. Of the $39.4 billion portfolio, credit cards accounted for $26 billion, unsecured lending $10.3 billion and secured lending $3.1 billion. Management attributed higher net interest margin partly to intentional risk expansion. Higher yield is economically attractive only when it more than compensates for eventual defaults, funding and capital consumption. Risk-adjusted NIM therefore carries more analytical weight than headline NIM alone.

Management attributed most of the early-delinquency improvement to seasonality and the late-stage increase largely to migration from Q1. That explanation is plausible but testable: new cohorts must perform as they season, and higher late-stage delinquencies must not become a persistent trend. The low-cost distribution model cannot eliminate a borrower’s inability to repay.

Operating leverage is already visible; expansion still costs money

ARPAC near $17 and an 83.5% activity rate show that Nu is extracting more value from an established customer base. Yet the efficiency ratio worsened sequentially as real estate and marketing expenses shifted into Q2 and international investment continued. A credible long-term thesis allows for that expense volatility instead of assuming every additional customer immediately produces mature Brazilian economics.

NuFormer, the company’s financial-behavior model, is an additional underwriting and service tool. Management says it is deployed in Brazilian and Mexican credit cards and Brazilian unsecured lending. The appropriate valuation benefit is better observed credit selection and service economics, not a separate AI premium awarded before the financial evidence arrives.

Value a bank as a bank

Stock Analysis places NU’s market capitalization at $73.33 billion, trailing P/E at 20.68 and price-to-book at 5.53. Its forward P/E of 15.58 is estimate-dependent. A lower forward multiple is not a discount already earned: it depends on future earnings materializing.

Book value is a useful second check. At more than five times book, the market expects sustained high returns on capital and meaningful reinvestment opportunities. A decline in profitability can hurt both earnings and the multiple investors are willing to pay. Generic enterprise-value or industrial free-cash-flow comparisons are less informative here because deposits, lending flows and liquidity assets are intrinsic to the banking business. Nor should a headline cash balance be treated as freely distributable capital.

Illustrative twelve-month earnings sensitivity

Editorial scenarios, not forecasts or analyst targets; value equals assumed EPS multiplied by assumed P/E
ScenarioIllustrative EPSP/EImplied valueChange from $15.18
Credit stress / valuation compression$0.7016×$11.20−26.2%
Profitable growth / stable premium$0.9520×$19.00+25.2%
Strong monetization / resilient credit$1.1023×$25.30+66.7%

The favorable case requires both earnings delivery and a premium valuation. The stress case shows why customer growth alone is insufficient: even a profitable bank can deliver a poor equity return if credit costs reduce EPS and investors lower the multiple simultaneously. No probability weighting is assigned, and these values exclude dividends, taxes and transaction costs. The middle case offers potential upside, but not such an overwhelming margin of safety that buying immediately after a 13% rally becomes necessary.

Risk Analysis and Portfolio Discipline

RiskHow it reaches shareholdersWhat to watch
Credit expansionHigher-risk lending can increase current revenue before losses mature.Early and late delinquencies, provisions, cohort behavior and risk-adjusted NIM together.
Brazil concentration and politicsCountry-risk repricing can reverse without a change in customer numbers.Runoff developments, fiscal policy, banking taxation and actual funding conditions.
Foreign exchangeLocal growth can translate into weaker USD earnings.Reported dollar results alongside FX-neutral growth; do not treat them as identical.
Deposit competition and liquidityHigher funding costs reduce spread; deposit flows constrain loan expansion.Deposit pricing, loan growth relative to funding, and country-level liquidity and capital disclosures.
International executionExpansion spending can arrive before durable profits.Mexico and Colombia engagement, risk-adjusted revenue and incremental expense trends.
Model, cyber and regulatory riskUnderwriting errors, service disruption or compliance failures can damage trust and economics.Observed loss performance, operating incidents and changes in regulatory requirements.
ValuationA high book-value premium amplifies sensitivity to lower expected ROE.Earnings quality and sustainable returns, not just consensus target revisions.

The U.S. recession discussion supplies a portfolio stress lens, not a forecast of Brazilian loan losses. PortfolioAI’s October 5 recession report argues that weak U.S. hiring leaves less cushion against shocks while expansion remains the base case. For NU, the relevant transmission would be tighter global financing, changing risk appetite and currency pressure. Actual borrower stress must still be assessed in its local markets. NU is a growth-oriented financial exposure, not a defensive cash substitute.

What would change the recommendation?

  • More constructive: subsequent quarters sustain strong risk-adjusted margins, deposit pricing remains disciplined, and customer monetization improves without worsening credit cohorts. A calmer entry price would independently improve prospective returns.
  • More cautious: persistent late-stage delinquency growth, provisions that overwhelm spread gains, or expansion expenses that outpace monetization. A political rally unsupported by earnings should not be mistaken for reduced business risk.
  • Next checkpoints: Brazil’s runoff later in October and the next quarterly release, which Stock Analysis lists for November 12 after market close. The operating release matters more to long-term value than the next headline-driven price gap.

Bottom line: Nu’s scale, engagement and funding franchise justify serious attention. Its price already demands that those advantages translate into durable credit-adjusted returns. Hold the thesis to that standard—and resist using a stronger share price as evidence that the underwriting risk has disappeared.

Sources and Reading Notes

  1. Nu Holdings Q2 2026 results announcement, August 13: June 30 operating metrics, profitability, funding, credit and business developments.
  2. Stock Analysis: NU statistics: October 5 closing quote, valuation and scheduled earnings date.
  3. Stock Analysis: NU analyst forecasts: October 1 consensus snapshot. Price targets are opinions; forward multiples depend on estimated earnings.
  4. 24/7 Wall St., October 5: election-related market reaction. Intraday quotes in that article are not closing prices.
  5. PortfolioAI’s October 5 Reddit watchlist: the NU-versus-MELI research question. Discussion attention is not a buy signal.

As of the October 5, 2026 close; financial results refer to Q2, not an estimate of Q3 performance. Price-series calculations use daily closes; scenario assumptions are illustrative. For informational purposes only, not personalized investment advice. Equity investments can lose substantial value.