People
The People
Governance grade: B–. A founder with $11B of personal Nu stock and a credible bench (ex–Brazil central-bank chief, ex–Square comp chair, ex–PayPal president), offset by a hard-coded super-voting structure that gives David Vélez 74.4% of the vote on roughly 19% of the economics — and lets him veto major capital actions even if his stake falls to 5%.
1. The People Running This Company
Nine senior executives. The decisions that move the P&L sit with three of them: Vélez (strategy/capital), do Lago (CFO and the credibility relay to the market), and Chanes (the Brazil P&L, ~80% of revenue). The new external hires — Campos Neto, Young, Eismann — are the genuine swing variables on whether Nubank can keep operating-leverage gains while building a global stack.
Why this team passes the trust test. Vélez has run NU for 13 years through the IPO, two LatAm launches, and a credit cycle. do Lago has owned the CFO seat since 2021 — long enough for the market's modelling assumptions to be tied to him personally. Chanes (Brazil) and Young (CTO) are recent appointments but bring proper enterprise-scale resumes (McKinsey/Itaú, Google/Snap/Amazon).
Where it gets uncomfortable. Hiring Roberto Campos Neto — Brazil's central-bank president until December 2024 — as Executive Vice-Chairman seven months after he left office is a textbook revolving-door optic for a regulated bank that holds the Central Bank's payment-license and capital framework. It is legal and was announced publicly; the question is whether outside shareholders feel comfortable that a former regulator's first private-sector role is the largest entity he previously oversaw. Cristina Junqueira is also still listed as "Chief Growth Officer & US CEO" while the US bank has not yet launched — succession into a real operating P&L remains theoretical.
2. What They Get Paid
Nu discloses aggregate compensation only — a Foreign Private Issuer exemption Vélez has used since IPO. No individual named-executive-officer numbers, no Form 4 filings, no Say-on-Pay vote. For FY2025 the company says aggregate compensation of all directors and key management was US$91 million.
Aggregate Comp ($M, FY25)
% of Revenue
% of Net Income
People Covered
Is the pay sensible? At $11.2B revenue and $2.87B net income, $91M of aggregate comp is 0.8% of revenue and 3.2% of profit — well below US bank-fintech peers. The 80%-plus stock weighting (per Nu's own 2022 clarification of the Contingent Share Award structure) means most of the headline number is at-risk equity that pays out only on hitting CSA performance hurdles. The honest critique is not that pay is excessive; it is that outside shareholders cannot see who gets what, so they cannot verify whether the CFO/CRO/Brazil-CEO are paid commensurate with the value they create.
3. Are They Aligned?
This is the section that decides the file. The arithmetic of alignment at Nu is unusually asymmetric.
Control versus alignment
Class B shares carry 20 votes per share and are not publicly traded. Vélez holds 88.3% of Class B through Rua California Ltd. (which he owns 100% of) plus a small spousal block. Junqueira holds another 11.7% of Class B via trusts. Combined founder voting power: ~84%. Outside Class A shareholders — collectively 79% of the economics — hold ~14% of the vote.
The shareholder agreement also entrenches Vélez's control past the point where his stake might fall: as long as he beneficially owns 5% of the vote, he retains the right to nominate one director and veto debt issuance beyond consolidated net equity. Below 25% he still gets two seats; below 40%, three. This is hard-coded; it is not a renewable agreement.
Controlled-company status. With 74.4% voting power Nu qualifies as a "controlled company" under NYSE rules and can opt out of requirements for a majority-independent board and a fully-independent compensation committee. In practice Nu has chosen to keep an 8-of-9 independent board and an independent comp committee anyway — but the right to drop those protections sits with Vélez, not the board.
Insider activity — sales, not buys
There has been no open-market insider buying disclosed in the period covered. Vélez's August 2025 33-million-share sale (~$462M at the day's $14 price) was the largest single insider transaction and triggered the 13G/A amendment showing Rua California's Class A stake stepping down. Junqueira added a ~$4M sale in March 2026. The April 2026 round of small "tax-withholding" dispositions on RSU vesting is mechanical and not signal-bearing on its own.
Read. The dollar magnitude is small relative to Vélez's remaining ~$11B stake — diversification, not exit. But the directional asymmetry (founders sell when they monetize equity; they never add) is normal for a still-young IPO and gives the bear case its only real ownership-trend talking point.
Dilution
Stock-based compensation under the 2020 Omnibus Plan and the founder Contingent Share Award programme remains the principal source of new Class A share issuance. The CSA was structured to pay Vélez additional shares only if performance hurdles are met; the rest of the team receives annual RSU grants. Total share count growth has been modest relative to the equity comp percentages typical at LatAm fintech peers — the bigger dilutive risk is the convertibility of Vélez's 901M Class B shares, which on conversion would become Class A but would not increase the total share count; it would, however, eventually collapse the voting structure on transfer.
Related-party behaviour
The 20-F discloses a registration-rights agreement with early holders, indemnification agreements for officers and directors that are broader than Delaware norms, and a related-party transactions policy updated in March 2024 requiring board or Audit & Risk Committee pre-approval. No material related-party contracts (consulting fees to founders, controlled-vendor agreements, family employment) are disclosed for FY2025. Sequoia (Doug Leone is still on the board) reduced its 13G position by 57% in May 2025 — that is a fund-cycle exit by an early backer, not a self-dealing event, but the optics of an insider VC unwinding while public holders sit through a 28% YTD decline deserve flagging.
Capital allocation
The board has not authorised a buyback. FY2025 net income of $2.87B has been redeployed into LatAm credit growth and the US/Mexico bank build-out. Capital allocation discipline rests entirely on Vélez and do Lago since the board has no meaningful counterweight on routine decisions.
Skin-in-the-game score
Skin-in-the-Game Score (1–10)
8/10. Vélez and Junqueira together hold ~$12.5B of NU equity at the current $12.18 price — among the highest absolute founder stakes of any newly-listed bank globally. The score is penalised two notches because (a) the dual-class structure means economic skin is decoupled from voting skin in a way that disenfranchises Class A holders, and (b) the only insider trading flow disclosed is selling, not buying, even after a sharp YTD decline.
4. Board Quality
Nine directors. Eight are independent under NYSE rules. Average age 56. Six committee chairs/members include a Brazilian CPA running Audit & Risk and a US bank CEO running Compensation & People — high-quality choices for a regulated LatAm bank.
Scores: 3 = deep expertise, 0 = none.
What the matrix shows. Two genuine banking-regulator brains (Calderón, Moreno) plus one ex–wealth-management COO (Sands) cover the audit and regulatory burden. Two deep engineering operators (Pham, Marcus) plus a recent additon (Piacentini, ex-Amazon International) give product-scale credibility. The gap: only Reses has hands-on US bank-charter experience — relevant because the company is now staffing up to launch a US bank. Outside Vélez/Junqueira, nobody on the board operates a LatAm-domiciled financial institution today, which matters for things like Brazilian central-bank examination handling and Mexico licensing posture.
Can this board challenge management? Probably yes on most operational decisions — Reses, Sands, and Calderón have all chaired or audited regulated entities and would not rubber-stamp risky underwriting. But on strategic decisions (M&A, where to deploy $2.9B/yr of profit, when to convert Class B shares), the shareholder agreement gives Vélez veto rights below even a 10% threshold. The board can advise; it cannot override.
5. The Verdict
Skin-in-Game (/10)
Founder Voting %
Indep. Directors (of 9)
Governance grade: B−.
The case for trust. A genuinely founder-led bank, 13 years into the build, with the founder's personal $11B stake still on the table. A senior management bench of ex–McKinsey/Itaú/HSBC/Google operators that grew through the IPO, the 2022 LatAm risk re-rating, and a recent profitable normalisation. A board that — on paper — has more independence and more relevant expertise than the controlled-company status would require. Compensation is heavily equity-weighted and well under 1% of revenue. There are no disclosed related-party shenanigans, no compliance restatements, no whistleblower events, no SEC or CVM enforcement matters.
The case against trust. Outside shareholders own ~80% of the economics and ~14% of the vote. Vélez's veto rights on debt, M&A, and major actions persist down to a 5% stake. Foreign Private Issuer status means no individual NEO comp, no Say-on-Pay, no Form 4. Cayman incorporation plus broad indemnification provisions make derivative litigation nearly impossible. Insider activity since IPO is sell-only, and the August 2025 $462M Vélez sale set the public-market tone for a stock that has fallen 28% YTD. Hiring Brazil's just-departed central-bank president as Executive Vice-Chairman is operationally smart but a regulatory-optics liability.
Upgrade trigger. Voluntary collapse of the dual-class structure on a defined sunset (e.g., the seventh anniversary of the IPO is December 2028), or the appointment of a fully independent Lead Director with explicit board-level authority to call meetings without the chairman, would re-grade the file to B+ / A−.
Downgrade trigger. A material related-party transaction (Rua California financing arrangement, founder-affiliated vendor contract), accelerated Vélez selling beyond diversification scale (more than 5% of remaining stake in any 12-month period), or any regulatory action involving Campos Neto's prior central-bank decisions that benefited Nu would drop the grade to C+ / C.