Read my Nu Deep Dive here to learn about the company in more detail. More recent content to read:

As you can see in the second chart below, FX had a larger impact on growth than it has in over 10 quarters. The 26-point headwind was a full 7 points worse than anything it has experienced in years. It’s always important to focus on foreign exchange neutral (FXN) growth for the company, but that’s especially true today. Deposits rose 55% Y/Y FXN; ARPAC rose 23% Y/Y FXN; purchase volume rose 20% Y/Y FXN; the credit portfolio rose 45% Y/Y FXN.

Note that the active customer rate fell for the first time in a long time due to Mexico and Colombia expansion. Nu doesn’t have its full product suite in those two nations like it does in Brazil. Furthermore, activity levels always start at low points for its users and build over time. So? Rapid customer growth outside of Brazil is an active customer rate headwind. They’re very confident in this being temporary as they roll more products out to those customers.

Nu’s bank has a $2.1 billion excess capital cushion on top of its $2 billion requirement. The holding company has another $2.2 billion ready to allocate when need be. The balance sheet is in fantastic shape.

Nu trades for 22× 2025 EPS estimates. The chart below says 26x because it’s using next 12 month EPS starting in Q4 2024 rather than Q1 2025. EPS is expected to grow by 33% this year and by 36% next year. Estimates should be pretty flat following this report.

Winning Latin America:

Nu is now the 3rd largest financial institution in Brazil by customer count. Despite having 58% of the entire adult population there as customers, it still delivered 16% Y/Y customer growth to keep that impressive penetration level rising. Deposits also rose at a solid 11% Q/Q FXN clip.

In Mexico, it crossed 10 million total customers thanks to 91% Y/Y growth. It already has 12% of the population in its customer base… 6 years after launching. Casual. Credit card customers rose 70% Y/Y with 10% higher Y/Y approval rates yet a 50% improvement in first payment default rates as its underwriting improves. It maintained its pace of $600 million Q/Q deposit adds while total deposits jumped from $1 billion to $4.5 billion Y/Y. 

In its newest Colombian market, it now has 2.5 million customers vs. 2 million Q/Q, while deposits jumped from $900 million last quarter to $1.3 billion this quarter. It is already top 5 there in terms of deposits. That was fast.

Cost to serve remained at a stellar $0.80 per customer per month. While this rose Y/Y due to more data and processing costs, it’s considerably below incumbents. That’s thanks to Nu’s branchless, tech-driven DNA. It remains fully confident in keeping cost to serve below $1 for the foreseeable future, which is dearly important within a commodity like banking. As I talk about constantly, the ways to durably outcompete the field in banking are with a better, more user-friendly product and cost advantages. Its product is best-in-class. Its cost advantages allow it to profitably offer more yield on deposits, lower rates on loans and more cash back in its marketplace to stand out from the pack.

“Our perspective is that it's still just day one.”

CEO/Co-Founder David Vélez

NIM & Risk-Adjusted NIM Headwinds:

As we work through this section, note that GPM and NIM are very tightly correlated.

NIM falling Q/Q and Y/Y is not ideal. But? The reasons for the decline are the same as in recent quarters and not concerning. First, cost of capital is rising as it rapidly expands into Colombia and Mexico with great success. This led to its cost of deposits rising from 80% of the blended interbank rate to 89% Y/Y.

Nu’s advantages in Brazil are far more developed than in the other two countries. For context, in Mexico and Colombia, it is paying more than 100% of the blended interbank rate. It knew expanding to other countries would have this initial impact, and that is a concession it’s eager to make. You can either optimize NIM in the near term, or maintain strong NIM, expand to the rest of Latin America and optimize down the road. The choice is clear in my mind. This is the correct decision.

Next, NIM is falling because it’s shifting its loan portfolio to more secured credit, with lower risk and lower yields. Furthermore, it has sharpened underwriting parameters a bit on the unsecured portfolio, which is improving borrower quality, lowering risk and lowering yields further. Beyond these items, Brazilian yields falling with rate cuts was a NIM headwind this quarter. Taking all of these items together represents 45% of the NIM decline this quarter. The remaining 55% was related to FX headwinds.

The 60 bps decline in risk-adjusted NIM was related to the same cost of capital deleveraging. As risk-adjusted NIM is NIM - credit loss allowance expense (CLAE), it’s important to dissect puts and takes for CLAE here. There were several offsetting factors at play. On the positive side, higher secured and unsecured borrower quality helps diminish CLAE intensity. That helps offset the hit to traditional NIM.

Conversely, brisk origination growth means more front-loaded loss provisioning, which has a negative impact on CLAE. Next, shifting its credit portfolio from credit cards to more personal loans hurts too. Specifically, 29% of its credit portfolio is now personal loans vs. 20% Y/Y.

Between these three items, cost of risk declined by about 10 bps to offset part of the traditional NIM decline.

More on Credit Health:

The shift to secured and higher credit quality borrowers helped the 15-90 day NPL rate fall from 4.4% to 4.1% Q/Q. Seasonality helped, but the improvement was beyond that typical pattern. Continued quarterly declines in 15-90 day NPL rate led to 90+ day NPL rate falling Q/Q. As you can see below, it continues to outperform its competition in terms of underwriting across every income demographic:

When asked if the same was true for personal lending, here’s what leadership had to say:

“So even though the analysis we presented here on Page 26 applies to credit cards, I expect the same dynamics to be playing out in unsecured loans as well.”

COO Youssef Lahrech

More on Secured Lending:

Payroll lending progress is palpable. Through agreements with the Brazilian Institute of Social Security (INSS), SIAPE (public payroll service for public workers), the Brazilian Army and a few other public partners, it now has payroll lending coverage for 70% of its addressable market. It has been hard at work on adding table stakes tools like loan portability and refinancing to drive adoption. Under Brazil’s open banking program, Nu is essentially able to tap into participating bank data to see what they’re charging customers for loans. It can then easily offer better terms to refinance that credit at better rates. Having loan portability and refinancing capabilities were vital prerequisites paving the way for this value creation. It thinks these tools have helped it quickly rack up 16% share of secured INSS payroll loans.

“We expect that just relying on the INSS and SIAPE collateral agreements, we will see an acceleration of our originations in the loan portfolio throughout 2025… in January 2025, we have seen historical high originations for public payroll loans and secured lending in general.”

CFO Guilherme Marques do Lago

As a reminder, while secured comes with lower ROE than unsecured, secured lending growth will not be an ROE headwind. These originations are not happening instead of unsecured or interest-earning credit card issuance. It’s using excess cash to fund this growth, which is NIM and ROE-accretive.

Balance Sheet Optimization:

Over the last few years, Nu has aggressively shifted to installment-based, interest-earning credit to maximize net interest income (NII). Its interest-earning portfolio (IEP) rose by 75% Y/Y FXN to give an idea of this expansion. At the same time, on a Q/Q basis, two things led to IEP falling from 28% to 27% of its credit card portfolio (23% Y/Y). First is strong seasonal growth for other non-interest-earning balances. Secondly, as discussed last quarter, it’s preemptively pulling back on PIX-style financing, which has been a key source of the IEP proliferation to date. As a reminder, PIX is essentially a nationalized Venmo in Brazil. Customers can borrow from Nu to conduct transactions similarly to a credit card. More on the reason for this pullback in the next section.

The overall credit portfolio rose by 45% Y/Y FXN. Within it, credit card receivables rose 28% Y/Y FXN to $14.6 billion; lending more than doubled Y/Y to $6.1 billion.

Balance sheet optimization is inherently tied to our other discussion on NIM and risk-adjusted NIM. If you’re extracting more yield from the same assets, NIM rises. That’s what balance sheet optimization aims to accomplish. While it has made great progress here over the last few years, it does not think it’s close to its NIM ceiling. Much more progress coming as it pushes its extremely modest 39% loan-to-deposit ratio higher.

More on the PIX Pullback:

The PIX hesitancy has nothing to do with credit performance. It has everything to do with driving better product-market fit and limiting friction for some of its less affluent cohorts. Following extensive testing, it reduced eligibility for lower credit bands to “address second order impacts related to NPS, churn and engagement.” It now has 12 tests live and is enjoying strong early results.

“It's important to highlight the overall profitability and resilience of the PIX…. I'm encouraged about resuming growth in terms of percentage of our credit portfolio over the coming quarters. I don't want to create any expectations that this resumption of growth will necessarily happen in the next one or two quarters.”

CFO Guilherme Lago

Between this retooling and “more challenging macro in Brazil,” it expects PIX to remain stable as a percentage of its overall portfolio for a few quarters (still will grow on an absolute basis). It should continue growing as a piece of the pie thereafter to keep optimizing the balance sheet with higher interest products like this one.

“If we were optimizing for short term profitability, we would continue growing significantly. But here we have a choice to make, which is prioritizing short term earnings over improving in some ways the user experience.”

CEO/Co-Founder David Vélez

ARPAC Headwinds:

Aside from FX challenges, growth in Mexico and Colombia is an ARPAC headwind too. This is because its product suite is less developed in those nations vs. Brazil, and most of its demand outside of Brazil is still for its savings accounts. That is among its lowest ARPAC products. Again, ARPAC rose 23% Y/Y FXN despite this. At the same time, Nu sees deposit growth in both nations as setting the stage of highly sticky customer growth. It’s confident in rapid credit cross-selling down the road. It did the exact same thing in Brazil, and we can see how well that worked out for the firm.

Macro:

As I’ve spoken about recently, macro risks in Brazil (and most of Latin America) are mounting'; that matters for Nu. Fortunately, it is “battle-tested” and always approaches credit underwriting the correct way. When it prices a loan, it is overly conservative with future assumptions and provisioning. That has served it well – including in this quarter while macro in Brazil was anything but favorable. Furthermore, it doesn’t rely on capital market liquidity to fund its growth. It has all the cash it needs. Capital markets routinely shutter amid poor macro regardless of repayment trends. They panic and ask questions later (which to be fair is often the correct decision). Nu doesn’t have to deal with this. It also reminded us that its loan durations are much shorter and loan sizes are much smaller than competitors. This will help it be more agile as macro headwinds potentially build.

“We've actually gone through severe recessions throughout our ten plus year history. So we have both a battle tested portfolio and a battle tested kind of underwriting framework around that.”

COO Youssef Lahrech

More Notes:

I thought this was a rock-solid quarter across the board. Growth remains stellar, gross margin is resilient thanks to strong underwriting and despite rapid geographic expansion. Net income margins continue their brisk ascents and the balance sheet is beautiful. Credit trends are strong and the PIX pullback has nothing to do with delinquency issues. It keeps outperforming everyone in terms of pricing risk. Nu continues to take more market share everywhere it operates, with reason to believe it will do the same in its telecom and ecommerce pushes. Mexico and Colombia have clearly turned into two highly compelling growth stories and more launches across the globe are inevitable. It has a massive runway to keep thriving and a world-class team to do just that. 

Sounds good to me. But? FX headwinds held everything back this quarter. That will continue to happen from time to time and will be noise that I try to tune out. Currency fluctuations from quarter to quarter have absolutely nothing to do with Nu’s long-term trajectory and ability to compound profits for decades to come. That’s what I really care about. When contextualizing that noise, everything that I want to look good looks great. I will not punish them for their currency weakening against the dollar… just like I won’t celebrate when growth looks abnormally good in the future due to currency favorability.

Pleased.