How often are Berkshire Hathaway and Cathie Wood invested in the same thing?

Today, we’ll be telling the story of one such company: Nubank, going through its incredible growth, still promising prospects, and trying to reconcile what it means for the company’s valuation.

Nubank is a digital bank that has literally changed the financial system in Brazil, upending decades of dominance by entrenched legacy banks that were notoriously not customer-centric, overly bureaucratic, and left large swaths of the population unbanked.

To say the company started from nothing in 2013 wouldn’t be an exaggeration, and yet, in just over a decade, it has grown to have more than 100 million customers, with 60% of Brazil’s adult population using at least one of its financial products.

The business is just now starting to flex its true earnings power a few years after IPOing, but much growth looms should they be able to continue expanding into countries like Mexico, Colombia, and perhaps the rest of Latin America.

More on one of the most promising, but also riskiest, companies we’ve explored — below.

— Shawn

Join a Community of Like-Minded Investors

In recent weeks, we’ve announced our Intrinsic Value Community, and the response has exceeded all expectations.

The Intrinsic Value Community is a place for sophisticated long-term investors to exchange ideas, challenge each other’s thinking, and build meaningful connections with like-minded individuals.

We give each other feedback on our latest research, post our best ideas, and bring in guest speakers, from industry experts to investment professionals, who share actionable insights.

We’re currently reviewing applications for a handful of remaining spots and will begin closing the process soon. If you’re interested in joining, now’s the time to apply!

Nubank: Banking on Latin America

The Purple Revolution

When Daniel and I first started researching Nubank, we thought we would find an exciting story, but one that was ultimately too hard to understand properly. After all, financial institutions can have hidden risks that can blow up their businesses. With the possibility of bank runs and operating in a developing country with high inflation rates, political instability, and a volatile currency, nothing about it looked like a simple investment.

But as we dug deeper, we saw that Nubank represents much more than we ever imagined. It's not just a digital bank; it's a full-stack financial services company that has built one of the largest (and most loyal) customer bases in Latin America, doing so with stunning efficiency.

Furthermore, it operates in one of the most structurally attractive banking markets in the world — a market defined by high spreads, high fees, and underserved customers. Nubank didn't just find product-market fit; it ran through the open door of customer dissatisfaction with a model built for scale, speed, and loyalty.

Still, it’s a company facing serious questions: Can it maintain credit quality as it scales? Can it sustain 30%+ returns on equity as competition heats up? How much does Brazil's volatile macro environment factor in? And what happens when the low-hanging fruit of Brazil's unbanked population is gone, and Nubank must compete head-to-head for wallet share among wealthier customers with traditional banks and new fintechs?

The State of Banking in Brazil

Understanding Nubank starts with understanding Brazil's banking environment a decade ago.

Banking in Brazil was dominated by a few large players. Think oligopoly: five major banks, all charging exorbitant fees for basic services like credit cards and checking accounts, the worst customer service imaginable, and interest rates on unpaid credit card balances frequently hitting north of 400%. I’ll say that again, 400% APRs on credit cards — roughly 20x the average rate in the U.S.

Opening a bank account often required going into a branch, waiting in line, filling out paperwork, and dealing with bureaucratic hurdles that felt frozen in time. And the process for freezing a card could be even more tedious. Meanwhile, roughly 30% of Brazilians didn’t even have a bank account, in a country with 2/3 the population as the U.S.

So, for tens of millions, the banking system wasn't just expensive and inefficient; it was exclusionary.

Into that environment walked David Vélez, a Stanford-educated, Colombian-born, venture capitalist who discovered the opportunity to disrupt a country’s financial system after working with Sequoia to expand into Latin America and learning firsthand how painful the customer experience with Brazilian banks was.

He quickly noticed that Brazil had massive smartphone penetration, a young population frustrated with traditional banks, and, as mentioned, a huge market of unbanked and underbanked consumers, paving the way for a nimble digital-first competitor to quickly enter the market and gain traction.

If you could build a mobile-first, low-cost banking product, you could win big.

The result was Nubank.

Nubank didn’t invent digital banking. But it was one of the first to realize that in emerging markets, the combination of smartphone ubiquity and customer frustration created an opening wide enough to build a generational company. While Western fintechs battled over features in already-banked markets, Nubank went where the need, and the margin, were greatest.

Nubank’s Wedge: Credit Cards

So, how did they do it? Rather than offering a full bank account upfront, Nubank started with a wedge product: a no-fee, digital-powered credit card.

It was brilliant for a few reasons:

• Credit cards were so heavily over-monetized in Brazil, customers were paying annual fees just for the privilege of having a card. A no-fee card immediately offered huge value to prospective users.
  

• By underwriting differently, using alternative data like cell phone usage, Nubank could more safely lend to customers that traditional banks ignored.
  

• The app-based experience was miles better than anything incumbents offered. Sign-up was fast, transparent, and seamless. Customers could manage the entire experience from their phone.

Despite the untraditional approach to lending, Nubank hasn’t necessarily taken undue risks to democratize finance in Brazil:

Customers loved it. The purple Nubank card became a status symbol — a way to signal, "I'm not getting ripped off by the big banks anymore." The start of a customer-first revolution, you might say.

From there, Nubank expanded into digital checking accounts, personal loans, investments, insurance, business banking, and more, branching out from being just a credit card company to a diversified (digital) financial services business. This journey was necessarily determined by laws at the time, which were meant to limit new competition by inhibiting new companies backed by foreign investors from acquiring banking licenses.

Thus, they started with credit cards, until they reached a scale where regulators couldn’t easily dismiss their request.

In short order, Nubank went from a trendy, if not groundbreaking, credit card to the trusted primary financial relationship for tens of millions of Brazilians who had never used another financial institution before.

And once Nubank had a foothold in a customer's financial life, expanding the relationship became much easier, and much cheaper, than competitors reliant on paid acquisition. Even today, the vast majority of Nubank’s customer growth comes entirely organically from user referrals, as opposed to paid marketing.

The Engine: Low-Cost Scale

One of the most striking things about Nubank is its cost structure.

Traditional banks in Brazil had sprawling branch networks, with tens of thousands of employees, and massive fixed costs. Nubank built everything digitally from day one, whereas legacy banks had a patchwork of different IT systems implemented asynchronously over several decades.

As a result of this late-mover advantage, and not being bogged down by physical branches or outdated IT systems, Nubank’s average cost to serve a customer is less than $1 per month. That’s estimated to be 85% lower than Nubank’s peers in traditional banking.

This cost advantage is durable; It’s a function of architecture, not just scale.

Low cost-to-serve allows Nubank to do something traditional banks struggle with: profitably serve low-income and lower-credit customers. In a country where financial inclusion was often a marketing slogan rather than reality, Nubank made it profitable at scale.

Another important point: Nubank has an 83% monthly activity rate. That’s an absurdly high level of engagement for a bank. It's what you'd expect from a consumer tech company, not a financial institution, and it’s telling of how much people love to use Nubank’s intuitive interface.

In Brazil, though, Nubank has already largely tapped its original addressable market, and now it wants to scale up the value chain at home, appealing to increasingly wealthier demographics.

On that point, we know that Nubank's mature customer cohorts show predictable monetization patterns. Early cohorts start low (around $5/month average revenue per active customer, aka “ARPAC”), but over a few years, they ramp to over $25/month. Once customers trust Nubank with one product, cross-selling becomes easier and extremely profitable.

If you believe that Nubank can simply converge the ARPAC across its customer base more with the revenue generated by its most mature customers, and traditional banks that have ARPACs north of $40 per month, then that alone would be enough to be very bullish on the company’s earnings prospects.

In other words, this cohort behavior gives Nubank a long runway of internal growth, even if new customer acquisition slows. As older cohorts "mature" into higher ARPAC brackets, Nubank’s revenue base becomes more resilient and more valuable.

Importantly, Nubank’s customer acquisition cost (CAC) remains low, around $5 to $6 per customer, meaning payback periods are fast, and as such, Nubank’s LTV/CAC ratios are among the best in the world for financial services. Meaning, it gets considerably more value from new customers than what it spends to acquire them.

Unit Economics: Now vs. Later

In 2024, Nubank delivered $1.97 billion in net income on $11.5 billion of revenue. That is…good(!) for a company that went public just over three years ago.

But what about other measures, like return on equity, that matter more for banks? Well, Nubank’s ROE comes in at an impressive 29%.

How about Nubank’s efficiency ratio (another critical measure of banks’ performance)? Again, Nubank leads its peers with a nearly 30% efficiency ratio.

For comparison, traditional Brazilian banks like Itaú and Bradesco historically posted ROEs in the 15-18% range, with efficiency ratios closer to 50%. FYI: Lower efficiency ratios are better; a bank’s efficiency ratio is calculated as: Non-interest operating costs / (Net Interest Income + Non-interest Income - Provision for Credit Losses). You want that to be lower because it means that overhead “non-interest” costs make up a lower percentage of operating revenues.

So, Nubank isn’t just "catching up" to incumbents in returns on equity and efficiency. It’s outperforming them.

That said, much of Nubank's revenue is still credit card-based. About half of its interest income comes from revolving credit. That’s a major risk.

New regulations in Brazil are capping how much interest can be charged on unpaid balances, and over time, credit card APRs will likely come down closer to global norms as Brazilians build out their credit history and incomes continue to rise. That also typically means lower default rates, as Daniel pointed out to me in our podcast together, offsetting some of the negative effects of collecting less interest, but still, given how elevated lending rates are in Brazil, rates will probably come down much faster than defaults in the coming years.

Nubank's strategy is already shifting, though. The company is growing its secured lending book (up 615% year-over-year), namely via payroll advances, and is attracting higher-end customers with its products like the Ultravioleta card, which comes with annual fees but higher rewards and a dedicated travel portal.

These secured lending products and loans to wealthier borrowers offer lower yields, but also lower credit losses, which is a necessary evolution as Nubank matures.

Additionally, Nubank’s non-interest income is growing as well. Interchange fees, insurance premiums, and investment platform fees are revenue streams that provide diversification away from pure credit spread dependency.

International Expansion: Mexico and Colombia

Nubank’s growth story in Brazil has been extraordinary, but its ambitions are far bigger.

Mexico and Colombia are the next frontiers, and thereafter, the rest of Latin America, at least that’s what the bulls hope. And while early signs are encouraging, success isn’t guaranteed.

In Mexico, Nubank already has more than 9 million customers. That’s an astonishing number when you consider that Nubank only entered the market in 2020, starting with the same wedge strategy using a no-fee, digitally-powered credit card.

On that point, Mexico actually shares many traits with Brazil: a highly concentrated banking sector, underbanked consumers, and widespread dissatisfaction with traditional institutions. But it's not a carbon copy. Credit bureaus are less reliable and fraud risk is higher, yet median incomes are higher, and per capita credit card usage is lower (24% of Mexico’s population use credit cards versus 40% in Brazil).

Nubank’s early traction suggests that the core insight — that a better product wins — holds true, but competition is fierce. Thousands of fintechs, and existing banking giants, are vying to better serve the Mexican market.

Colombia, while a smaller market, presents similar dynamics. As of early 2025, Nubank had over 1 million customers there. In a country of 50 million people, that's an early but meaningful footprint.

If Nubank can replicate even a fraction of its Brazil success in Mexico and Colombia, the payoff could be enormous. Combined, these markets could eventually double Nubank’s addressable market. And we haven’t even started talking about countries like Chile, Argentina, or Peru yet, either.

But scaling internationally isn’t just a matter of copy-pasting the Brazilian playbook. It requires localized product development, compliance with different regulatory regimes, and building brand trust from scratch.

Competition is fierce, but so is the upside, and Nubank has data, cost structure, and digital proficiency supporting it. Let’s just put it this way: I wouldn’t want to bet on any of the competitors who have to go against Nubank.

Credit Quality

Let’s not get too carried away with optimism, though. Mark Twain once described education as “the path from cocky ignorance to miserable uncertainty.” So, let’s embark toward miserable uncertainty for the sake of skepticism and learning.

One of the biggest bear cases against Nubank is that its credit performance could unravel at scale. Sometimes, people call this a “blow up,” and fast-growing financial businesses are, historically, a red flag — a sign that a company isn’t properly pricing credit risks. A few weeks ago, we talked about how Comfort Systems could juice growth by making unprofitable bids on construction projects, where sales would jump but profits would actually contract.

It’s sort of the same concern here. If Nubank isn’t properly underwriting consumer credit risk across Latin America with its credit card lending programs, they may be growing rapidly today, only to find said growth was built on a house of cards, as defaults surge during an economic downturn.

The argument goes: when you're underwriting millions of first-time borrowers, you're taking enormous hidden risk. As long as the macro environment is benign, everything looks fine. But when the cycle turns, defaults will spike, and lending profits will evaporate.

It’s a fair concern, and I think the weakest part of any bullish thesis on Nubank. But so far, the evidence doesn’t support it.

Nubank’s 90-day delinquency rate (% of people who haven’t made a payment in 3 months or longer) sits around 7.0%, which is high and recently has trended higher, but when you’re charging APRs north of 200% and 300% credit cards, it’s hard to say you are underpricing risk. You can still collect more than enough interest from those who pay to offset those who can’t.

And Nubank’s lending strategy — starting customers with low limits, aggressively managing exposure, and dynamically adjusting credit based on real-time repayment behavior — seems to be working.

Moreover, as we already discussed, Nubank’s mix is gradually shifting toward lower-risk products like payroll-secured loans, which should further stabilize credit quality.

Still, the real test will come in a deep recession. We haven’t seen how Nubank’s portfolio performs through a true credit crunch. If Brazil’s unemployment rate spikes or if inflation remains sticky, losses could rise.

Credit is the fulcrum risk for Nubank, and Nubank is racing to diversify its business. If losses stay contained, Nubank’s returns will likely remain extraordinary. If not, the entire bull case would be in peril.

With that said, Nubank isn’t making loans hand over fist. Its loan-to-deposit ratio is just 40% (for every $1 of deposits, it makes 40 cents in loans.) That’s abnormally low, as a more typical number would be 70-80%.

But this number is artifically low for two reasons: A regulatory kink in Brazil, where Nubank gets to sit on funds for longer as credit transactions don’t have to be settled for 30 days, meaning merchants don’t get paid until a month after something was purchased on credit (in the U.S., merchants receive funds in 1-3 days usually), and the fact that Nubank has accepted a lot of deposits from Mexican and Colombian customers, but isn’t lending as aggressively there yet as it does in Brazil.

Regardless, the point is that Nubank doesn’t appear to be recklessly lending to their maximum capacity to do so.

And then, when we look at Nubank’s net interest margin, it is again very inspiring. Net interest margin is calculated by dividing a bank’s net interest income, so the interest they earn on loans minus the interest they pay depositors, by the average amount of interest-earning assets they have.

Nubank’s net interest margin was around 17% last year, and I cannot emphasize how massive this is. The global standard is more like 3-5%, and the Brazilian banking industry is maybe on the higher end of that average, but still, Nubank’s net interest margin is an anomaly.

That leads into a whole conversation about trying to adjust net interest margin for the lending risks taken, since we know Nubank takes on more consumer loans than the average bank, but even then, by some measures, its risk-adjusted NIM is well above competitors, suggesting they’re lending at higher rates on average and managing the higher risks of that lending well.

The Competition Aren’t Luddites Anymore

Nubank’s early years benefited from a massive opportunity where incumbents were asleep at the wheel, and they could never respond as nimbly or as effectively as a Silicon Valley-powered fintech startup that quickly achieved widespread adoption under their noses.

A decade later, legacy banks have responded and are continuing to do so tactically. And while still nimble, Nubank is no startup anymore, either.

Traditional banks like Itaú, Bradesco, and Banco do Brasil have launched their own digital-only offerings. Banco Inter has grown rapidly into a credible challenger. C6 Bank, backed by JPMorgan, is investing heavily to grab market share. And Mercado Pago, the fintech arm of Mercado Libre, is aggressively expanding from merchant payments into consumer banking.

Competition isn’t just increasing; it’s professionalizing, and consumers have been the winners. Customer service, banking apps, and financial inclusion have dramatically improved nationwide in Brazil over the last decade. Looking forward, the financial players in this game aren’t necessarily guaranteed to be winners, at least not in the way they have been.

Yet, Nubank’s biggest moat is its cost structure, being digital first, but it doesn’t hurt that the company is beloved. Net Promoter Scores are an international metric to assess customer satisfaction and loyalty, and on that front, Nubank’s is reportedly 3x higher than incumbent banks and other major fintechs.

Still, competition will likely compress margins over time. It will also raise customer acquisition costs. Nubank’s organic growth flywheel — the word-of-mouth engine that powered its early rise — may slow.

Defending its position will require relentless reinvestment in product quality, brand strength, and customer experience. In other words, the playbook that got Nubank to 100 million customers won't be enough to win the next 100 million.

Valuing Latin America’s Biggest Fintech

Now, the question is: how does this picture come together in Nubank’s valuation, and does it make sense to add Nubank to our Intrinsic Value Portfolio?

Nubank doesn’t screen cheaply on traditional metrics, and certainly not compared to other banks. In fact, at a 27x P/E multiple, it trades at a premium to virtually every other bank in the world. But Nubank isn’t a traditional bank.

It has grown revenue at 75%+ per year over the last 5 years, has expanded margins along the way, and is compounding equity at 30%+ annually. It’s a category-defining platform with a long runway in massive, underpenetrated markets, and with considerable opportunity to cross-sell existing customers on more products while moving up the value chain, i.e., offering higher-margin services targeted at higher-net-worth customers.

If Nubank can sustain even half of its current growth rates (or even a third for that matter) over the next five years, while maintaining ROEs north of 20%, today's valuation will look fair, maybe even cheap.

Between continued growth in Brazil, expansion into Colombia and Mexico, and eventually other markets, and higher ARPACs as Nubank’s customer base matures, I have no trouble underwriting a 20% CAGR in revenue over the next 5 years, and really, 30% might not be that aggressive.

And that leads into the point that, while Nubank reports its results on the New York Stock Exchange in dollars, and dollar returns are what U.S. investors are concerned with, Nubank earns in Mexican/Colombian Pesos and Brazilian Reais. These are countries with much less stable economies and much higher inflation, and as such, their currencies tend to decline against the dollar over time.

In the last year alone, the Brazilian Real has depreciated by roughly 20% against the U.S. Dollar, and that’s a major headwind to foreign investors. Nubank’s growth, in local currency terms, is actually even higher than its already astronomical results.

Between these currency risks, the political risks associated with Brazil, which has faced serious challenges to its democracy and rule of law, on top of Nubank being a financial business taking explosive credit risks, we’re going to need a very high discount rate to value this business.

A discount rate reflects the required return you need to take a certain risk. For Treasury bonds, that rate is 4-5% — there’s little risk, if any, of not being repaid. For the typical large-cap stock, we usually use an 8% discount rate, reflecting an equity risk premium of 3-4%, which isn’t as high as it could be because there’s less risk of total capital loss with companies like Nike, Airbnb, and Alphabet.

With Nubank, however, I went with a 12% discount rate, reflecting an even higher risk premium for investing in a speculative company operating in an area with elevated currency and political risks.

With those assumptions, I assume Nubank can increase its ARPAC by 50%, and grow its active customer base by about 50 million, leading to revenue growth of 20% a year by 2029. I also assume their margins can improve slightly as they continue to scale and focus on wealthier customers.

These are somewhat optimistic projections, but not nearly as aggressive as what many analysts forecast for them. And, for further conservatism, I use a weighted-average exit multiple of roughly 22, assuming the company’s valuation multiple contracts from 29x earnings today as its growth slows in the coming years, which lowers our target buy price.

So, having said all that, I get a plausible target price range to buy the stock at between $9.50 (or lower if possible) and $12 per share. At the bottom of that range, the estimated return is over 19% a year, and at the top of the range, the expected return is closer to 13% a year — a reminder that entry price matters.

Depending on one’s risk tolerance and patience, an investor may be more or less inclined to initiate a position at the top of that range, which is where shares currently trade. (Just a reminder, as always, these are subjective estimates made for educational purposes, not financial advice!)

This is all speculation. Still, it’s the highest expected returns we’ve seen yet, for a stock that has recently traded between $10 and $12.50 per share, overlapping with that target buy range. And, I don’t think the model assumptions are as aggressive as they seem. For anyone who has actually looked at the company’s results over the last decade, they speak for themself.

In fact, some analysts think the company can achieve an ARPAC as high as $30 (more than twice what I estimate).

Conclusion

Nubank is very, very promising, but it’s not for the faint of heart. Regulatory crackdowns on lending in Brazil could disrupt Nubank’s golden goose, economic downturns could trigger waves of defaults that do serious damage to the company’s balance sheet, and even in less extreme scenarios, local currency weakness will likely continue to weigh on returns and become more consequential as growth slows generally.

This would be, without question, the riskiest business we’ve added to the portfolio, though I do believe there will be chances to buy this stock at prices that more than compensate for the risks taken.

This is a stock that is too risky to get a 5% weighting at current prices, so after chatting with Daniel, we settled on adding a 2% stake today at around $12 per share, and potentially growing that to a 5% position if the stock drops back toward $10 per share (where the risk/return profile becomes more appealing.)

This is well beyond our circle of competence, and for that reason alone, we should probably pass. Chalk the decision to move forward, then, up to youthful optimism.

After having dug into Nubank, though, I don’t believe the company’s results are “too good to be true,” and without overthinking things, I can see how this would look like an obviously attractive opportunity in hindsight.

It’s a company that has such compelling prospects that I’d be inclined to make it a small position if for no other reason than that we have an incentive to continue closely tracking it. However, buyer beware.

You can listen to our full episode on Nubank here, and access our valuation model of the company here. More updates on our Portfolio below.

Weekly Update: The Intrinsic Value Portfolio

Notes

• Nubank joins the portfolio at an average price of $11.90, with a 2% stake, making it tied for being our smallest holding with Nike. For both investments, we’re waiting for prices to reach a point where our conviction is high enough to increase the positions. Since we’re targeting a portfolio of approximately 20 long-term stock holdings, the average weighting would be 5%, but of course, some positions will have larger weightings, like with Alphabet, and some number of lower-conviction bets will have sub-5% weightings.
  

• Fortunately, we have enough cash in the Portfolio left that we don’t have to quibble over portfolio sizing too much yet, and we have plenty of “dry powder” to jump into the market aggressively if some of our favorite names become temporarily mispriced, as happened in April with tariff fears.
  

• If you have suggestions for our portfolio, please email us at [email protected]

Quote of the Day

"The biggest risk of all is not taking one.”

— Mellody Hobson

What Else We’re Into

📺 WATCH: Nubank: Inside one of the world’s largest digital banks

🎧 LISTEN: First Principles of Valuation with Tim Koller

📖 READ: Pernas Research on why Berkshire should shrink its investment portfolio and return capital to shareholders

———> You can also read our archive of past Intrinsic Value breakdowns, in case you’ve missed any, here — we’ve covered companies ranging from Alphabet to Airbnb, AutoZone, Nintendo, John Deere, Coupang, and more!

Your Thoughts

See you next time!

Enjoy reading this newsletter? Forward it to a friend.

Was this newsletter forwarded to you? Sign up here.

Use the promo code STOCKS15 at checkout for 15% off our popular course “How To Get Started With Stocks.”

Follow us on Twitter.

Read our full archive of Intrinsic Value Breakdowns here

Keep an eye on your inbox for our newsletters on Sundays. If you have any feedback for us, simply respond to this email or message [email protected].