Nu Holdings Ltd. (NU) September 2026 Conference Transcript

Okay, good morning, everyone. I'm Tito Labarda, a Latam financial analyst at Goldman. I have the pleasure of hosting Rob Livingston, the still new CFO of NewBank. So Rob, thanks for joining us. Thank you, Tito. And I guess just to get started right away, since it is your first conference with us and you're still relatively new, I thought maybe brief background on yourself and what had you come to NewBank? Sure, absolutely.

So I started my career at Capital One. I was at Capital One for 18 years, primarily in credit and market-facing roles. I did a short stint in finance there, but it was mostly on the credit card side of the operation in the U.S., U.K., and Canada. And then I joined Visa, and I was at Visa for 12 years. At Visa, I was a general manager for the first six years in Canada and mainland China, and then a divisional CFO for my last six years there. And I joined NewBank because I'd admired NewBank for many years from afar. I knew a few ex-Capital One people who had come here to NewBank, and they described it glowingly. And then when I was at Visa, NewBank was always the fish that got away because NewBank's always a MasterCard client. And so I'd seen it. And so meeting David, meeting the members of the management team and the board, hearing about the vision, hearing more details about the company, It just was a natural fit, and I'm very excited to be here. And with that, maybe what has surprised you, maybe both positively or negatively? I guess one thing that I knew coming into NewBank was that it moves at incredible speed. There's just a tremendous alacrity to the company and how quickly we can take advantage of opportunities. But it's different hearing about it in theory and experiencing in action, And this idea of just being willing to take bets when they make sense, when they produce the right risk-adjusted returns, without having the same level of sort of process necessarily that you would have at a traditional bank is fantastic. And that all exists within the construct of a credit risk management philosophy that's very familiar to me. It's based off of what Capital One did. It's this idea of being conservative, focusing all the time on risk-adjusted returns. And it's been great to get back into that space. In terms of negatively, I don't know if it's a negative surprise per se, but there are so many opportunities out there for Nubank right now. We've got 120 million customers in Brazil, 140 million across Latin America. And that degree of scale and customer love gives us the license where we could be launching new products every single day, but we don't have the bandwidth to do that. We have to be very focused, very disciplined, and my disappointment is that we can't do more faster.

And one question that's come up is, what, if anything, do you see could change in the CFO role as you envision it?

So Lago, who I'm succeeding as CFO, was a fantastic CFO for Nubank at the period that it was going through, really consolidating its growth in Brazil and then also expanding to Mexico and Colombia. My mandate is to continue the great work that he did, but also to prepare NewBank for expansion to further geographies and even more product complexity than we have right now. And so what that really looks like within the finance function is about platformization. It's turning the finance function into something that can expand across geographies and really be ready for this next chapter of NewBank's growth. Yeah, makes sense.

And maybe to get into the nitty-gritty, because the big question we've been getting is Brazil macro and a lot of changes. You have an election coming up. And you've been growing very fast, but can you sustain that growth given some of the credit cycle concerns, macro evolution in Brazil?

So our growth story has been very healthy. So we grew our loans at 37% year-over-year in Q2. Our customer base is still growing at the high teens percentage right now. And, of course, that translates into really solid net income growth just under 50%. And we see really good line of sight for that continuing even in the face of potential economic headwinds. So let me just state the bear story just to be sort of transparent about it. So we're in an environment in Brazil where consumers are more indebted than they have been in the past and with very high real interest rates, and so there's a big cost to managing that debt. They've also had the support of a government that has been generous in the lead-up to an election in terms of programs and campaigns to improve the banking lives of consumers, and that support may go away after the election. And so there are reasons to expect that we might have a downturn sometime in the next 12 to 18 months in Brazil. The truth is, though, is that we've had four downturns in Brazil in the 13 years that Nubank has been operating, And so it's not unexpected to have a downturn every few years. And we built our business to prepare for that. Every customer who we underwrite, we underwrite assuming that there might be a doubling in losses on day one relative to what we expect and what we've seen historically. And even with that doubling of losses, the customer has to be resilient in terms of having positive revenue or net revenue contribution. And so our approach to the next 12, 18 months of foreseeable future is that even though we don't see any signs of deterioration in our portfolio today, our underwriting philosophy means that we're going to be resilient no matter what comes over the course of the next year, year and a half. Now, the question then sort of triangulates from those first two observations into what does our growth look like next year. And from our perspective, we're not in the business of pushing or pulling on brakes, depending on our assessment of the economic environment. But if you see consumer distress happening and the consumers who are applying to New Bank for loans are in a worse financial position than they were before, that would naturally slow down our growth because we would be less likely to lend to them at that point.

That makes sense. And given that context, how do you see just credit quality for your clients and maybe for the system?

Our credit quality has been very stable. And so we've talked about that there's been some noise in the system because of either decisions that we've made or Desenrola, which is a government program for debt forgiveness that has lowered risk a little bit in Q2 by 5%. but the net result is actually quite stable. In the earnings call for Q2 that we had last month, I shared a series of graphs that showed our credit risk performance in Brazilian credit cards, which is our biggest business and our most mature business, and I showed what did it look like for us by income cohort relative to every other bank in the market or the other banks in the market. And what you can see there is not only for low income in particular Do we have lower losses for any income band than the other banks in this sector? But ours are actually stable or decreasing, whereas theirs are increasing over the past few months. And we believe that this is representative of the fact that we've got primacy. Sixty percent of our customers in the mass market segment have new bank as their primary banking relationship. When they're deciding who they're going to be paying back in a distressed environment, new bank is where they're going to be coming first because this is where they do all their banking business. Yeah, there was slide 20 in the presentation.

It came up a lot. The famous slide 20. But, okay, great. Maybe taking a little bit of a step back, right, because the other concern has been just in terms of saturating Brazil. You've been there for over 10 years now. How do you think about the long-term growth potential and drivers in Brazil?

In Brazil particularly. Well, I think there's been a question about whether or not we've achieved saturation in Brazil for several years. We were growing at one and a half million customers a month. Now we're growing at about a million customers a month in Brazil. And that should slow down over time. There's a law of large numbers. It means it has to happen, yet it hasn't really yet. We're still growing at a very good pace. And the way that I see that growth happening is in three different ways. One is the continued organic growth as we penetrate customers who are not already sort of in their wallets today. But that's going to be smaller. The second is going to be about sort of new segments. that we enter into and really focus on. And that includes small businesses, Peixotas in Brazil, where we have 6.8 million small business customers already, but we have lots of room to grow and especially to deepen those relationships. And the second is in the super core segment or the prime segment, where we have just introduced a product called Chroma, which is a fee-based product with amazingly good benefits, which customers have reacted very positively to already. And that's going to sort of deepen those customer relationships, hopefully ending up in significantly more revenue. And then there's the new products that we can offer. And the biggest product for us on the horizon within the core banking suite is private payroll. This is a product that has really sort of taken off in Brazil over the past two years and not necessarily in the healthiest of ways with the initial wave of private payroll loans. We've actually been testing and learning in that market for those past 12, 18 months. focused really both on credit risk but also operational risk associated with private payroll. And we're now at the point where we feel pretty confident that there are parts of this market where we can go after aggressively. And so I would expect to see us as one of the long-term winners in the private payroll space, but I think it's going to look different in a few years compared to where it looks today.

Makes sense. And maybe one follow-up on that, because I think a lot of other banks went kind of all in on private payroll. I know you weren't there at the time, but looking at NewBank's decision in hindsight was the right decision. But what do you think drove that decision?

So our hesitancy two years ago was on two fronts. One was on the consumer front, and the other was on the business front, their employer, in the payroll market. So historically, Brazil has a long history of payroll loans, right? There's public payroll loans where you have consumers who work for either the government at the state or the local or the federal level, and they've been working for the government for 15 years, and they're going to work for the government for another 15 years. That's a very stable payroll loan. There have also been private payroll loans historically with very large employers, think Volkswagen or Bradesco, in the market. Those are also relatively stable. This new class of loans were geared at the mass long tail of employers, and so employers who have one or two or 10 or 20 employees, and that's a very different beast altogether. And so there are two things that are happening. On the consumer side, we anticipated that there wouldn't necessarily be positive selection for these secured loans. Generally speaking, when you have a consumer who's running through hoops in order to get a lower price, they're good. People who are doing that are usually responsible with their money. In this case, we saw that it was more about looking for a loan of last resort, and we were worried about the fact that there would be adverse selection there. The second element is that we didn't feel like it was as secured as people thought it might be. So say, for example, you have a private payroll loan in Brazil a year and a half ago, 20% of your salary is being garnished for that loan. Well, the attractiveness at staying at that job compared to becoming an entrepreneur or working in the informal sector is not as attractive as it used to be. There's actually some pretty dangerous incentives then for consumers to essentially leave that employer and just abandon the loan at the same time. It's a significant chunk of their pay. And then on the employer side, there were both operational as well as credit concerns that we had at the time. The operational side was how would this work if a consumer went from employer to employer, and how would that debt sort of obligation be transferred? The government now has a process in place, but it wasn't there necessarily at the beginning. And the second is that employers may, either for operational reasons or business health reasons, not remit the money that they have garnished into the system. And this was something that required us to build a degree of comfort in evaluating the risk of employers with those two lenses, both credit risk and operational risk. And so if you think about all that, there's a lot of hair on that product, and we were very glad that we didn't jump in with both feet. And again, as I said before, with that being said, we've now found that there are parts of that industry and certainly big portions of the consumer base where we feel very comfortable going after the private payroll, but it's something that we're happy to do today and not a year and a half ago.

Seems like a good choice in hindsight, at least. You also talked in some of the other meetings I sat in on in terms of like fee income evolution. You've been very excited about Chroma. You have Ultravioleta. Maybe some of the drivers of that and how you see that evolving.

Yeah, so I think with Chroma and Ultravioleta, and both of these, just to take a little bit of a step back, so Chroma is a product for the mid-market SuperCore Prime segment, and Ultravioleta is a product for the up-market or SuperPrime segment. And essentially, with each of these products, what you can do is you pay a monthly fee, and then you get benefits. So for Chroma, your benefits are a ChatGPT subscription, an HBO Max subscription, great cell phone pricing through NuCell, which is our MDNO, our cell phone provider, and then some interest rates benefits as well on some of your savings accounts. And so there's real value to the consumer for that fee product. We didn't invent those products to shift our revenue mix from interest income to fee income. However, it's a very welcome byproduct of that. We invented these products to provide more value to consumers in these segments and to make our engagement with them stickier and move to be first-in-wallet, primary banking relationship. That's why we did it. But the fee income benefit is really valuable for us. And it certainly is something that I like as a CFO and which I want to see more and more of in terms of our profile over the next few years. If I can take a moment as well, NewCell is another example there of when we looked at NewCell as an offering, which is a very simple, easy to use mobile product in Brazil. We went into that saying, what would be the best product for consumers and and how do we get it to them, and can we build product market fit? We've got over a million customers already with us. We're the largest MBNO in the country, so it has been successful in that way, and it's a fee product. Again, our core principle is always to solve to the customer need and to earn fanatical customer love, and we do that with fantastic products and services, and our direction of travel is more towards fees as opposed to interest income, which is a good situation for us to be in.

Yeah, great. So it sounds like Brazil still has a long runway to go.

Brazil has a long runway to go. Right now, our RPAC, our average revenue per active customer, is at $17. Our oldest cohorts have RPACs north of $30. So if you just look at our existing customer base and our existing product mix, we would almost be able to double from where we are today in terms of revenue over the course of the next few years. And that's not even adding on the new segments and the new products that I mentioned earlier.

Yeah, makes sense. Great. Maybe you shift a little bit now to Mexico. You recently broke even. You continue to grow clients there at a quick pace. How do you see that evolution, particularly now that you have the banking license and what value that can unlock?

We're incredibly excited about our opportunity in Mexico. And if I just take a step back, Mexico is a story where Nubank has been very successful, even with one hand tied behind our back. So let me talk about that for a minute. So in Nubank, we have been applying for a banking license for several years. We now have it, which is fantastic. And we can offer payroll loans. And there are some other elements that are really helpful to us of being a full-service bank. Most importantly, the brand benefit. In Mexico, consumers are very leery of non-bank financial providers, and being a bank gives us more credibility, so we're excited about that. Even with that hand tied behind our back, our RPAC per customer is about twice in Mexico, just over twice in Mexico, what it is in Brazil, at the same point in the evolution. We achieve break-even two years faster, and it's in a market that's much less digital than Brazil is. So when I think about going forward, a few things to keep in mind. Number one is over the very near term, we're going to be taking this opportunity to test and learn our way into the new products that we can afford to offer now in Mexico with a banking license. And so we did break even in Q2. You might not see us with positive net income in Mexico next year as we invest more in the market. But the bigger story in Mexico is about digitalization. And yesterday, President Steinbaum and her government announced the Digital Economy Act. It's an act designed to take Mexico into sort of the top tier of nations in terms of digitalization. There are three elements of that act that are very, very important for New Bank and should provide tailwinds to us over the next few years. The first is the consolidation of the existing payment mechanisms that are mandated by the government into a single point-of-sale super app, let's call it, with QR functionality. Think of it very much like PIX in Brazil. And so the Mexican evolution towards digitalization had been hamstrung by the complexity of the market and the approach so far, and this would be very welcome for us. The second is that the government announced that there are going to be a number of sectors, such as gas stations and toll roads, where cash won't be accepted in the future. They're going to be banning cash altogether. If you think about the implication of that, that means for anybody who has a car, effectively you can no longer be purely cash-based. And the Mexican economy is incredibly cash-based today, and that's going to be a tailwind for us. And then the third is that the government is very focused on having the transferability and the sort of commonality of data structures across financial services providers. That's not quite the same thing as requiring open banking, but it's a relative of that. And that is also very, very good for us because the Mexican data environment isn't as strong even today as what Brazil was like in 2013, 2014. And so we see we're excited about Mexico based off of the shape of the market today, based off our track record and based off of the banking license. And then the potential of tailwinds from this law just makes it even more exciting.

Yeah. Great. And maybe how do you think about the competitive environment in Mexico? Because maybe somewhat similar to Brazil, you have a few large incumbent banks, but you're also seeing a lot more fintech the way we saw maybe in Brazil 10, 15 years ago.

Yeah, so there are a number of strong competitors in Mexico, as there are in Brazil, and we feel good about our chances to win despite that. So in Mexico, you've got Mercado Libre, you've got Plata. Both of them have very aggressive and strong business models that they're pursuing in the marketplace. And some of the incumbent banks are quite strong as well, and we respect them too. but what we have is a cost base that is unrivaled in terms of how we approach very disciplined expenditures and our product suite so that our efficiency ratio is best in market no matter what and our credit expertise I would posit is stronger than any of the other players in the market and so as Mexico becomes more of a consumer lending digitized market I think we're the best position to win.

Maybe, and I kind of wish I was asking this question on Friday instead of today, but you have a big announcement coming tomorrow. I know you can't give us all the details, but how do you think about the U.S. market? Any high-level thoughts that you can give us there? Sure.

So we have announced that we are going to be entering the U.S. market. We applied for a banking license and got conditional approval from the OCC in January, and there's a formation period after that, which is normal. We see the U.S. as a very good market for Nubank for many of the same reasons that Brazil was a good market 13 years ago. You've got an incumbent banking sector that is relatively profitable, which has very high cost structure. And so that means that there's going to naturally be a price umbrella that we can come in under. The second is that the banking market is fragmented in the U.S. So everybody talks about the long tail of credit unions, of course that exists, but you also have the market where people have different apps for different functions in banking. There's not one banking provider that makes it easy, makes it simple, and earns the sort of customer love that we have earned in Brazil. And so we believe that we can come to market with a proposition of a very, very clear, simple, easy-to-use suite of banking products at some point that will engender that sort of love and loyalty. And we don't need to do it for the entire population of the U.S., right? Our ambition isn't, well, our ambition is limitless, but it's not that limitless. We would be very happy with a sort of 5, 10 million customers in the U.S. that could be the same size as a Brazilian business for us because of the higher GDP per capita.

And maybe shifting a little on the AI front, you've talked about additional expenses needed this year. David recently joined the board of OpenAI. You have Hyperplane. I mean, what is the opportunity that you see with AI?

So the opportunity for AI is massive, both internally as well as externally. So let me cover both of those. So internally, where we've seen AI benefit us is in terms of productivity, first and foremost. So our developers, as with probably everybody here as sort of developers, are able to move much faster than they were before with 80% more throughput in terms of sort of co-delivered and in production. That's incredible. to be able to do. We've seen in terms of our customer service where our operations function now is able to support 40% more customers per person than they were 18 months ago. That's an incredible change to happen, and that comes from AI improving our ability to sort of offer customer support at the same CSAT levels as what they would have gotten from a human operator with AI customer service. And then the third area, and that's probably the most important area for us as a bank is in credit and in credit modeling, and AI is doing two things on the credit modeling side. The first is that with Newformer, which is essentially a macro data environment that is AI ready with every single piece of data that we've got within the bank, both in terms of internal transactions, but also everything that's happening on the app, everything that's happening on the phone, there's a lot there. What that allows is for much better credit risk models, much better revenue models, better marketing models and pricing models than what we had before. And when you're seeing step changes in our growth, and there were times last year where you saw quarters with significant step changes in growth, that's when a new generation of models were effectively launched internally, and we own all those models. Those are all internally generated. The second thing that AI allows us to do is to actually iterate after that first generation of models much more quickly than we did before. Previously, if we dial back, say, two years at Nubank, or if you dial back 15 years when I was at Capital One, it would take three months to validate a model. You build a model, then you validate it on the data set that you built it on, you validate it on a holdout, then you look at different time series and you make sure that it still works across all of those. And that is a laborious process if you've got statisticians doing the work with SQL Go, right, and fast. It's much, much faster with AI. We can do it within, like, we can do this within hours now at this point in terms of model validation and to do sort of model refinement plus validation. Let's say that it's going to take with the appropriate amount of human supervision closer to a week as opposed to three months. And so the speed of iteration is just so much faster than what it was before. And then on the external front, and this is something that we haven't talked about as much, but we've got an investor day coming in December. I hope everybody dials in for that. And we're going to be talking much more about the way that AI is transforming what we can offer customers, particularly in terms of this notion of putting a private banker in everybody's phone. So there's this idea that we could delight customers, help customers, by helping them understand how to use their money better than they are using it today. Almost everybody in this room probably has experience talking to a financial advisor, sometimes with good results, sometimes with less good results. The AI private banker, in our belief, is going to be stronger than any financial advisor, unbiased, and help customers find the best solutions for them, either within NewBank or outside of NewBank. When I think about NewBank and sort of surprises, that was your first question, Peter? The other one that I had is that at NewBank, every single decision that we make starts with this question of what will generate fanatical customer love. And so it creates very interesting sort of decision points throughout all aspects of the business. And as we're thinking about how to bring AI to bear to benefit consumers, that's where there's just tremendous excitement right now on the front lines of the company is how are we doing this to transform the daily lives of Brazilians, Colombians, Mexicans, Americans, wherever we go next, with this sort of product and this service that doesn't exist anywhere today.

Yeah, makes sense. And, I mean, you've also talked about global expansion. When you think about what are the potential countries beyond the U.S., you've got a question about Argentina, but how do you frame sort of which countries you would look at, what you consider?

Yeah, I mean, there have been two sort of headline models, I think, for the global sort of digital bank. There's our model, and then there's the Revolut model. Sorry. He who shall not be named. Yeah, I know. And those models are very different. Theirs is very sort of thin and wide and sort of very niche in terms of the sort of customer and use case that they're going after. Ours is very deep and built on customer loyalty and customer love. Where we're thinking about going next is going to be around how do we build products that resonate, and it could be wide, it could be deep, right? If it's going to be deep, it has to be in a country where there's the data availability that we need in order to build our models, where there's a regulatory environment that is friendly to NewBank, which doesn't want to do branches, which doesn't want to do cash out at ATMs, right? That's not our business model. And where our brand can translate and be powerful. And we've got a very high bar for that because there are so many opportunities today in Brazil, in our other markets, that it's just unlikely that we're going to be expanding with that deep full service or full suite of banking products that quickly in other markets beyond the ones that we've already announced. Yeah, makes sense.

Maybe to bring it back a little bit, some high-level guidance that you've given, like 20% efficiency for this year, but we know the additional expenses that you're expecting, but how do you think about the long-term efficiency for the business and operating leverage?

So having a market-leading efficiency rate is core to our strategy. It's fundamentally part of the cost base that we've got, where if you look at our efficiency ratio and you look at our credit expertise, those are two very powerful, sustainable advantages that create a moat for Nubank wherever we are. On the efficiency ratio, we don't have a specific target as a company. In Brazil right now, you can see that our efficiency ratio is 15%. It's hard to imagine it getting much better than that, but maybe it could, given what I shared about the progress of AI in terms of customer service earlier. As we look at new markets and new growth, what we don't want to do is we don't want to intentionally or unintentionally get into a space where we're a high-cost provider. You can always sort of argue why this will be an exception or this is why we need to do it in this one case. The answer for me is no. The answer for me is always going to be our value proposition depends on giving customers the single best price they could possibly get with fantastic service and a brand that they love. That's what we're going to be doing. And so we're constantly in each market going to be managing the efficiency ratio to be as low as it possibly can be. And I talked about the advantage of coming into the U.S. sort of earlier. As a company, our efficiency ratio this year should be around 20%. In the U.S., it might end up being slightly higher than that just because it's a higher-cost market to operate in. That's still less than half of where any of the leading U.S. banks are.

The other piece of guidance you sort of high-level given, the 12% risk-adjusted NIMH, it was a huge step up from what we had in the 1Q. One, what surprised you that it came up so quickly, so fast, and what gives you comfort that it's sustainable?

Yeah, those are great questions. So to answer the first, our risk-adjusted NIMH jumped almost 300 basis points quarter over quarter, and that was faster than expected, I think, both internally and externally. What we saw was really, really positive traction in the latter part of Q1 that we had quite anticipated in terms of balances building and matriculating into actually interest-earning balances. And then also early in Q2 at the very beginning, very strong growth, especially at the beginning of the quarter that helped the quarter overall. We also had a slight benefit in Q2 coming from Desenrola. So if you think about the breakdown of it, roughly two-thirds of the benefit came from better top-line growth with net interest margin, one-sixth of it came from underlying credit loss performance improvements, and one-sixth of it came from Desenrola, which is a government debt forgiveness program. So that's the evolution of what happened. In terms of where we look going forward, I haven't made any commitment about the long term in terms of risk-adjusted net interest margin, but for the next two quarters, we think it's going to be roughly in the same ballpark. If we had a new model generation that was launching in Q3 or Q4, then we might have to build up loan loss allowances, and that would be a drag on it. If we have outside growth, that could also be a drag on it, but we don't anticipate that as of today. And the strength of the top-line growth. The strength of that net interest margin growth, that for me is really the kicker and why I felt comfortable giving that guidance.

Yeah, that sounds good. Do you have a few minutes if there's any questions from the audience? Anybody want to ask? While we wait on that, maybe thinking about capital allocation as well, you announced the buyback program earlier this year. How do you think about your capital needs, where you allocate for the buyback?

Yeah, we're in a very healthy capital position right now. We're not capital constrained, so we're feeling good about that. Our buyback program had us able to buy back a billion dollars of shares. We've completed about half of that so far as what we announced in the last quarterly. And the way that we look at it is that our shares at this stage or at that stage were a very good sort of use of capital relative to the other opportunities we have. But my first choice would be to always invest capital in really, really good business opportunities, and we have them out there. What constrains us is just our sheer bandwidth as a management team, but also the readiness of our models in the marketplace. As those become ready, we're going to deploy capital each and every time towards what will give us the best risk-adjusted return.

Yeah, makes sense. Just checking if there's any questions. Anyone here, Mark? They're bringing you a mic if you want to wait. Just one second.

Thank you. So kind of a combined question. I was curious. I didn't know that Chromo was a fee-based product and had the features that you're talking about, which are actually pretty exciting. I hope that you bring that to the States. But Mercado Libre has been increasingly, I think, successful at transforming their business a little bit more into a fully integrated network, and they have an opportunity with their marketplace to begin to provide meaningful value to their credit card customers. And it's a ways down the line, but as you think about the evolving features and privileges of membership, what might that look like, and how do you think about that marketplace model?

Yeah, it's a great question. We did try a marketplace model a few years ago in Brazil. It didn't work as well as we would have liked because we discovered that people didn't necessarily want to buy a TV through their NewBank app. It didn't make sense to them. What we're very much focused now is expanding our brand and our customer love towards specific products that make sense to buy within an app like NewBank's app. And so NewSell has proved to be a very successful example of that. Our investment business is a very good example of that. I think that there's the opportunity for us to offer more complex banking products, such as mortgages or auto loans that we don't offer today. Insurance. Insurance. But all of those would be in a marketplace-type approach, right, where we would be getting a fee for selling the product and being a broker as opposed to actually doing that ourselves. The cost associated and complexity associated with those products isn't something that we're keen to take on. So our business model is very different, obviously, than MercadoLibre and with MercadoVago. It's just a very different model. And what we want to do is going to be systemic and more targeted, I think, than a full-service marketplace from day one. Does that make sense? Yeah. And as far as U.S. products go and what we're going to do there, I think Chroma is great, but it's also very new. So we have to see how Chroma performs over the medium term. It seems great so far. The customer love, the product market fit seems to be very strong so far. But, again, with anything in sort of consumer banking or especially in lending, we need 12 to 18 months to make sure that there aren't unintended side equipment. I think with that, we've reached the time limit. We have reached the time limit, but I just want to say that if you're not on our investor relations mailing list, it's a good place to be, and we did announce yesterday that there's going to be a live event from Miami tomorrow at 8 a.m. Pacific time, so you can find the details on our website, and I'd encourage you to tune in. It's going to be fun. Great. Thanks a lot, Rob. Appreciate it.