TIVP032: Remitly (RELY): Future Multibagger or Stablecoin Casualty? w/ Daniel Mahncke & Shawn O’Malley

27:19From 2019 to today, SG &A as a percentage of revenue declined from 54 % to just 37%. Nadi went from 25 % to 20%. Just over the course of this year, the marketing spend as a percentage of revenue again dropped by almost 5%. So they are saving. And I do think you see some scale advantages because I would argue the remittance space is just a small niche of the payment space. And the actual competitors are probably not as many as you would think at first glance. Again, it reminds me of our portfolio company, NewBank, where word of mouth has proven to be a much more highly effective driver of customer growth just because their product has been so much better than what the legacy banks could provide.

28:05And I guess that customer stickiness in this space is also higher than in other payment spaces. And as I said in our PayPal episode, I don't really know why I continue to use Venmo weekly beyond the fact that I started using it in college. And most people I know use it. And for me, it would be simple to change, but yet I'm still here using it. And so if you trust someone with money that could decide whether your family can go grocery shopping the next day. I mean, the stakes are way higher for what payment platform you're using. And like I said, there's no stakes for me and I already have trouble switching.

28:45And so the threshold is going to be so much higher for somebody whose family is on the line in terms of what platform they trust to remit their money with. I believe this is a factor that is underappreciated for the P2P business, especially in advanced economies and even more so in this sector. Technically, and you basically mentioned it, I could use another app than PayPal, but I've used it now for 10 years and have not once thought about actually changing it. And same for you with Venmo. And with Umitli, that's even more likely just for the reasons that you already mentioned. And generally, think about all the data Umitli has to get from its customers.

29:25That's sensitive data and you don't want to give that to just any app just for trying them out. Data is actually the second advantage that scale brings in this case. Customers have to list their names, date of birth, email, phone number, government idea, sometimes also proof of residence and all of that stuff. And as soon as they start sending money, that's when we immediately can start gathering all the payment data. This helps improve risk management, fraud prevention, float management, and of course, also customer support and satisfaction. And by now, 92 % of transactions are dispersed in less than one hour, and more than 95 % of transactions do not need any customer support along the way.

30:06And those are pretty strong stats. When you say 92 % of transactions are dispersed in less than one hour, you're talking about every single transfer, 24 hours a day, seven days a week across all corridors globally. That means that whenever someone sends money from one country to another, 92 % of the time, the recipients has those funds within one hour. Is that right? Exactly. And I would have said that sounds pretty insane, right? But since I know that you're way more knowledgeable on Bitcoin and the entire blockchain space, I don't think that's too surprising to you. But I think that's what we have to understand when we talk about this business.

30:45I know sometimes payment seems pretty abstract and that makes it harder to understand the competitive advantages in this space. And I wouldn't go as far as calling them moats, but there are definitely advantages in place also for companies the size of Remitly. And in this case, the complexity of the business lies in how remittances are handled. Every country is different. In the Dominican Republic, you still have cash deliveries to your doorstep. While many countries in South America are totally digital and only use mobile wallets, Remitly must have the right system in place for each of those countries.

31:20And perhaps that's the reason why we do not yet see stablecoins or Bitcoin handling the transaction. And that's yet another advantage of scale that you slowly build, and that's corridor optimization. I mentioned it before, but the more volume you put through key corridors like US to Mexico or US to India, the better you can manage the complexities of those markets. And part of that is pre-funding balances and negotiating with local partners. And all these advantages we listed here, they kind of reinforce each other. But once again, creating one of those flywheels that you and I love to see, more volume means more customers, which leads to better negotiating power with partners that reduces cost, which due to the scale economies shared model can then be passed on to consumers through, for example, lower pricing.

32:11That's the customer acquisition aspect. And that leads to the data advantage where more data improves your risk management, which leads to less fraud, which leads to more satisfied customers. and lower costs for Remitly. Again, I kind of think you see what I'm talking about here. When you talk about Remitly as a leader in the space and a beneficiary from the scale that comes with that relative to their competition, I'd be curious to hear your opinion on WISE. And I think, as you know, Daniel, here at TIP, internally for international payments, we use WISE. So I just would like to know why you think that they as a competitor won't benefit more from scale, given that they're about five times the size of Remitly.

33:02wise is probably the company that comes up most or is mentioned most as a competitor and it's the obvious answer if you look for a digital first competitor in the remittance space but on a closer look the two companies seem to be very different structurally it's not a perfect comparison but to perhaps keep the uber analogy going a bit wise might be what i don't know tesla is to Uber. Could Tesla's RoboTaxis theoretically displace Uber? I would say yes. Would it improve the ride-sharing market for consumers? Very unlikely. You would say definitely not. And it's not a perfect comp, but I'm giving my best to include some companies here as comps so that we can just, you know, better understand it.

33:47And to a lesser extent, I do think the same is going on with Remitly and Wise. The first major difference is the send amount. And with that comes also a different customer base. So just recently, Sean and I had the opportunity to invest privately in OpenAI. It's an opportunity that came on our desk thanks to one of our phenomenal members in the intrinsic value community. And in case you listen to this, thanks again for that opportunity. And we must say it was given to us on a pretty short notice and my bank wouldn't wire the money directly to the US. It didn't help that this was also on a weekend.

34:23So the obvious solution for me was to get my money onto Wise and then send it to the US account that way. I'm sure Wise is handling all kinds of transactions, but most of them are from people like you and me, and they tend to be higher ticket than most remittances, especially those of migrant workers. The average ticket size on Wise is between$3 ,000 and$4 ,000. On Remittly, it's just about 10 % of that. It's a totally different customer, and both companies want that different customer base. Wise wants to attract you and me, or better said, companies like TIP, so small and medium businesses that need international payment solutions to pay their employees.

35:02And I presume they focus on that market because it's objectively the bigger opportunity. Absolutely. The market has a size of about$12 trillion. And as mentioned earlier, the remittance market for migrant workers has a total addressable market of about$2 trillion. And still though, why can't WISE try to go for both? I mean,$14 trillion in TAM sounds a whole lot better than$12 trillion. That's true. But as you and Munger have both figured out, and I guess he copied you on that end, specialization wins most of the time. The user experience, the marketing, the payment network, the customer service, all of that is totally different depending on which customer base you want to attract and then also serve.

35:48So if I want to make an investment and the payment doesn't go through fast enough, that's definitely annoying, but it's a luxury problem. If someone sends money to their family so they can afford groceries for the next week, that needs a whole nother level of attention and customer service. One very practical difference, just to show you, Sean, how that difference matters, is that with WISE, you can only send remittances account to account. But as I said, in the Dominican Republic, for example, people want cash delivered to their homes. In the Philippines, people want their money in their mobile wallets.

36:21Every corridor is different. And for WISE to focus on that would mean a significant shift away from what currently is their biggest strategy. Another problem here, and we've seen it with larger and really more importantly, already profitable businesses, is that once they reach that stage of maturity, of recurring profitability, they rarely want to go back to spending a lot of money on these expensive growth plans that might support long-term growth, but if it's going to set them back on short-term profitability, it's really hard to sell shareholders on that. If investors don't like to see it, then it's less likely to happen.

36:59And I can imagine going after the migrant remittance market would mean a lot of investments in marketing and customer service and building new relationships and all that stuff. And ultimately, that would be just to get into a business with a smaller TAM and smaller transaction volume. Yeah, I think that's a big part of it. And you know, one thing that makes me also bullish on Remitly is a very simple one, and that is that customers love it. And I know this argument seems a bit weak and naive at first, or at least sometimes when I hear it, I roll my eyes when investors say that. But when there are certain data points that back this argument up, I would say most of the time accept it.

37:37When we talked about Newbank, you told me about their incredible customer acquisition through word of mouth. And we see the same also with Remitly. Marketing spend goes down relative to revenue, while the company keeps adding millions of users. The average account growth in the last five years has been almost 50 % a year. And the retention rates are also extremely high. After signing up and using Remitly for a year, the retention rate is about 90%. That's what management is telling us. And while I don't have any data on that, I would assume they are probably as high as 90 % right from the get-go, especially after you send perhaps two or three payments.

38:15Ratings on Apple's and Android's App Store are 4.9 and 4.7 with a combined rating of 3 million. So that's a whole lot of ratings. And remittances almost work like recurring revenues. They are sent regularly, often monthly or at least quarterly. And that's another advantage that you have compared to a company like Wise, where the main business is people like you and me who do not invest in OpenAI every single month on a quarterly basis. So if we assume Wise is not a close competitor, then I would assume we're talking about the legacy players like Western Union and MoneyGram as really being the entities that Remitly is trying to steal market share from.

39:01But can you confirm? I mean, is that who Remitly is really competing against and winning against? That is who they are competing against. And definitely that's not a very tough battle to win. On YouTube and Spotify, you can now see a chart that basically displays pretty well how Remitly is taking market share from Western Union and also how much more runway there is just by taking more share from that one company. In 2019, Western Union's revenue was About$5.3 billion compared to only$126 million for Remitly. And since then, Western Union's revenue has declined by a 5 % CAGR to barely over$4 billion.

39:44And at the same time, Remitly has grown revenue at a CAGR of 57 % to over$1.3 billion. That is not a good looking chart for Western Union. But perhaps it's time to now talk about Remitly's moat. or at least their competitive advantages since it sort of sounds like they're too small to really truly call it a moat. And ultimately, we have a fast-growing market here and we have these legacy players that are ripe for disruption. So you would think that that would attract a whole lot more competition than just Remitly. You would think it would bring in a lot of these digital first players trying to steal market share from Western Union.

40:28New competition coming in. might be what makes or breaks the investment case for remittly. Building deep modes in the payment industry, and I think I don't need to tell that to you, is incredibly hard to do since the barriers to entry and the potential to differentiate oneself are generally pretty low, at least early on. And as we talked about, once you reach scale, things change a bit. In the remittance space, there are about a handful of digital first players, actually a little less than that. The main players include Xoom, which is owned by PayPal. And it's by the way, Xoom, but with an X instead of a Z at the beginning, then it's Xe.com and Ria.

41:07But despite being digital first, all three come from a slightly different angle. Ria and Xe.com are actually both part of the parent company URNet and together they move over$50 billion in 2023. So combined, they probably are slightly larger than Remitly, but they're growing significantly slower. Remitly is growing at 38%, while they are only growing at 8%. And by now the companies might be somewhat similar in size, at least in terms of percent volume. And revenue wise, they are still a bit ahead at$1.7 billion. That's the latest figure I could get on the company, because like I said, it's private.

41:47So I would assume that take rate is a bit higher than Remitly's. I have to guess here, since like I said, the company isn't public, so I don't know the exact data. However, a higher take rate makes sense in this case because RIA still relies heavily on its cash payout network, which spans about 500 ,000 physical locations. So while it's big, it's not entirely a digital first player and it has a lot of legacy infrastructure. Because of the higher complexity and cost of such a business, it would just make sense to also charge a higher take rate, although margins generally would probably still be lower.

42:22Just to say it again for the listener, combined, Ria and XC.com are a bit larger than Remitly, but Ria is mostly still what you call an old-school remittance company, kind of like a Western Union. Exactly. Ria's digital app is growing, but it's still only a slice of their overall business. Remitly by Contours is 100 % digital, which means it doesn't have to pay agent commissions or run physical storefronts. And that creates a big difference in cost to serve and also in how scalable the business actually is. Xe.com is much more digital, but their customer focus is a bit different. I've also read through plenty of customer reports, probably too many.

43:06And Reddit was also used again, of course. And it turns out that Xe.com tends to be on the cheaper end of the spectrum, but it takes significantly longer to get your money. significantly longer means instead of an hour, as with Remittly, it can take a day and sometimes even a couple of days, at least when you trust the reports and the ratings that I've seen. It's also much more focused on expats and small businesses and everything is bank to bank. And that's also why it doesn't matter too much that remittances take longer. They are not really targeting immigrant workers who send money home for essentials like grocery or energy.

43:44Yeah. Okay. So it sounds like XE is closer to WISE than the Remitly's of the world. I would say so, yeah. I expect the average ticket size. And once again, I don't have the data to be much larger than Remitly's as well. So WISE is probably a better comparison. From what I've heard, XE is something an expat would use to, for example, pay rent in another country, not something a Guatemalan construction worker in Californian is going to use to send$100 home to his family. And then, like I said, there's Zoom. And since I didn't even mention it in the PayPal episode, you can probably imagine that it's not a vital part of PayPal's business.

44:23And it has been very neglected in recent years. There's even rumors that it will be sold at some point. And it's not a bigger part of the strategy. It wasn't back then. It isn't now. and it tends to be on the consumer side just a more expensive player than the others and it also takes longer so I don't think this will be a competitor at any time. PayPal acquired it in 2015 and I guess it was part of the broader strategy to get PayPal into as many markets as possible and increase its reach which is something that we discussed in our episode but it just didn't really work out. As I said earlier in a space like this you won't be successful doing remittances just on the side.

45:04This market will be dominated by companies that commit to this 100 % and just want to deliver the best service to the users. That's why I emphasized in the intro that Remitly, from the get-go, focused on the niche market of remittances for migrant workers and nothing more. And as we discussed before this call, if we would say that, let's say, stablecoins or any form of crypto will be used, I really do believe that the companies currently dominating the space can be the ones doing that. But to just jump on that bandwagon, you need to be the market leader. You need to have scale and you need to know what's going on in the market.

45:43And I do think that's more likely to be a company than remotely, than, you know, a little branch of PayPal. So that is the elephant in the room, though, is stablecoins. And I think you said you don't really see them as a threat to PayPal because there's just not that much reason to use stable coins for peer to peer transactions with friends. And there's no real incentive maybe for consumers to adopt them in e-commerce either because we already have credit card programs with great rewards and fraud protection and pretty wide acceptance. And for stable coins to be attractive, you would almost have to have retailers fund similar incentives to use them in the way that we use credit cards.

46:24And even then, probably only the biggest retailers like the Amazons and Walmarts of the world could even really seriously think about doing something with stable coins. And again, who funds those initiatives? So I would probably think Amazon and Walmart, it's not going to make a ton of sense for them because the cost of providing some sort of rewards for using stable coins would probably cancel out any of the savings they get from not having to pay fees to Visa and MasterCard. And merchants selling on those platforms are really probably not going to be happy to pay those either. So for anyone who isn't familiar with the term, stable coins do come from the crypto world where they use blockchain technology to process transactions.

47:06Unlike Bitcoin, the value of stable coins are pegged to designated fiat currency. So you might have a stable coin that uses the Ethereum network and is worth exactly one dollar at all times, at least in theory. And so let me ask you, how about international money transfers? I mean, this is a space where the traditional rails are slower and more expensive. And from everything I've ever heard about stablecoins, this is sort of exactly what they're meant for. So why don't you tell us, do stablecoins really make more sense for cross-border payments like remittances? Well, I have to give the same disclaimer as in the PayPal episode, which is I'm still no expert on this.

47:52and against some knowledge from you know the paper research and now I looked at it from a slightly different angle but it's hard to have a strong conviction on this because no one really knows how it will turn out and we both discussed how even 10 years ago people thought we might even pay with Bitcoin at this point but to the least people would use it to send money across borders so yes on paper stable coins make a lot more sense in cross-border transactions and if I'm not mistaken, the big improvement that stablecoins can bring is not necessarily just speed or lower costs. It's more about access.

48:29Banks do not serve everyone, especially not in emerging markets. And stablecoins need no banks and technically also no intermediaries. I say technically though, because there is a big difference between what's technically possible and what actually works for people who send and receive payments every single month. When we think about remittance services, one of the biggest problems they solve is the last mile problem and most recipients don't have crypto wallets they didn't have them for the last 10 years and perhaps they won't have them for the next 10 years although we don't know that right now they want cash or money deposited into a mobile wallet or a bank account that they already trust so unless there's a way to off-ramp stablecoins easily and locally that theoretical instant transfer kind of hits a dead end here at least to my knowledge and there are still some onboarding hurdles for senders as well you need to open a wallet verify your identity convert fiat to crypto and then pay gas fees and gas fees as you probably know way better than i sean are fees that you pay for transactions on certain types of blockchains compare that to remotely where you open the app enter a name and then hit send and if anything goes wrong there's and that's one of the most important things 24 7 customer support in your own language remember how long it took and still takes to just move away from western union to digital players i just don't buy at the moment that the same customers now switch over to stable coins in a matter of months when it took them 10 20 years to switch to what definitely is a better product than Western Union.

50:12And I totally see the use cases of using them as a store of value and highly inflationary currencies. But as we repeatedly said today, remittance payments are used to pay for essentials, not for savings or to accumulate wealth. So you got to know that the money is worth today what it's worth to mow and stable coins might do that. Bitcoin definitely couldn't do it. I think people probably get tired of hearing about crypto and they're maybe not expecting to hear about crypto on our podcasts and we're not trying to linger on crypto beyond the extent that this is kind of where the use cases make a lot of sense on paper and so you know we kind of have to talk about stable coins because it is a pretty legitimate alternative at least that has arisen and so it's not to say that stable coins won't eventually come or that they're not legit.

51:06But at the same time, 10 years ago, there were people who thought that everyday transactions would be all taking place in Bitcoin or Ethereum by now. And that hasn't happened. And the use cases for something like Bitcoin has evolved a lot. And yet here we are still using our credit cards and paying for things in dollars and using the Visa network. And it's maybe a tangent, but there are people who would argue that Bitcoin is more a store of value and an alternative to gold than a currency you would use on an everyday basis, at least currently. And so, again, that's not to say we should totally write off stablecoins because it's a completely different thing than Bitcoin.

51:40But at a minimum, I just think meaningful mass scale adoption, even if stablecoins are a superior technology, is probably going to take longer than people think. That's exactly where I'm coming from, too. I can also see how stablecoins will be used in payment and remittances especially. But I think it's just much more likely that today's payment giants will introduce them to customers. Now, is Remitly such a giant right now? Definitely not. Could it potentially become one at a certain point? Maybe. Remitly already runs a multi-rail system. So that's bank wires, card networks, pre-funded accounts.

52:23So if stablecoins can help them to move float between treasury accounts faster and cheaper, they will absolutely plug that in. And the end user won't even know or have any trouble with setting up wallets or anything like that. They will still get their money in an hour. And if they want it in cash, it will be in cash. If they want it on a mobile wallet, it will be on a mobile wallet. But maybe it's cheaper for a midly to make that happen. And they could pass those cost savings to the consumers, which in turn improves the service even more and therefore might also increase the adoption rate of digital versus offline remittances.

52:59Because, you know, there's still this big market that Western Union has. And Remitly has been investing in stablecoin space for years now, as so many of those payment companies do. So I think them implementing it instead of getting replaced by it is a more likely scenario. and the same goes for PayPal, but also of course for Visa, MasterCard and all of these major players. For anybody who listened to our PayPal episode, it is going to ring familiar again because it sounds like you don't see stablecoins as being a thesis breaker here with Remitly either. But I do want to ask you what that thesis breaker could potentially be then because this is still a relatively young and small company.

53:42So maybe that alone makes it more difficult to have high conviction and then being a major winner down the road. But still, there has to be some kind of risk lurking on the horizon that gives you pause about wondering whether they will be the winner long term. If you have to reach scale and be kind of the biggest player in the industry and you're still young and a small company, there's always plenty of risks that you have to pay attention to. And we mentioned how scale is the main advantage in this market. And that's why it's just so important to always be on top of mind to say, okay, is this company actually still on track to get there first?

54:22And the first possible risk is that the returns on future customer acquisition just simply goes down, which would hurt the thesis that Remitly is at an inflection point. Remitly has been growing accounts quickly. And we mentioned for NewBank that fast growth is not always a good sign, especially in the finance sector. but in this case it seems to be healthy growth new cohorts are still more profitable than the older ones and fraud rates are still very low but growth is expensive remitly used to spend over 40 percent of revenue just on marketing and while that has come down from past levels it's still high at about 25 percent of revenue so for comparison most of remitly's competitors the the ones that we named already in this episode, they spend about four to 5 % of the revenue on marketing.

55:10So that's significantly less. Now the payback time for an acquired customer is well below 12 months, according to the management. And the lifetime value of a customer currently is about six times higher than the acquisition costs. And I think you could even argue if you look at the numbers, that's probably eight to nine times higher. So the return on their marketing investments are huge, but it is still a temporary drag on profitability and therefore also on the stock. And if management continues to spend on marketing, it will probably be so for a while. And I think they will do that. But as they do it, I think they're smart about it.

55:46Marketing, where they really catch migrant workers. They have among the highest ROIs on their marketing dollars. So I'm more than fine with it if they do decide, you know, to keep marketing spend higher for longer. It's kind of similar to what booking.com did and the playbook they use. It's a company we haven't yet covered, but Clay did so on the We Study Billionaires show. And the travel booking industry doesn't have strong natural modes either, but booking spend on marketing, like nobody else in the industry did. And their marketing spend is still in the mid thirties, if you measure it as a percentage of revenue basis.

56:21So I think that's the playbook that admittedly is using and i think it's working out until now booking is like a good company for us to probably get more familiar with given that we have airbnb in our portfolio and it's just a fascinating case study in how you can build competitive advantages in a business that sort of looks like a commodity on the surface and at the end of the day booking doesn't own hotels it just lists the rooms but what it does own is demand and it got there by pouring billions of dollars into performance marketing. And a lot of that was on Google. And for years, this was called the Google tax where booking would spend 30 to 40 % of its revenue just to get in front of high intent travelers searching for hotels.

57:06But the key to building their moat has been and still is what happens after that first click. But booking turns those paid clicks into app downloads and then loyalty program members and repeat customers who later start their search directly on booking the next time they go around. And for Remitly, I'm not really sure exactly if they have the same level of customer lock-in and how that would work. But I imagine it mostly has to do with the trust in the brand that Remitly can kind of give them by consistently ensuring that payments make it to their destination. So I assume that's not something you would classify as a threat to the thesis, but what are the things that you see as existential threats here?

57:57Just on that last point, it is in fact more difficult for Remitly to turn customers into long-term customers. But right now that works without any problem as retention rates of over 90 % show. And as you said, it's mostly because they have what I would consider, and I just heard that from other people, I don't use it myself. They have the best service and customers trust them and that's worth a whole lot in this business. It's not a subscription service or rewards program but the payments are highly recurring. Many of them are even scheduled. So once everything is in place we typically don't switch it up.

58:31But yeah getting to the next big risk factor and one that actually has the potential to break the thesis and that's regulation. Just recently we saw the big beautiful bill causing the stock to tank. In fact the stock got hit repeatedly due to Trump's focus on illegal immigration, but that shouldn't matter for remittantly too much because as we said before, their customers are legal documented migrant workers with a US bank account and US debit cards. So it shouldn't matter that much. But in the bill, there was a section that proposed a 5 % access tax on remittances sent by US residents who are not US citizens.

59:10So also legal immigrants, green card holders, visa holders. And first of all, it would have been highly questionable if this is in line with the laws against discrimination of permanent residents. But fortunately, we don't need to worry about that much more because it's not part of the bill that actually passed. I guess this will probably not be the last headline we've seen about migration that could hurt Remitly in the short term. But the payment industry really hasn't been the only one that's experienced some unexpected headwinds, I wouldn't say. Definitely not. I mean, the last few months, we had to discuss politics way more often than I think both would like.

59:54But, you know, getting to another potential risk, which is not regulation, which is not politics, or Bitcoin and crypto, but perhaps the biggest risk that Remitly is facing is the one of competition and the potential margin pressure it can cause in the long run. Remittly's takeaway today is around 2.3%, but that number could get squeezed over time if other digital competitors start to gain more traction. And while Remittly build strong brand trust, remittances could become commoditized in the future if the consumer's preferences shift and price becomes more important. And you basically mentioned that.

1:00:30But I think what I currently see in the market is that a lot are willing to pay a little bit of a higher price if they know the brand that they're working with and that the money is actually getting there. That could change. So the fear is of what you said, a so-called race to the bottom that we also discussed with PayPal where take rates fall endlessly due to price competition. And while it's true that payment processing has natural pressure or natural price pressure over time, the data doesn't show a straight line collapse in take rates. For admittedly, as I said, take rates have actually been remarkably stable at around 2.3 % in recent years, despite growing volumes and also new market expansion.

1:01:11And of course, the company is young, so take it with a grain of salt. But I still think it's worth mentioning because it is kind of a trend that you see. in part it's also because there's just still so much more room to grow for all the players so fiercely competing already is it's just not what happens at the moment but the main reason is in my opinion that remittances and i said this before and i will say it again are not just a commodity transaction they come with all the complexities that we already discussed regulation fraud risk foreign exchange spreads and the need for you know local payout infrastructure and because of that the remittance space is i would say more insulated from pure price wars than domestic card payments for paypal which operates more in e-commerce and merchant payments takegrids have gradually declined over time but the drop has also been at least somewhat slow and manageable so supported by value-added services and network effects which is also something we talked about they could handle it pretty well in terms of profitability and the more likely long-term scenario to me is an oligopolistic market structure with a handful of dominant players where scale, trust, and infrastructure matter more than price alone.

1:02:26And in such a world, take rates might compress modestly, but it's not necessarily this race to the bottom that so many people are fearing. The winners will be those who can reinvest their scale advantages into cost reduction, fraud prevention, and user experience. And at least I think that not only the ones who undercut on pricing. When I think about it, I think I can imagine some pressure from Western Union here. I mean, Western Union clearly isn't going to roll over and just seed market share without a fight, even if they've done a pretty poor job of defending their market share thus far. And, you know, what if they did something like initiating a price for where they do try to drive out companies like Remitly by simply just choking them out at their own expense for a period of time.

1:03:12But once the dust clears, then they would be the only ones left standing because they have the cash to absorb losses for, I think, longer than Remitly probably can. That's a very good point. I think for everyone who ever had a class on game theory, and then they see the chart of Western Union declining, they would think, wait a minute, why should they just not get into such a price war and kind of destroy the competition that way and I think it's not so likely just because Western Union and their shareholders definitely have not an interest in Western Union doing that. If you hold shares today you do so betting on margins staying stable for as long as possible to then earn enough money to finance the dividend.

1:03:56If they start dumping prices to regain customers the dividend would have to be cut and margins would go down the drain and I'm not sure they could do that with their shareholder base. And also, if you think about it, we just talked about competition. There's so many little competitors coming up. So Western Union might be able to do it once and then with Midley is gone. But what about the 20 other companies that come up eventually and just take their space again? Even when we're not talking about politics, somehow it comes back to some kind of politics, even if we're just talking about internal shareholder politics.

1:04:28You can't cut that dividend, right? So some things never change. But jokes aside, in preparation for this episode, I did come across a payments industry report from a group called BCG. And they dug into the changes in the payment industry in recent years. And the biggest narrative has been that the fast growth phase is over. And payment providers are now focusing more on profitable growth. And I guess that description fits PayPal more than Remitly. But it would support the thesis of an oligopoly where the market leaders now go from growth to profitability. In that process, they start to stabilize their take rates.

1:05:12Yeah, that's a good point. I mean, it's kind of what I imagined how the payment industry could go on or could go forward. And it might be the reason why the payment industry also, which historically performed in line with the tech industry, if you look at stock charts, hasn't performed too well since 2021. I think investors still focus more on growth than profitability. That's what they do in the tech space. And perhaps they want the same in the payment space. Then again, Remitly isn't trading at high multiples either. Quite the opposite for a company growing that fast. So perhaps they just sour on the whole idea of payments.

1:05:46And maybe because they see stablecoins as more of a threat than I do. Well, talking about valuation and multiples, I think it's that time where you walk us through the details of Remitly's financials and what it means for the valuation ultimately. And I don't envy you because I valued companies at the earlier stages of their life cycle. And it's always exciting, but it's just so tough because there's so much optionality on the upside, but there's also so much more downside. It's definitely tough. When valuing a company like Remitly, it's important to distinguish between what the business looks like today and then what it might become once it reaches scale if it ever does so because of that right now Remitly is still investing heavily in customer acquisition and corridor expansions all the stuff that we mentioned but the unit economics are improving fast and the model is already starting to show signs of operating leverage and in my model you can see the process that I described when I compared Remitly to Uber it's a company that's right at the inflection point of transitioning from breakeven to meaningfully profitable.

1:06:57In 2024, we immediately processed around$55 billion in cent volume. And that's the total amount of money sent from customers or moved from customers through the platform. And over the next five years, I could see that volume increase to more than$140 billion, which implies a compounded annual growth rate of about 21%. And that's a meaningful slowdown compared to the over 40 % growth per year that we really posted in the last few years. But it's also a more sustainable trajectory for a business that is entering its next phase. And account growth is the other part of the revenue equation. And I see that growing at a similar rate, but slightly below since new cohorts tend to spend more than the older ones.

1:07:41And that causes cent volume to increase slightly faster than account growth. And revenue is just a mix of the two. And so that's why I would say, okay, it will compound at about 20 % as well. And again, that's about half the historic leverage, but we saw growth rates decline year by year. So I do think that's more realistic. And this implies that the take rate would decrease from 2.3 % in 2025 to slightly below 2.2 % in 2029. And I know that's a minor decrease and perhaps we see more of a decline, but it's more or less in line with the historical take rate decline. And I actually didn't want to model out the take rate myself just to see how business changes that I expect would actually impact the take rate organically.

1:08:25Well, I know it's always hard for people who are just listening to the podcast to follow when we're using a lot of numbers, which is why I say on Spotify and YouTube, we have a failure, we always say this, we have the charts and the models shown on screen so you can actually see the numbers. And to that point, you know the company better than I do, but it doesn't seem like the top line growth rates are incredibly unreasonable here. And from everything I've learned from you, it seems like the runway is huge. And just looking at the historical growth rates, I mean, this is not a ridiculous extrapolation of what they've recently done.

1:09:00And I know from my new bank episode, there is this, as the kind of value investors that we like to think we are, there's this kind of unease in your stomach where you're like, did I really just put a 20 % growth rate into a model for what, five years? Like, really? Am I okay with that? With Nubank, it's actually maybe a bigger assumption because they're working from a larger baseline. And from a lower baseline for Remitly, just as a numbers game, those growth rates are more palatable. I totally get what you mean. Whenever you have to assume 20 % growth annually, I think the investment rate isn't that great when you just assume it.

1:09:43And that's kind of the base case because historical data suggests that's a realistic rate, then you kind of do it. But as you said, it makes you pause for a minute. But when you look at the growth rates in the recent quarters and the last years, I think it's something I can be okay with. And perhaps getting to the margin side of the business, which is even more important than top-line growth, 2025 should be the first full year in which we admittedly is GAAP profitable. Free cash flow-wise, it's already been there. It has$188 million in 2024 and over$360 million in the 12 months. And the huge difference between net income and free cash flow margins comes from stock-based compensation and most importantly, changes in working capital.

1:10:26And that's actually something that you brought up before the episode, just looking at free cash flow margins and net income margins, why they are so different from each other. A lot of that working capital comes from capital that is tied up to pre-fund the balances and those accounts that we talked about earlier. Depending on when the quarter ends, this can heavily distort cash flow. So when customer liabilities rise faster than pre-funding, so it immediately receives cash while the funds are not yet withdrawn, that's a source of cash flow and inflates the cash flow numbers. And if Remitly pre-funds balances in advance and payouts hit before they receive any money, Remitly's cash flow drops significantly.

1:11:06And in my valuation, I actually adjusted for that by trying to reverse engineer how much of the cash flows were due to favorable timing of cash inflows in 2024. Since the difference was significant in that year, the overall difference was almost$81 million. And I won't go through it here because it would be way too many numbers, but perhaps I could do it in the newsletter. Well, I'll say that it makes sense. But also, I asked you this question offline about the differences between their free cash flow and their net income. And I think I had to read your message like five times before something clicked.

1:11:42And I was like, okay, I think I understand what Daniel was talking about here. So for any listeners, if you heard that, you might have to play it back a couple of times and you might still not understand exactly. But the point is just due to accounting differences, there are some real reasons that are valid for why there are these huge differences in the net income margins and the free cash flow margins. And it really has to do with the timing of the receivables and payables that are coming into the business because there are these large flows of cash for a money transmitter as opposed to a normal company that their primary business isn't just the flows of money.

1:12:20So that kind of complicates the accounting. But regarding the stock-based comp. You did also tell me before our call that they are trying to keep this down going forward. And that Remitly's CEO, Matt Oppenheimer, has passed on his last three stock grants in the hopes that that will help with dilution. Yes. And to me, that's a great sign showing that he is focusing on shareholder interests instead of just his own interests. Although I got to say they're pretty aligned because he owns about$100 million worth of shares. And I would assume that's the majority of his private net worth as well. So he's pretty aligned with the company and the shareholders in general.

1:13:01It's also not unusual for a company in its early stages to have a lot of stock-based compensation. Reddit, for example, which is a company that we have in our intrinsic value portfolio and you're a fan of, and that only IPO'd in 2024. So obviously they also still have a lot of stock-based compensation. the size of I think was 24 % of gross profit. Uber is about 11 % and we're currently stands at 19%. And once again, that's a lot of numbers. And hopefully it makes all more sense if you just, you know, see the charts on Spotify and YouTube, and perhaps otherwise you just dig into the newsletter when it's out.

1:13:38And in my model, I assume stock-based compensation to grow at a CAG of about 6 % in the next five years. To see what that could mean for dilution going forward, I projected at which prices this stock-based compensation could be exercised. And depending on that, you will see how much actual dilution is and the share count. Doing all of that, in my model, in the base case, you will see a diluted share count of growing about 4 % CAGR. And despite that dilution, I still have free cash flow per share growing at a CAGR of 28%. That might seem aggressive, but we will see all these scale advantages years and the inflection point only coming in and to fruition in the quarters and years ahead.

1:14:19And this growth rate only implies a free cash flow of$770 million in 2029, when we already saw, given all the positive headwind of these accounting changes, a free cash flow of$360 million in the last 12 months. Well, so something you want to always think about with a young company, and you know this, Daniel, is the steady state business and what it can look like when the company is at maturity and it's not just kind of growing at all costs. And so my question for you is how profitable or is there a way to figure out how profitable Remitly could already be if they were only focused on just retaining existing customers as opposed to the additional spending on acquiring new ones because they're in the growth phase.

1:15:05I know you said that marketing is a huge expense for them. So I do wonder, as we're trying to quote unquote normalize their earnings if it would make sense to adjust for that. I'm glad you asked because I spent way too much time coming up with what I thought of as being a steady state model. So only enough marketing expenses to sustain the current user base, no additional growth. And I'm afraid this doesn't make it less complex, but I will try to keep it short. So I'll walk through the math in more detail in our free newsletter, as I said, that we send out each Sunday and basically discussing the company that we also pitch on this show before.

1:15:43But basically what I did is to reverse engineer the average customer acquisition cost by using some of the management comments throughout the years to base my assumptions off. And doing that, I get to about$90. Now, new data from Q1 would suggest that the customer acquisition cost might be closer to$85 already, but let's stick with$90 for now, since those are the full year numbers of 2024. If we immediately ever stop chasing growth and just spend enough to replace churn, which is about 10 % per year, you would only need to acquire those 10 % per year. And on an 8 million active base, that's about 800 ,000 people.

1:16:23So replacing that with costs of about$90 in customer acquisition, that's about$70 million in maintenance marketing versus nearly$300 million that Remitly currently spends today. And after tax, we'll be talking about additional free cash flow of about$170 million. All of this was a long way of saying the financial power is already there and it's just about when Remitly will show it by spending less on customer acquisition, which by the way, I still think is the best use of their cash right now. It's a great way to think about a business that is in growth mode it and might seem less profitable on paper than it is in reality, which is kind of a weird concept.

1:17:05But it does remind me a lot of those conversations we had about early Amazon when you pitched that company a while back. And Amazon is sort of the famous case study on a company that has more earnings power than it looks on paper because they were investing so much in growth that actually suppressed their profitability. And it is interesting to me, though, that that you always hear people talk about the difference between maintenance and growth capex where maintenance capex refers to the minimum amount that needs to be reinvested just to maintain the business at its current size. Whereas in marketing, you just don't hear about that concept as much.

1:17:43And so I think it was really an insightful calculation for you to put together because I applaud you for breaking out maintenance marketing and growth marketing. I getting back to your base case model after hyping you up a little bit, Daniel, what fair price do you see for Remitly? All right. So first of all, thanks for the compliments. I think most of the time marketing is not as big of a thing for companies as it is in Remitly's case, but here it definitely makes sense thinking about this. So getting to the base case, I'm assuming a 14 times multiple of price to free cashflow, which is a bit higher than the current multiple of only 10.

1:18:22But in my opinion, more than fair for a company that I expect to grow revenues at 20 % and free cash flow at 30 % per year. So this would give me a fair value of$26. However, since the range of outcomes is just much more uncertain in this case and in any of our other cases, I do slap a 20 % margin of safety on it this time, which brings us to a current price target of about$20,$21 in the base case assumption. and this would still imply an IRR of 16 % from today's stock price. I feel like I know how wide the range of outcomes can be with these kind of companies and we've kind of alluded to that earlier.

1:19:02So I would presume that in your bear case model, the price target might be dramatically, dramatically lower in a worst case scenario. It is, it is. And in the bear case, I still expect double digit revenue growth but significantly less free cash flow growth than in the base case, only 15%, which, you know, still is materially growing. Honestly, I don't think this is the most likely outcome since the steady state math shows that we midly should already reach the free cash flow levels. This growth rate would imply for 2029, but who knows? Stock-based compensation could dilute shares way more than I modeled here, and perhaps retention rates decrease with more competition, leading to higher customer acquisition costs, less growth in the top line than I expect.

1:19:49And considering all of that, I get a price target of about$7 in my bear case. In my bull case, you see the variety of possible outcomes play out to the other side with a fair value in the high 30s, driven by slightly faster growth and better conversion to free cash flow, but mostly due to a significantly higher multiple. I assume 21 times price to free cash flow in this case. And it's not that hard to argue that a company growing at these rates would deserve those multiples. Some peers are already trading at those. But looking at Remitly right now, it does seem to be a bit of a stretch. And perhaps the bull case would be mostly about higher growth rates.

1:20:29But as you know, modeling these companies, at some point, you just don't want to assume more growth than you're already expecting. All in all, even more so than all other episodes, please take these models with a grain of salt. Forecasting five years into the future for a company that is so early in its journey is pretty much impossible. I would focus much more on the steady state method we did. And from there, you just have to be convinced that Remittly's product is actually better than the competition. But also the idea that this space will not be commoditized due to the complex nature of the Remittly's business.

1:21:04So while to say it in Wall Street language, the model would suggest a buy rating. I like to remember the investing wisdom. Valiation isn't one on spreadsheets. It's one on conviction about the range of possible outcomes. And this could still be a hit or miss. Despite many things currently hinting to the former, it's a bet that I might take my personal account, but I don't think it fits the risk profile of our intrinsic value portfolio. and I would just assume you agree with the concerns but tell me if that's not the case tell me if you are completely sold on Remitly I don't want to I don't know I don't want to say too much specifically on Remitly because I'm over here saying that I'm not a payments expert so who would I be to to argue with you on you know banging the table for why we should add Remitly but I'm not either yeah right well true exactly yeah so I I would just want to say that for anybody who's listening to this and they're saying, well, you know, it's kind of puzzling that your models say one thing and then you're doing something else.

1:22:07And it probably feels subjective to longtime listeners of the show because sometimes it might seem like we're choosing to closely adhere to the models we build. And then other times we're kind of whimsically dismissing the valuations. And as Drew's suggesting here with Remitly for non-quantitative reasons, I would say, and I've talked about this before, but it really is worth emphasizing that we are breaking down a different company every single week. And we're doing dozens of hours of research on these businesses before we distill it down to a model and then talk about it on the show. And the point is that we're churning through a lot of businesses and it's just not possible to be equally confident in every model we build.

1:22:48If somebody tells you they're equally confident in every model they build and they're doing 30, 40, 50 models a year, I wouldn't trust them because it's just it's just not possible and so I know from experience that sometimes I build a model and I feel like I have a very strong grasp of what's going on with the business and the industry and the competition and therefore I just I have so much more conviction and comfort in the assumptions in the model and then with other companies it's just not possible to feel the same way and circle of competence is a real thing and it's so much easier to value businesses that you've worked with or you've used before or you have some expertise into the industry for whatever reason.

1:23:29And with Remittances, this is a world that we both have very little familiar in with outside of the research we've done for this episode. And so when you layer that over a pretty fundamentally uncertain business that the market itself is unsure about, that just makes everything so much harder. And that really is why the concept of the too hard pile exists. It doesn't have to mean that the stock is banished and is always too hard, but it means at this moment in time, relative to our other opportunity costs, we would probably be better off spending the next hour of research looking at something new rather than going deeper on Remitly.

1:24:06That's how I think of the too hard pile. And that's what I say. I think we set aside Remitly for now. It's in the too hard pile. But, you know, a lot of our favorite writers on Substack and some of our friends and stuff follow Remitly closely. So I don't think this is a company we're going to just totally forget about. And maybe at some point down the road, and this is kind of a cop out, but it's true. Maybe we'll revisit it. I would also add, everything that you said is totally true. And I think the two hard pile should be exactly that. I would also add, though, that one thing that you have in a model is the price target, which is blended from all three scenarios.

1:24:39And the other thing is looking at, wait a minute, what downside does the bear case even have? So if I have a model which suggests a certain upside, but the downside might only be 20 or 25%, I can have a lot more conviction in that. But if my model shows a blended price target of, let's say,$20, and that is a pretty good upside, but the bear case is 60 % below the current market value, then I just also have to think about the risks coming from this. So we always talk about a range of outcomes. And when the downside isn't capped, I would also want, for one, more upside. But there is a limit where I would say, okay, if the downside is 60%, 70 % and it could actually happen, then I simply will not invest in this company.

1:25:23Because one thing that value investors want is to not lose money. And I might miss out on, you know, some upside here and there. but if I see a bear case to make and we talked about it for hours in this episode but also before, there is a bear case to make here and if that means the company might go to zero which is also totally possible at some point I just don't want to invest in it despite the blended price target might tell us well this is a buy target if you just look at your 12 % hurt rate I think that's very important as well and as you phrase it, those stocks would exactly go on the too hard pile As Munger would say, Daniel, nothing to add perfect all right then with that how about you give us your hints as always for the next episode yeah so next week we're going in a bit of a different direction and i can assure the audience they'll not have to listen to another payments business again because i kind of don't want to listen to any more about payments i appreciate the pitches daniel but i it just it makes me tired i don't know i it's just so much to wrap your head around um this actually my picture next week is a company that first came across my radar a year ago.

1:26:34And I liked it then, but not quite enough to buy it. But it was kind of at the top of my watch list. It's only gotten cheaper. The stock has fallen another 20 % or so. While at the same time, the fundamentals have continued to compound. So it's a company that, well, I guess it'll make more sense when you see the episode, but we'll end up comparing it with Nike a lot. And they're actually quite different though. And the fact that Nike sells primarily wholesale and the company I'm pitching mostly sells DTC through its own store. So there's some difference in the brand positioning. And I haven't given away too much as maybe a last hint.

1:27:13It's a company that's branding fundamentally leans very heavily on what I would say targeting the upper middle class. And so you're likely to see this brand worn at Starbucks, but also at the gym. I think those are pretty good hints. You could get it if you know the brand and perhaps you're also aware of the investment case. And with that, let me close it for today by quoting the one and only Warren Buffett, who gives us some wisdom on not neglecting the risks in an investment and appreciating the fact that there is a too hard pile and that there are more opportunities than just this one. Quote, the stock market is a no-call strike game.

1:27:55You don't have to swing at everything. you can wait for your pitch. And since I know next week's company already, I think there's a good opportunity we might strike then. So with that, I hope to see you there. Have a great day.