Georg Steiger is using BNPL to expand access to credit in the Philippines

The Philippines is home to one of Asia’s best-performing economies, but many of its 60 million adults may be missing out on the party thanks to their limited access to credit. In traditional lending models with heavily-manual processes, smaller loans were often simply too expensive to issue, but a new wave of entrepreneurs is looking to set that right.

Georg Steiger is one such entrepreneur. With BillEase, Georg and his team are leveraging cutting-edge data tools and a buy-now-pay-later business model to issue small value loans automatically and safely.

I speak to him about what first drew him to the Philippine market, and how they’re re-thinking data and data modelling.

If you have any feedback, questions, or if you would like to participate in the show, please feel free to reach out to me via the contact page on this site.

Regards,

Brendan

You can read the full transcript with timestamps here:

Georg Steiger 0:00

I think it's a very interesting market, Southeast Asia in general, but the Philippines in particular - and it's a great segment, it's the generation that will drive the economy in the next in the next 15 to 20 years. So it's a great way to acquire these customers and basically help them to get into the formal system, that's kind of our long term vision: you find these customers when they first get started and stick with them, build the products that they want... and the sky's the limit.

Brendan Le Grange 0:45

Welcome back to How to Lend Money to Strangers, the podcast about lending strategies across the credit lifecycle and around the world. Today, we're visiting the Philippines where I speak to Georg Steiger, an Austrian born entrepreneur, who co-founded First Digital Finance, a FinTech that's using advanced machine learning algorithms, and a buy-now-pay-later business model to extend credit to the under-banked populations of that country.

Unknown Speaker 1:11

So I'm originally from Austria, from Vienna, I actually studied law, then decided I don't want to be a lawyer. Did a bit of tech development on the side, so I actually was a bit of a programmer for a while, and then went to us to business school, I met my wife there. And then we moved both to Zurich, where I joined McKinsey, in the financial institution practice. And after two years, I wanted to see a bit more growth-focused markets so I transferred over the Singapore office, spending about seven years in Singapore, focusing on retail banking and risk management - across the region, so I've worked in Indo(nesia), in Philippines, a lot of time in Vietnam, Thailand, ja, even Dubai. So yeah, great, great opportunity to see different markets at different stages of development.

Brendan Le Grange 1:58

Yeah, let's pick up on that point about "growth-focused". The Philippines is a market that's not often spoken about, with the same excitement or in the same conversations when we talk about India and China, and their rapid growth. But the fact is, over the last decade or so, the Philippines has often been Asia's fastest growing economy. Now, there's certainly a lot of growth, that still needs to happen but this really is a fast growing environment. It is, though, also a market that has quite a lot of diversity in terms of its level of development - if we contrast some parts of Metro Manila, to some of the further flung islands and provinces, it's a very different environment.

So could you maybe give us a bit of an insider's view of what it's like to lend in the Philippines at the moment?

Georg Steiger 2:44

It's a very interesting market, actually. Because on the one hand, it's actually a bit more, I will say, Western in terms of feel, of the whole culture and and environment because of course, with a close relationship to the US and Europe, and large parts of the population are actually English speaking. So it's actually very accessible for say, Western companies, Western Western people.

On the other hand, it has it has received much less focus in terms of both, I would say multinationals, really pushing hard here and also in terms of investments in the in the FinTech scene. So if you look at at, like Asia, of course, China and India is where the big bucks go, and then one level below maybe is Indonesia, right. So if you talk to investors, and they're talking about Southeast Asia, that's 90%. Actually, Indonesia, Indonesia has actually developed very fast in the last, say, five years in terms of market, especially electronic transactions, ecommerce, but also FinTech. And Philippines, I think, is just really at the at the part where the curve gets gets steep. So there has been actually, just in last two weeks, a couple of exciting deals.

We have seen a huge push to online during the whole pandemic, because we've been on lockdown for more than a year now. So behaviour has really, really changed. And no amount of marketing money could have bought you that change in behaviour that we have seen. Like, for example, cash used to be by far the (most) prevelent payment method. And now, a lot of customers have just shifted to ewallet, basically overnight, because suddenly they were scared of touching cash, and buying stuff online. So now really everybody accepts, for example, payments via ewallet, that has really accelerated the pace of digital adoption. And I think we will see that trend continuing over the next few years. So that's an exciting place to be right now.

Brendan Le Grange 4:14

And just geographically, I think it's a country that stands to benefit a lot by the move to digital. I can't remember the exact numbers anymore, but I think it's over 7,000 islands that make up the Philippines. So a lot of people are really just hard to reach physically. So if we think about a branch network or even just where you might want to send out things like credit cards by mail, it can become very expensive to serve these populations, especially when we're talking relatively low loan values.

When you came to the Philippines, it was as a management consultant, and you were working with some of the more traditional banks, but you obviously saw something while doing that, that made you want to strike out on the entrepreneurial journey. So can you talk to us a little bit about what you saw in the market, and then also talk to us about some of the brands that you've launched? I introduced you as the co-founder of First Digital Finance, but that's not the name necessarily that people in the market would know your work by.

Georg Steiger 5:31

Yeah, so I spent two years working with a bank here, helped them set up their retail business and that was kind of my segue into the market, before I tried to start something on my own. But that was the long-term plan. And also, with my friends who were not my co founders, we saw that the processes were still very traditional, very paper-based. And what that means is that the operating costs per underwritten loan are just way too high to make sense for 90% of the population. So the ticket sizes, the tenors and so on, that would make sense for a bank to underwrite given the operating costs that they incur - because that is not different, right, if you make a $100 loan or $1,000 or $50,000 the cost of underwriting that is actually not so different, right, because you still need to do your DD and check on the customer and so on - so then given the high cost, it basically means that with the prospect like that you cannot serve 90% of the population profitably. That was the basic premise.

And if, if you apply modern technology, cloud based technology, quick development cycles, plus, back then the you already started to see like, a lot of the machine learning and AI advances emerging, so basically, if you apply all of this stuff, can't you make it cheaper and more efficient by at least one order of magnitude? That was the basic premise how we started, we started with really looking for segments that we felt were underserved by the banks. So one of the segments was OFWs (Overseas Filipino Workers). The Philippines has a large labour export industry - so think about 15% of GDP is actually remittances from from abroad. So, many Filipinos go work abroad for better opportunities and then remit that money back home. And to go abroad, often, this is quite the financial burden. And some of them get that from the family, but then others need to borrow for it. And it's a loan that's really life changing for many of them, because they go from earning, I don't know, $300-$400 here,when abroad they can earn a multiple of that, right. But again, for a bank to do this extremely complicated, because the documents are often in foreign languages. So for example, if a customer presents a document in Arabic - a lot of these customers going to Arabic countries, right - to a bank officer he would probably get it translated, right, we just use Google Translate and say 'okay, yeah, looks like t's right and it looks like the other five that we've seen in the last few weeks, so it's probably not not a problem', right?

So that was one product. And then another product was Loan Ranger, that was actually one of the first online lending sites in the Philippines - both are still running, but then in in 2017, we said, okay, this can't be it, because both are running profitable, but not really scaling and reaching the number of customers that we had hoped. So it's okay, let's think what we can do. And then we basically had a discussion with Lazada and said, hey, you guys are doing great stuff on rolling out ecommerce, but we all know that payments is still an issue. It's all cash on delivery, very few customers have credit cards and the ones who have, don't want to use it on a website - so why don't we do something together in payments?

And that's how we set up BillEase and, since then, we have basically invested most of our resources in building out that product and that platform, added now many more products - so it' now a full-fledged app that customers can get Lazada top-ups, they can get G Cash top-ups, PayMaya top-ups, all the major e wallets, they can get bank loans (but that's only switched on for loyal customers), they can now shop at more than 100 merchants where we're directly integrated. So you can, in the checkout, just pay with BillEase and you have mobile load, gaming loads, it's basically anything a customer could want. They can they can do directly from the BillEase app, right?

Brendan Le Grange 9:21

So you using buy-now-pay-later as a way to help that migration in the portion of the population that I guess would have been underbanked or sort of at least seen as difficult to lend to in the past. It's been about five years now since I've worked in the Philippine market, so my numbers may be a bit old, but from memory there were only about 2 to 3 million consumers with a credit card and that out of a population of 110 million people. So really low penetration. And if we think about how would you have done online commerce without a card, it would be really difficult without some sort of cash on delivery type process which again, very expensive to maintain for low value transactions. Instead, you're now stepping in and giving lots of small short term loans to these consumers, and doing the logistics of the cash, or at least the payment to the merchants to enable this population that existed outside of the reach of the big formal banks to be brought along, in the rush to online commerce.

Georg Steiger 10:19

Correct. Correct. So, card penetration still very low, around 2-3%. Even for our own customers, right? These are not unbanked or unbankable, I would say underbanked. So many of them have, or most of them have, bank accounts, at least for salary, but these are very rudimentary accounts. A typical customer will be 25 to 35 years old, so not like the first job but have a little bit of experience already, we have a lot of call centre agents, BPO workers, that's a very big industry here in the Philippines, kind of unloved a little bit by most of the banks because they change jobs very frequently, we have a lot of freelancers, even bankers... So it's, it's about, say, the young professional segment, the emerging middle class that we're trying to service.

And you know, credit card is a bit of a tricky product when you're new to credit, because a lot of customers take the credit card and then run up the limit, and then just keep that limit, which is of course very profitable for the bank, but not so great for the customer. So a term product, I think, is actually a better entry for many new to credit customers, because they know they have to pay x over three months, six months, whatever it is, right, and then they are done with it, right, but it's not like a balance that automatically always stays.

Brendan Le Grange 11:29

In my old role, I did some research on this, we called it the payment hierarchy: essentially, when a consumer with multiple debt obligations falls into financial trouble, which of those obligations gets paid first? And what we often found is that, although if you're looking at a normal portfolio view credit cards tend to be less risky than personal loans, when you look at these troubled consumers, invariably, it's an instalment loan that has been paid first, and the credit card is the one that's going into default. Now, I never looked at the data in enough detail to prove it scientifically, but my gut feel is aligned with yours, that a lot of this has to do with direction of travel of balances. So on day one, when you take out an instalment loan, you can budget in your head, it's kind of the most you're ever gonna have to pay per month, and you pay that same amount. But for a credit card it works the other way, you're starting with a zero balance and zero obligation and then, over time, that will slowly build up until maybe in a year's time when you're now having to pay the full balance back. And perhaps in that time you've taken on new debt or new obligations, or simply your situation has changed, and thus you've created all this affordability stress. So I do think instalment loans are a better solution in the space.

But I guess what was problematic in the past, when we think about those economics of making the loan that you spoke about before, we used to have to give a rather large lump sum to justify the costs involved. But if you give somebody a big lump sum of money, yeah, well (1) it is a bigger risk and (2) you're creating some of those same sort of affordability problems, they might have all good intentions, but once the money is in their wallet, it could get spent. So what you're doing with buy-now-pay-later is in some ways, I guess, like a bottom of the pyramid approach. When I was in business school we spoke about this idea that, you know, for us, we can go to the store, we can buy a big two litre bottle of shampoo, and we can use that over four or five months, and we can benefit from the savings of buying in bulk. But for low income groups, they don't have the money to make their big purchase, and would in fact, prefer to buy a lot of small bottles. Even though over time, it might be more expensive, in terms of the management of cash flow, they don't have that single big payment to make. And you're kind of doing that, in some way, from a lending point of view. So we're not going to give you one big lump sum, we're going to give you lots of small ones as you need it. And because each loan is smaller, you've got less risk at any one time, and you're creating all these opportunities to learn.

Georg Steiger 13:52

Yeah, it's I think they call it sachet products, right, these little shampoo andstuff, right? So it's like sachet loans. That's right, it's like training wheels, right? So we try to give the customer something where we're pretty sure they can afford it, and as long as they're willing, they can they can pay it back again. And you actually see that also, with customers they are quite happy to do that. Of course, the bigger the credit they can get often the happier they are, but they're absolutely fine with saying 'okay, I have to prove that I can actually handle this'. And at least you guys give me a chance, right? And you charge me fair rates. I mean, there's a lot of payday apps in the market that charge pretty, pretty high rates, which is not really helpful. We kind of peg it close to where the banks are, and cards are, but much smaller tickets. Then once they prove that they can actually meet the payments, we very quickly review the limits and adjust the limits and all of this done fully automated again, to keep the opex down.

Brendan Le Grange 14:45

You are, though, tackling the problem with some pretty sophisticated analytics as well. This is not just a case of 'well, it's a small loan, we'll take a gamble'. You've got machine learning, AI modelling coming in, and based off some rather interesting data fields - the traditional credit data fields, but also all sorts of alternative data sources.

Georg Steiger 15:04

Yes, absolutely right. So, I mean, without that it would not be a model that's sustainable. So we we do review every customer, we have managed now to automate about 90% of approvals fully, so that means without any human intervention.

Basically, what you do first is... there's always two aspects, right? One is fraud. And the other one is credit. On the fraud angle, just basically make sure you you can somehow tie that loan to a customer. ID is not very centralised here, so there's, I think, seven or eight different types of ID that they're accepting. Government is now launching a Philippines ID card that should be available for everyone and for lifetime. So once that is rolled out, I think the whole question of identity will be much, much easier to solve. I hope there's also some API, so you can check it out, similar to what India has done. So that would help a lot on the fraud angle and the identity angle. Until then, we look at biometrics, we look at fingerprints, and so on, right?

And then the second one is credit. And there we use, really all factors that we can find, right? Because it's very data poor, or there's very little traditional data, right, so only about 20% of our applicants have a credit bureau score, or credit bureau entry, we actually built our own bureau score, but it just requires a little bit less data than the traditional score. But even with that we only get about 20% decision rate, and 80% don't have that. And then how do you decide on on these customers? And so we use a lot of unstructured data that we collected over the years. And that that was the hard part, right? I mean, there's a lot of little things that have to come together, which push automation rates up to that to that level. So we actually just managed to do that last year. So took us took us a while here.

Brendan Le Grange 16:50

One thing Filipinos are famous for is a, sort of love of the phone, the selfie, and a high penetration of smartphones. Are you gathering all your data from the associated fields? Or are you asking consumers to add data in via traditional application forms?

Georg Steiger 17:08

Both, actually. So we use data from the smartphone, we use demographic data that they enter, we use other information, say on the mobile, email, IP, basically a lot of inferred information, we have some behavioural stuff that we're checking like how the filling out things and so on. So there again, you you have to be very creative in understanding what are factors that could hint at some aspects of credit quality. And then figure out how do you take these factors and quantify them, or what are proxies that you could look at, and then be very meticulous in collecting the data over time. And just running a lot of experiments and testing what what works and what doesn't work.

Brendan Le Grange 17:51

Actually, in one of the episodes coming up soon, I speak to some people that are doing big data analytics in Latin America, Africa, the developing world where they're concentrating on all these new data fields. And he says one thing to sort of switch your mind around with when you're thinking of modelling these fields is: we're used to having very few fields, but each of those would be really powerfully predictive but in the AI world, we don't have a lot of super predictive fields coming from these alternative sources, but we have so many of them that together, they actually can create the same effect, or something close to it. So it's a change of mindset on that front.

But for me, what's also interesting is that there have been a lot of lenders who've been having access to this data, some of them may have been gathering it, but very few have found a way to use it. Because if you're in a big developed market that has lots of traditional data on all your core customers, you haven't had the need for that uplift, perhaps people have tried it a little bit in marketing but from a credit point of view, they've said, well, delinquency is the gold standard, I'll just use that. But now it does feel to me, as I have these conversations, that more and more people are finding genuine analytical uplifts in these alternative data fields. And I think the big lenders who have been sitting back and not needing to explore that, are going to have to worry about their ability to use that data, if they're not doing so at the moment. How do they embrace these fields that can tell us some different stories about our customers in ways that those of us who have come up the traditional route of credit risk analytics have barely thought of.

Georg Steiger 19:25

So I mean, you need to collect a certain data point for quite some time before you can even start testing it, right, because you need to then see outcomes, you need to see trends. So you have to be thinking ahead quite a lot, of what could ever be a useful data point. And then then make sure that you already collected in a format that that you can actually crunch later on. That's actually been also quite a bit of our focus to make sure that we collect the right kind of data, that we collect it in good and consistent quality, so that we can always come back and say, okay, if we now change the model, we check how those come out.

Brendan Le Grange 20:02

And I think what's really helpful in this current movement, is that we're actually leveraging a whole lot of data that's already pre-captured and built by third parties who understand data. By which I mean to say, I was involved in, well not quite the launch of the credit bureau in the Philippines, but the early days of building the credit bureau out in the Philippines. And we were working with, you know, just the big banks who already had lots of traditional data, had it stored up well, and even that process of getting them all to agree on standardised layouts, and processes for getting the data to the credit bureau, you know, agreeing on things like special characters to use, was difficult. And then, you know, the government wanted to expand the reach of the credit bureaus, wanted to bring more people on board, which is great, exactly what we want to do but it is exponentially more difficult. How do you treat lenders where the data, well one, it doesn't exist or two, which is more common, it's patchy, so an account is there one month and falls away the next month, is back again. Are you better served, is the community better served, by bringing that data in, or by leaving that data out? It's really complicated. And so rolling it out in this way, takes a long, long time.

But actually, you've got all these organisations storing data already, the mobile phones are already capturing that data, it's in a few formats, through API's and things we can bring it in, and we can start using it without the need to try and convert all these businesses' data - it feels a lot more ready-made than we might have thought, you know, when we're thinking about ways to expand bureau reach in the past. Of course, you're going to have demographic fields and some unique fields that you're going to want a customer to fill in themselves.

Georg Steiger 21:39

Correct. Correct. So we we basically are always on the lookout for new data sources, or external providers. And whenever we see something that's interesting, we basically test it. Even if it's a small lift, if the data source itself is not giving you huge gini, right, as long as it's uncorrelated and it covers a new aspect, it can actually give you a decent lift. In the end, it comes down to what can we pay per gini point of lift? And there are some data sources are a bit too expensive, that they don't add as much as they would cost us, but then some are fairly reasonable at a few points of lift so we say, okay, fine, why not?

The difference when you go from linear regression - that's traditionally used - to, say, more machine learning type model building, is that with linear regression you always have to reduce factors, right, because if you want to go beyond say, I don't know, 7 to 10 factors you very quickly go into overfitting territory, right. But with machine learning, actually, the algorithms are quite hungry for, or some of the algorithms actually, are quite hungry for lots of factors. And also, you can throw in linear nonlinear types of data, right, and combines it very nicely. And what we have seen is, the more data points you use, or at least you consider in your model building effort, the more stable is the outcome.

Brendan Le Grange 22:56

One of the concerns you get when you speak to more traditional, more conservative lenders, when you talk about using AI and machine learning to build out credit models, is that there's a concern that it might create a sort of a black box where you don't know quite what it's doing with those 1,000 inputs. And the worry that either internal audit, or the regulator's gonna push back on that. (Yup) How do you think about that process of understanding the machine learning model?

Georg Steiger 23:22

I will say we focus on performance, rather than explainability. We validate our models, just like you should, with validation sets, we check them all the time. And, again, I think retail lending has to be a numbers game. And even if you don't understand exactly what's going on inside the model, as long as you can test it, and you can prove that the model and the process works. I think that that's what you should focus on not understanding the individual cases necessarily.

Brendan Le Grange 23:51

Yeah, that makes sense. So basically building the organisation so that it's really quickly reactive. So you're testing all the time, you're monitoring all the time, and if something goes wrong, you'll know about it immediately. And you've got the organisational culture to work quickly so you can implement a fix as quickly as needed.

Georg Steiger 24:09

Yeah, I mean, we look very closely into our vintages, by model, by product, by any segmentation you can think of. And so we catch it very fast if we see things not going the way they're supposed to. And then we can also act very fast. Yeah, then then we can just change the factors see what's going on. I mean, and I think that that's a bit of a myth, right. Even in a machine learning model, there's methods to say which factors are actually heavy versus not heavy. So it's just a different way of looking at it, right. So can you tell which factor or which five factors have the biggest impact on decision? Yes, no problem. You can tell that with XG boost you can tell it with random forests and also neural nets, I think. The thing is just you can't put it into a neat formula, right? But that doesn't mean that you don't understand what's going on.

Brendan Le Grange 24:59

That's actually a really good point, because most of the concern about this black box is the idea that we might perpetuate biases, human biases, but that's normally built around things like 'are we unfairly biassing on gender or on race or on some other factor like that?' And by knowing which fields are influencing the score, we can tell if we're doing that or not, it's not so important that we know the nuance, the complicated innards, as long as we can exclude fields that are of major concern, we know that they are not the ones influencing the model, we should be comfortable on that front. So this is actually something perhaps blown a little bit out of proportion.

Georg Steiger 25:34

Correct. I mean, and you're not only looking at credit outcomes, right? You can, at any time, look at score distributions, you can look at, has my approval rate jumped, right, if it goes from x to 2x overnight, then maybe there's something wrong.

Brendan Le Grange 25:48

And I also think that your business model lends itself to creating more data points, than perhaps a traditional lender would get. Or not so much data points, but data opportunities, so you got more chances to see whether a consumer's making the obligation or not. I used to do some work in collections on this sort of theme: often a collections agent would phone a customer who was in arrears, and get a promise to pay, but they would set that promise to pay for that customer's next pay date, which makes sense from the customer point of view, that's when the money's coming in, but if they missed that payment, they're basically missing two at once, so you've only got one opportunity to learn. I would always be encouraging them, particularly for higher risk customers, to move their promise to pay pre-pay day. Now, it's not always possible, but if you can do that, then you can see: first, do they make their promise to pay and then, second, do they make the next month's obligation? You've created two opportunities to learn. And you're doing that with short term loans with inter-month payments.

Georg Steiger 26:44

Absolutely right. So we see much more clearly if customers are sticking to the contract or not. Again, in cards, there's there's much more delay because they take some time to ramp up the balance. And then they can pay the interest for a while, right, but you don't see, you don't see credit outcomes necessarily with the speed that we will see it.

We actually set the payment frequency to align with a customer's paydays. So in Philippines, actually, most customers get paid twice a month. And we ask them that when they sign up, and that's how we set their due date. So it always aligns with their pay days. Because often delinquencies are just driven by the fact that by the time you're collecting that money is already spent. So it's much better in terms of convenience, and also helps customers to manage their credit easier.

Brendan Le Grange 27:28

Timing that strike date is something we can sometimes forget about when we're doing prime lending in developed markets. But when I was working in Africa, it was also a core part of our strategy when we set up an account. And then if they were a customer of our bank, we would model it as well - what is the right day in the month to run the debit order? And we would see huge differences around Christmas, for example. Let's say everybody was traditionally paid on the 25th of the month, what most companies would do for Christmas to give their staff some money to spend over the holidays, they would move pay days earlier, so maybe the 15th. And then all our debit orders will be running 10 days after payday. And the change in risk was dramatic. And if you tracked that over the months, it could take a customer five, six months until they were back all the way clear again. So I guess in that way, it's once again, you helping your customers to budget better, and particularly thinking of these customers often being new to credit, or relatively new to credit. It's of course better for risk as well, but you're avoiding a situation where a customer might be spending money they don't have.

But talking of new to credit or relatively underserved customers, do you see yourselves as a route into formal financial services for your customers? Or are you seeing your customers stay with you and you alone in this buy-now-pay-later space?

Georg Steiger 28:47

And so actually, we do see a lot of customers, I would say, graduating into more traditional products. So because we are a member of both TransUnion and CSC, so we report all our loans and so we do have some customers who then after being with us for a while, they're okay now applied for home loan or car loan, and was no problem, right, because they saw that I have very good payment record with with BillEase. And so that actually helped me to get into the financial system.

Brendan Le Grange 29:17

Yeah, that's really great to hear, particularly in the context of some of those industries you spoke about earlier, servicing the call centre workers, I know from when I was there, you know, it's the biggest part of the economy, the biggest employer in the economy, and yet all the formal banks say that 'we just, from a risk point of view, don't know who to lend to'. And so they're going to benefit too, because these are lenders with cheap capital where if they can find a low risk customer, they can reward them with things like a mortgage or a car loan, something that can really improve their lives, but they don't know how to find them. You're creating these customers, these good low-risk customers.

That said, you have caused not just a route into the form of financial services. You're also a full-on competitor because you provide a really convenient service and I think, when we talk about buy-now-pay-later in the developed world, it's all sold on that convenience and I saw in a press release that you're actually even looking at some zero APR buy-now-pay-later offerings in the Philippines, is that something you're able to talk about now or is it still being launched?

Georg Steiger 30:17

No, sure, we actually have a couple of offers live already. So we have a couple of merchants that have implemented this thing. It's not yet very common in the market, mostly because the expectation on the merchant side is that it's very expensive, it's actually not that bad, probably a little bit more than you would pay in, in more developed market, just because risk is higher here, the cost of funding is a bit higher here than, say, in Singapore or Europe or wherever. But we can do zero APR loans starting at single digit subsidy rates for merchants. And that's something that's really, really popular with customers, of course. These are people who have actually quite a bit of responsibility, they're the breadwinner, so used for like living expenses of the family, and so on so they don't have too much that stays with them. And so for them to save up for a mobile phone or computer, things like that, it's hard, because there's always some emergency that comes along. And so that money gets drained. And if you can offer them a nice instalment plan doesn't even have to be long, right? If you buy a phone for, I don't know, 10-15,000 pesos, you don't need a 24 months instalment, you do it three months, six months, and you can do that at zero APR, it's no cost to them and so it's a very attractive product to do something also nice for themselves.

Brendan Le Grange 31:26

It's such a simple idea but it really is a model that resonates well with consumers. And we're seeing in all the discussions we have in the UK, every credit card issuer is just asking, what do we know about buy-now-pay-later? how can they push back against these new competitors?

And I'm slightly putting words into consumers' mouths, but I wonder if it does reflect some of those concerns we spoke about earlier with the credit card model and the tendency to have this credit limit sitting there tempting you to spend. When you're doing buy-now-pay-later, there's none of that, the consumer is looking at an item and buying it, they don't have a limit available to them in the back of their minds. So it's a little bit more like an instalment loan. And I wonder if that's part of that appeal, as well as obviously the low cost and the ability to to shop online. So I'm not surprised to see it taking off, but I do love to also hear the spin that you've put on it as this route into the credit economy.

Georg Steiger 32:21

The difference is really that in in more advanced markets, you have, of course, a lot of providers for many different products already, right. So in Singapore, if you want a personal loan, you text the number with your ID and they'll give you a reply, right? Because Credit Bureau is like with high penetration, and they can score you very easily right. Here, that's not so easy, to get a loan it's a bit of a hassle. Usually you have to bring like documents and so on and so forth. Since they already have the account with us, it's very easy for them to just also migrate to other products, so the cross-sell or upsell becomes very natural in the app, which is very different from how it would look like for a card transaction or an offline transaction.

Brendan Le Grange 33:02

Indeed. Thank you so much, Georg, it's been an absolute pleasure having you on the show. If people want to learn more about your business, where can they look to find some more details?

Georg Steiger 33:12

Very easy. So they can look at our website, BillEase.ph, like Bill and Ease, or just look in the App Store, Play Store with the same name for our app.

Brendan Le Grange 33:23

Excellent. So thank you again, and thank you for listening. This has been How to Lend Money to Strangers, the podcast about lending strategies across the credit lifecycle and around the world. Focusing on that second part, in next week's episode I'm doing a deep dive into the consumer credit economy in China. China is obviously a huge market, and also a market that's delivering some really exciting FinTech innovation.

I've heard some of the headlines that have come out from that, but I wanted to go a little bit deeper. So I've brought in two experts, both of whom have worked for banks in North America, before moving back to China for the last decade or so, and both playing pretty active roles behind some of that growth that we've heard about. So join me for that show next Thursday, on the Apple podcast player, Spotify, or wherever you're listening to this one.