Nan Wang
Questions
- Let’s start with the market landscape. We see issuer processor focused providers, such as Lithic, Bay1, and all in one platform, such as Bond, Productfy, Unit, which integrate with third-party providers like Marqeta and then the FinTech would just only speak with this layer without managing other technical and regulatory elements. What differentiates these two types of providers, the vertical issuer processors versus the platform providers?
- Would the issuer processor integrate with Modern Treasury to provide that service?
- In terms of how these providers differentiate themselves, I'm hearing there's a differentiation to how much exposure the FinTech wants to have with the bank…
- Why is having a direct relationship with the sponsoring bank important?
- If you are starting out today and looking for a BaaS platform, given the increasing activity in the space, how would you select and decide who would be an ideal BaaS partner?
- What allows faster speed to market and what could impede it?
- In terms of interchange, if we're looking at the consumer segment, broadly speaking, we start with 1.35% interchange for the whole value chain…
- How's the interchange split between the FinTech and the BaaS platforms currently? How do you see this dynamic evolve in the future as the tech scales?
- How do you see the competition play out in the long term?
- Is there anything that you think is important and we haven't talked?
- What else is important in selecting a customer base?
Interview
Let’s start with the market landscape. We see issuer processor focused providers, such as Lithic, Bay1, and all in one platform, such as Bond, Productfy, Unit, which integrate with third-party providers like Marqeta and then the FinTech would just only speak with this layer without managing other technical and regulatory elements. What differentiates these two types of providers, the vertical issuer processors versus the platform providers?
Guest: I will say that the truth is the lines have been blurring a lot recently. I would say the biggest thing you'll notice is that either of these vertical, like issuer processor only platforms are trying to step more into BaaS-like features whether or not they've actually succeeded.
Galileo, as an issuer processor, was one of the first to really step into being more of a BaaS service with their instant program, which was not very well run. And I think they ran into issues with their bank or provider. We've seen a couple of companies switch out of that program and a good number of people have abandoned it. They seem to have failed to deliver on their promises.
I think Lithic is one of the better, upcoming issuer processors. They have a lot of rough edges. I think they're kind of moving into BaaS service as well, but in a lighter way and they miss out on some things, such as ledgering, so that's a key thing that makes it a little bit different.
I would say the key to a lot of these BaaS services is that they provide some form of ledgering service. Ledgering seems to be the core component of every BaaS. Your issuer processors often do not provide that.
Would the issuer processor integrate with Modern Treasury to provide that service?
Guest: That’s possible! You could build your own ledgering, or you use a service—it's really up to you on how you want to approach this. I think the cost can be pretty equivalent considering how expensive engineering salaries are. The one thing we've noticed across the entire industry is everyone's willing to negotiate like crazy if you push and you have competing offers. It feels like a very hotly competitive market because there are too many providers of BaaS services/issuing processing services to the number of viable neobanks out there.
A lot of these newer BaaS services are willing to throw the book of discounts at you, and there are a couple of services that are way overconfident because they’ve got a small batch of YC companies.
The real issue that’s not obvious right now is that there are not very many good neobanks out there. Most neobanks, if they even manage to raise or manage to build a strong team, copy each other and eventually only one can win. If you support multiple of these customers, you’d eventually kill your own customer base. Some providers are acutely aware of this, but many aren’t.
I think Bond is one of the more strategic providers, in the sense that they actually selectively pick and choose, and try to make each neobank they support a home run winner.
The one that I'm actually least excited about is Treasury Prime, based purely on my personal experience with their sales team. Talking with them, they're overconfident and very bullish on the services they provide despite not realizing how much they actually lack. They think they're the best out there because they released some basic debit card programs, but that’s now just table stakes. Just like many other providers, they glue together third-party services with their tech into one layer, and they bring the banks over. This is compelling up to a certain point, but you don't really want to deal with a bank that’s not prepared to work directly with a neobank as they can complicate things in regards to compliance. Treasury Prime seems to have a deadlock on YC neobanks, but half of them compete with each other or do exactly the same thing. Piermont Bank is their saving grace, doing what appears to be truly most of the heavy lifting.
In terms of how these providers differentiate themselves, I'm hearing there's a differentiation to how much exposure the FinTech wants to have with the bank…
Guest: Yup, that’s a key difference. Some providers like to lock away the bank relationship, which would be kind of risky to you as a neobank. For example with Unit from our conversations with them, they didn't really let us have contact with their bank. They didn't really acquiesce to somewhat basic requests, but this was from a while back.
I feel every BaaS service is trying to differentiate itself in some quirky way, which I don't think it's very well reasoned to their end-users. I don't think they are thinking: “Hey, I'm trying to build a new neobank today. What issues am I running into?” They’re either not re-assessing their value proposition often enough, or have blinders on to what their competition offers. Despite their best efforts to not be commodity services, it’s what they’ve become and will be down to who can execute the best.
Visa and MasterCard are of no help in the sector, and can actually be kind of detrimental if you’re not careful. You need to capture their attention first before you truly can negotiate, and you shouldn’t rely on them for anything critical.
Now if you have some “edge” interesting to them, such as you're supporting an underserved community, you're better off going with a Visa or MasterCard instead of like relying on the BaaS service to bring that relationship. Unless they somehow can bring you a very, very particular bank sponsor you need it doesn’t matter at the earliest stages. If they actually can deliver, which they sometimes falsely promise, and bring that relationship for you and let you control it, it’s a powerful support to have in the back pocket.
I can speak to BIN (bank) sponsors as well.
Why is having a direct relationship with the sponsoring bank important?
Guest: So the risk is, for example, if any of your platforms do something wrong, say Synapse violates some law and they have to pull the rug under you, you will lose that banking relationship if you don't own it.
It's important to own it because you can set the interchange terms directly with the bank. For example, from what I understand, Treasury Prime seems to take its own cut. They weren't super transparent or the salesperson didn't fully understand it, but we see a majority of BaaS platforms take a cut and do not do passthrough. You want to get the gross passthrough interchange rates or as close as possible to it. Each BaaS tries to make the billing complicated or obfuscated to increase their margins, but you can tell which service will work best based on how honest they are when you call them out.
The margins on debit are already crazy low as it is, so this is why you see a bunch of BaaS providers come out and say: we have only one card provider we work with and we need to charge you $5-7 per card with monthly per account fees of up to $2-4. That makes no sense, and you shouldn’t allow them to lock you into these ancillary services.
This is why all these debit cards startups have to charge their end-users, but newer cardholders don't realize yet that those startups offer a product with features that most people could get for free. Credit reporting can be done through debit-style credit cards and you see the incumbents in this space actually go through the hard work of setting up correctly. All these new neobanks that are popping up are taking way too many shortcuts that will actually trip them in the long run.
You could be paying millions of dollars easily per month. A lot of the early BaaS providers think they're going to be a hot ship, but they're really killing these companies in the long run. Because they're not going to be able to monetize in time.
If you are starting out today and looking for a BaaS platform, given the increasing activity in the space, how would you select and decide who would be an ideal BaaS partner?
Guest: What you want is a newer platform that is stable. I would avoid all the incumbents. I would not go to Galileo or i2c directly.
If I had to choose an issuer processor, I would select Marqeta or Lithic. I think those are pretty solid. Lithic is very rock-solid, but they are expensive. You're not going to get really good unit economics with them. And I think they need to start competing on pricing. Right now they're justifying their costs because people don't know any better. Eventually, people are going to catch on at scale.
Although I think your best bet is actually, get a term sheet from every single provider and counter negotiate and compare the feature set. So talk to all 10 BaaS providers, make sure you get a full picture of what they can actually do, push them to their limits on what they are offering. What is happening though is that a couple of key sponsor banks are actually the deciding factor.
Out of some of the BaaS providers, the farthest behind is actually surprisingly Stripe. Stripe is actually the most immature in terms of what they can offer. And they're not able to actually provide deep custom systems. Stripe powers Step as an issuer processor, but they are not ready for any customer other than Step. They built a system that's entirely custom and only works for Step. So in terms of selecting a BaaS provider, I would actually leave Stripe out until they actually reorient things in 2022. I'm probably not going to touch that until 2023 in terms of their stability.
What matters more is less the BaaS and more the BIN sponsor. It's a misnomer to focus on the BaaS. It's much better to focus on the actual BIN sponsor. Once you have the BIN sponsor down or you look at the provider and see if they work with a specific BIN sponsor, that's really when you can really make a true decision.
Another key thing on a BaaS provider is you own the cardholder relationship. If you do not own the cardholder relationship, avoid the BaaS. This is actually a very critical thing that people always miss.
What allows faster speed to market and what could impede it?
Guest: Speed to market has nothing to do with the BaaS. It has everything to do with the BIN sponsor. And that is actually a huge misnomer. Everyone thinks it's Treasury Prime who loves to brag how they're fast to market, but it's just because they have a buddy relationship with Piermont bank and Piermont bank is the fastest bank that you can get to market with on debit rails.
On credit rails, I think Evolve is amongst the fastest there, but you're never going to see anything faster than one to two months. That's just how the industry baseline is, with the regulatory process to make sure everything's followed through. That's what I would say is really the top constraint there, it's BIN-sponsored banks.
In terms of interchange, if we're looking at the consumer segment, broadly speaking, we start with 1.35% interchange for the whole value chain…
Guest: 1.3% on debit does sound correct specifically for exempted banks under the Durbin amendment, whether or not you actually see the 1.35% is a whole other question. You really need to push to negotiate. Some BaaS providers we've negotiated with, their model is that they take most of your interchange and hope you don't understand what interchange is, which is sneaky.
I'd say your true net is probably 1%.
The interchange really depends on your customer base. If your customer base is lower-income, they're probably spending at places that don't have nearly as high interchange, like grocery stores… or utilities which pay no interchange, essentially. You require a lot of volumes and you require a few whales.
I would be wary of debit interchange. What if there are some threats to the Durbin amendment? And that's something that you should actually factor in. Legislators are getting pressured by like merchant lobbyist groups to drop the Durbin amendment, which would kill 90% of these startups overnight time and they would fail immediately. The bet here is that everyone's looking up at Chime and Current to counter lobby.
How's the interchange split between the FinTech and the BaaS platforms currently? How do you see this dynamic evolve in the future as the tech scales?
Guest: When the FinTech scales, they better negotiate because BaaSes typically just take advantage of you because you're rushing to market. For example, Treasury Prime is just terrible. We looked at it, we thought it was such a poor baseline to negotiate from. For some context, every line on the term sheet you get from Treasury Prime must be negotiated. Everything should be slashed by like 25% minimum in terms of their fee costs, in my opinion. That's actually how bad and how highly overpriced these things are.
How do you see the competition play out in the long term?
Guest: I feel this will be a matter of who raises the most because there's so little and they can differentiate themselves on.
Everyone is saying the same things: do you provide better ledgering? Dashboards? Better ancillary services? How much control do you let me have with the bank? Will you actually support me through the bank? Do you provide legal support? How do you handle disputes and resolutions? There are literally 200 to 300 things you need to consider.
Another key thing you'll notice is does BaaS provide program management? Program management is a semi made-up term. What it really just means is they manage parts of the program for you. That's literally all it is. They would piece together Alloy, Unit 21. Jumio is another good provider for KYC, AML. But you're very limited in the options you really want to choose.
When you look at sub-program management providers, you want to see what the BaaS use in terms of their third-party providers? Basically, that's another big decision point.
Is there anything that you think is important and we haven't talked?
Guest: I would say if investors are looking at a BaaS, they shouldn't just be looking at the revenue numbers. There's a hidden disease amongst BaaS providers in which they're just taking advantage of the inflow of capital, but they're not looking at if their customers are healthy financially. I would say, look at the customers of the BaaS, because that very much will determine whether a BaaS is going to survive.
What else is important in selecting a customer base?
Guest: If the Fintech/brands want to scale and they want to have long-term positive unit economics, I think you should always pay attention to the ability to eject by building your own services. So all of the contracts will have a non-compete clause, such as you can't use this alongside another BaaS, which is understandable.
But the biggest thing you should have in there is that you need to make sure you don't get locked into using their services. Eventually, it will get to a scale in which, you will need to build your own version. For example, you don’t want to get stuck on their card printing services. You need to be able to negotiate that, always argue for the ability to expand out and eject. You should never get locked in.
Disclaimers
This transcript is for information purposes only and does not constitute advice of any type or trade recommendation and should not form the basis of any investment decision. Sacra accepts no liability for the transcript or for any errors, omissions or inaccuracies in respect of it. The views of the experts expressed in the transcript are those of the experts and they are not endorsed by, nor do they represent the opinion of Sacra. Sacra reserves all copyright, intellectual property rights in the transcript. Any modification, copying, displaying, distributing, transmitting, publishing, licensing, creating derivative works from, or selling any transcript is strictly prohibited.
