Hello. Good afternoon, everyone. Welcome to the Needham Growth Conference. My name is Kyle Peterson, one of the Fintech Analysts here at Needham. Thanks for all the persons that stuck around for the last session of day two here. I know it's been a long couple of months for people, but we're going to host a fireside chat with Greg Garrabrants, CEO of Axos. Greg, thanks for joining us.
Thank you.
I want to see maybe if you could just start out for people that might be a little newer, if you could just give a quick overview, kind of elevator pitch of Axos and the history and kind of how the strategy has evolved.
Sure. So Axos started as one of the first online digital banks, and it's evolved over my tenure of about 17 years to have three primary business segments. We have our consumer banking segment, which has an online financial institution that serves both consumers and small businesses. That platform that we use is an all-in-one style app that has self-directed trading, robo-advising. We also have a series of consumer lending products. We have auto. We have a mortgage business there. The commercial side is a series of specialty finance lending niches and deposit niches. They're software-powered in many instances. For example, we've got about 40-plus% share of the U.S. bankruptcy trustee market through enterprise software. And we do a lot of different and interesting types of national lending platforms in that business. And then we have a securities business.
That securities business has a custody operation and a clearing operation. The custody business serves registered investment advisors and provides enterprise software to them and a variety of other services. The custody business works with about 75 introducing broker-dealers to take deposits and clear for them. The synergies between the businesses are that the securities business has going on 500,000 customers, a little bit less than that, but around that. We're basically creating software to allow those clients to get access to the banking products because the RIAs don't want and lending products, because the RIAs and the independents don't want the big banks to get into their clients through either mortgage lending or SBLOC lending or different types of lending. So that is a significant synergy. And then we have interesting businesses.
And there's actually a decent bit of synergy between the small business side there as well. And then the commercial and consumer side have great countercyclicality because a lot of times when the mortgage business isn't good, a lot of the commercial businesses are good. So it's enabled us to achieve above-average growth rates and not to take undue credit risk. So we've had a really good run with respect to loan growth, credit risk, and being able to maintain above industry net interest margins for a long time.
Great. That's super helpful. And one thing I've always kind of thought is unique is you guys have been able to scale on the deposit side over time quite a bit faster than a lot of other banks. You do it without a brand strategy. I know there's been you've done some inorganic stuff in the past, but also it seems like you look on your presentation, you have depositors in all 50 states and all that. So I guess kind of what's been the main way and how have you been able to kind of engage and grow on the deposit side over the last several years?
Yeah, we have, I think, a very comprehensive strategy that tries to balance the commercial, consumer, and the securities side. So whenever a securities client comes in, whatever cash deposits are there through the sweep process, we can either take those on balance sheet or put them off at other banks. We have maybe 35% or 40% off at other banks earning fee right now. But as that business grows, it typically was about 8% of assets under custody that would end up in the depository side. It's kind of gone down now to more like 4%, but that's because rates are higher. But nevertheless, that's one mechanism of deposit gathering through what we call our universal digital banking platform, which is that all-in-one app.
We work to provide our clients value-added services, whether it's everything from free credit reporting to budgeting, planning, those sort of things, to try to bring those clients in and serve their needs on the consumer side there. That's a very tech-driven business and very digital. So we know if we spend a certain amount of money on marketing, we get a certain yield. So we can determine what that looks like. And that also, some of those accounts are rate dependent. So depending on where we want to be on a rate side, we can increase deposits there. And then the commercial businesses, we've been doing a lot there. As a lot of the banks have blown up in California, we've taken different teams. And so I would say that the banking landscape in California. It looks pretty open.
I mean, if you think about Silicon Valley, you think about City National was kind of a powerhouse before. They've been sort of de-entrepreneurialized by getting bought by their large Canadian parent. And so we just took a team, for example, out of City National for their entire venture business. We're not doing venture lending, but we took the team on the depository side, also their Washington, D.C. team. So we've been hiring these different commercial deposit teams. Some of them will have offices. We have a small office. We actually just signed the lease yesterday in the Valley. But we have a variety of niches there. And essentially pursuing that sort of diversified strategy has really allowed us to grow the deposits. And the high-yield savings can sort of be a fill when we need it. And that allows us not to have liquidity constraints.
Great. That's super helpful. And maybe just as kind of point of clarification, at least in the consumer deposit side, I know commercial is kind of a separate animal, but I think at least the perception that a lot of these digital-first banks are going after the unbanked or the underbanked that might have $500,000-$1,000. And it's a hard product to turn into a profitable relationship. So what does your typical customer look like for you?
The typical customer for us, the average checking account size is more around the $10,000 range. We actually, based on the criteria that we have, essentially the QualiFile scores, which is sort of a deposit credit score, end up excluding folks that have had issues with prior banks from an NSF perspective or whatever. And that tends to be a lot of the underbanked that are being pushed into some of these neobanks. I think it's a very hard thing to serve those clients well. If you're not doing some sort of lending component, whatever that is, it's very difficult to make money there. It's particularly difficult for us because when we cross the $10 billion mark, we lost all the interchange.
So if you look at a Green Dot or whatever, and they make this huge percentage of their income on interchange, and then we cross that, it's just a very difficult thing to do.
Yeah.
We lost all our. This was, I mean, now that's going on a half a decade ago, but we had to get out of even sponsoring prepaid card companies when we went over Durbin.
Yeah. Okay. Yeah, no, that makes sense and super helpful. If we could switch over to the asset side, I think you guys have been able to balance over several cycles where you guys are growing faster than the industry. Your credit has also been very strong. I guess what have been the focus areas and kind of how have you identified some of these and how have you viewed balancing risk management versus growth over time?
Yeah, it's been interesting because, I mean, clearly one of the big challenges has been a zero interest rate environment for a long period of time, obviously pushes up asset prices, and in some areas, some of the traditional spots that we were growing in and known for, such as sort of 5/1 ARM jumbo mortgages and 5/1 ARM multifamily mortgages, just really weren't good investments. So we did a bunch of stuff to shorten up the duration of our balance sheet well before rates went up. We emptied our bond portfolio essentially entirely. We had zero HTM securities. We shortened up our book dramatically and got everything super short. How we did that, even in our traditional businesses, which were 5/1 ARMs, is it was funny. Nobody really cared if we did a two-year fixed rate and we gave them a 15 basis point discount, so it's interesting.
Nobody really was focused on it. We were able to do that. Then the other thing that we did is we really took advantage of the rise of the non-bank financial institutions. So a significant part of our portfolio is originated in working with non-bank financial institutions. If you look at the significant part of what we do, whether it's capital call facilities, almost all the commercial specialty real estate, the non-real estate, and the real estate lender finance, it's really working with these non-bank financial institutions. What that looks like, I mean, just in a really quick way, because I don't know everybody's familiarity with us, but $100 million project, Madison makes a loan of 70. We make a loan to Madison of 70 on the 70. So we're at 49.
There's a series of covenants and transactions, depending on if it's a note-on-note or an intercreditor agreement or whatever. But essentially, whatever the structure is, the structure is that Madison is taking first loss. If anything happens with the underlying project and the borrower can't rebalance, Madison is required to rebalance. If something happens with the borrower, Madison has to become our borrower and pay us down and all this sort of stuff, which is why essentially now, today, we don't have losses in commercial real estate. We've not really had them, and it doesn't look like we're going to have them. So that's something that's very differentiated from other institutions, which really don't have the LTVs and the protections that we have. And I think we haven't had, other than we have one UBS, UBS O'Connor is behind us in one deal, and it's not their real estate guys.
And I think we're going to end up with that property back. But literally, that would be. That wasn't even in the commercial specialty real estate side. It was originated through a different group. But frankly, after all this, that's the only time we've ever had a sponsor ever say they don't think they're going to support anymore. And they've got $30 million in front of us, and our loan's going to be fine. I think we'll be able to get everything back on it. But I mean, so through all of this, and it just looks very good, and we're continuing to see that. So I think that structural element is really important as well. And I think sometimes what happens is you get these tech-enabled folks who really. They're good tech guys, but they're not good finance folks.
And so they screw a lot of stuff up with respect to that. And they also think that because they're tech guys, they actually know everything. And it's sort of interesting to watch that come and go. But I think we've shown that we're thoughtful risk managers, and then we also use technology in the right way too.
Yeah. Yeah, no, that's super helpful. And maybe kind of staying on the asset side and related to credit, I know there are obviously some concerns of exposure throughout different companies, especially in insurance and such with some of the fires that are going on in LA. You guys have a national kind of footprint in terms of your loan book. But I guess have you guys done any work, or is there anything you can share about potential exposure to?
Yeah. Well, so although we are national, those markets were core markets for us. And we really like those markets because of a variety of things, the stability there and whatnot. So we've done work so far, and we don't have a full assessment, but I think five of our mortgage properties have been totally destroyed. There may be others. I don't know. We're working. We've categorized houses that are in those areas. We have a lot of them that we said, well, they could be, and we've confirmed that they're not. But I think the core of it is that we don't believe we'll have any loss from this for the following reasons. First, we have insurance. If somebody was underinsured or not insured, we force-place the insurance, and we have access to those force-place carriers.
So we would force-place insurance for someone who didn't have it in case it would lapse, which in most cases, it doesn't. But if it did, we would do that. And then we have two separate policies, one with Lloyd's and another one with a different insurance company that essentially cover any gaps. And those policy limits are well above what we would need in the event that there would be any kind of issue. And let's say of those five, four of them have insurance coverage where we're the loss payee above the loan amount. And then one of them has a cross with a Beverly Hills property that's worth double the loan amount. Then I was crossed, and that guy has zero mortgage on that. So essentially, we're first on that. It was this sports guy. But where could something ever come up?
I think just theoretically, if you're thinking about this, is when you have insurance on a property and you have a mortgage, if the land value was high, somebody might get a mortgage just for the replacement cost of the house. So then you have a certain land value, right? If that land value was on the beach in Malibu and our lovely and efficient California government decides that they're going to take five years to approve a plan, which could easily happen with the Coastal Commission, which is interesting. Now, supposedly, that's something that's going to be done there. You could imagine that that value would be dissipated or something like that. That doesn't appear to be the case with any of the deals we have but and we have coverage that's global for that through these blanket policies.
But nevertheless, it's a theoretical risk that could arise, and it's important that you manage that as a bank just to think about it. And this presents a unique challenge now. Frankly, with respect to Pacific Palisades and those areas, they're going to get rebuilt. And frankly, if they waive the CEQA and all the other socialist sort of things that go on in California, they're going to get rebuilt very well because people are going to really be able to do that in a way that's efficient. If we let the free market operate there, it will eventually work. But here's the positive side of this. We have a big exposure to LA housing. Every other house we have, every construction loan we have, everything just went way up in value.
I mean, just being real, because we took a lot of the richest people in the country who have plenty of money, and we took their families and said, "Buy, you don't have a house." So everything in Beverly Hills is going up a lot. Everything all over is just going to skyrocket. And we're already seeing it, right? Literally, we'll have a $10 million construction loan on a guy who's trying to sell his house for $25 million bucks, and we're saying to him, "Okay, you've got a little more time, but it's not worth $25," and all of a sudden now he's getting a $35 offer. So that is what's happening now. So it's actually a net benefit to credit in things. Yes.
Just on the actual rebuilding, how would you kind of position yourself to kind of maybe take advantage of this?
That's a great question. I also was going to say, I think it's an opportunity. I mean, what we're looking at is saying, how can we roll out a construction project? We generally do a more spec construction because what we like about spec construction is we have a set of builders we know. We know they can do the job. We know how they do it. When you do owner-builder, it introduces a bunch of things, which one of those things is, gee, the spouses argue about the bathroom tile, and it takes three months because they want to ship the tile from Italy, and it gets lost in the, right, that kind of stuff. So you just have to have the discipline, and you can get around that. It is something you can work around, but it's a little bit of a pain.
But I think in this instance, the level of opportunity is going to be such that we're definitely going to want to be involved in that. We know that area very well. We have a lot of great relationships in that area with brokers and borrowers. And so we're already getting those requests in, and we're going to be really thinking about how to do that in a way that makes sense. And theoretically, we don't have this situation yet, but if you had a situation where somebody had a high land value and they wanted to sort of roll into that sort of program, you'd have to think about it.
I mean, normally the problem is that you don't do a construction loan until all the plans are done, everything's approved, you're sitting there ready, all the equity comes in, and then only after that you fund your first dollar. So it takes a while to get that money out. But I think it's going to be there. I think it clearly will be there as an opportunity. And we want to participate not only because that's our job, but also we want to do that just from a standpoint of it's the right thing to do. And I don't know how many mid-sized banks are going to be active in that area because there's just not a lot of mid-sized banks. But we think we can be agile and nimble and helpful to people there who've lost their homes. It's pretty awful.
I mean, it's negligence beyond almost any comprehension. But we've got to all kind of get it rebuilt, so.
Great. That's helpful, and maybe we'll move on. I know you made some comments about there's some tech guys that do lending that think they know kind of everything, but I view you guys as like you're kind of a bank first and at your DNA, but you guys are also pretty good tech guys yourselves. You guys built a universal digital bank, which a lot of other guys would have just used like an off-the-shelf software program. I guess what are some of the ways you guys have incorporated tech into your process? How do you guys approach it, and how do you integrate it into the bank as part of your value prop?
We have, I mean, a very ambitious strategy. It's interesting because I was at the Goldman Conference, and I sat in on the LPL presentation. They said, "Well, we've really got to figure out for all our brokers how to integrate banking and OneTouch SBLOC into it." The only problem is I can integrate banking and OneTouch SBLOC. I just don't have LPL's broker base, right? But what we did is our goal was to build this universal application. We've taken that universal application, that sort of, you call it, you cover SoFi, that SoFi style app. But we've also made it able to be utilized by all the end clients of our broker-dealers and RIAs. When an RIA comes in, what they get from us is they get an app that can have integrated banking in that app as well as 1-Touch SBLOC.
And so the idea was to be able to have our own development there so we could continually modify and have it work through a variety of distribution channels. And so being branchless, the idea of having brokers and RIAs sell our banking products because they're so well integrated, right? We haven't done this yet, but it's on the roadmap of somebody does a direct deposit, 10% of it goes directly to the RIA. Right now, when an RIA gets a check, they have to take the check. The check sometimes goes to the RIA. The RIA either scans it or, right? Well, it's just easier to take a picture of it. We don't have to hold it because we already know what the balance is. You clear it. You put it in the account that day. It can be traded that day.
So there's a lot of real synergy between having the bank and the brokerage stuff together. And so that's what we did on the consumer side. We built this. And then there's just a ton of really cool stuff you can do. We bought this, frankly, one of these fintechs spent like $30 million to build it. We bought it for $1.5 million when it was going out, and we got what we paid for on our robo. But it was useful because we learned how to do it. And now we're integrating the back end of that robo into the model management store that we have on the security side.
So we're going to have, I think, by the time I'll be talking to you next year, I'm going to say we have one of the coolest robos out there because you basically can pick your professional money manager out of your robo, which is pretty cool, right? And we're going to have little videos about, "Hey, take this guy, and here's why." And so really have that model management store out there for something like that. One of the things that I think is really exciting is we've been doing a lot of low-code work in OutSystems and also doing a lot of pushing our developers to use AI, and it's working. We're going to deliver a project for a new broker workstation because essentially the clearing and custody business are essentially SaaS businesses. We're delivering a new broker workstation because they don't like FIS WealthStation.
If we would have done that prior to OutSystems and AI, it would have taken like three years. It took like seven months and well less than half the people, and the ease of maintenance, the ability to change, do all this sort of stuff, and that's going to be the first time that we'll be able to have digital account opening for those brokers that can do it on the banking side too, so I mean, I think really it's a lot of being able to utilize technology to provide these end clients, both on the institutional and direct side, great tools.
Great. That's really helpful. Maybe kind of gearing back towards the asset side of the balance sheet, on the last call, you did talk about how there has been a bit more competition for some deals. And historically, I think of you guys as you find the niches before other people do, and you get creative and figure out ways to earn attractive spreads. So I guess where are you seeing a little more competition? Should we think of this as, are you willing to bend a little bit, whether it's on price or structure? I guess how are you guys kind of approaching it?
I think it's caught everyone. I'd say that in general, I describe the lending market as the spread compression has caught folks by storm a little bit. And so it's always hard for me to balance that out because every relationship manager, every client wants a lower rate, right? And so the question is, we didn't hit our loan growth targets last quarter. We've generally done pretty well. I guided this quarter to be like 300 or 400 up, which is still below what I would like it to be. We'll be around there, but I think you have to remember, we originate like $3.8 billion, and we have, let's say, $3.4 billion payoff or something. So it's hard too. You just have to be thoughtful, so I think there will be selected areas where we look to be a little bit less firm on pricing.
I mean, I think one of the things that's happening for us too is that our commercial deposit business has picked up so much that we've been able to maintain our spreads by doing better on the deposit side. So maybe that provides us a little more room. But I think it's an interesting balance. I mean, I've kind of liked the fact that we've been able to keep such great margins, and we've been able to kind of grow on that edge. But it was obviously slow for that quarter. And I do think you're seeing competitors just, when somebody asks for something, capitulate more. And I think that's just the result of there not being as much activity.
If we can get some pickup in the economy, which I think we will from deregulation and that sort of thing, I think maybe the fact that everybody thought rates were going to fall more than they are, that might be a headwind to that kind of otherwise good environment for getting some more demand.
Sure. I guess conversely, thinking compared to where we were on the last earnings call, we've seen at least the short end of the curve has come down a little bit, which should help deposits and funding costs. The long end has actually steepened up. Is that something where?
That helps a lot, well, right. I mean, if you take our businesses, the two businesses that we have that we fund out on the curve a little bit is both single-family and multifamily. They're 5/1 ARM products. So if you think about our growth, the growth has been done despite probably having a $400 million per quarter loss on those two portfolios. Why? Well, the problem is if you're essentially funding short, and just from a rate perspective, and you've got 100 basis points of negative spread in the one-month, five-year kind of range, it's just not a great time to originate 5/1 ARMs, right? And so getting a little positive shape to the yield curve helps a ton with respect to that business. Now, whether single-family originations come back, that's a demand-side thing. Multifamily is a totally different story, though.
The number of banks that have been just decimated in multifamily and can't do anymore, not for credit, but because they screwed up the interest rate side, right? Because I mean, what you saw is these guys were doing, let's say, non-recourse 70%. They'd say that there was a 3% cap rate, right? So they ended up with a 5.50 Debt Yield. And they did these deals, and they did seven-year deals at 3.5% or 4%. Those loans are now all trading in the 80s or something like that, just on interest rate, not on credit. So we didn't do that, but now that market has changed a lot. So now it's just about toeing in at the right place and waiting for the folks who own multifamily properties to get a real sense of the fact that cap rates have risen.
Even if you don't believe it because there's not a lot of transactions, you need to believe it. It's not just 150 basis points. It has to be a little bit more. Everybody was just basically thinking, well, rates are going to come down enough that you're going to be able to say, well, don't worry, this is all temporary. That kind of mindset, I think, is moving a little bit, which is going to start getting properties trading, which is what's important for volume. We're seeing good opportunities in multifamily now. If we get a little more shape to the yield curve, that'll help a lot.
Great. Yeah, that's good to hear. And you did kind of touch a little bit on deregulation a few minutes ago. I think that's something that has come up a lot as you guys are potential beneficiaries or likely beneficiaries of that. But how should we think about both whether it's direct impacts, like direct benefits for you guys, or indirect ways it could kind of spark loan demand or just generally things up?
I think I'll say in a couple of ways. One way is that obviously, I can't talk about regulatory conversations, but one of the first things that the Biden-Harris team thought was important right before rates went up like crazy, the top priorities that were put out, and they were put out publicly. They weren't interest rate risk or whatever. They were climate change and DEI and fair lending, OK? Those were the priorities prior to having a zero-rate environment for 15 years, right? That was the priority. It became something else. It swung to something that was much more focused on commercial real estate, things like that.
Look, largely, I think the regulatory environment, if a regulator has the idea that their job is to help a bank with safe growth and that the mechanism of doing that is a collaborative interchange where people in good faith approach problems in a manner that is productive to the ultimate goal of providing capital to our free pluralistic society, that's a really cooperative relationship. And regulation in banking is acceptable and fine. It's the trade-off we make for having uninsured deposits. If that regulatory environment becomes a mechanism for, let's say, the destruction of crypto, which it has been as an industry agenda, right? The pure destruction of crypto and anyone who touched it was absolutely on the regulatory agenda. And that was done in a variety of ways.
When it does things like that, whether it was Operation Choke Point 1 or Operation Choke Point 2.0 or whatever, that's destructive. So all that stuff ends, right? All that stuff ends. So then the question is, what does that do? And I think you can see the tone. I mean, the tone among all the executives and banks are like, hey, I like to work with people who are collaborative and not trying to destroy me. And that's a really great thing. And you see that change. And obviously, it's not just industry. I think people are feeling that. With respect to things like crypto, the question's going to be, I think you need guidance and regulatory change because you can't have a situation where the SEC's running around saying, well, everything is a security. And so any of these guys are unregulated exchanges.
You have to have a set of rules that come in that will then allow banks to interact with those clients in a helpful way. But we better figure that out because if it's a bunch of non-banks that figure out how to use stablecoin, and then people can use stablecoin for payments, at a certain point, banks will lose an edge in payments. I had a very high-level conversation with somebody who likely will be very important in the FDIC. And it was super important. They wanted to have a discussion about how to think about stablecoin, how to think about banks' role and things like that. You're just seeing a lot more collaborative interactions with folks trying to figure out how to do these things safely.
So I think you'll get good regulation out of it because you're going to have that good faith sort of goal that you're trying to basically grow the economy. And I think that that focus was something that was lost in the post-Silicon Valley panic, et c So yes.
So if you get this dialing back of regulations, that's great for growth and all that sort of stuff. Have you guys been able to put your head around what you would say from a cost perspective? I mean, how much of your costs are nonsense?
Yeah, it's a super interesting question because really, I think that's a really great question. I mean, I'll try to take it this way, is that let's talk about the type of costs. One of the type of costs are these precautionary sort of mechanisms that you're taking for something that's unnecessary, right? Because what you really want to do is you want to spend your risk management dollars on stuff that's actually useful, right? And so if what you're doing is you're spending all your time on things that are just not useful but are only useful from a regulatory perspective, that's a problem, right? And so those discussions and interactions, if you have those individuals coming together in good faith, can be a different interaction, right? And I think we're starting to see that come back. So I think that's very helpful.
I do believe that there could be the potential that you could get some clarity around certain types of issues. So clarity is helpful because then it allows you to really kind of hone down on what you're doing. We actually have, with respect to what we do on consumer compliance, for example, we have something called data-driven compliance, which what we've done is we've essentially digitized every compliance rule into essentially a rules engine. And all the data gets fed into it, and then it spits out any issues. And so essentially, we don't have those issues because we always solve them before they actually hit, right? So they're sending out alerts saying, hey, we didn't resolve this by this date, and you need to do it in a week. And then it gets escalated.
And eventually, you're going to have the head of compliance sitting over the person's desk, and it gets solved. So I mean, I think a lot of those things are there. I mean, look, we built a stablecoin platform, right? It's not a secret that you had to go get non-objection from regulators for it. I can't say what exactly happened, but that just sort of sat there and never got launched. So what did that cost? Well, it cost the cost to build it. It still sits on the shelf, and it also costs what you don't do with it. And so instead, you have it all offshore or whatever. But that would be an interesting thing because I think that would have been an interesting payment platform. It could have been useful in international, whatever. So I think it's a hard question for me to answer.
I mean, I think the biggest cost savings, though, I mean, maybe the way to say it is if I think if I look at where we get operating leverage, I think less about that, and I think more about IT and the AI impact. I mean, with respect to these low-code things, I'm super excited. I think in two years, when we get this low-code and this AI sort of utilization through, I think we're going to have quadruple the amount of innovation. We're going to be able to push through the same pipeline of IT people. I really do. I mean, I really start to see that curve sort of bend, and we saw it with this latest project, which is super exciting, and then I do think we have an AI task force.
And there's just a bunch of stuff that we end up having to do, taking data that comes in on appraisal and actually being able to sort of take it and put it into a structured database. And it's hard because you have so many different appraisers, and you've got to try to figure out, well, all of a sudden, boom, the machine can do that, right? I mean, there's just a lot of stuff like that that are immediate utilizations. And we're actively working on it. It's going to take a little time, but that's one. And then so we had a good cost quarter this quarter. We'll see that we didn't grow our cost as much. And that's something we need to do if we're making sure that we're matching our revenue growth to our cost, so.
Great. Well, it's been super helpful. I think we're about up on time. But Greg, thanks for joining us.
Thank you all for listening. I appreciate your time. And if you ever want to talk with any of the executives, we always open up our executive team. And Johnny is great too. He's been with us a long time. He knows everything that I know, mostly, most of it. So thank you.