Mortgage Tech Insights with Adam Carmel

Mike Simonson here. Thanks for joining me today. Welcome to the Top of Mind podcast. If you follow along with Altos Research, you're familiar with our weekly real estate market data video series with the top of Mind podcast, we seek to add context to the discussion about what's happening in the market from the leaders in the industry. Each week, Altos research tracks every home for sale in the country, all the pricing, all the supply and demand, all the changes in that data. And we make the insights available to you before you see it in the traditional channels. People really need to know what's happening in the housing market right now. So if you need to understand the housing market or communicate about this market to others, go to altos research.com. Book a free consult with our team. You can get started with a free local market report now.

(00:45)
So let's review your local market and talk about how to use market data in your business. Alright, let's get to the show today. I have a great guest today. We're talking mortgage markets today, Adam Carmel. Adam is the founder and CEO of Poly, a true mortgage technology Trailblazer. Adam founded Poly in 2019 to build a vertically integrated mortgage platform. We're going to learn about what that means. Adam's career has been focused on mortgage capital markets and we're going to hear today about the journey about getting where he is today at HousingWire. We love the poly mortgage rate data for our mortgage rate center, so we're going to talk about the poly technology and the data around mortgages today. Plus a lot of the dynamics in the mortgage markets and capital markets in general. Mortgage rates are driving so much of the housing industry today, so this is a critical topic. So Adam, welcome.

Thanks, nice to be here with you, Mike.

Great to have you. Okay. First tell me about Poly, where you fit in the mortgage ecosystem. What does vertically integrated mean? Tell me about what does that mean for your customers? Tell me about the company.

Yeah, so Poly was founded in 2019. We are exclusively focused in on the capital markets kind of ecosystem, starting with locking a loan all the way to selling a loan. And so when we say vertically integrated, that's everything in between. We were very intentional when we started the business to focus in on understanding all of the challenges and pain points of the legacy solutions out there and really rethinking how the industry should operate over the next 10, 20, 30 years. And certainly I personally think, and quite a large amount of lenders and banks agree that the way things have been done over the last 20 years needs to be truly changed. And so we set out to do that and here we are about five, six years later when you think about how profitability is booked in the industry and maximizing profitability and then ultimately driving the cost savings and as well increased profitability. Ultimately the goal is to be able to drive that to the transaction level between the loan officer or a mortgage broker and the consumer. And so everything we do every single day across all of our products and all of our features is to say, okay, how do we optimize profitability? How do we drive a revolutionary loan officer and mortgage broker experience and drive all that up ultimately at the transaction?

Got it. So ultimately to make a faster, cheaper, better transaction for the consumer,

That's the job of software. You would think that's what should have been going on here over the last several decades. But boy, the cost of originating a loan has gone from 2,500 to 12,000. So I don't know if we've done such a good job as an industry in the last

Industry

Five 30.

So you mentioned though you have some AI innovation, so what's happening with AI in that loan process?

We're real focused on a couple core things right now and then we're going to expand it out over time, but right now it's making recommendations on taking ineligible loans and how to make them eligible near mispricing. We're abstracting away and automating a lot of workflows, all with the idea of partnering with the loan officer or mortgage processor to enable them to get more loans done faster, higher conversion rates and just drive a much, much, much better experience between them and the consumer. There's hundreds of parameters and hundreds of dimensions in originating a loan and so we to ensure that there's full compliance and it's something that the loan officer can actually control with the consumer, we're making recommendations on things that within that set of parameters are able to be changed to take the loan from ineligible to eligible.

Got it. Okay. So that's real useful. That makes a lot of sense. I find it very compelling. There are some real, I think, compelling actual implementations of AI in the mortgage process that feels like active improvement for the consumer along the way.

I've been in the industry for almost 20 years or something like that, and so we have a real good feel for understanding what the pain points have historically been for the loan officers, the mortgage brokers and other people within the capital markets, lock desk, et cetera. Everything we do all day every day is just to understand how to solve those pain points where there has been limitations across other legacy antiquated software companies and solve them. I mean the industry has been conditioned to operating in a certain way for a long time as a result. I mean software's job is to drive costs down, make us more efficient, do more with less. It's not like, oh, the refi waves are here, let's hire a bunch of people. And then when the market turns, which it inevitably always does, there is this kind of unfortunate reckoning that's not good for anyone. It's not good for the people at the mortgage companies, it's not good for the mortgage companies themselves and banks. So we're trying to make this where no matter the loan volume, it scales up and down and in doing so, driving cost out of the system and like I said, increasing profitability.

Yeah, that's terrific. Okay. You mentioned that the cost of loans has gone from like 2,500 bucks to 12,000 in recent years. What drove that change?

Well, probably it depends on who you ask actually. I think there's a lot of things that have driven it up and I could point to hardly anything that's helped to drive it down until recently until modern solutions like Poly have come about. And so

Is it like documentation things like further compliance or more detailed things or I know that in some of those government backed pools, if the contract isn't done exactly right, you end up having to take it back out of the lender has to write back, is it those kind of things?

If you just look at every component of a loan and the associated expense of a loan, every part of it is up. Whether it's origination costs, compliance costs, technology costs up across the board. There's not one dimension that's driving it more than some other dimension. They all dimensions are up.

It helps me get the framework of what the technology is, what you're building and where you're focused though. Tell me about the origin story. How did you get to deciding to start Poly?

Yeah, I had started a mortgage company in 2013. The original idea was to figure out ways to drive down the cost of origination since I started to see it going up. But within a couple years, about a year and a half actually, I just decided that the much, much larger pain point in the industry was the pricing engine we were using. It lacked any amount of flexibility and configurability and granularity that I wanted and needed to be able to configure margins and pricing within how I wanted to do it. And as granularly as I wanted to do it, it lacked automation. The analytics was, everything about it just was old and bad and that was back in 2015 and not a whole lot's changed in 10 years. So we set out and we built our own proprietary pricing engine at this mortgage company and we saw a huge ROI in value.

(09:08)
And then a few years later I went to my board and I said, Hey look, I've got this much, much bigger vision, but if we were just to stay focused on in the pricing engine space, I think there's a real opportunity here. There's been basically one company for 20 some odd years doing the same thing over and over again and it didn't work for me and it doesn't seem like it's working so great for everyone else. Let's go try to build something that's completely cutting edge that can really be an agent of change and really transform things and really rethink things and from the bottom up. And so we sold the mortgage company, we took some of that money and we started Poly and that's how the company got started.

That's terrific. Some of my favorite entrepreneurial origin stories are like I have a need when I started Altos researchers because I needed to know what was happening to my home price and my little overpriced Silicon Valley mortgage. That's terrific. So it is, being the first customer is a powerful position to be in on that

And you understand what's out there and the pain points and you try to do your best to figure out a way to solve 'em. And it extends beyond just the product. I mean it also has to do with the service and the support and really being deeply passionate about the, we call 'em partners, customer partners and doing the right thing, always going out of our way to create ROI and value and success and happiness and so forth. And we are very, very passionate about that here.

Yeah, great. Okay, that's really useful. That's a great origin story. I love that about where that the company came from. So let's talk about some of the data for a little bit. We use at HousingWire we use poly rate lock data for our mortgage rates center so we can see the actual rates that people are locking on loans, we can see how much they're paying. So first tell me about the nuances of that data and then let's talk about other data that you have there and other insights we can see about the world that you can see by looking at a bunch of loans.

Yeah, so we're real excited about the partnership obviously with Housing Wire and it's been phenomenal and can't say enough about the team over there and in terms of it's differentiated, right? It's real time, it's actual lock data and we're really, really proud to be able to participate in it with you and HousingWire.

So loans come in, this is at the lock is like I'm buying the house. It is before the loan is actually executed. And so what does it tell us? Does it lead?

It's a leading indicator. A lot of the data in the industry has historically been off of a rate sheet. Well a rate sheet doesn't really tell you it's what's submitted by the industry. It's not necessarily what their actual rate sheet is,

What the actual close is, what the deal gets done at

Or what actually the rate sheet is. The actual price the loan locks at could be different than even sometimes than the rate sheet. It just depends on how people where they lock and it could be a different time of the day and a reprice and all these various things that are going on. And so having the lock data gives you the actual interest rate, lock commitment that the consumer is signing up for at that moment in time. And so it's a leading indicator is to where things are in the market versus as loans are sold into the market, which are two obviously different things because the market's moving up and down all the time

And it's always surprising how many infinite knobs there are to twist to actually get that final rate that you're getting. And we like to give it a one headline and rates are 6.85% this week, but what the hell does that mean?

It's funny when I talk to people not in the industry, they just think, oh, it's just this one rate. They don't realize there's all of this. It's like all this magic that goes into figuring out, okay, how do you get to that price or the rebate and the rate for the consumer? I think that's what makes one of the things that makes the industry so fun. I don't know. So it is interesting

And that's why you need a pricing engine.

That's right.

Which is sort of the core of what Poly is doing. And that was the pain that you were experiencing back when you were

That's exactly right.

Or a lender. Okay, there's a million knobs here and how do you turn the right ones at the right,

Exactly, exactly right.

Got it. Does the low volatility, the volatile, that range of the tighter range rates, does that impact, do you see consumer behavior changing? Do you see consumers responding to that?

Yeah, we've seen an increase in cash out, a little bit of rate and term and it has not been a major move. It's been a pretty subtle move in the market, but you could just see that, gosh, even small little changes are creating more refi opportunity. It used to be you need 50 basis points or some amount that actually was or maybe more. It is kind of small little changes that are making big differences right now.

That's a fascinating insight. Okay, that's really interesting. So even a small move in rates, people are ready to jump on it.

Well, small relative to the last 10 years, we're used to the last 10 years, what was it, 2013 or 14, we had that little flash crash. 2016 we had Brex at 2000, was it 18 summer we had the China trade war, 2000, was it 2021 was covid. And then of course the whole unwinding of covid. There's been crazy volatility up or down and for the last several years, yeah, we've been up, I mean since we've kind of gone and settled into this more recent range, it's been lower vol, there hasn't been major events and so maybe we're due for one, I don't know. So I'm saying on a relative basis, some would say this is a kind of a medium-sized move currently, but not compared to the last 10 years,

Not compared to the big crazy things. And so the question that comes to my mind is if we see consumers with hair trigger ready to jump, does that mean if we get lucky with mortgage rates this year and we get lucky with say inflation cools, and so the bond market rallies more and rates slide down, does that mean we are poised for even bigger activity than maybe we'd be expecting otherwise?

I learned a long time ago, I

Won't hold you to it. Yeah, don't these things, let's speculate with me here.

If anyone knew the answer to these things, they would be trading QIPs and Turks and Caicos or something, right?

(16:41)
I don't know. So my general opinion is that there are a lot of people on the sidelines waiting to buy homes, upgrade, their families are getting bigger, downgrade, there are kids have left the house, people who own homes that are locked in at these two and three quarters or three and a half percent rates where life has occurred and any number of things and they would like to move or whatever is going on in their life, but they're kind of stuck because these rates are so low. My personal view is I don't think rates need to get near where they were in 2020 21 to unlock all that purchase activity. I think it could be personally more in the fives, maybe lower fives, but in the fives and now to see major refi, your guess is as good as mine. I do think you'll see activity in cash out in the fives rate term. I don't know. I mean we have originated, I don't know, I'd have to look like 4 trillion or about three four some odd trillion since covid over six maybe. So there is some pent up probably refi just in what we've done here in the last three, four years. So personally I'm very bullish on myself, but I'm also an optimist. I have to be.

Yeah, well and that's terrific. It's funny, I've been doing the data in the housing market for a bunch of years and the traders always say, why aren't you trading on it? I've proven over and over again that I am not a trader. I do the data you trade and so I'm not going to hold you to that, but I do appreciate your take on where the activity is and seeing, for example, that even in the last few weeks we could see things like cash outs increasing and we're down maybe 40 basis points. So

Not a major move. Not a major move mean

And that's right. And on our data, we were just at the housing wire, we just host us, did the economic summit, the housing economics summit it the other day I did the talk where I showed we barely brushed 6% last September and that buoyed the market for four months in transactions and home prices. And so I see 6% as really a catalyst point for more transactions.

You're probably more right than me. I said fives, I don't know. Who knows?

Terrific. That's really interesting. Okay, so it is actually interesting though. So rates have been up for three years and you have an optimistic take on where they go from here. By nature, optimistic, do you build sort of rate forecasting into thinking about the business and do you have to build on it that way?

No, no. I've learned a long time ago to not make business decisions based on trying to forecast interest rates. Right? At Poly, we like to look inwards, we don't like to look outwards, we don't look at rates. We don't care what our competitors are doing. We don't want to be doing those types of things. We have a vision, we have a mission and who we bring along with that or our customer partners and we want to understand over the last 25 years what are all of their problems and how can we help solve them. And so that's the input we take in. We don't take in rate data, we don't care what our, we ignore. If we looked at all that kind of stuff, we'd be going backwards, not forwards. And so we're so focused on ourselves, we're so focused on continuous improvement, we're focused on the passion and culturally we have for our customer partners and enabling their success. And all we want to do is execute on this long-term plan. We have a 10 year plan, we then break back it down into five, then we back it down into two and then we back it down into quarter by quarter and that's who we are.

Awesome, terrific. Let's switch gears and so we've been talking about the consumer behavior at the rate level right now and that change on the other side, the capital market side, are there trends or there things going on there now that you are sensitive to or that I should be paying attention to? Changes on the capital market side or things like that that are important for your customers?

Not currently. I think that things are, like I was saying, it's like things are pretty shocking. I mean it's been a nice long kind of tranquil period of time unfortunately with an elevated rate environment. But there haven't been, in my opinion, at least material changes in the capital markets. I mean things are functioning liquid, things are real durable right now. The Fed hasn't done a whole lot really. I mean their forward forecasting is so buttoned up. They haven't moved much off the balance sheet. So I think no news is good news it seems. Have you seen or still heard of anything?

No. And I'm not a capital markets guy so I'm interested in insights about there are things, are there shifts for example suddenly in like we're doing sudden regulatory body changes, does that impact, are there things on the horizon they go that might unlock this or I always like to fish for those kind of insights that you might have that I don't get to see every day

Take the regulatory thing. There is so much activity in DC right now. I don't think anyone knows what to do, what to expect, how to react. I think everyone is in a wait and see type approach right now and you kind of react to what is coming or not coming. And so I just would not certainly at poll drive the business forward based on anticipating for example, what will or won't happen coming out of the regulatory framework. I mean there's been speculation for example, that the agencies get privatized. That would a major change for the industry obviously. But boy, when you start to actually do the math on it, I'm not sure, we'll see.

Not sure if they could pull it off. You

Mean wouldn't? I'm not betting on it, but I've heard about this.

Yeah, no, and that's useful for me to hear, right? There is top level jawboning about it and the question is can you really pull it off?

Look, I guess anything could be done. I could see scenarios where I could see scenarios where it's done and I could see a lot more scenarios where it's not done. For example, on that thing, I remember the first Trump administration, they were also talking about privatizing. It never happened. It just is a massive body of work and they're busy doing a lot of other things right now. I don't know if they have the capacity and bandwidth to take on this. This is not the biggest problem in their opinion, that needs to be solved today.

They ran out of time last time around

And by the time they get to this, they probably run out of time.

Do you have a take? So let's assume it happens in some way and then even that is full as fraught with assumptions, but do you have a take on what that might mean to your customers?

So what's interesting to me is we don't even know what would happen. We've only ever really operated with the agencies to have it minimum an implicit guarantee. Now it's like explicit, it's owned, right? The government's got it. And so now if the implicit guarantee is also removed and it's completely 100% privatized, I don't know, does the 30 year mortgage exist anymore? Maybe not. So I don't think the private markets would finance a 30 year fixed mortgage. It would probably move more to how things are done in Canada or Europe or Australia. And so that would have a major impact for consumers who have been conditioned to operating off of 30 year fixed mortgages for decades and decades and decades and decades and decades here. So yeah, that have a material impact on the consumer or the lender or the bank. That'd be fascinating. I could see banks owning more mortgage assets then everything moves to more arm. It would be a real shift. It'd be a real shift. You refi probably not happened for a long time. Anyone who's got a 30 year fix has not move into a three one arm.

Right. Well I've thought about that. That's a really great way to look at it. Like what a fabulous deal. The 30 year fix mortgage is,

By the way, no prepayment penalty. It's like we've got the best mortgage finance infrastructure in the world. Sometimes just the grass isn't greener on the other side of the road. We should just kind of enjoy what we've got here. It's functioning. It is liquid, it is resilient. There's backstops. I wouldn't touch it.

And of course 45 million of us are locked in at 3%,

Right? What are you going to trade your 3% 30 year for a 4%? Three, one. Why would you do what I'm just making

Numbers of. Yeah, exactly. Okay. That's a really, really interesting insight. It can be fascinating to see if they pull something off and if they do, what happens to all those pieces?

See the thing that this is what I've always thought on this topic, it's wonderful to talk about privatizing them and maybe it kind of creates some amount of liquidity for the government in the short term. But at the end of the day, the 30 year fixed mortgage has been one of the most effective tax cuts for the American consumer in the country that ultimately the US government finances but not themselves. The bonds are distributed everywhere else other than when the Fed bought them for these moments in time in 2008 and 2021. And so that's why I think the government will ever do anything. I think they'll say, oh well wait a minute. Having a 30 year fixed rate at whatever, 3%, 4% or whatever it is, is a massive tax. It's deflationary and we need all of that. We need as much of that as we can get right now.

Alright, that is terrific view. I appreciate it very much. Okay, so along those lines, one of the questions I'd love to get from my guests is based on your expertise, are there things that in that the headlines are getting wrong right now

In the mortgage industry or

Wherever you see it, wherever you see it that you like to point out? Because sometimes we read the headlines and we assume that's the case, but

The thing there is that kind of requires me to put an opinion on something that I don't really know honestly. Do I have views on certain things in the political world or the technology world or this thing? Yeah, I do, but there's really no upside to dispute any of those things because first of all, who am I to say? And second of all, it kind of doesn't matter for me.

Okay, fair enough. So alright, so you talked about your 10 year vision, 10 year vision for the company that's got to have some 10 year vision for the rest of the world, the mortgage world and the home buying world in the us. Can you tell me a little bit about your 10 year vision?

We want to do everything we possibly can to help drive down that. We want to help drive down on that cost hope ideally to less than a thousand dollars a loan. And I think over the next from 12 where now, yeah, and I think over the next, I mean that's kind of ambitious, but we are in the midst of arguably the greatest platform shift we've ever seen in the world. And I think the teams and companies that understand that and can execute on it are uniquely positioned. And there's a lot of things that go into this. It's not throwing some random chat bot thing into the world and saying, okay, that doesn't change anything. And so we are really, really focused in on, okay, how can we drive that cost down as much as we possibly can? Which includes obviously greater profitability for the lender, but ultimately because as long as there's a 30 year or just mortgages are, there's not more than a thousand products or something, it's still kind of a limited product set in the industry, it's still then becomes competitive back to the consumer. And so by driving those costs out and helping to increase revenue or profitability per loan, ultimately that will move up funnel into the transaction between the loan officer and the consumer. And so if we can do that, there's currently 12 trillion or whatever it is outstanding in the industry and that's a meaningful savings for the country in all the consumers that have loans. We are very, very focused on that and we're definitely a small part of it, tiny part of it right now, but we don't want to be small forever. So we're excited.

That's terrific. So you called it the greatest platform shift in the world and I assume that's an AI comment.

Yes.

So in the poly world, you weren't really doing AI in 2019, but there were some big problems you were tackling and now you have the AI platform layer also into that challenge. And so

Built, the system was built initially with a ton of machine learning and data science embedded in it. So it was never really not, it was not like chat PT comes out in, was it 2021 or 22? And all of a sudden we're like, oh we better do this. No, no. We had, we had some of this stuff back at the mortgage company back in 2000 16, 17, 18. It was always part of our roadmap. It was always part of the vision. This is not some new invention that we just kind of thought about after chat. BT came out absolutely not.

So you are doing things like with a machine learning environment. You're looking at a bunch of the knobs as we talked about earlier, and outputting profitable scenarios or optimal consumer.

The easiest way to put this is to pioneer something. You have to have the vision over the long term and you can't just be reactive or start talking about a bunch of stuff that because it's popular, all of a sudden it becomes popular because of the pioneers. And so in this space we were the first to market with it and we're going to keep on leading it and we're going to keep on tripling down on our investment in it.

That's great. So as you focus on the cost of the loan and that efficiency in the transaction, are there other dynamics in the housing market that, as you pointed out, you try to stay internal, but I'm interested in other things. Anything else that might come up to mind that this is really helped shapes my vision for the next 10 years? Tell me any other variables in there that I should know about it?

We're very, very, we've got so focused on where we want to take things and certainly variables come about daily, monthly, quarterly, annually, whatever. And you take that input in and you think about how that fits into the overall puzzle. But that doesn't change the vision. That just says, okay, that's a puzzle piece to the overall mosaic and we won't change that approach. In order to lead and pioneer and be the most cutting edge, you got to keep going down the path that you want to go down

For sure. And you got to hold that in the face of some of those variables. Are there any releases, poly releases or new things that your announcements that you're making that you want to make sure people pay attention to?

We do releases every couple two weeks. We're pushing stuff out very fast always. We do have some really, really cool big things coming this year, which we'll talk about throughout the year. We're really, really excited about them. They're truly transformational. It's been stuff that I've wanted to do since 2015, so I'm really, really excited. It's always been kind of building up to these kind of moments and we've got a couple of things coming that we're really excited about.

Excellent. Well we will look forward to those announcements when they come out. We'll make sure we amplify them for you. I appreciate your time today Adam. It's been really great. I understand the caution on making big pronouncements about the industry, but you really opened my eyes to a couple of things in there, which is really my whole desire with these

Conversations.

I love to learn from what you're seeing in the world and what you're building in the world. And so that's why I do these. I continue

To Great to be with you. Thanks very much

Polly.io. Any other place that you're like people reach you LinkedIn, anything like that?

LinkedIn email almost. It feels like everyone has my cell so I'm available 24 7. That's the way it goes.

I've had that for 19 years where people have my cell for the company. Yeah. Okay. Adam Carmel, Pauly is the company. It's been great talking to you today and really appreciated your thoughts and everybody, this is the top of mind podcast and if you appreciate conversations like I get to have with great folks, great leaders in the industry like Adam, we always appreciate review unless some nice five stars and wherever you get your podcast, that helps other people find us. So. Alright, thanks everybody. We'll be back more soon.

Thank you