The Exchanges

Q How do you construct a portfolio across these assets?

A There's a lot of ways to slice and dice the portfolio and to construct it. And so we, some of the computer system stuff helps us to stress test and shock the portfolio and see where does in aggregate the portfolio do well or do poorly. And we're always trying to figure out is can we modify our risk more cheaply internally? And what I mean by that, so like simple examples are some mortgage bonds do well when prepayments are high and some mortgage bonds do well when prepayments are low. So if you can buy the ones that do well when prepayments are high from people that are afraid prepayments are low and the opposite, you can get to a situation where you now have two bonds. One does great when prepayments are high, and the other one does great when prepayments are low, and they both do decent when prepayments are middle. If you can fuse those two together cheaply, you now own something that is somewhat indifferent to prepayments. So we're looking across collateral type, rating, structure, Are these bonds that do better when things are faster or slower from a prepayment from a default perspective? One interesting trait to do in commercial mortgages is to buy interest-only securities, where if there's a delay in the refinancing of a property, the value of those soars, because they only get interest, and as soon as the property is refinanced, they're shut out. Whereas if you can put s…

AI assessment note: “some of the computer system stuff helps us to stress test and shock the portfolio”

Q How did you get onto Wall Street originally?

A So it's kind of ironic or funny story, I guess, which is that my family was in the real estate business, small-time real estate business, and I assumed I would go into the real estate business, and I graduated college in 91, which was one of the times that was hard to get a job. So two things happened. One, it was hard to get a job, and two, I read the book Liar's Poker by Michael Lewis, and I said, wow, this sounds amazing. And I had never heard of, really, Wall Street career. I just didn't come from a background that, that was talked about, or I knew anyone who did that or anything. So I read the book, And I thought, this is interesting. And while still sort of trying to get jobs in real estate, I tried to get a job on Wall Street as well. And although I was Magnicum Loud, I graduated unemployed, because 91 was a bad year. I remember going to interviews and people saying, wait, you're Magnicum Loud economics, and you're unemployed? Like, I had a two and a half GPA, and I majored in biology, and I got a job in M&A banking in Goldman Sachs, like right out of college. I'm like, yeah, well, you graduated before the 87 crash, and I graduated after. So it was a different time. Just trying to find a job, living at home after college, trying to find a job and taking any sort of interview I got. And, you know, some of it is luck. And then eventually I was hired at a middle office type…

AI assessment note: “eventually I was hired at a middle office type job at Credit Suisse.”

Q How did you find your way into the trading desk?

A So I found that the trading people seemed to have a more interesting job, and maybe they made more money than some of the other people, and they weren't working the crazy long hours like I was as an analyst, so, so what I decided to do is, every day I worked on the 18th floor, and the trading floor was on the third floor. So I would go downstairs, and if anybody was traveling for work or out sick or on vacation, I would just sit in their desk. And I would do my job, and I wouldn't say anything to anyone, but I would just sit there and do my job. And so one, people would see me. Two, I would hear them talking and pick up something. And just, I thought I would be this way, top of mind, if they ever thought something. So when the BP's trader at the time needed a second person, I was sort of the logical person to do it. One, I was Known for being really good at my job. And I looked up in the database and I had structured more of these deals than all the other four people put together. And two, I was always around. So I was invited to join the BP's desk, which was the sleepier backwater part of the business. It wasn't really flow trading. There weren't a lot of trades. We sometimes didn't even do one trade in a day, but it was more about really understanding the risks inherent in these structures and the leverage in them. And it was not portfolio management. We would buy things Not …

AI assessment note: “if anybody was traveling for work or out sick... I would just sit in their desk.”

Q What were the different categories of these structured product assets?

A I think one of the things that makes it exciting to do it right now, and I think a reason that there's a lot of alpha that can be mined there, is you can sort of slice the universe a bunch of different ways. We have four sector teams, and each of them is manned by a person, so our sectors, the way we describe them, are consumer ABS, RMBS, CMBS, and CLOs, and that's sort of one really simple way to do it, and say, hey, we're 25% RMBS, and we're 35% CMBS, or whatever the case may be. And a different way to look at it would be to say, well, how many of your portfolio is double B-rated? Or single B rated or triple B rated or what have you, regardless of the type of collateral that it is, right? So that's a different way of slicing it and thinking about risk, which is not collateral based, but ratings based or leverage in the capital structure. Another is seasoning. Some of the deals were issued 1020 years ago, and the way those structures have evolved over time and the loan to value ratio of the underlying collateral is very different. So a double B that was issued in 2012 And it's still rated double B for whatever reason. Doesn't look anything like a double B that was issued in 2022. But they're both double Bs. So if you ask me how many double Bs do you own? One answer is, well, I own two. Another is, well, the 2012 one, it has none of the same characteristics as the 2022 one. So …

AI assessment note: “our sectors, the way we describe them, are consumer ABS, RMBS, CMBS, and CLOs”

Q So if you go full circle, you started getting into the business reading Liars Poker by Michael Lewis. You're doing this esoteric credit stuff. And a couple of years later, Ryan Gosling's playing you in a movie. What was that experience like going from trading stuff and fighting the crowd to getting something right, and then all of a sudden it's on the silver screen?

A Well, it didn't feel like all of a sudden it was on the silver screen for sure. What was an interesting experience was sort of fighting to maintain a position. And then just a few weeks after that being on the cover of the Wall Street Journal. So I went from being chicken little to cover the Wall Street Journal in just a matter of maybe not days, but certainly not more than weeks. So that I would say the time to Mr. Gosling playing me in a movie that was a little bit longer than that. And how does that feel? I mean, it's certainly, there's a lot less appealing people to play you in a movie than Ryan Gosling. So I guess I'm grateful for that.

AI assessment note: “there's a lot less appealing people to play you in a movie than Ryan Gosling.”

Q There was this wave at the time of all of these types of assets sitting on bank balance sheets and now sitting more in the hedge fund community. You effectively participated in that by switching over. What was it that caused you to want to form your own hedge fund?

A Sure. My father was a small businessman. I always had a certain amount of interest in starting a company. I think inertia is a very powerful thing, right? You have a job at a bank. It pays a relatively good salary. None of the risk is yours. If you lose a lot of money, you just get fired. You don't lose any money. You don't have the hassle and the risk of setting up a company. It was really For me in the beginning, for a while, and still to this day, like, this desk, I own this desk, and if we ever close Libra Max, like, I'm going to sell this desk for something, and when you work at a bank, you own nothing, and so that was different, but the inertia of not starting it is, one, I have a good salary. Two, I don't actually have to go find desks to buy. I don't have to rent office space. I don't have to hire a compliance department. I don't have to do anything, so while I had this sort of interest in starting a hedge fund for a long time, the inertia of working at a place where Where most of that stuff is taken care of for you is powerful, and you're making an income and whatnot. And then after the great financial crisis, we had one, I had a certain amount of notoriety that I thought could raise money, part of the Michael Lewis's book. Two, we thought that the asset class was really attractive in the aftermath of the GFC, that returns would be sort of easy in the beginning. And th…

AI assessment note: “after the great financial crisis, we had one, I had a certain amount of notoriety”

Q How does that underwriting process work such that a bank calls you up and you can have a good sense of what this less liquid collateral is worth that you can capture that for a seller or whatever the case may be?

A One of my partners, Glenn Perillo, runs an analytics division for us, and so we have a team of computer quant people that are building models that can quickly sift through the loans. Now, and I mentioned this before, back when I started, each bank had their own internal software that valued these bonds. Now there are companies that sell those, and the good news about it is that everybody is looking at the same thing, and occasionally our guys are able to identify that there's a mistake because somehow they've read the document and the way that this structure works in this third-party software. Is not consistent with that, and that's obviously great if we see that. Well, it's really great if actually the bond is worth more than the third-party software says, and it's great to know it if the bond's worth less and to not buy it. So we have a robust team that builds models that can sift through the loans quickly and make different kind of AI-related predictions about defaults, prepayments, whatnot, and so that, that model will say, hey, this bond's worth 80 or 90 or whatever the case may be, and we know where we can buy it, We're not a quant shop in the sense that we don't take that price, but what we use this for is a sifting mechanism, right? So we'll put 10 bonds through the system, and that system will say, hey, I think you should buy these five, and you should avoid these five…

AI assessment note: “we have a team of computer quant people that are building models that can quickly sift”

Q What steps did that take over the last 12 years from that initial focus on RMBS to where you are today?

A So I would say, I mean, there's two different things. One is the focus on our original hedge fund and growing the business, and two, on the asset classes. So like when we started, RMBS was maybe 75% of our assets, and it's mostly been somewhere between 15 and 30 for the last handful of years, right? As the RMBS trade became less attractive relative to other things, we moved into other things. I'll give you one example. Student loans really lagged the recovery, and we looked at the student loan opportunity a few years after we launched the company, and we said, this sounds like subprime. In the subprime case, people said, no one's going to pay these mortgages. 80% of the people are going to default, and we're going to recover nothing. And there was a period of time where people said, all these student loans are bad. Everybody owes 200,000 dollars, and they work at Starbucks. Again, it's not a question of good or bad. It's a question of how bad. So we first of all increased significantly. Student loans at one point were 30% of our master fund, and they were like zero when we started the company. We at one point did one of the first of our sort of bespoke funds, where we did a fund that was Just on student loans. It was a private equity more kind of structure in terms of like locked up capital with a date certain on it. We were able to raise some money for that, which was returned…

AI assessment note: “As the RMBS trade became less attractive relative to other things, we moved into other things.”

Q What is that fundamental human part of the underwriting of these assets beyond what the machine spits out?

A I think that's a combination of acumen and experience, right? So the team is reading the documents for themselves. The team has their own anecdotal view about different collateral types, loan types, geography. So we really like multifamily in the Southeast, and we hate malls in the Pacific Northwest or whatever the case may be. So the model also reflects that, but the human has a view of that as well. And then it's at the sector head level looking at The panoply of potential returns and saying, you know, we feel good or bad about the upside and the base case and the stress case, if you will. And then there's the direction for me on high about the things that I want us to look at at any given time.

AI assessment note: “I think that's a combination of acumen and experience, right?”

Q All right, so before I turn to these closing questions that you haven't answered, I have to ask, like, what was the experience of being portrayed in a movie like for you?

A Well, first of all, the experience is not over, and I still, to this day, occasionally get emails from people who were like, I read the book, or I saw the movie, and I want to be your friend, or can you give me advice? And I sometimes forward them to Mr. Lewis. So the beginning of the book, The Big Short, he said, well, he wrote Liar's Poker as a cautionary tale. He didn't want people to go into Wall Street, is what he thought. And people read that book, including me, and said, this is exactly what I want to do, right? So at the beginning, the foreword to The Big Short, he said, well, that's why I wrote The Liar's Poker to get people to not do it. My hope is with this book that people won't want to go into this industry. And so when some of the more interesting emails that I get, I forward them to him. I'm like, you did it again. You wrote another book. So I'm married. I have four children. I'm not really interested in being particularly famous or anything like that. I don't have a big media presence. And it's weird at times when I'm checking into the airport and the person next to me, here's my name. And they, they say, oh, you're the guy from the big short. Like I said, I get sort of emails from people that I've never met. I don't love that the parents of my children's friends feel like they already know something about me when we first meet. I don't love that, but I certainl…

AI assessment note: “I certainly don't think I was portrayed as somebody who wasn't smart”

Q How did you set that initial investment strategy? Whereas you mentioned you felt like you could own a bunch of assets that were going to be worth a lot more. And that sounds like a medium term kind of tactical trade compared to a more evergreen investment strategy.

A One of my pet peeves is people call structured products a trade as opposed to an asset class, which I disagree with. And I think that first of all, I've been doing it 30 to 33 years. So it's been a life for me, not just a trade. When you start something, you can dub yourself whatever you want. And either people give you money or they won't. But one thing I've learned in the 12 years, it's very hard to then change. You can claim anything you want to claim. But afterwards, if you try to change that, people say, well, that's style drift. So we, we said, if we're focused on what we do, we'll attract assets. Not really thinking that down the line, if we'd said, hey, we're really smart people, and we could do anything. And we could buy Greek bonds if we want, and we could buy Malaysian equities. We can't do that now because we didn't say back then that was something we would do. We were very focused on structured products, and we think there's this great opportunity in mortgage-backed securities right now. You can buy them to a 12 or 14% yield to really onerous assumptions that have a lot more convexity to being better than worse. You were buying them to a 12% yield to 85% of the people defaulting on their mortgages. Could it be more than 85%? Maybe. But, like, not everyone is going to default. And as it turned out, like, 40% of the people defaulted. A lot. But what you learn in inve…

AI assessment note: “people call structured products a trade as opposed to an asset class, which I disagree with”

Q Well, why don't we go all the way back to your background?

A Sure. Well, I don't know how far we want to go back. Ted and I went to high school together, and people are the same. My parents were cleaning out my bedroom when I was about 25, and they gave me some papers, and included in that was my second grade report card. And I was just then, at the same time, getting my review from my employer at the time, Credit Suisse vs. Boston, and And you put my twenty-five-year-old employment review next to my second grade report card, and the strengths and the weaknesses are the exact same. So people are who they are, and you can work harder to be a better person, or, or be happier, or calmer, or whatever it may be, but we each are the person that we were sort of made to be, and I haven't changed that much, so it's sort of interesting that you have some knowledge or memory. We went to a small high school, so we certainly knew each other, and to this day, I remember the names of your brother and your sister, so yeah, we know each other.

AI assessment note: “included in that was my second grade report card. And I was just then”