He details the impact in his book The Quants — a highly readable, very entertaining look at the new breed of mathematicians and financial engineers who got caught in the middle of it all — it was one of my favorite books of the financial crisis.
I spent some time with Scott chatting about the book, the players in it, and life after the WSJ.
Part II will appear tomorrow.
Barry Ritholtz: I am sitting here with Scott Patterson, author of the book ‘The Quants,’ staff reporter for the ‘Wall Street Journal…’
Scott Patterson: Former staff reporter. I’m on my own, I’m freelancing.
Barry Ritholtz: How do you like that?
Scott Patterson: I kind of love it.
Q: A lot of people have had a hard time making a go of it. I use Dan Gross as the standard-bearer for freelancing for the ‘New York Times’ and Slate and eventually ‘Newsweek,’ and when ‘Newsweek’ imploded, ended up at Yahoo Tech Ticker. Who else are you publishing for if you’re not regular with the ‘Journal’ now?
A: Whoever will take me, basically. Whoever pays the most. I’m mostly working on this new book and focusing mostly on that.
Q: What’s the new book going to be?
A: It’s about the history of electronic trading and markets, high frequency trading and where all this HFT stuff came from. It’s a very fascinating story, I’m learning a lot as I go.
Q: Let’s talk about your background. For a scribe, you have a lot of mathematical-leaning interests. Where did you go to school for undergraduate?
A: Completely no math, I’m an English liberal arts guy, I studied anthropology. I think maybe my connection to this comes from my anthropological background, because I sort of see these quants as being a subculture of Wall Street, they all hang out together, they have their strange little things that they do, their rituals, their language, so in a way, you could see this book as being like my journey into the heart of quant-land.
Q: So the Cro-Magnon essentially pushing aside the Neanderthals who spend their time doing silly stuff like fundamental research.
A: Right, gut traders.
Q: Well, that’s a whole different group. Where did you go to school?
A: I went to James Madison in Virginia.
Q: Did you keep going, any graduate work?
A: I’ve got a master of arts and English.
Q: So zero math…
A: I took math courses, I took statistics and some math in college, but it definitely wasn’t my focus. I can’t say I was really that good at it, but I understand the basics of how these guys operate, and what I find is you just ask them over and over again to explain what they’re doing, and you ask 10 different people and you’ll eventually get it. It’s really not that complicated, although some of the things they do do get so crazy and so complicated that I don’t think anybody really understands it.
Q: In my arc of bailout/crash books, “Quants” is the follow-up to “The Myth of the Rational Market’” — that is the economic, philosophical, “How did we get to this silly belief?” “Myth” approached this from the economic academic perspective, then your book comes at it from the mathematical traders perspective
There’s a terrific data point that James Montier used in one of his books on behavioral investing that says that IPOs are essentially giant money-losers, and yet people are always clamoring for them, and if the market was really rational, this wouldn’t happen, nor would any of these bubbles nor would any of these 100-year collapses that seem to come along every couple years. “The Quants” was a great counterpoint to this.
Q: I think this whole idea of rational markets, from an outsider’s perspective, when you come into the world of Wall Street and you see these theories, the efficient market hypothesis and whatever, you just scratch your head and wonder, “Where did these guys come up with this?” because it’s patently obvious that Wall Street is not about rational mathematical behavior, it’s about the polar opposite of that, it’s about greed and fear, and I think that basically, these quantitative models need to assume a rational market as a starting point, because if the market doesn’t act in some form of predictable, rational, mathematical way, you can throw it all out the window and forget it and go home and do something else.
Q: There’s a spectrum, on the one hand, you have the belief that markets are perfectly efficient and rational. Then there’s a belief that it’s completely and totally random, and the truth, as it often is, lies somewhere in the middle, where there are times that it’s somewhat predictable, the trend continues on. There are times when stuff is just insanely random from day to day, and there are times when you could use history as a guide up to a point, and others where it just veers away from it.
A: It works sometimes, I think a lot of times it’s self-reinforcing. You have these rational models that people started using, and lo and behold, the market starts behaving that way, because so much money is being put in these models, and yes, you can predict it, and smart people say, “This is how the market’s working, it’s rational, it’s efficient, it goes out of equilibrium and comes back into equilibrium.” I had this idea I wrote about in the book called the truth, which is shorthand for this belief that the market is rational, and people have told me, “That’s BS, these guys don’t believe in a truth,” but actually, some of them do, and I’ve had them tell me that, that they believe that there is this mythical truth, underlying how the market works, and that if you can figure it out, then you will be rich beyond your wildest dreams, so that’s not made up.
Q: They’ve drank the Kool-Aid and they ate their own dog food and really believe, as opposed to saying, “Hey, we’re not going to get this perfect, but here’s a way to kind of get an edge over everybody else.” Todd Harrison says this all the time: “Does technical analysis work because the levels matter or do the levels matter because everyone else is watching the same technical levels?” The question is, is it a function of true TA or that it’s the Keynesian beauty contest, you’re not voting for who you think is the prettiest girl, you’re voting for second derivative, who you think everybody else thinks is the prettiest girl.
A: Right, it’s game theory. A lot of these guys are poker players who I write about.
Q: Lets talk about the book. I was looking forward to reading this, and I was ready for something dry and tedious, and you start the book out with the big annual quant poker game, which really just grabs you. Please talk about that poker event, it really sets the table for the parade of characters and everything that comes across. You would think math geeks wouldn’t have quite as many characters, but they’re fascinating guys. How did you come across the big poker game?
A: I loved that scene, because it let me put all of these people in the same room together, which is right up the street here in the St. Regis Hotel (55th & 5th), and I was writing an article for the Wall Street Journal about the quants after the quant meltdown in August 2007 and focusing on this Morgan Stanley group, Process-Driven Trading, which is run by Peter Muller. Peter Muller, as I quickly found out, is a poker fanatic, and I learned just doing some web searches that the previous year, he had been in this poker tournament that is put on by Jim Simons of Renaissance Technologies, and Peter also helps run it, and he played against Cliff Asness, the founder of AQR, another big quant shop, in the finals of the tournament, and everybody had heard of Cliff. Cliff is a famous quant, but nobody’s ever heard of Peter Muller.
Q: Even though he’s running Morgan Stanley’s quant desk?
A: Right, very secretive guy, that’s one of the things that really fascinated me about him.
Q: And Morgan Stanley’s desk is not insubstantial, that’s a couple of billion dollars.
A: They, at the time, were the biggest equities trading desk at Morgan Stanley, they had five billion dollars or something like that, and nobody ever heard of them, I talked to Todd, who worked at Morgan Stanley, Todd Harrison, and I said, “Have you ever heard of PDT?” PDT’s been there since the early Nineties, back when Todd was there, and he never heard of him. Other people I talked to had never heard of them, they were secretive within Morgan Stanley, not just outside of it, but it was a super-secret group. That’s one thing I loved about it, was cracking inside of them. They love poker, and they equate that with trading and mathematizing it.
Q: It’s statistical, rule-driven, with elements of chance and you’re basically making probabilistic bets based on, “What are the odds based of this event occurring,” there’s some emotionality, because in poker, you are also playing the man. In many ways, there are enough parallels that people can say, “This is related…” I’ve heard guys say, “He’s a good poker player, I’ll hire him as a trader.” In reality, the skill set is very, very different. The touch points that are similar in a narrative make you feel like it is, but there are plenty of great traders who aren’t good poker players, and vice-versa. You can see that mano-a-mano competitiveness in the first chapter, you go over that personality-driven attitude. From “Liar’s Poker” we all know the expression ‘Big Swinging Dick’ (BSD), so there’s a little ‘whose is bigger?’ as they’re sitting down at the table.
A: I wanted to bring the people alive and show that this is about rational mathematical-driven quant strategies, which, like you said, can be really dry, but what I wanted to show is that these are real human beings behind the machine and there is a human element to how all this stuff works. Cliff Asness, I wrote how he gets really, really upset in 2008 when everything is blowing up, he’s punching computers, there’s real human emotion there, and I don’t think there’s anything wrong with that, I think it just shows that we have markets in which people are trying to predict what happens based on these mathematical principles, but at the end of the day, you have the real human elements of fear and greed and all of that stuff mixed in, which leads to blow-ups and a lot of times, this stuff just doesn’t work.
Q: A lot of the people that you’ve discussed in the book, it’s obvious you spent some time with them, interviewed them, but these guys are pretty below the radar…you mentioned the secrecy at Morgan Stanley, most of these quants just don’t do media, they’re not very accessible. Jim Simons has to be the most famous at Renaissance, who you tried to reach and just couldn’t…
A: Jim is the one character in the book who would never talk to me, much to my chagrin and frustration. Everybody else, I did speak with on one level or another.
Q: And some of these guys repeatedly.
A: Right, repeatedly.
Q: Phones, or are you flying to Chicago?
A: I met with some of them, I met with Ken Griffin right up the street here at the Four Seasons (57th between Park/Madison), that was a very interesting meeting in which we met at this restaurant in the Four Seasons, which was empty, and it was very bizarre.
Q: Was it off-hours?
A: It was in the middle of the day.
Q: Three or four in the afternoon?
A: Yeah, I think it was closed, but I’ve heard Ken has some ownership in the Four Seasons, so he can just do whatever he wants. It was an interesting meeting, and I emailed a lot of them – I spoke with most of them multiple times, and with Renaissance Technologies, I did have sources who knew the firm. I think I show I have a pretty good understanding, and I doubt anybody had ever written as much about Renaissance as this book.
Q: When I was an incoming math major in 1979, the outgoing chairman of the math department was Jim Simons, so I didn’t know who this guy was, he spoke at something where I got a glimpse of him for 15 minutes and that was it, with a cigarette dangling from his mouth, but nobody had a clue he was going to end up being not only this mad genius quant, but I think it’s safe to say his track record is the greatest hedge fund run over decades, no one is even close to him.
A: There’s no question, as far as anybody knows, I’ve heard about these secret funds that are operating completely in the shadows that have Medallion-like returns, — Medallion is Renaissance’s main hedge fund. So who knows about that, Medallion is 40, 45 percent a year, since 1989. It’s sick, it really is sick.
Q: It’s total madness.
A: It’s not Madoff, people keep saying they must be cooking the books or something. It’s ridiculous, they threw all the investors out, it’s only their own money, they’re definitely doing it, and it’s something I’m looking at in my new book, which is about electronic trading, but I’m also focusing on artificial intelligence, and in the early Nineties and mid-Nineties, Renaissance started hiring these guys from IBM’s voice recognition group who were doing artificial intelligence research on how to make predictions about what people are going to say and doing voice translation stuff. Somehow they figured out the machine-learning systems that they were using were applicable to financial markets.
Q: Sure, you have a decision tree where if you have this group of sounds, again, it’s back to probabilistic outcomes. If you have this group of sounds, the sound that’s likely to follow it is going to be one of these three options, and I wouldn’t be surprised if when you’re looking at this particular stock trading pattern, here are the likely outcomes, and what’s set up is it’s just making the highest probability bet with whatever limited information they have.
A: And they’re apparently very good at it, because as I point out in The Quants, if every voice-recognition guy could start a hedge fund, they would have done it, we’d have no voice recognition industry in this country.
Q: Simons started out as a code-breaker, so he doesn’t just have a math background, but he has a problem-solving background, which is somewhat unique versus some of the other quants that we see that are pure math-driven. I always thought that was something that made him special, although clearly a little confirmation biased, knowing his track record. You don’t know if that’s the factor.
Let’s talk about some of the guys you met and your impressions – Ed Thorp is seen as the father of the quantitative approach, he started out with ‘Beat the Dealer.’ This is a consistent theme, we see these guys playing blackjack, going to Vegas, playing poker. What was your takeaway with Ed Thorp, did you get to meet him?
A: Yeah, I met Thorp many times. He’s totally fascinating, he’s also probably…I don’t know if I respect anybody I’ve ever met on Wall Street more than this guy, just because of his demeanor. When you meet him, you know this is a guy who is completely straight-up, he’s going to tell you what he thinks, he’s a gentleman, he’s also just incredibly brilliant. He’s definitely old-school, he goes back to the days of the Great Depression when everybody had to pinch their pennies, so he’s not a gambler. The thing is, he learned how to beat these games in Las Vegas in a way that he didn’t have to gamble, that was his whole point, that I’m not gambling with my money. For him, it wasn’t about the money. So he was totally fascinating.
Q: Not about the money, but rather just demonstrating that a rule-driven, systemic approach to uncertain probabilistic systems can generate alpha. As long as we’re making the right bets, you’re not going to win every deal, but over the course of time, the high probabilistic bets are significantly going to out-perform random bets.
A: Right, it’s totally probabilities with blackjack, you never know if you’re going to win the next hand or not, but you know you’re going to win 55 percent of the next 200 hands, unless they’re doing funny stuff, which he also had issues with back in the early Sixties. The whole story of him going to Vegas backed by these mobsters in New York, Manny Kimmel, he didn’t even know these guys were mobsters, because he’s sort of this pie in the sky professor. It’s a great story, then he starts putting computers in his shoes. It’s a crazy story. He’s a mad scientist, he’s like a kid messing around in his garage, tinkering with things, he’s hanging out with Claude Shannon, who’s one of the most brilliant people of the 20th century, the father of information theory. I thought it was so fascinating, this guy who came out of this world, this MIT math world, hanging out with Claude Shannon, ended up becoming really the first quant fund.
And as I started picking apart all the pieces where all of those people came from, he kept coming up, I kept hearing people would say, “I bought Beat the Dealer, then I found out about Beat the Market,” which was his next book about convertible bond arbitrage. It was just this pervasive thing that really took off, and I never really thought he got the credit he deserved as being this monumental figure on Wall Street. I think most people have never heard of Ed Thorp.
Q: You make it clear in the book that Ed Thorp was the godfather of all the quants, and everybody from Cliff Asness straight down to…I don’t know what Jim Simons’ correlation is with him, but they’re almost peers, age-wise.
A: Right, Jim was also a student and teaching at MIT at about the same time that Ed Thorp was teaching there, and they have connections with information theory and code-breaking and things like that. I’m not aware of Jim Simons being inspired by Ed Thorp.
Q: They seem to be somewhat parallel, although Simons comes to the market somewhat later than Thorp does. Bottom line, you were just blown away by Thorp’s personality, his intellect, his whole persona.
Q: What I thought was kind of fascinating was early on, you start describing how things go awry in ’07, and if people are reading this and not understanding the timeline, the market didn’t really peak out until October ’07, and yet over the summer, we had this really…October 14, ’07, was the high point in the Dow and S&P, but we had this massive disruption over the summer. This is before Bear Stearns collapsed in March ’08, but after the Bear Stearns hedge funds started wobbling. The sense I got from reading through a lot of the book was all these guys, brilliant mathematicians, very clever guys, they pretty much had almost identical trades on. Is that a fair statement?
A: Yeah, it was a crowded trade, as they say.
Q: And so we see a lot of things start to take place. What I found surprising as this was going on was the inability of anybody to recognize the warning light on the dashboard flashing red, and I don’t know if it’s confirmation bias or just, “Hey, this has made us so much money over the last five years, we have to trust it,” because up until this point, nothing’s ever gone wrong. But it seems really smart guys with an ability to check for errors seemed to have just closed their eyes and jumped off the building.
A: It’s an example of, in a way, what it takes to do this kind of trading, because you really have to believe in it. That’s the impression I got from talking to Cliff Asness over and over again, was that he really believes in his models, and there’s nothing wrong with that. To put billions of dollars on the line based on your analysis of the way the market has worked over 30 years, and you think you’ve found this thing, this predictable pattern that’s going to work, then you go through periods of losses, you have to have the guts to be able to stick with it and say, “It’s going to come back.” So in a way, they have blinders, because they believe that no matter what happens, and even if you go through a period of de-leveraging, it will come back eventually, and the money will be back, the investors will be back, and everything will be fine.
Q: Prior to ’07, the draw-downs were pretty modest. They had months where they were down a couple of percent…
A: Cliff and AQR had been through a near-death experience in 2000 and in 1999 when the tech bubble was going up, because their models accurately predicted that all of these crappy dot-com stocks were worthless, and they were just going up and up, and they were shorting those things, so it was just killing them. They were months if not weeks from going under, so they had actually gone through a period where models were not working, they came out of it and it confirmed for Cliff that even if you hit a rough patch, things will come back, just stick it out.
Q: Any of the other quants have similar experiences? We can’t use Jim Simons as an example, but they had a good year in ’99, a good year in 2000, a good year in ’01…
A: There’s a difference between what a Medallion and AQR fund are doing, it’s important. Medallion and Pete Muller’s fund, PDT, had good years in ’99, 2000, 2001, because they thrive off of volatility.
Q: And their holding period is much, much shorter.
A: A very short holding period, so if the market is going up, and this is one of the great things about statistical arbitrage, is it has never failed in the history of the market until August 2007. There was a brief period when it first got started up in the late Eighties where it actually didn’t work for a while, but for the most part, it’s been one of the most consistent, reliable strategies ever created, in part because it’s really short term, and high frequency trading, which is really popular now, is a form of statistical arbitrage.
Q: Where you’re making very short-term bets. This book came out in early 2010. When did you hand the manuscript in?
A: I handed it in in late 2008.
Q: And it was a year later?
Q: Were you pulling your hair out and wanting to add stuff?
A: I was adding stuff up to the last minute when the editor told me, “No more,” and I was still trying to get more stuff in. As I was writing this book, the world blew up, and all of the sudden these guys…if you remember November, December 2008 was the end of the world, and that’s when I was supposed to turn this manuscript in, and things were just going crazy. So I was chasing to try to keep up with the story and changing the manuscript constantly, and I turned it in, and the editors wanted me to re-write it, because I turned in this really long thing with a cast of thousands, and I wrote about the CDO stuff in depth, and a lot of other hedge funds, I had far more extensive chapters on D.E. Shaw, and they wanted me to focus.
Q: So you had The Quants and The Big Short in one book.
A: I was shooting for that, which obviously I was trying to bite off more than I could chew.
Q: But the focus is terrific, you tell the story of the build-up and the collapse through the mathematical models, through the personalities.
A: And they all know each other, too.
Q: So if this went in in 2008, the question I was leading up to earlier was…let’s go through some of the characters and find out what happened to them, it’s now two, almost three years later.
Did anybody blow up, did anybody recover? What’s the net takeaway?
Part II will be published tomorrow morning . . .