In 1965, Esquire tasked Tom Wolfe with a profile of Junior Johnson, a whisky-runner-turned-stock-car-champion in the Deep South. Wolfe had a hard time going from his notes to a final piece, so he sent his rough sketches of Johnson and his milieu to his editor in New York. It started like this – “Ten o’clock Sunday morning in the hills of North Carolina. Cars, miles of cars, in every direction, millions of cars, pastel cars, aqua green, aqua blue, aqua beige, aqua buff, aqua dawn, aqua dusk, aqua aqua, aqua Malacca, Malacca lacquer, Cloud lavender, Assassin pink, Rake-a-cheek raspberry. Nude Strand coral, Honest Thrill orange, and Baby Fawn Lust cream-colored cars are all going to the stock-car races, and that old mothering North Carolina sun keeps exploding off the windshields. Mother dog!” – and got better from there. The editor changed next-to-nothing, and the stream-of-consciousness epic that ensued is now a New Journalism classic. While contributing editor Joe Flood may not be there yet, he is well on his way – and so here is his recent interview with Michael Lewis, in its totality, (basically) unedited.
Flood: The culture: when did it start to change?
Lewis: A lot of the books I write are set in culture shifts of one kind or another, or are actually trying to create those changes, which I think was the case with Moneyball. The new book is set at the back end of a shift that started right before I got to Wall Street in the 80s. So how is it that it changed when it’s these same old guys that are around already? Well there are a few answers—the obvious new ingredient was the computer. Computing technology changed what was possible on a trading desk for people like Louie Ranieri and John Meriwether—Gutfreud was not so much of an innovator, he just happened to be on top of a shop where the changes were happening, though the bond market was already intellectually open place and naturally suited for those changes. Along with that it’s really nice to have a Black-Scholes pricing model, the ability to price options—or the belief in the ability to price options—had a huge effect on the quantity of options in finance world. But you don’t get Black-Scholes, and certainly not in regular use on trading desk, without the computer. We take it for granted because it’s now the heart and soul of Wall Street, but in 1970 it wasn’t computers, it was a lot of guys with paper. But in 1985, when I showed up, the computer was there. And I was oddly on the edge of the business, with John Meriwether, responsible for explaining and creating a lot of these new derivatives….I had at my fingertips what was really a breathtaking technology, where you could just put in information on the option stuff and it would spit out a price. It made the financial world a lot more complicated, and became the reason that it was suddenly useful to have a PhD in mathematics from MIT. That’s completely new…it’s hard to believe what a putatively ‘sophisticated’ trading desk was like just a few years earlier, a typical trader didn’t even have a college degree.
Moneyball is a more visible shift in culture, but it’s the same thing. It’s replacing the wisdom of experience and intuitive judgment learned over a career with precise metrics only made possible by computer power, and different industries have assimilated that technology at different speeds.
Flood: Thoughts on the COCHRANE-KRUGMAN Debate?
Lewis: I think there’s an eternal dialog that will never end, between the value of intuitive judgment and supposedly ‘common’ sense, and the value of sophisticated analysis that in some cases defies common sense. It is true that when you start with an environment with essentially no statistical sense—one which has not assimilated models into their thinking, where essentially the thinking is that same as in 1920s—in that environment when you plop down this new way of thinking, this new way of thinking is very useful. Intuition can lead you astray when not checked by analysis, whether in baseball or on Wall Street. The example on Wall Street is the early 80s, when publicly traded options and futures existed for the first time—they started in around ‘79 or ‘78 in Chicago and it’s an interesting story in itself because Milton Freidman was involved—when you look at how things were priced. When the Japanese bond future is introduced there were huge arbitrage opportunities because people were trading futures like they weren’t related to anything else. When they liked the market they bought futures and when they didn’t the sold them. In that environment there were ridiculously fat opportunities because so few people understood the relationship between cash and futures, they didn’t get that with something like Japanese bond futures there’s very little risk, almost none in fact. The same was true in baseball in 2002. On-base-percentage was a huge, fat opportunity that intuition missed for a number of reasons (walks are boring, hitting was emphasized by coaches, etc.) when these analytic tools were first introduced. So then what happens? Well, the analytic becomes fetishized to the point where Black-Scholes and the various tinkerings with it were treated unquestionably, when in fact they were mispricing tail risk. Now an old fashioned trader running a firm would have shied away from that kind of catastrophic risk. He may not have understood it all, but he wouldn’t have taken on risk that would put firm out of business if things went bad. The idea that you can pin a number on tail risks leads to an environment where models and analysis are fetishized to the point where there’s no judgment check. There’s a dialog here that will never go away and you can’t ignore it. The big question in psychology and behavioral economics is where does intuition fail and where is it useful? Right now nobody is quite sure, and the answer is different in different industries, and people overshoot on both sides [of the issue]. There are different environments where you’re going to get further with statistical analysis, because what’s going on is measurable to varying degrees…What happened on Wall Street was sort of Moneyball gone wrong, people who were so sure of their analytics that it enabled them to not think about things, or to use it as a tool for justifying what they were doing. It’s just as easy to misuse statistics as to misuse intuition, and in a weird way it’s more dangerous because there are fewer people out there to challenge your statistics, whereas if you make a mistake of common sense or obvious logic, people are going to see it and say something.
When you use statistical arguments as an excuse for not thinking about problems or when it becomes camouflage for what you are doing it’s ironic, because presumably you’re only interested in statistical analysis because you have some kind of higher mental capacity or interest. This is the story of The Big Short. You see these people—the Morgan Stanley trading desk is one of the best examples—you see these people who think they know better…it’s not the Long Term Capital Management story because in a lot of ways they did know better. These were smart, thinking people, they just missed how the rest of the world was responding to what they were doing, making similar trades to them, and how they were exposed to other people. In the subprime mortgage market, the people running the [Morgan Stanley] desk weren’t even thinking, they were using the models blindly to take enormous risks w/o thinking for themselves. It wasn’t just that they were exposed to other people an human irrationality, they were thinking that AAA rated bonds backed by subprime mortgages were actually AAA…None of it is possible without the models….Obviously this is a subject very dear to my heart. I don’t know why, but this borderline between human beings making decisions based on raw data, and the way it eludes human experience.
Flood: What’s going on now?
Lewis: I think my sense is that we’re in this weird intermission in financial markets, where the bigger firms where the problem was centered are still being bailed out by the federal government, so they’re in a business where they don’t have to think very had to make money: get 0% interest from fed, plow the money back into securities and pocket the spread. But things are changing. The people running, say Morgan Stanley, when they saw what traders were doing, they only saw two pieces of data. One was the value-at-risk, and one was just the size of the buckets, how many AAA or AA bonds were on the balance sheet. Now they see more elaborately stress-tested positions, “If X happens in subprime mortgage default rates, this will be the effect on these positions…” They’re seeing better info, and they’re less trusting of the traders than before. So in that sense yes, things are changing. But what I don’t think is happening, and I could be wrong about this, is right now the market has not become simplified, and I don’t think that’s going to happen without regulation, it’ll take something big to simplify markets. The point of complexity in the end isn’t to better value risk, although that’s what they all tell you. The actual purpose is to actually make it complicated so customers don’t understand. All these institutions have been forced to become seemingly more transparent. There aren’t smoke filled room, everyone has glass-paneled rooms, but the response of the financial world to surface transparency is to make it more deeply opaque…Talk to anybody on a trading desk, so much of profit—in fixed income, especially—is from an asymmetry of understanding between customers and investment banks. An awful lot of fail switches come down ‘to this is our analysis, trust us…” This happened with subprime. Investment banks who design securities will not have an incentive to reduce the level of complexity until there is regulation.
Flood: So the other option besides government regulation is the market itself. At what point do customers say, “We’re tired of you screwing us”? Where is the customer revolt?
Lewis: This is the part I didn’t understand at Salomon Brothers. You would think that a firm that traded against its customers would be penalized, and instead the firm that trades against customers the most was rewarded. Goldman is legendary for its duplicity, but customers are afraid not to deal with Goldman. Why is that? There’s no great answer but I can remember being in a meeting with Tom Strauss, the President of Salomon Brothers, and we were with our biggest customers outside of the US and they were all mine, I’d established the relationship. Now the customer was a smart guy, and he said ‘You guys are always fucking us. Why are people putting up with this?’ And Strauss said, “We’ve found customers have very short memories.’ If you think about it this way, with a sufficient asymmetry of info—if the dealers know more than customers—the customers feel they need to have a relationship with dealers, even if it is info that nobody should have to have…It’s a weird dependency, this relationship, but it’s part of the culture of the financial business world. People don’t have careers in single institutions, nor do they expect to. If you ask people where they expect to be in five years, everyone assumes they’re going to be somewhere else. These institutional relationships are much more fluid and temporary. If I’m a salesman at Goldman who screwed the trader at a fund who’s a big customer, the problem is between [the salesman and trader], not the hedge fund and Goldman. And [the hedge fund trader] may want to work at Goldman down the road anyway, and if not, he wants to be in the golden graces of everyone at Goldman…It’s gotten to the point where if you’re a customer that is fucked by Goldman you’re embarrassed by it rather than angry, it’s sort of your fault because you didn’t know. This whole thing of buyer beware struck me as an odd notion in 1986, never mind today. In every industry—every decent industry—there’s some sense of obligation between the person selling and the person buying, but the bond market is interesting because there isn’t.
Flood: So regulation. Where are we at with that?
Lewis: The first thing that needs to happen is the very first thing that should have happened a year and a half, two years ago. When Bear went down, the Treasury should have instantly acquired powers of receivership. So if Lehman fails, it doesn’t fail the way it did. It goes into receivership by Treasury. Shareholders and bondholders are wiped out, but other creditors, like with credit default swaps, aren’t wiped out. The second thing that’s needed is that if you had to have reserve capital against credit default swaps, like you do with insurance, the market would be largely shut down because it wouldn’t be profitable.
Then if you want to fiddle with things more, remove implicit federal guarantees. Some institutions would fail right now; Citigroup would fail, I think, which they should. Then you attack the structure of the business. This should be the goal: that a firm cannot trade for its own account in securities it is selling to its customers. That’s but roughly what Volcker is trying to say. You have brokers and you have dealers. You don’t have broker-dealers. You have Schwab and Citadel, some give financial advice, others take proprietary risks, and they aren’t under the same roof, and then everything takes care of itself. Not completely, brokers have other problems, but if the two are not in the same place, it’s amazing how many problems you would get rid of. Even in my day it was shocking the extent to which my job as salesman was to get bond traders or derivatives traders out of their shitty positions and use the customers as the dumb money. It’s a perversion of how capital markets are supposed to work. If you just do what I said, the end result is a far less profitable financial system, one that no longer has the status it has now, one where the best students from MIT and Harvard aren’t flocking there. There is an obvious political fight that has to happen, a brutal one between financial services and the rest of the country. If we’re going to have a sane society this has to happen…it’s not a sign of a healthy society when the smartest people want to be in a middleman role.
In reference to Calvin Trilling’s article in NYT about how the biggest problem on Wall Street was that there were too many smart people: A lot of the most trenchant and useful observations come from people who are at arm’s length, if you’re inside a system, you buy why it is the way it is. Calvin Trilling can write that because he’s at a remove.
There is value in the financial system, but it’s essentially parasitic. It’s only profitable if there is something that is generating actual wealth to profit from. It’s necessarily secondary. If nothing else is going on in the world, you don’t have a financial system, you have a casino where everything is zero-sum. I don’t think there shouldn’t be a Wall Street, but it needs to be placed back in the box. In a functioning capital system, you only get rich when you’re contributing and increasing the size of the pie. The problem with the financial system is that you can get very rich shrinking the pie. In 2007, Wall Street paid out more than ever in its history, and they almost sunk the world. It’s not even benign, it’s toxic and they’re making money from this. The incentives need to be better aligned with the larger economy. It’s not happening right now, but I think reform of this magnitude always takes time. Crises are sharp and fast, but the political process is slow. Look back to the Depression, it was four years after the Crash before Glass-Steagall. Right now there’s a dawning realization of what needs to happen. A year ago Volker was laughed at, now he’s at the center of everything, and that’s a reflection of that change.
Flood: Is there a risk of unintended consequences here, as per Volcker fixing the money supply and freeing up interest rates, sort of accidentally setting the bond market aflame, in 1979?
Lewis: It would be hard to make it worse than it is now. If they did something as crude as banning credit default swaps, the world would be a better place. There’s a better solution than banning them, but the incentive system is so screwed up right now it would be very hard for regulation to make it worse. There is low hanging fruit out there. I agree that it’s possible down the road for regulation to be introduced that would make it worse, but I don’t hear anything being bandied about that is in danger of doing that. Most of what is being thrown about doesn’t go far enough, but it is moving in the right direction.
aiCIO Editorial Staff