Jim Simons (full length interview) - Numberphile
Should we start at the beginning?
Okay.
As a child, were you good at mathematics? Was mathematics a natural thing for you?
Which, yeah, it was very natural. I always liked it. I liked counting. I liked continually multiplying things by two.
Although by the time I got to 1024 or whatever it is, I had enough of it.
But I liked math. I discovered as a very young kid, maybe four, something called Zeno’s paradox. You ever hear of Zeno’s paradox? My father told me that the car could run out of gas, and I was disturbed by that notion. That never occurred to me. But then I thought, well, it shouldn’t run out. You could always use half of what it has, and then you could use half of that, and then half of that, and it could go on forever. So it would never run out.
Now, it didn’t occur to me, yes, but it wouldn’t get very far either. But the idea that in principle you didn’t have to run out of gas was kind of a profound thought for a very little boy.
Were you a talented student? Were you always getting really good grades?
Well, I was a pretty talented student. I knew I was very smart. Somehow, on the other hand, I was very careless. So I sometimes screwed up arithmetic tests just because I did it too fast or sloppily. But I liked everything about math. I liked everything about math; I loved learning the formulas for the volume of a sphere—four-thirds pi r cubed. I always thought that was a great formula.
When I got to high school, we started with plane geometry, proofs and theorems. That’s where I really got gripped. I really loved that. I loved working on some of the problems. Some were harder than others, and I liked doing that. I was never the fastest guy in the world, but I would plod through it with determination. I just liked it.
Did it feel like this was where your career was going to take you, or were you like another boy who dreamed of being a baseball player or something?
No, no, no, no. The only thing I thought about was I would be a mathematician, whatever exactly that meant. I didn’t know quite what it meant except that mathematics was the only subject I really liked. Science was okay, but it wasn’t very well taught where I went to high school, at least I don’t think so.
And history and English—I loved to read—but I wasn’t a good writer. I read a lot. When I went to MIT, I majored in mathematics, of course, and skipped the first year because I had some of those courses in high school. I even took a graduate course in my freshman year of second semester in math because it said no prerequisites required. Oh no prerequisites. I’ll take that course. I had a very hard time getting my arms around that course. I finished it okay, but I was very puzzled by some of the concepts.
However, that summer I got another book. I got a book on abstract algebra, and then in a week, everything came together for me. I understood. I was no longer puzzled. I understood why they did everything. Other parts of math got me stymied for a while, but typically it would take some time for me to sort of grasp what it was they were trying to do.
Then it would go very smoothly, so I graduated early from MIT in three years.
Did you imagine you would be a professor sitting in a university, or like what?
I didn’t have a grasp before. By the time that my freshman year was over, I kind of realized that’s what it would mean—that someone would pay you for thinking about mathematics and creating mathematics. What I thought earlier was I would do something that involved mathematics, but it was mathematics that I wanted to major in. I could see quickly that if you’re going to do that, you’re going to be a professor somewhere.
You also talked about this week of algebra where everything sort of fell into place. Is that a common thing? Is there like—is this a thing that always happens? There’s this moment where all the cogs move into place, and it’s like, oh, I get it now.
I think, yeah, there are moments like that. Sometimes you’re introduced to a new concept, and you wonder why—why is this concept interesting? And then you’re thinking, you see some examples, and they say, oh yeah, okay, this is really good. A lot in mathematics is making definitions. You’ll see things that really could come together under one aegis. You can define this set of notions and sometimes just a structure which follows all these notions.
That’s a great way that mathematics has advanced because when you put a lot of things together that have similar characteristics, then you can try to prove theorems about the general set of things rather than some particular example. That turns out to be a very powerful approach. So, making a good definition is a good thing.
They wanted me to go to Berkeley and get away from MIT—to meet some new faculty. Because I was quite close to the MIT faculty, and they thought—I think they didn’t want to get rid of me so much as they thought I was probably pretty good, so I should get exposed to a certain guy named Chern who was just coming to Berkeley that year.
I got this very nice fellowship, and I went there. I was very eager to work with Chern, except he wasn’t there. He celebrated his first year at Berkeley. He had just come to Berkeley, and he celebrated that first year by taking a sabbatical, so he wasn’t there.
So I worked with another guy, which was fine. By the time Chern came in the second year, I was already pretty far along with the thesis project. I was giving a seminar at the beginning of my second year at Berkeley, and in walked this tall Chinese guy, and I said to Kai next to me, “Who’s that?” I said, “That’s Chern.”
I didn’t know he was Chinese. I thought Chern was probably short for Chernowski or something, or he was probably some Polish guy who shortened his name to Chern. If it had been Chen or Chan, I would have known it was Chinese, but Chern with the “r”—I know.
But anyway, I met Chern then, and we became friends because I was much younger than he, but we became friends and later collaborators.
Can you give some idea about what your area of specialization is at this point? What are you zeroing in mathematically on?
Well, at that point and pretty much thereafter, the field that I worked in was called differential geometry. Differential geometry is geometry, but it’s the geometry of curved spaces. It’s not flat things in the plane. It’s any shape or form but typically with some distance function so that you know how far apart things were. The basic object to study is what’s called a manifold.
A manifold is a lot of pieces of space glued together. For example, the surface of a sphere is a manifold. Now, you can’t just take one little sheet of paper and make a sphere out of that, but you could glue a bunch of pieces of paper together and make a sphere. So a manifold is something that locally in your neighborhood looks like it’s just regular space. Maybe it’s curved a little bit, but globally it could be quite different.
The surface of a tire or a doughnut, for example, is fundamentally different from the surface of a sphere. You can’t deform one into the other. What really got me onto it is a certain theorem called Stokes’ theorem.
Now, Stokes’ theorem is the ultimate, I think, generalization of what was originally called the fundamental theorem of calculus. Now, I don’t know if you studied calculus, but the basic theorem is if you integrate the derivative of a function—so you take a function, you take its derivative, first derivative, and then you integrate that, let’s say from a to b, well, that gets you back the original function. The answer when you integrate the derivative is f of b minus f of a.
That’s the answer. If you integrate the derivative of a function, you’ll get back the value of the function. Well, that was the fundamental theorem of calculus. Presumably, Newton knew that.
Now, that had been generalized into higher dimensions, so there were other things you could integrate and differentiate and go from boundaries to interiors and so on. The general statement of that theorem was to me almost breathtaking. It was so beautiful, that theorem. And that’s a bit one of the basic theorems in differential geometry.
It’s about integrating something over the boundary, and that’s the same as integrating something else over the interior, something related to its derivative—in effect, it’s called its differential. So there’s a whole calculus of things called differential forms and integration and all that.
That takes place in manifolds in general, and anything that looks like pieces of space. So I love that theorem, and that’s what got me into differential geometry.
So was it a beauty or an elegance you saw, or was it a trophy and a prize you saw that could be chased?
No, it was just a beauty. It was just the beauty of it. I just liked it. I learned a fair amount of algebra, and that was fine, but it didn’t grab me. But the geometry did, so that’s what I specialized in.
When I got to Berkeley, I was fooling around and started. I’d made some observations, and I mentioned them to my thesis advisor—the guy I was going to work with. He said, “Oh, that’s interesting. That makes me think about such and such.”
There was a problem floating around that people had been trying and failing to solve, but I thought, oh, that’s very interesting. So I started working on that, and he advised me not to because so-and-so had tried and someone else had tried it, and those were big shots, and I was just a little shot. But I didn’t pay much attention to him, and I solved it.
I just worked through it and, you know, had a little help along the way from people, but basically, I drove that. I had done some other mathematics after my thesis work on something called minimal varieties, and it was kind of a fundamental paper. It took me six years to write that paper.
People, I think, thought, how could it take six years? But anyway, it did, and that turned out to be a very good paper. That was where I was coming from at roughly that time frame—or a year or two afterwards, when I was 30, I went to Stony Brook to be chair of their math department.
Stony Brook University was a very new university. They had a rather poor math department; in fact, it was very bad, but an excellent physics department. They offered me this job being chairman, and I thought that would be a lot of fun.
In that time frame, I had decided to try to learn something about an area of topology, which is related to geometry—it’s the kind of handmaidens of each other—called characteristic classes, whatever that might be. And I thought, I’m going to just learn these classes because I didn’t really know it, and it was important, and I wanted to learn about it.
So I kind of started from the beginning and worked my way up. I was trying to solve a problem that, as I later discovered, had intrigued many people, and it’s never been solved satisfactorily.
Since I say that, I didn’t solve it either, but in the course of trying to do that, certain terms came up—certain functions of a sort—that began to look just very interesting to me. I saw, well, they were a pesky term that I couldn’t get rid of. I needed to get rid of that term in order to make this formula that I wanted, but the term seemed to have a life of its own.
One thing led to another. I defined a certain invariant of a three-dimensional manifold and used that to prove some what’s called immersion theorems—whatever those are. This was a very handy thing. I showed it to Chern and said, “Look at these results in three dimensions.”
He looked at it and said, “Oh, well, we could do this in all the dimensions,” because it was an area that he really knew about—not the results that I’d gotten, but the general area. So we worked together, and we came up with these results, this whole structure.
In fact, there’s the slides of the presentation that Chern made at the International Congress in the early ’70s. It was very nice geometry. I pushed on with it, and we defined some things called differential characters, which was another chapter with working with a guy named Chiger.
But the Chern-Simons invariants—about ten years later, the physicists got a hold of it, and it seemed to be very good for what ailed them, or whatever might have ailed them. And so, it wasn’t just string theory as I subsequently developed. It was kind of all areas of physics, including condensed matter theory.
Even some astronomers seemed to want to look at those terms. That’s really what’s great about basic science—in this case, mathematics. I mean, I didn’t know any physics. It didn’t occur to me that this material that Chern and I had developed would find use somewhere else altogether.
This happens in basic science all the time—that one guy’s discovery leads to someone else’s invention and leads to another guy’s machine or whatever it is. You know, basic science is at the—you know, it’s the seed corn of our knowledge of the world.
What did it feel like at the time you were coming up with it? Did it feel obscure and just like a little diversion, and then ten years later it felt special?
But it never felt like a little diversion. I really liked the results. I loved the subject, but I liked it for itself. I wasn’t thinking of applications.
You know, I mentioned to C.N. Yang, who was at Stony Brook at the time, who was a Nobel Prize-winning physicist of great renown, that I’ve got these things. Maybe they’d be useful in physics because I knew the physicists were looking in the same places, using certain mathematical structures. But he didn’t bite, and I didn’t know how it would be useful, so I just left it alone.
But I was very pleased with the math, and then it led to another set of definitions and some math that I did with this guy Jeff Chiger, which started as sort of a mini-field—although I didn’t realize it at the time—called differential cohomology.
There’s a fair amount of people working in that area. What does that make you feel? Does it become like vindication, or what I did was more useful than I thought? Or like what do you think when it gets used ten years later, or by then are you divorced from it, and you’re like, do what you want with it?
Oh, it’s always made me feel very good. Of course, naturally, you like to think that something you did had far-reaching ramifications. You know, it was nice, but it wasn’t—I didn’t go to bed dreaming about, ah, now I’ve revolutionized physics. First of all, I didn’t revolutionize physics, but I did some stuff there, and sure, it made me feel good.
Actually, in the middle of my mathematics career, which ended when I was about 37 or 38, I spent four years at a place called the Institute for Defense Analyses down in Princeton, which was a super-secret government-based national security agency—based a place for code cracking, trying to break the enemies—whoever it was—Russia, I guess, code machines and cipher machines.
I spent four years there in Princeton. During that time, I was working on this—remember I said it took me six years to work through all this stuff in minimal varieties? I did a lot of it while I was there, but I also learned about computers and algorithms in the code cracking field.
I was no good as a programmer—terrible—but I was pretty good at coming up with algorithms and trying to, you know, and I found that very exciting to think, oh, this might help crack this code. Here’s an algorithm that could work; someone else would program it up, then it would run on the computer, and maybe it would work, and maybe it wouldn’t.
I did one thing there that was quite good, but I can’t tell you what it was—it’s all classified. So I had a good career there, both doing mathematics and learning about the fun of computer modeling.
Was that specialization that you’d had during your PhD, the manifolds and the topology, related to the algorithms—were the two or three?
Oh, completely! So why were you drafted in for that in the first place?
Then they paid money! I was getting kind of bored simply being an academic, and also this work—this six-year project—was maybe making me feel, gosh, other guys are publishing papers, and I’m not. But I wasn’t thrilled with being an academic.
This place hired mathematicians—a handful. You could do mathematics half your time, and the other half of your time you were supposed to work on their stuff. There was no teaching, so it was like, oh, I could do as much mathematics as I was doing anyway.
And they paid better. I thought this would be interesting and a nice change, and it was! I enjoyed it. I did a lot of good math or that I thought was good and it did good work for them. But that ended after four years because I got fired.
And well you can’t just leave it at that you got fired. Okay, why did you get fired? Well, I got fired because this was in the middle of the Vietnam War, which I didn’t like. The war, not that we were doing anything to support the war in this particular organization, but nonetheless I just didn’t like it and the head of the organization, he was down in Washington. He was a big shot named Maxwell Taylor, General Maxwell Taylor. He wrote an article in the New York Times magazine, the section of a Sunday magazine, a cover story, how we’re really winning the war in Vietnam. We just have to stay the course and it’s all going to be great, and I thought that’s a lot of baloney.
So I wrote a letter to the Times saying not everyone who works for General Taylor supports his views and in my opinion blah blah blah. It was a good letter and they published it right away, of course, because it was unusual that someone would write such a letter in his situation. I didn’t hear a peep. No one said anything at the company. They didn’t say you shouldn’t have done that, but I was clearly on the watch list. When you wrote it did you know you were doing something reckless? No, I didn’t really think it was reckless because it was my opinion and it really didn’t have anything to do with my work.
But then about three or four months later, a guy from Newsweek magazine came to see me and he said he’s doing an article about people who work for the Defense Department who are opposed to the war. He says I have a lot of trouble finding such people but could I interview you? Well, I was 29 years old; no one had ever asked to interview me before. It sounded like hey, maybe I’ll be interviewed. Okay you can interview me.
So we asked this and that, but he really wanted to know so what is your policy? So I sort of made up a policy although it was pretty much true. I said well you know here at IDA you have to spend at least 50 percent of your time on their stuff. In those days it was secret that it was even codes and ciphers so I just said their stuff. But you could spend 50 percent of your time in mathematics and so my algorithm now is until the Vietnam War is over I will spend all my time on mathematics and then after it’s over I’ll spend all my time on their stuff until the two things match up again.
And that was kind of largely true. I was doing mostly mathematics but I was doing a little of their stuff. So then it occurred to me to tell my local boss that I gave this interview and he said, well what’d you tell him? I said, well I told him what I just said to you. He said, you did? He says, oh I better call Taylor, his boss.
And he called Taylor and I came back into his office and he said, you’re fired. He said I’m fired. Yes, you’re fired. Taylor fired you. Taylor told me to fire you. So, well I was fired. I said you know I don’t know how you can fire me my title is permanent member, which was I started as a temporary member and then I became a permanent member. The boss was very funny. He said, well here’s the difference between a temporary member and a permanent member: a temporary member has a has a contract; a permanent member doesn’t. I had no contract so I was out of there.
And it was amusing. I mean I wasn’t worried; I had solved this big math problem that I told you about the six-year thing. I knew I was going to get a job very easily so I wasn’t really worried about that.
You’ve honed your skills with algorithms and you’ve learned a lot about computers. So I’m seeing the writing on the wall here for where things go next. How do we progress to yeah so at a certain point, now I wasn’t then I went to Stony Brook after this. It was chair and did this work with Churn and did some work with Chiger.
And we got stuck. We were trying to prove something, a mathematical thing and it was very frustrating. We worked on it for about two years; we got nowhere. I mean it’s okay to work for a long time if you feel you’re getting somewhere but we got absolutely nowhere on this problem.
And my father had made a little bit of money and I have had the opportunity to try investing it and that was interesting. I thought, you know, I’m going to try another career altogether. And so I went into the money management business so to speak.
So you started with some of your dad’s money and that got you a taste of an interest in it. Some family money and then some other people put up some money and I did that. No models for the first two years. So what were you doing then? You were just using cunning and you know just no, like normal people do.
I brought in a couple of people to work with me and we were extremely successful. I think it was just plain good luck but nonetheless we were very successful. But I could see that this was a very gut-wrenching business. You know you come in one morning you think you’re a genius, the markets are for you. We were trading currencies and commodities and financial instruments and so on, not stocks but those kinds of things.
And the next morning you come and you feel like a jerk, the market’s against you. It was very gut-wrenching and in looking at the patterns of prices, I could see that there was something we could study here, that there may be some ways to predict prices mathematically or statistically. And I started working on that and then brought in some other people and gradually built models.
And the models got better and better and finally the models replaced the fundamental stuff. So it took a while. I would have thought with your background as a mathematician this would have almost occurred to you immediately. Like you would have straight away said this. What was the two-year delay?
Well, two things. I saw it pretty early and I brought in a guy who was a wonderful guy also from the code cracking place. He was, I thought together we’d start building models. That was fairly early, but it wasn’t right away. But he got more interested in the fundamental stuff and he says the models aren’t going to be very strong and so on and so forth.
So we didn’t get very far, but I knew there were models to be made. Then I brought in another mathematician and a couple more and a better computer guy and then we started making models which really worked.
But you know, there’s something called the efficient market theory which says that there’s nothing in the data—let’s say price data—which will indicate anything about the future because the price is sort of always right. The price is always right in some sense, but that’s just not true. So there are anomalies in the data even in the price history data.
For one thing, commodities especially used to trend—not dramatically trend but trend, so if you could get the trend right you’d bet on the trend and you’d make money more often than you wouldn’t, whether it was going down or going up. That was an anomaly in the data. But gradually we found more and more anomalies. None of them is so overwhelming that you’re going to clean up on a particular anomaly because if they were, other people would have seen them.
So they have to be subtle things and you put together a collection of these subtle anomalies and you begin to get something that will predict pretty well.
How elaborate are these things? Because in my head I’m imagining you know some equation like Pythagorean equation; you put a few numbers in and something spits out. But are these giant beasts of equations and algorithms or are they simple things?
Well, the system as it is today is extraordinarily elaborate, but it’s not a whole lot of quite—you know, it’s what’s called machine learning. So you find things that are predictive. You might guess oh such and such should be predictive, might be predictive, and you test it out in the computer.
And maybe it isn’t, maybe it isn’t; you test it out on long-term historical data and price data and other things. And then you add to the system if it works, and if it doesn’t you throw it out.
So there aren’t elaborate equations, at least not for the prediction part, but the prediction part is not the only part. You have to know what your costs are when you trade. You’re going to move the market when you trade.
Now the average person will make a buy of 200 shares of something and he’s not going to move the market at all because it’s too small. But if you want to buy 200,000 shares, you’re going to push the price. How much are you going to push the price? How are you going to know, are you going to push it so far that you can’t make any money because you’ve distorted things so much?
So you have to understand costs and that’s something that’s important. And then you have to understand how to minimize the volatility of the whole assembly of positions that you have.
So you have to do that, that last part takes some fairly sophisticated applied mathematics. Not earth-shattering, but fairly sophisticated. What discipline of mathematics or disciplines is it? Multi-disciplinary or are we talking mostly statistics?
It’s mostly statistics and some probability theory, but I can’t get into you know what things we do use and what things we don’t use. We reach for different things that might come, that might be effective. So we’re very universal. We don’t have any but it’s a big computer model for one thing.
There is a capacity to the major model. It can manage a certain amount of money, which is rather large, but it can’t manage an enormous amount of money because you’re going to end up pushing the market around too much. So it was kind of a sweet spot as to how much it’s reasonable to manage.
Therefore, it would never grow into some behemoth which would take everybody out and you’d be the only player. I mean, well of course you were the only player that’d be known to play against. There are limitations, at least the way we see it, but we keep improving it.
We have about a hundred PhDs working for the firm. That’s what I mean. I mean how did you get to that point? Did you start to think we need this, we need that? What did we just hire? Smart people, my algorithm has always been you get smart people together.
You give them a lot of freedom, create an atmosphere where everyone talks to everyone else. They’re not hiding in a corner with their own little thing; they talk to everybody else. And you provide the best infrastructure, the best computers and so on that people can work with.
And make everyone partners. So that was the model that we used at Renaissance. We would bring in smart folks and they didn’t know anything about finance, but they learned.
What was your employment criteria then if they knew nothing about finance? What were you looking for? Someone with a PhD in physics who’d had five years out and had written a few good papers and was obviously a smart guy or in astronomy or in mathematics or in statistics.
Someone who had done science and done it well and was interested in applying his mind or her mind, although it was mostly his, to modeling markets and making money. But it’s a very good spirit now. I’ve been six years away from the company so I’m not running it anymore, but I’m the chair of the board and I go to a monthly board meeting.
I think the morale is very good, the spirit is good. And it’s really a very good way to work scientifically. It’s a big collaborative effort and everyone is happy to see someone else come up with something good because the first person is going to share in that, because everyone’s shared in the profits.
So, okay, you might wish it was you to show how smart you were, but nonetheless good he did it; I’m going to make money from it. I would imagine a lot of people want to be financially successful. Most people want that, of course I suppose, and lots of people are good at mathematics and know a lot about computers like you know at your level I would imagine.
Why did you do it? Why didn’t someone else do it? I don’t know. Well, first of all some other people have done it. I think that we’re, our firm is better, but nonetheless I’m pretty sure of that. But nonetheless other people have done some very good modeling and so we’re not alone.
But it’s not easy to do and there’s a big barrier to entry. For example, huge data sets that we’ve collected over the years, programs that we’ve written to make it really easy to test hypotheses and so on. The infrastructure is exceptionally good, so everything is tuned right.
It took years to learn how to do that. People don’t leave our firm or if they do they believe just to do something else altogether. Everyone has signed a forever non-disclosure agreement and so on because we’re very secretive about what we do because you can’t patent that stuff or copyright it.
Because then everyone would see what it is and someone will just work around it and say oh that patent’s no good; look I’ve made this twist and that tweak and it’s different. So what you have is your intellectual property and you have to keep that to yourself.
So yes, we were very successful and continue to be. As I say there are others, but very few investment operations are a hundred percent because this is a hundred percent model driven. It’s not ninety percent or eighty percent; some people have models and for advice.
So what does the model say? Oh yeah let’s go. Oh I don’t forget that, I don’t want to pay attention to that. But Renaissance is 100 percent model driven. No trade is ever made because someone walks into the trading room and says hey let’s buy IBM; it’s a sure winner or anything like that.
You know we got too much Google; we got to show it. Nobody does that. It may be that we had too much Google, but nonetheless he might have been right. But it’s just what the model says and that religious sticking to the model is the only way you can run such a business because you cannot simulate that guy who walked in and said hey let’s—Google’s too high, let’s sell it.
How can you simulate that? You don’t know what might have happened. But you can simulate. You can come up with a model or a new predictor and you can simulate it in the past and see how did it do.
So you have to stick to it. I know you guys made the model so you do have ownership of it and feel proud of it. But is it hard to follow the model religiously? Like is it hard for your ego to think all the successes because of the computer? Like I just sat there and watched? No, the computer is just a tool that we used to, I mean it’s— a good cabinet maker doesn’t say it’s all because of my wonderful chisel.
You know, you may have great film equipment but that’s not why you’re a success at doing what you’re doing. You’re working with good equipment, but another guy would make a mess of it with the same equipment.
So no, we don’t feel, oh the computer is doing everything. The computer does what you tell it to do. Mathematics is very collaborative, and things that are learned are built upon by sharing. Are you learning things behind the locked doors at Renaissance that could help mathematics but you’re not sharing because you want to keep them a secret to help the business?
No, there’s nothing that we’re learning behind locked doors in Renaissance that would help the general field of mathematics or the general field of science as far as I can tell. Do you think there’s something about your personality that made you able to do it or is it just luck?
Well, I think a lot of it’s luck. I probably have a good personality for running a group of people, but there are other people who maybe even have a better personality for doing that. We underestimate the role of luck.
It’s typical that if someone fails at something he’ll say I had bad luck and if he makes a success, he’ll say I was a smart guy. And oh I was just lucky. People don’t usually say that. If they make a big success, and of course obviously not everyone can make a big success, but I think luck played a role. I was in the right place at the right time.
But I really think what I’m good at is getting good people to work together. So it was more your managerial ability than your… Mathematical genius that resulted in it definitely wasn’t my mathematical genius. I think I’m a pretty smart guy; I can understand this stuff. But I wasn’t, and I did come up with a few predictive algorithms, so that was fine. Many people did.
I knew I wasn’t going to come up with all of them, so that’s why we have all these folks. It wasn’t my mathematical genius, but I think people respected me because I was a good scientist, so they said, “Well, he may act stupid, but we know he’s really smart.”
Given that you will put some of it down to luck, what are you more proud of: all of this in the business or the mathematics from that first half of your career? To the extent that I’m proud, I think I’m proud of both. I’ve done some mathematics, and some of it’s had a positive effect, and I guess I’m proud of that. I’ve built a nice business, and I’m proud of that. I don’t say I’m prouder of one than the other.
For the last number of years, I’ve been working with my wife on this foundation, which she started actually in ‘94 with my money. But nonetheless, she started the foundation, and then I got more and more involved with it as time went on. That’s my main thing, and I’m pretty proud of the foundation.
Let me focus you more on the mathematics versus the business, then. Would you trade any or all of your business success for being the man who cracked the Riemann hypothesis or something like that? No, that’s a good question. Would I trade that? Well, I probably would trade some of it. I mean, for the Riemann hypothesis, it would certainly be a thrill to solve it. I’m pleased mostly with the way my career has gone.
So, would I trade part of it for something? I don’t know. I’ve never looked back and said, “I wish,” at least in business, “I wish I hadn’t done that,” or “I wish I had done this and not that.” Whatever it is, I’ve never looked back that way. I guess the thing I’m getting at is: what do you define yourself as—a businessman or a mathematician?
Well, I would not be very good at running an ordinary business. If I had to run a big manufacturing business or something like that, I don’t think I’d be very good at it. There’s a level of detail that I would find tedious. I’m not the best organized person in the world, and frankly, I would be bored.
I’m not your typical businessman. The kind of business that I did run was very natural to me. So, in that sense, I was a businessman. I like business. I always say what’s the important mathematics skill in business is subtraction. You have to understand when your revenues are going to exceed your costs.
Some people just look at revenues: “Hey, we’re growing!” Yeah, but you’re losing money every day. “Well, that’ll take care of itself after a while.” I don’t think I’d be very good. I once had to run a small business in the computer area for about six months. It was a business that we had started or invested in, and it wasn’t going well. We had to get rid of the head and find someone new, and I found myself running it for about three to six months, commuting to Philadelphia.
I came to the conclusion that I was doing a better job than the guy who was there before, but I wasn’t doing a very good job. I found myself in business meetings with a distributor from St. Louis, and I remember thinking, “What am I doing spending my time with some distributor from St. Louis? This is not the way I want to run my life.” Finally, we got someone in who really knew how to run a business, and then that was his kind of a business.
The foundation is focused on the support of scientific research, primarily basic science but not completely, because we have a large autism project which involves a lot of basic science. Treatments are a goal, but the rest is support of mathematics, physics, computer science, biology, neuroscience, genetics—we support basic science, and that is what we like to do.
There’s a certain amount of outreach; we have a Math for America program. We spend maybe 10 or 15 percent on outreach and education, but 80 or 85 percent is support of basic science. The nature of basic mathematical research is that you can’t really know where it’s going to go.
What are you doing? Are you throwing money at a wall and seeing what sticks, or how targeted are you? To some extent, yes. We have investigator grants that give outstanding mathematicians, physicists, and computer scientists 10-year support. They can use that money for a postdoc or something like that to help their work along. We don’t know what work it’s going to be. We know what the guy has done, but it’s very competitive. These are very competitive grants, and they’re just the best people.
We also have collaborative projects that we support. There is a goal, but it can be a rather vague goal. For example, one of our goals is the origins of life. We’d like to know how we all got here. How did life originate? What was the path to life? There are about 40 or 50 people working on that, and it’s a very exciting project. We’re making some progress, so there’s a goal. But do I think that in the next five years, we’re going to know everything about the origins? No, I don’t think so. But we’ll learn more about this path that started with primitive organic molecules and ended up with something like this interview.
There is a goal, but it’s a relatively big goal. This model of philanthropy—supporting basic research and not knowing where it’s going to end up—almost seems the opposite of what you were talking about with your business. Totally. You were talking about revenues and costs and making sure.
Yes and no. When we would bring smart guys into Renaissance, we didn’t know what they were going to do. We hoped that they’d come up with some good ideas to make the firm better, but I didn’t say, “Oh, this guy is going to discover this thing.” I don’t know what he’s going to discover.
Of course, it’s very focused on a narrow set of outcomes, whereas the other kind of basic science—anything goes—is very different from that. What’s your attitude to risk? Because you’ve almost created a business that’s based on removing risk, reducing it. And yet you took a risk by leaving tenure to start a business. I mean, are you a risk taker, or are you a risk minimizer?
You probably don’t need to worry about it. Well, I suppose as we get older, we get a little less risk-averse. I don’t know. I’ve always been something of a risk taker. I’ve had good taste, especially in people, and I’ve pretty much always partnered up with people, so that cuts the risk.
But I did, when I started the company, finance it myself, and I didn’t know whether it would succeed or not. I was pretty sure it would, but I didn’t know. You put a lot of money into mathematics, so you’ve got some right to comment on it. How are you feeling about it?
Oh, I think mathematics is really going quite well worldwide. The research is a bit… a lot of new ideas are coming up, new fields are flourishing. It feels like a pretty healthy enterprise to me. What’s not healthy is the state of mathematics education in our country. That’s very unhealthy for young people.
That’s why we started this thing called Math for America. We don’t have enough mathematics teachers who know the subject, and even in science, that’s for a simple reason. Thirty to forty years ago, if you knew some mathematics enough to teach in high school, there weren’t a million other things you could do with that knowledge. You could become a professor, but let’s suppose you’re not quite at that level. Being a math teacher was a nice job.
But today, if you know that much mathematics, you could get a job at Google, you could get a job at IBM, you could get a job at Goldman Sachs. There are plenty of opportunities that are going to pay more than being a high school teacher. There weren’t so many when I was going to high school.
The quality of high school teachers in math has declined simply because if you know enough to teach in high school, you know enough to work for Google. They’re not going to pay that much in high school. So how do we redress that as a country?
A person works for a combination of financial reward and respect. Right? A guy becomes a Supreme Court justice. He’s not doing it because he’s going to make a fortune. He’ll be well-paid, I suppose, but everyone says it’s a big deal. You have a lot of respect, and you respect yourself, presumably.
So you can’t pay high school teachers of math as much as they would get at Google, but you can give them a bump. We give people fifteen thousand dollars a year more than their regular salary, but we also create a community of math and science teachers. They love it, and it makes them feel important, and they are important.
They interact with each other; they are not forlorn, stuck in some high school with no one to talk to. We’ve created a community with a great esprit de corps. These people don’t walk out of the field, but turnover is very low, whereas ordinarily, the turnover is quite high.
In Finland, for example, which has very good mathematics teaching, those are real professionals, and they have a lot of status—societal status. A teacher in Finland, here in the United States, teachers don’t have such good societal status. And of course, the whole thing of getting rid of bad teachers—there’s been some bad teachers, but nothing is said about “Let’s reward and recognize the really good ones.”
If you’re running a business, yes, you want to get rid of the deadwood. If there are people who just aren’t doing the job, but more importantly, you want to recognize the people who are doing a good job and reward them, extol them, make them feel good in one way or another. We don’t do that at all with teachers; we’re just bashing them over and over again.
Why has that happened in the United States but not in Finland? Well, it hasn’t happened in most European countries, either. For one thing, we have teachers’ unions, but that shouldn’t be the reason, because Finland probably has unions too. I don’t know.
There’s been a shift that started about ten years ago where we have to measure these teachers by outcomes. Alright, fair enough. What outcome? We’re going to give their students standardized tests. If they have value added and so on, well, then they’re good. If they don’t have value, well, it turns out judging people on standardized student tests is a disaster.
It’s a very weak statistic; it’s not very highly correlated with how they’re going to do the next year by the same measure. Somehow, the theory is we have to measure these people. If they’re not doing a good job, we have to get rid of them somehow, or we have to get these people out of the schools.
There’s an old saying in the British Navy: “Floggings will continue until morale improves.” That’s kind of what this is; we’re just beating these people up. How did it start? I don’t know. Someone got the idea that we can measure output, but we can’t. Or at least we’ve imposed a system that stinks for measuring output.
The longer we stick with this system of rating teachers on performance of their students on standardized tests, the worse it’s going to get. The solution sounds to me like just respect them a bit more and make them feel good about themselves. Yes, and recognize and reward the very best ones, and that encourages people to say, “Hey, I could rise up and get one of those, become a master teacher,” or whatever.
Someone could even come into the field and say, “Hey, my cousin is a master teacher in New York; he’s having a good life.” We’re doing it in New York pretty successfully, but it needs to be all around the country, and it needs to be bigger than it is in New York.
You seem to be a believer in markets. Is this not something that would correct itself over time? If the teaching drops and Google can’t get the mathematicians, they have to improve the team. Will this not take care of itself?
Here’s how we’re taking care of it now. We’re importing people on H-1 visas, and we’re bringing in people from other countries who do have this knowledge. All the technology companies only want more and more H-1 visas because they can’t get enough home-grown talent.
That’s okay for a while, but about 12 or 13 years ago, for example, 80 percent of the graduates from the Indian Institute of Technology left India and would come to Europe and the United States to get jobs. Now it’s 20 percent because India is doing much better. There are plenty of technical jobs in India.
I can stay home; I don’t have to go to England or the United States to work. There are people from other countries who are coming, and we’re taking advantage of that, but the day will come when we just won’t have enough people here of our own folks who know enough to do this work. Basic research is, as I’ve said, the sort of wellspring of human knowledge about our world.
However, the federal funding for basic research has become restricted. For one thing, federal funding for science has gone down anyway, overall. Second of all, there’s an increasing tendency for these agencies to fund something called translational research rather than basic research. Translational means, “Okay, you’re working on cancer; great! We’re going to have a cure for this in three years.”
Okay, fine, we’ll give you money, but if you’re working on how the basic cell is working, well, you know, that application’s too far away. There’s less funding; Congress pushes the NIH and the NSF to say, “We want to see some return on this money; we want to see results.”
So they’re more conservative; wild ideas less often get funded because the government’s not doing a good job at supporting basic science. There’s a role for philanthropy—an increasingly important role for philanthropy.
Do you give your money to basic research because you feel somehow indebted to it for your own success, or do you do it just out of a belief? Or do you feel like you’re giving something back to what gave something to you? That’s an interesting question. I do it because it feels good. I like science; I like to see it flourish. I like to be around scientists; I like to learn new things. My wife feels the same way; she loves science.
We’re very happy to support this. Do I feel I’m giving back? Not especially. You know, I could give back in a lot of ways. There are a lot of things I could do besides support science.
Do you have a favorite number? Seven. Next question. Do you have a favorite mathematician? Well, Archimedes and Euler are my current favorites. Maybe you met somebody—I’m very impressed with those two guys.
Thank you so much for your time. Alright, well, this was kind of fun.