'My Life as a Quant: Reflections on Physics & Finance'(2)
Z: Two decades ago was a boom time for physicists to pursue a career in finance. Do you think today physics can still contribute to quantitative finance as was presumed back then?
D: Yes, but in a limited way. Finance is not physics, and markets are not physical systems. They keep changing as new experience accumulates, unlike physical systems. The biggest contribution physics can make to finance right now is the example that it sets of the very close interaction between experimental observation and theoretical insight, which play leapfrog with each other. The coupling between observation and theory in finance is much much weaker, unfortunately.
Z: You are now the director of financial engineering program in Columbia. In your experience, what kind of students will have the best performance in this program? Directly from undergraduate education? Work a couple of year and then go back to school? Or some math, physics or engineering Ph.D. students?
D: Best is a good solid education in the sciences or mathematics, a few years of experience with real markets, and then a return to school. But the best is often the enemy of the good, and many other choices are good enough, and will work. There isn’t one path that is right for everyone, and temperament matters a great deal.
Z: Let me keep pushing on the question. What kind of qualification will be most helpful for a student now to find a job in financial industry? Once a college student asked a senior partner what kind of courses did he need to take in order to better prepare a finance career. The partner smiled, it would be best if you can take a psychology course because remember, we are a people business. Do you agree?
D: No, I don’t. Psychology is overblown in explaining people’s behavior. Best, as I said, is a good understanding of modeling and its techniques in the natural sciences, a good knowledge of the business world, and a great deal of common sense, because it is a people business.
Z: OK. Let’s get into the finance topic. You once mentioned there is a massive language gap between quants and traders. Could you please be more specific?
D: Traders are the experimentalists – they have to deal with the world as it is and react to it over a short period. Quants are the theorists – they have the leisure to think more deeply, but they have to, because they are a step or two further from the real world. They therefore talk different languages, of necessity. Traders think about what will hurt them in the next instant, about risk. Quants tend to think more about value. Both are important of course.
Z: You consider the Black-Scholes formula and Merton’s continuous-time finance are strokes of genius. They’re like great discoveries in Mathematics. Now with the time passing by, more and more innovation happened in finance. Do you think they are still shining today? Do you think there will another revolution in finance just as big as Black-scholes model?
D: It’s hard to imagine another one that great. Black-Scholes and Merton are not perfect, and real markets are more complicated and violate many of their assumptions. But the insights of BSM, that you can replicate complex securities out of simple ones, and hence value them, has been the great driver of innovation of the past thirty years, and I’m probably short-sighted but I can’t imagine anything comparable. In any event, progress has diminished since then, though the applications of BSM have grown dramatically. One area crying out for great ideas is a better model of how simple assets behave, stocks, bonds, etc. Most of finance is built on geometric Brownian motion, and it’s not correct, and we need something better and reliable.
Z: The graveyard of financial services is littered with the tiny corpses of corporations who thought that what the world needed was a better model they could charge for. What do you mean by saying that?
