He walked quietly into the lecture room, tall, thin and pale. His suit hung loosely off his frame. He spoke softly: “This is 15.439, Problems in Finance. There is a list of 50 problems for this course. If we do all the problems, we will average 2.27 problems per class. The materials are in Graphic Arts. I see no reason why we shouldn’t begin. Problem one, Richard Gluck.”
Summer break had just ended. It was the first day of class. I hadn’t looked at the question or done the reading, much less bought the two foot high stack of articles from Graphic Arts. And this was Fischer freaking Black, co-creator of the Black-Scholes model, who had just picked me out of the sixty five or so people in that room to answer the first question in his class.
I scanned the question, something about the Capital Asset Pricing Model. One detail that sticks was a sub-question asking whether interest rates could go negative.* I answered as best I could. My classmate Steve, sitting next to me, raised his hand and backed me up. (I’m grateful to this day.) The arguments raged back and forth. I was losing. Fischer called for a vote. Nine for. He didn’t bother counting the nos. He just wrote “lots” on the board. Fischer allowed for a bit more debate. Then he summed up the problem and gave his view. Finally, he crossed off the 9, putting a 10 in its place. I’ve never been so relieved in my life.
Many, nay most, of the finance courses I took at MIT were technical in nature. The tests and the problem sets had definitive answers, no matter how labyrinth the paths to those answers. Problems in Finance was different. Yes, we read plenty of technical articles, but the nub of the course were those 50 questions whose answers required wrestling with ideas and applying them to the messiness of the real world. Fischer was teaching us how to think; he was doing it like a Zen master, his fifty questions like fifty koans to be debated by us, his disciples, twice a week for a semester. He’d often sum up a question with an aphorism. “Finance is a language” was one. It encapsulates what Problems in Finance was about.
George Box wrote: “The most that can be expected from any model is that it can supply a useful approximation to reality: All models are wrong; some models are useful”. It was in 15.439 that I came to understand that models and financial theory were languages, useful for framing real world problems, but not strictly true. That insight has served me well throughout my career as an analyst and investor. Fischer freed me to think about what I was observing, to apply models as needed, but to understand that there are limitations. It was a useful mindset to develop because financial markets generate enormous amounts of data, but very little signal. In such an environment, models are not beacons, they are candles. It’s important to understand that.
Fischer used to stop during class, pull a yellow post-it pad from his pocket and jot down notes. We thought, when he did it as you were talking, this can’t be good. There goes my grade. That wasn’t it at all. I was lucky enough to join him at a couple small group meals. He did it at lunch. He did it at dinner. He was constantly writing down ideas as he heard them or they occurred to him. I have always been intellectually curious, but this was taking it to a new level. When I look back, the depth of Fischer’s intellectual curiosity always amazes me. We could all benefit from a dose of it.