Transcript
Eric Crittenden: Hold on a minute. This seems like it's a game changer, right? If you can get the returns to go higher and the drawdowns to get smaller, and the risk to get a lot smaller, isn't that the whole point of the industry? Isn't that what all these hundreds of thousands of people are spending billions of dollars a year trying to do?
Eric: I think it was my junior year, and it was the futures and options class, and the task at the time was to implement modern portfolio theory, but we had to write the code ourselves and I think it was Fortran back then. This would have been 1996-ish, something like that. The task was to go out and collect historical data and build what we consider to be an optimal portfolio, and then submit it and then justify, defend your portfolio choices. At the time I had a lot of friends that worked locally in Wichita Kansas in commodity trading and they had access to all kinds of terminals and data that a regular person doesn't have access to.
We had a little quantitative finance club on the university at the time and we shared data. I decided to collect all of the data, so I had this enormous database of hundreds and hundreds of different sectors and asset classes and strategies. And I thought at the time, "Well, this is what I'm supposed to be doing, expand the scope, do this thing right and try to learn something." I did the analysis and something interesting happened. The optimizer, which is trying to maximize your historical return, and minimize whatever your metric of risk is-- for most people it's standard deviation, other people drawdowns. There are many many different ways to estimate what risk is, I don't remember what it was back then.
This optimizer kept saying to put a big chunk, 30% to 50% of your money in this obscure thing, which I believe was called systematic global macro back then, which I wasn't that familiar with back then. The optimizer didn't really like anything else, it really liked that and so it was saying to put nearly half your money in this asset class. I did what I thought I was expected to do. I built the optimal portfolio, and it really looked great. It had a sharp ratio of about 1.0, it appeared to have a lot of capacity and it was very durable and stable through changing market conditions.
Remember, this was 1996 or 1997 and there'd been a lot of chaos. There was the crash of 1987, there was the S&L crisis, Orange County going bankrupt, the peso crisis, a lot of stuff. An interesting thing happened, though. When I submitted the assignment, the teacher flagged it and said, "I think your results are a little too good, I'd like to know what you did here." When I explained it to him, he said, "Oh, I see what happened here. Get rid of this systematic global macro, that's not a real asset class. Just use stocks bonds, real estate, small cap stocks, emerging markets things of that nature."
I said, "Okay." I did it and the returns came down, the volatility shot up and the drawdowns essentially tripled. I thought, "Well, that's no good. It's interesting," and I talked to him about it and he said, "No, that's fine. Turn it in and move on." But I wanted to talk about it some more, I was curious. I thought, "Well, hold on a minute, this seems like it's a game changer. If you can get the returns to go higher and the drawdowns to get smaller and the risk to get a lot smaller, isn't that the whole point of the industry? Isn't that what all these hundreds of thousands of people are spending billions of dollars a year trying to do?" No interest, crickets. Fellow students didn't care, teacher didn't care, but I never stopped caring.
From that point forward, I kept looking into this asset class. Some people call it managed futures, systematic global macro, rules based, global trend. There are various different names for it and there are different flavors, but the beta that comes from that industry back then appeared to be incredibly valuable. When added in with stocks, real estate and bonds it made a bigger difference than any of the other asset classes. That is still true today. In fact, the optimal portfolio when I rerun such an exercise comes up with the exact same numbers, and it shows the same level of benefit.
