Enthralling video. I am going to watch it again. The contrasts between what the experienced trader was saying and what the academics were doing was great. LTCM doubled their margin because they were not as profitable one year, not a good thing to do, not even a rational thing to do. Of course it's easy for me or anybody to be a Monday morning quarterback.
Thanks for the link. It is interesting that when I first saw it in '00, I was doing off-floor analytics, and I have never seen an exchange floor, seriously. I was one of those quants who believed that the models will eventually be proven to be right, almost without exceptions. Then I went to Chicago, and the exchanges (CME, CBOT, CBOE, etc) hit me like a pile of bricks. I think I must have put on a trading jacket, and went to the CME SP pits, and CBOE's SPX pits and just stood there watched the action for about a week straight. Honestly, I think I learned more about the philosophy of trading in that week than a full year of studying graduate level finance. If anything, I realized that my trades are not trading against the "market" per se, they are trading against the crowd. If anything, this piece just made me realize my respect for people like Leo Melamed (not really Ben Schwartz, I knew him, long story). From the show, I think I picked up on what day he was trading the yield curve, and the little flash of SPM9 limit book. But his sayings are right on the mark. I love this quote: "The fear in the eyes of the traders around you, that too is information"
Interesting video. Thanks for posting it, Equalizer. I'm amazed that Scholes continues to think the model is fundamentally sound.. Frankly, it is a matter of time that the academic finance community will acknowledge that the Black-Scholes model for pricing options is fundamentally flawed. Not just the fact that the underlying stochastic process is incorrect in that they made unwarranted sweeping assumptions such as log-normal distribution, continuous time Wiener, constant volatility through the life of the contract etc etc. There's much written on these topics. ironically, the so called great breakthrough in the Black-Scholes- that investors preferences toward the option do not matter, and the only thing that matters is the risk free interest rate.. is fundamentally flawed. In any case, for all of the adulation, LTCM fellows were doing classic yield curve spread trades, swaps and straddle spreads, and some outright speculative positions-- all with enormous leverage and all in the same direction- betting that the spread would converge.. So much so for dynamic portfolio hedging..
Just had to drop a note and say thanks for the video. I watched the original airing of the TV program over 5 years ago. Good viewing...
Hi Guys, you're very welcome. Circle, I'm not sure that Scholes was referring to the Black-Scholes model in particular, rather the models that LTCM were using. Frightening leverage though. Regardless, what is surprising from the film is Scholes - for the brilliant man that he is - just didn't get it. I kinda like Merton though, the man is not afraid to show some emotion, and he really is one of the nicest guys you could meet. Hey rufus, you must have some interesting stories from the exchanges you spent some time in. Cheers, EQ.
Actually I am mostly an off-floor electronic trader (running mostly automated strategies for the past 4-5 years), so I have never spent significant time on the floor. But my observations on the floor did show me a different perspective, a very useful one. I used to argue that the core LTCM convergence strategy never did that badly (lost about $500-600M in total), it is the directional trades (such as the absolutely disastrous direction trade on S&P500 vol, that lost $1.3B) that killed the fund. So I think there isn't a black / white whether their "models" worked, they have a large variety of models, it is the correlation between the models that took the fund down to its knees, and then the directional trades are the nail on the coffin. To use my own experience as a small example. I was a part of a smallish hedge fund in '98-'01, the fund did very well in '99 (up >200%), so investor money flood in (we grew from about 90-100M to over 220 in less than 2 months). So we sat on top of a pile of money that honestly we didn't know what to do with, our core strategy, risk-arb and value rotation, faced some scale problems (heck, how many stocks can you call "value" in 2000 anyways). So we started experimenting with different strategies, such as stat arb, etc. While none of the experiments did particularly badly, it took the fund's attention away from the core strategies, and the combination, along with the equity market bubble bursting in Apr/May/Jun in 2000, took the fund down 35-40%. So instead of hoping to hit the high watermark again, we folded the fund, returned the investor money. The partners wasn't going to sit around and make 800k, when they took home 8-10M the year before. So stick with the core competencies is key, and only expand rationally. I think LTCM suffered from the same type of over-extention, and over-confidence in their models in markets they are not familiar with.
Thank you for posting this, it was an excellent video (though not quite as detailed as I might have liked). One thing I found interesting about this video were the interviews with various big names in finance. For example, Miller (half of the Modigliani-Miller theorem team) sure had a messy office. Bodie co-wrote one what was probably one of the best finance textbooks I've used in University: http://highered.mcgraw-hill.com/sites/0070897689/