How accurate were options prices BEFORE Black-Sholes?

The Black-Sholes method was invented in the 70s to figure out what an option should be worth in a perfectly efficient market. But how close to perfect were option prices before the method was invented? Did the traders of the day generally get the prices more-or-less correct by intuition? Or were there wild differences between what options were trading at and what they should have been worth?

I don’t know if Black Scholes is ‘accurate’ in terms of always getting the right price. It requires a lot of assumptions some of which aren’t necessarily accurate in real life. I would say that it was the first attempt at a consistent pricing model, and in that way was revolutionary. Their work has also allow a whole slew of other folks to add to their research and perhaps reduce some of the assumptions that go into the model. I think it helped in the development of other derivatives after future academics added to their work.

I think quant jocks for years were using their own methods which were roughly based on their own analyses of volatility and rates. Some of my old materials have a reference to a book in 1900 called the Theory of Speculation by Bachelier. I think Put Call Parity was in use prior to B-S. Not being a quant guy myself, I can’t even begin to go into all that. But the bottom line is that the models used were not consistent and probably yielded divergent values on any given option.

Someone with far more brains that I have can discuss the uses of Brownian Motion…blah blah blah in pricing derivatives.

There were options of course, before Black Scholes. But the Chicago Baord Options exchange began the first ogranizes trading of options in April 1973 and the Black Scholes paper was published in the May-June 1973 issue of Journal of Political Economy. The model priced correctly (i.e. gave prices quite close to the market prices until about 1987). Subsequent to that it’s not been quite so accurate.

The big question, of course, is how do you know which is right? The answer shoudl be, well if the model helps me make money I wouldn’t otherwise then it has useful information in it. Suppose it helped you “guess” the right way to invest 51% of the time as compared to blind luck of 50%. How many flips of that coin would it take to convince you?

There is, however, a nice story circulating called something like "How I help Fischer Black (the Black of Black Scholes) a million dollars. When the Kansas City Board of Trade began trading Value Line Futures contracts. They were priced wrong because traders didn’t understand tha a geometric index behaves differently from a regualr arithmetic index like the S&P or Dow. You can show that the geometric index in the Value Line will underperform a portfolio of the same stocks by one to two % per year. Traders were buying the Value Line index to take advantage of the “Small Firm” effect and shorting other futures to hedge themselves. But they lost to Fischer on the mispring.