To: Robert Douglas who wrote (95361 ) 1/7/2000 2:27:00 PM From: GVTucker Respond to of 186894
OT***way off topic and horribly boring to everyone but us I am familiar with the Maginn/Tuttle passage. In fact, when I was studying for the CFA, I pointed out the error to the AIMR. This particular text was written in 1983, when understanding about stock index futures pricing was still in its infancy. One passage that I find particularly dated is toward the end of their Stock Index Valuation summary that states, "this contract has several features that make the underlying arbitrage difficult..." Back in those very olden days, when professionals couldn't use the DOT system for the most part, a buy program was indeed very difficult. The early arbs would have a pre-printed stack of 500 cards ready to go. When the buy program would hit, ten different runners would literally run around the NYSE, depositing the cards at their respective stations. Execution risk was huge, and thus the arbs needed a giant premium before trying a buy program. Obviously, the world has changed a little since then, and index arbitrage execution is now a very simple process. If you want a more detailed analysis, you can go back to the early work of Modest and Sundaresan, going back to 1983 in the Journal of Futures Markets, or Figlewski in Financial Analysts Journal in '84. (As a caveat, I stopped paying attention to this back in the late 80's, so interpretations could have changed since then. It was a mathematical proof that led to the risk premium conclusion, however, so I would think that the odds favor that this theory has remained intact.) Just a final thought. The one aspect that Maginn and Tuttle did not realize is that the S&P Futures contract in one of the few contracts that actually has an expected return. The expected return on the S&P Futures contract is the risk free rate. Thus, an arb that is short the future and long the cash must also factor into the model an expected loss on the future. For everyone else, I promise if this goes any further that I'll take it to PM. Although I thoroughly enjoy such a discussion (and I suspect that you do also, Robert), I get the feeling the most people do not. Kind of like when I get into a discussion of macroeconomics at cocktail parties. It's a pretty quick way to disperse a crowd.