AQR Working Paper
From 1970–1996 the compound return on a portfolio of stocks mimicking the S&P 500 was better than the return from rolling over 1-month T-bills or from investing in long-term bonds. An investment in an “S&P 500 portfolio” of stocks for the full time period minus the stock market’s six best months, would have led to a return equal to that of long-term bonds. Being out of the stock market for the best 13 months would have led to T-bill return.
In others words, missing only 13 months out of 27 years (324 months) would have eliminated the entire return premium for stocks versus T-bills, while retaining all of the stock market’s risk!
Severe consequences of this type are often used as an argument against “market timing,” often by those with a vested interest in keeping investors in stocks. However, there is a problem with the above line of reasoning. Examining only the potential harm from timing is entirely one-sided. The potential benefits to timing are of similar magnitude to the potential losses.
This paper does not advocate market timing. In general, like all active management, market timing should be undertaken only to the extent an investor feels his skills overcome the hurdles. If investors believe they have no skill after hurdles (call this net skill), we believe they should not time the market at all. Investors with a small amount of net skill should only allow small deviations from the otherwise optimal passive market portfolio. Only for investors who truly believe they have a large amount of net skill (a presumably rare event) are large market timing bets rational.