The mere mention of the years 1987 and 1929 should serve as a reminder of this, but market history is replete with other gross discontinuities in asset class behavior. My personal favorite is the performance of bonds before and after 1984. For the 50-year period from 1934 to 1983 the return of the long treasury was 3.48% annualized. Had you depended on the historical record for an estimate of expected bond returns you'd have guessed wrong about the 11.34% return over the next 16 years. (And on October 19, 1987 things got spectacularly singular-a minus 23 daily-standard-deviation fall in stock prices. For those of you unfamiliar with statistics, 23 standard deviations is about the same odds as your computer suffering spontaneous decomposition and reassembly on one of Jupiter's moons, or of my starting at cornerback for the 49ers next year.)

Even the supposedly immutable long-term relationship between debt and equity returns is not written in stone. From 1802 to 1900 the return of US stocks and bonds was nearly identical at 5.89% and 5.87%, respectively, compared to 10.30% and 4.01% in the 1900s. Remember that inflation was close to zero in the 1800s, but about 3.3% in the last century. Thus a large real return was earned for both stocks and bonds in the 1800s, but only for stocks in the 1900s. What will be the relative returns of stocks and bonds in the next century? If you think you have the answer, please tell me. I'd love to know.

One thing is clear, though-leveraging gargantuan sums without a proper appreciation of the capriciousness of the capital markets is the financial equivalent of skydiving while drunk. And if your models are largely based on the last few years of data you've just left your parachute on the plane.
<<Previous                       Next >>

Printer Friendly Page