In Predicting the Markets of Tomorrow author James O'Shaughnessy offers his ideas on the investment environment we are likely to encounter over the 20 years from 2006 through 2026. He selected twenty years as this time horizon based on extensive analysis of market behavior over approximately the last 200 years. His logic goes something like this:1.When calculating returns from any investment strategy, it is essential to focus on the real return, after accounting for inflation.
2.Approximately two hundred years of stock market data (1809-2004) show that real returns have been highly erratic, especially when analyzed over periods of a few years or less.
3.However, when one calculates returns using overlapping periods of 20 years, they become much smoother. Stocks have rarely lost value over a 20 year period.
4.There are probably some underlying factors that cause returns to be smoother over 20 years. O'Shaughnessy suggests two. First, many investors don't really get started saving and investing until their mid 40s, giving than about 20 years to accumulate assets before retiring. Second, retirement at 65 together with a life expectancy of 85 suggests retirements (and asset depletion cycles) that last about 20 years.
5.If one decomposes the 20 year average returns of the S&P into the returns of the growth and the value stocks that comprise the S&P, these two groups have tended to move out of cycle with each other. Growth stocks occasionally have produced the higher return, as they did in the 1980s and 1990s. More often, value stocks have outperformed value stocks.
6.The returns of these three groups (S&P, Growth, and Value) all seem to revert to their mean rates of return. Any group that has outperformed in a 20 year interval is likely to underperform in the next 20 year period.
7.Since growth stocks outperformed in the 1980s and 90s, it's now their turn to underperform while value stocks outperform.
8.One can also segment the market by the capitalization (the total value of all the shares of a company). This analysis suggests that small cap stocks are likely to outperform large cap stocks over the next 20 years.
9.The average 20 year real returns /standard deviations of the key market groups between 1947 and 2004 have been:
Large Cap Growth: 6.26%/3.83%
S&P 500:7.30%/3.76%
Large Cap Value: 10.32%/3.42%
Small Cap: 10.42%/2.94%
10.As seen in the figures above, Large Cap Value and Small Cap stocks have higher returns with lower standard deviations. When you add on the fact that these two groups have underperformed over the last 20 years, O'Shaughnessy appears to have a compelling argument for focusing on these two groups. To hedge his bets slightly, he recommends a preferred portfolio allocation of 50% large cap value, 35% small cap growth, and 15% large cap growth.
11.Fixed income securities, even inflation protected treasuries (TIPS) are currently producing returns that, at best, break even. They are "Return-free risks, not risk-free returns".Avoid them except as a place to park cash they you will need in the next few years.
Reviewer's Comments: I agree with O'Shaughnessy's approach and conclusions but would have liked a better justification for using 20 year average returns. One could argue that generations are separated by about 25 years which might make that figure the logical interval for averaging. Perhaps someone has (or should) compare the results of averaging over different periods such as 10, 15, 20, 25 and 30 years. Or, even better, use a Fast Fourier Transform to determine the power spectral density of each time series.
Click Here to see more reviews about: Predicting the Markets of Tomorrow: A Contrarian Investment Strategy for the Next Twenty Years (Hardcover)
