would contribute to mean reversion of returns. Stock market returns are influenced by growth in corporate earnings and changes in average P/E ratios. Each has tended to fluctuate around long-term averages. For example, periods of investor exuberance produce higher than normal P/E ratios, while periods of investor pessimism can produce lower than normal P/E ratios. Stock market returns can fluctuate significantly from one year to the next as economic growth and P/E ratios oscillate around their long-term averages. Such oscillations would contribute to mean reversion in market returns. Figure 30.6 shows how longer horizons narrow the range of annualized returns. Annual stock returns have varied widely, ranging from a negative 54 percent during the 12 months ended 6/30/32 to a maximum of positive 140 percent, which occurred in the subsequent 12-month period, ending 6/30/33. That's volatility! That is also an extreme example of mean reversion. Over longer periods of time, cumulative returns have converged toward their long-term average. During this time period, which included the Great Depression, there was never a 25-year period in which the diversified stock market failed to increase real after-tax wealth. There is a similar argument for mean reversion of bond market returns. Over the past 76 years, bond yields have generally been in a range of about 5 percent to 7 percent. A period of rising market yields will produce negative bond returns. However, if market yields subsequently decline to their normal range, there will be a period of unusually high bond returns. The distribution of bond and bill returns exhibits a similar pattern, although the one-year returns are far less volatile that those of stocks. However, a long-term investor should recognize that the real, after-tax returns of bonds and bills have converged toward low or negative averages. Figure 30.7 shows the distribution of 25-year returns for all three asset classes. Stocks consistently provided growth in real, after-tax wealth; bonds rarely provided any growth in real, after-tax wealth; and Treasury bills have never provided an increase in real, after-tax wealth. This leads to the trade-off between safety and effectiveness. In nominal terms, all asset classes have positive expected returns and thus would appear to contribute to the goal of wealth accumulation. But, when we shift to wealth stewardship focused on real, after-tax results, we find that money market and fixed income securities may be ineffective and possibly counterproductive. This is an important consideration for the investor who is interested in long-term results. Investors who base investment decisions on nominal, annual return and risk data are likely to conclude that T-bills are riskless, bonds have modest risk, and stocks are very risky. And it is absolutely true that during short investment horizons, stock and bond returns can vary, with stocks being particularly volatile. However, an investor who is focused on the long-term preservation and growth of the real value of an estate may come to a very different conclusion. Since 1926, there has never been a 25-year period in which stocks failed to increase real wealth, there have been few periods in which bonds increased real wealth, and "safe" Treasury bills almost always reduced real wealth. To be more precise, it was the issuer of the Treasury bills, the government, that destroyed the wealth through taxation and monetary policy. Individual investors often use risk analysis to get an idea of potential worst-case scenarios. An understanding of the relationship between time and risk is very important in this regard. Riskier asset classes have higher expected returns but