ESTIMATING FUTURE BOND AND STOCK RETURNS

Spending rules have to be based on estimates of future bond and stock returns. As explained above, it’s the estimates of the real returns that matter for spending rules. In Chapter 2, we discussed how to formulate estimates using long-run historical data. In that chapter, we developed estimates of the real returns on Treasury bonds and the S&P 500 using data beginning in the 1950s. There we discussed returns on both the medium-term (five-year) Treasury bond and the long-term (20-year) Treasury bond, but here we will focus just on the medium-term bond.3 Since 1951, the medium-term bond has earned a real compound return of 2.4 percent.

In the case of the S&P 500, we developed two different estimates of the long-run real return. The first was based on the actual real returns earned from 1951 to 2009. According to Table 2.2, the real compound return has averaged 6.7 percent since 1951. That estimate reflects the actual capital gain experienced during the period. Fama and French (2002) developed alternative estimates of stock returns based on the rate of growth of earnings rather than the actual capital gain. Table 2.4 updated their estimates to obtain an average (real) compound return for the S&P 500 based on earnings growth of 5.3 percent.4

The first row of Table 14.1 lists these estimates of the real compound returns. Because we will need nominal arithmetic averages later in this chapter, the table also shows the nominal compound average calculated ...

Get Portfolio Design: A Modern Approach to Asset Allocation now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.