Here are two more links to charts that add to the earlier discussion of normalized P/E ratios, interest rates, and long-term returns:
[charts may take a few minutes to load]
As you can see using the recent past as a guide to the relationship between interest rates and market valuations is suspect, because you're really looking at the run from the 1982 normalized earnings multiple bottom to the 1999 normalized earnings multiple top. This period matches well with interest rates – at least in terms of direction.
However, we are talking about a specific time period where a lot of that movement could be coincidental. If you look at the chart, you'll notice that there have been other periods where the Dow's 15-year normalized P/E ratio changed a great deal despite a lack of any concomitant move in interest rates.
Again, this supports the idea that the absolute level of interest rates (high or low) is not as important a determinant of how the Dow's earnings power is valued by investors. Rather, it is the direction of interest rate movements that seems most likely to coincide with the direction of changes in investor perceptions of the Dow's earnings power. All of this may simply mean that investors tend to project the recent past into the distant future.
This may help explain why the multiple placed on the Dow's rather steady long-term earnings potential has varied quite a bit over the years – as the earnings of the recent past, the trend in earnings growth, interest rates, and the rate of inflation over the past few years come together to form projections for the future that are often too pessimistic or optimistic given the tendency for such extreme trends to mean revert.
The second chart shows the distribution of various normalized P/E ratios for the Dow from 1935-2005. The percentage difference between actual earnings and normalized earnings is also shown. Numbers in parentheses are negative.
If you have any questions about the charts, please send me an email.