Bill, your new book is excellent, thank you for your efforts. One of the two or three must haves for retirees! My question is this. Since the book was written and given the timeline of the data, does the fact that Schiller’s CAPE landing at about 35 now have a material impact on SAFEMAX? With all the data, I will likely have to read the book a second time and I may have missed this, but it appears that your limitation on CAPE in your charts runs up to about 22. I do recognize CAPE also was at this level in 2021-22, and it did not change things materially, so I assume it will not, but just curious on your brief thoughts on this generally. Thank you, sir.

 

Hi B.,

Thanks for your excellent question and interest in my research.

I did engage in a discussion of higher CAPES in my book, in section 13.5. Unfortunately, we have very limited data on what happens to portfolios which begin with CAPES above 30. Prior to 1997, 30 had been broached on once, in July 1929 (31.48). That was associated with a SAFEMAX of 7.41%, which is surprisingly high. However, that retiree experienced high deflation rates, which allowed them to actually reduce their withdrawals the first five years of retirement. Had normal inflation of 2% to 3% prevailed, it might have been a very different picture.

The peak CAPE of 43.23, achieved by the 4/1/2000 retiree, resulted in a SAFEMAX of 5.7%. However, there are only 25 years of actual data available for this retiree; the remaining five years are estimated using a “historical average” investment return and CPI. Thus, the 5.7% SAFEMAX is subject to change in the future.

Since July 1997, when the 30-level on CAPE was broached again, there have been a total of 19 instances of CAPE rising above 30, through January 1, 2015. This is the last quarterly retirement date for which there is 10 years of actual, not extrapolated, data. The average CAPE for all 19 instances is 37.2, and the average SAFEMAX for all 19 instances is 6.09%. Using averages is not my preferred method of analyzing data, but we have limited data to work with in this instance.

Since 1/1/2015, there have been an additional 18 instances of CAPE above 30, but there is insufficient actual data to compute SAFEMAX for them.

As I mentioned in my book, I believe a SAFEMAX of 5.25% to 5.5% seems like a reasonable, conservative estimate for current retirees. Inflation is a bit higher now than for many of the earlier above 30 instances. The Shiller CAPE is 38.2 as I write this, toward the high end of its range. I hope that retirees will eventually be able to increase their withdrawals from these low levels, unless inflation becomes a problem.

Hope this analysis helps.

Best regards, Bill Bengen