Valuation-Informed Indexing #128

by Rob Bennett

I wrote about an article (“Investing Based on Market Valuation”) in the Oblivious Investor blog in last week’s column. There’s one more point that Mike Piper made in the article that I would like to take a look at today.

Mike says: “If P/E10 was at a historical high at a time when TIPS yields were also very high, I wouldn’t fault somebody for moving more of their portfolio to TIPS. Similarly, I think it would be reasonable to use a lower withdrawal rate if you retire at a time with unusually low TIPS yields and an unusually high P/E10. On the other hand, I’d be extremely reluctant to suggest either moving more of a portfolio to stocks or using a higher withdrawal rate from a retirement portfolio just because P/E10 is low by historical standards.”

I know where Mike is coming from. There was a time when I would have agreed wholeheartedly with these words. I don’t agree with them today. In fact, my view today is that a good argument can be made that these words are irresponsible (I don’t believe that Mike holds any intent to be irresponsible or is aware that his words here are irresponsible, however). I’d like to explain why.

I am the person who discovered the errors in the Old School safe withdrawal rate studies. I began writing about the issue in May 2002. So I have had a lot of time to think about all of the issues that come up in discussions of safe withdrawal rates. My thinking has had a chance to evolve. I now believe that it is critical that investors be advised to take higher withdrawal rates from retirements beginning at times when P/E10 levels are low.

Mike is trying to be reasonable. He believes in Buy-and-Hold. So he doesn’t like the idea of investors changing their stock allocations in response to changes in valuation levels. But he understands that valuations matter. So he is seeking out a middle ground. He is suggesting that some small changes at the edges might be okay while ruling out big allocation shifts. If the safe withdrawal rate is something in the neighborhood of 4 percent (Buy-and-Holders now acknowledge that the numbers in the Old School studies are wrong but they remain unwilling to acknowledge how wildly off the mark they are), it would be dangerous for retirees to go with a withdrawal rate of 9 percent, even when prices are at rock-bottom levels and the research shows that that withdrawal rate is perfectly safe.

I remember when John Walter Russell informed me of the 9 percent number. It shocked me. I had felt comfortable saying that taking a 4 percent withdrawal was dangerous in retirements beginning at times of insanely dangerous valuations. This was something different. Could a 9 percent withdrawal ever truly be safe? For a time, I downplayed the 9 percent number. I didn’t want to be responsible for anyone going with such a withdrawal rate and then experiencing a failed retirement.

The reality is that a 9 percent withdrawal rate taken in a retirement beginning at a time when valuations are one-half of fair value is every bit as safe as a 1.5 percent withdrawal rate taken in a retirement beginning at a time when valuations are three times fair value. The safe withdrawal rate that applies for the former retirement has to be six times larger than the one  that applies for the latter retirement since the valuation level that applies for the latter retirement is six times larger. Why are we all so reluctant to acknowledge this obvious truth?

It feels funny. Adjusting portfolio values to reflect the effect of valuations is not yet the conventional thing to do. Humans are social creatures. When logic and our desire for social acceptance push us in different directions, we feel more comfortable yielding to the pressures exerted by our desire for social acceptance.

This widespread reluctance to accept what logic demands may be causing us all a whole big bunch of misery in the not too distant future.

We are today at a P/E10 level of 23. There has never been a secular bear market that has come to an end without us first falling to a P/E10 level of 8. That’s a 65 percent price drop. That sort of price drop would translate into additional investor losses in the trillions, losses large enough to put us in the Second Great Depression.

When we see P/E10 levels dropping below fair-value levels, we need to find some means to encourage investors not to sell stocks but to buy them. It is imperative that we not see valuations fall to insanely low levels.

What would persuade investors? Research showing the returns that stocks purchased at low or moderate price provide to the investors making their purchases at such price points!

But investors will be wary of claims made on behalf of stocks after the next crash. If we are going to persuade them, we need to be able to make a convincing case.

Mike’s case — that prices might kinda sorta matter sometimes but not really too much — won’t get the job done. If Mike believes that it is not safe for investors to choose high withdrawal rates for retirements beginning at times of low P/E10 levels, he is not really convinced that valuations matter all that much. If the people trying to persuade investors to buy stocks following the next crash don’t believe what they are saying, the investors hearing their words won’t believe them either.

If valuations matter, they matter. There’s now 30 years of research showing that valuations matter a great deal indeed. We need to stop pretending that valuations might sorta, kinda matter and get about the business of exploring the research-based finding that the price that applies at the time a stock purchase is made is by far the biggest factor determining whether that purchase will end up providing a strong value proposition or not.

Rob Bennett created The Stock Investment Strategy Tester. His bio is here.