An earlier blog entry described the background of my recent correspondence with Michael Kitces on safe withdrawal rates (SWRs). Set forth below is the text of an e-mail that I sent Michael’s way in the evening of Thursday, August 21.
Thanks again for providing stimulating feedback.
There’s a punctuation mark that you use in your study that I found entirely appropriate in one sense and extremely odd in another sense. It comes at the end of the sentence in which you put forward the following words: “Those extended periods of expanding or contracting P/E ratios (producing long-term real returns above or below historical averages) can often be anticipated in advance — by looking at the valuation of the aggregate market at the beginning of the time-period.” You employed an exclamation point at the end of that sentence.
Your exclamation point expresses your surprise at the claim being made. That’s entirely appropriate because most investors and most investing experts today do not think of stock returns as being predictable. But It’s also extremely odd that a finding that stock returns are largely predictable would come as a shock to anyone in the year 2008. Whether stock returns are predictable or not is a question of fundamental importance. Can it really be that this reality comes as a surprise to many smart people at this late date?
It really does come as a surprise to many. I have interacted with tens of thousands of middle-class investors on internet discussion boards and most are shocked to hear that long-term returns are to a large extent predictable. But the historical data leaves room for virtually no doubt on this question. Why do people not know this? How is it that this critically important reality is not more widely appreciated?
I believe that the reason why the reality is not better appreciated is because Passive Investing is today’s dominant model for understanding how stock investing works. The Efficient Market Theory tells us that returns should NOT be predictable. If the market is efficient, the market price should always be roughly right. If the market price is always roughly right, it should not be possible to predict future returns by taking into account mispricings.
Your correct statement that knowing the P/E10 of the market helps us anticipate future returns is in conflict with what the the dominant model for understanding how stock investing works tells us to expect from stocks.
The root puzzle is — why were the initial SWR studies set up in the way they were? Why did anyone think it would be a good idea to describe the SWR as a stable number? Why was it not understood from the start that the SWR must change with changes in valuations (as do long-term returns)? The explanation is — the early SWR researchers believed in the Efficient Market Theory and in Passive Investing.
If you are right that returns are predictable (I strongly believe that you are), then the Efficient Market Theory is wrong and the Passive Investing model has failed to properly explain how stock investing works in the real world.
Do you see it that way? Do you agree that the stakes here are huge?