Set forth below is the text of a comment that I recently posted to the discussion thread for another blog entry at this site:
Shiller quite clearly said don’t time the market. Unless you can produce a quote that contradicts that (and you never have) there’s no need to keep asking him the same question.
Anyway, here’s a new peer reviewed paper: “Shiller’s CAPE: Market Timing and Risk”
“We find that stocks outperform 10-year U.S. Treasurys regardless of CAPE, except when CAPE is very high (as mentioned above, the reference point is 27.6). In other words, consistent with market efficiency, ‘buy and hold’ is generally a good strategy even when stock valuations appear to be high by historical standards. Despite the strong negative relationship between CAPE and long-horizon stock returns documented by prior researchers, it is difficult to take much advantage of this apparent market inefficiency. The stock market as a whole appears closer to being efficient than not.”
That’s much for the link to the research paper, Anonymous. That’s obviously a huge help.
There’s a good chance that I will write a column re this paper. The comments below are in response only to the abstract. So this is just an initial reaction.
I don’t agree even a tiny bit that the results shown here are consistent with market efficiency. If the market is efficient, there cannot be ANY overvaluation. “Efficiency” means that all factors are being considered. If all factors are being considered, the market is properly priced. If there is any overvaluation at all, the market is not properly priced. So I don’t get that statement at all.
It’s an analytical mistake to make the comparison to 10-year Treasuries. That greatly biases the result. TIPS were paying 4 percent real in 2000. That was the risk-free return at the time. No investor would have chosen Treasuries as an alternative to stocks at a time when a 4 percent real risk-free return was available. Make the comparison to the rational alternative to stocks and you of course get a very, very different result.
However, I do agree with the general point that the valuation level for stocks has to be very high for stocks to be beat by a risk-free asset class. John Walter Russell did research at the SWR Research Group showing this. He showed that an investor who went with a heavy stock allocation whenever the P/E10 level is below 20 would do fine.
That doesn’t mean that all investors should follow that strategy, only that it is one reasonable option. But it does show how powerful the long-term value proposition is for stocks in ordinary circumstances. One of the reasons why the benefits of long-term timing (price discipline) have remained hidden from many of the experts in this field for so long is that it is a rare event for the P/E10 value to go above 20. So there have not been many circumstances in which long-term timing was required or even that beneficial.
Recent years (the time-period from 1996 forward) have very much been an EXCEPTION. This is the first time in the history of the U.S. market in which we have seen so long a time-period in which going with a low stock allocation was a good idea according to the peer-reviewed research in this field. It doesn’t change the reality to point out that today’s reality is an unusual one. But it does go a long way toward explaining why lots of good and smart people have had a hard time recognizing this important (and otherwise easy to understand) reality.
I of course very much disagree with the conclusion that it is not possible to take advantage of market inefficiency. But it appears to me that this is an important study, one that everyone who works in this field needs to take a look at. I am most grateful to you for having brought it to my attention, Anonymous,.
Please take good care.