Rob Bennett emails to tell me about a handy little tool he’s constructed, which looks at stock valuations and calculates what kind of real return you can expect to get from investing in US equities over the next decade or six. The tool is based on one of Robert Shiller’s favorite indicators, P/E10, or price to previous ten years’ average earnings. Here’s how P/E10 evolved between 1881 and 2006:
Right now, P/E10 is at about 22, which mean that stocks are expensive, by historical standards: Rob reckons that fair value, for stocks, is about 14 or 15, and that anything above 20 is "truly dangerous":
There have been three occasions when we have gone to
P/E10 levels above 20. On those three occasions, the average loss in the years that followed was
68 percent. So the general rule is to lower your stock allocation when the P/E10 level goes above 20.
Think of Rob as a buy-and-hold kinda guy with very infrequent reallocations, just like most sensible financial advisers. But Rob’s reallocations are really infrequent: only once a decade or so.
The key point is that all investors should be changing their stock allocations a bit when valuations
go through dramatic shifts. There cannot be one rule for all because different investors face
different financial circumstances and have different risk tolerances. But the conventional idea that
investors should stick with a single allocation at all price levels just does not make sense.
I favor Stay-the-Course investing, but I reject Passive Investing. I say that investors should be trying to keep
their risk/reward ratio roughly stable. That does not require lots of allocation shifts. It generally requires
one adjustment every 10 years or so. But it is critical that the long-term investors make those shifts
when valuations go to the sorts of extreme levels that apply today. Investors who try to practice
buy-and-hold when prices are going down by 68 percent are eventually going to become so discouraged
that they abandon stocks altogether, which is the worst possible thing to do.
This plan isn’t really about market timing: it would have had you underweight equities for pretty much all of the big 1990s boom. But Rob’s sure that over the long term looking at valuations does make sense:
Those lowering their stock allocations in the mid-1990s
obviously did not pick the top! It doesn’t matter. If they stick with a valuation-informed strategy,
they will end up ahead of those following Passive Investing in the long run. We have already
reached a point where those who lowered their stock allocations in late 1997 are ahead
of the game. As prices continue downward, those investors will move farther and farther ahead.
Eventually, even investors who lowered their allocations in 1995 will be ahead. It often
takes time for rational prices to reassert themselves, but they always do.
I think Rob’s approach has a lot to be said for it, but I do have a few problems with it. For one thing, I just don’t think that many investors have anything like the requisite amount of patience needed – where you can happily sit back for a decade or two waiting for the P/E10 to come down to the mid-teens before going overweight equities.
And for another thing, I’m far from convinced that the behavior of the stock market a century ago, when both the world and corporate capital structures were very different from how they look now, can really teach us much about stock-market investing. I don’t believe that the stock-market asset class has rules which govern its long-term behavior, and I can easily believe that we will never again see the S&P 500 trading on a P/E10 of less than 15. (It’s already been 20 years, and I certainly don’t believe that discount rates are likely to go back up to their 1988 levels any time soon.)
All the same, for people who like to take a 30,000-foot view of investing, this is a very handy little tool. And it gives me pause, too: I’ve worried for a while that stocks aren’t an attractive asset class any more. The problem, of course, is that there’s not very much in the way of alternatives. And in any case, according to Rob, if I put all my money in stocks today, I can expect a real return of 5.65% over the next 30 years; 4.65% would be unlucky. I’d be happy with that, I think.
(Thanks to Bob’s Financial Website for the historical P/E10 data.)