Valuation-Informed Indexing #301
by Rob Bennett
Robert Shiller’s 1981 finding that valuations affect long-term returns changed the process of developing investing strategies in a fundamental way. We used to believe that stock investing risk was static and that it thus made sense to stick with the same stock allocation at all times (the one exception being to lower one’s stock allocation as one approached retirement age because one’s tolerance for risk had changed). Now we know (or at least some of us think that we do!) that risk is variable and that investors seeking to maintain a stable risk profile need to adjust their stock allocations in response to big valuation shifts.
I have of course only been able to scratch the surface re the new way of thinking about stock investing strategies brought on by the Shiller Revolution. We now know that the safe withdrawal rate is not a steady 4 percent but drops to as low as 1.6 percent at times of high valuations and rises to as high as 9 percent at times of low valuations. And that the long-term return on our stock investments is highly predictable so long as we are willing to look at least 10 years out rather than focus on the results that we will obtain in a year or two or three (which really are unpredictable, just as the Buy-and-Holders say). And that there are times when stock prices rise so high that far safer asset classes offer better long-term value propositions. And so on.
The purpose of The Investor’s Strategy Tester is to permit the investor to explore strategic questions on her own, using the 145 years of historical return data available to us today as a guide to the pros and cons of various strategic choices. The investor is permitted to set up various strategic options and compare the odds of various outcomes under each of those scenarios given the range of results produced by those choices in the past.
Today’s P/E10 level is 26. That’s very high. So most investors should be going with stock allocations a good bit lower than those they would choose if valuations were at fair-value levels. But there are of course no black-and-white circumstances. If you go too low with your stock allocation, you might miss out on gains that those who follow Buy-and-Hold strategies will be enjoying. There are always judgment calls to be made. It makes no sense to stick with the same stock allocation at all risk levels. But the short-term is not at all predictable. You need to avoid going to extremes. It is a rare investor for whom a zero stock allocation ever makes sense.
The purpose of the Investor’s Strategy Tester is to reveal how things are likely to play out in the long term, presuming that stocks continue to perform in the future at least somewhat as they always have in the past. A key feature of the calculator is that it never offers a single recommendation to the investor seeking to identify his best strategy bet. The Strategy Tester assigns probabilities to various outcomes. It lets you know that Strategy Choice A is likely to perform better than Strategy Choice B. But it never lets you forget that, even when Strategy Choice A is clearly the better choice, there is at least a small chance that those following Strategy Choice B will put up better numbers. That’s the reality of long-term stock investing, like it or not.
Another important feature of the calculator is that it reports results at six time-periods: (1) five years; (2) ten years; (3) fifteen years; (4) twenty years; (5) twenty-five years; and (6) thirty years. Too much of today’s investing commentary is rooted in a belief that one strategy is superior for all time-periods. This is often not the case. A strategy that is likely to work well for five years may be likely to work poorly 20 years out. Investors need to know what long-term risks they are taking on by following a strategy that may well make them happy enough for five or ten years. Some strategies work well at all time-periods but are only a small bit superior in the early years and then become far superior only over a long period of time. Investors are more likely to be able to exercise patience if they know in advance that that is how things are likely to play out.
Say that you want to examine a very simple strategy question: Does it make sense to invest heavily in stocks when they are selling at fair-value prices? Pretty much everyone knows that the answer is “yes.” Stocks offer a better long-term return than most alternatives. But there are times when even the best informed investors lose confidence in stocks. Following price crashes, investors let the short-term pain they are experiencing draw them away from the generally superior asset class. The Strategy Tester makes clear with graphics that the value proposition of stocks purchased at fair prices grows stronger and stronger as time passes.
I entered a scenario where an investor has a $300,000 starting-point portfolio value and assumes that he can earn a 3 percent real return on money not invested in stocks. The P/E10 level is at fair-value at the beginning of the 30-year time-period to be examined. And the investor will be adding $20,000 to his portfolio each year. I compared four stock allocations: (1) 80 percent; (2) 60 percent; (3) 40 percent; and (4) 20 percent.
It’s certainly no surprise to see that the 80 percent stock portfolio performed best. Even the best possible results produced by the 20-percent allocation were barely as good as the worst possible results produced by the 80-percent allocation. The middle result for the 80 percent allocation beat the best possible result for both the 20 percent and 40 percent allocations and beat all but 25 percent of the results generated by the 60 percent allocation. Go out far enough in time and it is not true that stocks purchased at reasonable prices are risky; stocks purchased at reasonable prices always deliver outstanding long-term results.
What fools people is that that is not so at shorter time-periods. At five years out, the worst result produced by an 80-percent stock portfolio is worse than the worst result produced by a 20-percent stock portfolio. The best result is a good bit better. So it could be argued that it is a mistake even for an investor with a low risk tolerance to avoid stocks when they are selling at reasonable prices. But at five years out, the risk of stocks is real. If an investor truly cannot afford short-term underperformance, he needs to go with a lower stock allocation than what would be advisable for most other investors.
These are not startling insights. But I think the calculator serves a useful purpose if it helps investors to visualize the possibilities that may occur over time and identify the probabilities associated with each of the various possibilities.
Rob Bennett’s bio is here.
