Valuation-Informed Indexing #89
by Rob Bennett
People have a hard time getting their heads around the idea that market timing can be a risk-diminishment strategy. When people hear the word “timing,” they think of short-term timing, guessing in which way stock returns are headed for the purpose of obtaining higher returns. This is properly viewed as a high-risk strategy. People understand that short-term timing can produce good returns. What they don’t like is the added risk. Their dislike for the added risk that comes with short-term timing causes them to react with distaste to long-term timing too.
But short-term timing and long-term timing are not similar strategies. They are very, very, very different things. There is no way to know which way returns are headed in the short term. So short-term timing is guessing. That’s why it is risky. Long-term timing is not guessing. Long-term timing is a sure thing.
It’s that sort of statement that causes people to label my work “controversial.” Long-term timing is a sure thing? Huh? That cannot be. That’s an outrageous statement That’s another case of Rob being Rob.
I don’t mean by that statement that anyone can know precisely when the market is going to turn. That’s not possible. But it is not necessary to know precisely when the market is going to turn to profit from long-term timing. That’s simply not the point of the exercise. The point of the exercise is to keep your risk profile constant. It is possible in many circumstances to make an allocation shift with certainty that doing so will improve your risk-adjusted long-term return.
Say that you were going with a low stock allocation at some point between the year 1975 and the year 1995. Stocks were at very low, moderate, or somewhat but not insanely high price levels for that entire time-period. Going with a too low stock allocation is risky. Why? Because you only get so many years to finance a retirement and only stocks provide returns high enough for most of us to do the job in the time allotted. Let your fear of stocks cause you to underinvest in them and you are increasing the risk that you will fail to meet your financial goals.
So increasing your stock allocation was a positive move during that entire time-period.
Did it matter whether you made the change in 1975 or in 1980 or in 1985 or in 1990? Yes. It would have been better to have made the move in 1975. Once stock prices dropped to low levels, increasing your stock allocation was the right move and it’s better not to delay making a right move.
But it’s not right to say that there was risk in making the right move in 1980 or 1985 or 1990. The right move is the right move. The payoff for making it was not as great in the later tine-periods. But there was still a payoff. And obtaining a payoff is not taking on a risk. There was more risk attached to a decision not to increase your stock allocation in those years. So it cannot properly be said that a decision to increase your allocation was risky. Such a call was a risk-diminishing call, not a risk-enhancing call.
But wait. I can hear the objections of my critics coming through the speakers of my MacBook Pro as I type these words. “Writing these words in 2012, you know how things turned out, Rob. No one knew how things were going to turn out in 1975 or 1980 or 1985 or 1990 or 1995. Those were risky calls at the time they were made.”
Not if you believe that Shiller’s finding that valuations affect long-term returns is legitimate.
If valuations affect long-term returns, we can know in advance (by looking at the valuation level that applies) the range of long-term returns that applies for a given purchase of a broad index fund. If all or almost all possible returns beat the returns available from the non-stock asset classes, there is no risk in going with stocks. That was the case for most of that 20-year period. For the years in which that was the case, it is not fair to say that going from a low stock allocation to a moderate stock allocation increased risk.
The opposite has been true for the past 16 years. Stock prices have been insanely high for most of the time-period from 1996 forward (except for a brief period of time in early 2009). In the more recent time-period, the odds that the long-term return on a broad index fund will beat the long-term return provided by super-safe asset classes have been poor. In those circumstances, the opposite of the claim put forward above applies: Lowering your stock allocation diminished risk.
Market timing earned a bad reputation in pre-Shiller days, when people did not know how to do it right. But many investors have drawn the wrong conclusions from their negative timing experiences. It’s not timing that is bad. It is guessing which is bad. The form of timing that involves guessing (short-term timing) really is bad. The kind of timing that does not involve guessing (long-term timing) is very, very, very good.
If valuations affect long-term returns, it is staying at the same stock allocation at all times that increases risk to unacceptable levels. If valuations affect long-term returns, stock risk is not a constant but an ever-changing thing. Close your mind to the need to engage in long-term timing and you insure that your stock allocation will be wildly off the mark through most of your investing lifetime.
That’s not smart or stable or reasonable investing That’s dangerous investing. You should want to keep your risk profile roughly constant. To do that, you must be open to making shifts in your stock allocation percentage in response to big valuations swings.