William J. Bernstein
Mammas, Don’t Let Your Babies Grow Up To Be Timers
As with the swallows’ return to Capistrano, the change in investment season has brought the warble of the full-throated market timer. Freshly aware that stocks can actually lose money, the battle cry grows from a burgeoning army of investment advisors and market strategists: "Future stock returns will be low! Only the nimble will survive!"
Now they tell us; just where were these geniuses three years ago, when prices were 50% higher? During the late-nineties bubble, how many of these newly devout timers pointed out that stocks do, from time to time, devastate their owners? Or that there were three twenty-year periods during the last century when real stock returns were nearly zero?
Recently, praise of timing has also come from a highly credible source—Peter Bernstein (no relation). Speaking before a group of analysts in January, he lashed out against the concept of the "policy allocation," such as the hallowed 60/40 stock/bond mix used by most pension pools, and came out in favor of a more opportunistic approach. (Mr. Bernstein discusses his views on the matter in an interview with Kathryn Welling of Weeden & Company.)
The gravitas of this famed author and economist makes him difficult to ignore. Back in the Alice-in-Wonderland world of the late 1990s, his voice was one of the few that questioned the relevance of high historical stock returns. Yes, he said, equity has always beaten bonds and bills over the long term, but that was simply because of an anomalously high risk premium that has long since gone the way of disco and the five-cent cigar.
A few samples from Mr. Bernstein’s recent remarks:
. . . we are going to have to learn to live without the crutch of things like policy portfolios.
. . . we’ve reached a funny position where the long run doesn’t work . . . the old long run, she ain’t what she used to be . . . the long run here is not necessarily going to bail you out . . . equities are not necessarily going to be the best place in the long run.
. . . you can wake up every morning saying, Well, how does the world feel? It’s a much more interesting way to live.
You have to be much more unstructured, opportunistic and ad hoc than you have been in the past . . . I am talking about that dirty word, market timing.
Mr. Bernstein also hammered away at the iron laws of expected returns, which he first wrote about in the 1990s as a voice in the wilderness: historical real per-share earnings growth is minuscule, as is dividend yield. In such a world, a positive equity risk premium is not a sure thing, even over many decades. But he is vague as to what alternative he is proposing beyond waking up in the morning, smelling the coffee, and trading like mad. He offered just one hint:
. . . you can do the market timing with a lot of quantitative stuff, tactical asset allocation. Wells Fargo, First Quadrant, they’re all doing tactical asset allocation on a quantitative basis. If I were going into that, that’s how I’d do it.
Now we’re getting somewhere. Not only do I like and admire Rob Arnott, chairman at First Quadrant, but this hedge-fund manager possesses another sterling quality: he’ll talk to me. So, unable to restrain my curiosity, I rang Rob up and asked him how he does things. The short answer: he focuses on risk-adjusted expected returns. If the expected return of an asset class goes up/down, so does his allocation to it. If its risk goes up/down, his allocation does the opposite. He also pays attention to money supply and a few other parameters, but basically, he’s a concave investor: when Mr. Market has been selling hard and long enough to seriously cheapen an asset, he’s buying, and when Mr. Market has a prolonged manic break, he’ll take the opposite side.
If that’s market timing, then I too must plead guilty. Sure, I believe in the efficient market hypothesis, but that doesn’t translate into ignoring the risks and returns of asset classes and failing to act accordingly. In the last days of the bubble, the expected equity risk premium was close to zero, and if you believe that TIPS represent a risk-free asset (some do, I don’t), the risk premium was decidedly negative. Today, the expected risk premium (ERP) is probably in the range of about 4%—the difference between the expected stock return and that of 10-year treasuries. Does the prudent investor own more stocks at an ERP of 4% than at an ERP of zero? You bet. How much more? That depends upon the mission; if it's a hedge fund, probably quite a lot. If it’s a conservatively managed tax-sensitive account, relatively little. In other words, if you were a 65/35 person in 1999, you might be a 70/30 person now. (Except that now you’re four years older, so maybe you’ll slip right back to 65/35.)
The rub is that "timing" is an inflammatory six-letter word—a veritable bomb in the staid world of portfolio management, and one that Mr. Bernstein threw with wonderful effect. Its spectrum stretches all the way from large and rapid changes in allocation based on things like macroeconomic parameters, relative strength, volume, sentiment, and overall gut feeling—certifiable behavior, in my opinion—to slow and relatively slight changes in allocations based on valuation and expected return. This latter strategy, involving very small and infrequent policy changes opposite large market moves, more often than not improves overall portfolio performance.
The latter concept is a bit difficult to grasp. Think of it this way: even the most devout efficient marketeers rebalance; trimming a portfolio back to policy is nothing more, and nothing less, than a bet on mean reversion. Taking the process one step further and adjusting the policy allocation itself opposite valuation changes is merely a way of amplifying a rebalancing move—"overbalancing," if you will.
It all comes down to predictability. Just how predictive are the usual market strategist staples—sentiment, volume, moving averages, and so forth? Near zero. On the other hand, how predictive is an S&P dividend yield of 1.2% or an average REIT yield of 9%, as we simultaneously saw in 1999? Since we have only one sample of history, we can’t know for sure, but I’d be willing to give odds of at least 60/40 in favor of below and above average future returns, respectively. There is, in fact, a mountain of data from the markets of many nations pointing to modest predictive value to such balance-sheet parameters. Over the lifetime of a portfolio, enough judicious allocation changes based on such 60/40 bets almost certainly adds value.
Many investors, at least the ones who could add, shaved back their allocation to large-cap U.S. stocks in the late 1990s—even the normally bullish Jeremy Siegel wrote a delightful piece at the height of the madness in March 2000 entitled "Big-Cap Tech Stocks are a Sucker’s Bet." Ten years before that, an even bigger bubble inflated the Japanese stock market to a 67% cap weighting of the EAFE. Had there been an EAFE index fund over the past fifteen years, it would have been beaten by 92% of actively managed funds, since few active managers were crazy enough to match the Japan-heavy EAFE country mix. This is most certainly not a case against foreign indexing—passively managed European, Pacific Rim, and emerging markets index funds, in general, have blown the doors off their actively managed cousins. (Over the past ten years, for example, both the MSCI-EAFE-Pacific-ex-Japan index and the DFA Pacific Rim Small Company portfolio have beaten all nineteen actively managed Pacific Rim funds—it speaks volumes when both large-cap and small-cap indexed approachs simultaneously beat all of the active competition. The same is true in the emerging markets arena.)
Why this disconnect between actively managing policy allocations and actively managing individual security selection? The explanation gets to the difference between market efficiency and rationality. Here are two proposed operating definitions:
- "Efficient": Among a group of similar securities, there is no payoff to careful selection.
- "Rational": The existence of reasonable risk premia, relative both to risk-free asset classes as well as among risky asset classes.
Market irrationality is like pornography: difficult to define, but you know it when you see it. One doesn’t have to look very far to find violations which most reasonable investors can agree upon and which can be surprisingly long lasting. A quote apocryphally attributed to Keynes has it that "The market can remain irrational much longer than you can remain solvent." For example, the junk-treasury spread can hover around 300 bp for years, despite a long-term historical junk-bond loss ratio in the 400 bp range. Other asset classes can also become hopelessly overpriced or underpriced, oft times simultaneously, as happened to large-cap U.S. stocks and TIPS in late 1999. The zero ERP that characterized the "new era" made sense only to semi-delusional right-wing financial columnists.
Simply put, although the individual investor will likely come to grief manipulating the selection of individual securities, the judicious adjustment of policy allocations according to expected returns—increasing an allocation slightly when its expected return is very high, decreasing an allocation slightly when it is very low—will on average slightly enhance long-term results. This is simply an amplification of normal rebalancing.
Varying allocations—"timing," if you will—is similar to the consumption of alcohol. It can either enhance or degrade portfolio health; it all depends upon the circumstances and the quantity. When partaken in small, infrequent amounts from a concave vessel, its benefits are small but perceptible. When chugged indiscriminately, it is deadly.
Copyright © 2003, William J. Bernstein. All rights reserved.
The right to download, store and/or output any material
on this Web site is granted for viewing use only. Material may not
be reproduced in any form without the express written permission of William
J. Bernstein. Reproduction or editing by any means, mechanical or electronic,
in whole or in part, without the express written permission of William J.
Bernstein is strictly prohibited. Please read the disclaimer.