One thing that unites most financial managers these days is the assumption that asset allocation will be challenging in the years ahead, as valuations for both stocks and bonds all but guarantee muted returns in the decade ahead. Unfortunately, most analysts who remind us of this reality fall short in terms of useful suggestions for just what to do about it, and when allocation advice is actually proffered, little is mentioned of the hidden bets those allocation suggestions are actually making.
Take, for example, the relatively simple math of bond returns. As everyone knows, rising interest rates are a negative for bond prices, particularly long-term bonds, which have considerably higher duration than short- and intermediate-term bonds. This is especially true in today’s low-rate environment, in which long-term government bonds now have essentially the same interest rate sensitivity as equities. It goes without saying, then, that a recommendation to allocate to long-term bonds today carries with it quite a bit more risk than it has at almost any other time previously. Given this, would-be long bond buyers should first ask themselves what is required for their bets to be right. At the very minimum, a profitable long bond bet at these starting yields requires continued anemic economic growth and below-target inflation.
Those seem like heroic assumptions, however, to this writer. Furthermore, while we are told that the bond market is the stock market’s older and wiser brother, its track record in predicting inflation is relatively dismal. As Yale professor and Nobel Laureate Robert Shiller noted, the reality is that the bond market does a better job of telling you what has happened recently, and a rather poor job of giving any indication of what is to happen:
[B]ond markets act as if they think inflation can be extrapolated. Long-term interest rates tend to be high when the last decade’s inflation was high. US long-term bond yields, such as the ten-year Treasury yield, are highly positively correlated (70% since 1913) with the previous ten years’ inflation. But the correlation between the Treasury yield and the inflation rate over the next ten years is only 28%.
Given little margin for error, then, bond buyers betting on more stagnation should inform their opinions with this evidence and ask themselves whether they are really predicting the future, or just suffering from the recency bias of a strong bond bull market.
The process of assessing what bets one is actually making is no different in the equity market. The math of equity valuations is not as straightforward as with bonds, and the inherent riskiness of equities (of which opportunity cost is an underappreciated risk), proper assessment is crucial.
Consider, for example, the case of portfolio manager John Hussman, of whom I have written extensively before. There is no need to revisit my criticisms of him here, but a bit of a post mortem on his fund’s terrible performance is helpful in understanding my broader point. For years now, Mr. Hussman has been citing rich valuations as a reason to bet against equity gains, and his being continuously wrong has not just cost his investors the opportunity cost of missing out on bull market gains, it has also cost them capital as his bets against the market have cost his shareholders their capital.
What is useful in studying his performance, however, is that it can be argued that while he may actually be right in terms of stock valuations being rich, his analysis seems to have gone only skin-deep, leaving his framework bereft of any context that might have tilted his bets another way. For example, Mr. Hussman either did not consider or he ignored the fact that valuations, in order to be useful guides, cannot be considered in a vacuum, and must be viewed in light of prevailing inflationary and economic regimes. Adjusting for these factors would have told him that valuations – while looking rich in absolute terms – have until only very recently been quite normal. This is all the more puzzling given that a 2010 post on his site discussed just this very thing. So, while investors in Mr. Hussman’s fund were placing a bet on mean reversion for valuations, they failed to consider that stocks do not rise simply because they are cheap, and they do not drop because they are expensive. Rather markets typically fall because of exogenous events such as rapid inflation or disinflation, or recessions and expansions, making mean reversion obvious only in hindsight. Finally, the magnitude of the mean reversion for which Mr. Hussman positioned his portfolio, in order to pay off, seems to have required either a return to either severe inflationary conditions, or a descent into a prolonged period of deflation, both of which were low-probability outcomes. It seems unlikely that investors in his fund considered that this was the bet they were really making when they invested in it.
It is not just Mr. Hussman who is guilty of faulty analysis. As I have written elsewhere, over the duration of the current bull market, many allocators, lured by the superficial cheapness of many foreign markets, have failed to appreciate the material differences in global stock market composition, and thus have been frustrated (until the last year or so) as their globally-tilted portfolios – which were in reality just a bet on cyclical stocks – have underperformed markets with a tilt to secular growth industries, such as the US, which tend to enjoy premium valuations because of their growth-orientation. Similarly, any US investors who gave up on foreign investments after a period of dismal returns and so went all-in on domestic stocks have underperformed recently, not least they failed to appreciate the impact of currency on global equity market bets, and so abandoned a good hedge against a weakening dollar.
In sum, in today’s challenging environment for asset allocators it is critical for investors to assess fully the nature of the bets they are taking and to move beyond superficial analysis for a better appreciation of the myriad risks and possibilities poised to affect their portfolios.