Most people who have studied investing believe that they should possess the market portfolio. With their U.S. stocks, they shouldn’t pick and choose; rather, they should own as many positions as possible, presumably through broad index funds. Ditto for their bonds as well as for their overseas holdings. This belief comes courtesy of Professor William Sharpe, who won a Nobel Prize for his troubles.
Of course, most investors don’t follow this precept precisely. They don’t hold a single broad-market index for each asset class. Nonetheless, if they invest through funds rather than through stocks directly, investors will end up owning something that approximates Sharpe’s recommendation. Their portfolios will fluctuate around those benchmarks, but their results shouldn’t stray too far from the mark.
As my colleague Paul Kaplan (to whom this column owes a debt) reminds me, this faith in the market portfolio deserves some rocking. Critically, Sharpe’s conclusion when developing the Capital Asset Pricing Model—that the market portfolio was the single best portfolio for all investors—depends upon the assumption that investors can and will borrow to leverage their portfolios. In addition, the interest rate for their borrowing costs must be the risk-free rate.
In practice, of course, most investors—even among institutions—don’t borrow, and if they do attempt such a thing, they are not granted the risk-free rate. This has a dire effect on the CAPM. Writes fellow Nobel Laureate Harry Markowitz (who received his award on the same day as Professor Sharpe), if we “take into account the fact that investors have limited borrowing capacity, then it no longer follows that the market portfolio is efficient.”
In fact, continues Markowitz, “This inefficiency of the market portfolio could be substantial and it would not be arbitraged away even if some investors could borrow without limit.” He then defends that statement formally, for those who enjoy such treatments. Suffice it to say that while the equations are past the scope of this column, the upshot is not: Markowitz was correct on the math. His results have not been challenged.
Markowitz draws a straightforward investment conclusion. In the theoretical world of the CAPM, investors achieve extra return by borrowing to buy additional equities. That is, they increase their portfolios’ betas by putting more money to work at a beta of 1.0 (the market portfolio), rather than investing the same amount of money in stocks that have higher betas. In the actual world, not so much.
That is because without borrowing, only the latter strategy is possible. Those who accept the CAPM’s conclusion that beta alone determines a stock’s expected future returns, and who are willing to accept above-market risk in exchange for potentially above-market gains, must purchase higher-beta equities. No ifs, ands, or buts. That is their only choice. Such investors must reject the market portfolio by doubling down on volatile stocks and skipping the tame ones.
The problem is, such behavior introduces pricing distortions. If enough investors adopt this mindset—a likely occurrence, given how many assets are held either by institutions that have very long time horizons, or individuals who have decades of expected life remaining—then higher-risk stocks will become overbought. Too much money will crowd into the same subset of securities, which will reduce their expected returns. Conversely, the low-beta stocks will be relatively neglected, and will outdo their forecasts.
Those results have indeed occurred since the CAPM was introduced, leading some to speculate that Markowitz got it right, even if he didn’t form his argument as a prediction. Readers appreciated the CAPM’s logic, which earned Sharpe fame, but the publication of the theory didn’t much change their habits. For the most part, those who sought greater returns continued not to leverage their portfolios, and continued to seek extra profits by purchasing riskier stocks. The stock market acted largely as it always did.
The effect of restricting leverage doesn’t stop there. For every pricing distortion, the financial markets have a potential counter-distortion—a reaction against the original movement. That is, if higher-beta stocks did indeed become overpriced due to excess demand and lower-beta stocks underpriced (only a hypothesis—this is not a topic that lends itself to proof), and investors perceive that pattern, then stock buyers may change their collective behavior. Perhaps they will flock to low-beta stocks.
Which then might cause further reverberations. None of which matter for the purposes of this discussion. The point is, once the door is opened such that some participants should refuse to hold the market portfolio, then all manner of beasties can slip into the room. The theory does not permit the halfway—either it holds, or it doesn’t hold. And it doesn’t hold.
None of which, of course, demonstrates that buying the market portfolio is a mistake, or that thinking of such a position as a “neutral” starting point is wrong. The market portfolio may well be an appropriate choice, and it surely is the most sensible of starting points. The prospective investor in U.S. stocks has several thousand publicly traded alternatives. Absent further information, the most-logical portfolio is that which owns them all, in proportion to their worth.
However, there is further information. In addition to Markowitz’s observation, which affects all market participants (the pricing of high-beta securities being independent of one’s individual circumstances), there are a host of personal factors that could lead the logical investor to deviate from the market portfolio. I will cover those in Tuesday’s column.