June 21, 2010
A DEATH EXAGGERATED
Is modern portfolio theory (MPT) dead? Yes, according to many pundits and strategists. There have always been skeptics of modern finance, although membership in this club has risen sharply in recent years, thanks to the surge in market volatility and the steep losses posted by the major asset classes during late-2008 and early 2009. A popular argument is that multi-asset class diversification didn’t spare investors from unusually big declines, ergo, MPT failed. By that standard, the case for abandoning conventional asset pricing theory looks compelling. There’s just one problem: It’s wrong.
MPT never promised to limit losses, short-term or otherwise. Yet it’s become popular to assume that modern finance will offer investors a silver bullet by delivering tidy returns for little or no risk. No theory or trading system can assure that, but somehow those unreasonable expectations have crept into thinking about MPT.
Untangling the misinformation about standard finance is complicated, a topic worthy of a book and ongoing analysis when managing money. Meantime, let’s consider one piece of the misinformation pie: MPT failed to save investors from the hefty losses in late-2008 and early 2009.
That sounds like a reasonable complaint, but a careful reading of MPT reminds that no such promise was ever made. In fact, MPT is a body of research that provides an investing framework that offers to minimize if not eliminate one type of risk: idiosyncratic risk. This is the risk tied to picking individual securities. That risk can be harnessed for boosting return, of course. But it’s also the source of unusually big losses at times. The Warren Buffetts of the world needn’t worry about idiosyncratic risk, but for the rest of us it’s rash to dismiss it out of hand.
In the long run, idiosyncratic risk isn’t likely to generate a risk premium, at least not for the average investor. In the short run, of course, this risk can be harnessed for profit, but that’s a lot harder than it sounds, especially after adjusting for taxes and trading costs, as numerous studies remind. That not-so-subtle implication: avoid idiosyncratic risk. What you're left with is the market risk, which does generate a risk premium over time. A cost-effective way to earn a market risk premium is with conventionally designed index funds and ETFs. And if we consider a formal interpretation of MPT, the concept should be applied across all the major asset classes. In short, a simple, investable proxy of the “market” portfolio should be hold stocks, bonds, REITs, commodities weighted by relative values of those markets. Over time, this benchmark will earn a modest risk premium, according to MPT. Here’s one way to slice and dice the market portfolio, although minds will differ as to the best approach for defining the crucial components.
What do you get when you buy the market portfolio? Part of the answer is emphasizing what you don’t get: idiosyncratic risk. That leaves us with systematic risk, or the market’s beta risk. MPT forecasts that the market portfolio is the best mix of risky assets in risk-adjusted terms for the average investor over the long haul. In fact, that seems to be the case, based on my analysis of markets as published in The Beta Investment Report. Numbers don't lie. Alas, this information isn't widely published, certainly not in the usual suspects of financial journalism. Instead, the media likes to look at, say, the S&P 500 by itself over the last year or two, or even ten, and draw sweeping conclusions about MPT based solely on that massively misinformed perspective.
And since even properly defined systematic risk can be volatile in the short run, a naive review of MPT convinces many that the theory's worthless. But there is a grain of truth here. If you’re investment horizon is measured in days, weeks or even months, MPT’s not going to help you. No one who understands MPT ever claimed otherwise.
But over the medium- and long run periods, MPT’s still quite valuable. It’s still no short cut to big gains with little or no risk. But as a model for helping us think about designing and managing portfolios, MPT is still useful, if not essential. Yes, it’s an evolving theory, as I explain in my book, Dynamic Asset Allocation: Modern Portfolio Theory Updated for the Smart Investor. Financial economists have uncovered a number of risk factors beyond the broadly defined market beta that offer opportunities for enhancing the market portfolio’s risk-adjusted performance. On the surface, these opportunities look like evidence of MPT failures, although the risk-based explanations are compelling counterpoints.
One quick example is related to the research that shows that the market beta doesn’t fully account for risk and return. But the smoking guns on this front aren’t as cut and dry as the critics would have you believe. For instance, some researchers make a case for dismissing market beta as a robust tool for explaining return. A number of these studies are based on crunching the numbers on a monthly basis. But analyzing the connection between risk and return via beta over longer time frames shows a stronger linkage—strong enough to give pause before throwing out MPT entirely.
The conventional interpretation of MPT that’s widely embraced in the media is based on finance research through the mid-1960s. By that standard, buying and holding the market portfolio and letting it ride is the embedded wisdom. But research over the last several decades tell us that risk and return are more complicated, which implies doing something other than holding the unmanaged market portfolio. In the long run, the broad market portfolio is still likely to perform as theory predicts and generate middling to slightly above middling returns for relatively little risk compared with the various efforts to beat this index.
At the same time, let's recognize that MPT has evolved, even if it’s easier to claim that this theory is dead. What does an updated view of MPT tell us to do? In the coming days and weeks, I’ll take a closer look at some examples.
Posted by jp at June 21, 2010 10:42 AM
I agree that some advisors/consultants/managers abuse MPT for commercial gain. Of course, that's true of every other investing strategy, so we're back to square one.
I'm all for doing your own research, but if you'd diversified across the major asset classes over the past decade or two, you'd probably outperformed half of the "professionals," perhaps a bit more, after taxes and trading costs. As to the question of why put your money into a "losing sector," well, that's a loaded question. No one knowingly invests in losing securities/sectors/asset classes. The problem is that no one really knows what's going to be losing ex ante. Sure, we have some clues about future returns/risk, and so maybe it's a good idea to customize. But there's a limit to this game, as the bias toward middling historical investment returns (or worse) attest.
Posted by: JP at June 21, 2010 8:17 PM
MPT is simply a way for fund managers to convince prospective investors that they should hand over their money and it will be safe and sound...except when there is trouble ala 2008. Since so many managers tout MPT, then during market routs, they all lose and they can simply say everyone is losing.
Bottom line: There is no substitute for doing your own research and making your own decisions. (why would anyone put money into a losing sector just to be "diversified"?)