Do You Really Believe in Diversification?

I was talking to an advisor recently (“Joe”) who was considering the replacement of a manager for what he claimed was poor performance.  Based upon my inquiries, however, it became clear that there was no problem with relative performance or with risk-adjusted returns.  Nor was there any problem with “mission creep” either.  The manager was doing what he was supposed to be doing in the way he was supposed to be doing it.  Joe was simply dissatisfied with the returns that were being generated on a nominal basis.

Most fundamentally, what this conversation revealed was that Joe neither understood nor believed in diversification.

Few advisors argue with the idea that establishing a well-diversified asset allocation plan consistent with one’s goals, investment horizon, and risk tolerances should be the first priority of most investors. Joe surely wouldn’t.  The key advantages of broad and deep diversification are the capture of a healthy share of available returns, smoother portfolio performance and lower volatility.  Especially in a secular bear market like the one we have been suffering through since 2000, those are worthy goals.  Unfortunately, either Joe doesn’t believe what he claims or doesn’t understand what he believes — perhaps both.

As I have pointed out before, the total return of any portfolio consists of three components, each of which may be positive or negative: (a) returns from overall market movement; (b) incremental returns due to asset allocation; and (c) returns due to timing, selection, and fees (active management). The latest research suggests that, in general, about three-quarters of a typical portfolio’s variation in returns comes from market movement (a), with the remaining portion split roughly evenly between the specific asset allocation (b) and active management (c). To the extent that research differs from that stated above, it concludes that asset allocation is more important and active management is less important.  The exercise of allocating funds among various investment vehicles and asset classes — providing diversification — is at the heart of investment management.

The theory supporting diversification is simple: Don’t put all of your eggs in one basket. A single holding has huge potential for gains if the right instrument is selected, but even bigger risks (because investing “home runs” are so hard to come by).  In general, the greater a portfolio’s diversification is, the lower its risk.  Lower risk is a good thing, but only if the portfolio’s potential return is healthy enough to meet the client’s needs.  Fortunately, a well-diversified portfolio captures most of the potential upside available with much lower volatility, thus providing higher risk-adjusted returns.

A diverse portfolio ensures that at least some of a portfolio’s investments will be invested in the market’s stronger sectors at any given time – regardless of what’s hot and what’s not and irrespective of the economic climate. At the same time, a diverse portfolio will never be fully invested in the year’s losers. For example, according to Morningstar Direct, about 25% of U.S. listed stocks lost at least 75% of their value in 2008 but only four of over 6,600 open-ended mutual funds lost more than 75% of their value that year. Thus a diversified approach provides smoother returns over time (even if not as smooth as desired!).

Accordingly, portfolio diversification requires careful management of the correlations between and among the asset returns and the liability returns, issues internal to the portfolio (the volatility of specific holdings in the portfolio), and cross-correlations among these returns. With day-to-day volatility very high and correlations remarkably high too, diversification is now more valuable than ever.

The table immediately above will be familiar to most of you.  It shows the annual returns of various asset classes over the past 20 years.  Most people (and, sadly, most advisors – like Joe) pick investments based upon what has been “hot” — as if looking in the rearview mirror while driving.  If, from 1991-2010, one had invested in the previous year’s top performer, s/he would have received 3.88% average annual returns.  But since investing tends to be mean reverting, a smart contrarian who invested in the previous year’s worst performer would have averaged returns of 10.91%. Even so, an investor who created a more diversified (if not optimal) portfolio of 45% domestic large cap stocks, 10% domestic small cap stocks, 10% international stocks and 35% aggregate bonds would have seen average annual returns similar to those earned by the contrarian (9.66%) but much lower volatility (12.61% versus 21.32% for the contrarian). 

The key advantages of broad and deep diversification, then, are the capture of a healthy share of available returns, smoother portfolio performance and thus less volatility.  Especially in a secular bear market like the one we have been suffering through since 2000, those are worthy goals.

But let’s get back to Joe.  Just as a diversified portfolio ensures that at least some of a portfolio’s investments will be invested in the market’s stronger sectors at any given time, that portfolio will also – of necessity – be invested in some of the market’s weaker sectors too.

We all like the idea that it’s possible to forecast what sectors will do well and allocate assets to them and away from those that won’t do well – market-timing in other words.  The idea is that we don’t need to diversify if we just learn to predict what the market will do next week, next month, and so on.  Unfortunately, there is little evidence that, as a general matter, it can be done successfully (for example, see here, here, here and here, momentum investing notwithstanding).

As reported by Bespoke, Bloomberg surveys market strategists on a weekly basis, and along with asking them for their year-end S&P 500 price targets, Bloomberg also asks for their recommended portfolio weightings of stocks, bonds and cash.  These alleged experts are generally wrong, often spectacularly so.  The peak recommended stock weighting came just after the peak of the internet bubble in early 2001 while the lowest recommended weighting came just after the lows of the 2008-2009 financial crisis. Can anyone say “recency bias”? As John Kenneth Galbraith famously pointed out, we have two classes of forecasters: those who don’t know and those who don’t know they don’t know. 

Joe wants to fire a good manager whose sector is underperforming.  He is thus making the classic mistake of “driving” his investment analysis while looking in the rearview mirror. Don’t do that! 

The next time you are tempted to fire a manager on account of poor sector performance, remind yourself that a portfolio without assets that aren’t performing well on a nominal basis is not diversified and that diversification is a very good thing indeed.