The Results Are In: Diversification Works

So far we’ve covered the reasons why diversification is essential to sound portfolio management. Now that we’ve graduated from residency, it’s time to put all of this into practice.


Click here or the image to view the high-res pdf graphs that go along with this article.
So far we’ve covered the reasons why diversification is essential to sound portfolio management. Now that we’ve graduated from residency, it’s time to put all of this into practice.

Let’s look at the 30-year period from 1979-2008. This period includes multiple bull markets, multiple bear markets, recessions, expansions, and major political events. I’ve created 5 different portfolios, with each one adding another asset class (see figure 1).


Over the 30-year period, if you had invested only in US stocks (Portfolio 1), you would have had the highest returns about one third of the time, the highest of any portfolio. However, by adding other asset classes within the global equity markets (Portfolio 4) you would have increased your return above the single asset class portfolio. Further by adding bonds (Portfolio 5) you would have reduced both your average return and your highest calendar year return, but still achieved the same total amount of wealth over the 30-year period. As I’ll discuss below, this is because Portfolio 5 had the least fluctuation in returns. Looking at the ten year period 1999-2008 which included two ravenous bear markets, only Portfolio #5 had a respectable annual return of 3.3%, but the other portfolios had returns less than 3% a year and two of them actually had an overall negative return for the decade. As an investor if you are not broadly diversified, are you willing to accept losses for 10 years? I think this would test the patience of just about every investor out there, including myself.

Risk Business

We know that returns come from risk. While there are several measures of risk, most individual investors are concerned with losses. In terms of losses, the most diversified portfolio (Portfolio 5) really shines in several ways: its worst loss was about 20% less than the other portfolios worst losses, it had the least number of negative years, and fewer percent of months with negative returns. In the six years when Portfolio 1 had negative returns, Portfolio 5 had about half the losses as Portfolio 1. Its performance during the two bear markets in the past decade was far better—losing less than one third as much as Portfolio 1 from 2000-2002 and about 20% less during 2008. Digging further, if you look at the variations in annual returns (another measure of risk) you find that Portfolio 4 actually had LESS risk than Portfolio 1 and HIGHER returns. Finally Portfolio 5 had about the same total return as Portfolio 1 but with almost 25% less risk.

Think about this: Portfolio #5 is a really boring investment: a significant chunk of the portfolio is invested in bonds. During a slow shift you’ll never turn to your colleague and say, “I’m so excited to tell you about the government bonds I bought last year!” But for most investors this is a trade-off you should be willing to make. Check out figure #2. You’ll see that the more broadly diversified portfolios had less fluctuation in returns than the single asset class portfolio. It’s the reduction in volatility—the cornerstone of diversification—which leads to greater wealth.


It’s important to note several limitations in this analysis. First, the time period is crucial. For example ending the analysis in 2007 would have changed the results. Portfolio 5, with its bond position, would have had the lowest return but also much lower risk. Its return per unit of risk would still be the highest. Portfolio #4 still had the highest return with lower risk than Portfolio #1. Second, the analysis assumes that you stuck with the plan and didn’t jump ship during market downturns.  The conclusions are still the same: diversification works because combining risky asset classes together creates a portfolio which is more than the sum of its parts.

Setu Mazumdar MD practices EM in Atlanta, GA and has passed the CFP® Certification Examination.




  1. How often should one rebalance the portfolio?

    My guess is for illustrative purposes your analysis rebalances on Dec 31st each year. However, if one rebalanced in March, then again in April of 2009 one would come out ahead of “holding tight” until the end of the year.

    So again, when should the portfolio be rebalanced? A passive method would likely be best to take the emotion out (which always seems to doom most investors).


  2. Setu Mazumdar, MD--President, Lotus Wealth Solutions on


    Thanks for your question.

    There are numerous academic studies looking at the question of how often a portfolio needs to be rebalanced. In the analysis I assumed rebalancing on a yearly basis at the end of every year. There is no definitive conclusion on the frequency of rebalancing. The time method of rebalancing is somewhat arbitrary. A better method is to set rebalancing ranges among different asset classes and then to rebalance when an asset class falls outside of that range. This way you don’t rebalance too frequently and thus avoid unnecessary fees and especially taxes if in a taxable account. What rebalancing does is to make sure that your overall portfolio adheres to your particular risk preferences over time. Of course this is much harder to do because most investors do not have he discipline to stick with this strategy. But there are other issues to consider as well: future cash flows, dividends and capital gains distributions, taxes, fees, and of course discipline.

    Regarding the question of rebalancing in March 2009–while it is true that rebalancing at the bottom generates higher returns, the problem is that no one can predict when the bottom is. Therefore it is best to stick with a disciplined rebalancing strategy through up and down markets. We can only know the perfect time to rebalance in retrospect. Hope this helps you.

Leave A Reply