It’s Time to Fix Your Undiversified Portfolio

t’s time to fix your undiversified portfolio with some history, data, and strategy-of course.

Modern Portfolio Theory (MPT) is the brilliant mathematical model behind the investment concept of diversification. Nobel Prize winner Dr. Harry Markowitz introduced the theory in 1952 (1), and forever changed the world of finance and investing.

It’s Time to Fix Your Undiversified Portfolio

He proved mathematically why holding a bundle of non-perfectly correlated assets is superior to holding a few concentrated positions. He showed that by holding an appropriately diversified portfolio of assets, investors could maximize returns for any given level of risk.

I don’t want to get too far into theory, because most people just don’t care. And I don’t want to address the underlying assumptions of MPT which have been questioned in recent years. Those things don’t matter to most investors. What matters is minimizing investment risk and maximizing returns.

Any primer on Modern Portfolio Theory must begin with a recognition that investors do not like risk and need to be compensated for bearing it. Investment risk is often defined as the variation in possible returns from the average.

Higher risk investments carry a wider range of possible outcomes, but also carry higher expected returns, compensating investors for withstanding the volatility. In contrast, investments that have relatively stable returns are expected to produce lower returns.

Historical data strongly supports this assumption. For example, from 1926 to 2011 the average geometric return on U.S. Treasury Bills was 3.6%. Over the same period the average return on large domestic stocks was 9.8%, while small domestic stocks returned 11.2%. This additional return is also called the equity risk premium, and is the result of the increased risk (more ups and downs) accompanying stock ownership.

Perhaps more importantly, Modern Portfolio Theory showed that not all investment risks are rewarded with higher returns. The total risk of any individual security consists of two uncorrelated components: the market, or systematic risk and the firm-specific, or idiosyncratic risk.

The entire capital market and all investors therein are exposed to unavoidable systematic risks, such as currency problems, fluctuating interest rates, war, recession, inflation, and government interventions. You can’t get away from these, but you do get compensated for bearing the risk.

The idiosyncratic component simply reflects the reality that each company is exposed to a unique set of risks specific to its own business and financial situation. Idiosyncratic risk can include competition, lawsuits, fraud, bad management, financial constraints, etc. This type of risk is not rewarded, because it can be eliminated through proper diversification.

In broadly diversified indexes such as the Wilshire 5000 and the Lehman Brothers Aggregate Bond Index, idiosyncratic risk approaches zero. Investors whose portfolios are based on such broad indexes are subject only to systematic risks. By contrast, a portfolio made up of just a handful of securities, no matter how carefully researched, is subject to high levels of idiosyncratic risk. This is why stock pickers and sector investors almost always hold inefficient portfolios that underperform index investors on a risk adjusted basis.


We can observe the average investor’s inclination to hold an undiversified portfolio quite easily in the research that has been conducted on the topic.

One excellent study on the topic,” Equity Portfolio Diversification” (2), was performed by Dr’s William Goetzmann and Alok Kumar. The authors analyzed the portfolio decisions of more than 60,000 individual investors at a large U.S. discount brokerage house from 1991 through 1996. In general, most investors were grossly under-diversified.

An average investor holds a four-stock portfolio (median is three).

Over-all, the evidence indicates that most portfolios have significantly higher volatility levels relative to the market portfolio, and investors are not compensated for their higher risk exposures.

In addition to widespread under-diversification, the authors find evidence to further support my argument against stock picking. If you read the arguments of many individual stock pickers, you’ll notice a very strong (and foolish) belief that they are somehow superior to other investors. Even though there is absolutely zero evidence supporting the idea that individuals can outperform the market over the long haul after accounting for taxes and fees, they’ll argue about it. But there isn’t an argument. It’s fact versus myth.

Investors whose trading decisions are consistent with stronger behavioral biases exhibit greater under-diversification. Furthermore, investors who overweight certain specific industries or stock characteristics such as volatility and skewness are less diversified.

We also observe that less diversified investors trade more frequently and pay considerable transaction costs. The average annual trading cost for investors in our sample is 1.46% of their annual income. Using the brokerage data, Barber and Odean (2001) estimate that the average trading cost of active investors is 3.90% of their annual income. These transaction cost estimates indicate that investors in our sample pay considerable transaction costs but still fail to diversify appropriately.

What comes next should not surprise you. The investors who held the least diversified portfolios earned far less than the most diversified group of investors.

The unexpectedly high idiosyncratic risk in investor portfolios results in a welfare loss as measured by the Sharpe ratio of individual portfolios. This evidence in itself is not very surprising. More surprising is our finding of significant differences in the portfolio alphas.

The least diversified (lowest decile) group of investors earns 2.40% lower return annually than the most diversified group (highest decile) of investors on a risk-adjusted basis. The economic cost of under-diversification is higher for the group of older investors, where the risk-adjusted performance differential between the least diversified and the most diversified investors is 3.12%

Because less diversified investors trade more frequently, these performance estimates indicate that the net returns earned by under-diversified investors are likely to be even lower. Consequently, the net performance differential between the least diversified and the most diversified investor groups is likely to be higher.

The moral of the story is this: a well diversified portfolio tends to outperform a portfolio constructed by picking stocks. In other words, own index ETFs and stop trying to buy individual stocks. Or as stated by the authors:

Most investors could have improved the performance of their portfolios by simply investing in one of the many available passive index funds.

How You Can Properly Diversify

Many investors have the wrong idea when it comes to diversification. I commonly hear people talk about buying 10-20 dividend paying, domestic blue chip stocks as a form of proper diversification. This is foolish. As is holding a portfolio comprised of 100 technology companies, or 5000 health care related companies.