Diversify, Diversify, Diversify(?)

In their February 2005 paper entitled “Portfolio Concentration and the Performance of Individual Investors,” Zoran Ivkovic, Clemens Sialm and Scott Weisbenner test the dictum for diversification by examining the stock trades of a large number of individuals during 1991-1996 through a discount broker. They find that:

• Consistent with prior research, on average across the entire sample, stocks bought by individual investors significantly underperform stocks sold. In other words, they consistently lose to the “pros.” However, individuals with large, concentrated portfolios generally make good trades.
• Concentrated accounts outperform diversified accounts, especially among investors with large portfolios. The less diversified the account, the better the outperformance.
• The greatest excess returns of concentrated portfolios are associated with stocks that are geographically local and are not in the S&P 500 index, suggesting the advantage of private information.
• The concentrated portfolios carry higher levels of risk and lower Sharpe ratios than the diversified portfolios.

In summary, some risk-tolerant investors successfully exploit significant informational advantages by concentrating their portfolios in a few stocks that they know well.