How much do equity investors leave on the table by focusing on categories of stocks (industry or style) and paying little attention to individual stocks? In her July 2013 paper entitled “Categorical Thinking in Portfolio Choice”, Swasti Gupta-Mukherjee investigates whether such simplifying categorical thinking is economically beneficial and what factors magnify or diminish it. She focuses on managers of U.S. equity mutual funds. She defines a Categorical Thinking Index (CTI) as relative emphasis on category-wide information relative to stock-specific information for making portfolio adjustments. She implements CTI based on two-digit SIC codes (broad industries) as categories. Each quarter, she measures CTI for a mutual fund based on the explanatory power of past industry returns for portfolio changes relative to the explanatory power of past individual stock returns for these changes. Using monthly returns, quarterly holdings and style (benchmark) representations for 2,812 U.S. equity mutual funds, along with associated industry and individual stock returns, during 1990 to 2011, she finds that:
- Categorical thinking as defined and measured varies considerably across funds, indicating the potential for meaningful explanation of fund performance.
- In aggregate, categorical thinking tends to be high when investor sentiment is extreme and to increase with market uncertainty (as measured by VIX), asset-level information uncertainty (as for growth stocks) and level of portfolio diversification across industries and styles.
- Categorical thinking tends to be stronger among fund managers who head contrarian and more expensive funds, have longer tenure and rely more on public information in the form of analyst forecasts.
- Portfolio managers who exhibit strong categorical thinking significantly underperform those who exhibit weak categorical thinking.
- The funds with the lowest tenth of CTI outperform those with the highest tenth by a net annual average of 3.7%, with average net annual four-factor (market, size, book-to-market ratio, momentum) alpha 2.2%.
- Outperformance of low-CTI managers is attributable largely to stock picking and not style timing.
In summary, evidence suggests that U.S. equity investors who focus strongly on industries/styles and weakly on individual stocks underperform those with the opposite approach by roughly 4% net per year.
Cautions regarding findings include:
- The definition of categorical thinking is abstract, and its calculation numerically intensive. Investors therefore may consider findings more relevant to development of their own investment approaches rather than selection of mutual funds.
- The study uses equities and equity categories (mostly industries and somewhat styles) to investigate categorical thinking. Results may differ when applied across asset classes.
See also “Stock Picking or Industry Picking?” and “Sector Rotation vs. Stock Picking”.
