In its chapters on investing, Nudge puzzles over the “home bias puzzle,” in which investors in a given country tend to overweight their portfolios with stocks from that country. So for instance, although U.S. equities make up less than half of the global stock market, most U.S. investors’ portfolios are dominated by them. This kind of geographical proximity in investing is often explained by differences in regulation, culture, and taxation between nations, as well as differences in understanding about home versus foreign companies. These frictions can occur within nations as well, according to Tobias Moscowitz (of Chicago’s Booth School of Business) and Joshua D. Coval, leading to what might be called the “hometown bias puzzle.”
Using a unique database of mutual fund managers and company locations, identified by latitude and longitude, we find that the average U.S. fund manager invests in companies that are between 160 to 184 kilometers, or 9 to 11 percent, closer to her than the average firm she could have held. Alternatively, one out of every ten companies in a fund manager’s portfolio is chosen because it is located in the same city as the manager. With a variety of measures used, the null hypothesis of no local equity preference (or local bias) is consistently rejected, demonstrating that the distance between investors and potential investments is a key determinant of U.S. investment manager portfolio choice.
Why the hometown bias? Familiarity and understanding.
We find that local equity preference is strongly related to three firm characteristics: firm size, leverage, and output tradability. Specifically, locally held firms tend to be small, highly-levered, and produce goods not traded internationally. These results suggest an information-based explanation for local equity preference, since small, highly levered firms, whose products are primarily consumed locally, are exactly those firms where one would expect local investors to have easier access to information and are firms in which such information would be most valuable.
A pdf of the working paper is here.