Do Finance Professors Invest Like Everyone Else?

Abstract

Comparing the results of the Fed’s Survey of Consumer Finances with those of a survey of finance professors at U.S. universities, the authors found that finance professors are significantly more likely than others to invest in equities. They also found that finance professors are less prone to behavioral biases because their decision not to invest in equities is based on neither the outcome of their past investments nor their short-term expectations of the market.

See comments and response on this article.

Standard rational economic models predict that in the presence of a positive risk premium, all investors will hold some equity. However, there is evidence in the finance literature that a number of households in the United States hold no equity. Results from the Fed’s Survey of Consumer Finances (SCF) show that even within the top quintile of income distribution, a significant number of households do not invest in equities. In our study, we examined the investment pattern of finance professors to investigate whether they participate in the stock market to a greater extent than the general public.

Finance theory suggests that in order to achieve optimal diversification, a portion of every portfolio should be invested in equities. Because finance professors are advocates of traditional finance theory on equity market participation, we would expect all finance professors to invest in equities. To test whether this group of experts practices what it preaches, we surveyed finance professors at universities across the United States. During the summer of 2007, using the University of Texas at Austin list of all regionally accredited universities, we manually collected the names and e-mail addresses of finance professors at these institutions. We used a questionnaire to collect data on actual portfolio holdings and demographics from each finance professor selected. We investigated whether finance professors, compared with households in the Fed’s 2007 SCF sample, are significantly more likely to invest in equities. Because our investigation focused on individuals who provided detailed information on their financial asset holdings, we included 4,160 respondents from the SCF sample and 1,368 respondents from the finance faculty sample.

Unsurprisingly, we found that finance professors participate in the stock market to a greater extent than do members of the SCF sample. However, our counterintuitive finding is that a significant number of the finance professors do not participate in the stock market. Arguably, those finance professors who choose not to hold stocks are aware of the “rational” arguments for investing in stocks. Our findings thus raise the question of whether it is an oversimplification to suggest that not holding stocks is an investment “mistake.” Therefore, advising those less knowledgeable in finance always to hold stocks may not be beneficial.

We also investigated whether some behavioral biases related to nonparticipation in the stock market are present in our sample of finance professors. Researchers have broadly characterized these biases as overconfidence and considering the past. An investor’s overconfidence in the future prospects of an investment can lead to an increased affinity for risk taking. With respect to the bias of considering the past, two alternative manifestations have been popularized: (1) Investors are willing to take on more risk after experiencing a gain because they believe they are using the house’s money, and (2) investors are willing to take on more risk after experiencing a loss because they are trying to recoup their prior losses (i.e., break even).

We first examined whether finance professors are so confident in their own stock market predictions that they will invest in equities only if they expect a bull market in the short term. We then investigated whether our finance professors consider their past investment outcomes in deciding whether to hold equities and are thus prone to either the house-money effect or the break-even effect. Specifically, we tested whether members of this group decide whether to invest in equities on the basis of either their short-term market predictions or the outcomes of their past investments. We found no support for overconfidence (in predictions), the house-money effect, or the break-even effect. Taken together, these results suggest that our sample of finance professors is less prone to certain behavioral biases than the general public.