Though the causes of the 2008 financial crisis are multifaceted, many believe that it was driven in part by a gendered culture of risky trading behaviour. The authors examine if there is any credence to this claim by asking whether men and women behave differently in experimental asset markets. They find that as the number of women in a trading market increases, the magnitude of price bubbles decreases. In addition, women’s price forecasts are lower compared to men’s.
This study tests the assertion that men are more likely than women to engage in risky financial trades that result in pricing bubbles. The authors were motivated by the 2008 financial crisis and the suggestion that it was partly attributed to the overrepresentation of men on the trading floor, and their tendency to take irresponsible risks. They use an experimental research design that simulates a market for long-lived assets, which are assets such as real estate that a business expects to retain for at least one year.
Simulating a market for long-lived assets provides a conservative estimate of gendered differences in risky trading since presumably trade involving long-term investments would be approached more cautiously (though, the financial crisis calls this assumption into question).
In addition, because women who work in finance-related fields may have adopted traditionally masculine attributes, such as aggressively competitive behaviour or a tendency to engage in risky trades, an experimental design helps distinguish the relationship between trading practices and gender from the masculinized work cultures found in actual financial markets.
An experimental design helps distinguish the relationship between trading practices and gender from the masculinized work cultures found in actual financial markets.
The authors recruited students to participate in a series of experimental trades. Students were placed in all-male, all-female, or mixed-gender trading groups. The authors were particularly interested in whether all-male market trading was more likely to result in asset bubbles, whereby asset trading occurs at prices that deviate significantly (either above or below) from the assets’ true market value.
At the start of each trading session, traders were given a predetermined number of assets. The assets were shares of stock in a company that traders could choose to sell. In addition, traders were given a predetermined amount of cash to buy more assets. In each session, participants traded 18 assets during a sequence of 15 double-auction trading periods. A double auction is where potential buyers of assets submit bids at the same time that potential sellers of assets submit their asking prices. An auctioneer then decides on a fixed price point. All sellers who were asking for less money than the auctioneer’s chosen price point may then sell. And all buyers who submitted higher bids may then buy at the lesser price point.
The authors also asked participants in mixed-gender trading groups to forecast average asset prices prior to each trading period. This allowed them to rule out the possibility that trading behaviour is driven by gendered beliefs about how markets operate. They did not find any consistent pattern of differences in the average forecasts of men and women, making it less likely that differences in forecasts were the source of the price patterns observed in the single-gender trading groups.
The experiment revealed four main findings. First, bubbles in all-male trading groups occurred more frequently than in all-female trading groups. Second, the all-male trading groups overpriced assets to a greater extent, meaning that the bubbles in the all-female markets were smaller in comparison. Third, the bubbles in 67% of all-female markets were negative, meaning that the all-female groups more often traded assets at prices that were lower than their true market value.
Lastly, in their market simulation using mixed-gender groups, the study found that, as with the all-female trading groups, increasing the number of women in the trading group resulted in smaller bubbles. Compared to the all-male markets, bubbles were more likely to be the result of underpricing rather than overpricing and overpricing occurred less frequently overall.
The study shows that it is not necessarily that women are more risk averse but that men may be overly risk seeking, especially in contexts that are reinforced by male peers on the team.
In summary, the study found that all-male markets produced price bubbles, while all-female markets produced prices that were below the assets’ true market value. These results are corroborated in the experiments using mixed-gender groups, since a higher frequency of women in the trading groups reduced the magnitude of pricing bubbles. The findings support the idea that financial markets might operate differently with higher rates of female trading participation.
- Increase diversity – Part of the reason that masculinized attributes, such as hostile competition and risky trading behaviour, are so pervasive in finance is because there are more male traders than female traders. Trading firms would benefit from increasing the diversity of their trading teams. This will require them to decrease the entry barriers for women and increase the inclusiveness of the working culture. Doing so may help to moderate the effect of masculinized work cultures that foster needlessly dangerous competition and risky trading.
- Re-direct funding – Investors can direct their funds to investment houses that have greater diversity in their trading teams. This will not only lead to more responsible investing (avoiding participation in bubbles), but may also be an incentive to firms with low diversity to change the mix of their traders.
Thar SHE Blows? Gender,
Competition, and Bubbles
in Experimental Asset Markets
Catherine C. Eckel,
Sascha C. Füllbrunn
Texas A&M University,
American Economic Review