Harvard Business Review’s new research published in the Journal of Banking and Finance offers data-driven evidence that gender diversity can improve the accuracy of high-stakes decision-making.

In a study of more than 63,000 stock price forecasts issued by over 3,400 analysts at 364 investment banks, researchers found that teams with more women consistently showed lower optimism bias, especially when analysts were under pressure to inflate projections. 

Optimism bias, or the tendency to overstate positive outcomes, is a known issue in finance, particularly when firms have vested interest in the companies being evaluated. The study compared analysts’ one-year target prices to actual outcomes and consensus estimates, and then analyzed the effect of team composition on forecasting accuracy. 

The results showed that increasing the share of women on a team by just 8 percentage points was associated with up to a 12 percent reduction in optimism bias. Importantly, the effect wasn’t linked to women being more cautious individually. Instead, mixed-gender teams saw a shift in group behavior overall. Both men and women issued more realistic and less inflated projections when working in diverse environments. 

To rule out other explanations, the study tracked analysts who moved from less diverse teams to more diverse ones. Their forecasting bias decreased after the move, reinforcing the conclusion that it was team composition, not individual ethics or firm culture, driving the change.