Scientists show positive effects of affirmative action policies promoting women

The study has been published in Science.

"Many economic studies provide evidence that women tend to opt out of participating in competition even when they are equally or even better qualified than their male colleagues, " says Univ.-Prof. Matthias Sutter from the Department of Public Finance of the University of Innsbruck. The weaker inclination to enter in competition is one of the reasons why even highly qualified women are at a disadvantage on the labor market regarding promotion opportunities and subsequently wage equality, while their expertise and skills are missing in relevant positions. Companies and public institutions try to counteract by implementing various incentives and interventions. But the actual effects of these policy interventions are still unclear.

"It is nearly impossible to gather reliable field data about this topic," says Matthias Sutter, who, together with his colleague Loukas Balafoutas, conducted a laboratory experiment with 360 undergraduate students about women's and men's immediate reaction to various interventions and how their implementation affects subsequent coordination and teamwork. "Nobody before us had dealt with the question scientifically of how interventions may affect later cooperation within a team even though it is one of the main aspects of the whole discussion," underlines Sutter this innovative aspect of the study.

Attracting high-performing women to competition

In this multi-stage experiment the subjects were randomly assigned into groups of six each consisting of three men and three women. The task of the subjects was to add as many sets of five two-digit numbers as possible within a timeframe of 3 minutes. The subjects had to choose whether they wanted to be paid for each correct calculation or to receive three times as much per correct calculation if they chose the competitive scheme and were part of the two best performing competitors. Balafoutas and Sutter examined five different competition treatments: In one treatment gender was not taken into account. The second treatment corresponded to a minimum quota regulation, often used in public institutions -- there had to be at least one woman among the two winners. In the third treatment women received preferential treatment and automatically received an additional point. In the fourth treatment women were strongly favored by two additional points. Treatment 5 allowed for a repetition of the competition in the case that no woman was among the two winners.

Sutter sums up the main results of the experiment: "Without intervention the number of women willing to compete was only half the number of men. In three of the other four treatments the frequency of competing women is significantly higher, whereas there was no significant effect for men." The option of repeating the competition was the only treatment that had no significant effect on women's entry choices. The frequency of women opting to enter competition was particularly high when they were strongly favored. "The most exciting result of the experiment is that, all in all, overall performance did not suffer. Of course, we deducted the additional points from the average performance of the winners and we found that the female winners were qualified enough so that without the additional points they would have performed better or at least as well as their male colleagues," explains Sutter. "The results of this stage suggest that interventions to support women do indeed have a positive effect, in particular with regard to most highly qualified women."

Teamwork is not threatened by interventions

In the last stage of the experiment Balafoutas and Sutter examined whether the implementation of interventions to promote women results in a later discrimination when presented with a cooperation task and, thus, in decreased efficiency of team performance. The economists found that none of the interventions affected later teamwork negatively. "We set up a simple coordination game, which was about the efficient coordination of actions, for example, mutual information exchange," they explain. In this game it was completely clear to all subjects who participated in which intervention beforehand and who the winners were. "It would have been simple to discriminate through inefficient actions against someone who won because of a certain intervention," says Sutter. "The fact that this did not happen was a surprise." The scientists hope to be able to eventually confirm their results by field data from the private and the public sector. "But that seems a long way off," says Sutter.