In the last century, the situation of women in the labour market with respect to that of men has improved considerably. However, despite this narrowing of the gap, men still have better salaries, occupy more management posts than women and have a lower unemployment rate.

Faced with this situation, the question is obvious: Why do women not fare as well as men in the labour market? Furthermore, even in cases where a man and a woman occupy similar positions, there continues to be an important pay gap in most developed countries. Knowing the reasons behind these differences holds the key to being able to tackle them, for example, with adequate employment policies.

Labour economists have been assessing these differences for a long time and trying to shed light on why they occur. Frequently put forward as a factor is the differential impact that decisions on maternity/paternity have on the working lives of men and women. This is a much-studied subject and explains an important component of the gender gap in wages and employment.

However, stereotypes about women being weak at certain tasks and the overvaluation of the competitive versus the collaborative character could explain another portion of the discrimination that women face in the labour market. Factors such as gender stereotypes, the type of tasks in which they compete, who their adversaries are or prior information about their performance in a task, have an influence in the form of a disadvantage for women when competing which, in turn, may have repercussions on their situation in the jobs market.

This article shows that factors of a cultural nature exist that demand a different political approach to that of policies on maternity. They are factors more related with the roles traditionally associated with men and women and even with the way that both perceive themselves, than with labour legislation and the effective organisation of work. 

In recent years, a series of studies have appeared, not without controversy, that aim to explain the reason for this gap in the labour market through differences in men and women’s psychology and preferences.

One of the arguments common to these studies revolves around competitiveness. It is based on the fact that the labour market is an eminently competitive environment where workers strive for better positions and higher remunerations. In such an environment, the differences between men and women, according to the indications of these studies, could be explained, partly, by the fact that many women may have a lesser preference for competing. Moreover, even if they had to compete, they would be at a disadvantage because they have a lower competitive capacity. The conclusion is that these women who are less interested in competing would be less likely to fight to obtain more competitive posts. Equally, if they did worse when competing, they would obtain a lower salary in those jobs where competitiveness is especially important for determining productivity and salary.

Obviously, in order to be credible and not spur greater discrimination, statements that highlight differences between men and women in terms of preferences and capabilities must be supported by rigorous evidence. To obtain this evidence, a good option is to resort to laboratory experiments where participants are asked to carry out certain simple tasks, for which they are offered financial compensation.

In general, the results of many such experiments confirm the existence of differences between the attitudes and competitive aptitudes of women and men. Thus, when allowed to choose how they want to be paid, women frequently choose a salary that is not dependent on competition and when forced to compete, they are less productive than men. However, these experiments do not fully explain whether women choose to compete less because they feel that they are bad at it, or they compete worse because they are not accustomed to doing so.

Meanwhile, other experiments have also been conducted regarding the existence of gender differences in competitiveness with results that are not very clear, and a debate exists on whether these possible gender differences could be due more to cultural factors (which everything seems to indicate) than to genetic factors.

This study has focused precisely on understanding which cultural aspects have the greatest influence on the different competitiveness levels that men and women display. First and foremost, it is important to find out how it is possible that gender differences effectively appear in some experiments and in others they do not. It is interesting to confirm that the experiments that show the greatest differences are related with tasks about which very deep-rooted stereotypes exist, according to which men are better than women at performing these tasks. In contrast, studies where no differences were found referred to tasks about which no stereotype exists, or the stereotype is the opposite.

All this leads to the suspicion that the effects could be related with what in psychology is known as the “stereotype threat”. This says that the mere existence of a negative stereotype around a group makes the individuals of that group feel themselves subject to a scrutiny that puts them under pressure, which causes them to perform worse, thus confirming the stereotype. From this psychological concept, it is possible to pose the following questions: Do women compete worse when they do so in tasks where a stereotype exists that affirms that they are not as good as men at performing this task? What happens when the belief in stereotypes is reinforced with allusion to aspects related to gender?

To be able to clarify this issue, an experiment was conducted with university students of both genders where different tasks were proposed, some with a strong male stereotype (“men do this better than women”) and others without. For this, masculine and feminine tasks were chosen, based on evidence originating from social psychology (figure 1). For example, the “masculine” task consisted of checking whether two pairs of geometrical figures were identical or not, which requires good spatial vision, an ability considered to be fundamentally masculine (and which underpins stereotypes such as women being “worse at parking” or “not knowing how to read a map”). In contrast, the feminine task required symbol coding skills, a task that is more “gender-neutral”, but to a certain extent associated with the belief that women are better at these types of activities. The experiment participants, before they knew their real performance results, believed that the task given to them was done better by men or by women or equally well by both.

In the first phase, participants in the experiment, men and women, performed each of the two tasks without being compared with anybody, and they obtained a remuneration. In the second phase, they were told that they were going to perform new examples of those same tasks, but this time in partnership with another participant, and that they would only obtain remuneration if they did the task better than their rival. In other words, they were placed into a competitive situation.

Furthermore, they were given certain information before they started competing, and this information was divided into eight different groups. The control group knew that they were competing. Some participants were informed regarding the gender of their opponents, others were told whether in the previous phase they had performed better or worse than their rivals, and one group were informed of both of these things. Finally, the last group were only reminded that they were competing against one other person.

Graph 1 shows a summarised version of the results of the experiment. It must be taken into account that participants were doing the task for the second time and, therefore, all of them improved their performance. But, as can be observed in the graph, in the task considered “masculine” (spatial vision), women competed worse than men only when their rival’s gender was revealed, or when they were informed that they were competing against somebody. In contrast, the equivalent graph for the symbols coding task (the one considered “feminine”) shows no significant gender difference.

When a task is competitive, the differences in performance between men and women depend on the prior information that they have.

As a final phase of the experiment, and following the tests, a survey was conducted among all participants. The survey asked them, among other things, whether they believed that the tasks that they had performed were done better by men, by women, or by both equally. What was the result? That women who believed that one of the tasks was done worse by their own gender than by men, performed it worse themselves.

The overall result of this investigation is that we only observe important gender differences in the performing of a task under competitive conditions when three situations occur: 1) a strong stereotype exists about women being worse at the said task; 2) the stereotype is reinforced by reminding participants of aspects that make them think about the gender issue, such as telling them the gender of their opponent; and 3) women who are more negatively affected by competition are precisely those who believe that the stereotype is true.

Is it possible to find similar results in a situation outside the laboratory? Do these experiments adequately reflect what happens in real life? As far as this investigation is concerned, the answer to both questions is “yes”.

Every year the Community of Madrid organises the Spring Mathematics Contest, a competition which enjoys the participation of around 40,000 secondary and sixth-form students. One of the advantages of this competition was having access, beforehand, to a measure of the ability of each student in this subject: their mathematics grade at school. If these grades are observed, it is appreciated that differences barely exist between boys and girls. However, as the students go through the elimination rounds of this contest, the proportion of females gradually falls, to the point that barely any of them reach the final round.

As shown in graph 2, the proportion of female students who participate in each round of the mathematics competition falls significantly from round 1, where they represent between 35% and 45% of participants, to round 2, where the proportion of females is around 30%. The result is even more pronounced for the proportion of winning females in the final round. Furthermore, if the students are divided into four age groups, the gender difference in the proportion of participants in one round and another grows as the age of the students increases, to the point that in the level with the oldest students (16-17 years) there is no winning female student in the final round of the contest. The conclusion of this experiment is especially powerful because it enables the observation that, as competitive pressure increases, gender differences in the results also increase, as is reflected in the average scores (graph 3).

Aspects of socialisation exist that affect men and women differently when competing, and these aspects ultimately have an impact in competitive settings such as the working environment.

Stereotypes affect the decisions and performance of any person; therefore, reducing their effect is of social importance. It is advisable, whenever possible, to avoid the classification of tasks into female and male tasks, as well as stereotypes that talk to us of “male” disciplines (for example, Sciences or Engineering disciplines) and other “female” disciplines (for example, the Letters in general or scientific careers such as Nursing). Much progress has been made in this area, but there is still work to be done to increase the visibility of women in historically male fields. If studies on the importance of role models are taken seriously, raising the visibility of women who work and excel in fields that are historically male will be important in helping to change these associations and stereotypes.

Furthermore, it is important to take into account that some measures proposed to resolve these problems could have counter-productive effects if thorough work is not done on their implementation. In this sense, caution must be exercised in how policies that attempt to mitigate differences in the labour market are applied and, in any event, it is essential to design very carefully what message is sent to society with each policy. The reason? The very existence of a policy that reinforces women can lead to a negative stereotype, whether true or not, being perpetuated, precisely jeopardising the gender it aims to protect. To offer a current example, when applying a gender quotas policy, it is important to consider that although in the short term it may be useful for raising visibility of the gender at a disadvantage, in the medium term it could serve to reinforce another stereotype, which says that women who hold certain posts are there to cover the quota, not because of their skills and merits. This could lead to women deciding not to occupy such positions, and the policy would cause the contrary effect to that sought.

For this reason, and keeping in mind future generations, it is fundamental to put an end to stereotypes, by emphasising and raising the visibility of women who, despite the difficulties, manage to occupy important positions and who, once in them and with equal pay, manage to perform in them as well, or as badly, as men.

To finish, a relevant debate would be one regarding whether or not competition should be encouraged in educational and working environments. It could be said that the important thing is either to participate or alternatively to win, but what must be avoided is insisting on it being important for females to participate but important for males to win.

IRIBERRI, N., e REY-BIEL, P. (2017), “Stereotypes are only a threat when beliefs are reinforced: On the sensitivity of gender differences in performance under competition to information provision”, Journal of Economic Behavior & Organization, volume 135, 99-111.

IRIBERRI, N., e REY-BIEL, P. (2018), “Competitive Pressure Widens the Gender Gap in Performance: Evidence from a Two?stage Competition in Mathematics”, The Economic Journal.

BERTRAND, M., C. GOLDIN e L. F. KATZ (2010), “Dynamics of the Gender Gap for Young Professionals in the Financial and Corporate Sectors”, American Economic Journal: Applied Economics, 2 (3).

BERTRAND, M., e K. HALLOCK (2001), “The gender gap in top corporate jobs”, Industrial and Labor Relations Review, 55, 3-21.