Auswirkungen von Risikobereitschaft, sozialen Präferenzen und Wettbewerb auf die weibliche Berufswahl

Eine Studie zu der Frage wie sich bestimmte Präferenzen auf die Berufswahl auswirken:

In the current survey we study preference differences between men and women, focusing on three factors that are relevant in the labor market: Risk taking, social preferences and reaction to competition.  If women prefer jobs that are less risky, more socially virtuous and less competitive, then this could explain part of the gender differences in the labor market.

Das ist ja eine bereits häufiger besprochene These, die eben nicht auf reine für den Beruf erforderliche Eigenschaften, sondern eher auf die Präferenzen abstellt.

Der erste Bereich ist die Bereitschaft ein Risiko einzugehen:

The findings in this section show clear evidence that men are more risk-taking than women in most tasks and most populations. Some important caveats are, however, needed. Perhaps the most important one for labor markets is that women who chose (or who were trained) in jobs such as financial advisors, are not less risk-taking than their men colleagues. But why do we see so many more men in these positions than women? One answer might be the riskiness of the compensation in these positions. But there are 18 other possibilities as well

Der zweite Bereich sind soziale Komponenten:

First, we identify experiments that have demonstrated gender differences and look for evidence that women 37 are more responsive than men to the conditions of the experiment. We find such evidence in a wide variety of settings.

In dictator games, we find that women’s decisions vary with the size of the pie more than do men’s (Andreoni and Vesterlund), and that women’s decisions are sensitive to the gender (and home state) of their counterpart while men’s are not (Ben-Ner, Kong and Putterman).

In trust decisions we find that the amounts women send varies more than the amounts men send with the identification (and gender) of their counterpart (Buchan, Croson and Solnick), and with the existence of a picture of their counterpart (Eckel and Wilson). Similarly, female trust is sensitive to the size of the pie, the social distance in the experiment and the ability of the second player to respond, while male trust is not sensitive to any of these factors (Cox and Deck). In reciprocal decisions, we again find that women are more sensitive to conditions of the experiment. Men are less likely to punish (reward) a partner who had previously been unfair (fair) than women are (Eckel and Grossman). Women are influenced more strongly than men by the first-mover’s decision in sequential dictator games as well (Ben-Ner, Putterman, Kong and Magan). And women were more reciprocal in trust games than men (Croson and Buchan; Buchan, Croson and Solnick; Chaudhuri and Gangadharan; Snijders and Keren; Schwieren and Sutter)

In social dilemma settings women’s decisions are more sensitive to the ability to communicate than are men’s (Stockard, van de Kragt and Dodge). In the prisoner’s dilemma setting, female behavior varied more than males as the gender composition of their group changed (Ortmann and Tichy).

Second, we look between studies and compare the differences in male and female behavior. Between-study comparisons of levels is always tricky, thus we are more careful in our interpretations here. If our explanation is correct, we will see more variability in female behavior across related studies than in male behavior. We find between-study evidence for our explanation in three different settings. In responder behavior in ultimatum games, we compare the Eckel-Grossman and Solnick papers and find that rejection rates by women differ by 18.6% while rejection rates by men differ by only 8.7%. In dictator giving we compare the Eckel-Grossman and Bolton-Katok papers and find that male giving differed by only $0.62 while female giving differed by $1.46 between the two studies. Finally, in VCM games, we find that gender differences are caused by female contributions changing by 9.9 percentage-points on average, while male contributions change by only 4.6 percentage-points on average.

We believe, as suggested by Gilligan (1982), that men’s decisions are less context-specific than women’s. Participants of both genders are likely maximizing an underlying utility function, but the function that men use is less sensitive to the conditions of the experiment, information about the other party, and (even) the other party’s actions, than the function that women use. This causes what appear to be inconsistent results in our experimental studies; sometimes men appear more altruistic than women and other times, women appear more altruistic than men. But primarily what we see is women’s behavior is more context-dependent than that of men.

These results (and our organizing explanation of them) have important implications for the labor market. If, as the evidence suggests, women are more other-regarding than men they may be more likely to choose jobs that create benefits for others (e.g. in “helping” 39 professions) which are traditionally lower-paid. This may contribute to the wage gap. Similarly our organizing explanation suggests that women’s social preferences are more sensitive to subtle cues than are men’s. This can lead women to choose professions which they think are socially appropriate for women, based on the cues they observe about the workforce (for example, what proportion of women are in this given profession). In contrast, men’s choice of profession would be less sensitive to these cues.

Also stärkere Auswirkungen von sozialen Gegebenheiten bei Frauen. Diese würden sich eher Jobs suchen, in denen sie anderen helfen und dabei anscheinend auch eher darauf achten, was andere Frauen machen.

Zuletzt geht es dann um den Wettbewerb:

First, a higher fraction of men choose competitive environments than women. Second, men and women were more likely to choose competitive environments when they have an advantage in performing the task than when they do not. Women who choose competitive environments perform just as well as men in those settings. Under this more sophisticated view, the source of the observed gender differences in reaction to competition is driven by the fraction of competitive types, which is higher among men than among women.

Meiner Meinung nach passt das sehr gut zu den biologischen Theorien: Testosteron steht in einer gewissen Verbindung mit der Wettbewerbsbereitschaft. Da mehr Männer als Frauen die entsprechende Schwelle in dem Bereich überschreiten sind Männer auch häufiger in Wettbewerbssituationen zu finden. Die Frauen, die aufgrund ihrer biologischen Grundlagen allerdings ebenfalls damit gut zurechtkommen, schlagen sich genau so gut.

Dann geht es innerhalb des Wettbewerbs um die Präferenzen beim Verhandeln:

In a direct measure of attitudes rather than behavior, Babcock, Gelfand, Small, and Stayn (2003) asked several hundred people about their negotiating experience. They found that men place themselves in negotiation situation more often than women, and regard more of their interactions as potential negotiations. This difference is robust to age. Why do we see this difference in attitudes and behavior? We believe that it reflects differences in men and women’s attitudes toward competitive situations. As in the research above, when people can choose men are more likely to choose competitive interactions (like negotiation) than women are. This type of sorting can have strong implications for labor market outcomes (e.g., Lazear, 2000). Differences between career choices and promotion speeds can be caused by the desire to avoid competitive situations. High status positions are usually highly competitive. If men are more eager to participate in competitive environments than are women, this could explain a large part of the gender gap.

Dass würde dann dazu führen, dass die Frauenquote nur den Frauen hilft, die eh Freude an diesem Wettbewerb haben.

Aus der abschließenden Besprechung:

In general, this literature has documented fundamental preference differences between the two genders (with exceptions noted in the text). These differences are consistent with preferencebased explanations for the gender gap in wages.

For example, most lab and field studies indicate that women are more risk-averse than men (section 2). This risk-aversion can lead to the attraction of women to jobs with lower mean, lower-variance salaries. This preference-based explanation is consistent with some gender-gap evidence without resorting to discrimination arguments.

A number of studies also indicate that women’s social preferences are more sensitive to subtle cues than are men’s (section 3). This can lead women to choose professions which they think are socially appropriate for women, based on the cues they observe about the workforce (for example, what proportion of women are in this given profession).

Finally, a third stream of literature suggests that women’s preferences for competitive situations are lower than men’s. This can lead women to choose professions with less competition (and again to end up receiving lower wages on average). It can also lead to women earning less or advancing more slowly within a given profession.

Of course, the fact that the gender gap can be explained by preference differences does not mean that discrimination does not occur, nor can we conclude that social policy affecting this domain would not be welcome or value-creating

Hier müsste man also, damit Frauen bereit sind in entsprechenden Jobs zu arbeiten und für diese zu arbeiten, gerade diese Eigenschaften fördern. Und natürlich auch prüfen, inwieweit sie auf Biologie beruhen.

Dazu ist auch noch etwas in der Studie:

One important and interesting question about these differences is whether they are ingrained or taught. By taking a step back and asking what causes the gender difference we can also connect some of the different elements discussed in this paper. For example, a large body of literature in evolutionary biology and socio-biology documents differences in competitiveness between males and females, in many species. 10 This literature argues that the differences in competitiveness arise because, due to differences in the cost of reproduction, competitive males will attempt to mate at every opportunity. Females, on the other hand, are inherently choosy, reserving their favors for the strongest suitor. Connecting competitive behavior with risk taking, Dekel and Scotchmer (1999) developed an evolutionary model of preference-formation, to investigate to what extent evolution leads to risk-taking in winner-take-all environments (like reproduction). They show that winner-take-all games are related to the survival of risk-takers and the extinction of risk-averters. Since in many species a winner-take-all game determines the males’ right to reproduce, the argument suggests that males will evolve to be risktakers. Similar evolutionary explanations are consistent with women being more sensitive to social cues than men. In exercising choosiness about mate selection, women who were sensitive to these cues could, on average, produce more fit offspring than those who were not. In contrast this increased sensitivity did not affect a male’s chance of reproduction, and was thus not selected for. Indirect evidence on the nature/nurture question comes from the studies with children (before nurture has full impact) and cross-culturally (when nature remains constant but nurture changes). The evidence we have here suggests that gender differences in preferences remain among children and in different cultures, providing support for the nature over nurture explanation. Of course, some cultural differences have been identified, suggesting that nurture has an effect as well.

Also eine klassische evolutionspsychologische Erklärung. Wem diese zu vage ist, der kann auch eine rein medizinisch-hormonelle Erklärung suchen. Die Präferenzen, die hier geprüft worden sind, stehen meines Wissen nach jeweils mit Testosteron (Risikobereitschaft und Spass am Wettbewerb) bzw. Östrogen (Empathy und soziale Punkte) in Verbindung.