Gender stereotypes are simplified expectations of individuals’ traits and capabilities based on their sex. Such beliefs are pervasive in social relations.Footnote 1 In international relations and political science theory, however, diplomacy and interstate negotiations are generally assumed to be driven by other factors, such as material conditions, power asymmetries, institutional rules, social norms, and the personal capabilities of individual negotiators.
We examine the possibility that international negotiators’ willingness to seek cooperative solutions is also affected by gendered perceptions of themselves and the other party. We argue that international negotiations involve social interactions among diplomats who—to varying degrees—are constrained by deeply rooted gender roles and beliefs about appropriate behavior based on sex differences. Such constraints are activated intuitively, through “fast cognition” processes, rather than with careful rational deliberations, and are therefore potentially consequential even in a highly professionalized environment such as the international negotiations environment. Diplomats with a cultural background in contexts where gender differences are strongly emphasized are more likely to be affected by gender stereotypes than those socialized in cultures with less pronounced gender roles. We propose that the impact of stereotypes may be counterintuitive, and not uniformly to the disadvantage of female negotiators.
Our argument builds on negotiation theory and research in psychology, communications, and economics. Scholars in these fields have repeatedly found masculine traits to be associated with competitive distributive bargaining, whereas feminine characteristics are linked to cooperative problem solving.Footnote 2 Furthermore, gender stereotyping in negotiations tends to connect an effective negotiator with the so-called male attributes of being “strong, dominant, assertive and rational,” while an ineffective negotiator is associated with the supposed female attributes of someone who is “weak, submissive, accommodating and emotional.”Footnote 3 This is assumed to place female negotiators at a disadvantage: women are more likely to be perceived as a weak opponent and less likely to strike a good deal for themselves.Footnote 4
Gender stereotypes are usually assumed to be self-reinforcing—with men and women confirming the idea of difference by repeatedly taking on the social scripts prescribed for them. However, exposure to stereotypical behavior may also lead to contrary reactions. Specifically, we examine two possible effects of the female gender stereotype that have been found in social psychology research—male chivalry and female stereotype reactance. Female stereotype reactance may occur when women have realized and acknowledged the potentially destructive effects of a stereotype for their ability to act efficiently. Empirically, this would entail observing women reacting negatively to other women who behave in a stereotypically feminine way. Male chivalry, on the other hand, occurs when men suppose an obligation to compensate a female counterpart in the negotiations who is perceived as ineffective and inferior in line with the stereotype. The combined effect of these two mechanisms is to reverse the difference in behavior between male and female negotiators compared to what the stereotype prescribes; men are made more and women less cooperative and relational.
Empirically, we test the significance of these mechanisms in a survey experiment with 201 diplomats in the Council of the European Union. This contrasts to most of the experimental negotiation analyses that rely on college students or participants in opinion labs, including studies focusing on international negotiations. Our respondents are active, elite negotiators.Footnote 5 The respondents to the survey were randomly assigned to a scenario where, in some, conditions stereotypically feminine behavior was displayed by one party in the negotiations. The results indicate that stereotyping may indeed affect the behavior of international negotiators. Although we did not find reliable support for the female reactance mechanism, our results indicated that male respondents who encountered a female negotiator acting in a stereotypically feminine way tended to display a chivalry reaction in which they became more accommodative than they would have been otherwise. In accordance with our theory, the mechanism is triggered in particular among diplomats from countries with relatively low levels of gender equality, that is, where gender stereotypes are likely to be stronger.
We conceive of the chivalry reaction as a double-edged sword for female negotiators. On the one hand, it confirms and nurtures an image of women in negotiations as weak and emotional. On the other hand, women may gain advantage from men falling into the chivalry “trap” and offering compensating treatment beyond what they would otherwise do. By emphasizing the cognitive biases triggered by individual diplomats’ reactions to gender stereotypes, our study is situated in the broader research field that Hafner-Burton and colleagues call the behavioral revolution in international relations.Footnote 6 It is also related to the growing body of research on communication of emotions and empathy in international diplomacy.Footnote 7 We contribute to this research by emphasizing the potential intermediate effect of gender stereotypes on negotiators’ perceptions and actions.
Gender Stereotyping and International Negotiations
Negotiations between state representatives are one of the most frequently occurring and consequential practices of international relations. The idea that negotiations have masculine and feminine characters, where the “feminine” includes stronger emphasis on cooperative problem solving, features frequently in policy debates on the role of women in diplomacy and international organizations. Both scholars and policy actors have argued that increasing the number of women is not just the right thing to do but, as Madeleine Albright put it, “frankly, it is the smart thing to do.”Footnote 8 Empirically, the trend points toward a less biased descriptive representation of men and women in international affairs.Footnote 9 Although the research on gender in international relations has made significant progress,Footnote 10 very few studies in IR apply gender theory to the study of international negotiation.Footnote 11
Instead, research on interstate negotiations has predominantly been conducted within a rational choice framework, with parameters set by material and political conditions and patterns of interdependence at the domestic and international level.Footnote 12 The space for individual negotiators to make a difference is limited in this framework.Footnote 13 Constructivist accounts of international negotiations, on the other hand, have emphasized the role of norms, persuasion, and social interaction among negotiators.Footnote 14 These studies underline the fact that diplomats are motivated by perceptions of material and political conditions, rather than the conditions as such, and that perceptions can be transformed through socialization and communicative action. However, the focus in constructivist research on international negotiations has been on the transformation of preferences, interests, and identities, assuming negotiators make decisions within the frame of what Kahneman calls “slow” (or System 2) cognition.Footnote 15 This assumes a relatively careful deliberative process of decision making, affecting negotiation behavior through shifts in the perceptions of the negotiators. The theoretical argument that we propose regarding how gender stereotypes may matter in international negotiations departs from both these traditions.
The way we conceive of gender stereotypes’ role is to trigger more intuitive and immediate decisions, “fast” (or System 1) cognition to use Kahneman's notion. The stronger the gendered eyeglasses of the diplomat are, the stronger the reflex will be. This is akin to the research on communication of emotions and empathy in personal face-to-face interactions in international diplomacy.Footnote 16 For example, Holmes and Yarhi-Milo argue that diplomats with an ability to express empathy are more likely to convince the other party that they understand their motivations and the interests underlying their positions, which increases the chances for successful conflict resolution. They describe this line of research as highlighting “how individual behaviors—in particular, signals sent through expressive behaviors, such as emotional expression—are perceived, which in turn affects outcomes.”Footnote 17 Our study contributes to the “emotional turn” in IR by adding the gender perspective to this research.Footnote 18 We believe that the perception and reception of individual negotiators’ expressive behaviors are likely to be moderated by the gender-negotiation stereotype. Thus, male and female negotiators will perceive different types of emotional and rational expressions and signals through their respective stereotype lenses. This, in turn, may affect their negotiation behavior. To what extent effects on behavior translate into differences in negotiation outcomes will depend on a host of other factors particular to the specific negotiations. Thus, our argument and empirical study confines itself to the level of cooperative negotiation behavior.
The Gender Stereotype–Negotiation Link
Gender stereotypes originate in gender belief systems, which socialize men and women into different roles based on their sex. Stereotyping occurs when socialization generates rigid, simplified expectations of difference regarding men's and women's characteristics and capabilities. Research focused on negotiation behavior that spans communications studies, psychology, and economics offers a vast evidence base for expecting gendered effects among negotiators that potentially affect their inclination to seek cooperative solutions.Footnote 19 Based on this broader negotiation literature, we explore a set of common factors found to drive gender differences in negotiations that fall under the “gender stereotype–negotiation link.”Footnote 20 The literature has found that stereotyping along the lines of masculine and feminine traits affects negotiators’ expectations of, and reactions toward, their negotiating counterparts, and ultimately their performance in the negotiations.Footnote 21 Under such stereotyping, masculine traits falling along the lines of being strong and assertive become matched with effective negotiation skills, whereas feminine traits falling along the lines of being weak and emotional are considered ineffective at the negotiating table.Footnote 22
While observers of international relations have emphasized the positive aspects of the feminine stereotype for creating value in negotiations, including both a more cooperative and relational view of the other party and an ability to express emotions and empathy,Footnote 23 the gender stereotype–negotiation link is usually assumed to place female negotiators at a disadvantage. According to the results of two meta-analyses covering the larger negotiation literature,Footnote 24 men tend to be more competitive and reap better outcomes in negotiations in comparison to women. Stereotyping is considered key to understanding what drives this male advantage.Footnote 25
The stereotype thus prescribes different roles, and may have unequal distributive effects, for men and women. This means that male and female negotiators react differently to communicative signals expressing stereotypical behavior. Furthermore, social-psychological research has shown that, under some circumstances, being exposed to stereotypical behavior may lead to contrary reactions toward the prescribed social script. Because gender stereotypes tend to disadvantage women in negotiations, female negotiators may react negatively toward feminine stereotypical behavior. This is seen in some experimental research when women tend to react with resentment to stereotypes that are activated by researchers. The mechanism is described in the literature as a psychological process related to stereotype threat. According to Steele and Aronson “stereotype threat is concern and anxiety over confirming, as a self-characteristic, a negative stereotype about one's group.”Footnote 26 Under stereotype threat, women either perform worse in negotiations, conforming to the stereotype, or they exercise “stereotype reactance” under which they “behave in a manner inconsistent with the stereotype.”Footnote 27 Women are especially likely to engage in stereotype reactance when they are primed explicitly with a female stereotype in a negotiation situation.Footnote 28 Under such priming, according to Kray, Thompson, and Galinsky, women engage in stereotype reactance “by engaging in behaviors that are counter to those prescribed by the stereotype.”Footnote 29
Contrary to women, when primed with a gender stereotype in a negotiation situation, men are usually assumed to react in a way that confirms the stereotype.Footnote 30 Under stereotype confirmation, many studies find that men perform better when primed with gender stereotyping. Because of the advantages that male stereotypes garner in perceptions of negotiation capability, instead of feeling threatened, men feel enhanced when faced with stereotyping in negotiations, and this improves their performance. However, in addition to performance enhancement, a few studies find that men may also engage in stereotype confirmation by acting chivalrously toward a female negotiating partner. Here they show a higher level of cooperation with women but not men in negotiations when the feminine gender stereotype is primed.Footnote 31 Under the chivalry reaction, men seem to play into the stereotype that women are less skilled negotiators as a result of disadvantageous feminine traits by offering special, compensating treatment to them but not to male partners. Thus, as women may display stereotype reactance against a female stereotype, men may also react by abandoning the hard-bargaining male stereotype and instead display a more accommodating approach.
Stereotypes are likely to harm women in perceptions of capability. As a consequence, men tend to confirm stereotypes through enhanced performance, or by exercising chivalry toward female partners, while women encounter stereotyping as threatening and perform poorer, or react by resisting. Chivalry reactions indicate that female stereotyping may have somewhat contradictory consequences for women. While confirming an image of female negotiators as weak and vulnerable, at the same time, women may gain advantage from men being more accommodative than they otherwise would be.
This means that although gender stereotypes are usually assumed to reinforce gender roles, by setting expectations on individuals’ behavior, priming feminine stereotypical behavior by a female partner may actually reverse the behavior of both men and women. The female reactance and the male chivalry mechanisms both tend to pull men and women away from their ascribed gender roles. When faced with the female negotiation stereotype these two mechanisms potentially push negotiators toward a situation where women become less willing to be accommodative than men rather than the other way around (which is what the stereotype would predict). We design our experiment to test the outcome of these two processes—where men become more and women less willing to take a cooperative deal. Empirically, we test the following hypotheses on our sample of EU diplomats:
H1
Feminine stereotypical behavior shown by a female negotiating partner affects the willingness to engage in cooperative bargaining by making men more willing to take a cooperative deal (male chivalry effect).
H2
Feminine stereotypical behavior shown by a female negotiating partner affects the willingness to engage in cooperative bargaining by making women less willing to take a cooperative deal (female stereotype reactance effect).
Gender stereotypes are socially constructed beliefs about individuals’ characteristics and appropriate roles based on their sex. Such beliefs are not static—the extent to which people ascribe to them varies. We expect that the effects we discussed are more likely to be triggered among negotiators with a background in countries where gender stereotypes are stronger. To our knowledge, the negotiation literature has not yet evaluated gender stereotypes across cultural contexts. The samples of college students, opinion lab participants, and job candidates that are most often used in experimental negotiation studies tend to be homogenous in terms of nationality. In international negotiations, however, diplomats are likely to have varying perceptions of gender relations depending on their national backgrounds. Gender stereotypes depend on socialization processes into gender belief systems, which differ between countries. Support for and achievements in gender equality vary widely among the EU member states. Countries differ in regard to more or less gender equality in resources, capabilities and achievements, more rigid or progressive gender role socialization, and greater or fewer women-friendly policies.Footnote 32 Our third hypothesis, therefore, is:
H3
The reaction to stereotypical behavior is stronger for representatives from countries with stronger gender role socialization than for representatives from countries with weaker gender role socialization.
Research Design
Our empirical case is the Council of the European Union, which is arguably the most powerful political body of the EU.Footnote 33 All EU legislation has to pass the approval of the council, which also has important executive functions within foreign and security policy. Most research on decision making within the council uses models that are based on rational choice assumptions. Council negotiators are assumed to be promoting preferences relating to national sector-specific,Footnote 34 party-political,Footnote 35 or bureaucraticFootnote 36 interests. Studies with a more constructivist orientation have emphasized the role of socialization and the potential generation of common group norms in the course of iterative interactions.Footnote 37 Similar to the general IR literature on interstate negotiations, gender has hardly been considered in these studies.Footnote 38
We conduct a survey experiment of state representatives in the council's preparatory bodies. The people involved in these negotiations are career diplomats based in Brussels or, for some of the higher-level committees, in the national capitals. For most of the dossiers, the basis for the negotiations is a proposal tabled by the European Commission. The negotiations usually have two phases: a coalition-building phase, where states strive to gather support for their preferred position, and a brokering stage (which may also include the European Parliament) where the final agreement is hammered out under the mediation of the state holding the presidency. In the latter stage, the ministers may become involved, but most of the negotiations are in practice conducted in the preparatory bodies.Footnote 39
The experiment focuses on the reaction to stereotypical feminine behavior shown by a fictitious partner in a scenario described to the respondents. More specifically, we investigate the willingness to give support to a specific policy proposal favored by the fictitious partner, and in exchange receive a promise of support on a future occasion. This type of cooperative bargaining, which includes issue linkages—where one party agrees to vote in favor of another party's preferred position in exchange for reciprocal support on a different issue—is a common feature of council negotiations.Footnote 40 It is part of the broader class of negotiation behavior that is often referred to as integrative bargaining.Footnote 41 In our scenario, the reciprocity is diffuse rather than specific,Footnote 42 which means that the favor exchanged is not specified in detail, but rather comes in the form of an “IOU” (I owe you), to be checked in at a later time. Previous research has found that this is the most common type of reciprocity in the Council of the EU.Footnote 43 Thus, by choosing this particular type of bargaining we prime the respondents with a form of negotiation that they have likely experienced as EU negotiators. We do not specify where in the negotiation process the request is coming, or whether the respondent is pivotal to the outcome, but the exchange proposal could refer to either the coalition-building or the final brokering stage. In either case, a state with highly salient preferences needs to get others on their side to succeed.
The Survey Experiment
During the fall of 2015, we approached all twenty-eight member state representatives of a selection of eleven committees and working parties in the council, which means 308 diplomats in total (28 member states × 11 groups). We were able to complete the survey experiment by means of telephone interviews with 201 of these (a response rate of 65 percent). The selection includes the most important high-ranking committees on the one hand, and a number of lower-level working parties (among several hundred) on the other. It targets negotiators in a broad range of policy areas, including foreign and security policy, economic policy, internal market regulations, environmental policy, agriculture, and more.Footnote 44 We preprogrammed the questionnaires and supplied interviewers with a link to a web-based survey that appeared on a screen in front of the interviewer, allowing for computerized randomization of respondents into treatment groups. The experiment was embedded in a survey that took on average fifteen to twenty minutes to complete.
The experiment consisted of three scenarios to which the respondents were randomly assigned. Some were given a scenario where another negotiator (a “she” in group 1 and a “he” in group 2) used stereotypical feminine behavior when suggesting a cooperative deal in a bargaining situation. The remaining respondents were assigned to a scenario where the same suggestion was given, but where the other negotiator did not use the stereotypical behavior, and was not identified either as a man or a woman (group 3). Our main interest is in whether men and women react differently to the scenario where a female partner acts stereotypically feminine (group 1). However, by using three groups we can compare the gender difference in group 1 to gender differences in group 2, where a man acts stereotypically feminine, as well as to the gender differences in group 3, where there is no stereotypical behavior at all. If we find a gender difference in group 1, but not in group 2 and 3, the stereotype priming is likely to be at work. We expect no gender differences in group 2 because neither the male chivalry mechanism nor the female stereotype reactance mechanism should be at work when the partner is a man. Moreover, group 3, where there is no stereotypical behavior and where respondents are not primed on gender, indicates whether any difference found between men and women in group 1 is likely to be caused by the stereotype priming, or whether the difference exists a priori.
The stereotypical feminine behavior is operationalized as someone showing neediness, emotions, and distress.Footnote 45 We strove to formulate the treatment to be reasonable in the context of professional international negotiations, which means that it is a fairly moderate treatment. Specifically, in our main scenario (group 1), a female negotiator “turns to you for support,” and stresses her “concern” and “fear” for negative reactions in case of failure to reach agreement on a particular proposal. The exact treatment was:
Now, I would like you to think about a situation where a representative from another member state contacts you concerning a particular proposal, which is of high importance to this member state. This colleague turns to you for support, in what she describes as a very problematic situation. She is very concerned about being unsuccessful on this particular proposal, as she fears strong negative reactions in case of failure. Now, suppose that this person proposes that you give your support to this proposal, and in exchange promises to support your member state on another occasion. How likely would you say it is that you would accept the proposal? I would like you to indicate the likelihood on a scale from 0–10, where 0 means that it is very unlikely that you would accept the proposal, and 10 means that it is very likely that you would accept the proposal.Footnote 46
Group 2 received the same scenario, but the “she” was exchanged for a “he.” In the scenario for group 3, the stereotypical behavior (including the identification of the other party as a “she” or “he”) is removed from the treatment (the section in italics). We assign treatments based on perfect randomization between three groups, where the two groups that include feminine stereotypical behavior have sixty-three (group 1) and fifty-seven (group 2) respondents respectively. Eighty-one respondents were assigned to the treatment without stereotypical feminine behavior (group 3).Footnote 47
Results
There is no main effect of stereotypical feminine behavior on the willingness to accept the proposal when we analyze the whole sample, including both men and women. This is what we would expect given the assumption that stereotypical feminine behavior will have an effect when used by women, but not by men, and then in different directions for men and women. In all three groups the respondents are on average more willing than unwilling to agree to give their support in the scenario we present to them. The mean for the treatment without feminine stereotypical behavior (group 3) is 6.10, which is above the midpoint 5 on the eleven-point scale. The means for group 1 and group 2 are 6.43 and 6.05 respectively. None of the differences are statistically significant.
Figure 1 graphs the means for the three groups by sex. Men and women react differently to the scenario in group 1, but the same is not true for the other two groups. When a “she” uses female stereotypical behavior in the scenario, men are significantly more likely to take the suggested deal compared to women, while the same difference is not found in the other two scenarios. The difference in group 1 is statistically significant; the mean for the men under the “she” treatment (6.76) is one scale point larger than the mean for women in the same group (5.76), (p = 0.048). As expected, when a man acts stereotypically feminine (group 2), no difference between men and women emerges. The mean for women when the man uses feminine stereotypical behavior is 5.92 and for men it is 6.15, and the difference is not significant (p = 0.710). We also see that when there is no priming of gender stereotypes at all (the neutral group 3), there is no significant difference between women and men (6.33 for women versus 5.98 for men, p = 0.431). Thus, these findings support the conclusion that the gender difference found in group 1 is indeed provoked by the stereotype priming of the scenario.
Figure 1. Mean likelihood of accepting the deal
We proposed two mechanisms driving the difference in group 1. On the one hand, we suggested that men may become more willing to cooperate when a woman acts stereotypically feminine (the male chivalry reaction) while, on the other hand, women become less willing to cooperate (the female reactance reaction). There are two relevant comparisons to point out when evaluating whether both of these mechanisms are at play. First, there is a modest but significant difference between men in group 1 (mean = 6.76) and men in group 3 (mean = 5.98) (p = 0.037), which is consistent with the male chivalry mechanism (hypothesis 1). Stereotypical feminine behavior seems to have a positive effect on men's willingness to accept a bargaining proposal when expressed by a woman. Thus, men appear to react with chivalry to a woman who displays stereotypical feminine behavior by offering special, compensating, and cooperative treatment. Second, we do not find credible support for the female reactance mechanism (hypothesis 2). When comparing women in group 1 (mean = 5.76) to women in group 3 (mean = 6.33) the difference is indeed half a scale point, and it is in the other direction compared to men. So, while men became more cooperative, the women in our sample become less cooperative when facing feminine stereotypical behavior. However, the difference between women in groups 1 and 3 is not significant (p = 0.339).Footnote 48 We can therefore not say that women react to feminine stereotypical behavior shown by women by becoming less willing to cooperate.
The Moderating Impact of Socialization into Gender Belief Systems
Our third hypothesis proposed that the effect of gender stereotyping is moderated by the national-cultural background of the negotiators in terms of gender norms. Gender stereotypes are the products of socialization processes, which vary between cultural contexts. We assume that stronger socialization into gender roles makes diplomats more likely to interpret the negotiation behavior of the other party through a gender-stereotype lens. Respondents with a more gender-egalitarian background, on the other hand, are less likely to react intuitively to gender stereotypes.
To test for this possibility we conduct a multivariate regression with the respondents who received the feminine stereotypical behavior displayed by a “she” (group 1). We include in the model an interaction variable with the sex of the respondent and the value of his/her state on the gender equality index of Humbert and colleagues.Footnote 49 Socialization into gender roles is difficult to measure but their index is a useful proxy for distinguishing between different degrees of gender role socialization in Europe. It was developed specifically to compare the status of gender equality in EU member states. As such, it captures gaps between women and men across a range of areas: in work, financial resources, knowledge attainment, health status, access to power, and use of time. Such gender gaps are likely to be both the sources and the products of values and beliefs about the roles of men and women in society. The index is a composite, including a range of indicators relating to these factors. We find it reasonable to assume that gender stereotypes are weaker in countries with smaller gender gaps across these areas of work and life.Footnote 50
The model also includes two control variables that have been found to matter for EU diplomats’ tendency to engage in cooperative negotiation behavior in the EU.Footnote 51 The variable member state size controls for power asymmetries in the negotiations. It is a continuous variable indicating the population size of the respondent's state. Previous research has found that larger states are less willing to compromise in EU negotiations.Footnote 52 The variable voting rule denotes whether the dominant voting rule in the council for the policy issues of the respondent's committee/working group is unanimity (0), mixed (1), or qualified majority (QMV) (2). The existing research indicates that the voting rule matters for the degree of competition in the negotiations, although there is some disagreement on whether a unanimity requirement leads to more or less cooperative behavior.Footnote 53
Table 1 displays the results. Models 1 and 2 confirm the previous finding of a gender difference of about one scale point, and when including the controls for power asymmetry and committee voting rule.Footnote 54 Models 3 and 4 include the interaction with the gender equality index. Although the interaction term as such is not significant, the appropriate test is to plot the marginal effect of sex at different levels of gender equality.Footnote 55 Figure 2 shows that national background should be taken into account when evaluating the gender stereotype effect (left-hand panel). The figure, based on Model 4, indicates that the gender difference found in group 1, regarding willingness to agree to a cooperative deal when a “she” displays stereotypically feminine behavior, is statistically significant only up until about 54 on the gender equality index.Footnote 56 So for countries with higher levels of gender equality we cannot confirm any effect of gender stereotyping. The right-hand panel of Figure 2 shows which countries this refers to. The EU member states vary from 34 (Romania) to 74 (Sweden) on the Gender Equality Index. The Nordic countries and the Netherlands stand out as the most gender equal in the EU, while Romania, Slovakia, Portugal, Greece, and Bulgaria are found at the other end of the scale. The countries to the left of the line in the figure (pointing out 54 on the scale) are the ones where the gender stereotype priming can be statistically confirmed, according to Model 4. These are mainly countries in Eastern and Southern Europe.
Figure 2. The conditional effect of gender equality at the national level
Table 1. The effect of women's feminine stereotypical behavior on willingness to cooperate
Notes: Results from ordinary least squares regressions. Robust standard errors in parentheses. *p < .05; **p < .01; ***p < .001.
Conclusion
Interstate negotiations are the primary means by which international conflicts are resolved and states reach mutual agreements in international affairs. The willingness of negotiators to seek cooperative solutions is essential for conflict resolution, and for the realization of mutually beneficial international treaties. Although our empirical study was not designed to test the relative weight of gender stereotypes compared to other factors that affect international negotiations, our findings indicate that deeply embedded ideational constructs of masculine and feminine behavior deserve to be taken into account also in this context.
Our findings point to a non-intuitive paradox. By playing into the gender stereotype, women negotiators may actually reverse the negotiation behavior of men compared to what the stereotype itself prescribes. This is the case when women negotiators act stereotypically feminine by displaying emotions and vulnerability in the negotiations. When exposing our respondents to such a scenario, we found that some men acted with chivalry, becoming more willing to agree to a cooperative proposal. The study tells us little about how common such stereotypical behavior is in real-world international negotiations. Furthermore, the effect sizes are relatively small and the power of the tests limited, which calls for replications. Nevertheless, we confirm a mechanism at the elite level that previously has been found in laboratory experiments. Although we did not find support for the female reactance mechanism, the response of some male negotiators in our study shows that they are receptive to gender stereotypes. This indicates that gender stereotypes—and strategic or nonstrategic action building on such stereotypes—may affect international negotiators’ behavior.
The idea that women are more cooperative in negotiations and more willing than men to engage in integrative rather than distributive bargaining is widespread, both in the negotiation literature and in the policy world. In 2000, the United Nations adopted Security Council resolution 1325, which underlines “the important role of women in the prevention and resolution of conflicts and in peace-building.”Footnote 57 The resolution builds on the idea that more women entering the traditionally male-dominated area of diplomacy will make a difference for if and how conflicts are resolved. Gender stereotypes, on the other hand, are structural behavioral constraints that originate in gender belief systems in which men and women are socialized into different roles based on sex differences. These constraints are not primarily constructed at the international level, but brought to the negotiation table by diplomats given their deeply rooted predispositions from home. Our findings indicate that such predispositions are stronger for some diplomats than for others, depending on the type of gender relations they have internalized. One implication of our study, therefore, is that an increase in women participating in diplomacy and international affairs may not mitigate gendered stereotyping if it's not accompanied by improvements at the domestic level in terms of norms and behaviors that support such stereotyping.
Our study only scratches the surface of the potential role gender stereotypes play in international negotiations. For example, we investigate only feminine (and not masculine) stereotypical behavior, and we focus on the effect of these stereotypes on only one type of negotiating behavior (integrative bargaining). Still, we believe our results have important implications for scholars and policy practitioners. International relations scholars should take the message that gender matters in international negotiations, not just for normative reasons relating to the descriptive representation of women, but also potentially for negotiation behavior. The fact that we were able to trigger reactions to a gender stereotype in this context, with a fairly moderate treatment, indicates that the dominant rational choice perspective on international negotiations may be missing out on a social dynamic that affects negotiations.
Our findings speak to the significant role of “nonstandard behavior,”Footnote 58 emotions, and intuitive cognition in international relations. In speaking to the negotiation literature, our evidence contributes on at least two fronts. First, we confirm a gender stereotype–negotiation link among elite-level negotiators where previous research has been based mainly on laboratory work. Second, we add evidence to an understudied mechanism relating to men—the chivalry reaction—and illustrate the conditional nature of chivalry, as relating to national-cultural background in terms of socialization into gender roles.
To women in international affairs, the chivalry reaction indicates that stereotypes may be a double-edged sword. It demonstrates that the perception of female negotiators as weak and vulnerable exists, even at the elite level, which in many circumstances is likely to be a disadvantage. It also shows that there are strategic opportunities for female negotiators to take advantage of the stereotype by appealing to the protective nerve of some male representatives. To what extent such sophisticated strategic behavior is deliberately performed by female negotiators is beyond our study.
Supplementary Material
Supplementary material for this research note is available at <https://doi.org/10.1017/S0020818319000043>.
