Abstract
Existing scholarship points to gender patterns in diplomacy. This study examines such gender patterns in new ways, expecting women to be less likely to be ambassadors in states with more economic clout and in conflict-affected states, but more likely to serve as ambassadors in more gender-equal states. Most importantly, we examine whether these gender differences diminish over time. New data on ambassador appointments for Denmark, Sweden, the UK and the US spanning the period 1970–2015 reveal that whereas there are no gender differences with respect to postings to gender-equal states or states with domestic conflict, women are indeed less likely to be ambassadors in economically significant states and in states in inter-state conflict. Crucially, these patterns are not diminishing over time. This study opens up for future investigation into underlying mechanisms that explain the persistence of some gender patterns and the absence of others.
Key messages
Gender differences in ambassador appointments persist over time.
Female ambassadors are less likely than men to be posted to states with higher economic status.
Female ambassadors are less likely than men to be posted to countries with inter-state conflict, but there are no such patterns for intra-state conflict.
There are no gender patterns in ambassador appointments to more gender-equal states.
Introduction
The number of women in diplomacy rose rapidly in the 2000s (Towns and Niklasson, 2017), and some ministries of foreign affairs (MFAs) have reached gender parity among their diplomats. This development is in itself a dramatic change to diplomacy, an institution with a history of formal bans on women. When these bans were lifted, generally between the 1920s and 1950s, marriage rules pertaining to women were often put in their place. These marriage rules were only lifted in the 1970s (Aggestam and Towns, 2018). Recruitment to and career development in MFAs have sometimes continued to favour men (see, for example, Conley Tyler et al, 2014; Enloe, 2014; Calin and Buterbaugh, 2019). However, exactly how gender patterns in diplomacy have evolved over time and what the current rapid surge in women actually means for gender equality in diplomacy remains under-studied. This article is a step towards filling this gap, examining gender patterns subsequent to the entry of women into diplomacy.
Previous studies have shown that within many MFAs, there are functional differentiations and hierarchical stratifications between men and women (see, for example, McGlen and Sarkees, 1993; Sjolander, 2001; Conley Tyler et al, 2014; Towns and Niklasson, 2017; Calin and Buterbaugh, 2019). Thus, men tend to occupy more senior leadership positions and are over-represented in positions dealing with security and trade, while women cluster in human rights and administration. Existing scholarship also analyses how gender shapes the diplomatic professional role, for example, how masculinities and femininities are associated with different careers and network approaches (Dobson, 2012; Neumann, 2012; Niklasson, 2020).
These studies of individual MFAs are vital for examining the institutional rules, norms and networks that generate gendered divisions of labour and gendered hierarchies within MFAs. However, we lack more systematic longitudinal or comparative insights. Specifically, we know very little about changes in these gendered divisions of labour and gendered hierarchies over the past few decades. As the diplomatic career has now been open to men and women alike in most MFAs for several decades, what has happened to these gender patterns in diplomacy? Do they dissipate over time as the number of women in diplomacy increases?
We subject some of those patterns – those related to ambassador postings – to closer scrutiny, examining what happens as the share of female ambassadors grows. Rather than a focus on the internal rules and gender organisation of single MFAs, our study shifts attention to bilateral ambassador postings (that is, ambassador postings to other states) and potential gender patterns in terms of where male and female ambassadors were posted between 1970 and 2015. Postings by four Western liberal democracies with relatively small overall gender gaps in society and politics are in focus, that is, Denmark, the UK, the US and Sweden (World Economic Forum, 2021: 20). This limited case selection allows us to take contextual factors, for example, bilateral trade relations, into account. These countries share many characteristics, but they also differ in interesting ways (see the ‘Data and methods’ section later). Do the gender patterns still follow the same trajectories in the four countries as the share of women grows, or do they differ?
Expected changes in gender patterns
The concept of gender has been understood in several ways. We follow Scott’s (1986: 1067) classic formulation, seeing gender as ‘a constitutive element of social relationships based on perceived differences between the sexes’. Gender thus concerns the norms, identities, discourses and practices that socially differentiate between male, female, third sex, non-binary people and so on. Social and political institutions, such as those of diplomacy, are generally infused with gender. That said, gender is contextually produced and malleable across time and space. We thus expect gender differences in diplomacy to vary historically and contextually.
Bilateral ambassadors serve as heads of diplomatic missions, which typically take the form of a resident embassy with diplomatic staff in a foreign capital. Ambassadors have the highest diplomatic rank, serving as key figures in the representation and promotion of their state and government policy abroad. It is a position that carries a lot of prestige.
That said, not all ambassador positions are of equal kind or equal weight. Some positions entail working in militarised contexts, whereas others primarily involve promoting national business interests or managing foreign aid relations in non-securitised environments. Some positions are in foreign capitals characterised by a high involvement of women in politics, whereas others are in political contexts where men dominate political life. Ambassador positions are clearly not of equal clout, as ambassadorships in states of high significance to the home state generally carry more status.
This study takes its starting point in research which suggests that male and female ambassadors may not be randomly distributed across these different kinds of ambassadorships. In the following, we discuss our expectations informed by this research. We begin by describing three overall gender patterns found in previous studies – male over-representation in ambassador postings of economic importance; male over-representation in ambassador postings to states in violent conflict; and female over-representation in ambassador postings to more gender-equal states – and state our expectations that these over-representations diminish over time. We then discuss why we expect the strength of these patterns to lessen over time. Our aim is to examine the presence and change of these patterns. Establishing the mechanisms through which they are produced is a worthwhile extension in future research but beyond the focus of this article.
Male over-representation in postings to states of economic importance
As stated earlier, gender concerns the distribution of social status and power among actors. Gender scholarship has demonstrated that women tend to cluster in institutional positions considered less important, whereas men tend to be over-represented in assignments of greater clout (see, for example, Bashevkin, 1993; Studlar and Moncrief, 1999; Connell, 2006). Such patterns have also been identified in prior diplomacy scholarship (Farias and do Carmo, 2017; Rumelili and Süleymanoğlu-Kürüm, 2018; Calin and Buterbaugh, 2019). For example, cross-sectional analyses of all ambassador postings made by the 50 largest national economies in 2014 reveal that female ambassadors are under-represented among postings to capitals with the highest economic clout (Towns and Niklasson, 2017).
In this study, we dig deeper into the question of where male and female ambassadors are posted with respect to the economic clout of the receiving country. The first expectation that we explore empirically is that male ambassadors are over-represented (and women under-represented) in postings to states of greater economic importance. Crucially, we expect that these gender differences diminish over time.
Male over-representation in postings to states afflicted by violent conflict
A wealth of scholarship examines how physical violence and military conflict are gendered, exposing how understandings of femininities and masculinities feed into the expectations, options and practices of men and women. For instance, prevalent constructions of femininity entail notions of physical vulnerability, inability to handle physical violence and risks of being victimised. Thus, narratives about violent conflicts often portray women as weak, inept and in need of protection by male warriors (see, for example, Elshtain, 1995; Kinsella, 2005). Masculinity, on the other hand, is often associated with physical violence, aggression, heroism and physical protection (see, for example, Cohn, 1987; Kinsella, 2005; Sjoberg, 2011). As a result, military affairs have been male dominated around the world. If such ideas are at work in diplomacy, we would expect female ambassadors to be under-represented and men to be over-represented in violent contexts.
Niklasson and Towns (2017) have indeed shown that in 2014, men were over-represented among ambassadors posted to capitals in militarised states, as captured by military spending, number of personnel in the military and involvement in armed conflict. In interviews with MFA staff, Rumelili and Süleymanoğlu-Kürüm (2018) have likewise shown that MFAs may be hesitant to expose female diplomats to violent conflicts. What remains unclear is whether these patterns can be observed over time and whether they decrease in magnitude.
Thus, the second expectation we explore empirically is that male ambassadors are over-represented (and women under-represented) in postings to states afflicted by violent conflict. We expect that these gender differences diminish over time.
Female over-representation in postings to more gender-equal states
Third, we examine whether female ambassadors are over-represented in postings to more gender-equal states. There are good reasons to believe that this may be so. For one, MFAs and sending governments may assume that female ambassadors face fewer professional obstacles when networking in more gender-equal contexts (Legue-Feilleux, 2009; Rumelili and Süleymanoğlu-Kürüm, 2019; Niklasson, 2020). What is more, posting a female ambassador might be a way of signalling political alignment with a receiving state (Jacob et al, 2021).
The third expectation we explore empirically is that female ambassadors are over-represented (and men under-represented) in postings to more gender-equal states. Centrally, we also expect that these gender differences diminish over time.
Women in all kinds of professions have been found to advance at a slower pace than their male colleagues, even when they make up a significant share in a profession (see, for example, Wynen et al, 2015). Diplomatic career paths are sometimes described as ‘leaky pipelines’ for women (see, for example, Conley Tyler et al, 2014; Farias and do Carmo 2017), characterised by disproportionate attrition and blocked by glass ceilings (Cotter et al, 2001). The over-representation of men can thus be stable and slow to change. We nonetheless expect gender differences in ambassador postings to lessen over time, for three main reasons.
Decreasing gender patterns due to the nature of the diplomatic career
The first reason for expecting weakened gender patterns is related to the nature of the diplomatic career. In Denmark, Sweden and the UK, the most important postings are generally filled with experienced and senior diplomats who have advanced in their diplomatic careers through other, less central postings (Birkan, 2019). This process takes about 15 years in the Danish MFA (Munk-Larsen, 2011). About 90 per cent of the Swedish, 99 per cent of the British and all of the Danish ambassadors are career diplomats of this kind (Stein, 2007; Syal, 2021). A ‘lag’ between when women first start appearing as ambassadors, primarily from the 1970s onwards, and when they gain access to top postings may therefore be expected in a meritocratic system. With time, we should therefore expect the under-representation of female ambassadors to decrease.
In contrast with the other three states, 30–35 per cent of US ambassadors are typically political appointees (Diplopundit, 2016; Fedderke and Jett, 2016). These tend to cluster in higher-status postings to wealthy states where the standard of living is high and the risk of military conflicts with the US is low, that is, European states, Australia, Canada and Japan (Fischer, 2014; Hollibaugh, 2015; Fedderke and Jett, 2016). There are no studies on whether men and women are distributed equally between career and politically appointed ambassadors, but there are indications that early female ambassadors were more likely to be political appointees than career diplomats (Morin, 1994; Klingvall and Winai Ström, 2010). Even so, US posting patterns should decrease over time as well, as the gender balance has improved in the State Department and female career diplomats have had enough time to advance and become eligible for high-status ambassadorships.
Decreasing gender patterns due to women’s changing status as tokens
The second reason for expecting weakened gender patterns is that women’s status as tokens (Kanter, 1977; McGlen and Sarkees, 1993) within MFAs should decrease as their numbers grow. When there are more women, they stand out less because of their gender and may instead be seen as individuals with a range of different capacities and strengths. When gender stereotyping decreases, so may ideas about certain kinds of postings being suitable for female ambassadors and others for men. The three gender patterns discussed earlier should thus be less likely to persist as the number of female ambassadors grows.
A sceptic may object that women and men may have different career preferences and that gender patterns should thus persist or even grow as the number of women increases. For instance, women may have a stronger engagement in redistributive issues (Calin and Buterbaugh, 2019) and may thus prefer postings in developing countries. Since diplomats in our four countries can influence where they are posted (Birkan, 2019; personal communication with Birgittha Widman, Chief of Unit, Human Resources Management [Manager Recruitment], 8 February 2021), it is possible that women cluster in low-status positions because of their own choices.
We find this objection plausible but unlikely. Studies in foreign affairs (Scott and Rexford, 1997) and in politics (Wängnerud, 2005) show that women’s responsibilities become more diverse and less in line with traditional gender divisions of labour as the share of women in organisations grows. Furthermore, we consider it unlikely that senior female diplomats at the peak of their careers would turn away from the most prestigious postings in favour of postings in developing countries with lower diplomatic status and higher levels of hardship. MFAs are hierarchical organisations composed of exceptionally driven and ambitious people, with female recruits equally ambitious as men (Conley Tyler et al, 2014). That said, if, contrary to our expectation, our results show persistent gender patterns in postings over time, then women’s and men’s different career preferences would be one important hypothesis to examine in future scholarship trying to unpack the mechanisms behind persistent patterns.
Decreasing gender patterns due to increasing gender equality
The third reason why we expect changing gender patterns is increasing gender equality overall in the four countries (World Economic Forum, 2021: 20). Although much work remains to be done, reports from different professional spheres indicate that career opportunities are becoming more gender equal (see, for example, Öhberg, 2017; Staiger et al, 2000). Ideas about fundamental differences between men and women have been challenged (even if they still remain), as the number of women in sectors handling violence has increased in all four countries. We thus expect the level of over-representation of men posted as ambassadors to countries involved in violent conflicts to lessen over time.
Changing gender norms more broadly may also affect the over-representation of women to gender-equal states. The responsibility to promote gender equality may increasingly be expected to fall less on women alone – for states recognised as committed to gender equality, a male ambassador’s embrace of gender equality may be a deliberate strategy to signal strong commitment to these issues (Niklasson, 2020; cf. Government Offices of Sweden, 2018). It may also become less important to appoint female ambassadors to signal political alignment with other gender-equal states when the sending state’s reputation as gender equal is firmly established.
In sum, there are several reasons to expect gender patterns in ambassador postings to change with time. Whether they have indeed changed is an empirical question. In the following section, we describe the data and methods we use to examine this question. Again, identifying underlying causal mechanisms to the patterns we observe is beyond the scope of this article.
Data and methods
To examine the strength of the patterns and whether they persist over time, we use descriptive statistics and visualisations, complemented with inferential statistical analyses.
Case selection
Our data cover information about all ambassadors appointed to bilateral posts by the US, the UK, Denmark and Sweden during 1970–2015. The Scandinavian countries are small states where women have constituted a large share of politicians and a highly gender-segregated labour force for a long time (Hernes, 1987; Kantola, 2006; Teigen and Skjeie, 2017). In comparison, the two Anglo-Saxon states are larger, enjoy more international power and prestige, and women’s share in politics and of the labour force is lower, though less gender segregated. They are also characterised by market feminism rather than state feminism, as defined by Kantola and Squires (2012). To be sure, these differences have lessened over the past two decades. We have thus selected variations of relatively gender-equal states in order to argue that if there are persistent gender patterns in these four states’ ambassador postings, there are strong reasons to suspect that there are also persistent gender patterns in the postings made by other, less gender-equal states.
Coding of postings
Only main posts and no side accreditations have been coded. We have thus coded the gender classification of 5,133 ambassadors in total: 1,839 from the US; 1,449 from the UK; 955 from Denmark; and 890 from Sweden. The larger number of ambassadorships of the US and the UK is due to their global power position.
Finding reliable data on ambassador postings over time is difficult and time consuming. MFAs do not generally record and publish the gender distribution of their diplomatic staff, or, as in the case of the UK, they do so at an aggregate level for a more recent time period. We thus rely on manual coding, drawing information from various sources: the Office of the Historian, Department of State (US)1; the British Oral History Programme and the British Diplomats Directory (UK)2; and the government directories containing all leading public officials (Kongelig dansk hof- og statskalender and Sveriges Statskalender) for Denmark and Sweden, respectively. For Sweden, we have also used the MFA directory (UD-kalendern).
There is no standard tenure for all ambassadors and the typical tenure has shortened over time (Arias and Smith, 2017). Typically, ambassadors from our four states serve two to five years per posting.3 For all countries but Denmark, we have consulted our sources every fifth year. The Danish data contain more datapoints but sometimes for different years than for the other cases; we were not able to obtain the Kongelig dansk hof- og statskalender for 1995, for example, so we included 1994 and 1996 instead. For the same reason, the data cover 2014 for Denmark instead of 2015. We have no reasons to believe that these deviations have affected our results substantially, especially as ambassador postings usually last for longer than a single year. Still, our data do not cover all ambassadors appointed by the four countries under study during 1970–2015; a few are likely to have escaped us, but as long as we have not missed female or male ambassadors systematically, this should not affect the analysis.
Coding of indicators
Gender is coded into only two categories – man or woman – based on names, gendered titles and gender pronouns. We recognise that this dichotomy does not capture the multidimensionality and complexity of gender. Still, previous studies (Towns and Niklasson, 2017) show that few, if any, diplomats are classified professionally as non-binary.
There is no given way to measure the ‘economic clout’ of receiving states, and the importance of a posting varies among sending states. We therefore focus on economic indicators that should be of relevance to all four of our sending states: (1) economic and trade power in terms of overall gross domestic product (GDP) (World Bank4); (2) export and import volumes in relation to the sending state (Correlates of War trade data); and (3) membership in the Group of Seven (G7) or Group of Eight (G8) (Britannica5). The trade data only run until 2014, which means that the analyses of these indicators end in 2010 for all countries but Denmark.
We use two indicators of violent conflict in the receiving country: the presence of (1) intra-state conflict and of (2) inter-state conflict (Uppsala Conflict Data Program Dataset Armed Conflict6). It is difficult to find reliable measures of gender equality or gender-disaggregated data over space and time. However, the share of women in the lower house of the national legislature (V-Dem7) is a useful and well-documented indicator for capturing our theoretical expectation. It is also likely to be of much more relevance for the posting of ambassadors than other gender-equality indicators (for example, literacy rates), as ambassadors primarily interact with public officials and other elites in the host country. In some of the analyses, we have also included two control variables: regime type, in the form of the Polity score (Polity IV project8); and aid dependency, specified as the net official development assistance and official aid received in current USD (World Bank’s World Development Indicators9).
Results
The overall share of female ambassadors remained low in the 1970s (see Figure 1). Clearly, more equal access to the MFAs took a while to translate into a marked increase in female ambassadors, as the supply began to increase in the lower ranks first.
Share of female ambassadors over time
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Share of female ambassadors over time
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Share of female ambassadors over time
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Notably, it is not until the 2000s that the share of female ambassadors reaches even 20 per cent in Sweden and the US. Sweden has had the largest share of female ambassadors across the four countries since 2005, consistently exceeding 30 per cent, with the US reaching that mark in the mid-2010s.
While we see a steady increase in the share of female ambassadors across the four countries, our data reveal considerable cross-national variation, despite all four having lower general gender gaps than most other states (World Economic Forum, 2021: 20). Denmark maintains a low share of female ambassadors – well below even the 10 per cent mark through the late 2000s – only reaching 20 per cent in 2014. Sweden reached that mark ten years earlier. As for the UK, its pattern over time closely mirrors that of Denmark, consistently lagging behind the US by about 10 percentage points since the 1990s. This difference may be related to the fact that many of the early US female ambassadors were political appointees (Morin, 1994).
In what follows, we subject these trends to closer scrutiny. Of particular interest is the question: does the rising share of female ambassadors translate into less gender-differentiated posting patterns over time? Or, do gender patterns exist and persist in terms of the kinds of countries to which male and female ambassadors are sent?
Postings to states of economic importance
The first expectation we explore empirically is that female ambassadors are more often posted to countries of lower economic importance than are men, expecting these gender differences to diminish over time as the share of women increases. The first indicator of economic status is the total GDP of the receiving countries, measured in billions of current (as of 2019) US dollars. The two darker lines in the graphs in Figure 2 represent the average GDP of the receiving countries to which male ambassadors versus female ambassadors were posted during each year for which we have data. With the exception of Denmark in the year 2000 – where the only two female ambassadors (out of a total of 80) were posted in India and Brazil – the average GDP of countries to which female ambassadors are posted is consistently lower than that of the countries to which their male counterparts are deployed. A statistical analysis with year fixed effects and controlling for regime type and aid dependence reveals these patterns to be statistically significant across all four countries.10
Average GDP of the receiving country by ambassador gender, 1970–2020
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Average GDP of the receiving country by ambassador gender, 1970–2020
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Average GDP of the receiving country by ambassador gender, 1970–2020
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
The patterns thus persist over time, and they are consistent across sending countries. As Figure 2 shows, the gender gap actually increases over time in Sweden and Denmark, especially once the total number of female ambassadors exceeds the single digits. In the UK and US, by contrast, the gender gap is less pronounced over time and shows signs of diminishing in the 2000s and 2010s, respectively, but then widens again in the UK by 2015. In sum, the gender differentiation in postings is more pronounced in Sweden and Denmark, the two countries with some of the highest levels of domestic gender equality. Our expectation that the gender gap in ambassador appointments would diminish over time clearly does not find support in the data.
For Sweden, Denmark and the US, similar patterns emerge for the placement of female versus male ambassadors in countries with which the sending country has stronger trade ties (see Figures 3 and 4). For both imports and exports, the gender-differential placement patterns are statistically significant in a baseline model with year fixed effects.11 In both the US and UK, female ambassadors are significantly less likely to be placed in important export countries, even when controlling for regime type, overall GDP and aid dependency. This pattern also strongly increases over time, particularly in the US. For the import data, however, the results are statistically significant in models including controls for regime type, overall GDP and aid dependency only for the pooled sample, probably because of the larger N in that analysis.
Average import volume from receiving to sending country by ambassador gender, 1970–2020
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Average import volume from receiving to sending country by ambassador gender, 1970–2020
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Average import volume from receiving to sending country by ambassador gender, 1970–2020
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Average export volume from sending to receiving country by ambassador gender
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Average export volume from sending to receiving country by ambassador gender
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Average export volume from sending to receiving country by ambassador gender
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
There are no clear indications of a widening gender gap in ambassador placements in Denmark and Sweden, with the possible exception of Sweden’s import data. This small increase is primarily driven by the important trade relationship with Germany, however. Between 1970 and 2015, all Swedish ambassadors to Germany were men, except for Ruth Jacoby in 2010. The same goes for the Swedish ambassadors to other large economies that are also important trade partners, for example, the US, the UK, France and Italy.
However, Sweden also trades a lot with smaller Nordic and North European countries. Thus, the trade indicators capture something other than the GDP measure for Sweden, which may explain the difference in gender patterns between Figures 2 and 3. This may be interpreted as female ambassadors having access to postings of economic importance to Sweden but less so when these posting are in countries of high economic status in general, like Germany.
The trading pattern of the US differs somewhat from that of Sweden. Several of its most important trading partners (for example, Canada, Mexico, China, Japan, Germany, the UK and France) are also large economies in general. Just like in the Swedish case, these prestigious ambassadorships were exclusively appointed to men until quite recently; it was only in 2013 that Caroline Kennedy was appointed as the first US female ambassador to Japan.
As a final indicator of economic status, we examine the placement of male and female ambassadors in G7/G8 countries (see Figure 5).12 For most years, there are no Danish, Swedish, US or UK female ambassadors posted to G7/G8 countries at all. The share of female ambassadors only reaches the equivalent of their male counterparts in individual years. A baseline model with year fixed effects supports this trend for the US and Sweden, where women are only 40 per cent (significant at the .05 level) and 42 per cent (significant at the .1 level) as likely to be placed in G7/G8 countries as men, respectively.13 However, the data suggest that the share of female ambassador postings in G7/G8 countries has increased in recent years once the share of female ambassadors in each MFA also grew. We thus see the gender gap evening out from the late 2000s onwards.
Share of male and female ambassadors placed in G7/G8 countries, 1970–2020
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Share of male and female ambassadors placed in G7/G8 countries, 1970–2020
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Share of male and female ambassadors placed in G7/G8 countries, 1970–2020
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
The indicators thus point in slightly different directions with regards to our first expectation, that is, that men remain over-represented in states of economic importance. This gender pattern is confirmed but primarily when economic importance is measured by GDP and G7/G8 membership. Ambassadorships to important bilateral trade partners may be more accessible to women, depending on the economic status of those states internationally. In turn, the higher the international economic status of the receiving state, the less likely women appear to be posted there. This pattern does not obviously weaken over time. Although the G7/G8 variable might indicate the beginning of such a change, the GDP one does not.
Postings to states affected by violent conflict
We also consider the overall security situation in the countries to which ambassadors are posted. If views of women as more vulnerable and in need of protection generate concerns about placing women in violent conflict settings, we should expect fewer women to be posted to countries with ongoing armed conflict within their territory, whether international or domestic.
The graphs in Figure 6 reveal no clear time-consistent patterns in Sweden, Denmark or the UK. In some years, the share of women placed in countries with ongoing domestic armed conflict is higher than the share of male ambassadors, and in other years, it is lower. The early spikes in the women’s share in Denmark and Sweden owe to the very small number of women ambassadors at the time. In 1985, only two out of 46 ambassadors in Sweden were women, both placed in conflict-affected states (Colombia and the Philippines). In Denmark, the only female ambassador in 1975 was posted to conflict-affected Argentina. The gender-differentiated postings flatten out in both Sweden and Denmark from 2005 onwards, though the gender gap grows again in Sweden in the 2010s. In the UK, meanwhile, the graph indicates that a gender gap may be emerging since the late 2000s, when the share of female ambassadors rose above 10 per cent. As these are relatively recent trends, however, we cannot make definitive statements about whether they constitute random fluctuations or actual time trends.
Share of male and female ambassadors placed in countries with intra-state conflict
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Share of male and female ambassadors placed in countries with intra-state conflict
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Share of male and female ambassadors placed in countries with intra-state conflict
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
In the US, the share of female ambassadors posted to countries affected by intra-state conflict is persistently below that of their male counterparts, though the gap has been smaller since 1995. Statistical analyses with year fixed effects and controls for regime type, GDP and aid dependence show that these results are statistically significant (at the 0.05 level). US female ambassadors have only a .61 probability of being placed in a country affected by intra-state conflict, compared to male ambassadors.
In contrast with domestic conflict, female ambassadors from all four countries are less likely to be placed in countries involved in inter-state conflict (see Figure 7). With the exception of Denmark in 2000, where one of only two female ambassadors was placed in India,14 none of the four countries placed a single woman ambassador in a country affected by inter-state armed conflict during the period under investigation. The lack of variation in women’s (non-)postings to states involved in inter-state conflict does not allow for a statistical analysis to corroborate this trend. In conjunction, however, the results lend support to the claim that female ambassadors are less likely to be placed in countries that can be deemed insecure.
Share of male and female ambassadors placed in countries with inter-state conflict
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Share of male and female ambassadors placed in countries with inter-state conflict
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Share of male and female ambassadors placed in countries with inter-state conflict
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
As expected, the difference to male ambassadors may also be decreasing over time, though not because women are sent to these postings more frequently, but because men are sent there less frequently. In 2015, male ambassadors were only posted to one country involved in inter-state conflict, that is, India; in 1990, 25 years earlier, it was India and Iraq; and in 1980, it was China, Ethiopia and Iran.
Postings to more gender-equal states
Female ambassadors may also be over-represented in receiving countries that are characterised by greater gender equality and women’s presence in public life overall. Here too, our expectation is that the gender patterns should diminish over time. We use the share of women in the national legislature as an indicator. The visualisations and analyses are, for obvious reasons, restricted to receiving states that have a legislature in place.
Figure 8 shows the average share of women in the national legislature for those receiving states to which female ambassadors are posted versus those where male ambassadors are posted. The visualisations suggest no clear differences between where male and female ambassadors are sent; in the US, for example, the gender gap has been consistently minimal over time. The UK, on the other hand, stands out by posting women to countries with higher women’s legislative representation, which is a trend that has persisted since the late 1990s,15 though it appears to weaken. The peak in 1980 is caused by the posting of Anne Warburton – one of two British female ambassadors that year – to Denmark. When the second peak occurs 20 years later, there were six female ambassadors employed by the UK. Three of them were posted to states where the share of female legislators exceeded 30 per cent: South Africa, the Netherlands and Finland.
Average share of women in the legislature in receiving countries by ambassador gender
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Average share of women in the legislature in receiving countries by ambassador gender
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
Average share of women in the legislature in receiving countries by ambassador gender
Citation: European Journal of Politics and Gender 5, 3; 10.1332/251510821X16473315423448
The overall impression, however, is that our third expectation is not met. The over-representation of female ambassadors to more gender-equal states is not decreasing over time since it was never apparent to start with, at least not in three of the four cases and during the time period studied here.
Concluding discussion
The results meet some of our expectations but challenge others. Women are indeed under-represented in postings to states with greater economic status: the average GDP of countries to which Danish, Swedish, UK and US female ambassadors are posted is almost consistently lower than that of the countries where men are posted; female ambassadors are under-represented among postings to states with which the sending states engage in more bilateral trade; and women are under-represented as ambassadors to G7/G8 states. Stated differently, men are over-represented and more likely to end up in bilateral postings to states of economic clout. Women are also less likely – much less likely – to be posted to states embroiled in inter-state violent conflict.
However, contrary to our expectations, few of these gender patterns seem to diminish over time. The gender gaps in the average GDP and trade volume of the states to which male and female ambassadors are posted do not dissipate; in fact, the gaps even seem to increase in some cases. The only gender pattern that seems to even out over time is the posting of women to G7/G8 states. The fact that these gender patterns have emerged and persisted among some of the more gender-equal states in the world is an indication that one cannot assume that gender gaps will disappear automatically over time.
Nonetheless, some of the gender patterns we expected to see did not appear in the analysis. For one, there is no systematic pattern with respect to posting women and men to states suffering from internal conflict, other than in US postings. Given our hypothesis stemming from the gendered myth of protection, this is curious: for three of the four states, there is no indication that female ambassadors are kept out of harm’s way. It is also curious given that female ambassadors are highly unlikely to be placed in states involved in inter-state conflicts. Could it be that the postings are driven more by the status of the post than by the gendered protection myth? Two of the inter-state conflict countries are China and India, which are both very powerful economically and militarily. States affected by internal conflict, by contrast, are often developing countries and more peripheral in international society, with capitals that constitute the opposite of status postings. Inter-state conflicts are of a different standing, often generating more international attention and likely producing higher status for a diplomat. Future studies could dig deeper into the mechanisms that channel female and male ambassadors away from or towards international conflict postings versus internal conflict postings.
We also found no systematic gender pattern with respect to postings to more gender-equal states. Upon closer reflection, this may not be so surprising. Given that we selected four states with lower general gender gaps in international comparison, it may be that representatives of these four states do not feel that they have anything to prove or signal with respect to gender equality.
There are many fruitful avenues for future scholarship on gender patterns in diplomacy generally and in ambassador postings more specifically. Our results are similar for all four countries: there are no consistent differences among the four or between the Anglo-Saxon and Scandinavian countries. However, more research is required on how gender patterns in ambassador postings develop in different kinds of contexts. Scholars could look more closely at gender differences in political appointments versus career ambassadors among states that rely on relatively large shares of appointees, such as the US. Another obvious central question concerns the drivers of the persisting gender patterns we have identified with respect to economic status postings and inter-state conflict. Clearly, there must be mechanisms at work that produce gender differences and keep them in place in diplomacy – mechanisms that are not merely lagging effects of women having entered the diplomatic profession late. Future research could therefore examine why some states have so many more female ambassadors than others and why male and female ambassadors tend to be posted to different kinds of places: what are the mechanisms driving the gender patterns in ambassador postings? Are they primarily located at the individual level (with women and men making different career choices) or at the institutional level? If institutional, then what precisely are the institutional factors that steer women to some postings and men to others? Are the patterns a function of ‘leaky pipelines’ in the shape of organisational gender cultures, human resources (HR) staffing, institutional networks and mentorship relations, rules and benefits for accompanying family members, or something else within the MFA? Are there international factors – such as international status hierarchies – that interact with gender and help shape where men and women are posted? There is indeed a wealth of fascinating questions to ask of gender in bilateral diplomacy, and we have only just begun exploring them.
Notes
Office of the Historian, Department of State, available at: https://history.state.gov/departmenthistory/people/by-year.
British Diplomats Directory, compiled by Colin Mackie, available at: https://issuu.com/fcohistorians/stacks/4bf815ad95c94cedbf8780d67845976b.
Information gathered from: Arias and Smith (2017); personal communication with Birgittha Widman, Chief of Unit, Human Resources Management (Manager Recruitment), 8 February 2021; and ambassador CVs on the webpages of the British Foreign, Commonwealth and Development Office (FCDO) and of the Danish MFA.
World Bank, available at: https://data.worldbank.org/indicator/NY.GDP.MKTP.CD.
Britannica, available at: www.britannica.com/topic/Group-of-Eight.
Uppsala Conflict Data Program Dataset on Armed Conflict, version 19.1, available at: https://ucdp.uu.se/downloads/index.html#armedconflict.
V-Dem dataset, version 10, available at: www.v-dem.net/en/data/archive/previous-data/v-dem-dataset/.
Polity IV Dataset, available at: www.systemicpeace.org/polityproject.html.
World Development Indicators, available at: https://databank.worldbank.org/source/world-development-indicators.
The dependent variable of the receiving country’s GDP is logged. It is statistically significant at the .01 level in Sweden, Denmark and the US, and at the 0.1 level in the UK. Observations with missing values are dropped in the statistical analyses; therefore, the N in the analyses presented in this article is reduced relative to the overall data on female and male ambassadors. All statistical analyses are run with robust standard errors. In the visualisations and statistical analyses, only placements to other countries are included; placements to international or regional organisations are omitted.
They are statistically significant at the .01 level, except for the export models for Denmark, where the significance is at the .1 level.
In the year of its establishment in 1975, the Group of Six (G6) consisted of the six countries of the US, the UK, France, Germany, Italy and Japan. Canada joined in 1976, transforming the group into the G7. With Russia’s membership, the group became the G8 from 1997 until 2014.
Lack of variation in the share of women ambassadors over time precludes an inferential statistical analysis with additional controls.
As per the definition of the Uppsala Conflict Data Program Dataset, India was involved in an inter-state armed conflict with Pakistan in 2000, even though the conflict was not characterised by military confrontation at the time.
These findings for the UK are corroborated in a statistical analysis with year fixed effects and controls for regime type, GDP and foreign aid.
Funding
This work was supported by Vetenskapsrådet (Swedish Research Council), under Grant 2013.00638 (‘Women and Diplomacy’), and the Knut and Alice Wallenberg Foundation, under Grant 2020.0186.
Acknowledgements
The authors gratefully acknowledge the research support of Sara Anderson, Ketevan Bolkvadze, Lars Crusefalk, Linus Forsman, Felicitas Höck, Felicia Robertson, Tove Selnes and Signe Stahlenius. They also thank the anonymous reviewers and the editor for constructive comments that helped improve the manuscript.
Author biographies
Anne-Kathrin Kreft is a postdoctoral fellow in the Department of Political Science at the University of Oslo, Norway. Her research focuses on armed conflict, political violence and gender. Kreft’s work has been published in, among others, the Journal of Peace Research, the Journal of Conflict Resolution, International Affairs and Comparative Political Studies.
Birgitta Niklasson is an associate professor at the Department of Political Science at the University of Gothenburg, Sweden. Her research focuses on gender and careers in political and administrative contexts. Niklasson’s work has appeared in, for example, Public Administration, West European Politics, Foreign Policy Analysis and the Hague Journal of Diplomacy.
Ann Towns is a professor of political science at the University of Gothenburg, Sweden, a Wallenberg Academy fellow and director of the GenDip programme on the gender of diplomacy. Her research centres on norms and hierarchies in international politics, often with a focus on gender and diplomacy.
Supplemental data
The data used for the visualisations and complementary statistical analyses are available on the website of the Gender and Diplomacy (GenDip) project, available at: www.gu.se/en/gendip
Conflict of interest
The authors declare that there is no conflict of interest.
References
Aggestam, K. and Towns, A.E. (eds) (2018) Gendering Diplomacy and International Negotiation, London: Palgrave.
Arias, E. and Smith, A. (2017) Tenure, promotion and performance: the career path of US ambassadors, Review of International Organizations, 13(1): 77–103. doi: 10.1007/s11558-017-9277-0
Bashevkin, S. (1993) Toeing the Lines: Women and Party Politics in English Canada, 2nd edn, Toronto: University of Toronto Press.
Birkan, O. (2019) The history of appointment process of American ambassadors and its comparison with the appointment process of the UK ambassadors, Journal of Economics and Administrative Sciences, 9(17).
Calin, C. and Buterbaugh, K. (2019) Male versus female career ambassadors: is the US Foreign Service still biased?, Foreign Policy Analysis, 15(2): 205–23. doi: 10.1093/isafpa/ory005
Cohn, C. (1987) Sex and death in the rational world of defense intellectuals, Signs, 12(4): 687–718. doi: 10.1086/494362
Conley Tyler, M.H., Blizzard, E. and Crane, B. (2014) Is international affairs too ‘hard’ for women? Explaining the missing women in Australia’s international affairs, Australia Journal of International Affairs, 68(2): 156–76. doi: 10.1080/10357718.2013.860582
Connell, R. (2006) Glass ceilings or gendered institutions? Mapping the gender regimes of public sector worksites, Public Administration Review, 66(6): 837–49. doi: 10.1111/j.1540-6210.2006.00652.x
Cotter, D., Hermsen, J.M., Ovadia, S. and Venneman, R. (2001) The glass ceiling effect, Social Forces, 80(2): 655–81. doi: 10.1353/sof.2001.0091
Diplopundit (2016) Obama’s career ambassadorship appointments: highest on record at 70.8%, https://diplopundit.net/tag/career-vs-political-appointees/
Dobson, H. (2012) Where are the women in global governance? Leaders, wives and hegemonic masculinity in the G8 and G20 summits, Global Society: Journal of Interdisciplinary International Relations, 26(4): 429–49.
Elshtain, J.B. (1995) Women and War, Chicago: University of Chicago Press.
Enloe, C.H. (2014) Bananas, Beaches and Bases: Making Feminist Sense of International Politics, 2nd edn, Berkeley, CA: University of California Press.
Farias, R.S. and do Carmo, G.F. (2017) Brazilian Female Diplomats and the Struggle for Gender Equality, Cham: Springer International Publishing.
Fedderke, J. and Jett, D. (2016) What price the court of St. James? Political influences on ambassadorial postings of the United States of America, Governance (Oxford), 30(3): 483–515. doi: 10.1111/gove.12254
Fischer, M. (2014) This very telling map shows which U.S. ambassadors were campaign bundlers, Washington Post, 10 February.
Government Offices of Sweden (2018) Sweden’s Feminist Foreign Policy: Examples from Three Years of Implementation, Stockholm: Ministry for Foreign Affairs.
Hernes, H.M. (1987) Welfare State and Woman Power: Essays in State Feminism, Oslo: Norwegian University Press.
Hollibaugh, G.E., Jr (2015) The political determinants of ambassadorial appointments, Presidential Studies Quarterly, 45(3): 445–66. doi: 10.1111/psq.12205
Jacob, S., Scherpereel, J.A. and Adams, M. (2021) Gender-balanced representation and the erosion of global liberal norms, Representation, 1–10. doi: 10.1080/00344893.2021.1982758
Kanter, R.M. (1977) Men and Women of the Corporation, New York: Basic Books.
Kantola, J. (2006) Feminists Theorize the State, Basingstoke: Palgrave Macmillan.
Kantola, J. and Squires, J. (2012) From state feminism to market feminism?, International Political Science Review, 33(4): 382–400. doi: 10.1177/0192512111432513
Kinsella, H. (2005) Discourses of difference: civilians, combatants, and compliance with the laws of war, Review of International Studies, 31(S1): 163–85. doi: 10.1017/S0260210505006844
Klingvall, M.I. and Winai Ström, G. (eds) (2010) Från Myrdal Till Lindh: Svenska Diplomatprofiler, Möklinta: Gidlund.
Legue-Feilleux, J.R. (2009) The Dynamics of Diplomacy, London: Lynne Rienner.
McGlen, N.E. and Sarkees, M.R. (1993) Women in Foreign Policy: The Insiders, London: Routledge.
Morin, A.M. (1994) Do women make better ambassadors?, Foreign Service Journal, 71(12): 26–30.
Munk-Larsen, T. (2011) Vil du vide, hvordan man bliver ambassadør?, Globalnyt, https://globalnyt.dk/content/vil-du-vide-hvordan-man-blir-ambassador-her-er-opskriften.
Neumann, I. (2012) At Home with the Diplomats, Ithaca, NY: Cornell University Press.
Niklasson, B. (2020) The gendered networking of diplomats, Hague Journal of Diplomacy, 15: 13–42. doi: 10.1163/1871191X-BJA10005
Niklasson, B. and Towns, A.E. (2017) Gender, status and ambassador appointments to militarized and violent countries, in J. Cassidy (ed) Gender and Diplomacy, London: Routledge.
Öhberg, P. (2017) Ambitious Politicians: The Implications of Career Ambition in Representative Democracy, Lawrence, KS: University Press of Kansas.
Rumelili, B. and Süleymanoğlu-Kürüm, R. (2018) Women and gender in Turkish diplomacy: historical legacies and current patterns, in K. Aggestam and A.E. Towns (eds) Gendering Diplomacy and International Negotiation, London: Palgrave, pp 87–106.
Rumelili, B. and Süleymanoğlu-Kürüm, R. (2019) Gender and governance in Turkish diplomacy: a feminist institutionalist approach, paper presented at the Gender and Diplomacy Workshop, Gothenburg, June.
Scott, J. (1986) Gender: a useful category of historical analysis, The American Historical Review, 91(5): 1053–75. doi: 10.2307/1864376
Scott, J.M. and Rexford, E.A. (1997) Finding a place for women in the world of diplomacy, Review of Public Personnel Administration, 17(2): 31–56. doi: 10.1177/0734371X9701700204
Sjoberg, L. (2011) Gender, the state, and war redux: feminist international relations across the ‘levels of analysis’, International Relations, 25(1): 108–34. doi: 10.1177/0047117810396990
Sjolander, C.T. (2001) Adding women but forgetting to stir: gender and foreign policy in the Mulroney era, in N. Michaud and K.R. Nossal (eds) Diplomatic Departures: The Conservative Era in Canadian Foreign Policy, 1984–93, Vancouver: UBC Press, pp 220–40.
Staiger, D., Auerbach, D. and Buerhaus, P. (2000) Expanding career opportunities for women and the declining interest in nursing as a career, Nursing Economics, 18(5): 230–6.
Stein, R. (2007) Diplomatic training around the world, in S. Kishan Rana and J. Kurbalija (eds) Foreign Ministries: Managing Diplomatic Networks and Optimizing Value, Malta and Geneva: DiploFoundation, pp 235–64.
Studlar, D. and Moncrief, G. (1999) Women’s work? The distribution and prestige of portfolios in the Canadian provinces, Governance: An International Journal of Policy and Administration, 12(4): 379–95. doi: 10.1111/0952-1895.00110
Syal, R. (2021) Former Tory MP’s posting as UK ambassador to Cuba raises fresh cronyism claims, The Guardian, 26 January.
Teigen, M. and Skjeie, H. (2017) The Nordic gender equality model, in O. Knutsen (ed) The Nordic Models in Political science: Challenged but Still Viable?, Bergen: Fagbokforlaget, pp 125–48.
Towns, A.E. and Niklasson, B. (2017) Gender, international status and ambassador appointments, Foreign Policy Analysis, 13: 521–40.
Wängnerud, L. (2005) A step-wise development: women in parliament in Sweden, in J. Ballington and A. Karam (eds) Women in Parliament: Beyond Numbers, Stockholm: IDEA Handbook, pp 238–48.
World Economic Forum (2021) Global Gender Gap Report 2021, Geneva: World Economic Forum.
Wynen, J., op de Beek, S. and Ruebens, S. (2015) The nexus between gender and perceived career opportunities: evidence from the US federal government, Public Personnel Management, 44(3): 375–400. doi: 10.1177/0091026015588194
