The gendered logic of cabinet reshuffles

Why might we expect different gender logics for reshuffles? Two conditions shaping cabinet formation following elections are absent for reshuffles and may affect women's cabinet inclusion. First, media attention to ministerial selection is greater during the immediate post-election period, when prime ministers or presidents form their first cabinet. ‘The presentation of a team of ministers after an extended election campaign (and, in many countries, lengthy negotiations between different parties) regularly marks a genuine highlight of the government-building process in terms of media attention and public excitement’ (Helms and Vercesi Reference Helms and Vercesi2022). For leaders who have made public commitments to diversity, activist women and the media pay close attention to women's inclusion in the new cabinet. The features of the cabinet as a collective body are closely scrutinized at the initial stage, with attention to the proportion of women and to the proportion of ministers representing other collectivities such as region, race or generation. Cabinet reshuffles, however, attract less media attention, providing presidents and prime ministers more leeway in selecting ministers from their (generally homogenous) personal networks and in slighting cabinet representational requirements.Footnote ¹ Second, the timing of elections is relatively predictable while timing of cabinet reshuffles is not. Reshuffles often result from random ‘shocks’, such as a scandal, ‘social protests’ or other unanticipated events (Fischer et al. Reference Fischer, Dowding and Dumont2012: 515). Where reshuffles are unforeseen, feminist activists lack the opportunity to raise issues of women's cabinet inclusion, enabling a president or prime minister to replace female ministers with men.

Despite these differences, research on cabinet reshuffles offers little guidance on how such inter-election changes in cabinets might be gendered differently from cabinets formed following an election. Jörn Fischer et al. (Reference Fischer, Dowding and Dumont2012: 507) suggest in passing that gender might affect the frequency with which events lead to changes in cabinet that result in ‘ministerial exit’, but they do not speculate on women's inclusion or exclusion in a reshuffled cabinet.

Some researchers, however, find that the proportions of women in cabinet are indeed lower for reshuffles compared to initial cabinets. There is suggestive evidence that leaders pay less attention to representation during reshuffles than when forming their cabinets after an election, except in cases of reshuffles motivated by an upcoming election. Studying presidential appointments in the United States, Janet Martin finds that ‘all presidents make a greater effort at the beginning of their term in office to recruit and appoint women than is the case during the course of an administration’ (Reference Martin1989: 165). MaryAnne Borrelli's study of gender and presidential cabinets (Reference Borrelli2002: 43) similarly finds women more likely to be appointed to initial cabinets, noting that greater media scrutiny leads newly elected presidents to be more concerned with symbolism. Not all reshuffles are the same, however. When leaders seek to refresh their image ahead of an election, as Marcelo Camerlo and Aníbal Pérez-Liñan (Reference Camerlo and Pérez-Liñán2015: 608, 609) note, ‘the electoral calendar decisively shapes cabinet dynamics’, particularly in presidential systems with fixed terms of office, providing ‘incentives to mobilize voter support’. These motivations for cabinet reshuffles have gendered implications. Because they are intended to signal a government's direction, if re-elected, such reshuffles increase the likelihood of bringing new faces into cabinet, creating opportunities for women, who are more likely to be political newcomers. Rebecca Davis's study of women's cabinet appointments in Western Europe examined pre-election reshuffles, defined as those occurring 12 months or fewer prior to an election. She found no evidence that women are brought in as ‘fresh faces’ for pre-election manoeuvres. Instead, like Martin and Borrelli, she found women's appointments more likely immediately following elections than within a government's term (Reference Davis1997: 79).

Many reshuffles are provoked by economic or political shocks, but the gendered consequences of such reshuffles are not straightforward. Although economic or national security crises provide advantages to men, given the association of masculine traits and competence on economic and security issues, crises may also advantage women if the government in power is linked to serious mismanagement or corruption scandals. In such cases, senior men in government circles will suffer tarnished reputations if they are considered political insiders. Women's status as outsiders may confer opportunities during such crises. Karen Beckwith (Reference Beckwith2015) explains that political crises pose gendered opportunities for women to access party leadership because senior men are likely to be too close to the unpopular leaders and junior men may prefer to delay seeking leadership until their party's fate improves. A similar gendered logic may apply to cabinet reshuffles provoked by political crisis, such as widespread discontent with the government or mass protests against specific policy initiatives.

Few scholars have examined gender and ministerial exits specifically, but the evidence from those studies is mixed. Studying five presidential democracies, Maria Escobar-Lemmon and Michelle Taylor-Robinson (Reference Escobar-Lemmon and Taylor-Robinson2016: 183; 193) found few gender differences in how men and women left their posts, suggesting that, once appointed, women ministers experience similar treatment while in office. The absence of gender differences could mean that the motivations for including women in post-election cabinets do not disappear with reshuffles. Where ‘rules specify group representation’, a reshuffle may require replacement of an outgoing minister with another who fits the group representational requirements (Annesley et al. Reference Annesley, Beckwith and Franceschet2019; Fischer et al. Reference Fischer, Dowding and Dumont2012: 511).

This dynamic fits what Claire Annesley et al. (Reference Annesley, Beckwith and Franceschet2019: 252–255) call the ‘concrete floor’: the number or proportion of women deemed necessary for a cabinet to be perceived as legitimate. Concrete floors are established over time at different thresholds, but there is evidence that, once established, they are maintained by subsequent leaders when forming post-election cabinets. Following work by Annesley et al. (Reference Annesley, Beckwith and Franceschet2019), we ask if the pattern of concrete floors for initial cabinets holds for reshuffled cabinets.

We further extend this work by examining gendered patterns of portfolio allocation in initial and reshuffled cabinets. Previous research shows that women are less likely to receive high-profile appointments (Goddard Reference Goddard2019; Kroeber and Hüffelman Reference Kroeber and Hüffelmann2022) and more likely to head ministries associated with feminine (and less prestigious) policy areas (Krook and O'Brien Reference Krook and O'Brien2012). Regarding reshuffled cabinets, ministers serving in prestigious posts are more likely to survive in cabinet and, in a study of seven west European countries, men were found to ‘have significantly higher survival rates’ than women (Bright et al. Reference Bright, Döring and Little2015: 442, 456). We further consider how gendered patterns of portfolio allocation affect women's prospects for remaining in cabinet during reshuffles.

In sum, scholarship on gender, ministerial recruitment and cabinet reshuffles has paid little attention to women's representation in reshuffled cabinets. We address these lacunae in cabinet research by examining cabinet appointments across time in two countries with multiple gender-parity cabinets following elections, France and Spain, and we provide a hard test of the concrete-floor model by focusing on gender-parity cabinets.

Case selection and methods

We explore the gendered dimensions of cabinet reshuffles through a paired comparative case study of cabinets in France and Spain, since the first parity cabinet appointed in each country.Footnote ² We seek to identify the mechanisms or events ‘that alter relations among specified sets of elements in identical or closely similar ways over a variety of situations’ (Tarrow Reference Tarrow2010: 239) to understand how initial gender-parity cabinets establish a concrete floor of gender parity in cabinet reshuffles (or fail to do so). In focusing on gender-parity cabinets – those with equal numbers of male and female cabinet ministers and arguably the most challenging to construct, we provide a hard test of the concrete-floor model.Footnote ³

France and Spain differ on two structural factors with implications for ministerial selection: (1) fixity of government term and (2) degree of centralization. The term of government – fixed for the French presidency, or flexible, as in the Spanish parliamentary system – establishes the level of political (and potentially electoral) risk of cabinet reshuffles. French presidents enjoy fixed terms of five years and hence have clear information about when the next election will take place, providing opportunities for strategic pre-election cabinet reshuffles.Footnote ⁴ In Spain's parliamentary system, prime ministers must win an investiture vote and maintain the support of parliament throughout their term;Footnote ⁵ hence, prime ministers must consider the possibility that a cabinet reshuffle could lead to a fall of government before the next scheduled election.Footnote ⁶ Fixity of government term should affect the number of cabinet reshuffles throughout a government, with opportunities for gendered cabinet changes. Multiple cabinet reshuffles require continuing reassessment of women's places within each cabinet and a recommitment to gender parity.

Second, France is a unitary state with no politically significant territorial cleavages. Spain, in contrast, is a multilevel state with 17 regions enjoying varying degrees of autonomy and several regions with their own distinct party systems (Field Reference Field2016). The territorial dimension of Spain's politics has become increasingly conflictual, with one of the largest and wealthiest regions, Catalonia, pursuing independence. The territorial cleavage has exacerbated left–right conflicts and made government formation more difficult. Representational requirements to include cabinet members from specific regions further complicate cabinet formation (Annesley et al. Reference Annesley, Beckwith and Franceschet2019: 155–175) and make it more likely that prime ministers focus on cabinet as a collective rather than individual ministers when reshuffling their teams.

France and Spain share similarities on a range of factors relevant to cabinet formation and women's inclusion. Both the French president and the Spanish prime minister have autonomy in cabinet formation. Except during cohabitation, the French president appoints the prime minister and the cabinet (Bucur Reference Bucur2017; Grossman Reference Grossman2009: 269–270). The Spanish Constitution gives the prime minister sole power to select ministers (Rodríguez-Teruel Reference Rodríguez-Teruel2011); ministers are formally appointed by the monarch once a successful investiture vote for the prime minister has taken place (Ajenjo Reference Ajenjo, Bjørn, Martin and Cheibub2015).

Second, the pool of eligible candidates for ministerial appointment extends beyond national parliaments in both countries; there is no requirement that ministers be drawn from the legislature. In France, deputies appointed to cabinet must leave their legislative seats on becoming minister; nonetheless, ‘the recruitment pool for ministers remains dominated by parliamentarians’ (Kam and Indriðason Reference Kam, Indriðason, Dowding and Dumont2005b: 47). In Spain, while many ministers are drawn from the lower house of parliament, non-parliamentary ministers have become more common in recent years (Rodríguez Teruel and Mir Reference Rodríguez-Teruel, Mir, Pinto, Cotta and de Almeida2018).

An additional similarity concerns women's political representation. Although neither country mandates gender parity for cabinet appointments, each country has legislative gender quotas for elective offices. France adopted a gender-parity law for candidates for all political parties in 2000.Footnote ⁷ In 2008, quota requirements were extended beyond elected office to include ‘professional and social responsibilities’ (Jacquemart et al. Reference Jacquemart, Revillard and Bereni2020: 54). By 2012, the Sauvadet law extended quotas to senior executive positions in the bureaucracy (Jacquemart et al. Reference Jacquemart, Revillard and Bereni2020: 53), and in March 2021, France introduced ‘gender quotas on the Executive Teams and leadership pipelines of companies over 1,000 people’.Footnote ⁸ The steady spread of gender quotas over time heightened French presidents' awareness of gender in cabinet composition, and not just in post-election cabinets (see also Claveria Reference Claveria2014). According to Valentin Behr and Sébastien Michon (Reference Behr and Michon2013: 335, 332), ‘gender parity has progressively become a mandatory rule of cabinet composition … At every cabinet reshuffle, political leaders, journalists and analysts measure representativeness … on the basis of the number of women.’

In Spain, too, quota laws apply to elected office, corporate boards and the public sector (Verge and Lombardo Reference Verge, Lombardo, Lépinard and Rubio-Marin2018). Gender quotas were used by left parties, including the Partido Socialista Obrero Español (Spanish Workers' Party – PSOE), for over 30 years, but the 2007 Equality Law made them mandatory for all political parties. The law established a ‘principle of balanced presence’, meaning that neither sex could exceed 60% of candidates for office, members on corporate boards or appointed positions in the public sector (Verge and Lombardo Reference Verge, Lombardo, Lépinard and Rubio-Marin2018: 134–144). Although the right-wing Partido Popular (PP) opposed the law, it has complied with the rules, leading to a steady increase in the proportion of women in parliament. Although electoral quotas were slower to succeed in France, women's representation has reached similar levels in the two countries, with women holding 39.5% of seats in the French National Assembly in 2020 and 44% of seats in the Spanish Congress of Deputies following the 2019 elections. These similarities – autonomy of the selector in cabinet formation, a ministerial eligibility pool beyond members of parliament and a history of gender quotas in other political arenas – should facilitate gender parity in reshuffled cabinets.

A final similarity concerns the types of political change and crises that serve as the backdrop for cabinet reshuffles in the two countries. Each country has experienced profound changes to its party system as well as several political crises in the past decade. France has experienced party system changes across the board. The French post-war party system had conventionally been constructed as: a combination of parties and political elite groupings of the centre right (e.g. the Union pour un Mouvement Populaire, UMP); the left anchored by the Parti Socialiste (PS); with various minor parties of the right and left, including a Green Party (Les Verts).Footnote ⁹ Recent French party system changes have included the confirmation of the right-wing nationalist Front National (FN, now Rassemblement National) as an established political party.Footnote ¹⁰ Following its 2012 presidential re-election loss, the UMP re-established itself in 2015 as Les Républicains; this transformation followed several years of renaming and recombining of political elements of the centre right in France. Most important, France saw the rise of a new political party formulation, En Marche! (subsequently renamed La République en Marche – LREM), which produced in 2017 a successful presidential candidate in Emmanuel Macron, who had never held elected office. In those same elections, the PS imploded, its presidential candidate eliminated in the first round.

Spain's politics have likewise undergone profound transformation in the past decade. A stable two-party system with the leftist PSOE and the rightist PP normally holding a majority of parliamentary seats has been transformed by the emergence of new state-wide parties and the reinvigoration of territorial crises (Vidal and Sánchez Vítores Reference Vidal, Sánchez-Vítores, Hutter and Kriesi2019). The declining combined vote share received by the PSOE and PP and the emergence of a new centre-right party, Ciudadanos, and a radical left party, Podemos, added to increased regional conflict and created significant volatility, with frequent elections after failed attempts by party leaders with narrow pluralities unable to gain enough support to win investiture votes in parliament.Footnote ¹¹

These party system changes in France and in Spain, not yet solidified in either country, and the various related crises faced by their governments, are factors that should arguably mitigate against attention to gender equity and women's equitable inclusion in reshuffled cabinets. Because party system change is common across Western Europe (Emmanuele and Chiaramonte Reference Emmanuele and Chiaramonte2018), our focus on France and Spain can serve as a basis for hypothesis-generating comparative cases in other countries, providing the foundation for further research.

Data and hypotheses

We constructed an original dataset of all cabinets from the first gender-parity cabinet in each country, including both initial and reshuffled cabinets, to 15 January 2022. For France, data include ministers appointed throughout the Sarkozy, Hollande and Macron presidencies; for Spain, all ministers appointed during Zapatero's, Rajoy's and Sánchez's governments are included. In addition to the sex and portfolio for each minister, we also recorded the dates they entered and exited.Footnote ¹² Following Christopher Kam and Indriði Indriðason (Reference Kam and Indriðason2005a: 329), we define cabinet reshuffles ‘as any change in ministerial personnel or responsibilities that affects more than two officeholders and at least two portfolios’ during the lifetime of an existing cabinet. Initial cabinets are defined as those formed: (1) following a general election (for French presidents and Spanish prime ministers); or (2) appointment of a new prime minister of a different party affiliation from his or her predecessor, in the absence of an election (for Spain).

With data for initial and reshuffled cabinets in the two countries, we make within-case comparisons across time, as well as cross-case comparisons between France and Spain. We derive our hypotheses about the gendered logics of cabinet reshuffles based on the literature discussed in the previous section. Regarding within-case comparisons across time, we hypothesize a concrete-floor impact: the proportion of women in cabinet should persist across reshuffles (H1). Because both countries have legislated gender quotas, we expect that leaders in both countries will maintain that proportion of women in cabinet reshuffles. Individual women may of course be removed, but we expect the gender balance to be maintained (i.e. other women will be brought into cabinet).

Second, although there is no research on party system change and gendered cabinet formation triggered by political crises, we anticipate potential impacts. Because changes in actual parties and party formations create uncertainty and unpredictability, and increase competition in the party system (Kitschelt Reference Kitschelt and Goodin2011), presidents and prime ministers may be reluctant to appoint inexperienced or relative newcomers to cabinet (although see Ennser-Jedenastik Reference Ennser-Jedenastik2021). Historical patterns of men's political over-representation mean that women are more likely to be newcomers and thus outside party leaders' inner circles. We expect that when reshuffles are motivated by political crises, the proportion of women will change (H2), but the direction of change will depend on the type of crisis. In France and Spain, new parties challenged political elites on unemployment and austerity but also highlighted corruption and the need for political renovation. Gender and politics research finds that women benefit from gendered expectations that women are more concerned with social welfare and are more honest than men (Barnes and Beaulieu Reference Barnes and Beaulieu2014; Esarey and Schwindt-Bayer Reference Esarey and Schwindt-Bayer2019). We therefore hypothesize that corruption crises will lead to increased numbers of women in cabinet (H2a); as will crises provoked by austerity and unemployment (H2b). However, when crises are provoked by terrorism, we expect the proportion of women to decline (H2c), given men's presumed competence on national security issues.

Leaders may also reshuffle their cabinets ahead of an election, with an eye to improving their electoral position. Previous research finds no gendered effect of pre-election cabinet reshuffles (Davis Reference Davis1997). Theoretically, it makes sense to assume that leaders may wish to signal inclusiveness by bringing in more women during a pre-election reshuffle but also that leaders might bring in well-known heavyweights to signal the strength of their political team. As such, we hypothesize that pre-election reshuffles will not change the gender balance, even if individual men and women are replaced (H3).

We explore other gendered patterns in cabinet reshuffles by examining portfolios and whether women are more likely than men to exit within a term. For portfolios, we distinguish between high-prestige, mid-prestige and low-prestige portfolios, following coding schemes common in cabinets and gender scholarship (Goddard Reference Goddard2019; Krook and O'Brien Reference Krook and O'Brien2012). High-prestige posts are the inner-circle posts, including interior, finance, presidency, justice, foreign affairs and defence. Low-prestige posts include sport, culture, science, women and youth, while mid-prestige posts include all the economic and social welfare posts such as energy, industry, education, housing and the like. Based on consistent findings of women's lower likelihood of receiving high-prestige portfolios (Goddard Reference Goddard2019; Kroeber and Hüffelmann Reference Kroeber and Hüffelmann2022; Krook and O'Brien Reference Krook and O'Brien2012), we expect to see similar patterns, with men more likely to hold high-prestige posts for both initial and reshuffled cabinets in France and Spain (H4).Footnote ¹³

Research on whether women are more likely to exit during a term is limited; hence, we have no firm basis on which to predict gender differences in cabinet survival rates. That said, precisely because we hypothesize that norms of gender balance in the two countries will minimize reductions in the proportions of women when cabinets are reshuffled, we do not expect to find differences between men and women when it comes to exiting cabinet early (H5).

Gender parity and cabinet reshuffles in France

Three French gender-parity cabinets were constructed following national elections in 2007, 2012 and 2017, respectively, by French presidents Nicolas Sarkozy (UMP), François Hollande (PS) and Emmanuel Macron (LREM). Each commanded a governing majority in the National Assembly and reshuffled their cabinets several times. Sarkozy made six major changes to his cabinets; François Fillon continued as prime minister across all cabinet reshuffles. Hollande thrice changed the composition of his cabinet, with three prime ministers (Jean-Marc Ayrault 2012–2014; Manuel Valls 2014–2016 and Bernard Cazeneuve 2016–2017). Macron, elected in 2017, reshuffled his cabinet three times.

Confirming our first hypothesis, gender parity was generally maintained across reshuffles (see Figure 1), with the exception of Sarkozy's cabinets. Sarkozy's initial cabinet was the first gender-parity cabinet in French history. However, his first reshuffled cabinet fell slightly below parity, keeping the number of female ministers constant (n = 7) but increasing the size of the cabinet (from 15 to 16). Thereafter, women's cabinet inclusion declined. Although cabinet size remained generally constant, only three women survived to serve in the reshuffled cabinets of June 2009 and March 2010, constituting less than 20% of ministers. By Sarkozy's fourth cabinet reshuffle in November 2010, seven women were included in a cabinet of larger size (20), but women's numbers declined in the two subsequent reshuffles (to six and five, respectively),Footnote ¹⁴ even as cabinet grew in June 2011 to 23 ministers. Although women's numbers were relatively constant, Sarkozy increased the number of ministries, appointing more men (but not more women) to cabinet, hence decreasing the percentage of female ministers across time.Footnote ¹⁵

Figure 1. Percentage of Women in Cabinet, France 2004–2021, by President

Presidential candidate François Hollande promised gender parity in his 2012 campaign; nine women and ten men provided such parity in his initial cabinet; minor cabinet adjustments on 18 June 2012, not constituting a cabinet reshuffle, increased the number of women in cabinet to ten, equal to the number of male ministers.Footnote ¹⁶ Hollande reshuffled his cabinet three times thereafter. Seven women and nine men served in the first reshuffled cabinet of 31 March 2014, just missing parity. A major reshuffle, on 27 August 2014, resulted in a cabinet of eight women and nine men, meeting the standard for parity. Hollande's final reshuffle, 11 February 2016, achieved parity, with ten women and ten men. A final adjustment, in December 2016, resulted in a gender-parity cabinet of nine women and ten men.

In 2017, presidential candidate Emmanuel Macron promised to form a gender-parity cabinet if elected and, like Hollande, he did so, appointing nine women and nine men to his initial cabinet. Macron reshuffled his cabinet three times, and generally maintained gender parity across these reshuffles. His first, on 19 June 2017, gave women a majority: eleven women and nine men. His second reshuffle, in October 2018, violated gender parity, with twelve men but only nine women, the sole Macron cabinet with fewer women than men. The July 2020 reshuffle restored gender parity, in a slightly smaller cabinet. Fourteen women served in the Macron cabinets, three of whom served continuously. After his initial cabinet, Macron reshuffled most of his female ministers out of cabinet but brought in four new women as ministers.

Gender-parity predictions and variation in France

We hypothesized that cabinet reshuffles resulting from different types of political shocks would affect women's inclusion differently: corruption and austerity crises would increase women's cabinet inclusion (H2a and H2b) but terrorism would reduce women's inclusion (H2c). Recent French presidents have had to address multiple political shocks: corruption, terrorist attacks, high unemployment and party system changes.

All three presidents examined herein have confronted corruption charges involving their governments and cabinet members. Contrary to our prediction, corruption crises had no impact on women's cabinet presence in France (H2a). Most issues were resolved by ministerial resignation, although in some cases, a president used the opportunity to engage in a broader reorganization of his governing team. Sarkozy's minister of employment, social relations and solidarity, Éric Woerth, accused of a conflict of interest, lost his post in the March 2010 cabinet reshuffle (Erlanger Reference Erlanger2010); Sarkozy did not increase the number of women in cabinet with this reshuffle, and hence did not employ women's presence strategically to signal opposition to or resolution of corruption. Hollande confronted several corruption scandals, most of which did not involve cabinet ministers and none of which involved women (BBC News 2013a, 2013b). Near the end of Hollande's presidency, in March 2017, PS interior minister Bruno Le Roux resigned in response to claims that he had employed his children in government positions for which they did little or no work, paying them with public funds (Morenne Reference Morenne2017). Le Roux's resignation prompted his replacement but not a cabinet reshuffle or change in the number of women in the cabinet.

At least three of Macron's cabinet ministers had been accused of corrupt activities, including illegal nepotism (Vinocur Reference Vinocur2017), misuse of public funds and conflict of interest (Reuters 2021). Most seriously, Macron survived a vote of no-confidence in July 2018 (Young Reference Young2018) related to corruption charges (Le Monde 2021; McAuley Reference McAuley2019). Despite the no-confidence vote, which Macron easily withstood, no cabinet reshuffle ensued.Footnote ¹⁷ Charges of corruption neither triggered increases in women's cabinet appointments to signal a clean sweep nor removed women from office.

Recent French presidents have also grappled with the economic shock of serious unemployment challenges, but these generally have not affected cabinet gender balances (H2b). Only for Sarkozy, approaching the midterm of his presidency, confronted ‘with an economy reeling from the euro crisis and nearly zero growth [with an] unemployment rate of 9.9%, a 12-year high, and rising’ (Erlanger Reference Erlanger2012), were women removed from cabinet. Sarkozy reshuffled his cabinet in June 2009 and March 2010, keeping only three women in each. Hollande, who ‘had publicly announced … that his candidacy for a second mandate would be conditioned by the decline in unemployment’ (Durovic Reference Durovic2019: 1488), nonetheless maintained a relatively constant number of women in cabinet. In general, Hollande did not reshuffle women out of cabinet, even in response to unemployment challenges.

Although unemployment hit a decade low by 2019 (Vaudano Reference Vaudano2019), Macron's economic policies provoked massive civil resistance in the Gilets Jaunes protests against a proposed gasoline tax, involving hundreds of thousands of protesters across France (Le Monde 2019). Macron's approval ratings dropped, with ‘only 26 percent of the French having a good opinion’ of their president, a condition under which a cabinet reshuffle might have been expected.Footnote ¹⁸ Nonetheless, the only change of cabinet was the removal of Interior Minister Christophe Castaner, ‘widely criticised for his handling of the … protests’ (Henley Reference Henley2020). Macron refrained from reshuffling his cabinet, and minor changes of ministers had no impact on female cabinet members. In sum, at a collective level, women's inclusion is not affected by political shocks as motivations for cabinet reshuffles; the proportion of women in cabinet remains generally constant.

With respect to terrorism-related crises (H2c), France experienced several politically motivated violent attacks between August 2007 and August 2019.Footnote ¹⁹ Nonetheless, such threats have not reduced the proportion of female ministers. Some of the worst attacks occurred during the Hollande presidency, including the violent attack on staff of the satirical newspaper Charlie Hebdo in January 2015, a deadly assault on a Jewish kosher grocery, a series of coordinated violent attacks across Paris on 13 November 2015, resulting in more than 100 deaths, and an attack on a crowd of celebrants in Nice on Bastille Day, 2016, which killed 86 people, with an additional 433 injured. These attacks had no impact on women's representation in Hollande's cabinets. Terrorist attacks have continued during the Macron presidency, including an attack on 3 October 2019 on the Paris police headquarters, resulting in five deaths, and an attack on 29 October 2020 on the Notre Dame basilica in Nice. These events did not lead the president to remove women ministers and replace them with men. None of these events has had any impact on the proportion of women in Macron's cabinets.

Does the approach of an election encourage a president to increase the number of women in cabinet, to remove a potentially contentious issue from his campaign? Conversely, does a president with discouraging re-election prospects reshuffle his cabinet to include the ‘big beasts’ who might increase his re-election chances? Limited research suggests that as an election approaches, leaders will abandon gender parity among ministers in a pre-election reshuffle (H3). Following Davis (Reference Davis1997), using a 12-month lead time for French presidential elections, the evidence for France is inconclusive. Two of the three French presidents reshuffled their cabinets within a year or less of the scheduled presidential election. Sarkozy's 2010 cabinet reshuffle, more than a year before scheduled elections, was considered to have been a strategic move in anticipation of the 2012 elections (Erlanger Reference Erlanger2010). In the 2010 reshuffle, Sarkozy increased the number of women ministers. In his final reshuffle in June 2011, the number of women stayed constant (although their percentage of cabinet posts dropped to 21.8%). Hollande, declining to run for re-election, reshuffled his cabinet in February 2016, more than a year before the election, but made minor adjustments in December 2016 that shifted the ultimate cabinet below parity. Macron, facing re-election in April 2022, had not reshuffled his gender-parity cabinet since July 2020. This most recent reshuffle, however, was nonetheless targeted towards re-election (Henley Reference Henley2020). Given the relative brevity of cabinet duration across the three presidencies, cabinet reshuffles in a year preceding an election appear to be part of the regular pattern of cabinet renewal; changes within cabinet support a conclusion of cabinet member durability with little disadvantage for female ministers or women's collective presence in cabinet, even when presidents strategically reshuffle cabinet ministers to bolster re-election chances.

The allocation of cabinet posts does, however, follow a clear gendered logic, confirming our prediction that women are less likely than men to receive high-prestige portfolios (H4). Of 36 high-prestige cabinet appointments in France since 2007, men received 16 (69.6%) of them (see Table 1). Women are even less likely to be appointed to a high-prestige post during a reshuffle: 11 men and just two women were appointed to high-prestige appointments in cabinet reshuffles.

Table 1. Gender and Portfolio Allocation in France in Initial and Reshuffled Cabinets

We also examined whether women and men are equally likely to exit cabinet within a term (H5).Footnote ²⁰ We found that women are as likely as their male counterparts to continue in cabinet and are not more likely than men to exit cabinet early, confirming our hypothesis. Although nearly half of all cabinet ministers exit early, there is no statistically significant gender difference. Some 70% of women appointed to initial cabinets exit early as do 70.6% of men (see Table 2).

Table 2. Gender, Exits and Ministerial Survival in France

Notes: Phi = 0.006, not significant at p < 0.951.

In sum, we do not find that reshuffles follow a gendered logic that is different from considerations around women's inclusion for initial cabinets. With the exception of Sarkozy, French presidents have generally maintained gender parity. Nor have French presidents used political shocks – not even terrorist attacks – as opportunities to remove women from cabinet. Instead, the political autonomy of French presidents permits them to form cabinets as they wish; the ease of constructing gender-parity cabinets and the diffusion of gender quotas combined to sustain gender-parity cabinets across time in France.

Cabinet reshuffles and gender parity in Spain (2004–2021)

Two Spanish prime ministers, José Luis Rodríguez Zapatero and Pedro Sánchez, both leading the PSOE, formed cabinets in which women held either an equal number or a majority of ministerial posts. During conservative leader Mariano Rajoy's term, men made up a majority of ministers. Across all three leaders, the gender ratio did not change substantially from the initial cabinets, despite reshuffles and ministerial exits (see Figure 2). Women's inclusion persisted even as reshuffles occurred against the backdrop of economic and corruption crises and far-reaching party system change.

Figure 2. Percentage of Women in Cabinet, Spain 2004–2021, by Prime Minister

During his first term (2004–2008) Zapatero reshuffled his cabinet twice, maintaining gender balance in the first reshuffle but reducing women's inclusion to seven of 16 ministers (43.7%) following the second reshuffle. Upon re-election in 2008, Zapatero's cabinet grew to 17 ministers (nine women and eight men). Two reshuffles occurred during this (shorter) term; when Zapatero left office in 2011, there were seven women and eight men in cabinet. Using our definition of gender parity, particularly when cabinets contain an odd number of ministers, only one reshuffle (in July 2007) fell below parity. Gender parity was maintained for all other cabinet reshuffles during Zapatero's term.

Becoming prime minister in December 2011, Mariano Rajoy constructed a much smaller initial cabinet, with four women (30.7%) and nine men. Although eight members of the initial cabinet exited, two due to scandals, Rajoy avoided any large-scale reorganizations of his team. At the end of his first term, women's cabinet representation had dropped to 23.1%, with three women and 10 men in cabinet. Rajoy's second term, beginning in November 2016, ended when he lost a confidence vote in parliament on 1 June 2018. The proportion of women remained steady throughout (38%), with only one minister exiting during the shortened term.

Following Rajoy's defeat, PSOE leader Pedro Sánchez became Spain's new prime minister and formed the world's first cabinet where a super-majority of ministers were women (11 women and six men). The number and percentage of women in cabinet held steady throughout Sánchez’s first term in office, despite the exit of three ministers. The second Sánchez government, still in office at the time of this writing, is the first coalition government since the return of democracy. After failing to win an investiture vote on his own, Sánchez entered into coalition with Podemos, forming a cabinet with five ministers from that party. The 22-member cabinet included an equal proportion of men and women. No reshuffles occurred in its first year in office, but three reshuffles took place in 2021, leading to larger proportions of women. The first reshuffle, in January 2021, was small and maintained gender parity. The second and third reshuffles were much larger, and in both, the number of men declined. A reshuffle in March 2021 changed the gender balance to 12 women and 10 men and in the large-scale reorganization in July 2021, Sánchez increased the number of women to 14 and reduced the inclusion of men to eight (36%).

Gendered expectations and Spanish cabinet reshuffles

Consistent with H1, the proportion of women is generally maintained during reshuffles, with two exceptions where the level of women's inclusion declined and two where it increased (see Figure 2). Contrary to our expectations about the impact of different types of political crises on women's cabinet inclusion, our data do not show Spanish prime ministers including more women in reshuffles provoked by corruption or austerity crises (H2a, H2b) and terrorism threats have not led to fewer women in cabinet (H2c).

Political crises driven by corruption intensified in 2013 during Rajoy's administration. Although Rajoy did not reshuffle his cabinet, the crisis of confidence in Rajoy's government was serious, leading to significant declines in voter support for the PP (Delgado and López Reference Delgado and López2014: 290). In 2014, a corruption scandal led to the exit of the health minister, Ana Mato, who was linked to a kickback scheme (Delgado and López Reference Delgado and López2015: 279). The PP was not alone in being linked to corruption scandals; the monarchy and other political parties were likewise facing charges, and public sentiment was soon viewing corruption as an all-encompassing problem among the country's elite. Perhaps given the degree of risk, Rajoy avoided reshuffling his cabinet. As such, we cannot confirm H2a: corruption crises did not lead to cabinet reshuffles that increased the proportion of women.

During Zapatero's first term (2004–2008), Spain's economy grew, but his second term coincided with the onset of a global financial crisis. The prime minister reorganized his cabinet three times, but our data do not confirm H2b: these reshuffles did not lead to larger proportions of women to soften the image of the government's austerity measures. The first and largest reshuffle, occurring against the backdrop of anti-austerity demonstrations, involved six ministries changing hands, with embattled ministers replaced by party heavyweights (Aizpeolea and Díez Reference Aizpeolea and Díez2009). Yet gender parity endured, and a woman, Elena Salgado, already one of Zapatero's top ministers, was moved from the public administration portfolio to the economy and finance portfolio. A second reorganization took place about 18 months later, amidst a worsening unemployment crisis and declining government popularity. Zapatero proclaimed the reorganization to be about renewing and strengthening the government in order to bring about the reforms needed for economic recovery (Romero Reference Romero2010). The prime minister eliminated two portfolios (equality and housing), which were absorbed into larger departments. These posts had been led by women, leaving the cabinet with seven women and eight men, still meeting our definition of gender parity.

Regarding terrorism, we expected that gender stereotypes would lead prime ministers to favour men as ministers given their assumed competence on national security (H2c). Yet this was not the case in Spain. The country's first-ever parity cabinet was formed in the wake of the deadly Madrid train bombings, occurring just three days before the general elections. Unlike France, which suffered several instances of international terrorism, Spain has long faced the threat of domestic terrorism linked to the separatist Euskadi Ta Askatasuna (ETA). During Zapatero's and Rajoy's terms, some attacks occurred but there were also instances of ceasefires and positive developments. None of these events led to a cabinet reshuffle.

As with the French case, the Spanish case confirms H3: pre-election cabinet reshuffles do not change the proportion of women in cabinet. Of the three Spanish prime ministers under study, only Zapatero undertook a reshuffle in July 2007 that was clearly targeted at the next election. The portfolios that changed hands were not part of the cabinet's inner core, however, but instead included health, housing and culture. The gender balance remained the same.

Turning to expectations about portfolio prestige and duration in cabinet, our data confirm our hypothesis that men are more likely than women to serve in high-prestige portfolios (H4). Combining all appointments since 2004, only 48 ministers held high-prestige posts, but 32 of them were men (66.7%) and only 16 (33.3%) were women. A somewhat different pattern emerges when we disaggregate by initial and reshuffled cabinets. Although the numbers are smaller, men are considerably more likely to be appointed to high-prestige posts in reshuffled cabinets than are women (see Table 3).

Table 3. Gender and Portfolio Prestige in Spain: Initial and Reshuffled Cabinets

When it comes to exiting early versus serving a complete term (H5), we find no significant differences between men and women in Spain. Of women appointed to initial cabinets, 48.9% exited early, compared to 40.4% for men (see Table 4).

Table 4. Gender, Exits and Ministerial Survival in Spain

Notes: Phi = −0.085, not significant at p < 0.389.

In sum, reviewing cabinet reshuffles in Spain across the last three prime ministers confirms that, regardless of the factors prompting the reshuffle, the proportion of women does not generally decline. Nor are women ministers individually more likely to exit cabinet during a term than men.