When is Access to Political Capital Gendered? Lessons from the Czech Parliament

This article compares access to bill making and senior legislative ofﬁces among male and female MPs when their respective parties are in government or in opposition. Using an original dataset with all Czech MPs elected between 1996 and 2017, the article ﬁnds that female legislators face a more restricted access to these important legislative assignments when their party is part of government and when their value as a means of generating political capital is high. This points to the speciﬁc conditions under which female MPs might be arbitrarily held back by their parties in the parliamentary workplace.


Introduction
While there is evidence that female legislators face various forms of structural bias that affect their political careers, the full extent of this bias is yet to be explored (Erikson and Verge, this issue).This article examines whether female MPs have a more limited access to those legislative tasks and duties that can enhance their political capital.This is done by analysing gender differences in access to bill making and senior legislative positions among government and opposition legislators.As the legislative assignments and duties that are distributed by government parties are more clearly connected to political decision making, they naturally come with greater exposure and visibility-not only for nominees themselves, but also for their party (Stambough and O'Regan, 2007;Volden et al., 2013;Louwerse and Otjes, 2016).
This article assumes that if senior party officers who oversee the distribution of political goods have an (implicit) preference for male legislators, it will most likely manifest itself when stakes are high and the party is at its most visible.In developing its expectations, the study makes use of the gendered workplace approach which highlights the role of informal rules-in-use that govern political institutions in bringing about outcomes that disadvantage women (Erikson and Josefsson, 2019;Erikson and Verge, this issue).An original dataset containing all Czech legislators elected for six major parties between 1996 and 2017 is used for the purposes of empirical testing.The Czech Republic offers an ideal setting for investigating these expectations because of its relatively stable-yet polarisedmultiparty system, wherein most political parties take regular turns in being part of government and being in opposition (Mansfeldova ´, 2011;Outl y and Prouza, 2013).The Czech case is also interesting for its fairly recent return to democracy, which enables this study to broaden our knowledge of the ways in which relatively new political institutions can produce gendered outcomes (Mackay and Waylen, 2014).
This article finds that male legislators have a more privileged access to important legislative duties when their party is part of government, but not when it is opposition.These results support the study's expectation that masculinityinfused models of an 'ideal politician' limit female MPs' access to important political assignments when the value of these as a means of generating political capital is at its highest.

Gendered access to political capital in the parliamentary workplace
In most democracies, political parties have a major say in determining who gets elected and who is given important political assignments that are sources of political capital (Bjarnega ˚rd, 2013;Kenny, 2013;Kenny and Verge, 2016;Verge and Claveria, 2018).Understanding the preferences of political parties is thus a key to understanding the distribution patterns of important sources of political capital in the parliamentary workplace (Erikson and Verge, this issue).Political parties can produce gendered outcomes even if they are governed by formal rules that are seemingly gender-neutral (Gains and Lowndes, 2014;Verge and Claveria, 2018).Past research shows that the informal rules, norms and rituals that govern the day-to-day running of political parties favour male politicians (Mackay et al., 2010;Bjarnega ˚rd and Kenny, 2016;Lowndes, 2019).These gendered informal rules-in-use have their roots in the long history of male over-representation in politics, which has given rise to a stereotypically masculine model of an 'ideal politician' (Bjarnega ˚rd, 2018).Therefore, stereotypically masculine traits and values are seen by parties as highly meriting for a successful political career and are also believed to be in demand by voters (Murray, 2010;Kenny, 2013;Kenny and Verge, 2016).Women are attributed stereotypically feminine traits and values that are considered to be incompatible with the nature of the political arena (Franceschet and Piscopo, 2008;Murray, 2010).These resilient constructions of masculinity and femininity determine which political opportunities are available to men and women (Gains and Lowndes, 2014).Men who possess the soughtafter masculine qualities are seen as competent, dependable and deserving of access to informal networks of power where they can be groomed into becoming future party insiders (Bjarnega ˚rd, 2013;Verge and Claveria, 2018).Women are often seen as potential liabilities in terms of electoral popularity but also ability to satisfactorily carry out legislative duties (Niven, 2006;Pruysers and Blais, 2017).Thus, women are less likely to be recruited as candidates and, once elected, more likely to be given tasks and assignments that are of secondary importance or that lie on the fringes of their party's interests (Bjarnega ˚rd, 2013;Kenny, 2013;Kenny and Verge, 2016;Verge and Claveria, 2018).
The prevailing model of a 'perfect politician' leads to a demarcation of institutional tasks and roles as either masculine or feminine (Murray, 2010).The typically 'feminine' political assignments tend to be more labour-intensive and come with less exposure, while the stereotypically 'masculine' tasks and duties generate more visibility and are more clearly connected to the traditional understanding of power as physical force (Rosenthal, 1997).It has been shown that female politicians propose more bills on issues that are considered to fall within the 'feminine domain' (Swers, 2005;Schwindt-Bayer, 2006;Bektas and Issever-Ekinci, 2019) and fewer in issue domains that are stereotypically masculine, such as economy or defence (Rosenthal, 1997;Schwindt-Bayer, 2006).Past research also finds evidence of a gendered distribution of senior legislative positions (Rosenthal, 1997;Heath et al., 2005) and gendered appointments into senior executive office (O'Brien, 2015;Stambough and O'Regan, 2007).Women are less likely to be selected to high-ranked legislative (Heath et al., 2005) or executive posts (O'Brien, 2015;Verge and Astudillo, 2019).
While there is evidence that some political tasks and posts are defined as either predominantly masculine or feminine, we need to better understand whether female politicians are less likely to be entrusted those legislative assignments and duties that come with greater decision-making power and higher visibility, as access to these political goods increases legislators' chances to get re-elected and gain political seniority (Folke and Rickne, 2016;Da ¨ubler et al., 2018;Papp and Russo, 2018).A better understanding of the distribution patterns of legislative assignments can thus improve our knowledge of the extent to which gender bias shapes the political careers of female politicians.This article attempts to bridge this gap by examining gender differences in access to bill making and senior legislative offices in low-and high-stakes conditions created by the party's government status (for a similar approach, see Volden et al., 2013).The expectation is that if the party officers who are responsible for the distribution of legislative tasks and duties hold gendered ideas about who is a prospective and worthy representative of the party, they are most likely to manifest themselves when the party is in power.This article focuses on three different kinds of legislative tasks and offices the distribution of which is firmly controlled by political parties, namely bill making, legislative committee chairpersonship positions, and parliamentary party group (PPG) leadership positions.
Political goods that are available to the parties that are part of government come with greater decision-making power and exposure (Stambough and O'Regan, 2007;Louwerse and Otjes, 2016).This is because government parties are more likely to see their bill proposals passed or to influence the decisions made by those legislative committees they preside over.It is in these cases that the link between political assignments as a means of enhancing political capital is most obvious (Louwerse and Otjes, 2016).Political parties that are part of government are thus likely to carefully consider who is to be given access to these potentially potent sources of political capital.It is in these conditions that the gendered ideas about who is an ideal representative of the party might provide male legislators with an arbitrary advantage (Bjarnega ˚rd, 2018).
A successful passage of bill proposals is a good source of political capital both for the initiator and his/her party, as it often generates considerable media attention.Since the bill proposals submitted by government legislators have a greater chance of becoming laws, the parties that comprise the government are most likely to be involved in the process of deciding who should be allowed to submit bill proposals in their name (Volden et al., 2013).The bill proposals submitted by opposition legislators face much bleaker odds of attracting majority support (Louwerse and Otjes, 2016).Instead, the power in the hands of opposition MPs lies in numbers.The more proposals they submit, the higher the chances that one of them will attract media attention (Bowler, 2010).And it is only when opposition proposals attract attention that they become a source of political capital for the initiator-at least in terms of enhanced name recognition.Given the relatively low political salience of individual opposition bill proposals, it is unlikely that opposition parties are as involved in regulating access to this important legislative activity as their government counterparts.This lack of effective gatekeeping is likely to benefit hardworking and productive opposition legislators, regardless of sex.Accordingly, our first hypothesis posits the following: H1: Political parties that comprise the government will favour male MPs when deciding who should be allowed to submit new bill proposals.Opposition parties will not favour any particular sex.
Similar dynamics can be expected to influence the distribution of senior legislative offices (Rosenthal, 1997;Heath et al., 2005).Government parties are likely to fill the chairpersonship positions allocated to them with high-ranked party insiders who will act as the public face of their respective committee's decisions (Verge and Astudillo, 2019).Opposition parties, which do not control a voting majority in any of the committees they preside over, might want to choose to fill the posts with those legislators who represent a visual alternative to the sitting government.In such settings, female legislators and less experienced legislators might be given a chance (Stambough and O'Regan, 2007).Hence, we posit that: H2: Government parties will be more likely to nominate male legislators as legislative committee chairs.Political parties that are in opposition will not favour any particular sex.
Finally, appointments to PPG leadership positions are also expected to be governed by high-and low-stakes dynamics, this time created by the relative importance of the positions themselves rather than the party's government status.The high-stakes position of PPG chairperson comes with considerable influence over agenda setting and policy formation, while the vice-chairpersonship position is largely symbolic and thus low stakes.It is thus expected that male party insiders have better chances of becoming PPG chairs; while the distribution of PPG vicechairpersonships will be relatively gender equal.Therefore, our last hypotheses posit that: H3a: Political parties, regardless of their government status, will be more likely to nominate male legislators as parliamentary party group chairs.H3b: Political parties, regardless of their government status, will not favour any particular sex in the selection of parliamentary party group vice-chairs.

Case selection
The Czech Republic is, in many ways, a typical representative of a polarised, multiparty system where major parties take turns in leading a government and being in opposition.This allows us to examine the effect of being in power on the distribution patterns of legislative duties.The political scene is ideologically fragmented and coalition governments of three to four political parties are the norm (Linek and Mansfeldova ´, 2007).In the 2017 election, nine political parties reached the 5% threshold necessary for securing parliamentary representation.The political system was for nearly two decades dominated by two ideological rivals-the Civic Democratic Party (ODS) and the Czech Social Democratic Party ( CSSD).Up to the 2017 election, these two parties have regularly taken turns in leading the government, either alone or in a coalition with other parties, until they were challenged by a new party-ANO 2011.The current minority government consists of two parties, ANO and CSSD, and it relies on the support of the Communist Party of Bohemia and Moravia (KS CM) to get its policies through the legislature.
The relationships between government parties and opposition tends to be confrontational rather than collaborative 1 .While opposition parties get to nominate chairpersons to a number of legislative committees, government parties normally control a majority of seats in all the committees as the seats are assigned proportionally to party size.The descriptive statistics presented in Table 1 lend initial support to this study's expectation regarding access to bill making.Government legislators submit fewer bill proposals than their opposition colleagues, on average, but see more of their proposals passed2 .
A good number of Czech political parties are relatively stable and institutionalised (Outl y and Prouza, 2013).Since the 1996 election, seven parties have defended their parliamentary presence in more than one parliamentary term and six of those have gone through the process of party leader change without splitting or falling apart (Mansfeldova ´, 2011).This level of institutionalisation ensures a degree of routinisation in the distribution of legislative tasks and posts which is firmly in the hands of political parties (Outl y and Prouza, 2013).After the 2006 elections, new parties began to appear on the Czech political map, although many of them have not survived more than one parliamentary term.To ensure that attention is paid to institutionalised and routinised internal functioning, only those parties having been re-elected at least once during the period under study are included in the empirical analysis.The following parties are included in this study: ANO 2011, CSSD, Christian and Democratic Union (KDU-CSL), KS CM, ODS and TOP 09.
The Czech Chamber of Deputies is also a typical representative of the newly democratised legislative bodies of Central Europe (Saxonberg, 2003;Linek and Mansfeldova ´, 2007).The wave of democratisation that swept through the region after 1989 opened up a window of opportunity for a major institutional overhaul.
The ensuing democratic reforms witnessed a significant empowerment of the proportionally-elected parliament and the establishment of a multiparty system.For the newly created political parties, there was a window opportunity to put in place internal gender rules that would be in line with the spirit of the times that called for a greater inclusion and political emancipation.However, the process of democratisation witnessed a sudden collapse of women's descriptive representation, following a swift scrapping of the 30% seat quota that kept women's representation in the communist-era parliament comparatively high and stable (Saxonberg, 2003).As a result, the share of women in the parliament dropped down to 10% in 1990 and it increased only incrementally over the last two decades from 15% in 1996 to 22% in 2017, the last election held to date (Stegmaier et al., 2014).
The reasons why women fell through the cracks of this democratic reform have probably roots in the communist regime's approach to the emancipation of women.While outwardly committing themselves to the ideals of feminism, the communist regime did little in practice to redress the main sources of inequalities between men and women.What is more, it created new avenues where such inequalities could flourish.Women who took up full-time jobs as a part of the regime's resolve to achieve universal employment faced the double burden of still being seen as the primary caretakers in the home (Saxonberg, 2003).On the eve of the 1989 revolution, men were left unchallenged in their dominance of political decision making at every level despite affirmative action.Given this half-hearted commitment to women's (political) emancipation, it is perhaps not surprising that the fall of communism was accompanied by an anti-feminist backlash that saw the return of conservative views that consider home and family as an appropriate realm for women (Saxonberg, 2003).In this context, it is possible that a stereotypically masculine model of an 'ideal politician' has quickly emerged in the newly established (or, in some cases, re-established) political parties.While the relatively low levels of women's descriptive representation support this assumption, little is known about the political fortunes of those women who get elected (Stegmaier et al., 2014).By bridging this gap, this article sheds new light on the opportunity structures faced by female politicians in the relatively new political parties that emerged after the demise of communism.
Owing to its features, the Czech case is likely to generate results that are both generalisable to a larger subset of cases and also of value to those wanting to learn more about the gendered opportunity structures faced by elected officials in postcommunist countries.The systemic features described above are by far not unique to the Czech political system which means that this study's results are likely to be of relevance to other cases where institutionalised parties take turns in being part of government and being in opposition.Moreover, the relatively recent democratisation of the Czech system together with the collapse of women's descriptive representation that accompanied the demise of communism provide an interesting setting for studying the ways in which new democratic institutions can produce gendered outcomes (Gains and Lowndes, 2014).

Data and methods
An original dataset forms the backbone of empirical analysis.The Czech incumbent legislator dataset contains all the lower chamber legislators elected between 1996 (the first post-independence election) and 2017 (the last election held to date).To avoid the risk of including political parties that have not properly institutionalised their internal functioning, only those parties that have been reelected at least once during the period under study are included in the analysis. 3 The resulting dataset comprises 1032 observations at legislator-electoral term level.The dataset is described in greater detail in Tables A1 and A2 in the Online Appendix.
The three hypotheses are tested using linear regression modelling with fixed effects for political party and electoral term.Two count variables are used as dependent variables to test H1, namely the number of bill proposals submitted by each legislator, either individually or as part of a group, and the number of their proposals that are voted through as new laws.A binary indicator for holding a committee chairpersonship for at least six months during any given electoral term is used as the response variable of interest when testing H2.And to test H3, binary indicators for holding PPG chairpersonship and PPG vice-chairpersonship are used.

3
The following parties are excluded from the dataset: ODA and SPR-RSC which were both elected in 1996 but not re-elected in 1998, the Greens that were elected in 2006 but not again in 2010, VVelected in 2010 but dissolved by the next election and U ´svit-elected in 2013 but not contending again in 2017.Unie svobody (US)-though re-elected once in 2002, is also excluded because of the party's gradual disintegration following its re-election.
When is Access to Political Capital Gendered?
The same battery of independent variables is used in all the models to ensure cross-model comparability.The main explanatory variable is legislator sex, measured as a binary indicator where the value '1' represents female MPs.Control variables include legislator pre-election ballot position as a measure of which place within the party hierarchy the legislator currently occupies (Crisp et al., 2013).While preference voting is becoming increasingly popular among the Czech voters, the ballot position one holds still remains the most decisive factor determining one's election chances (Stegmaier et al., 2014).The more electable ballot spot one is given, the clearer the signal that the party wants her/him to be elected.The models also include the following control variables: MP's age; a dummy for having a college degree; and the number of years of legislative service as a measure of legislative experience.These controls are included to make sure that the differences found between male and female MPs are not merely attributable to their background competence or political experience.Descriptive statistics on all the variables included in the forthcoming analysis can be found in Table A3 in the Online Appendix.Fixed effects facilitate meaningful comparisons between female and male legislators that are as free of omitted variable bias as possible.By including fixed effects for legislative term and party, the models only compare those legislators who work together in the same parliament and belong to the same party.In other words, all the term and party-specific factors-such as the size of the party, share of female legislators or particularities of any kind that characterise each particular legislative term-are thus held constant.For example, the models can account for parties that are overly productive or those that might boycott law making as part of their political strategy, or even control for legislative terms that were characterised by unusually low levels of law making.
Since the party's government status is specific and constant for each party and legislative term, this variable cannot be included together with party-term fixed effects due to perfect collinearity.This makes unfeasible the estimation of standard interaction models where the party's government status is one of the controls.This is why two models are presented for each dependent variable of interest, one for opposition and one for coalition parties.Note that this approach does not come with any methodological shortcomings compared to a traditional interaction model and is perfectly equivalent to running an interaction model where all the variables, including the fixed effects, are interacted with the party's government status.The benefit is the possibility to include party fixed effects, which is a necessity given that the political assignments under study are distributed at party level.A series of traditional interaction models together with relevant marginal effects and marginal effects graphs are presented in the Online Appendix (Tables A7 and A8 and Figure 1).These are fully consistent with the results reported below.
Despite the nature of the dependent variables, linear regression models are used.This is done for the sake of having more easily interpretable coefficients.Corresponding negative binomial (for count outcomes) and logistic (for binary outcomes) models are included in the Online Appendix to show that they are consistent with those reported here (see Tables A4-A6).The standard errors used in all models are robust and clustered at party-district level.

Is access to law making gendered?
Hypothesis 1 presented above is that parties comprising the government will favour male legislators when deciding who should be allowed to submit new bill proposals.Since the proposals submitted by government legislators have a good chance to become new laws, they are also more likely to attract attention-not only to the initiator but also his/her political party.For this reason, bill making can be a potent source of political capital for government legislators in terms of access to policy making as well as better visibility.Government parties are therefore more likely to be involved in deciding who should be given this important political assignment.We expect that if the party officers who are in charge of making this decision have an implicit preference for male legislators, it will manifest itself in these conditions.Unlikely to win majority support, the relative value of the bill proposals submitted by opposition legislators as a means of generating political capital through greater exposure is more limited.Opposition parties are therefore expected to be less involved in regulating who gets to submit proposals, which makes the playing field more open.These conditions are likely to benefit hardworking opposition legislators, regardless of gender.Given that past research finds female legislators at least as hardworking as their male colleagues (Anzia and Berry, 2011;Fulton, 2012), it is expected that both female and male opposition legislators will have the same access to bill making.
The models shown in Table 2 by and large confirm this study's expectations concerning access to bill making (H1).Specifications one to three feature the number of bill proposals per legislator as an outcome of interest; while specifications four to six capture the number of new laws per legislator.A quick glance at specifications one and four, where both opposition and government legislators are pooled together in a single model, highlights the limitations of analysing billmaking activity without taking into account the party's government status.The statistically non-significant point estimates for the female indicator in the pooled models would have us believe that there is no difference between female and male legislators in terms of access to bill making.Stratified models captured in specifications two to three and five to six challenge these results, however.Specification two shows that female coalition lawmakers submit, on average, two fewer bill proposals than their male colleagues, while specification three suggests that female and male opposition legislators are equally active on this front.The pattern is the same when new laws are considered.Male government legislators see two more of their proposals passed, on average, when compared to their female colleagues.In case of opposition lawmakers, no gender differences are observed.The models include important controls for legislators' political experience and education.This means that the observed gender differences are not easily attributable to legislators' background qualities.

Is access to senior legislative and party positions gendered?
We now take a closer look at access to the senior legislative committee (H2) and party (H3a and b) offices that tend to come with greater visibility and, in some cases, agenda-setting power.The analysis first zooms in on the appointment patterns to committee chairpersonships (H2).These senior positions come with higher salary, larger budget for staff and better name recognition; and are thus in high demand.However, while the committee chairs appointed by government parties have real influence over their respective committee's decisions; their opposition counterparts do not control a voting majority in any of 'their' committees and have therefore mostly administrative responsibilities.The chairperson positions distributed by government parties can thus more easily be turned into political capital.Table 3 shows that the committee chairpersonship appointment patterns appear to be in line with this article's expectations (H2).Female government legislators are less likely to be nominated by their parties as committee chairs, while their opposition counterparts are as likely as their male colleagues to secure a nomination.The point estimate for female coalition legislators is less than a tenth of a decimal point shy of reaching the conventional statistical significance at 10%.But the trend set by both stratified models is in line with the study's expectations.
Finally, there are two kinds of PPG leadership positions that this article looks at: chairpersonship (H3a) and vice-chairpersonship (H3b).The former is a highstakes position within each party regardless of the party's government status that comes with considerable influence over agenda setting and party position formation.The latter is largely symbolic and hence low-stakes.Specifications two to three in Table 4 confirm that both female government and opposition legislators are approximately five percentage points less likely to become chairs of their PPGs when compared to their male colleagues.In the case of PPG vicechairpersonship, which is more of a symbolic position, female opposition legislators seem to have an advantage over their male colleagues (specification six), while female government legislators seem to be as likely to get the position as their male colleagues.Note that this is also in line with this study's expectation that opposition parties might be more willing to give those senior positions that come with more limited formal power to female legislators as a way of visually signalling that they represent an alternative to the sitting government.

Conclusions
The parliamentary workplace can produce gendered outcomes even if it is governed by a comprehensive set of formal rules that are seemingly gender-neutral (Erikson and Verge, this volume).Past research shows that a long history of male dominance in politics has given rise to informal norms and rules that help to perpetuate the status quo by making male politicians appear as more qualified and more worthy of being invested into in the form of trust and/or additional political opportunities.This study has investigated whether these dominant masculine constructions of an ideal politician limit female MPs' access to important legislative assignments when the value of these as a means of generating political capital is high.Utilising the variation in influence over political decision making and visibility brought about by political parties' status as either government or opposition, this article has examined gender differences in access to bill making and senior legislative offices in high-and low-stakes conditions.The study anticipated that if the senior party officers who oversee the distribution of political assignments and duties have an (implicit) preference for male politicians, it will most likely manifest itself when their party is power.The empirical results lend support to this expectation.Male legislators appear to have more privileged access to important sources of political capital when the party they belong to is in government but not when their party is in opposition and when the relative value of this access is limited.This study's results improve our knowledge of the extent to which gender bias affects the political careers of female politicians.Understanding the dynamics of access to important legislative assignments from a gender perspective can, for instance, help us to better understand whether the pathway to political seniority is gendered (Folke and Rickne, 2016).Existing literature provides ample evidence that legislators' parliamentary activity can improve their chances of being renominated by their parties and getting re-elected (Da ¨ubler et al., 2018;Papp and Russo, 2018).Securing a good renomination outcome is particularly important in PR settings where the party-determined ballot position still is the most critical determinant of one's (re-)election chances (Crisp et al., 2013).If access to important legislative assignments is more restricted for female MPs, as this study's results suggest, then women might have to work harder to maximise their chances of getting re-elected.This is problematic because re-election tends to be a gateway to political seniority and therefore political influence (Folke and Rickne, 2016).Also, female MPs' ability to gain meaningful political power and to act on behalf of women is likely to be limited if they lack access to important legislative assignments.
While this study shows that female MPs are disadvantaged in their access to important sources of political capital, it comes short of identifying which exact combination of mechanisms governs this association.Three tentative mechanisms are offered here.The first one centres around the existence of social norms that define politics as a social sphere that is unsuitable for women (Gains and Lowndes 2014).If the political arena continues to be viewed by society as a hostile masculine domain where backstabbing and conflict are common and where masculine traits and values are the best form of ammunition, then choosing visible party representatives that embody these views can be a rational political strategy.Second, masculinity-infused ideas about what constitutes an ideal politician can also be rooted in the dominant culture of the political workplace that is gendered (Erikson and Josefsson, this issue).This holds true especially in those cases where the political arena has traditionally been dominated by men.If senior political positions have for a long period been predominantly held by men, then the traits these office holders possess can easily become synonymous with the office itself (Verge and Claveria, 2018).Finally, the preference for male politicians by senior party officers in high-stakes conditions can also boil down to homosocial reproduction.Homosociality, that is, the preference for members of one's own sex, is common in social contexts which are characterised by high levels of unpredictability and volatility and where one particular sex is over-represented.It has been shown that, in such conditions, critical actors develop insider networks of trust along homosocial lines (Bjarnega ˚rd, 2013).The more senior members use these networks to solidify their grip on power by distributing political opportunities to the junior members who are expected to reciprocate by being loyal.It is not counterintuitive that these homosocial networks will be at their strongest when stakes are high and when the party officers need to make sure that potent sources of political capital are placed in dependable hands.It goes beyond the scope of this study to establish how these mechanisms interact with one another in producing the observed outcomes or whether additional mechanisms might also be at play.Instead, the article opens avenues for further research, which can, through indepth qualitative study, identify exactly which informal rules produce the observed outcomes.
How do the results travel?This study takes the entire Czech post-independence democratic history into account and is thus comparative in nature.It comprises six political parties, six electoral terms, seven elections and more than 1000 observations, which makes its results particularly robust.The nature of the Czech case as a typical representative of a polarised multiparty system where the relationship between government parties and opposition is relatively confrontational makes plausible the expectation that the dynamics identified here can be found in other similar cases.This holds true especially in those political contexts where politics continues to be seen as a domain that is unsuitable for women.The added value of the Czech case lies in its relatively recent transition to democracy, which allows this study to contribute to our knowledge of the ways through which relatively new political institutions can produce gendered outcomes (Gains and Lowndes, 2014).Finally, the approach developed by this article, which exploits the high-and low-stakes conditions created by being in or out of power as a means of chiselling out informal norms, can be of use in other political systems where the relationship between government parties and opposition is confrontational.

Table 1
Descriptive statistics on bill making in the Czech Chamber of Deputies

Table 2
Access to bill making

Table 3
Appointment patterns to committee chairpersonship positions