How Are Political Institutions Gendered?

Introduction

What does it mean to say that political institutions are ‘gendered’? This article identifies the mechanisms at work, how they are consolidated over time and how they are disrupted via resistance or reform. The article proposes research strategies for investigating the micro-foundations of gendered political institutions.

A focus on institutions enables gender scholars to study the ways in which gender gets inscribed into the very rules of political life, shaping the behaviour of political actors (male and female) over time, often in opposition to the content of specific policies (Chappell, 2010). Feminist institutionalism has focused attention upon the ways in which institutions shape political behaviour and outcomes in gendered ways (Krook and Mackay, 2011). Political institutions reflect but also help constitute the roles, relations and identities of women and men in the political arena (whether voters, campaigners, service users, public servants or politicians). Research has focused primarily on those institutional arrangements designed to promote gender equity and women’s interests, such as women’s policy agencies (Outshoorn and Kantola, 2007) or gender quotas (Krook and Norris, 2014). There exists, however, a major research agenda on the ways in which the organisation of political life more generally is gendered and how gender is implicated in processes of institutional design and change. As Fiona Mackay (2011: 181–193) argues, ‘apparently gender-neutral political institutions have differential effects on women and men’, noting also that gender remains a ‘relevant analytical category’ in institutional contexts where women themselves are absent.

Feminist scholars’ interest in institutions relates also to their importance within strategies for gender equality. Changing institutions is a prize beyond that of securing a particular policy change or placing more women in political office. It offers feminists the opportunity to recalibrate the ‘rules of the game’ and thus shape the parameters for future decision-making. As Htun and Weldon (2010: 212) argue, ‘institutional capacity affects policy no matter who is governing’. And, while changing institutions is not an easy task, it is certainly a more plausible (or immediate) option than dismantling deep structures of patriarchy. Bo Rothstein (1996: 138) explains the opportunity thus: ‘If social norms . . . vary with the character of political institutions, then we can at least to some extent decide which norms shall prevail in the society in which we live’.

To assert that political institutions are ‘gendered’ suggests that concerns of gender run right through – like the rings inside a tree trunk – and cannot be pinpointed in one place (a women’s policy agency, for instance) or with one group of people. But, if gender is everywhere, where do we go to look for it? While much institutionalist scholarship is focused on big structures (legislatures, executives, transnational agreements) or entire policy systems (welfare states, social security, taxation), this article takes on the challenge of developing analytical and methodological strategies for researching the micro-foundations of how political institutions are gendered. A focus on micro-foundations means studying the encounter between actors and institutional rules within particular political settings (while recognising that these are influenced by wider institutional contexts). The analysis uses a definition of political institutions as the formal and informal ‘rules of the game’ or, more specifically, the ‘shared prescriptions about what actions are required, prohibited or permitted’ (with associated sanctions) (Ostrom, 2011: 7). The contribution of this article is to develop strategies for unearthing institutional rules that distribute power along gendered lines. We are concerned with strategies for dragging such rules out from within complex and embedded institutional networks and subjecting them to scrutiny.

The purpose of the article is theory generation, and it employs a process of reasoning that iterates between theoretical propositions and empirical observations – drawn from gender scholarship in comparative politics – with the goal of finding the simplest and most likely explanation for them. However, while yielding plausible conclusions, the article does not seek to positively verify them. Rather, our model identifies key variables and posits relationships between them, as a precursor to new programmes of empirical research. Gender is understood as shaping power relations but within the context of intersecting social categories and identities, notably associated with race, social class, disability and sexuality. However, a limitation of this article is that it does not provide a systematic analysis of such intersections (although they are referred to), pointing to the need for further theorising and research. At the same time, transgender studies offer a more complex understanding of gender than the one offered in this article, opening up avenues for future research.

The article starts by putting forward a simplified model of how political institutions are gendered, identifying the part played by gendered rules, gendered actors and gendered outcomes (building on Gains and Lowndes, 2014). The article then asks a series of questions about the component parts of the model and their inter-relationships. First, we ask how it is possible to identify gender-relevant rules within political settings, elaborating both threshold criteria and varied presentations (including informal rules). Second, we ask how these rules work to constrain and empower actors along gendered lines, specifying mechanisms of regulation, obligation and persuasion. Third, we ask how seemingly neutral rules can generate ‘gendered effects’, focusing on their interaction with institutions outside the official domain of politics (in wider society and the economy). Finally, we focus on gendered actors themselves, looking at how they reproduce and consolidate gendered institutions through enacting rules but also generate change through adapting, resisting or intentionally reforming institutional rules. Exploring these questions enables us to build a second, more comprehensive version of the model.

How Are Political Institutions Gendered? A Simplified Model

The simplified model presented in Figure 1 identifies four sets of variables: rules about gender, rules which have gendered effects (but are not specifically about gender), the gendered actors who work with rules and gendered policy outcomes. These are explained in turn below:

OPEN IN VIEWER

Figure 1.

The Simplified Model.

By distinguishing between these four variables, the model offers comparativists a set of concepts that are applicable in a wide range of political arenas, in more than one country and across different time periods, thus facilitating the exploration of similarities and differences. In the simplified model, relationships between the variables are represented in a non-directional cycle to show their iterative and mutually reinforcing character. The four factors represent different entry points for comparative research or indeed research within a single political system. Having established the basic model, we now consider in more detail its component parts, how they might be operationalised, and how they interact. We start by asking how, in the context of political institutions, we can identify gender-relevant rules on the ground.

How Can Gendered Institutional Rules Be Identified? Threshold Conditions

We work here with core concepts from new institutionalism, rather than pinning our colours to the mast of any particular school. Institutions are defined as the rules of the game, within which both individual and collective actors are the players (who may at times seek to change those rules). As such, institutions are not the same as organisations. Organisations constitute collective actors which are subject to (and influencers of) institutional rules. The rules of a specific legislature, for instance, shape the gendered behaviour of both political parties (e.g. candidate selection) and individual politicians (e.g. committee membership). However, on departing from a ‘brass name plate’ conception of institutions, researchers are left with the challenge of how to identify their object of analysis. The meso level of analysis, characteristic of historical institutionalism, sees institutions as broad policy regimes, which are ‘big, slow-moving and . . . invisible’ (Pierson, 2003). New institutions are expected to emerge only at critical junctures where path dependency is ruptured. There is a considerable body of work from a gender perspective that operates within this frame; for instance, Waylen’s (2011) comparative work on the gendered implications of democratic transitions. But we argue here for the advantages of a micro-level approach for unpicking how gender becomes inscribed over time in political institutions like legislatures, executives, municipal governments or public services (and, indeed, how these processes can be challenged). While aggregate effects are of course important, a focus on micro-foundations helps us, in Joni Lovenduski’s (2009) memorable phrase, to ‘nail the bias’ – to surface and scrutinise those institutional rules that distribute power along gendered lines. Our focus is upon what Ostrom (2005) calls the ‘action arena’, where actors encounter institutional rules within specific political settings; such rules shape actors’ behaviour but are also subject to challenge and change under particular conditions. Reflecting long traditions in feminist analysis, we locate these action arenas within a broad institutional context, examining interactions with gendered institutions in wider society.

Identifying institutional rules requires the specification of threshold conditions – ways of determining what is (and is not) an institutional rule. Specifying threshold conditions is an important theoretical and methodological task because it allows us to address the critique that, by admitting informal institutional elements, institutionalism becomes non-falsifiable. Feminist institutionalists need to take seriously Rothstein’s (1996: 145) challenge that, if the concept of institution ‘means everything, then it means nothing’. We need to be able to distinguish political institutions from other social facts that shape gender in/equality. The following threshold conditions apply equally to formal and informal institutional elements and are operationalisable in diverse research contexts. Such institutional rules are:

These threshold conditions allow us to counter the criticism that institutions can ‘mean everything’ and therefore lack leverage in explaining gendered processes and outcomes in politics. The conditions have been formulated to cover both formal and informal rules. It is important to reject the assumption, sometimes present in the literature, that new institutionalists are only concerned with informal institutions (in contrast to the old institutionalists who studied formal organisations and structures). A key insight of new institutionalism has indeed been the recognition of the role of informal conventions in structuring political life. We argue, however, that the conceptual and empirical prize is to understand the relationship between formal and informal elements, how this differs between contexts and over time, and the ways in which actors seek to shift this relationship for strategic reasons (whether through reform, resistance or gradual change). Nowhere is this formal/informal relationship more significant than in the gendering of political institutions. We need to record and analyse the varied, and interacting, presentations of gendered rules. Researching political recruitment, Bjarnegård and Kenny (2015: 751) argue that ‘informal rules may reinforce change, ensuring that formal procedures are actually complied with’, noting that the effective implementation of legal quotas is more likely among ‘parties that have had informal arrangements for alternating male and female names on candidate lists’. But they also point to ‘informal practices of quota subversion, ranging from running women in “no hope” seats . . . to practices of local patronage’.

It is important to challenge a view of formal and informal institutions as separate entities, existing in a binary relationship. Scholars discuss informal rules as ‘raw material’ for formal institutions (e.g. Knight, 1992, from a rational choice perspective) or as significant but outside the scope of their research (e.g. Streeck and Thelen, 2005, from historical institutionalism). Comparativists Helmke and Levitsky (2004) consider the ways in which informal institutions interact with formal arrangements through complementary, accommodating, competing and substitutive relationships. In all these cases, the epistemological starting point is formal institutions, and there is an assumption of a clear demarcation between formal and informal elements. Ostrom’s concept of rules-in-use provides an alternative approach, and one that is particularly suitable for analysing how political institutions are gendered. Rules-in-use are not just informal; some kind of opposite to rules-in-form. Rather, they are a combination of both formal and informal elements (specific to a political setting) that can only be determined empirically. Rules-in-use are best described as the distinctive ensemble of ‘dos and don’ts that one learns on the ground’ (Ostrom, 1999: 38). They cannot be deduced theoretically. The challenge for feminist institutionalism is less to broaden the institutionalist lens to include the informal, and more to investigate the interaction between formal and informal institutional elements, and how this (shifting) interaction operates to advantage or disadvantage women (and can constitute a point of intervention for gender equality reform). Indeed, research has shown how informal gendered conventions may operate to frustrate or dilute the impact of formal gender equality reforms like quotas (Freidenvall and Krook, 2010). The interplay between formal and informal rules is a dynamic process; for instance, male-dominated political elites may shift the locus of power from formal to informal mechanisms in order to counteract women’s increased access and presence in formal decision-making sites.

In seeking to better understand the components and dynamics of such ensembles, we need to move beyond ‘formal’ and ‘informal’ as descriptive labels and towards clearer explanatory statements about how different types of institutional rules do their work. Having established threshold criteria for identifying rules, we now seek to classify relevant rules according to the ways in which they shape actors’ behaviour (and may reinforce or pull against one another). Such a classification provides a conceptual map for gender scholars seeking to identify and challenge relevant institutional rules (and compare them across different political settings and over time), while also deepening our understanding of the nature of institutional constraint itself.

How Do Rules Shape Behaviour on Gendered Lines? Regulation, Obligation and Persuasion

The purpose of our investigation is to establish how political institutions differentially shape action, opportunities and benefits along gendered lines and how such patterns become (relatively) systematic, stable and predictable over time. When rules meet the threshold conditions specified in the previous section, they have the capacity to regularise the gendered distribution of power – over and above the preferences of particular actors or, indeed, the stated intentions of specific policies or organisations. Institutional rules that empower certain actors generally limit opportunities for other actors; in short, institutional constraints have distributional effects. Every rule that specifies who has power (to be a member of a political community, exercise political rights, make political decisions or receive public services or benefits) also signifies who does not have power. Such power settlements are not ‘once and for all’ but are rather contested and renegotiated over time. While the rules of political institutions are, in so many cases, exclusionary of women and/or women’s interests (and reinforcing of masculine advantage), institutional reform also has the potential to ‘rule women in’. At the same time, when institutional privileges are extended to certain groups of women, they may – via formal or informal means – exclude other women on the grounds of, for instance, marital status, social class, sexuality, ethnicity or age. For example, initial moves towards female suffrage in many countries included only married women of property-owning classes; a pattern reproduced in early legislation on abortion, adoption and fertility services. Research on the suffrage movement in the United States has documented the exclusion of women of colour by white activists who argued that challenging ‘rules about gender’ would be easier if ‘rules about race’ were put to one side. Interestingly, nineteenth-century civil rights activists also argued that promoting women’s rights could be detrimental to gaining institutional reform in favour of black men (Buechler, 1990: 22).

Historically, and across political jurisdictions, women have struggled (and continue to struggle) to access political opportunities – not on a one-off basis, or dependent on special pleading or special circumstances – but as a matter of institutional entitlement, which itself is not just formal but substantive. Understanding the circumstances within which inclusionary rules ‘stick’ or ‘slip’ is an important task. We need to understand more about how institutions shape behaviour in order to grasp the conditions of both their tractability and their intractability. Gender reform, after all, requires an ongoing programme of both institutionalisation and deinstitutionalisation. Reformers need to understand what makes institutional rules stick, in order to design and embed new inclusionary political institutions. But they also need to appreciate how rules slip, in order to guard against the attrition of new rules (through adopting appropriate forms of institutional maintenance) while simultaneously working to dislodge those tenacious rules that fortify male privilege (in whatever formal/informal mix).

Understanding these gendered processes requires that we analyse the different ways in which institutional rules shape behaviour and how different types of rules may combine to reinforce one another (or not). Rejecting the invitation of the dominant institutionalist literature to ‘choose between’ causal mechanisms, we argue that institutions actually shape behaviour through a combination of regulative, normative and discursive constraints, a mix that varies in different cases and contexts. Given that political actors and institutional designers are themselves subject to mixed motivations, we have to be aware of the artificial nature of ‘pure’ ontological constructs. As actors encounter institutions, their response is likely to be shaped by (some combination of) their selfish interests, their ‘need to belong’ and their comprehension (of not) of institutional discourses (Lowndes and Roberts, 2013). Understanding how political institutions are gendered is only possible if we recognise these simultaneous and interlocking modes of constraint. For instance, the communication of persuasive discourses about the impact of sexual harassment on women in public life via the ‘Me Too’ movement may turn out to be more effective at dislodging embedded patterns of behaviour than changing laws or policies; or such discursive mechanisms may serve over time to deepen the impact of these policies and help ensure compliance.

Rather than trying to win an argument about ‘what counts most’, the real prize for feminist institutionalism is to better understand how distinctive modes of constraint work together in practice and to establish what this means for processes of institutional design and change. It is the combination of multiple modes of constraint that gives institutions their capacity to endure over time and to secure actors’ loyalty rather than having to rely on force- or incentive-based compliance mechanisms. What Scott (2008) calls the ‘over-determination’ of institutions is a key problematic when considering gender. Research shows time and again the limitations of formal regulation as a means of combatting male privilege or, indeed, of promoting gender justice. Because institutions also shape behaviour through normative and discursive mechanisms, the impact of new ‘positive’ rules (like the introduction of women’s policy agencies or party quotas) or ‘negative’ rules (like the removal of legal obstacles to women’s political advancement or recognition) are likely to be filtered through, or at least articulated with, the ongoing effects of other more subtle forms of constraint. This overdetermination presents a challenge not just for scholars seeking to unpick the ways in which political institutions are gendered over time (and in the face of formal rule change) but also for activists and reformers. Interventions that target normative and discursive, as well as regulative, forms of institutional constraint have a greater likelihood of shifting institutional power settlements in women’s favour. An appreciation of the differentiated texture of institutional constraint can help us explain (and even predict) opportunities for institutional change.

Our ecumenical theoretical approach enables us to distinguish, rather than choose, between three types of institutional rule:

It is through these three different types of rules that actors know what is expected of them; they are the (different) means through which institutions shape behaviour. If we are not alert to the potential role of each of these forms of constraint/empowerment, our research is unlikely to unveil the gendered way in which political institutions work or to generate effective strategies for resistance or reform. Research is needed in three main areas. First, how do institutions regulate actors’ behaviour to produce gendered effects? We need to consider how laws, standards and protocols produce such effects and identify the rewards and punishments associated with compliance/non-compliance. Second, how do institutions obligate actors to behave in gendered ways? Here we should examine how gendered logics of appropriateness operate through routines and practices and how are they enforced through informal means (e.g. displays of disapproval, social isolation, threats or evens acts of violence). Third, how do institutions narrate forms of gendered behaviour and hence legitimise gendered political outcomes? How is political life framed in such a way as to ‘organise in’ or ‘organise out’ certain arguments on a gendered basis? Different research strategies are needed to investigate each type of rule, and mixed methods may be most likely to capture their uneven and compound effects.

There is considerable theoretical and methodological value in separating out different types of institutional constraint (Lowndes and Roberts, 2013: 69). The typology helps us to differentiate the causal mechanisms at work, even though their effects are likely to overlap in practice. Indeed, we need to consider how they combine to produce stability in gendered institutions over time. When all three types of institutional rule work together to embed gender inequality, it will be very hard to shift the resulting power settlement. Where institutional designers do have the chance to promote gender equality, they are well advised to consider more than the construction of regulatory rules, seeking also to model new obligatory and persuasive rules (backed up in each case by appropriate sanctions or enforcement mechanisms). To take an example, Piscopo (2017) shows how female politicians and activists in Mexico have set out to ‘leverage informality’ in order to make election quota and parity laws more numerically effective, thus responding to the informal conventions of ‘quota subversion’ used by party elites to maintain male dominance (e.g. running women in losing districts and manipulating rank-order rules on electoral lists). Reformers have sought to establish obligatory and persuasive rules that are powerful enough to change party behaviour (enforced by sanctions of ‘shaming and punishing’), while also pressurising electoral officials to rewrite regulatory rules to compel parties to distribute the top candidacies evenly between women and men. Waylen (2011: 214) points to the ‘potential dynamism and mutability’ of informal institutions in securing greater gender equality; our analysis adds further depth through its plea to disaggregate the ‘informal’ according to the causal mechanisms at work.

How Do Rules with Gendered Effects Emerge through Institutional Interactions Beyond The Political Domain?

Our simplified model (Figure 1) shows that political institutions shape action, opportunities and benefits along gendered lines not only through rules that are specifically about gender but also through ‘rules with gendered effects’, which emerge from the interaction between rules in the political domain and those in the wider institutional context. As noted earlier in the article, while direct discrimination is based upon having a different rule for men and women, indirect discrimination arises when the same rule is applied to everyone but has different and negative impact for women. While this is a familiar formulation, our contribution is to provide a specifically institutionalist explanation of how such gendered effects are sustained and how they can be disrupted as part of gender equality strategies. Rather than rely on generalised notions of culture or social norms, we need to investigate articulations between political and ‘non-political’ institutions. ¹ Our argument is that these articulations are not necessary or inevitable but have to be actively maintained in order to secure gendered institutional settlements. By the same token, gender equality campaigners can act strategically to disrupt such connections and forge other more productive linkages.

Rather than seeing political institutions as organisations with hard boundaries, we need to look at them as sets of rules that are intertwined with rules emanating from beyond the political domain. As Ostrom (2005: 115) points out, ‘rules operate configurationally rather than separably’. Processes of regulation, obligation and persuasion operate across these boundaries. Cutting loose the concept of institution from its traditional organisational moorings has particular advantages for gender scholars. Such an approach encourages research on the ways in which political institutions span different action arenas and shape behaviour beyond the confines of particular organisations – as in, for example, Annesley and Gains’ (2010) work on the ‘gendered disposition’ of the UK core executive, which is made up of multiple organisations responsible for the coordination of central government policies. We are also reminded that feminist strategies for institutional change need to be directed not just at individual organisations but also at the rules and conventions that are shared between them or shape their interaction.

Such an approach enables us to explore the interconnections between rules-in-use in political and ‘non-political’ domains. In his classic text, Karl Polanyi (1992) establishes that the economy is embedded in ‘institutions economic and non-economic’; similarly, we argue that politics is embedded in ‘institutions political and non-political’. While institutionalists commonly point to the vertical ‘nesting’ of political institutions – with operational rules constrained by higher level legislative and constitutional rules (Ostrom, 2005) – we also need to pay attention to the ways in which they are articulated horizontally with institutions that structure wider social, economic and cultural life. Sylvia Walby (2004: 22) makes a similar point regarding national-level ‘gender regimes’, arguing that ‘the different domains’ of any regime are ‘interconnected in practice, not sealed into separate compartments’. Taking a micro-level approach, our focus is on the way in which these interactions generate ‘rules with gendered effects’ in the domain of politics, whereby seemingly neutral rules of political life have systematically differential effects upon women and men. Through the threshold conditions explored earlier in the article, we exclude from our definition of political institution both personal habits and generalised notions of culture and social norms. It is important that gender scholars do not ‘smuggle in’ such elements to their understanding of political institutions. Instead, we should problematise the relationship between the rules of political institutions and those of the wider economy and society; thus, we can examine how these relationships generate gendered effects in politics and why they should be of concern to equality reformers.

Feminist institutionalists have shown how institutional rules (generally informal) about caring responsibilities and domestic labour, which are rooted in the domestic arena, influence the development and interpretation of political institutions. To take a specific case, Meryl Kenny’s (2011, 2013) research on candidate selection for the Scottish Parliament reveals that the relevant rules-in-use were quite different from the rules-in-form and, moreover, had strikingly gendered effects. Kenny found that formal job descriptions and person specifications for candidates, which had been developed by the Scottish Labour Party, were technically still in place but were not actually distributed (despite requests from candidates). Kenny argues that female candidates often failed to be selected not because they were less ‘qualified’ but because of the way in which qualifications for political office were understood. Kenny observed that selectors’ preferences for a ‘local’ candidate were far from objective; experiences in community activism (e.g. related to health and childcare) were judged as less significant than a track record as a local government councillor or experience in a business association or trade union. The research reveals a highly gendered consideration of what constituted relevant links between political and ‘non-political’ domains.

Because institutional rules exist within, between and around particular organisations, we can expect to see rules inside and outside the political domain interact and co-evolve. Indeed, change in ‘non-political’ domains may be an important source of institutional change – either destabilising or reinforcing political rules, or providing novel templates for their redesign or adjustment. The same actors move between different institutional arrangements, bringing with them their personal values and ideas and their experience of negotiating institutional constraint in ‘non-political’ domains. Institutional resistance, as well as redesign, may emerge from these trans-institutional encounters; indeed, ‘institutional recombination’ (Crouch, 2005) is a typical outcome. The introduction of same-sex marriage in many jurisdictions is witness to the mutability of even the most embedded public institutions, under pressure from changes in wider social sensibilities and sustained social movement activism. Yet this reform has taken place within an institutional terrain marked by the privileged status of heterosexual marriage. Actors, gay and straight, confront the gendered roles of ‘husband’ and ‘wife’ (and associated power distributions) within the new recombinant institution of more inclusive marriage – roles which may over time be reinterpreted or further embedded. In short, political institutions are not insulated from changing gender relations in the home, workplace or civil society. As Krook and Mackay (2011: 7) note, ‘political institutions are themselves constituted by these embodied social practices of “doing gender” on a daily basis. Recognising these reciprocal relationships, in turn, opens up possibilities for agency and change’.

Returning to Kenny’s research on candidate selection, it is clear that attempts to broaden the field were obstructed by gendered practices of boundary construction, which sought to ‘determine an institutional “inside” and “outside” by privileging certain institutional connections over others’ (Lowndes and Roberts, 2013: 167). It is not a matter of chance when a female candidate’s experience is discounted; rather, it has to be discounted in order to establish what does count. As argued throughout this article, political institutions have dense and sticky connections to ‘non-political’ institutions; challenging these connections, or seeking to establish new ones, is a subject of strategic action and political struggle. In UK government, there has been a close and enduring articulation between cabinet membership and specific fee-paying schools and elite universities, which has favoured upper class men (and some women). Challenging this connection could help to disrupt traditional gendered leadership styles and priorities, alongside the crafting of new institutional articulations that are more empowering of (more) women. While it is important not to essentialise women’s engagement with caring institutions, the absence of close institutional connections between care and politics is one way in which women are endlessly ‘ruled-out’ of political opportunities. We can only imagine a situation where the experience of combining work and caring responsibilities was seen as an asset rather than a liability for representatives and leaders (female or male), or where caregivers’ privileged access to street-level public services (schools, clinics, parks) was recognised as a source of specialist knowledge and expertise.

The character of these institutional articulations is potentially as important for gender equality as the design of new institutional rules specifically ‘about’ politics. Institutional configurations (as well as individual rules) distribute power in gendered ways but in a dynamic context in which actors seek to influence how institutions are articulated with each other. Such action may seek to consolidate the gender status quo or to challenge existing connections and forge new ones, which are more empowering of women in political spaces. To deepen our understanding of how seemingly neutral rules have gendered effects, we need to problematise institutional boundaries, recognising what Crouch (2005: 16) calls the continuum between ‘internal and external, endogenous and exogenous’. Indeed, feminist institutionalists advocate the use of highly contextualised research strategies which recognise that gender equality innovations are ‘nested’ within a dense institutional environment and subject to important interaction effects (Mackay, 2014). Research of this sort can inform feminist strategies for institutional change. Political institutions will never realistically be ‘de-gendered’, but there may be opportunities for their ‘re-gendering’ through the cultivation of new, empowering institutional linkages, alongside the progressive subversion of the established inside/outside boundary. Comparative research (between countries and over time) shows not just that ‘institutions and politics are gendered but also that they can be gendered’, pointing to the work of political actors – representatives, activists, movements and parties – ‘to instate practices and rules that recast the gendered nature of the political’ (Beckwith, 2005: 133, emphasis added). This is not (just) about engaging more women in politics or even about ensuring ‘women’s interests’ are addressed; it is, far more fundamentally, about investigating how masculinities have been inscribed into public institutions and how strategic agency can be mobilised ‘to feminise political structures, rules, and norms . . . literally to regender state power, policymaking, and state legal constructions and their interpretations’ (Beckwith, 2005: 133).

In summary, to understand how political institutions are gendered we need to research their articulation with ‘non-political’ institutions in society and the economy, the ways in which these articulations create gendered effects and the opportunities for disrupting and reconfiguring institutional linkages in women’s interests. Just as the broader institutional environment shapes the political domain in gendered ways, politics also influences the ongoing evolution of gendered institutions in the wider society and economy.

Gendered Actors Working with Rules: Interpretation, Adaptation, Resistance and Reform

So far, we have focused upon the ways in which institutions shape political behaviour and outcomes along gendered lines. But, while institutions constrain behaviour, they are at the same time human constructions. We can see this most clearly in grand foundational moments (writing constitutions, designing legislative structures, crafting international agreements) and in acts of major institutional reform. Yet it is also apparent in the day-to-day enactment of political institutions by agents. Rule-in-use are only ‘in-use’ because actors use them. In using rules, actors inevitably engage in acts of interpretation. In forging the new institutionalism, March and Olsen (1989) explained how institutional precepts were continuously interpreted in relation to the different cases that institutional actors encounter, in a manner similar to legal reasoning. Actors reproduce rules as they use them, thus generating the regularities and stability associated with institutions; but interpretation and enactment are also sources of institutional change. Actors have to fit cases to rules and, in so doing, are likely to adapt those rules or even rely on informal rules where formal rules are inadequate or inappropriate. In politics, both institutional change and institutional stability emerge out of the dynamic encounter between institutional actors and institutional rules (within contexts that are crammed full, awash even, with other institutions). We argue that gender scholars need to engage specifically with the interaction between actors and rules, which is inherently creative but also contested. Power does not just lie in the active enforcement of rules, but also in everyday contests over their interpretation, and in the context of struggles to adapt or change rules in favour of greater equality.

The most obvious research question here is whether the gender of actors who work with institutional rules ‘matters’. A feminist institutionalist approach reminds us that institutional actors are real people (Mackay et al., 2011). All rules are encountered by gendered and embodied agents, with attendant variation in how they are understood and interpreted. It is important to pay attention not only to how individuals act, but also what these actions mean for them. Enacting formal and informal gendered rules may have a different meaning for women than for men; for example, women parliamentarians are often expected to ‘do’ the performance of a politician – which is coded as masculine – while also ‘managing’ their femininity (Lovenduski, 2005).

Drawing on Pitkin’s (1972) notion of substantive representation, feminist scholarship has moved away from assuming an automatic link between an increase in the ‘descriptive representation’ of women (e.g. in legislatures) and policies to benefit women or promote gender equality. The assumption that women will automatically ‘act for women’ is problematised, acknowledging that not all women will take this route and that male actors can also advocate for gender equality. However, the literature consistently shows improved outcomes for women where descriptive representation is improved, suggesting that at least some women political actors will seek to act for women either through pressurising decision makers in executives or directly through legislative activity (Annesley et al., 2015). Research also shows that formal or informal collaboration among women representatives can increase their influence within legislatures (Holman and Mahoney, 2018), although this impact may be mediated by the character of electoral systems, links with wider women’s movements and the role of individual critical actors (Celis et al., 2016). Outside of the legislative environment, Gains and Lowndes’ (2018) research found that female police commissioners in England were twice as likely as male commissioners to prioritise policies to tackle violence against women. However, those commissioners who adopted a strong interpretation of their gender equality duties (what we have called here ‘rules about gender’) were 2.4 times as likely to prioritise such policies, regardless of their own gender. Such conclusions suggest that maximum analytical leverage can be gained from looking at how gendered actors work with institutional rules (rather than focusing on women’s absence or presence alone).

Moving beyond a focus on women representatives, Judith Squires (2008) has argued for a focus on the ‘constitutive representation of gender’. This approach stresses the multiplicity of gendered identities among both women and men and asks ‘not whether but how gender relations are constituted through representative processes’ (Squires, 2008: 188, original emphasis). Squires (2008: 195) argues that ‘gender experts’ and ‘femocrats’ within women’s policy agencies have ‘come to be the privileged speakers for women’s interests’, thus shaping policies that benefit small female elites. She shows how the ‘institutionalised feminism’ of the Swedish state has constructed gender in such a way as to benefit dominant groups of women and invisibilise the identities and experiences of migrant women (Squires, 2008: 195). Focusing on intersectionality in a US context, Wendy Smooth (2011) asks the question: ‘Standing for women? Which women?’ Her research with African American state legislators argues that the very definition of ‘women’s interests’ is subject to intersecting effects of racism; she cites one legislator as arguing that criminal justice has become a ‘women’s issue’ because of the impact on mothers, partners and other female family members of the high incarceration rate among Black men (Smooth, 2011: 436).

Thus, the gendering of political institutions is a relational phenomenon; it is about the distribution of power between and among gendered actors. At the same time, ‘relational’ does not imply a binary or zero-sum game. Institutions reflect, and also constitute on an ongoing basis, multiple gender identities among (as well as between) women and men, as demonstrated by Connell and Messerschmidt’s (2005) classic work on hegemonic/subordinate masculinities and by more recent research on ‘institutionalised femininities’ (Paechter, 2018). Observing British women parliamentarians, Puwar notes that ‘while adhering to the social rules of femininity . . . they have to make sure that they are not too feminine’. She argues that many cultivate a ‘fine balanced fusion of femininity and masculinity’ when seeking to operate within an institution that is based on ‘masculine performance’ (Puwar, 2004: 75). We can add to Mahoney and Thelen’s (2010: 8) description of political institutions as distributional instruments ‘laden with power implications’, by noting the gendered character of those power dynamics. The gendering and re-gendering of institutions are ‘active processes with palpable effects’ (Hawkesworth, 2003: 531), animated by embodied institutional actors (with intersectional identities) who interpret, adapt and contest dominant rules.

With her collaborators, Kathleen Thelen has breathed new life into studies of institutional change with her plea that researchers investigate forms of change that take place outside moments of breakdown/replacement. Because ‘dynamic tensions and pressures for change are built into institutions’, transformation can occur through strategies that fall short of all-out reform, like institutional layering or conversion (Mahoney and Thelen, 2010: 15–18). Such ‘below the radar’ change may be particularly important from a gender perspective, especially for those policy areas (like reproductive rights) that stimulate ‘doctrinal resistance’ on the public stage (Htun and Weldon, 2010). In her social policy research in Chile, Staab (2017) shows how gradual yet transformative gender change has been possible despite the constraints of neoliberal principles and enduring social conservatism. Institutional change may be emergent, reflecting the compound effect of many small acts of institutional resistance as well as creative acts of ‘bricolage’ in which institutional materials ‘to hand’ are recombined by actors in new ways.

None of this is to deny the importance from a gender equality perspective of processes of intentional institutional design or reform. Although reform attempts rarely satisfy in full the intentions of their architects, they are a crucial part of the process whereby actors develop ‘an understanding of what constitutes a good society, without necessarily being able to achieve it, and how alternative institutions may be imagined to contribute to such a world’ (March and Olsen, 1989: 91). This is exactly the motivation behind what Mackay (2011) calls ‘feminist guided change’, for instance via the introduction of gender quotas or equality commissions. Of course, such projects may meet with resistance in the face of other reforming discourses. Joan Grace (2011: 111) has shown how in Canada, ‘the feminist framing of childcare as a women’s equality issue has been effectively supplanted by a focus on the family, and more specifically, on children’s issues and early childhood education’. But, attempts at reform do stimulate debates about the values that generally lie hidden beneath the surface of extant political institutions (March and Olsen, 1989: 91). Because institutions embed power and values, their dis-embedding (in processes of democratic transition, for instance) opens up important political spaces and creates opportunities to inscribe political institutions with new commitments to gender equality (Waylen, 2011).

Feminist institutionalists can play an important role in the wider movement to ‘bring actors back in’ to institutionalist theory. Feminist scholarship on performance and embodiment can help deepen our understanding of how gendered actors work with rules in reflexive ways. While we should not underestimate the very harsh sanctions faced by women contravening gendered rules in many contexts (see, for example, Krook’s (2017) research on political violence), it is also true that there are many instances where gendered actors work creatively with rules and, in so doing, disrupt and loosen their hold. Rules change through adaptation but also because they fall into disuse (when actors fail to actively maintain them), leading to institutional change through processes of drift and displacement (Mahoney and Thelen, 2010: 16). In her research on the Indian Parliament, Rai (2015: 1191–1192) shows how women parliamentarians started to congregate with men (and other women) in the Central Hall rather than using the allocated ‘Ladies Room’, which led to greater visibility and more opportunities for networking. Rai argues that the ‘non-use’ of this gendered space was ‘a powerful way of resisting’ established social conventions and, in so doing, developing new rules for everyday life in parliament.

In summary, research needs to focus on how rules constrain actors in gendered ways, but it also needs to investigate how actors interpret and remake such rules in their efforts to recalibrate gendered power settlements.

Conclusion: An Elaborated Model

Thirty years ago, March and Olsen (1989) argued that ‘the organisation of political life makes a difference’. This article has sought to explain the gendered character of that difference – how it is that institutions shape political opportunities, behaviours and outcomes in gendered ways. Understanding the way in which gendered institutions structure action helps to explain why equality reforms so often stutter or go into reverse and why the efforts of gender equality advocates are so frequently frustrated. At the same time, a deeper understanding of the role of institutions in perpetuating gender inequality (in the practice and outcomes of politics) can guide potentially transformative strategies to change the underpinning rules of the political game. Such strategies aim to set new parameters for action, going beyond the goal of getting more women into politics (important as this is) or adopting specific policies.

Realising this goal requires an appreciation of the micro-foundations of gendered institutions. Rather than treating institutions as organisational systems or broad policy regimes, we have focused on the encounters between actors and rules in specific political settings (action arenas), which are nevertheless connected to other institutions within and outside the political domain. Our simple model (Figure 1) identified the way in which political action is shaped both by ‘rules about gender’ and by seemingly neutral rules that have systematically gendered effects. Using a wide range of examples from comparative politics, we have operationalised our theoretical propositions by establishing threshold conditions for relevant rules (formal and informal) and specifying the different ways in which such rules constrain political action. Distinguishing institutional rules from both personal habits and broader social norms, the article has shown how rules operate to influence the gender power balance through regulation, obligation and persuasion. Understanding how political institutions are gendered requires an appreciation of the distinct but combinative effects of such rules upon behaviour. Separating them out conceptually and methodologically provides guidance for researchers seeking to understand why the ‘masculine affairs of government’ (Mazur, 2002: 10) prove so tenacious, even in the face of formal rule change. It helps researchers to know ‘where to look’ and also how to challenge those who criticise institutionalist explanations for being vague and non-falsifiable.

In conclusion, Figure 2 sets out an elaborated model of how political institutions are gendered, having interrogated the component parts of the simple model and how they relate to one another and the wider institutional context. At the centre of the model is the encounter between gendered actors and gendered rules, with arrows indicating the two-way (constitutive) relationships that link actors and rules. Rules constrain actors’ behaviour in gendered ways through mechanisms of regulation, obligation and persuasion; but actors are also capable of generating change through adapting, resisting or intentionally reforming institutional rules. Institutional actors encounter the rules of the political game through embodied gendered subjectivities, which may also be mediated by gendered identities ascribed to them by others. Such positionalities are constituted within a far wider institutional context, including domestic, economic and sociocultural domains. Political institutions are articulated with ‘non-political’ institutions, which tend to reinforce gendered power settlements. And, as Figure 2 shows, the gendered outcomes of political and policy processes feed into that very same gendered institutional context.

OPEN IN VIEWER

Figure 2.

The Elaborated Model.

The model could be criticised for putting ‘gendered’ in front of each variable, and thus muddling dependent and independent variables in constructing an explanation. We argue, however, that gender does indeed operate as both a dependent and an independent variable in politics – hence the intractability of gender injustice in the process and outcomes of politics. The purpose of the model is to break down the totalising proposition that political institutions are gendered and put in its place a specification of the relevant factors and relationships that underpin this overdetermined and iterative phenomenon. In so doing, the model offers comparativists a set of concepts and propositions that are potentially applicable in a wide range of political arenas, in more than one country, and across different time periods. The goal is thus to facilitate the exploration of similarities and differences in how institutions are implicated in the relationship between gender and politics.

The model can also inform strategies for institutional change in favour of gender equality. The model identifies a number of distinct points for intervention and shows how intervention at one point is likely to influence (and be influenced by) other factors. It is not enough to focus only on ‘rules about gender’, whether abolishing those that discriminate against women or putting in place new rules to promote equality. It is also vital to identify and call out seemingly neutral institutional rules that have gendered effects because of their interaction with wider institutional configurations, which differentially shape roles and benefits associated with men and women. Being aware that rules shape behaviour through obligation and persuasion as well as regulation can lead to a richer appreciation of what is required to deinstitutionalise masculine privilege and, conversely, to institutionalise gender equality. Reform strategies may prioritise one type of rule, depending on strategic opportunities, but should be undertaken in the knowledge that the most robust institutions shape behaviour through the combined impact of regulation, obligation and persuasion. Equalities legislation, gender quotas or women’s policy agencies (for instance) will have limited impact upon political behaviour and outcomes if they only put in place new structures and regulations. Similarly, the effects of abolishing discriminatory or anti-women regulations is inevitably undermined if sexist norms and discourses remain unchanged. Designing new regulations has to be accompanied by work to build and embed shared normative commitments and new discursive frames. At the same time, changing ‘rules about gender’ in the political domain is not enough. There is also a need for strategies to subvert the connections between political and ‘non-political’ institutions that lead to apparently neutral institutional rules having systematically gendered effects. Forging new connections that bolster gender equality in politics is equally significant.

Focusing on micro-foundations, we have built a model that identifies concepts and methods for research on how political institutions are gendered, within one political system and across many. Moreover, our model identifies points of intervention, and associated challenges, for those who seek to build more gender-just political institutions. Building a synoptic framework necessarily sacrifices nuance. It is for specialists to add this nuance in respect of their own particular research areas, and to draw upon that knowledge in dialoguing with the basic premises of the model. Moreover, the limitations of this article in relation to intersectionality highlight the need for future research on how gender intersects with multiple social categories and identities within different institutional settings.