Gender, devolution and political representation:

Evidence from the UK and Spain

Meryl Kenny † and Tània Verge

‡

(First draft. Do not cite or quote without authors’ permission)

Abstract

This paper seeks to assess the impact of state architectures on strategies to increase

women’s representation. To do so, we compare candidate selection reform over time in

two decentralized European countries: the UK and Spain. The article has three main

goals. First, we aim to disentangle how the relationship between territorial reform and

women’s representation is mediated by party organization, particularly by the multi-

level party dynamics which arise from the structure of the state. Second, we analyze the

ways in which party feminists, in pushing for candidate selection reform, have

transferred their activism across party and institutional levels. Finally, we seek to assess

under what circumstances can positive gender outcomes – in this case, numerical

increases in women’s political representation – be achieved

Keywords

Decentralization, political parties, candidate selection, women’s representation, gender

quotas

Authors’ information:

† Meryl Kenny (meryl.kenny@unsw.edu.au). She is postdoctoral researcher at the

School of Social Sciences and International Studies, University of New South Wales

(Sidney, Australia). Her research interests bridge the intersection of gender politics,

party politics, territorial politics, and institutional approaches to the study of politics.

‡ Tània Verge (tania.verge@upf.edu). She is visiting professor at the Department of

Political and Social Sciences, Universitat Pompeu Fabra (Barcelona). Her research

interests focus on political representation, political parties, gender and politics, and

multi-level systems.

Paper presented at the X Congreso de la Asociación Española de Ciencia Política,

Murcia, 7-9 September 2011, Section 2.4 ‘Género, instituciones y políticas’.

1

Introduction1

Political decentralization has been one of the most notable worldwide trends in recent

decades. In Western Europe, for example, the dichotomy between federal and unitary

states has been increasingly eroded, as decentralized state architectures in the form of

federations – such as Germany, Austria and Switzerland – coexist with ongoing

processes of decentralization in a number of countries, older and newer democracies

alike – including France, Italy, Portugal, the United Kingdom and Spain (Keating,

2009). Indeed, as Hopkin argues (2009: 179), ‘the distribution of important policy

competences to the regional level has become the norm in established democracies’. As

questions of constitutional and institutional restructuring have become ever more

politically relevant, these developments have led to various innovations in political

participation (Krook 2004). Globally, the most common reforms have been provisions

for the increased representation of women, including constitutional, legal, and party

quotas. Gender quotas have been debated or adopted in more than one hundred

countries; most within the last two decades (see Krook 2009).

In this article, we explore the hitherto under-researched relationship between

political decentralization and women’s representation. Specifically, we examine the

impact of decentralizing reforms on the organizational dynamics of political parties and

how this relationship, in turn, affects the adoption and implementation of gender quotas

to increase women’s descriptive, or numerical, representation. In evaluating this

relationship, we adopt a ‘party-centered’ approach (Hopkin and Bradbury 2006),

arguing that political parties are a key mediating variable between the formal

institutions of the state and gender equality outcomes. Although political

decentralization strongly affects party systems, party organization and electoral

strategies at different tiers of government (Hough and Jeffery 2006), most federalism

scholars pay only superficial attention to political parties (see Fillippov et al 2004).

Those scholars that do address the role of parties often assume that processes of

federalization and decentralization are exogenous to party system dynamics (Chhibber

and Kollman 2004), failing to acknowledge that parties themselves can initiate

1 Acknowledgements: The authors are grateful to the participants workshop on “Feminism and State

Architectures: Devolution, Federalism, Regionalism and (Gender) Equality”, ECPR Joint Sessions of

Workshops, Saint Gallen, 12-17 April 2011, for their richful comments to a previous version of the paper.

The authors are also thankful to Xavier Coller and Helder Ferreira do Vale for sharing their database on

Spanish regional parliaments.

2

decentralizing reforms, or that parties’ internal dynamics can ‘absorb and even overturn

the pressures arising from institutional change’ (Hopkin and Bradbury 2006: 136).

Crucially, we conceive of these dynamics as gendered (see also Kenny and Mackay

2011). Not only have political parties traditionally been male-dominated – that is, men

have typically monopolized parties’ decision-making and candidate selection bodies –

but, political parties have historically been built around ‘unacknowledged traditional

conceptions of gender relations’ which shape both formal party rules and informal

norms and practices (Lovenduski 2005: 58). Thus, by highlighting the importance of

(gender) power relations within parties and focusing on political parties as actors in

their own right, we can achieve a more comprehensive understanding of the complex

relationship between decentralizating reforms, party organizational dynamics, and

gender equality outcomes.– in this case, numerical increases in women’s representation.

We evaluate this relationship in the context of a qualitative comparative case

study of quota adoption and implementation in Britain and Spain, two European

parliamentary democracies where the electoral and legislative arenas are party-

determined and which have relatively recently undergone significant devolution of

power to sub-state institutions. Indeed, both the British and Spanish cases have been

identified as ‘federations in the making’ or ‘quasi-federations’ (Bogdanor 2003; Moreno

2001). In keeping with our party-centered approach, the empirical analysis focuses on

two state-wide political parties: the British Labour Party and the Spanish Socialist

Workers’ Party/Partido Socialista Obrero Español (PSOE). Both are well established

parties which have a strong presence across their state territories and which have had to

respond externally and internally to major decentralizing reforms (Hopkin 2009). Both

present candidates for state-wide and regional elections and have tight organizational

linkages between party levels (Fabre 2011). We exclude regional parties from our

analysis, as they do not face the challenge of maintaining a coherent organization across

a multi-level polity. Finally, both parties have taken leading roles in promoting

women’s representation in their respective countries. These similarities provide a useful

foundation for examining the impact of decentralizing reforms on party organizational

structure and for evaluating the effects of that relationship on the adoption and

implementation of gender quotas.

The remainder of the paper proceeds as follows. The second section discusses

the shortcomings of the three seats of literature we engage with, namely federalism,

party politics and gender politics, and builds on an integrated framework from which

3

our theoretical expectations are drawn. The third section briefly presents Spanish and

British decentralization processes and locates debates on women’s representation at the

outset of territorial reform. The fourth section examines the interaction of

decentralization with party (gendered) legacies in candidate reform processes. The last

section presents the main findings and concludes.

Decentralization, party politics and women’s descriptive representation

Our starting point is that we can only understand the relationship between

decentralizing reforms and gender equality outcomes – in this case, numerical increases

in women’s representation – by investigating the inner lives of political parties. While

there is a growing body of research on the impact of state architecture – that is, vertical

and horizontal power allocations – on women’s politics (see for example Chappell and

Vickers 2011; Haussman et al 2010; Vickers 2010), the role of political parties as actors

is downplayed in much of this work. In analyzing whether particular forms of state

architecture serve as a barrier to or an opportunity for women’s equality seeking, most

scholars focus on the impact of formal institutional structures on policy outcomes (see

for example Chappell 2002; Grace 2011). Indeed, even those scholars who do directly

address women’s descriptive representation continue to ignore or underplay party

political factors while over-privileging systemic variables such as electoral systems or

the role of women’s movement organization (Vickers 2011).

In contrast, research on women’s representation has increasingly focused on the

internal dynamics of political parties as a crucial area in need of further investigation.

While systemic variables structure the overall context of recruitment, political parties

are the main ‘gatekeepers’ to political office in most countries, particularly in strong

parliamentary democracies. Research in the field provides considerable evidence that

aspiring women candidates face significant obstacles in the recruitment process, for

example, highlighting widespread incidences of direct and indirect discrimination by

candidate selectors (see for example Norris and Lovenduski 1995). These gendered

interactions take place within a wider framework of formal and informal party rules and

practices which are shaped and structured by masculine gender norms (Lovenduski

2005). To a limited extent, these scholars have addressed the relationship between state

architecture, party organizational dynamics and women’s descriptive representation.

4

While some argue that decentralized candidate selection processes are more likely to

result in numerical increases in women’s representation, on the basis that localized

concerns engage more women (Lovenduski and Norris 1993) or that a selection process

taking place closer to ordinary party members might more easily attract previously

excluded groups (Norris 1997), research on gender quotas demonstrates that these

measures are most successfully implemented in highly centralized candidate selection

processes where there are effective sanctions for non-compliance (Caul 2001; Cowell-

Meyers 2011). However, much of the work on gender quotas continues to focus on why

quotas are adopted rather than how they are implemented, thus, failing to thoroughly

examine the intra-party dynamic (see Davidson-Schmich 2006). Moreover, the majority

of research in this area focuses largely on national elections, disregarding the multi-

level dynamics of the political setting in which parties operate.

Meanwhile, the literature on comparative federalism similarly fails to

comprehensively engage with the ‘inner lives’ of political parties. Work in the field

generally assumes that political parties in multi-level polities – particularly in federal

states – will ‘mimic’ the organization of the state, adapting to the distribution of power

and competences across levels of government (see for example Riker 1975; Fillipov et

al 2004). For example, some scholars argue that decentralizing reforms can lead to the

redistribution of power from central party organs to regional party branches, resulting in

a decentralized candidate selection process (Chhibber and Kollman 2004). However, a

growing body of research has challenged many of these assumptions, examining in

more depth the ways in which decentralizing reforms shape party organizational

dynamics, including distributional conflicts over resources, definitions of platforms, and

electoral and governing strategies (Deschouwer 2006; Hopkin 2003; Swenden and

Maddens 2009). Actually, the relationship between state architecture and party

organization is not necessarily straightforward. Party characteristics cannot simply be

read off from the territorial configuration of the state, nor are decentralizing reforms

necessarily accompanied by party organizational change (Hopkin 2009: 183). Yet, in

analyzing the relationship between decentralizing reforms and political parties, these

scholars have generally focused on ethno-regionalist conflicts, largely ignoring issues of

women and gender, as well as other marginalized groups. This omission is particularly

surprising given global developments such as the rise of candidate gender quotas and

the key role of women activists as agents in recent processes of institutional and

constitutional restructuring.

5

This article takes the first step towards integrating these different bodies of

literature in order to gain a more comprehensive understanding of the complex

relationship between decentralizing reforms, party organizational dynamics, and gender

equality outcomes. In doing so, we adopt an approach that is both party-centered and

gendered. Following Fabre and Méndez (2009: 103), we apply Elazar’s (1987)

distinction of the distribution of powers in federal systems to an analysis of the vertical

integration of state-wide parties. Vertical integration consists of the links established

between the central and regional levels of a political party (Smiley 1987; Thorlakson

2009). The ‘self-rule’ dimension addresses the degree of organizational autonomy

allocated to sub-national party branches in the management of regional party affairs,

including autonomy on candidate selection processes and party policy for regional

elections. ‘Shared-rule’ addresses the degree of regional party influence in central party

boards, measured through the participation of regional branches in central party organs

and decision-making processes.

In this article we argue that the dimensions of ‘self-rule’ and ‘shared-rule’

provide a useful framework through which to evaluate the success or failure of party

gender quotas. We posit that the stronger shared-rule and the weaker self-rule are, the

more successful quota adoption and implementation will be. On the one hand, for

voluntary party quotas to be effectively implemented at the various elections the party

competes in, all party levels should be committed to the measure adopted in order to

enforce its application. This relates to the extent to which regional branches have

participated in the particular decision-making process involved in the adoption of party

quotas (‘shared-rule’). If the national leadership board contains several regional

representatives and/or party conferences include a significant number of regional

representatives, then, as Threlfall puts it (2007: 1085), ‘policy from above is also policy

from below’. Conversely, if regional representatives are absent or excluded, quota

reforms might raise suspicions of illegitimate party centralism.

On the other hand, the implementation of gender quotas is more effective the

more centralized the candidate selection process is, which means that ‘self-rule’ is in

this aspect low or at least moderate. First, a strong and centralized party organization

means that the national party level has more control over its local and regional branches

and is, therefore, better able to implement and enforce quota reforms. Second, the party

central leadership is more likely to be well educated and liberal in their attitudes

towards women (Randall 1987). Third, centralized selection processes neutralize local

6

power monopolies and avoid self-nomination (Baldez 2004: 238), which tends to

disadvantage women because they are less likely to promote their own candidacies (see

Fox and Lawless 2004: 275). And, finally, the more centralized and institutionalized the

candidate selection process is, the easier it is for women and other outsider candidates to

determine what needs to be done to run for office (Davidson-Schmich 2006).

In addition, we posit that the success of quota adoption and implementation is

mediated by the (gendered) power relations and organizational inertias of the parties

themselves. Political parties are gendered organizations – in other words, formal party

rules and informal norms and practices, as well as historical legacies and ideological

traditions are shaped and structured by gender power relations (see Lovenduski 2005;

Kenny and Mackay 2011). A focus on (gender) power relations allows us to gain a

fuller picture of the tensions raised by decentralizing reforms. While parties may change

their organizational structure in response to decentralizing reforms, these changes are

filtered by established party procedures, structures and traditions (Hopkin 2009: 182). In

other words, developments within political parties may be ‘more path-determined by

past commitments, institutional legacies and practices’ than by self-conscious

programmes of change (Mackay 2004: 113).

Decentralization in the UK and Spain and women’s political representation

Whereas Spain’s democratization and political decentralization were concurrent

processes in the late 1970s and early 1980s, in the UK the devolution campaign gained

momentum in the 1990s in a context of a long established democracy2. A common

concern in these devolution processes was to address the will of self-government

expressed by national minorities, namely the Scotts and the Welsh in the British case3

and the Catalan and Basque in the Spanish case (Moreno 2001, Greer 2007).

In Spain, the constitutional model established two different procedures for

devolution justified on historical and political grounds, which differed both in speed of

competence transference and institutional development, though it permitted any region

as well, the so-called autonomous communities (comunidades autónomas), after a

2 This followed several unsuccessful attempts to introduce devolution in Scotland and Wales in the 1970s,

a discussion of which is beyond the scope of this paper. 3 Northern Ireland is not examined as it is marked by a unique political setting and has a territory-specific

party system in which no ‘mainland’ state-wide parties seek election (see Hough and Jeffery 2006).

7

period of five years, a broad leeway to choose, through its regional statute of autonomy,

the degree of devolution desired within the limits established by the Constitution

(Colino 2010). This incremental adaptation gradually transformed Spain from an

asymmetric decentralized system into a largely symmetric quasi-federation (Aja 2003).

The seventeen regions have their own elected legislative and executive institutions.

The UK system of devolution is asymmetric, in that the devolved regions have

different sets of competences, institutional forms, and ways of working. For example,

while the Scottish Parliament can pass primary as well as secondary legislation, the

National Assembly for Wales has, until the recent March 2011 referendum, only been

able to pass secondary legislation in devolved areas. Other matters remain reserved to

the Westminster Parliament, which does not normally legislate in devolved areas

without consent.

Whereas in Spain, in a mimetic reproduction of central institutions, regions

adopted the same electoral system in place for the national lower house, namely PR

D’Hondt method with candidatures highly controlled by party leaderships (through

closed electoral tickets), in the UK devolved institutions also aimed at installing a new

political culture, which facilitated the incorporation of a gender perspective on political

representation. Nonetheless, in both countries under study debates over women’s

representation have largely centered on candidate quotas for women in political parties.

Additionally, the adoption of quotas has been driven by the main left-wing parties – the

British Labour Party and the Spanish Socialist Workers’ Party (Partido Socialista

Obrero Español, PSOE). This is why we will focus largely on them, although we also

address the wider party systems and political contexts.

In the Spanish case, by the late 1970s, women’s representation was an issue of

very low priority in the political agenda. In the three first national legislative elections

(1977, 1979, 1982) women’s representation oscillated between 5% and 6% (around 20

out of 350 seats). On average, the first regional elections held in 1983 yielded the same

results – an average of 5.6% (Verge 2011). Democratization and decentralization were

not accompanied by any measure to foster women’s representation nor was there a

public debate on the issue as these proportions were in tune with European standards

(excluding Scandinavian countries)4. Women-friendly measures would be dependent on

political parties’ will, encouraging feminists to establish alliances with those of the Left.

4 For example, in 1980, women’s representation was 5.3% in France, 7.9% in Italy, 7.5% in Belgium and

3.1% in the UK (see Mossuz-Lavau, 1988).

8

Over the past three decades, increases in women’s representation in the Congress of

Deputies have been paralleled by those found in regional parliaments. In 2003-04,

before the statutory quota was introduced, women’s representation reached 36% at both

levels. After the quota law was enforced, female deputies mounted to 43% at the

regional level while no increase was observed at the national level.

Conversely, in the UK processes of decentralization and campaigns for home

rule for the historic nations of Scotland and Wales and the contested territory of

Northern Ireland opened up new possibilities for regional feminist actors to push for

gains in women’s descriptive representation (Kenny and Mackay 2011). Whereas the

devolution debate in Wales was largely top down, in Scotland, women activists working

inside and outside political parties – in particular, the Scottish Labour Party, as well as

civic reform groups such as the Scottish Constitutional Convention and its Women’s

Issues Group – successfully gendered these reform debates (Russell et al 2002). In

response, all Scottish parties declared concern with the under-representation of women

in politics. In Wales, in contrast, opportunities for gender equity entrepreneurs to

campaign for women’s political representation were constrained by the ‘general

ambivalence about the devolution project’ (Mackay 2003: 84). The “story” in Wales,

then, is not one of grassroots mobilization, but rather one of a strategic coalition – a

group of powerful women in influential positions, including academics, trade union,

party officials, and party activities - who pushed for positive action to promote women

candidates (Chaney 2008: 273). In the results of the first elections to the devolved

institutions, women comprised 37.2% of the new Scottish Parliament and 40% of the

new National Assembly for Wales (rising shortly afterwards to 42%). Compared to

state-wide institutions, these are very good outcomes for women’s representation as

female deputies are currently stagnated at less than 20%.

Positive action in candidate selection in the Labour Party and the PSOE came

only after women had obtained institutionalized spaces within the organization5. Party

feminists then used these as a springboard to achieve wider influence in party-decision

making, pushing for further and farther-reaching reforms in candidate selection

(Astelarra 1990; Lovenduski 2005). As a result of sustained strategic campaigning

within the Labour Party and the PSOE, organized women’s activists were able to put

5 In the PSOE, party feminists organized into a women’s caucus until in 1985 it was transformed into an

executive secretariat, acquiring the highest status in the federal party board.

9

women’s representation on the agenda in the early 1980s, linking arguments for gender-

balanced representation to wider programmes of party modernization as well as to

bridging in the gender gap in the parties’ electoral support (Verge 2006: 128-9; 2007:

170; Kenny and Mackay 2011). In both cases, women were also able to take advantage

of international debates about quotas, especially those taking place within the Socialist

International Women, the sister organization of the Socialist International (SI) Labour

and PSOE are adhered to, and by the adoption of quotas by other parties of the SI

(Eagle and Lovenduski 1998; Threlfall 2007; Valiente 2005).

Decentralization, party vertical integration and gendered legacies

Quota adoption

Whereas political decentralization has clearly shaped Spanish politywide parties’

electoral and governing strategies and they all organize according to the structure of the

state, their internal decentralization has remained to a great extent isolated from the

broad process of institutional change. Indeed, the degree of political decentralization of

the state is greater than intra-party decentralization (Montero 2005: 68). As devolution

began after forty years of dictatorship rule, Spanish parties were primarily concerned

with establishing strong, united, and cohesive organizations, and the centralization of

power was conceived of serving these aims.

In the UK, politywide parties have responded to the challenges posed by

devolution, adapting their organizations to the decentralized context of government and

electoral competition. Given its asymmetrical nature, devolution had a stronger impact

on party politics in Scotland and Wales, where the territorial salience is more relevant,

particularly on those parties with electoral strongholds in these territories, namely the

Liberal Democrats and especially Labour. Yet, there is no ‘perfect correlation’ (Riker

1975: 137) between changes in state architecture and party organizational change:

politywide parties have avoided large-scale internal reform, leaving many long-standing

structures and practices intact (Hopkin and Bradbury 2006: 150).

When comparing current degrees of vertical integration of British and Spanish

politywide parties significant differences emerge. For example, the PSOE presents

significantly higher levels of ‘shared-rule’ than the British Labour Party, incorporating

regional branches and leaders in central decision-making to a much greater extent

10

(Fabre and Méndez-Lago 2009). While the British Labour Party increasingly recognizes

the territorial autonomy of the Scottish and Welsh parties, no seats in the National

Executive Committee (NEC) are reserved for regional representatives, and the NEC

retains its central role in the British Labour Party’s organizational structure. While

Scottish and Welsh members can attend the national Labour Party conference, they are

not formally recognized and constitute a small proportion of total delegates (Hopkin

2009: 188; Laffin and Shaw 2007). In the PSOE, the territorial inclusiveness is a

criterion always applied in the composition of party organs in the quasi-consensual

negotiations which take place during party conferences, including the main decision-

making body, the Federal Executive Commission (Comisión Ejecutiva Federal).

After the women’s section displayed several lobby activities, in the party

conference held in 1988 the PSOE passed a 25% quota in party positions and electoral

candidatures. Although the floor discussion was much contested the whole party

structure assumed the measured once it was approved. The fact that regional

conferences are celebrated in cascade after the national one has been held implies that

the principles adopted at the federal level are subsequently translated at the lower

echelons of the party organization. PSOE’s regional party branches have their own

bureaucracy and organs of self-rule. However, they replicate the central party structure

since central party statutes lay out how party branches must be organized and reforms of

regional statutes must be approved by the central organs. So, the 25% quota was

incorporated too in regional party constitutions.

In the approval of this first quota party feminists also managed to transfer

activism across party levels. Gender quotas were actually pioneered by a regional party

federated with the PSOE in the region of Catalonia, that of the Party of the Catalan

Socialists (PSC)6. In 1982 the PSC guaranteed women 12% of the places on party

committees and on candidate lists and in 1987 proportions were raised to 15%, a

number equivalent to the party’s female membership at the time, although Catalan party

feminists already asked for 25%. Once the sub-national sister party had established the

quota the vertical (upward) contagion took place. Simultaneously, the endorsement of

the 25% quota by the PSOE eased its introduction in the PSC in 1990, observing then a

new vertical (downward) contagion. In both cases support of party leaders was a key

6 The PSOE has sub-national units in each region, with the exception of Catalonia. The PSC operates as a

sister party rather than as a territorial unit with a federal-type agreement. It is completely autonomous to

select its candidates and the reform of its statutes does not need PSOE’s approval.

11

factor to surmount opposition from medium-level cadres. The quota provisions were

gradually enlarged until reaching a gender neutral formulation of parity between sexes

(neither less than 40% nor more than 60% of representation for any sex) in 1997 in the

PSOE (again in all party levels) and in 2000 in the PSC.

In the Spanish case, although constitutional (territorial) reform did not create at

the onset a window of opportunity for women’s demands in the field of political

representation, its effect, that is the creation of a multi-tiered polity, has provided new

opportunities for party feminists to ‘play the two-level game’. Particularly, PSOE’s tight

vertical integration across party levels (i.e. the fact that central party decision-making

prevails) has produced a contagion effect from the party to the institutional level in the

following manner. Efforts for the introduction of a statutory quota were initiated in the

national lower house in 1996. All bills presented by left-wing parties, among them the

PSOE, to make parity mandatory were rejected due to the opposition of the conservative

Popular Party (PP), which held the majority of seats (Verge 2011). The blockade to

gender quotas at the national assembly was circumvented at the regional level. Between

2002 and 2005 three regions led by the PSOE (Castile-La Mancha, Balearic Islands and

Andalusia) incorporated zipping into their regional electoral laws (men and women

alternate from top to bottom of the candidates’ tickets). Although the politywide PP-led

government lodged an appeal to the Constitutional Court – effectively suspending the

implementation of these reforms, the PSOE voluntarily adopted zipping in these regions

in the subsequent regional elections, which resulted in the highest rates ever of women

elected by a party (about 50%). In 2004 the PSOE won the general elections and

announced the preparation of a bill to promote gender equality in different policy areas,

including political office. The so-called Equality Law, passed in March 2007, forces

parties to incorporate a minimum of 40% and a maximum of 60% of any sex into

candidates’ lists for all elections. The Law allows for more favorable measures

introduced by regional reforms, supporting the legality of zipping in the aforementioned

regions. In 2007 women’s presence at the regional level mounted to 42.8%.

In the UK the first major programme of reform for women was a package of

internal party quotas, crafted by a broader coalition of women in trade unions and in

senior offices in the Labour Party. The campaign for gender quotas in parliamentary

selections met with considerable resistance as members generally opposed ‘any action

from the centre’ to enforce positive action mechanisms in parliamentary selection (Short

1996: 20). Therefore, all-women shortlists (AWS) was seen to be one of the only

12

measures that could ensure that Constituency Labour Parties (CLPs) selected women

candidates, requiring at least some of them to chose from ‘women-only shortlists’

(Russell 2005: 109). The National Executive Committee (NEC) did adopt the policy of

‘one woman on a shortlist’ in 1988 and actively encouraged CLPS to select women

candidates, but progress on women’s representation continued to move slowly. In this

setting, in 1993 a compromise was drawn up by the same group of trade union women

who had campaigned for internal party quotas and who now sat on the newly reformed

NEC women’s committee: AWS would be applied in half of all winnable seats, and half

of all seats where MPs were retiring. However, legal challenges stalled the AWS agenda

at UK level (Eagle and Lovenduski 1998).

Parallel to this, campaigns for devolution opened up new possibilities for

regional feminist actors to push for gains in women’s representation. In the case of the

British Labour Party, because of its highly centralized structure, once the centre was

persuaded, the party was able to enforce quota reforms in both Scotland and a

recalcitrant Wales to ensure gender balance in candidate selection (Russell et al 2002).

Both the Scottish and the Welsh party branches adopted the ‘twinning’ mechanism

devised by Scottish Labour women activists and academics which required all

constituencies to be paired on the basis of geographical proximity and winnability, with

each pair having to select one male and one female candidate.

Quota implementation

The aspect of self-rule which most clearly affects women’s representation is candidate

selection, that is, the degree of autonomy party regional branches have to autonomously

decide who they field in electoral tickets. PSOE’s regional branches have a consultative

role in national elections as it is sub-regional (provincial-constituency) parties which

draft candidate tickets and submit them to a national electoral commission which holds

veto power and may rearrange the order of the candidates. The composition of this

electoral commission takes territorial balance into account. This organ subsequently

presents the lists to the central leadership for final approval. The regional branches of

the party select candidates to regional parliaments, though the central party organs must

give its approval too. Thus, self-rule when it comes to candidate selection is rather

limited. This helps explain why the quota was successfully implemented even in a

context of electoral decline, risking a backlash among male incumbents (Threlfall 2007.

1079).

13

After the approval of the parity quota, the PSOE elected 35% of women in the

1999 regional elections (approximately a 52% net increase) as opposed to PP which

elected 28%. Further measures were adopted to ensure that parity reaches the winning

positions too. Until the 1999 regional elections, party regional branches complied with

the overall quota provisions but many women candidates occupied positions in the party

list which did not stand the chance to get elected. Under the Spanish proportional

electoral system, seats are allocated through the d’Hondt method to closed party lists.

Candidates need to be ranked sufficiently high up to win a seat in their constituency. So,

for a while, gendered practices in candidate selection impaired quotas to produce more

substantive gains on women’s representation. Since 1999 a representative of the

Women’s Section sits in the national electoral commission and holds veto power: lists

failing to comply with the quota provisions are amended (Verge, 2007: 172).

Although the PP strongly rejects quotas, tight vertical integration has also

allowed the party to feminize its institutional representation once a vague

recommendation for gender balance was issued by the central leadership at the end of

the 1990s. Following the adoption of PSOE’s parity quota, the PP countered by making

a rhetorical commitment to gender equality in party positions and institutional

representation and experienced a net rise in women’s representation in the majority of

regional assemblies as well as in the national lower house (Verge 2011). The

candidacies regional branches put forward for both national and regional elections also

need to be ratified by the party’s central organs. However, PP’s commitment with

equality in representation is less strong than in the PSOE. Even after 2007 once the

statutory quota was introduced, which the PP also appealed before the Constitutional

Court cross-party differences still exist: whereas the PSOE averages 46% of women

deputies at the regional level, the PP averages 39%, having failed to reach 40% of

women’s representation in seven regions, as Table 1 shows. In the lower house, in 2008

the PSOE elected 43% of women and the PP 32%, as illustrated in Table 2. As Threlfall

(2007: 1088) argues, ‘the Spanish case shows laws that are not fully backed by party

leaders [PP] can be less effective than a party rule that commands a consensus among

party leaders [PSOE]’.

14

Table 1. Women in regional parliaments Spain, 1983-2007 (%)

1983 1987 1991 1995 1999 2003 2007

PSOE 6 10 16 22 35 42 46

PP 5 4 11 15 26 35 39

Total 6 7 14 20 30 36 43

Source: Own elaboration based on Coller et al. 2007.

Table 2. Women in the national parliament Spain, 1979-2008 (%)

Election year PSOE PP Total

1979 5 11 5

1982 7 1 6

1986 7 6 8

1989 17 10 13

1993 18 15 16

1996 28 14 22

2000 37 25 28

2004 46 28 36

2008 43 32 37

Source: Based on Verge (2006) updated by the authors.

Candidate selection in the British Labour Party presents a more complicated

picture. Despite its highly centralized structure, the British Labour Party has a long

tradition of decentralized constituency-based selection (Denver 1988), due in part to the

single-member, first-past-the-post electoral system. In the late 1980s and 1990s,

however, the process became increasingly centralized as part of a wider reform

trajectory of party modernization. In the run-up to the 1999 elections for the devolved

institutions, questions were raised as to the devolution of decision-making within the

party. However, for the party leadership, the priority was not on devolving selection

procedures to the Scottish and Welsh parties, but rather on attracting a wider pool of

applications to ensure that seats in the new devolved institutions would not simply go

the ‘usual suspects’ (Russell 2005). The party introduced a pre-selection approval

procedure, which established a central panel of approved constituency and list

candidates. The selection boards that drew up these approved lists included politywide,

Scottish and Welsh party members as well as non-party members (Bradbury et al.

2000). While constituency selection was left to one-member-one-vote, the electoral

boards overseeing list selection were made up of members from both the regional party

branch and the politywide party branch.

This highly centralized approach proved to be controversial, with some concerns

that these changes were an attempt to consolidate New Labour support within the

Scottish party. Party officials were accused of implementing an ‘ideological test’ of the

15

New Labour credentials of prospective candidates, rejecting several prominent party

figures on the nationalist-left wing of the party (Bradbury et al 2000). The twinning

scheme also generated controversy, with accusations that some candidates were

‘twinned out’ of the selection process through central party intervention.

The controversy over the 1999 selections created significant pressure for internal

party decision making to be devolved downward, leading to a distinct ‘change of tone’

as well as several ‘organizational concessions’ on the part of the British Labour Party

(Hopkin and Bradbury 2006: 142). In 2000, control over leadership selection was

devolved to the Scottish and Welsh parties while a year later control over candidate

selection was also devolved, although these decisions take place within a framework of

centrally prescribed principles (Laffin and Shaw 2007). This signals a return to the

party’s past tradition of decentralized constituency-based selection in which participants

have been left with considerable leeway to circumvent or subvert reforms and to return

to past practices of local patronage and privileging of ‘favorite sons’ (Kenny 2011). In

this vein, the Scottish Labour party has been extremely reluctant to implement strong

equality guarantees, with a detrimental impact on trends in the recruitment and election

of female candidates (Mackay and Kenny 2007).

While Labour remains the only party to implement strong equality guarantees in

both regions, there is some (limited) evidence of a “contagion effect” in 1999. In

response to Scottish Labour’s use of equality measures, the Scottish National Party

implemented unofficial measures to encourage women to stand for election; similarly,

in Wales, Plaid Cymru implemented a “gender template” on the regional lists (Russell

et al 2002). Post-1999, however, women’s descriptive representation has been pursued

far more vigorously in Wales than Scotland, with momentum being maintained by key

‘champions’ within the main parties (Mackay 2003). In contrast, the issue of women’s

descriptive representation has not retained high salience for Scottish political parties

since 1999, nor has it remained a matter of party competition (Mackay and Kenny

2007). In contrast, gender-balanced representation was highly politically salient in the

UK general election in 2010, although ultimately none of the British politywide parties

met their own targets for women’s representation, nor have they come close to matching

the levels attained in Scotland and Wales (Kenny and Mackay 2011).

As it can be seen in Table 3, despite sustained efforts by the Labour Party

central leadership, the percentage of female MPs in the party’s parliamentary group at

16

Westminster, 31% after the 2010 general election, still falls short of the adopted

benchmark for women’s representation.

Table 3. Women in the national parliament UK, 1979-2010 (%)

Election year Labour Cons LibDem Total

1979 4 2 0 3

1983 5 3 0 4

1987 9 5 6 6

1992 14 6 10 9

1997 24 8 7 18

2001 23 8 10 18

2005 28 9 16 20

2010 31 16 12 22

Source: Based on Shepherd-Robinson and Lovenduski (2002), updated

by the authors.

With regards to regional parliaments, various aspects can be highlighted. First,

the introduction of larger proportionality in the election of deputies has clearly

facilitated women’s representation. Second, to a large extent, women’s significant gains

in both the Welsh assembly and the Scottish parliament are attributable to progress

made by the Labour Party. Third, as our examination of candidate selection has shown,

increasing self-rule by Labour’s regional party branches has been disadvantageous for

women. Although women’s representation in Labour parliamentary benches are very

high and surpass the levels internationally defined as parity between the sexes, an

important decrease has been observed since the first regional elections, as indicated in

Table 4.

Table 4. Women in regional parliaments UK, 1999-2011 (%)

Region Election year Labour Cons LibDem Main NSWP Total

1999 50 17 12 43 37

2003 56 22 12 33 40

2007 50 29 13 26 33

Scotland

2011 46 40 20 28 35

1999 54 -- 50 35 40

2003 63 18 50 50 50

2007 62 8 50 47 47

Wales

2011 50 29 40 36 42

Source: Based on Burness (2011). Main NSWP in Scotland refers to the Scottish National Party

(SNP) and for Wales to Plaid Cymru (PC).

17

Concluding remarks

This paper has addressed the under-researched relationship between political

decentralization and women’s representation. In particular, we have examined the

impact of decentralizing reforms on the organizational dynamics of political parties and

how this relationship, in turn, affects the adoption and implementation of measures

which seek to increase women’s representation, namely gender quotas.

As Sawer and Vickers argue (2011: 9), “the timing of the creation of political

institutions is of major importance in determining the kind of gender norms that are

incorporated in the new political architecture”. In line with recent work on institutional

(re)design, we see some evidence that gender concerns are better integrated at the start

of processes of institutional and constitutional restructuring (Chappell 2011; Mackay

2009). Yet, at the same time, we have shown that Spain and the UK present a more

complicated picture.

In Spain, although gender concerns were not integrated at the ‘start’ of

devolution, substantial progress on women’s representation has been made over time.

The strong shared-rule combined with a limited self-rule has allowed PSOE’s gender

equality reforms to be successfully implemented and effectively enforced at both central

and regional party levels. Integrated party policy processes (high shared-rule) and

centralized candidate selection (low self-rule) have helped overcome the potential

fragmenting effect of multiple levels. Conversely, in the UK, the devolved institutions

are ‘new’ institutions but simultaneously they are ‘nested’ within a wider institutional

system, as well as within past historical legacies, which limits the potential of gender

equality reforms as tensions between central intervention and local control are ongoing.

In the British case, shared-rule is rather limited and regional branches self-rule has

significantly increased with devolution, thus reducing the capacity of the Labour Party

central leadership to effectively monitor and implement positive action in candidate

selection.

Overall, political parties need to be considered as independent actors in any

analysis of the relationship between gender and institutional reforms. Their internal

dynamics and organizational inertias filter the pressures arising from institutional

change. Crucially, gendered legacies shape parties’ history, so intra-party (gender)

power relations are relevant to understanding the tensions raised by territorial reforms in

relation to women’s political representation and candidate quota reform processes.

18

References

Aja, Eliseo (2003). El Estado Autonómico. Federalismo y hechos diferenciales. Madrid:

Alianza Editorial, 2nd ed., pp. 207-256.

Astelarra, Judith (1990). Participación política de las mujeres. Madrid: CIS.

Baldez, Lisa (2004). Elected Bodies: The Gender Quota Law for Legislative Candidates

in Mexico. Legislative Studies Quarterly 2:239-258.

Bogdanor, V. (2003). “Asymmetric Devolution: Towards a Quasi-Federal

Constitution?”, in Dunleavy, P et al. (eds) Developments in British Politics, 7,

(Basingstoke: Palgrave Macmillan).

Bradbury, Jonathan, James Mitchell, Lynn Bennie and David Denver (2000).

“Candidate Selection, Devolution and Modernization: The Selection of Labour

Party Candidates for the 1999 Scottish Parliament and Welsh Assembly

Elections”, British Elections and Parties Review, 10, 151-172.

Burness, Catriona (2011). Women and Parliaments in the UK. Edinburgh: Scottish

Parliament Chamber.

Caul, Miki (2001) “Political Parties and the Adoption of Candidate Gender Quotas: A

Cross-National Analysis”, Journal of Politics, 63 (November), 1214-29.

Chaney, Paul (2008). “Devolved Governance and the Substantive Representation of

Women: The Second Term of the National Assembly for Wales, 2003-2007”,

Parliamentary Affairs, 61 (2), 272-290.

Chappell, Louise (2011). “Nested Newness and Institutional Innovation: Expanding

Gender Justice in the International Criminal Court”, in Mona Lena Krook and

Fiona Mackay (eds) Gender, Politics and Institutions: Towards a Feminist

Institutionalism. Basingstoke: Palgrave: pp. 163-180.

Chappell, Louise (2002). Gendering government: Feminist engagement with the state in

Australia and Canada. Vancouver, B.C.: University of British Columbia Press.

Chappell, Louise and Jill Vickers (eds) (2011). “Critical Perspectives on Gender,

Politics and State Architecture”, Politics & Gender, 7 (2), 251-293.

Chhibber, P. and K. Kollman (2004). The Formation of @ational Party Systems:

Federalism and Party Competition in Canada, Great Britain, India, and the

United States. Princeton: Princeton University Press.

19

Coller, Xavier, Helder Ferreira do Vale, and Chris Meissner (2007). “Political Elites in

Federalized Countries: The Case of Spain (1980-2005).” Center for European

Studies Working Papers Series no. 146, Harvard University.

Colino, César (2010). “Constitutional Change Without Constitutional Reform: Spanish

Federalism and the Revision of Catalonia’s Statute of Autonomy”. Publius: The

Journal of Federalism 39(2): 262–88.

Cowell-Meyers, Kimberly (2011). “A Collarette on a Donkey: The Northern Ireland

Women’s Coalition and the Limitations of Contagion Theory”, Political Studies,

59, 411-431.

Davidson-Schmich, Louise K (2006). “Implementation of Political Party Gender

Quotas: Evidence from the German Lander 1990-2000”, Party Politics, 12: 211-

232.

Deschouwer, Kris (2006). “Political Parties as multi-level organizations”, in Richard

Katz and William Crotty (eds.) Handbook of Party Politics, pp. 291–300. London:

Sage.

Eagle, Maria and Joni Lovenduski (1998). High Time or High Tide for Labour Women?

London: Fabian Society.

Elazar, D.J. (1987). Exploring Federalism. Tuscaloosa: Univ. Alabama Press.

Fabre, Elodie (2011). “Measuring party organization: The vertical dimension of the

multi-level organization of state-wide parties in Spain and the UK”, Party Politics,

17 (3), 343-363.

Fabre, Elodie, and Mónica Méndez-Lago (2009). “Decentralization and Party

Organizational Change: the British and Spanish Statewide Parties Compared”, in

Wilfried Swenden and Bart Maddens (eds) Territorial Party Politics in Western

Europe, pp. 102–18. Basingstoke: Palgrave Macmillan.

Filippov, Mikhailm Peter C. Ordeshook, and Olga Shvetsova (2004). Designing

Federalism. A Theory of Self-Sustainable Federal Institutions. Cambridge:

Cambridge University Press.

Fox, Richard L. and Jennifer L. Lawless (2004). “Entering the Arena? Gender and the

Decision to Run for Office”, American Journal of Political Science 48(2): 264-80

Grace, Joan (2011). ”Gender and Institutions of Multi-Level Governance: Child Care

and Social Policy Debates in Canada”, in Mona Lena Krook and Fiona Mackay

(eds) Gender, Politics and Institutions: Towards a Feminist Institutionalism.

Basingstoke: Palgrave, pp. 95-111.

20

Greer, Scott L. (2007). @ationalism and Self-Government: The Politics of Autonomy in

Scotland and Catalonia. Albany: State University of New York Press.

Haussman, Melissa, Marian Sawer, and Jill Vickers (2010). Federalism, Feminism and

Multilevel Governance. London: Ashgate.

Hopkin, Jonathan (2003). “Political Decentralization, Electoral Change and Party

Organizational Adaptation: A Framework for analysis”, European Urban and

Regional Studies 10: 227–37.

Hopkin, Jonathan (2009). “Party Matters: Devolution and Party Politics in Britain and

Spain”. Party Politics 15 (2): 179–198.

Hopkin, Jonathan and Bradbury, Jonathan (2006). “British statewide parties and

multilevel politics”. Publius: The Journal of Federalism, 36 (1): 135–152.

Hough, Dan and Charles Jeffery (eds.) (2006) Devolution and Electoral Politics, pp.

37–53. Manchester: Manchester University Press.

Keating, Michael (2009). “European Devolution”, In The Oxford Handbook of British

Politics, edited by M. Flinders, A. Gamble, C. Hay & M. Kenny. Oxford: Oxford

University Press.

Kenny, Meryl (2011). “Gender and Institutions of Political Recruitment: Candidate

Selection in Post-Devolution Scotland” In Gender, Politics and Institutions:

Towards a Feminist Institutionalism, ed. Mona L. Krook and Fiona Mackay.

Basingstoke: Palgrave Macmillan, 21-41.

Kenny, Meryl and Fiona Mackay (2011). “Gender and Devolution in Spain and the

United Kingdom”, Politics & Gender 7: 280–6..

Krook, Mona L. (2004). “Gender Quotas as a Global Phenomenon: Actors and

Strategies in Quota Adoption”, European Political Science, 3(3): 59–65.

Krook, Mona L. (2009). Quotas for Women in Politics: Gender and Candidate

Selection Reform Worldwide. New York: Oxford University Press.

Laffin, M. and E. Shaw (2007). “British Devolution and the Labour Party: How a

National Party Adapts to Devolution”, British Journal of Politics and

International Relations 9: 55-72.

Lovenduski, Joni (2005). Feminizing Politics. Cambridge: Polity Press.

Lovenduski, Joni, and Pippa Norris (1993). Gender and Party Politics. Thousand Oaks,

California: Sage.

Mackay, Fiona (2003). “Women and the 2003 Elections: Keeping up the Momentum”,

Scottish Affairs 44: 74-90.

21

Mackay, Fiona (2004). “Gender and Political Representation in the UK: The State of

the ‘Discipline’.” British Journal of Politics and International Relations 6: 99–

120.

Mackay, Fiona (2009). “Institutionalising ‘New Politics’ in Post Devolution Scotland:

‘Nested Newness’ and the Gendered Limits of Change”. Paper presented at the

Political Science Association Annual Conference. Manchester, April 7-9 2009.

Mackay, Fiona and Meryl Kenny (2007). “Women’s Representation in the 2007

Scottish Parliament: Temporary Setback or Return to the Norm?”, Scottish Affairs

60: 25-38.

Montero, Alfred (2005). “‘The Politics of Decentralization in a Centralized Party

System: The Case of Democratic Spain”, Comparative Politics 38: 63–82.

Moreno, Luis (2001). The Federalization of Spain. London: Frank Cass.

Mossuz-Lavau, Janine (1988). Women of Europe: mirroring the course of women's

rights in Europe (1977-1987). Brussels: Commission of the European

Communities.

Norris, Pippa, and Joni Lovenduski (1995). Political Recruitment: Gender, Race and

Class in the British Parliament. Cambridge: Cambridge University Press.

Randall, Vicky (1987). Women and Politics: An International Perspective. Basingstoke:

Macmillan.

Riker, William H. (1975). “Federalism”, in F.I. Greenstein and N.W. Polsby, eds.,

Handbook of Political Science, Volume 5: Governmental Institutions and

Processes. Reading: Addison-Wesley, pp. 93-173.

Russell, Meg (2005). Building @ew Labour: The Politics of Party Organization. New

York: Palgrave Macmillan.

Russell, M., Mackay, F. and L. McAllister (2002). “Women’s Representation in the

Scottish Parliament and National Assembly for Wales: Party Dynamics for

Achieving Critical Mass”, Journal of Legislative Studies, 8 (2), 49-76.

Sawer, Marion, and Jill Vickers (2010). “Introduction: Political Architecture and its

Gender Impact”, in Federalism, Feminism and Multilevel Governance, ed.

Melissa Haussman, Marian Sawer, and Jill Vickers. Ashgate, pp. 3–18.

Smiley, David V. (1987). The Federal Condition in Canada. Toronto: Mc-Graw-Hill

Ryerson.

Swenden, Wilfried, and Bart Maddens (eds) (2009). Territorial Party Politics in

Western Europe. Palgrave: Houndmills, Basingstoke.

22

Thorlakson, Lori (2009). “Patterns of Party Integration, Influence and Autonomy in

Seven Federations”, Party Politics 15 (2): 157–77.

Threlfall, Monica (2007). “Explaining Gender Parity Representation in Spain: The

Internal Dynamics of Parties.” West European Politics 30 (5): 1068–95.

Valiente, Celia. 2005. “The women’s movement, gender equality agencies and central-

state debates on political representation in Spain”. In State feminism and political

representation, ed. Joni Lovenduski. Cambridge: Cambridge University Press,

pp. 174–94.

Verge, Tània (2006). “De la cuota a la democracia paritaria: Estrategias partidistas y

representación política de las mujeres en España”, Política. Revista de Ciencia

Política 46: 107–39.

Verge, Tània (2007). Partidos y representación política: Las dimensiones del cambio

en los partidos políticos españoles, 1976-2006. Madrid: CIS.

Verge, Tània (2011). “Institutionalising Gender Equality in Spain: Incremental Steps

from Party to Electoral Gender Quotas”. West European Politics (forthcoming).

Vickers, Jill (2010). “A Two-Way Street: Federalism and Women’s Politics in Canada

and the United States”. Publius: The Journal of Federalism 40 (3): 412–435.

Vickers, Jill (2011) “Gendering Federalism: Institutions of Decentralization and Power-

Sharing”. In Mona L. Krook and Fiona Mackay eds. Gender, Politics and

Institutions: Towards a Feminist Institutionalism. Basingstoke: Palgrave, pp. 129-

46.