For Richard Rose (ed) Encyclopedia of Electoral Systems Washington DC: CQ Press. Women s Representation and Electoral Systems

Transcription

1 For Richard Rose (ed) Encyclopedia of Electoral Systems Washington DC: CQ Press. Women s Representation and Electoral Systems Pippa Norris (Harvard University) Debates about electoral reform have revolved around the practical impact of changes to the status quo. Underlying these arguments are contested visions about the fundamental principles of representative democracy, the central normative criteria that an electoral system should meet, and whether the virtues of strong and accountable government are more or less important than social representation, including women and ethnic minorities. As is well known, women continue to be strongly underrepresented in elected office despite the scale of trends in the family and work-force, transforming almost every aspect of women and men s lives during the postwar era, and the growth of the women s movement altering the political culture. In spring 1999 there were just under 5000 women in parliament worldwide, representing 13.3% of the membership of lower houses and 10.6% of upper houses (Inter-Parliamentary Union 1999). The situation varies markedly by region (see Table 1); women do best in the Nordic region where they are on average 38.3% of MPs in the lower house, topped by Sweden, compared with only 3.6% in Arab states, with few or no female representatives in parliaments such as Djibouti, Kuwait and the United Arab Emirates. The proportion of women members elsewhere fell between these extremes, including in the Americas (15.4%), Europe (15.0%), Asia (14.3%), Sub-Saharan Africa (11.6%) and the Pacific (11.5%). The rank order of countries reveals that the level of socioeconomic development is not a necessary condition for women s success; in South Africa, Mozambique and Costa Rica, for example, female representation is far higher than in the United States, Italy and Japan. [See Table 1] Within this context the key issue is how electoral systems affect the structure of opportunities for women s representation. Ever since the seminal work of Maurice Duverger (1954) and Douglas Rae (1971), a flourishing literature has classified the main types of electoral systems and sought to analyze their consequences (see for example Lijphart and Groffman 1984; Taagepera and Shugart 1989; Lijphart 1994). The most important factors affecting women s representation are twofold: (i) the basic electoral formula that determines how votes are counted to allocate seats and (ii) the use of legal strategies of affirmative action. The Electoral Formula Electoral formulas are commonly categorized into four main types. Proportional representation (PR) includes open and closed party lists in regional or national constituencies using largest remainders and highest averages formula. Semi-proportional systems includes the single transferable vote, cumulative vote and limited vote. Majoritarian electoral systems include plurality, second ballot, and the alternative vote. Lastly, mixed systems are those like the Additional Member System combining majoritarian and proportional elements. The claim that party list PR facilitates the entry of women into elected office in established democracies has been confirmed in a series of studies (Norris 1985; Rule and Zimmerman 1992; Lijphart 1994; Matland 1998). The relationship between the electoral system and women s representation is influenced by many factors and is not automatic. In Western democracies the difference in women s representation between PR and majoritarian systems was quite modest until the 1970s, but the gap subsequently expanded (see Inter-Parliamentary Union 1995). District magnitude is commonly seen as particularly important with more women commonly elected under large multimember constituencies. Three reasons are commonly given for this phenomenon. First, in multimember districts, parties have an electoral incentive to produce a 'balanced' ticket in attempting to maximize their appeal, including all social groups in lists of 1

2 candidates for each district, such as northerners and southerners, farmers and shopkeepers, and women and men. Second, balancing the slate in PR systems also avoids internal conflict between party factions, like trade union or women s branches struggling to influence the selection process. Lastly, PR facilitates the use of measures of affirmative action and gender parity to encourage the selection of more women candidates. Some of these qualities are also present in semi-proportional systems, such as the single transferable vote used in Ireland, but these systems usually have a smaller number of candidates (4-5) in each district. In contrast in majoritarian systems, candidates are selected for single member districts. Where the selection process is in the hands of the local constituency party, there is no incentive for each to pick a ticket that is balanced at the district or national level. In these systems, party members often want a member who will maximize their chances of winning in that constituency, irrespective of the broader consequences for parliament (Norris and Lovenduski 1995). In mixed systems, such as in elections to the German Bundestag, we would expect far more women to be elected via party lists rather than through single member districts. Given the earlier literature, the key question is whether the relationship between electoral systems and female representation continues to hold in a wider range of newer and established democracies in the late 1990s. We can compare the representation of women based on the proportion of women elected to the lower house in May The comparison is restricted to fifty-three democratic states, defined as those countries with a combined Freedom House index of four or less, thereby excluding parliaments in authoritarian systems or under military rule that cannot exercise more than a symbolic role. The comparison confirms that women continue to be better represented under PR List systems. As a simple rule, women proved twice as likely to be elected under PR than majoritarian electoral systems. Women were 10.8 percent of MPs in majoritarian systems, 15.1 percent in mixed or semi-proportional systems, and 19.8 percent of members in PR systems. The pattern was far from linear (see Diagram 1) due to intervening systemic factors like the structure of opportunities to enter legislative office. In particular the process of recruitment within parties, including the informal and formal rules and the party organization, strongly influences the opportunities for women in elected office (Lovenduski and Norris 1993; Norris 1998). More women were elected in some majoritarian systems like Australia and Canada than in some highly proportional party list systems such as in Israel and Belgium. By itself the electoral system is neither a necessary nor a sufficient condition to guarantee women s representation. Nevertheless the electoral system functions as a facilitating mechanism, which expedites implementation of measures within parties, like affirmative action for female candidates. [Diagram 1 about here] The role of the electoral system is independent of levels of socioeconomic and political development. A regression model was run predicting the proportion of women in parliament the 53 democracies where the electoral system was coded from majoritarian (0), to mixed (0.5) and proportional (1). The model controlled for levels of socioeconomic development (measured by per capita GDP), providing an indication of the growth of the service sector and women s employment in the paid workforce, and the level of political development based on the Freedom House Gastil democratization index. The results confirmed that the type of electoral system continued to prove significant (standardized beta regression coefficients of.29 with sig. P.016), even after controlling for socioeconomic development (beta.28 with sig.09) and level of democratization (beta.24 not sig). Statutory Regulations of Affirmative Action Statutory quotas are also part of the electoral system where women must make up at least a minimal proportional of the legislature or of parliamentary candidates. In a handful of states by law women must make up a certain proportion of the list ballot: such as 50% in Italy and 30% in Argentina. France has actively debated the use of gender parity in elected office through constitutional reform (Reynolds and Reilly 1997). In India 33% of seats on local municipal elections are reserved for women and this practice has also been proposed for the Lok Sahba. Reserved seats are also employed in Bangladesh (30 out of 330 seats), Eritrea (10 out of 105) and Tanzania (15 out of 255). Affirmative action strategies like these can also serve political minorities based on regional, linguistic, ethnic or 2

3 religious cleavages, although the effects depend upon the size and spatial concentration of such groups. Reserved seats are used in countries as different as Jordan (Christians and Circassians), Pakistan (non-muslim minorities), New Zealand (Maori) and Slovenia (Hungarians and Italians) (Reynolds and Reilly 1997). Alternatively affirmative action or positive discrimination strategies have been commonly introduced through party rules to promote the training, funding or selection of women candidates (Lovenduski and Norris 1993; Norris 1998; Dahlerup 1998). Gender quotas adopted in party rules have been used all over the world, including by the ANC in South Africa, the PRD in Mexico and throughout Scandinavia. The British Labour party, for example, used all-women shortlists for nomination in half its target marginals in the run up to the 1997 general election. Although this policy was subsequently dropped under legal challenge, it still proved highly effective, contributing towards the doubling of the number of women in the Westminster parliament from A worldwide review of practices in the early 1990s, by the Inter-Parliamentary Union (1993), found twenty-two parties employed gender quotas for legislative elections, while fifty-one parties used them for elections to internal party posts. These measures were commonly introduced in Western Europe during the 1980s by parties of the left (Social Democratic, Labour, Communist, Socialist and Greens). In contrast in the first free elections parties within Central and Eastern Europe moved in the opposite direction, abandoning gender quotas largely in reaction to the old Communist regimes. Conclusions: A twin-track strategy We can conclude that debates about electoral reform have often produced conflict about means (what would be the effects on party fortunes of alternative systems?) but even more fundamentally about ends (what is the primary objective of the electoral system?). Those concerned with increasing women s representation have advocated two strategies: the adoption of proportional representation with large multimember constituencies and the use of affirmative action rules in candidate selection and in legislative bodies. The electoral system influences the incentives facing party selectorates when deciding whether to choose female candidates. In large multimember districts, parties have greater incentives to create a balanced ticket in the attempt to maximize their electoral appeal. In single member districts this incentive is absent at the local level, where the choice of candidate is often decided. Affirmative action strategies can be adopted in any electoral system but they may be easiest to implement when applied to balancing the social composition of party lists, for example zippering (designating every other position on the list for male or female candidates), or balancing the list by region, occupation, or religion. Although insufficient by themselves, proportional representation electoral systems, in combination with affirmative action strategies adopted by law or by party selection procedures, can therefore serve to increase social representation, producing parliaments that look more like the people they serve. 3

4 Table 1: Women s Representation Worldwide, 1999 Rank Country Lower or single House Elections Seats Women % W Upper House or Senate Elections Seats Women 1 Sweden Denmark Finland Norway Netherlands Germany * South Africa New Zealand Argentina " Cuba Austria Viet Nam Iceland Mozambique Seychelles Australia Monaco " Namibia China Spain Lao People's Democratic Rep Switzerland Canada Dem. People's Rep. of Korea Luxembourg Costa Rica Guyana United Kingdom * Turkmenistan Uganda Lithuania Ecuador " Mexico % W 4

5 31 Rwanda Latvia El Salvador United Rep. of Tanzania Dominican Republic Suriname Angola Bahamas " Czech Republic Eritrea Zimbabwe Jamaica " Saint Kitts and Nevis " San Marino " United States of America Venezuela Poland " Portugal Slovakia Guatemala Philippines Mali Senegal Azerbaijan " Belgium " Congo " Ireland Colombia Kazakhstan Bolivia Indonesia Cape Verde " Italy " Saint Lucia " Trinidad and Tobago France Bulgaria " Chile " Peru

6 57 Syrian Arab Republic Russian Federation * Guinea-Bissau Nicaragua Zambia Dominica " Honduras Malta Bangladesh " Sao Tome and Principe Ghana Republic of Moldova Guinea Botswana Gabon " Hungary Cambodia ?? " Samoa Burkina Faso " India Côte d'ivoire " Madagascar Croatia " Mongolia Malaysia " Slovenia " Ukraine Mauritius Israel " The F.Y.R. of Macedonia Tunisia Romania Andorra " Uruguay Belize " Georgia Central African Republic " Iraq

7 82 Armenia " Greece " Maldives Burundi " Uzbekistan Brazil Cameroon " Malawi " Thailand Cyprus Sri Lanka " Sudan Albania Yugoslavia Iran (Islamic Rep. of) " Kiribati Saint Vincent & the Grenadines Japan Fiji " Singapore Turkey Liechtenstein Lesotho " Mauritania Republic of Korea Haiti ?? " Kenya Algeria Swaziland Tajikistan Paraguay Chad Lebanon " Pakistan Bhutan 1997 * " Egypt " Ethiopia ?? " Gambia " Solomon Islands Papua New Guinea

8 107 Kyrgyzstan Yemen Morocco Djibouti " Jordan " Kuwait " Micronesia (Fed. States of) " Nauru " Palau " Tuvalu " United Arab Emirates " Vanuatu * Data valid as of or for the year indicated Source: Women in World Parliaments. Geneva: Inter-Parliamentary Union. May

9 % Women MPs by Electoral System % Women Lower House Pakistan Nepal UK USA Philippines Mali France Chile Zambia Bangladesh India Ukraine Thailand Malawi South Korea Turkey Germany New Zealand Mexico Ecuador Venezuela Ireland Bolivia Italy Russia Hungary Japan Taiwan Australia Canada Sweden Denmark Finland Norway Netherlands South Africa Argentina Austria Mozambique Spain Switzerland Costa Rica Czech Rep Portugal Poland Belgium Colombia Bulgaria Madagasgar Israel Uraguay Greece Brazil Majoritarian Mixed Proportional 9

10 References Dahlerup, Drude Using Quotas to increase women s political representation. In Women in Parliament: Beyond Numbers. Edited by Azza Karam. Stockholm: IDEA. Duverger, Maurice. (1954) Political Parties New York:Wiley. Inter-Parliamentary Union (1995) Women in Parliament Geneva: IPU. Inter-Parliamentary Union (1993) Electoral Systems: A Worldwide Comparative Study Geneva: IPU. Interparliamentary Union Women in National Parliaments. Lijphart, Arend. and Groffman, B. (1984) Choosing a Political System New York: Praeger. Lijphart, Arend. (1994) Electoral Systems and Party Systems. Oxford: Oxford University Press. Lovenduski, Joni and Norris, Pippa. (1993) Gender and Party Politics London: Sage. Matland, Richard E Enhancing Women s Political Participation: Legislative Recruitment and Electoral Systems. In Women in Parliament: Beyond Numbers. Edited by Azza Karam. Stockholm: IDEA. Norris, Pippa Women in European Legislative Elites, West European Politics, 8(4): Norris Passages to Power. Cambridge: Cambridge University Press. Rae, Douglas. (1971) The Political Consequences of Electoral Laws New Haven: Yale University Press. Reynold, Andrew and Ben Reilly. Eds The International IDEA Handbook on Electoral System Design. Stockholm: IDEA. Rule, Wilma and Zimmerman, Joseph. F. (1992) United States Electoral Systems: Their Impact on Women and Minorities New York: Praeger. Taagepera, Rein and Shugart, M.S. (1989) Seats and Votes New Haven, Yale University Press. 10