The Combination of Gender and Ethnic Quotas in Electoral Politics
The Combination of Gender and Ethnic Quotas in Electoral Politics
Abstract and Keywords
Around the world, countries are increasingly using quotas to enhance the diversity of political representatives. This chapter considers the histories and policy designs of ethnic and gender quotas that regulate national legislatures. Most countries with quotas target only one type of under-represented group—for example, women or ethnic minorities. Even in countries with both gender and ethnic quotas (called ‘tandem quotas’), the policies typically evolved separately and work differently. Women and ethnic minorities are treated as distinct groups, ignoring the political position of ethnic minority women. However, a handful of countries have ‘nested quotas’ that specifically regulate the political inclusion of ethnic minority women. The second half the chapter focuses explicitly on nested quotas. It lays out how nested quotas work, where and how they have been adopted, and the prospect for their spread to new countries in the future. The chapter concludes with reflections on the promises and pitfalls of nested quotas as a vehicle for multicultural feminism.
In 2001, the rule of the Taliban—one of the most repressive regimes towards women in modern history—came to an end. Shortly thereafter, women’s political representation in Afghanistan increased dramatically. A country that had never before elected even 5 per cent women to its national assembly suddenly elected the 24th highest share of women national legislators in the world (IPU 2016b). To get there, Afghanistan used a quota—a law or policy requiring candidate lists or representative bodies to include members of targeted groups (Hughes et al. 2017). Specifically, the Afghan constitution sets a minimum threshold quota of 27 per cent of seats in the House of the People for women, including three of the ten seats reserved for Kuchis, a nomadic Pashtun group from southern and eastern Afghanistan. With this policy, Afghanistan became the first country in the world to guarantee national legislative representation to women from a particular ethnic, racial, religious, or tribal group.
Afghanistan is far from alone in using quotas to shape who is represented in legislatures. As discussed in Chapter 1, pressure for gender equality has taken a ‘participatory turn’, whereby ending legal discrimination against women is no longer sufficient. Gender quotas have become a common way that governments can guarantee women’s participation as political leaders. Indeed, by 2015, some seventy-three countries had at some point written national gender quotas into their electoral laws or constitutions (Hughes et al. forthcoming).1 Although less frequent, legislative quotas for ethnic, racial, religious, and tribal groups (hereafter simplified as ‘ethnic quotas’) are also used increasingly throughout the world to stave off ethnic conflict or to ensure the diversity of (p.98) elites in multicultural societies (Bird 2014; Htun 2016; Lublin 2014; Reynolds 2005).2 Examples include seats reserved for Turks in Cyprus, ‘Black communities’ in Colombia, Christians in Iran, and Māori in New Zealand.
Though uncommon, some countries have legislative quotas that target both women and ethnic minorities—referred to as ‘tandem quotas’ or ‘dual quotas’. These contexts are logical sites to investigate what happens when the participatory turn in gender equality and multicultural accommodation collide. Yet, gender and ethnic quotas are often treated as separate systems. Even in countries with tandem quotas, gender and ethnic quotas typically come into being at different times, use distinct mechanisms, and operate independently of one another (Bjarnegård and Zetterberg 2014; Bird 2016; Htun 2016; Htun and Ossa 2013; Hughes 2011; Krook and O’Brien 2010). Furthermore, countries often give little or no attention to groups positioned at the intersection of gender and ethnic quotas—ethnic minority women (Hughes 2011).
However, like Afghanistan, a handful of countries have policies that explicitly target ethnic minority women. ‘Nested quotas’ (also called ‘subgroup quotas’) require that some portion of seats or candidate list positions go to a targeted subset of members, typically a marginalized subgroup (Hughes 2014b). In theory, nested quotas for ethnic minority women could ensure that the participatory turn in gender equality does not leave ethnic minority women behind, and that accommodating ethnic minority groups with quotas does not empower solely men to lead and represent the group’s interests. In practice, however, nested quotas may be more a tool of the powerful than a means to advance multicultural feminism.
In this chapter, I investigate gender quotas, ethnic quotas, and how these policies interact with one another, focusing only on policies that regulate the single or lower house of a national legislature. I begin by trying to make sense of the patterns we see today by looking into the past. I consider the varied histories of ethnic and gender quotas and discuss when and why these policies have been adopted. Next, I identify differences between ethnic and gender quotas and survey the arrangements that exist in countries with both policies. (p.99) In the third section, I turn my attention to nested quotas, laying out three different institutional arrangements for nested quotas, reviewing how the countries with nested quotas adopted them, and speculating how other countries could adopt them through examples of India, Venezuela, and Brazil. I conclude with a discussion of the promises and pitfalls of nested quotas as a way to ensure that ethnic minority women benefit from attempts to guarantee women’s participation and accommodate multiculturalism.
4.1 Ethnic and Gender Quotas: When and Why Are They Adopted?
To understand why nested quotas are so rare, it is essential to first appreciate that ethnic and gender quotas have distinct histories. They have been adopted in different times, places, and under different circumstances. Figure 4.1 shows the number of independent countries having ever adopted each type of quota, beginning in 1900. Ethnic quotas—shown as the grey solid line—came first and spread slowly and steadily.3 Ethnic quotas have become more popular over time, but the pace of their adoption has not been particularly remarkable. Gender quotas—shown as the black dotted line—were adopted later but after 1995 spread rapidly. By 1997, the number of countries having adopted gender quotas surpassed those having adopted ethnic quotas, and gender quotas have continued to multiply at a faster rate. Finally, the number of countries having ethnic and gender quotas together—shown as the line of small white circles—did not start to spread until the mid-2000s. In this section, I briefly explain these different patterns, focusing first on ethnic quotas, then on gender quotas, and finally on their co-evolution.
4.1.1 Ethnic Quotas
For many countries, ethnic quotas have been around since sovereignty. Indeed, the majority of countries that have adopted ethnic quotas used them to elect their first national legislature after independence. Therefore, the timing of quota adoption is driven, in part, by when countries became independent or (p.100) established representative institutions. Iran (1906) and New Zealand (1908) are the first countries to show up in Figure 4.1, but Vietnam (1946), India (1950), Cyprus (1960), Fiji (1970), and Djibouti (1977) are other examples where ethnic quotas date to independence.
Many ethnic quotas have their roots even farther back in history than is shown in Figure 4.1—back to colonial times. Across Africa and Asia, the British Empire frequently instituted quotas to guarantee representation for White European settlers, immigrant workers, and/or indigenous populations (Htun 2004; Krook and O’Brien 2010; Reilly 2001; Reynolds 2005). For example, in Fiji, during early colonial times, the British government instituted separate communal rolls for indigenous and Indo-Fijian voters; a variant of this system was then institutionalized in independent Fiji in the 1970 constitution (Reilly 2001). Other quotas were put in place by colonial settlers. For example, in New Zealand, reserved seats for the indigenous Māori were adopted in 1867 by British settlers, who by that time were self-governing (New Zealand Parliamentary Library 2009).4
(p.101) If not to ‘deal with the natives’, ethnic quotas were often devised to address or prevent ethnic conflict. Take the case of Cyprus, ruled in turn by the Turks and the British. In the 1950s, the majority Greek population in Cyprus began to resist British rule, demanding that the island became part of Greece; the Turkish minority wanted instead a partition into Greek and Turkish parts of the island. After years of violent conflict, the British helped broker an agreement for an independent Cyprus that split political positions among Greek and Turk ethnic groups (Schneckener 2002).
These two quota logics—to provide representation for indigenous or minority populations and to stem ethnic conflict—continue to explain ethnic quota adoption in more recent times. Beginning in the 1990s, Latin American countries set aside legislative seats for indigenous and Afrodescendent peoples. Venezuela came first, followed by Colombia, and most recently Bolivia. Reserved seats for autochthonous (indigenous) populations also popped up in Eastern Europe, for example, in Slovenia and Montenegro. Reserved seats have also been adopted in countries exiting violent conflicts, including Burundi and Iraq. And even though the British Empire may no longer be driving the adoption of reserved seats, international actors continue to shape ethnic power-sharing agreements, especially in the wake of ethnic conflicts (Krook and O’Brien 2010).
This is not to say that the reasons for ethnic quotas did not evolve over time. Some early quotas were designed quite explicitly to ensure the rule of privileged minorities. India, South Africa, and Zimbabwe all had quotas for Whites or ‘Anglos’; they were a tool of the powerful to maintain their power. More recent policies are seen instead as a way to shore up the representation of small and/or geographically dispersed marginalized groups (Lijphart 1994; Lublin and Wright 2013; Reynolds 2005). Using ethnic quotas to enhance the political representation of marginalized groups remains controversial, but the ethnic quota has generally evolved ‘from an obstacle to an instrument of democracy’ (Htun 2016: 3).
4.1.2 Gender Quotas
Compared to ethnic quotas, gender quotas are generally understood as a more recent phenomenon. But, the origins of gender quotas actually trace back to the 1930s (Krook 2009). In 1935 in British India, the Crown allocated women small seat shares (less than 4 per cent) in national assemblies (Afzal 1999). After Pakistan’s independence, a similar seat share was set aside for women in its Constituent Assembly (1947) (Krook 2009). The same year, women’s activists in China won a decade-long battle for a gender quota, as 10 per cent of seats in the National Assembly and Legislative Yuan were set aside for women (Edwards 1999). Reserved seats for women proliferated slowly across (p.102) the next few decades to Asia (including Bangladesh and the Philippines) and Africa (Sudan, Ghana, Tanzania, and Egypt) (Abou-Zeid 2003; Chou and Clark 1994; Chowdhury 2002; Tripp et al. 2006).
National gender quotas remained relatively rare until the 1990s. Before 1990, just seven independent countries reserved seats in their parliaments for women, often through temporary measures (Abou-Zeid 2003; Chou and Clark 1994; Chowdhury 2002; Tripp et al. 2006). These quotas tended to be like Pakistan’s and China’s, allotting women a modest seat share. In 1989, Uganda became the first country to reserve more than 10 per cent of seats for women in its national legislature (Hughes et al. 2015).
Then, the landscape of gender quotas shifted. In 1990, Nepal became the first country to adopt a candidate quota, requiring 5 per cent women on all political parties’ candidate lists (Hughes et al. 2017). The following year, Argentina followed suit, requiring that women be a minimum of 30 per cent of candidates on all party lists (Jones 1998). Candidate quota adoption took off: between 1995 and 1999, thirteen countries adopted them (Hughes et al. forthcoming). Although more slowly, reserved seats for women also continued to spread, including to Kenya (1997), Somalia (2001), Morocco (2002), and Rwanda (2003). In recent years, ‘quota fever’ has persisted (Paxton and Hughes 2016). Indeed, fourteen countries adopted gender quotas for the first time between 2011 and 2012 alone (Hughes et al. forthcoming; Dahlerup et al. 2014). Today, national gender quotas touch all corners of the globe and have been adopted by countries at all levels of economic development and democracy (Dahlerup 2006; Krook 2009; Muriaas et al. 2013).
How and why did national gender quotas spread so quickly during the last twenty-five years? Or, as Kymlicka and Rubio-Marín (this volume) might ask, how do we make sense of this ‘participatory turn’ in gender equality? Gender quota adoption has been driven by forces working at international, regional, and local levels (Hughes et al. 2017). Increasing international pressure to include women in political decision-making by embracing gender quotas has been both normative and economic: countries adopted gender quotas to signal their status as ‘modern’, and to satisfy explicit demands from international donors and foreign governments (Bush 2011; Hughes et al. 2015; Krook 2009; Swiss and Fallon 2016; Towns 2010).5 Regionally, diffusion has played a key role, leading countries to adopt policies like those of their neighbours (Bush 2011; Hughes et al. 2015). For instance, Argentina’s candidate quota innovation set up Central and South America to become the global leader in candidate quotas (Htun 2016; Krook 2009; Piscopo 2015; Schmidt and Saunders 2004; Schwindt-Bayer 2010). Locally, women activists (p.103) have often been at the forefront of the push for gender quotas (e.g. Bruhn 2003; Krook 2009).
Although the histories of ethnic and gender quota adoption differ, two key drivers of ethnic quota adoption also help explain gender quota adoption: armed conflict and colonialism. The introduction of gender quotas in post-conflict countries is not necessarily designed to prevent future conflicts (although that argument is sometimes made). However, as with ethnic quotas, conflict opened up space in which international actors were able to press for gender quotas (Anderson and Swiss 2014; Hughes and Tripp 2015; Tajali 2013; Tripp 2015). Colonialism also factors into gender quota adoption, but not only because colonial powers instituted them in colonies. After decolonization, gender quotas diffused between former colonizers and colonized. For instance, women in Argentina first learned about gender quotas from women in Spain, where the Spanish Socialist Party had been using an internal quota since the late 1980s (Krook and O’Brien 2010). And, the ‘parity’ measures adopted by France in 1999 spread to her former colonies of Senegal and Tunisia in 2007 and 2001, respectively (Hughes et al. 2015).
4.1.3 Tandem (Ethnic and Gender) Quotas
So, when and why did tandem quotas come on the scene? The first country to have both gender and ethnic quotas in place at the same time is Pakistan.6 Remember that Pakistan was one of the first countries to adopt gender quotas in 1947, a policy they carried over from colonial times. Although quotas for religious minorities also trace to the colonial period, Pakistan did not adopt ethnic quotas until 1973. East Pakistan had just split off to become Bangladesh, and the new government introduced reserved seats for non-Muslims to safeguard the representation of the few religious minorities who were left (Rais 2004).7 Notably, from 1973 until today, Pakistan’s pioneering use of gender and ethnic quotas together has received little acknowledgement or attention.
The use of ethnic and gender quotas together did not really start to spread across countries until the mid-2000s (see Figure 4.1). One reason is that gender quotas have only recently become a widely accepted tool to promote women’s representation. During the earliest years of ethnic quota adoption, women were not generally expected to participate widely in elite politics. Second, when gender quotas began to take off, quota advocates articulated (p.104) the need for policies for women in particular, in some cases actively distancing their campaigns from affirmative action for other groups (Bird 2003). Overall, different identities have been politicized successfully at different times and in different places (Krook and O’Brien 2010).
The lack of overlap between ethnic and gender quotas is not simply a product of separate reform movements. Research also suggests that when one or more marginalized groups are granted policies that promote their representation, other groups may have a harder time achieving the same success (Hughes 2015; Krook and O’Brien 2010). India is a quintessential example; although their ethnic quota dates to before independence, legislators have refused to enact a Women’s Reservation Bill for more than two decades (see more on the Indian case in Section 4.3.3).
Despite pressures against adopting gender and ethnic quotas in the same country, the number of countries with tandem quotas has been rising during the last fifteen years. Between 2000 and 2015, eleven countries moved from having either gender or ethnic quotas to having both. In fact, today, the number of countries with tandem quotas is about the same as the number of countries with quotas alone.
Why are tandem quotas adopted? In many cases, the presence of both in a single country seems to be largely happenstance. Consider Samoa, which had ethnic quotas from independence, but the exclusion of women from public office remained severe and persistent. For instance, in the 2011 election, women won only two of forty-nine seats (4 per cent) in the Legislative Assembly (IPU 2016a). It was not until 2013, amid mounting international pressure for change, that Samoa adopted quotas for women. The longstanding use of ethnic quotas may have shaped the type of gender quota that was ultimately adopted (reserved seats), but seemed not to factor into the ultimate decision to adopt a gender quota.
Ethnic and gender quotas do not need to be adopted at distant points in time to evolve separately. Consider the case of Venezuela. In 1997, Venezuela passed the Suffrage and Political Participation Act, which included a provision requiring that women make up 30 per cent of candidates on closed party lists. The following year Hugo Chávez was elected president. Chávez spearheaded the writing of a new constitution, which established protections for indigenous peoples—including a provision of three reserved seats in the Chamber of Deputies—and recognized women and men as equal citizens. Ironically, the 1999 constitution—the very document that advanced the cause of women’s liberation in Venezuela—also laid the groundwork for the end of the gender quota (Wagner 2005). Before the 2000 elections, the Supreme Court of Justice ruled that the quota was incompatible with the principle of full equality enshrined in the new constitution (International IDEA and UCAB 2015). Activists called for a new parity measure to replace the repealed provision, (p.105) but it would be some fifteen years (and after Chávez’s death) until that parity was implemented alongside ethnic quotas during a Venezuelan election.8
Not all ethnic and gender quotas evolved separately, however. In 2003, Burundi began to transition out of a ten-year ethnic civil war. As part of the peace process, a power-sharing quota system was put in place in the national legislature, the cabinet, and the military to ensure balanced political representation across Hutu, Tutsi, and Twa ethnic groups. During the transition, women also successfully lobbied for a national gender quota. Ultimately, the 2005 constitution divided up seats in the National Assembly by ethnicity (Hutus 60 per cent, Tutsis 40 per cent, and Twa three seats) and gender (women 30 per cent).
Although Burundi is the only example of ethnic and gender quotas being first adopted at the exact same time, wars are a common backdrop for tandem quota adoption.9 Take the case of Iraq. Following the US invasion, women’s groups pushed for a 25 per cent gender quota for elections to the Transitional National Assembly, a share that was ultimately codified in the 2005 constitution and electoral law (Krook et al. 2010). In 2009, in the midst of ongoing war, the legislature adopted quotas for religious minorities—Christians, Yazidi, Sabeans, and Shabak (USDS 2011).10 Thus, war and the transition to peace facilitated both gender and ethnic quota adoption within a few years. Or consider Djibouti, which began guaranteeing representation in its National Assembly for ethnic Somalis, Afars, and Arabs before independence. In 2001 a civil war came to an end, and the following year the president issued a decree (p.106) requiring women make up at least 10 per cent of party lists (Decree 2002–0253, Article 2).11
Next, I focus more closely on how quotas are designed. Especially when gender and ethnic quotas appear together, what do these policies look like, and how do they interact with one another? I also consider the effects of gender and ethnic quotas on legislative diversity.
4.2 Gender and Ethnic Quotas: Apples and Oranges?
Given their distinct patterns of development, it is perhaps unsurprising that ethnic and gender quotas are often different from one another (Bird 2016; Bjarnegård and Zetterberg 2014; Hughes 2011). Even when countries have both gender and ethnic quotas, policies are typically designed differently and implemented as separate systems (Bjarnegård and Zetterberg 2014; Htun 2004; Krook and O’Brien 2010). In their study of countries with tandem quotas, Bjarnegård and Zetterberg (2014: 317) find that ethnic minorities are often guaranteed presence in legislatures through the creation of special districts, whereas gender quotas are more commonly added on to existing constituencies:
In general, policymakers designing quotas for minority groups appear to perceive separation from the dominant political actors as the priority. Quotas for minorities are designed to grant the group certain self-determination … Quotas for women, on the other hand, seem to be designed with their integration into the regular (s)election process in mind. Quotas for women are designed to ensure that women are elected by—and often also compete with—both men and women.
In part because of their different designs, ethnic and gender quotas often do not interact with one another. Some reserved seats for ethnic groups are exempted from gender quota laws, and others target too few seats for gender quotas to have any effect. For example, in Colombia, lists submitted for election of five or more seats must include 30 per cent women (Dahlerup et al. 2014), but only one seat is set aside for indigenous peoples, precluding any interaction between the two policies.
(p.107) Gender and ethnic quotas both enhance the diversity of political elites (Hughes 2011; Hughes et al. 2017). However, when gender and ethnic quotas are regulated differently and do not interact with one another, it may have negative consequences for those affected by both policies—ethnic minority women (Hughes 2011; Bird 2016). Gender quotas might primarily help ethnic majority women get elected, and the benefits of ethnic quotas might accrue predominantly to ethnic minority men. The economic, cultural, and political circumstances ethnic minority women face often put them at a political disadvantage, and quotas may do little to change the status quo.
Yet, in some countries, ethnic and gender quotas operate similarly, and may interact in ways that help ethnic minority women. Burundi provides one such example. Remember that in Burundi, ethnic groups and women are allotted a percentage of total seats in the Chamber of Deputies: ethnic Hutus 60 per cent, ethnic Tutsis 40 per cent, and women 30 per cent.12 (The ethnic split in the legislature is more balanced than in the general population, where Hutus are 85 per cent and Tutsis are 14 per cent.) The quota is regulated through placement mandates for candidates on party lists: for every three candidates on a party list, no more than two may be from one ethnic group, and out of every four candidates, at least one must be a woman (Burundi Electoral Code, Article 108). Then, if desired percentages are not achieved, the Independent National Electoral Commission co-opts legislators after the election to balance out the numbers. The number of co-opted legislators has varied from one election to the next (eighteen in 2005, six in 2010, and twenty-one in 2015).
In cases like Burundi, where the same mechanism regulates the election of both ethnic minorities and women, ethnic minority women may reap the benefits (Hughes 2011, 2016). In Burundi, Tutsi women can satisfy both the quota for Tutsis and the quota for women at once, and are thus the most likely group to be co-opted.13 As a result, Tutsi women are overrepresented in the legislature relative to their share of the population. In fact, in 2015, Tutsi women won more seats in the legislature than either Hutu women or Tutsi men (Hughes 2016).
Aside from using similar mechanisms to balance ethnicity and gender, Burundi is unusual for other reasons. One is the scope of its ethnic quota. Only a handful of countries have laws that carve up the entire legislature into shares for particular ethnic groups. The few other cases—Cyprus and Lebanon—lack gender quotas. Most countries with tandem quotas look more (p.108) like Colombia, where the ethnic provisions regulate a small share of seats, typically fewer than 5 per cent of the total seats in the legislature (for example, in Iraq, Pakistan, Samoa, Slovenia, and Venezuela). Such a configuration is unlikely to boost the representation of ethnic minority women. Indeed, research suggests that particularly if gender or ethnic quotas influence only a small share of seats, benefits to ethnic minority women may be minimal (Hughes 2011). What can be done to ensure that ethnic minority women do benefit from quota provisions? In Section 4.3, we consider one possible approach: nested quotas.
4.3 Nested Quotas
Nested quotas are still a relatively new and rare phenomenon. Consequently, researchers have not agreed on a definition of what constitutes a ‘nested quota’. In this chapter, I define nested quotas as those that require some portion of seats or candidate list positions to be explicitly set aside for subgroups—here, ethnic minority or indigenous women (Hughes 2014b).14 This is a narrower definition than advanced by Karen Bird (2016), who authored the most comprehensive study of nested quotas to date. To Bird (2016), nesting occurs when the gender quota rule applies to any portion of designated ethnic seats, or when the ethnic quota rule applies to seats designated for women.
Although a bit technical, I want to take a moment to address the differences between Bird’s (2016) definition and mine, because these differences have empirical consequences for what counts as a nested quota. Although her definition may sound similar to mine, consider the case of Burundi, which Bird counts as a nested quota, whereas I do not. In Burundi, the electoral law has rules for ordering ethnic groups and women on the same party lists (meeting Bird’s definition of nesting), but does not require explicitly that a certain share of Hutu or Tutsi candidates are women, or that a certain share of women candidates are Hutu or Tutsi (failing to meet my definition of nesting).15 Djibouti, which also uses candidate requirements for both ethnicity and gender, is similar; Bird counts Djibouti as having a nested quota, whereas I do not. (p.109) Using my narrower definition, I identify only four independent countries with nested quotas in 2015: Afghanistan, Bolivia, Jordan, and Nepal.16
4.3.1 How Do Nested Quotas Work?
Although there are only a handful of nested quotas, there is important variation among them in how they operate. To better make sense of nested quotas, I develop a typology for thinking about how they work—or could work, in theory. I categorize nested quotas by the mechanism through which they operate. Focusing on two main types of quotas—reserved seats and candidate rules—I lay out three different arrangements for nested quotas: (1) nested reserved seats, (2) candidates nested in seats, and (3) nested candidate quotas.
First, in nested reserved seats, some share of women’s reserved seats must go to ethnic groups or, similarly, some share of seats for ethnic groups must go to women. Afghanistan and Jordan both have nested reserved seat systems. In Afghanistan, the constitution guarantees women are a minimum share of seats, at least two per province. In a second provision, ten seats are set aside for Kuchis, including three for women. The quota system in Jordan works differently in that some ethnic quotas are exempt from the gender quota. Women are guaranteed seats in each of twelve multi-member districts, including in three districts reserved for Bedouins. Yet, the twelve seats reserved for Christians and three for Circassians/Chechens need not include women. To make sense of the difference between Afghanistan and Jordan, we can invoke Bird’s (2016) useful distinction between ‘fully nested’ and ‘partially nested’ systems. Afghanistan has a fully nested system; the gender quota rules apply to all seats, including seats reserved for Kuchis. Jordan has a partially nested system, where only some ethnic groups are subject to gender provisions (Bird 2016).
In a second option, one group has reserved seats and another group has candidate quotas. That is, some share of candidates for ethnic reserved seats must be women, or some share of candidates for women’s reserved seats must (p.110) go to a particular ethnic group. Although both arrangements are plausible, only the former has existed in the world so far. In 2010, Bolivia adopted a parity law, requiring that women comprise half of all candidates, including seats reserved for indigenous groups.
The third and final option is a nested candidate quota: some share of ethnic candidates must be women, or some share of women candidates must be members of a certain ethnic group or groups. Nepal’s 2007 interim constitution and election law established this type of quota.17 For large parties, candidates on electoral lists must include 31.2 per cent Madhesi, 13 per cent Dalits, 37.8 per cent indigenous, 30.2 per cent Khas and Aryan, and 4 per cent from backward regions, and for each of these groups, half of the candidates must be women.18
4.3.2 The Adoption of Nested Quotas
How do nested quotas come about? Although we only have a few cases to study, there is evidence that the same factors that drive tandem quota adoption, in general, seem to lead to nested quotas, in particular. For one, the narrative of Bolivia’s nested quota adoption shares some similarities to the story of tandem quota adoption in Venezuela. In Bolivia, the first self-declared indigenous president, Evo Morales, was elected in 2005. His leftist political party, Movement toward Socialism (MAS) adopted a 5 per cent quota for indigenous groups while also introducing parity for women, upping women’s share of positions on electoral lists from 30 per cent to 50 per cent (Htun 2016; Htun and Ossa 2013). In both Venezuela and Bolivia, then, the use of gender and ethnic quotas together followed the rise of an ideologically leftist and transformational leader.
As with all other quotas, war and peace also loom large in the stories of countries with nested quotas. As discussed at the beginning of this chapter, Afghanistan adopted quotas for women and Kuchi nomads following the US invasion in 2001. Gender quotas were new, but Afghanistan had reserved seats (p.111) for Kuchis before.19 Nepal is a similar case. After a ten-year civil war with Maoists came to an end in 2006, Nepal laid out requirements for the representation of women, lower castes, and indigenous tribes in its interim constitution (Interim Constitution of Nepal 2007). In Nepal’s case, a gender quota had been in place (the 1991 candidate quota of 5 per cent was increased to 33 per cent), but the ethnic quotas were new.
Next, we can return to the path to quota adoption of Jordan. In many ways, Jordan looks to have followed a path to its nested quota similar to Samoa’s path to its tandem quota. In both countries, ethnic quotas date to independence, women’s representation remained low for a long period, and gender quotas were finally adopted in the 2000s. Like Samoa, pressure from domestic women’s movements and international organizations is likely part of Jordan’s quota adoption story. However, Jordan presents an interesting case to look at more carefully, because the addition of nested quotas came nearly a decade after Jordan began using tandem quotas. Jordan first adopted a gender quota in 2003 and reformed the policy in 2010, but the nested quota for Bedouin women was not added until 2012.
Why did Jordan adopt a nested quota for Bedouin women? At the time, women’s and human rights activists were arguing that the six seats reserved for women in the 120-seat National Assembly were too few. Doubling the quota to twelve seats, one woman per district—including the three districts for Bedouins—threw these activists a bone (Hughes 2014b). However, it is also important to know that Bedouins have formed a bedrock of support for the monarchy that has ruled Jordan since independence. Expanding Bedouin women’s representation may have also been a way to increase the chances of electing loyalist women (Hughes 2014b). Interestingly, the same argument could be made in Afghanistan, since Kuchi nomads are ethnically Pashtun, Afghanistan’s historically ruling group (Hughes 2014b).
4.3.3 Prospects for Future Nested Quotas
Quotas are continuing to spread, and tandem quotas are increasingly popular. As countries weigh options for how to design tandem quota systems, nested quotas may also spread. Nested quotas are possible in countries with just about any ethnic, racial, and religious composition. Here, I consider the prospects for nested quotas in two cases that already have some form of quotas—India and Venezuela.
Take first the case of India, one of the most diverse countries in the world. Since independence, India has been committed to ensuring diversity in its (p.112) parliament through reservations for Scheduled (low-ranking) Castes and Scheduled Tribes (SCSTs). India reserves 18.4 per cent of seats in the Lok Sabha (lower house) for SCSTs. Which groups are eligible to run in SCST districts—and whether quotas in the Lok Sabha should be extended to other groups—has been bitterly disputed (Randall 2006). To date, however, groups designated as Other Backward Classes (OBCs) and other marginalized groups such as Muslims are not guaranteed seats in the national legislature and must contest the ‘general’ seats.
Gender quotas, too, have a long and contested history in India. Legislators first formally introduced a constitutional amendment to reserve seats for women in the Lok Sabha in 1996 via the Women’s Reservation Bill (WRB) (Nanivadekar 2005). By design, the WRB includes a nested component; the bill would set aside one-third of seats for women, including among SCST seats.20 However, the quota has been criticized on the basis that it would benefit upper caste, elite women, forcing OBC and Muslim men out of the general seats (Dubochet 2014; Menon 2000; Randall 2006). In response, some have called for alternative arrangements such as nested quotas for OBC and Muslim women within the general seats reserved for women (Randall 2006).
Currently, it is difficult to predict if a national gender quota in India will ever pass. The bill passed the upper house in 2010, and many speculated that the 2014 elections would facilitate the quota’s adoption, but the bill has been shelved time and again. Still, looking globally, India stands out as a likely candidate for a nested reserved seat quota. If a gender quota passes in India, it is most likely to introduce some form of nested reserved seats.
Venezuela is also an interesting case to which to return. Like India, Venezuela already has a history of quotas. But, the similarities between Venezuela and India end there. The number of distinct ethnic groups in Venezuela is significantly fewer than in India. In Venezuela, 3.7 per cent is Black or Afrodescendant, 2.7 per cent is fully Amerindian, 1 per cent is ‘Other’, and the rest are White or Mestizo. Also different than India, Venezuela has a tandem quota system at the national level already. As already discussed, Venezuela has had both reserved seats for indigenous peoples and a parity quota since 2015.
Given Venezuela’s existing provisions, adopting a nested quota would be complicated. Reserved seats for indigenous peoples are elected through three single-member districts that each span multiple states. Venezuela sets a high bar for contesting these seats, requiring a candidate to: ‘have exercised a position of traditional authority in her or his community; have a recognized history in the social struggle for the recognition of cultural identity; have (p.113) engaged in behavior to benefit indigenous peoples; and be [a] member of a legally recognized indigenous organization that is at least three years old’ (Angosto Ferrández 2011: 21; cited in Htun 2016: 42).
What is the clearest way forward for Venezuela to guarantee the representation of indigenous women? Venezuela could amend its constitution to create nested reserved seats, perhaps doubling the number of seats and requiring one indigenous woman and one indigenous man from each of the three districts. Or, Venezuela could adopt a system of candidates nested in seats, adding a provision requiring parties contesting indigenous reserved seats to put forward equal numbers of men and women candidates.
Overall, the cases of India and Venezuela are fairly straightforward to consider because they already have ethnic quotas in place. In some ways, it is easier to theorize how to add gender quotas to existing policies than vice versa. Ethnic quotas generally exhibit a much wider degree of policy variation than do gender quotas. However, it is certainly possible that countries with gender quotas could add ethnic quotas to the mix. Consider Brazil, which has a nominal gender quota already in place. Brazil had been debating quotas for Afro-Brazilians for congressional races for more than a decade (Htun 2004, 2016). If they do, Brazil should consider nested candidate quotas as a way to ensure Afro-Brazilian women’s inclusion as well.
The spread of quotas is one of the most important political trends in the last quarter century (Hughes et al. 2017). Most countries in the world today use some form of ethnic quotas, gender quotas, or both, to ensure diversity among political elites. Yet, looking at ethnic and gender quotas together suggests that these two types of policies almost always come about independently, without clear evidence of cross-pollination, and are often regulated in different ways.
Where ethnic and gender quotas exist together, how each policy is designed and how the two policies interact with each other could matter most for those directly affected by both policies—ethnic minority women. Especially where quotas regulate a small share of legislative seats, gender quotas may benefit ethnic majority women, and ethnic quotas may benefit ethnic majority men, leaving ethnic minority women with minimal or no representation whatsoever. Alternatively, where mechanisms for filling seats for women and ethnic minorities are similar, the ability of an ethnic minority woman to satisfy both ethnic and gender provisions in the law may make them especially desirable candidates, leading to their overrepresentation relative to other marginalized subgroups. Overall, as stand-alone policies or in conjunction with one another, gender and ethnic quotas may fail to guarantee an equitable allocation of seats across marginalized groups and subgroups.
(p.114) To address this limitation, a handful of countries are already using nested quotas to ensure a particular balance of ethnic majority women, ethnic majority men, and ethnic minority women. Nested quotas exist in quota systems that use reserved seats (Afghanistan, Jordan), those that require diversity among candidates (Nepal), and systems that use both types of policies (Bolivia). The diversity of these few nested quota arrangements show that nested quotas could be implemented in a much broader range of countries than we see them today. Nested quotas are an innovative solution that could ensure that quotas do not just promote the representation of marginalized groups, but marginalized subgroups as well.
However, it is also clear that nested quotas are not a magic fix for the complex challenges at the intersection of multiculturalism and the ‘participatory turn’ in gender equality. Details of some of these nested quota cases suggest that—like gender and ethnic quotas more broadly—nested quotas may ultimately help those in the majority maintain their power. Specifically, adding seats for women from an ethnic group loyal to the ruling regime may satisfy international demands for women’s participation while doing little to promote genuine power-sharing across groups. Guaranteeing ethnic minority women a handful of seats in the national legislature may amount to little more than a symbolic gesture.
Overall, quotas are an imperfect solution to enhancing the diversity of legislatures, often failing to change the underlying structures that create inequalities among groups. At the same time, quotas have also led to real and significant transformations in some countries, literally changing the faces of political power. We also know that the impact of quotas is shaped by how they are designed; not all quotas are created equal. I suggest these insights about quotas overall will likely apply to nested quotas. We should not assume every nested quota will have a transformative effect, but they do provide a solution to the pervasive exclusion of ethnic minority women from political decision-making, and could have a substantial impact on the lives of ethnic minority women in some contexts. In practice, it will be important to attend to the details of how nested quotas function, to look to the mechanisms of how candidates and legislators are selected. Quotas that encourage debate and discussion within marginalized groups to select women representatives will be those that best address the concerns of multicultural feminism.
Abou-Zeid, Gihan, 2003. ‘Introducing Quotas in Africa: Discourse in Egypt’. Paper presented at the meeting of the International Institute for Democracy and Electoral Assistance, 11–12 November, Pretoria, South Africa.
Afzal, Nabeela, 1999. Women and Parliament in Pakistan, 1947–1977, Lahore: Pakistan Study Centre.
Angosto Ferrández, Luis Fernando, 2011. ‘La competencia por la representación indígena en las elecciones venezolanas (2004–2010)’. Cuestiones Políticas 46: 13–54.
Bird, Karen, 2003. ‘Who Are the Women? Where Are the Women? And What Difference Can They Make? Effects of Gender Parity in French Municipal Elections’. French Politics 1: 5–38.
Bird, Karen, 2014. ‘Ethnic Quotas and Ethnic Representation Worldwide’. International Political Science Review 35(1): 12–26.
Bird, Karen, 2016. ‘Intersections of Exclusion: The Institutional Dynamics of Combined Gender and Ethnic Quota Systems’. Politics, Groups, and Identities 4(2): 284–306.
Bjarnegård, Elin and Pär Zetterberg, 2014. ‘Why Are Representational Guarantees Adopted for Women and Minorities? Comparing Constituency Formation and Electoral Quota Design within Countries’. Representation 50(3): 307–20.
Bochsler, Daniel, 2010. ‘Electoral Rules and the Representation of Ethnic Minorities in Post-Communist Democracies’. European Yearbook of Minority Issues 2007/8 7: 153–80.
Bruhn, Kathleen, 2003. ‘Whores and Lesbians: Political Activism, Party Strategies, and Gender Quotas in Mexico’. Electoral Studies 22: 101–19.
Bush, Sarah Sunn, 2011. ‘International Politics and the Spread of Quotas for Women in Legislatures’. International Organization 65: 103–37.
Carey, John, 2015. ‘Here’s How the Opposition Got a Two-Thirds Supermajority in Venezuela’. The Washington Post. Retrieved 4 August 2016. <https://www.washingtonpost.com/news/monkey-cage/wp/2015/12/13/heres-how-the-opposition-got-a-two-thirds-supermajority-in-venezuela/>.
Chou, Bih-Er, and Janet Clark, 1994. ‘Electoral Systems and Women’s Representation in Taiwan: The Impact of the Reserved-Seat System’. In Electoral Systems in Comparative Perspective: Their Impact on Women and Minorities, edited by Wilma Rule and J. Zimmerman, 161–70. Westport, CT: Greenwood.
Chowdhury, Najma, 2002. ‘The Implementation of Quotas: Bangladesh Experience—Dependence and Marginality in Politics’. Paper presented at the regional workshop of the International Institute for Democracy and Electoral Assistance, 11–12 November, Jakarta, Indonesia.
COA (Council of the Americas), 2015. ‘Explainer: All About Venezuela’s Upcoming Elections and New Gender Quota’. Retrieved 4 August 2016. <http://www.as-coa.org/articles/explainer-all-about-venezuelas-upcoming-elections-and-new-gender-quota>.
Dahlerup, Drude (ed.), 2006. Women, Quotas and Politics, London: Routledge.
Dahlerup, Drude, Zeina Hilal, Nana Kalandadze, and Rumbidzai Kandawasvika-Nhundu, 2014. ‘Atlas of Gender Quotas’. International Institute for Democracy and Electoral Assistance, Inter-Parliamentary Union, and Stockholm University. Retrieved 17 December 2014. <http://www.idea.int/publications/atlas-of-electoral-gender-quotas/>.
Dubochet, Lucy, 2014. ‘Why India Needs the Women’s Reservation Bill’. Oxfam India.
Edwards, Louise, 1999. ‘Feminist Campaigns for Quotas for Women in Politics, 1936–1947.’ Twentieth-Century China 24(2): 69–105.
Foschini, Fabrizio, 2013. ‘The Social Wandering of the Afghani Kuchis: Changing Patterns, Perceptions, and Politics of an Afghan Community’. Afghanistan Analysts Network.
Htun, Mala, 2004. ‘Is Gender Like Ethnicity? The Political Representation of Identity Groups’. Perspectives on Politics 2(3): 439–58.
Htun, Mala, 2016. Inclusion Without Representation: Gender Quotas and Ethnic Reservations in Latin America, Cambridge: Cambridge University Press.
Htun, Mala and Mark P. Jones, 2002. ‘Engendering the Right to Participate in Decision-making: Electoral Quotas and Women’s Leadership in Latin America’. In Gender and the Politics of Rights and Democracy in Latin America, edited by Nikki Craske and Maxine Molyneux, 32–56. London: Palgrave.
Htun, Mala, and Juan Pablo Ossa, 2013. ‘Political Inclusion of Marginalized Groups: Indigenous Reservations and Gender Parity in Bolivia’. Politics, Groups, and Identities 1(1): 4–25.
Hughes, Melanie M., 2011. ‘Intersectionality, Quotas, and Minority Women’s Political Representation Worldwide’. American Political Science Review 105(3): 604–20.
Hughes, Melanie M., 2014a. ‘Crossing Intersections: Overcoming the Challenges of Cross-National Research on the Legislative Representation of Women from Marginalized Groups’. In Political Inequality in the Age of Democracy: Cross-National Perspectives, edited by Joshua Dubrow. London: Routledge.
Hughes, Melanie M., 2014b. ‘Quota Design and the Political Representation of Women from Indigenous and Tribal Groups’. Robert Schuman Centre for Advanced Studies Research. Paper No. RSCAS 2014/105.
Hughes, Melanie M., 2015. ‘Single-Axis Politics: Explaining the Persistent Political Overrepresentation of Men from Majority Racial and Ethnic Groups’. ECPR Joint Sessions of Workshops. Warsaw, Poland, 29 March–2 April.
Hughes, Melanie M., 2016. ‘Double-Duty Politics: How Electing and Appointing Minority Women Can Keep Majority Men in Power’. International Political Science Association. Poznan, Poland, 23–8 July.
Hughes, Melanie M., Mona Lena Krook, and Pamela Paxton. 2015. ‘Transnational Women’s Activism and the Global Diffusion of Gender Quotas’. International Studies Quarterly 59(2): 357–72.
Hughes, Melanie M., Pamela Paxton, and Mona Lena Krook. 2017. ‘Gender Quotas for Legislatures and Corporate Boards’. Annual Review of Sociology 43: 331–52.
Hughes, Melanie M., Pamela Paxton, Amanda Clayton, and Pär Zetterberg. Forthcoming. ‘Global Gender Quota Adoption, Implementation, and Reform’. Comparative Politics.
Hughes, Melanie M. and Aili Mari Tripp. 2015. ‘Civil War and Trajectories of Change in Women’s Political Representation in Africa, 1985–2010’. Social Forces 93(4): 1513–40.
International IDEA (Institute of Democracy and Electoral Assistance) and UCAB (Universidad Católica Andrés Bello). 2015. ‘IDEA-UCAB Study Mission’s Preliminary (p.117) Report on Conditions for the 2015 Electoral Process.’ Retrieved 4 August 2016. <http://www.idea.int/americas/upload/Preliminary-report-UCAB-IDEA.pdf>.
Interim Constitution of Nepal. 2007. Retrieved 7 August 2016. <http://www.consulat-nepal.org/IMG/pdf/NEPAL_CONSTITUTION_2007.pdf>.
IPU (Inter-Parliamentary Union). 2016a. Parline Database. Last accessed 17 February 2016. <http://www.ipu.org/parline-e/parlinesearch.asp>.
IPU (Inter-Parliamentary Union). 2016b. ‘Women in National Parliaments: Situation as of December 1, 2015’. Last accessed 17 February 2016. <http://www.ipu.org/parline-e/parlinesearch.asp>.
Jones, Mark, 1998. ‘Gender Quotas, Electoral Laws, and the Election of Women: Lessons From the Argentine Provinces’. Comparative Political Studies 31(1): 3–21.
Kefale, Asnake, 2013. Federalism and Ethnic Conflict in Ethiopia: A Comparative Regional Study, London: Routledge.
Khel, Muḥammad Nazeer Kaka, 1984. ‘Status of Non-Muslim Minorities in Pakistan’. Islamic Studies 23(1): 45–54.
Krook, Mona Lena, 2009. Quotas for Women in Politics: Gender and Candidate Selection Reform Worldwide, New York: Oxford University Press.
Krook, Mona Lena, and Diana O’Brien, 2010. ‘The Politics of Group Representation: Quotas for Women and Minorities Worldwide’. Comparative Politics 42(3): 253–72.
Krook, Mona Lena, Diana O’Brien, and Krista Swip, 2010. ‘Military Invasion and Women’s Political Representation’. International Feminist Journal of Politics 12(1): 66–79.
Lijphart, Arend, 1994. ‘Proportionality by Non-PR Methods: Ethnic Representation in Belgium, Cyprus, Lebanon, New Zealand, West Germany and Zimbabwe’. In Electoral Laws and their Political Consequences, edited by Bernard Grofman and Arend Lijphart, 113–23. New York: Agathon Press.
Lublin, David, 2014. Minority Rules: Electoral Systems, Decentralization, and Ethnoregional Parties, Oxford: Oxford University Press.
Lublin, David and Matthew Wright, 2013. ‘Engineering Inclusion: Assessing the Effects of Pro-Minority Representation Policies’. Electoral Studies 32(4): 746–55.
Menon, Mivedita, 2000. ‘Elusive “Woman”: Feminism and Women’s Reservation Bill’. Economic and Political Weekly 35(43/44): 3835–9.
Muriaas, Ragnhild, Liv Tønnessen, and Vibeke Wang, 2013. ‘Exploring the Relationship between Democratization and Quota Policies in Africa’. Women’s Studies International Forum 41(2): 89–93.
Nanivadekar, Medha, 2005. ‘Feminist Fundamentalism over Women’s Reservation Bill: Lessons from the Quota Debate in India’. Paper presented at IWPR International Women’s Policy Research Conference. Washington, DC, 19–21 June.
New Zealand Parliamentary Library, 2009 . ‘The Origins of the Māori Seats’. Parliamentary Library Research Paper. Retrieved 14 March 2015. <http://www.parliament.nz/>.
Paxton, Pamela, and Melanie Hughes, 2016. Women, Politics, and Power: A Global Perspective, 3rd edition, Thousand Oaks: CQ Press.
Piscopo, Jennifer, 2015. ‘State and Gender Equality Activists: The Evolution of Quota Laws in Latin America’. Latin American Politics and Society 57(3): 37–49.
Rais, Rasul Bakhsh, 2004. ‘Islamic Radicalism and Minorities in Pakistan’. In Religious Radicalism and Security in South Asia, edited by Satu Limaye, Mohan Malik, and Robert Wirsing, 447–64. Honolulu: Asia-Pacific Center for Security Studies.
Randall, Vicky, 2006. ‘Legislative Gender Quotas and Indian Exceptionalism: The Travails of the Women’s Reservation Bill’. Comparative Politics 39(1): 63–82.
Reilly, Benjamin, 2001. Democracy in Divided Societies: Electoral Engineering for Conflict Management, Cambridge: Cambridge University Press.
Reilly, Benjamin, 2006. Democracy and Diversity: Political Engineering in the Asia-Pacific, Oxford: Oxford University Press.
Reynolds, Andrew, 2005. ‘Reserved Seats in National Legislatures: A Research Note’. Legislative Studies Quarterly 30(2): 301–10.
Schmidt, Gregory, and Kyle Saunders, 2004. ‘Effective Quotas, Relative Party Magnitude and the Success of Female Candidates: Peruvian Municipal Elections in Comparative Perspective’. Comparative Political Studies 37: 704–34.
Schneckener, Ulrich, 2002. ‘Making Power-Sharing Work: Lessons from Successes and Failures in Ethnic Conflict Regulation’. Journal of Peace Research 39(2): 203–28.
Schwindt-Bayer, Leslie, 2010. Political Power and Women’s Representation in Latin America, Oxford: Oxford University Press.
Swiss, Liam, and Kathleen Fallon, 2016. ‘Women’s Transnational Activism, Norm Cascades, and Quota Adoption in the Developing World’. Politics & Gender 13(3): 458–87.
Tajali, Mona, 2013. ‘Gender Quota Adoption in Postconflict Contexts: An Analysis of Actors and Factors Involved’. Journal of Women, Politics & Policy 34(3): 261–85.
Towns, Ann, 2010. Women and States: Norms and Hierarchies in International Society, Cambridge: Cambridge University Press.
Tripp, Aili Mari, 2015. Women and Power in Post-Conflict Africa, Cambridge: Cambridge University Press.
Tripp, Aili Mari, Dior Konaté, and Colleen Lowe-Morna, 2006. ‘Electoral Gender Quotas—Sub-Saharan Africa’. In Women, Quotas and Politics, edited by Drude Dahlerup, 112–37. London: Routledge.
Tuttle, Gray, 2005. Tibetan Buddhists in the Making of Modern China, New York: Columbia University Press.
United Nations. 2011. ‘Djibouti, Reporting for First Time to Women’s Anti-Discrimination Committee, Boasts Progress in Health, Education, Women’s Participation in Politics’. Retrieved 3 August 2016. <http://www.un.org/press/en/2011/wom1874.doc.htm>.
USDS (US Department of State). 1997. ‘Background Notes: Kiribati, May 1996’. US Department of State Archive. Retrieved 14 March 2015 <http://1997–2001.state.gov/www/background_notes/kiribati_0596_bgn.html>.
USDS (US Department of State). 2011. ‘Iraq’. 2010 Country Reports on Human Rights Practices, 8 April 2011. <https://www.state.gov/j/drl/rls/hrrpt/2010/nea/154462.htm> accessed 12 April 2018.
Wagner, Sarah, 2005. ‘Women and Venezuela’s Bolivarian Revolution’. Venezuelanalysis.com, 15 January. Retrieved 4 August 2016. <https://venezuelanalysis.com/analysis/877>.
(1) I focus in this chapter on national gender quotas that apply to all political parties in a system. Many countries also use party gender quotas, where one or more political parties have adopted voluntary measures to facilitate women’s inclusion.
(2) Studies use different terminology and criteria for ethnic quotas (Bjarnegård and Zetterberg 2014; Bochsler 2010; Bird 2014, 2016; Htun 2016; Hughes 2011; Lublin 2014; Protsyk 2010; Reynolds 2005). I focus here on laws and rules that guarantee representation for one or more marginalized ethnic groups in the single or lower house of the national legislature. Therefore, I exclude policies that create constituencies in islands and other minority-dominated areas (e.g. Denmark), that draw district boundaries to enhance the representation of certain groups (e.g. United States), or that relax electoral thresholds for ethnic parties or organizations (e.g. Germany), unless representation for a targeted group is guaranteed. I do not count separate legislatures for targeted groups (e.g. the Sametinget in Norway), quotas for state or local legislatures (e.g. Peru), or quotas for upper houses (e.g. Bosnia and Herzegovina). I also exclude policies that guarantee representation for dominant ethnic groups (e.g. Zimbabwe ) or for other types of identities such as youth, persons with disabilities, workers, and nationals living abroad (e.g. Uganda).
(3) Estimates and figures based on ethnic quota data are based on best available information. Presently, there is no publically available data source that tracks ethnic quotas, and published sources sometimes disagree. For example, Reynolds (2005) and Htun (2016) identify quotas only in Ethiopia’s upper house, but Kefale (2013) reports that the constitution reserves twenty seats for ‘minority nationalities and peoples’ in the House of Peoples’ Representatives, the lower chamber.
(4) Even when the British government did not directly influence the passing of ethnic quotas, colonialism may still have played some role in adoption. Consider the case of Kiribati. After World War II, British colonial authorities moved most of the population of one of Kiribati’s islands, Banaba, to Fiji, where most Banabans continue to reside today (Lublin 2014). Since independence in 1979, Kiribati has fought against Banaban Island’s cessation to Fiji, in part, by reserving a seat in its National Assembly for a Banaban (USDS 1997).
(5) Internationally, many scholars see a turning point in 1995 around the Fourth World Conference for Women in Beijing, which helped to spread the idea that representation in political decision-making was women’s right (Htun and Jones 2002; Swiss and Fallon 2016).
(6) The same time that China guaranteed seats for women, seats were also set aside for overseas Chinese, occupational groups, Mongolians, and Tibetans (Edwards 1999). However, seats for the Mongolians and Tibetans were based on geography (not ethnicity) and were typically filled by ethnic Chinese (Tuttle 2005).
(8) Venezuela’s gender quota has only been partially implemented. In 2015, Venezuela’s national elections board (Consejo Nacional Electoral, CNE) instituted parity through a gender balance quota, ruling that parties must nominate at least 40 per cent women (IPU 2016b). However, the decision came a month after the opposition, the United Democratic Roundtable (MUD), held its primaries for the upcoming elections (COA 2015). MUD fought against the parity provision, and the CNE ultimately did not require full implementation of the law (International IDEA and UCAB 2015). In the December elections, MUD won a super-majority of seats in the legislature, in part because they secured all three seats reserved for indigenous peoples (Carey 2015). With MUD now holding the reigns, the future of the parity law is uncertain.
(9) Notably, I do not count the Philippines, where the 1987 constitution required that: ‘For three consecutive terms after the ratification of this Constitution, one-half of the seats allocated to party-list representatives shall be filled, as provided by law, by selection or election from the labor, peasant, urban poor, indigenous cultural communities, women, youth, and such other sectors as may be provided by law, except the religious sector’ (Article VI, Section 5(2)). Although referencing both ‘women’ and ‘indigenous cultural communities’, there are no set minimums for these groups. Legally, all seats could be filled by any sectoral interests, including peasants, urban poor, youth, or particular groups of workers such as fishermen or farmers (Reilly 2006).
(10) Iraq also reformed its gender quota in 2009. The initial law required a minimum of 25 per cent women by regulating the order of men and women candidates on party lists. The new law enhanced these placement mandates and introduced a procedure for governates to follow if women did not win at least 25 per cent of seats through regular channels.
(11) International influences were also important in Djibouti’s adoption of a gender quota. Through much of Djibouti’s history, women’s political participation was almost non-existent. It was not until 1986 that women could even stand for election, and as the twentieth century drew to a close, no woman had ever been elected to the national legislature. In 1999, Djibouti ratified of the Convention on the Elimination of All Forms of Discrimination against Women (CEDAW) (United Nations 2011). CEDAW ratification set the stage for a series of quota reforms, including the adoption of a 10 per cent gender quota in 2002.
(12) In addition to the ethnic quotas for Tutsis and Hutus, Burundi also sets aside three seats for Twas, or pygmies. These seats are filled differently (after the elections through co-optation).
(13) Identifying the Tutsis as a ‘minority group’ is questionable (Hughes 2014a). Tutsis numerical minority in Burundi, but they may not be a ‘minority’ in the sense of marginalization. Tutsis held economic, political, and military power in the decades after Burundi’s independence. Since democratization in 1993, however, political power has rested squarely in Hutu hands.
(14) This chapter focuses on gender and ethnicity, but nesting is used for other identities as well. For example, in 2010, Egypt required some share of women’s reserved seats to be labourers/workers.
(15) By my calculations, across the last three elections (2005, 2010, and 2015), the gender quota of 30 per cent was only met consistently for Tutsis. In 2010, all three seats reserved for Twa went to men, and in all three elections, women never reached 30 per cent of Hutu seats.
(16) My counts of quotas differ from Bird’s (2016) for other reasons as well. First, we look at different countries. Because I include only countries recognized as independent by the United Nations, I exclude three of Bird’s (2016) cases: Kosovo, the Palestinian Territories, and Taiwan. Second, I operationalize ethnic and gender quotas more narrowly. Because I do not count threshold relaxations for ethnic parties as ethnic quotas, I do not code Montenegro, Poland, or Serbia as having ethnic quotas. Because I do not count gender regulations for primaries and internal elections as gender quotas (see Hughes et al. 2015), I do not code Panama as having a gender quota. Overall, then, of the seventeen countries Bird (2016) identifies as having both gender and ethnic quotas, we only agree on ten cases: Afghanistan, Bolivia, Burundi, Colombia, Djibouti, Iraq, Jordan, Niger, Pakistan, and Slovenia. I add to this list Nepal (2007), Samoa (2013), and Venezuela (2015).
(17) These laws were used to elect a Constituent Assembly (CA) in 2008. However, the CA failed to agree on the terms of a new constitution. In 2013, elections were held for a second CA, and this time the quota laws were largely ignored. They were not repealed, however, so I still treat Nepal as having a nested quota in 2015. Notably, the only quota specified in the 2015 constitution is a 33% minimum for women in the Federal Parliament. The constitution called for an ethnic quota but did not establish one. Ultimately, the 2017 election law adopted nested quotas for candidate lists similar to those used to elect the 2008 CA.
(18) Percentages add to more than 100 per cent because these are not mutually exclusive groups. For example, a Dalit from a backward region could satisfy two requirements of the quota while occupying only a single seat. Nepal’s nested quota only applies to parties whose candidate lists include at least 30 per cent of the total number of candidates elected under the list PR system. In practice, in the 2008 elections for the Constituent Assembly, this was parties with one hundred candidates or more.
(20) Women’s groups have proposed amendments to the Bill, for example, to increase the size of the legislature and for all districts to become dual-member constituencies with one man and one woman representative. Again, such a plan would require women’s election from among the SCST reserved seats and would thus constitute a nested quota.