Democracy and ethnic inequality: a comparative case study

ABSTRACT

Recent studies have found that democratization and reductions of socioeconomic ethnic inequality generally show a positive association. Yet, the causal mechanisms behind the co-variation are not well understood. Moreover, the identified relationship covers considerable heterogeneity, with some transitions to democracy being followed by significant distributive change and others not. To shed additional light on the relationship, this article examines the political and social dynamics related to five democratic transitions. Three led to reduced socioeconomic inequality across ethnic groups, two did not. The “positive” cases (Bolivia, South Africa, and Nepal after 2006) reveal that political empowerment of socioeconomically disadvantaged groups and consequent changes in policies constitute important drivers of reduced ethnic inequalities. In contrast, the “negative” cases (Guatemala and Nepal 1991–2002) demonstrate that democracy only leads to reduced ethnic inequality if democratization is followed by effective mobilization by the previously excluded group(s), and this step might be obstructed by traditional elites.

Keywords:

• democracy
• ethnic inequality
• horizontal inequality
• mobilization
• inclusion
• egalitarian policies

Does democracy ensure greater social and economic equality for its citizens than autocracy? Theoretically, democratization provides opportunities for previously excluded and socioeconomically disadvantaged citizens to mobilize in favour of policies that enhance their relative well-being.¹ Despite strong expectations, studies have found only limited evidence that democracy has a consistent, negative impact on inequality.² This has led scholars to explore different conditioning factors that might influence the relationship, including social cleavages, elite capture, and the nature of the pre-democratic regime.³

Moreover, scholars have begun to explore different dimensions of disparity, focusing on socioeconomic inequalities between ethnic groups rather than between individuals or social classes. On the one hand, ethnic inequalities can spark collective grievances and mobilization under democracy to address perceived injustices. On the other hand, historically dominant groups may strongly resist redistribution favouring disadvantaged ethnic (out)groups.⁴

Recent global studies show that democratic transitions are indeed associated with reduced socioeconomic ethnic inequality.⁵ To understand the relationship, however, we should investigate not only the average statistical association but also whether it works through the stipulated causal mechanisms. There are currently no qualitative comparative assessments of whether the proposed dynamics underlie the identified empirical regularities. A related challenge is that the statistical findings cover a great deal of heterogeneity at the case level: Whereas democratic transitions in highly unequal countries have clearly been followed by increased redistribution between groups in some countries (e.g. South Africa), others have seen much less pronounced changes (e.g. Guatemala). In all these cases, previously excluded groups have a clear collective interest in more redistribution, yet outcomes differ significantly. It is far from clear why democratic transitions sometimes set in motion successful pushes for more equality and sometimes do not.

To study how and under what conditions democracy reduces ethnic inequality, I proceed in two steps. First, I develop a causal chain that I investigate in three “positive” cases: the democratic spells of Bolivia (1982–), South Africa (1994–), and Nepal’s second democratic spell (2006–).⁶ This analysis reveals that democratic transitions in highly unequal Bolivia, South Africa, and Nepal were followed by increased political influence of socioeconomically disadvantaged groups. Although the speed and comprehensiveness of the political inclusion process varied, it has led to important reductions in ethnic inequalities through a range of policies, including universal social policies, targeted (“affirmative action”) policies, and initiatives aimed at reducing discrimination.

Second, I consider two “negative” cases: Nepal’s first democratic spell (1991–2002) and Guatemala (1986–). In these cases, democratization was not followed by a strong and coherent political mobilization by previously excluded groups, and traditional elites were able to retain significant political influence. The absence of a strong and coherent mobilization was itself a product of high fractionalization within previously excluded groups. These disconfirming cases allow me to refine the theory and adjust the scope conditions of the causal mechanism⁷ emphasizing the necessity of solving collective action problems among previously excluded group(s).

This study has a number of implications for the literature on democracy and inequality. By specifying and assessing mechanisms and scope conditions, it contributes to our understanding of when and how democracy reduces ethnic inequality. It highlights that democratization does not automatically lead to successful egalitarian pushes as collective action problems may squander movements towards more equality. More broadly, the findings contribute to the extensive body of research on the effects of democracy⁸ and support the idea that democracy, at least under certain conditions, promotes political dynamics that benefit less privileged groups.

Defining democracy and ethnic inequality

I understand “democracy” as a political regime in which leaders are chosen through competitive, regular, and inclusive elections.⁹ This electoral conception is inspired by Przeworski and Schumpeter but adds inclusive suffrage as a criterion. It is suitable for my purposes because it focuses on what is distinct about democracy and avoids conceptual overlaps with ethnic inequality and potential mechanisms.¹⁰

I understand “ethnic inequality” as socioeconomic disparities between ethnic groups. Such group differences in standards of living are shaped by variations in labour income, wealth, as well as access to essential public goods such as health and education.¹¹ Ethnicity is broadly defined as an experienced sense of commonality rooted in a belief in common descent, which can be indicated through various identity markers including language, phenotype, religion, tribe, or caste.¹² The existence of ethnic inequality assumes that there are clearly defined groups with relatively durable boundaries over time.¹³ Ethnic inequalities are less pronounced in some multiethnic countries, e.g. Iran and Switzerland, and more pronounced in, e.g. Guatemala, Peru, Bolivia, Côte d’Ivoire, South Africa, Nigeria, Kenya, Myanmar, and the United States.¹⁴

A closer look at the cases

Prior research indicates that democratization empowers formerly excluded and socioeconomically disadvantaged groups to gain political influence and leverage it to advance policies that enhance their group’s relative well-being.¹⁵ Furthermore, this work argues that democratic transitions should have the strongest effect in the initially most unequal countries.¹⁶ The basic idea is that in new democracies with substantial inequalities, marginalized groups have deeper grievances, which they have an opportunity to address. Therefore, the most significant reductions in ethnic inequality should be observed in countries that exhibited pronounced inequality prior to democratization.

Table A1 in the Online Appendix shows a list of countries that have democratized at high levels of ethnic inequality and their subsequent reductions in ethnic inequality. Even within this subsample of countries where we should see the strongest effect of democratization, the dynamics clearly differ and range from large reductions (e.g. Namibia, South Africa, and Bolivia) to virtually no changes (e.g. Bangladesh, Myanmar, and Sudan). This raises important questions, including how democracy enables reductions in socioeconomic disparities between ethnic groups, and why it only does so in some cases.

To examine this puzzle, I begin by examining three positive cases – Bolivia (1982–), South Africa (1994–), and Nepal (2006–) – in which democratization in a high-inequality context was followed by substantial reductions in ethnic disparity. I thus look at cases where the causal mechanism should be present.¹⁷ While my focus is on the within-case analysis of each case, the positive cases also conform to a Most Different System Design logic which allows me to consider similar causal relationships across a variety of cases (see Table 1 below). This enables me to assess the generalizability of the mechanisms, and expand or limit the scope conditions in which I expect the causal relationship to occur.¹⁸ To answer the second part of the puzzle – why democracy only enables inequality reductions in some cases – I explore two negative cases in which democratization was not followed by significant reductions in ethnic inequality: Guatemala (1986–) and Nepal’s first democratic spell (1991–2002).

Table 1. Case overview.

	Bolivia	South Africa	Nepal I and II	Guatemala
Year of transition	1982	1994	1991 and 2006	1986
Level of ethnic inequality at transition	High	High	High and High	High
Reduction of ethnic inequality*	High	High	Low and Medium	Low
Major groups’ population share** (Previously excluded in bold)	Indigenous (∼60%) White/Mestizo (∼40%)	Black (76.4%) White (9.1%) Colored (8.9%) Indians (5.6%)	Indigenous (∼36%) Caste hill Hindu elite (∼31%) Madhesis (17%) Dalits (∼15%)	Ladino (∼60%) Indigenous (∼40%)
Autocratic interlude	2019–2020	–	2002–2006	–
Level of development in year of transition***	USD 3816	USD 6173	USD 1361 and USD 1884	USD 5136
Political history: colonial status and previous regime****	Settler colony Military dictatorship	Settler colony Ethnic oligarchy	Never colonized Hindu monarchy	Settler colony Military dictatorship
Politico-geographic region	Latin America	Sub-Saharan Africa	South Asia	Latin America

Notes: *Based on Online Appendix B (cross-national data) and country-specific sources presented in the case studies. If cross-national data and country-specific sources are not in agreement regarding the magnitude of the inequality reduction, country-specific sources have been given more weight. This was the case for Nepal. **Important nuances to these group categories are discussed for each given case below. Population data are contemporary yet reflect the patterns at the time of transition. ***GDP per capita in USD 2011 ****This refers to the regime just prior to the democratic transition. Historically, Bolivia and Guatemala are characterized by more regime instability than the other cases, with prior spells as exclusive democracies and multiparty autocracies.

Mechanisms and empirical manifestations

When and how should we expect democratization to be followed by reductions in ethnic inequality? Using an existing theoretical argument¹⁹ as point of departure, the following section outlines a causal chain, which is summarized in Table 2. Transitions to democracy allow previously excluded and socioeconomically disadvantaged groups to participate in the political process (Link 1) and to gain influence through political representatives (Link 2). Formerly excluded groups can secure representation through both ethnic and non-ethnic parties, and these frequently align with leftist agendas.²⁰ Even if the parties representing marginalized groups do not secure positions in government, they can still promote inclusion by exerting pressure on established parties to include representatives from marginalized groups.²¹

Table 2. Theorized mechanism and empirical manifestations (successful reduction in inequality).

	Cause	Link 1	Link 2	Link 3	Outcome
Theorized mechanism	Democratic transition	Democratic transitions allow previously excluded groups to participate politically	Disadvantaged groups gain political influence directly or indirectly	Egalitarian policies are implemented	Reduced levels of socioeconomic inequality between ethnic groups
Empirical manifestations	Change in political regime type as indicated by foundational free elections	– Formation of new parties – (Increased) voter participation among previously excluded groups – Greater representation of previously excluded groups in existing parties	Parties representing disadvantaged groups (1) secure government or parliamentary seats or (2) promote inclusion by pressuring dominant parties to become more inclusive	Implementation of policies: universal or targeted policies; discriminatory policies are curbed	Reductions in measures of ethnic inequality attributable to policy changes

Scope condition: Significant socioeconomic disparities between relatively clearly defined ethnic groups.

Political parties advocating for marginalized communities can attain government or parliamentary positions, enabling them to shape legislation and oversee its implementation (Link 3). Various types of policies can help to mitigate ethnic inequality.²² First, universal policies distribute benefits according to universal categories, and they include income transfers as well as public services, such as health and housing, that disproportionately benefit disadvantaged citizens. Second, targeted policies offer advantages to specific groups and include investments in disadvantaged regions, educational scholarships, legal aid, group distribution requirements, land reform, and asset redistribution.²³ Third, historically marginalized groups can use their increased political influence to combat ethnically based discrimination, for instance, in hiring procedures for state jobs or in the labour market. Overall, this leads to more equal economic opportunities, which help to mitigate group disparities.

Each step of the outlined causal chain in Table 2 is a necessary condition, and to the extent that the full chain plays out, democracy should reduce ethnic inequality. However, democratic transitions do not necessarily set in motion the proposed causal chain. Rather, democracy can be considered a permissive condition that provides opportunities for previously excluded groups to mobilize in favour of more egalitarian policies. If, for instance, excluded groups are too fractured to exploit opportunities under democracy (Link 1 absent), the causal chain breaks, and democracy should lead to limited or no distributive change.

Table 2 also presents the empirically observable implications of each step, i.e. what kind of evidence would be in line with the proposed mechanism. To analyse the cases, I mainly rely on case-specific literature and data but also include cross-national data. To the extent possible, I have triangulated different sources to detect potential disagreement about specific developments. Temporally, I study the mechanism from the time of democratization and 20–30 years after these events, based on the idea that democracy’s effects are not immediate and it takes time to mobilize voters and for policies to take effect.²⁴

Bolivia: indigenous mobilization and egalitarian policies

Following the resignation of long-time dictator Hugo Banzer in 1978, Bolivia experienced a turbulent period of elections, labour conflicts, and military coups. In 1982, General Garcia Meza was ousted in a coup, and the coup leaders reinstated the legislature and named Hernan Silez Suazo to the presidency. This choice was later ratified through competitive elections. Throughout the period between 1978 and 1982, large, militant, and well-organized unions together with peasant movements were decisive in defeating repeated military attempts to reassert control.²⁵

Bolivians who self-identify as Indigenous make up around 60 percent of the population and, as a group, fare consistently worse on a range of socioeconomic parameters, such as education, poverty, and occupation, than people who identify as White or Mestizo.²⁶ Bolivia was long dominated by class-based politics although the class cleavage had acted as a proxy for identity politics as lower-class citizens were much more likely to be Indigenous.²⁷

Political mobilization of previously excluded groups

Indigenous movements and parties did not emerge in Bolivia until the transition to democracy began in the late 1970s.²⁸ Since democratization in 1982, social class and ethnicity have become significant political issues.²⁹ By the early 2000s, two parties, the Movimiento al Socialismo (MAS) and the Movimiento Indígena Pachakutik (MIP), campaigned on Indigenous issues and explicitly aimed to represent the Indigenous population in the presidential elections.³⁰

The most successful Indigenous party was the MAS, which combined leftist ideology with a call for Indigenous rights.³¹ Its original platform emphasized a combination of leftist economic policies and reforms aimed to increase political equality, including proposals to reform state institutions and promote decentralization.³² Much of the party’s discourse centred on redistributing income and addressing the needs of marginalized sectors of the population.³³ Indeed, Evo Morales, the leader of the MAS who later served as Bolivia’s first Indigenous president from 2006 to 2019, identified inequality as a fundamental reason for the party’s formation.³⁴

Disadvantaged groups gain political influence

Originating from labour unions and Indigenous interest groups, the MAS’ success can partly be attributed to its strong foundations in social movements and its inclusive ethnic and populist appeals. The fluid nature of ethnic boundaries and limited ethnic polarization earned the MAS support not only from those who identify as Indigenous but also from a significant portion of voters who generally identify as White or Mestizo.³⁵ With the exception of a political crisis in 2019, which saw an opposition candidate as president of the interim government, MAS candidates have held the presidency since 2005. In response to the MAS’ electoral achievements, several non-ethnic political parties have recruited a greater number of Indigenous candidates for political positions and integrated specific Indigenous demands in their agendas.³⁶ This trend of ethno-political inclusion is also visible in Online Appendix Figure A1, which builds on two different cross-national datasets.

Implementation of egalitarian policies

Has the emergence and rise of Indigenous parties following democratization led to policy changes that reduced ethnic inequalities in Bolivia? At the political level, there is little doubt that representation of Indigenous peoples in national policymaking improved.³⁷ During the 13 years of Morales’ administration (2006–2019), the government enacted a wide range of policies and programmes, including direct cash transfers and broad redistributive policies.³⁸

Many of these policies did not specifically target Indigenous people but nevertheless benefitted them disproportionately.³⁹ One example is the administration’s literacy programmes, which taught more than 800,000 people to read between 2006 and 2008 – at least according to the MAS’ own estimates.⁴⁰ Other programmes that have disproportionately benefitted the Indigenous population are the Bono Juana Azurduy programme, which makes small payments to pregnant women and children under two years of age who see the doctor regularly, and the conditional cash transfer Bono Juancito Pinto, which pays families to keep their children in school.⁴¹ Indigenous people, who live disproportionately in rural areas, have also benefitted from the Morales administration’s agrarian reform programme, which redistributed almost a million hectares of land between 2006 and 2009.⁴²

The Morales administration also introduced a range of policies that specifically seek to promote Indigenous culture, eliminate discrimination, and improve the socioeconomic circumstances of Indigenous people.⁴³ For example, the government strengthened the law against job discrimination and sought to expand the teaching of Indigenous languages, culture, and history in the education system.

Moreover, a new constitution introduced in 2009 recognizes the country as “plurinational,” makes the Indigenous flag – the Wiphala – one of Bolivia’s national symbols, recognizes various Indigenous tongues as official languages, and requires that the central and departmental governments use at least two official languages.⁴⁴ The constitution also grants numerous rights to Indigenous people, including the right to collective land ownership and the right to benefit from the exploitation of natural resources in their territories.⁴⁵ Figure A2 supports the finding that there are fewer barriers to state jobs and contracts for Indigenous people after the MAS came to power in the mid-2000s.

Although it is hard to make a definitive assessment of the impact of the described policies, preliminary results are encouraging. Illiteracy, infant mortality, and extreme poverty have declined under the Morales administration, and the rural poor’s share of national income has significantly increased.⁴⁶ The income gap between Indigenous and non-Indigenous households was reduced by roughly 25 percent during Morales' time in office.⁴⁷ Moreover, Indigenous languages and practices have a greater official standing, and Indigenous people enjoy more rights and protection from discrimination. Overall, the policies pushed by the MAS and the smaller MIP have reduced ethnic inequalities.⁴⁸ Finally, it bears mentioning that ethnic inequality seems to have indirectly destabilized democracy in Bolivia, which was particularly visible during the political crisis of 2019–2020. Due to space constraints, I discuss this dynamic in Online Appendix C.

South Africa: adressing deep-rooted racial inequality

In 1948, the Afrikaner-dominated National Party imposed a policy of apartheid that entrenched and extended an already comprehensively racialized and segregated state.⁴⁹ Apartheid was a widespread system of racial segregation meant to preserve the privilege of the White minority population. In many ways, an individual’s socioeconomic position under apartheid therefore depended on their official racial classification.⁵⁰

Political mobilization of previously excluded groups

The African National Congress (ANC) and its allies led a highly organized struggle against apartheid, including demonstrations, strikes, and armed resistance, which was decisive in forcing the White elites to accept a negotiated transition.⁵¹ By the 1980s, South Africa’s racial policies had begun to encounter opposition from business elites and Western political movements that pressed for economic sanctions against the regime. In the view of Frederik Willem de Klerk, who served as South Africa’s president from 1989 to 1994, the only way to end South Africa’s isolation was through an understanding with the ANC, which meant holding inclusive elections.

Disadvantaged groups gain political influence

South Africa’s long history of racial division and apartheid formally ended in April 1994 when, for the first time, the entire country voted in national elections.⁵² Nelson Mandela became the first president of the post-apartheid era, and a government of national unity was formed with the participation of the Afrikaner-dominated National Party. The election ensured full Black participation in politics and representation at the highest levels of government.⁵³ The ANC received 62.5 percent of the vote, the National Party led by de Klerk 20.4 percent, and the Inkatha Freedom Party, mainly supported by ethnic Zulus, 10.5 percent.⁵⁴ Employing cross-national data, Figure A3 shows a very abrupt change towards more inclusion and better representation following the transition to democracy.

Since then, South Africa’s party system has largely followed racial divides, pitting the ruling ANC against the mainly White-based Democratic Alliance (DA).⁵⁵ After the National Party collapsed in 1997, most White voters (as well as Indians and Coloreds) moved to the DA, which became the formal parliamentary opposition. Now, the DA also receives increasing electoral support from middle-class Blacks.⁵⁶

Implementation of egalitarian policies

Democratization in 1994 brought to power a government with a clear public commitment to mitigating inequality.⁵⁷ Indeed, the centrepiece of the ANC’s 1994 election platform was the Reconstruction and Development Programme (RDP), a poverty-alleviation scheme that enjoyed broad support.⁵⁸ For Nelson Mandela and other key leaders, a central priority was to balance high expectations for socioeconomic improvement among Black South Africans with both fiscal constraints⁵⁹ and fears of political exclusion among the White minority.⁶⁰

The new South African constitution affirmed fundamental political, civil, and property rights and prohibited discrimination based on race and ethnicity. More ambitiously, it also declared a set of socioeconomic rights, such as housing, healthcare, and social security, “within its available resources.”⁶¹ How and to what extent did democratization and the ANC government lead to policy changes?

The apartheid state did introduce a welfare system, but it discriminated along racial lines, and especially poor White people benefited.⁶² The first post-apartheid government completed the process of removing racial discrimination from labour market and welfare policies.⁶³ Moreover, pro-poor policies were represented as ways to reduce racial inequality.⁶⁴

Since 1994, the ruling ANC has raised taxes, introduced welfare grants for pensioners and low-income children, and expanded public healthcare and education.⁶⁵ Post-apartheid policies thus improved redistribution to the poor via expansion of social services.⁶⁶ Moreover, the ANC government significantly improved Black housing and access to water and electricity.⁶⁷ These policies have narrowed the profound racial gaps in access to such services and raised living standards.⁶⁸ Today, South Africa has a high level of redistribution, which entails a progressive tax system, a system of public welfare provision, and pro-poor spending on public health and education.⁶⁹

Moreover, labour market policies have been advanced, including removal of labour market discrimination, extension of employment protection, skills development, strengthening of unions and centralized bargaining, as well as extension of minimum wage setting to unorganized sectors.⁷⁰ While wages for semi-skilled and skilled workers have increased, there are huge challenges with unemployment concentrated among rural African job market outsiders.⁷¹

The ANC government also pursued policies designed to open new economic opportunities for Black South Africans via policies of affirmative action, such as “Black Economic Empowerment” (BEE).⁷² The BEE, which came into effect in 2004, aims to redress inequalities of apartheid by allowing Black citizens to participate more fully in South Africa’s economy.⁷³ Figure A4 shows how access to state jobs and contracts for all ethnic groups has improved markedly since the end of apartheid.

How have these policies implemented by the democratically elected government affected racial inequality in South Africa? Despite improved basic public services and cash transfers, unemployment, poverty, and inequality remain high and heavily correlated with race.⁷⁴ At the same time, the share of total income received by Black South Africans has continued to rise, and the share received by White South Africans has continued to fall.⁷⁵ Average White incomes were seven times higher than average Black incomes in 2005–2009 but only four times higher in the period 2015–2019.⁷⁶ Underlying the declining interracial inequality is the growth of the Black middle and upper class, which was supported by the affirmative action policies.⁷⁷ In short, while racial inequalities remain high in absolute terms, there are clear indications that the democratic transition paved the way for a significant reduction in interracial disparities.

However, the gap between better-off and poorer Black South Africans continued to grow. This pattern helps to explain the absence of major reductions in income inequality between individuals despite reduced inequality between Blacks and Whites.⁷⁸

Nepal II: slow steps towards more equality

In Nepal, the most politically salient sociocultural groups are the “high-caste” Hindu elite from the hills, the “low-caste” Hindus (Dalit), Indigenous nationalities (or Adivasi Janajati), and the Madhesi (peoples from the Tarai region who share culture with North Indian societies).⁷⁹ The high-caste hill Hindu elite (CHHE) constitutes around a third of the population, yet they have overwhelmingly dominated politics and enjoy higher standards of living than other groups in Nepal.⁸⁰ The non-CHHE groups – Indigenous nationalities, Dalits, and Madhesis – combined constitute two thirds of the population but have historically been politically excluded and discriminated against.⁸¹

In April 2006, a “people’s movement” reclaimed democracy, and a peace accord with the Maoist rebels was signed in November 2006. After ten years of Maoist insurgency, an alliance of seven major political parties and the Maoists launched the movement to end four years of direct rule by the king.⁸² The new ruling coalition declared Nepal a secular state, created a new interim constitution, and held elections in 2008 for a constituent assembly tasked with drafting a new constitution. This assembly, which became the most representative legislature in Nepali history,⁸³ abolished the monarchy and declared Nepal a federal democratic republic.⁸⁴

Political mobilization of previously excluded groups

During the earlier period of democratic governance from 1990 to 2002, the substantial political control exerted by the CHHE group prevented major reforms (see below), a tendency that was reinforced by majoritarian principles that favoured the party/group in power.⁸⁵ In contrast, the post-2006 period saw major reforms that increased political inclusion, including a federal system and a proportional representation electoral system.⁸⁶ This regime change was enabled by the unified efforts of reform-oriented forces, which encompassed civil society organizations and political parties, including ethnic organizations and parties. These efforts were further strengthened by the Maoists, who pledged to join the democratic process.⁸⁷ Underlying this joint mobilization was the gradual strengthening of ethnically based organizations made possible during the 1990–2002 democratic years. In this sense, the first democratic years provided enabling conditions for reform-seeking forces to grow and raise awareness about their demands for a more inclusive polity.⁸⁸

Nepal’s transition in 2006 further broadened participation by different ethnic and minority groups, including traditionally marginalized Madhesis, Dalits, and Indigenous groups, thereby significantly increasing the number of people invested in the political process.⁸⁹ The 2008 elections saw a surprising victory for the Maoists, who secured a majority through a coalition with the Madhesi parties and were in power until 2012. The largest parties since 2006 have been the Communist Party of Nepal (Maoist Centre), CPN (M), the CPN (UML), and the NC. Moreover, smaller Madhesi parties have formed.⁹⁰ The changes towards more political inclusion are shown in Online Appendix Figure A5.

Implementation of egalitarian policies

To what extent did the democratic transition reduce group inequalities? Tackling the socioeconomic disparities among different groups has been challenging due to Nepal’s extreme poverty and the resulting dependency on foreign donors to fund the necessary policies.⁹¹ Despite the challenges, national governments have implemented several initiatives aimed at reducing ethnic inequality.⁹²

From the mid-1990s onwards, electoral politics began to shape the delivery of social protection, and various cash transfers increased in size and eligibility as parties pursued votes.⁹³ However, as noted above, the most fundamental changes happened after the transition in 2006.⁹⁴ Policies such as free primary education, strengthened decentralized management, and support for students from poor and marginalized groups have increased opportunities for Dalits and Indigenous groups (Janajatis).⁹⁵ There have also been numerous improvements in health services over the years,⁹⁶ and today significantly more Nepalis have access to water and sanitation services, secondary schools, public hospitals, and electricity than two decades ago.⁹⁷

Nepal has integrated the reduction of ethnic inequalities as a significant goal in its development strategy.⁹⁸ The alleviation of economic inequalities linked to ethnic identities was implicit if not explicit in many social protection programmes.⁹⁹ Since 2007, the government’s three-year plans (2007–2010, 2010–2013) have taken a number of steps to aid Dalits, Madhesis, and Janajatis as well as other recognized marginalized groups, and most recently, the stated aim is a minimum social protection floor for all.¹⁰⁰ The 2015 constitution and an amendment to the Civil Service Act contain provisions for reservations and quotas to expand opportunities for the historically disadvantaged groups,¹⁰¹ and Nepal has implemented a range of affirmative action policies to address group disparities.¹⁰² Moreover, legislation from 2011 aims to reduce caste-based discrimination,¹⁰³ and there is evidence that discriminatory practices against certain castes are being reduced, though more in the public than in the private sphere.¹⁰⁴ This improved access to state employment and contracts for previously excluded groups is suggested by the data in Figure A6.

Overall, Nepal has made great progress on improving living standards and reducing poverty during this period. Trends in the Human Development Index (HDI) and Human Poverty Index (HPI) have been positive across all castes and ethnic groups,¹⁰⁵ and data suggests that these developments have reduced ethnic inequalities.¹⁰⁶ Differences in the average household well-being index among castes and ethnic groups are shrinking,¹⁰⁷ and poverty levels have decreased more rapidly among disadvantaged groups between 1995 and 2010.¹⁰⁸ More and more members of historically marginalized groups who have completed an education or earned high wages in tourism, high-tech, or working abroad have entered the middle class.¹⁰⁹

Despite the continued political dominance of the CHHE, low development levels, and lack of cohesiveness among disadvantaged groups, universal as well as targeted policies seem to have improved the situation of the disadvantaged groups. While inequalities remain pronounced,¹¹⁰ there is evidence that disparities in human development by caste and ethnicity have been reduced over time, especially in more recent years.

Nepal I: democracy without inclusion

The analysis of Bolivia, South Africa, and Nepal’s second democratic spells showed how democratic transitions led to reductions in ethnic inequality in a fashion consistent with the theoretical mechanism. In the following, I discuss two cases showing that beneficial effects of democracy in the form of more ethnic equality are far from given. I examine two negative cases, Nepal (first democratic spell, 1991–2002) and Guatemala (1986–), where the democratic transitions led to limited distributional changes.

Political mobilization of previously excluded groups

In late 1989, the Nepali Congress Party (NC) leadership sought to initiate a mass movement to introduce democracy, which was joined by left and communist parties. Following mass demonstrations, King Birendra lifted the ban on political parties in 1990 and allowed for an interim government headed by a coalition of opposition leaders. Subsequently, a Constitution Recommendation Commission drafted a new constitution, setting the stage for parliamentary elections in 1991.

Democratization in 1991 was not sufficient to bring about broad-based political inclusion and significant socioeconomic advances for disadvantaged groups. Even though non-CHHE groups demanded power-sharing institutions, the new constitution adopted a first-past-the-post method, a unitary state structure, contained discriminatory constitutional articles, and maintained Nepal’s status as an official Hindu kingdom.¹¹¹ This helped to perpetuate the long-standing dominance of the state by the CHHE, which had promoted the idea of Nepal as a relatively homogenous Hindu nation of Nepali speakers.¹¹² Unsurprisingly, non-CHHE groups were generally unsatisfied with the 1990 constitution.¹¹³

The CHHE groups continued to dominate every sector of the government.¹¹⁴ The key political parties during this period, i.e. the Nepali Congress Party (NC) and the Communist Party of Nepal (Unified Marxist-Leninist) or CPN (UML), predominantly consisted of individuals from CHHE backgrounds. Despite securing the third highest number of seats in the 1991 parliament, the Maoists faced continuous sidelining by the government, ultimately leading them to go underground.¹¹⁵

However, the democratic transition did provide political and civil liberties that gave marginalized groups a greater platform to organize, highlight inequalities, and make political demands.¹¹⁶ The Madhesi movement, the Dalit movement, and the Indigenous nationalities movement all sought access to economic resources, political influence, equal treatment, and especially in the case of the Madhesi, more regional self-governance.¹¹⁷

Ethnic politics became more salient after competing political forces (Maoists, monarchy, and parliamentary parties) began to respond to and exploit ethnic grievances.¹¹⁸ The parliamentary parties recognized that without promising significant reforms, they risked losing the support of disadvantaged groups to either the rebels or the king.

Disadvantaged groups gain political influence and the implementation of policies

During the 1990s, the Indigenous nationalities and Dalits often supported the CPN (UML), while the Madhesis supported the NC in higher numbers. Although the two parties were in government consistently throughout the 1990s, they had only initiated minimal reforms to tackle the discrimination and poorer living conditions experienced by the non-CHHE groups.¹¹⁹ The newly emerged social movements had not gained enough strength to pressure the parliamentary parties to reform the polity substantially.¹²⁰ Part of their limited success could be due to the lack of effective cooperation. While these movements shared some common grievances, coordinated political actions between them were limited,¹²¹ and they clearly had exclusionary tendencies. For instance, the Indigenous nationalities movement largely ignored the issues faced by low-caste Hindus and Madhesis, and vice versa.¹²² Moreover, the development of broad-based welfare was limited by clientelist and patronage party politics, which continue to dominate Nepalese politics.¹⁴

Democracy was fragile in Nepal and already started to erode with the onset of violent conflict. In 1996, an armed conflict between Maoist rebels and the state erupted that lasted ten years and left more than 13,000 people dead and hundreds of thousands displaced. While the rebels’ rhetoric was primarily ideological, mobilization had clear ethnic and regional patterns.¹²³ In the midst of the war, the murder of King Birendra and his immediate family in 2001 created further instability. The new king, Gyanendra, dissolved an elected government in 2002 and in February 2005 seized direct control of the country, in effect dismantling democracy until 2006 when a new “people’s movement” forced him to resign. Several factors are likely to have contributed to the democratic breakdown 2002–2006, some directly tied to group inequalities, which I discuss at greater length in Online Appendix C.

To summarize, Nepal’s first democratic spell 1990–2002 led to minimal political and socioeconomic inclusion of marginalized groups (see Online Appendix F for further discussion). Nevertheless, the period did help to mobilize social forces, setting the stage for the reintroduction of democracy in 2006, after which Nepal’s political system became more inclusive, as discussed above.

Guatemala: limited reductions in ethnic inequality

Guatemala had a very high level of ethnic inequality at its transition to democracy in 1986, and the country has the second-highest proportion of Indigenous inhabitants in Latin America.¹²⁴ The Guatemalan civil war from 1960 to 1996 concluded with a peace agreement negotiated between the government and the guerilla coalition URNG. Ethnic inequality had played a major role in the conflict, but it was not explicitly an ethnic conflict in terms of pitting Ladinos against Indigenous.¹²⁵ The Guatemalan Truth Commission (Comisión de Esclarecimiento Histórico, CEH) identified three broad sets of historical causes of the conflict: marginalization and subordination of Indigenous communities, economic exclusion, and authoritarianism.¹²⁶

Political mobilization of previously excluded groups

The protracted peace negotiation process (1987–1996) took place in a context where civilian politicians had returned to government in 1986. Increased political freedoms were crucial for the reemergence of civil society, including the Indigenous movement.¹²⁷ By 1992, a visible Indigenous movement took a central role in shaping the Agreement on the Identity and Rights of Indigenous Peoples (AIRI). This agreement became a cornerstone of the Guatemalan peace accords.¹²⁸

Despite the Indigenous Maya movement’s significant role just after the end of the civil war, Guatemala has yet to witness the rise of substantial Indigenous political parties or the emergence of leading presidential candidates from the Indigenous communities.¹²⁹ National-level Indigenous mobilization in Guatemala has been markedly less robust than in other Latin American nations with substantial Indigenous populations.¹³⁰ By the 2000s, the Indigenous movement had even weakened and lost its ability to mobilize people outside the capital where most of the Indigenous population lived.¹³¹ Today, Guatemala’s Indigenous movement is “fragmented and demobilized.”¹³² This absence of effective mobilization and subsequent political inclusion of the Indigenous population is shown in Figure A7.

While the exact causes of the relatively weak mobilization remain unclear, some scholars attribute it to social and political fragmentation within the Indigenous community – reinforced by a legacy of violence and inhibitive institutional rules – which hampered unified action based on Indigenous identity (see Online Appendix E for an elaborate discussion). A significant number of Indigenous leaders have endorsed leftist political parties,¹³³ and while leftist political mobilization could, in principle, also lead to significant reductions in ethnic inequality, leftwing parties have failed to implement egalitarian policies. Moreover, none of the major parties has been responsive to the grievances and demands of the Indigenous population.¹³⁴ This makes sense given the context of a very weakly institutionalized party system with parties that are only vaguely guided by ideology or programmatic agendas.¹³⁵

Implementation of egalitarian policies

When the Indigenous movement and other pro-poor actors gained political influence, conservative forces were effective in counter-mobilizing and blocking implementation. In direct recognition of how deep inequalities contributed to the conflict, the aforementioned peace settlements in Guatemala addressed socioeconomic themes.¹³⁶ Specifically two agreements could have addressed ethnic inequalities. The “Agreement on the Socio-Economic Aspects and the Agrarian Situation (1996)” could have reformed social service delivery and land inequalities, and the more targeted “Agreement on the Identity and Rights of Indigenous Peoples” (AIRI) included provisions for elimination of discrimination and social and economic rights for Indigenous peoples.¹³⁷

The more ambitious aspects of the peace accords were not implemented, however.¹³⁸ Most clearly, constitutional amendments concerning the AIRI and 50 other issues were rejected following a well-orchestrated campaign by conservative forces.¹³⁹ This made apparent the weakness of the party system compared with the strength of conservative forces that opposed at least parts of the peace accords.¹⁴⁰ In the same vein, pro-poor tax reforms have been blocked by “the oligarchy,” which retains extraordinary influence in a Guatemalan society with a highly skewed power structure.¹⁴¹ Reductions in socioeconomic inequalities between Indigenous and non-Indigenous groups have thus been minor¹⁴² as illustrated in Figure A8.

In sum, the visible Indigenous movement in Guatemala after democratization was fragmented and lacked serious political leverage. In particular, the “popular wing,” which prioritized socioeconomic matters, lacked connections to established political actors, which constrained its ability to influence policymaking.¹⁴³ In this sense, the Indigenous population was less successful in Guatemala than in Bolivia (and Ecuador) at mobilizing politically, gaining national representation, and subsequently improving its relative socioeconomic condition. The status quo was further secured as traditional elites in Guatemala were effective at blocking the most important reforms.

Conclusion

This paper has demonstrated how democratic transitions in Bolivia, South Africa, and Nepal (after 2006) led to reductions in ethnic inequality through political empowerment of disadvantaged groups and egalitarian policies. Table 3 summarizes the empirical developments in the analysed cases together with the main theoretical steps. Although the countries differ in important respects, their political developments following democratization largely coincide with the proposed empirical manifestations. In Bolivia, the link between the transition to democracy and inequality-reducing policies is far from immediate or mechanic. However, democracy worked as a precondition for the MAS’ mobilization and electoral success. In South Africa, the transition to full democracy led to the ANC’s electoral victory in 1994. Over several years of ANC government, a range of policies were introduced that helped to reduce the gap between Black and White South Africans. Nepal’s first democratic spell (1990–2002) did not immediately lead to empowerment of previously excluded groups, but things gradually improved after the second transition in 2006, which has been followed by some inequality-reducing measures.

Table 3. Summary of findings.

	Cause	Link 1	Link 2	Link 3	Outcome
Theorized mechanism	Democratic transition	Democratic transitions allow previously excluded groups to participate politically	Disadvantaged groups gain political influence directly or indirectly	Egalitarian policies are implemented	Reduced levels of socioeconomic inequality between ethnic groups
Bolivia	1982	– MAS forms and Indigenous voters are mobilized by MAS	– MAS in power 2005–2019 and 2020– – Existing parties pressured to address Indigenous demands	Universal, targeted, and anti-discrimination policies implemented	High
South Africa	1994	– Black-supported ANC already mobilized strong support prior to transition	– ANC in power 1994–	Universal, targeted, and anti-discrimination policies implemented	High
Nepal II	2006	– More effective mobilization after 2006	– Better rep. and more influence after 2006 – New and existing parties	Universal, targeted, and anti-discrimination policies implemented	Moderate
Nepal I	1991	– Some level of mobilization in 1990s	– Disadvantaged groups not well-represented in 1990s – Continued dominance of traditional elite	Limited policy change (e.g. cash transfers)	Low
Guatemala	1986	– Limited Indigenous mobilization	– Indigenous movement unable to influence national politics – Effective counter-mobilization by traditional elite	Limited policy change	Low

In Guatemala and Nepal’s first democratic spell, democratization did not lead to strong political mobilization by previously excluded groups. In both cases, lack of mobilization was linked to high fractionalization within these groups, and traditional elites were able to retain significant political influence. As such, the Guatemalan and Nepali (I) cases show how the mechanism linking democratization to increased group equality strongly depends on coherent mobilization with the goal to reduce ethnic inequality. Such mobilization patterns were much clearer in Bolivia and South Africa, but it would be too simplistic to portray them as completely straightforward and without internal tensions (see Online Appendix E for nuances).

The findings call for further examination of the deeper causes of the variation in the collective mobilization of disadvantaged groups to achieve distributive change. Building on the analysis presented above, I offer several hypotheses in Online Appendix H, which could be examined in a broader set of cases. These include the negative legacies of conflict and repression; democratic age and quality, features of the transition process (i.e. whether it is led by incumbents or by disadvantaged groups), the strength and institutionalisation of the party system, including the level of clientelism, as well as ethnic demography, such as subgroup fragmentation and size of the previously excluded group.

Note that the employed minimalist conceptualization of democracy also holds implications for my findings. In Online Appendix G, I show that where democracy had the least impact on reducing ethnic inequality, the quality of democracy was lowest.¹⁴⁴ This fits with a broader pattern regarding the consequences of democracy, showing that the beneficial effects mainly manifest in enduring, high-quality democracies.¹⁴⁵

More broadly, this paper speaks to the large-N literature on the effects of democracy on various outcomes of interest,¹⁴⁶ including inequality (see Online Appendix A), as I have demonstrated the value of using comparative case studies to investigate mechanisms and scope conditions.

Acknowledgements

I thank Svend-Erik Skaaning, Kees van Kersbergen, Stephan Haggard, Gerardo Munck, as well as participants at the APSA 2022 panel “Determinants of Support for Authoritarian Rule in Democracies and Autocracies” and the “Ethnicity, Inequality, Conflict, and Democracy” workshop in 2022 at Aarhus University.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Lasse Egendal Leipziger

Lasse Egendal Leipziger is an Assistant Professor at the Department of Political Science, Aarhus University. His research focuses on democratization, socioeconomic inequality, and ethnic politics.

Notes

1 E.g. Meltzer and Richard, “A Rational Theory of the Size of Government”; Huber and Stephens, Democracy and the Left.

2 E.g. Knutsen, “Reinvestigating the Reciprocal Relationship between Democracy and Income Inequality”; Acemoglu et al., “Democracy, Redistribution, and Inequality”; Gradstein and Milanovic, “Does Libertè = Egalité? A Survey of the Empirical Links between Democracy and Inequality with Some Evidence on the Transition Economies.”

3 See Scheve and Stasavage, “Wealth Inequality and Democracy”; Dorsch and Maarek, “Democratization and the Conditional Dynamics of Income Distribution”; Albertus and Menaldo, Authoritarianism and the Elite Origins of Democracy; Keefer, “Clientelism, Credibility, and the Policy Choices of Young Democracies”; Jensen and Skaaning, “Democracy, Ethnic Fractionalisation, and the Politics of Social Spending.” See Appendix A for detailed review.

4 See Appendix A for an elaboration.

5 Huber et al., “Democracy, Targeted Redistribution and Ethnic Inequality”; Leipziger, “Does Democracy Reduce Ethnic Inequality?”

6 By “case” I mean a country’s democratic spell, which constitutes the spatio-temporal bounds of the operation of the mechanism. By “mechanism” I understand the entities and activities that link cause and outcome.

7 See Goertz, Multimethod Research, Causal Mechanisms, and Case Studies, 66–8.

8 See Gerring et al., “Does Democracy Matter?” for a review.

9 See Skaaning et al., “A Lexical Index of Electoral Democracy,” 1495.

10 See Appendix G for a full conceptual discussion.

11 Stewart, Horizontal Inequalities and Conflict, 3–23.

12 Vogt, Mobilization and Conflict in Multiethnic States.

13 Canelas and Gisselquist, “Horizontal Inequality as an Outcome,” 307–8.

14 Ibid.

15 E.g. Leipziger, “Does Democracy Reduce Ethnic Inequality?”; Huber and Stephens, Democracy and the Left.

16 See Dorsch and Maarek, “Democratization and the Conditional Dynamics of Income Distribution.”

17 Beach and Pedersen, Process-Tracing Methods: Foundations and Guidelines, 11; Goertz, Multimethod Research, Causal Mechanisms, and Case Studies, 64–5. With this procedure, I clearly “select on the dependent variable.” However, this procedure is only a problem if the analysis were purely cross-sectional, that is, one draws inferences across cases rather than based on within-case analysis.

18 McAdam et al., Dynamics of Contention, 10.

19 Leipziger, “Does Democracy Reduce Ethnic Inequality?”

20 Gisselquist, “Ethnic Politics in Ranked and Unranked Systems,” 398.

21 Vogt, Mobilization and Conflict in Multiethnic States, 71; Madrid, “Indigenous Parties and Democracy in Latin America,” 167.

22 Stewart et al., “Policies toward Horizontal Inequalities.”

23 Ibid.

24 See Huber and Stephens, Democracy and the Left.

25 Dunkerley, Rebellion in the Veins, 251–67.

26 Flesken, “Why Ethnic Parties Form,” 1168; Gisselquist, “Ethnic Politics in Ranked and Unranked Systems.” Mestizo refers to mixed ancestry with white European and Indigenous background. Within Bolivia, both the Indigenous and non-Indigenous populations can be subdivided into several subgroups. Among the Indigenous, the primary language groups are Quechua and Aymara, with a smaller group representing Lowland communities. Bolivians who do not identify as Indigenous typically categorize themselves as Mestizo or White (Gisselquist, “Ethnic Politics in Ranked and Unranked Systems: An Exploratory Analysis”).

27 Faguet, “Latin America’s Shifting Politics.”

28 Madrid, The Rise of Ethnic Politics in Latin America, 41; Van Cott, From Movements to Parties in Latin America; Yashar, Contesting Citizenship in Latin America.

29 Gisselquist, “Ethnic Politics in Ranked and Unranked Systems,” 388.

30 Madrid, “Indigenous Parties and Democracy in Latin America,” 161; According to Faguet, “Latin America’s Shifting Politics,” the old party system based on left–right cleavage collapsed, a change that was catalysed by decentralization and urbanization. The new party system was more in line with Bolivia’s social structure, which includes a large Indigenous population that fares consistently worse on socioeconomic parameters.

31 Hirseland and Strijbis, “‘We were Forgotten’,” 2020.

32 Gisselquist, “Ethnic Politics in Ranked and Unranked Systems,” 391.

33 Flesken, “Why Ethnic Parties Form.”

34 Madrid, The Rise of Ethnic Politics in Latin America, 64.

35 Ibid., 72.

36 Ibid., 166.

37 Gigler, “Poverty, Inequality and Human Development of Indigenous Peoples in Bolivia.”

38 Farthing, “An Opportunity Squandered?”; Hunter and Brill, “‘Documents, Please’”; Niedzwiecki and Anria, “Participatory Social Policies.”

39 Madrid, The Rise of Ethnic Politics in Latin America, 175–6.

40 Movimiento al Socialismo, “2010–2015 Programa de gobierno: Bolivia, país líder. La Paz : Movimiento al Socialismo,” 35–6.

41 Gray Molina, “The Challenge of Progressive Change under Evo Morales,” 66–7.

42 Madrid, The Rise of Ethnic Politics in Latin America, 176.

43 Agrawal et al., “Political Representation & Social Inclusion,” 13–21; Flesken, “On the Link between Ethnic Politics and Identification,” 160; Madrid, The Rise of Ethnic Politics in Latin America, 176.

44 Madrid, The Rise of Ethnic Politics in Latin America, 176.

45 Ibid.

46 Gray Molina, “The Challenge of Progressive Change under Evo Morales,” 73; Madrid, The Rise of Ethnic Politics in Latin America, 176.

47 Hicks et al., “Identity, Patronage, and Redistribution.”

48 Madrid, The Rise of Ethnic Politics in Latin America, 175.

49 Welsh, The Rise and Fall of Apartheid, 47; “Afrikaner” denotes a South African individual whose family has Dutch origins and whose primary language is Afrikaans.

50 Seekings and Nattrass, Class, Race, and Inequality in South Africa, 2.

51 Wood, “An Insurgent Path to Democracy.”

52 Campbell, “South Africa,” 129; Herbst, “The Nature of South African Democracy,” 210.

53 Campbell, “South Africa,” 129.

54 Herbst, “The Nature of South African Democracy,” 210.

55 Alence, “South Africa After Apartheid,” 81; Campbell, “South Africa,” 112–4; Gethin, “Extreme Inequality and the Structure of Political Cleavages in South Africa, 1994–2019,” 2.

56 Campbell, “South Africa,” 114.

57 Herbst, “The Nature of South African Democracy,” 212; Seekings and Nattrass, Class, Race, and Inequality in South Africa, 3.

58 Simkins, “South African Disparities,” 105.

59 Alence, “South Africa After Apartheid,” 84.

60 Lieberman and Lekalake, “South Africa's Resilient Democracy,” 104; Simkins, “South African Disparities,” 109.

61 Alence, “South Africa After Apartheid,” 80, 87; Campbell, “South Africa,” 120.

62 Seekings and Nattrass, Class, Race, and Inequality in South Africa, 18. Apartheid policies were somewhat relaxed before 1994, which helps to explain, why a Black middle class slowly emerged in the late apartheid period (Seekings and Natrass, Class, Race, and Inequality in South Africa, 34).

63 Ibid., 37–8, 343.

64 Ibid., 371.

65 Chatterjee et al., “Can Redistribution Keep Up with Inequality?”; Seekings and Nattrass, Class, Race, and Inequality in South Africa, 369.

66 Seekings and Nattrass, Class, Race, and Inequality in South Africa, 37–8, 358.

67 Campbell, “South Africa,” 112; Lieberman and Lekalake, “South Africa's Resilient Democracy,” 111; Simkins, “South African Disparities,” 116.

68 Lieberman and Lekalake, “South Africa's Resilient Democracy,” 111.

69 Ibid.; Seekings and Nattrass, Class, Race, and Inequality in South Africa, 3.

70 Seekings and Nattrass, Class, Race, and Inequality in South Africa, 37–8.

71 Lieberman and Lekalake, “South Africa's Resilient Democracy,” 111; Seekings and Nattrass, Class, Race, and Inequality in South Africa, 369.

72 Seekings and Nattrass, Class, Race, and Inequality in South Africa, 343.

73 Campbell, “South Africa,” 129.

74 Alence, “South Africa After Apartheid,” 85; Simkins, “South African Disparities,” 107.

75 Seekings and Nattrass, Class, Race, and Inequality in South Africa, 39, 45, 341.

76 Chatterjee et al., “Can Redistribution Keep Up with Inequality?”

77 Lieberman and Lekalake, “South Africa's Resilient Democracy,” 111; Seekings and Nattrass, Class, Race, and Inequality in South Africa, 45, 344.

78 Chatterjee et al., “Can Redistribution Keep Up with Inequality?” 1–3, 34; Seekings and Nattrass, Class, Race, and Inequality in South Africa, 45.

79 In practice, the distinction between ethnic group and caste is often blurred in Nepal (Hangen 2010, 26). By “ethnic inequality” I also refer to caste inequality.

80 Lawoti and Hangen, “Introduction,” 14; UNDP, “Nepal Human Development Report 2014,” 35.

81 Lawoti, “Informal Institutions and Exclusion in Democratic Nepal,” 18.

82 Hangen, The Rise of Ethnic Politics in Nepal, 1.

83 Lawoti, “Introduction to Special Section,” 11.

84 Drucza, “The Politics Behind Social Protection in Nepal,” 315; Hangen, The Rise of Ethnic Politics in Nepal, 1.

85 Lawoti, “Introduction to Special Section,” 12.

86 Ibid.

87 Ibid., 14; Kantha, “Nepal's Protracted Democratization in Terms of Modes of Transition,” 66.

88 Lawoti, “Introduction to Special Section,” 13.

89 Kantha, “Nepal's Protracted Democratization in Terms of Modes of Transition,” 67; Lawoti and Hangen, Nationalism and Ethnic Conflict in Nepal, 5.

90 Drucza, “The Politics Behind Social Protection in Nepal,” 319–23.

91 Brown, “Nepal” 292–3; Tribhuvan, “State of Social Inclusion in Nepal.”

92 Brown, “Nepal,” 292–3.

93 Webster, “Social Protection, Caste and Ethnicity in Nepal,” 19.

94 Kantha, “Nepal's Protracted Democratization in Terms of Modes of Transition,” 371; Lawoti, “Introduction to Special Section,” 13.

95 Subedi, “Caste/Ethnic Dimensions of Change and Inequality,” 2008.

96 Ibid., 11.

97 Ibid., 2, 11.

98 Brown, “Nepal”; Fukuda-Parr, “Correcting Horizontal Inequality as a Development Priority,” 99.

99 Webster, “Social Protection, Caste and Ethnicity in Nepal,” 9.

100 Ibid., 7.

101 Tribhuvan, “State of Social Inclusion in Nepal,” 4.

102 Fukuda-Parr, “Correcting Horizontal Inequality as a Development Priority,” 99; Subedi, “Caste/Ethnic Dimensions of Change and Inequality,” 11.

103 Tribhuvan, “State of Social Inclusion in Nepal,” 15.

104 Subedi, “Changes in Livelihood and Caste Relations in Udipur,” 98.

105 UNDP, “Nepal Human Development Report 2014,” 66.

106 Ibid., 17. Mean gaps widened during 1995–2003 and started narrowing during 2003–2011, perhaps due to a more inclusive political settlement in the second period. Conflict (1996–2006), disturbances, and economic upheavals had a more serious negative impact on the well-being of poorer and more vulnerable communities (UNDP, “Nepal Human Development Report 2014”).

107 Ibid., 35–6.

108 Subedi, “Caste/Ethnic Dimensions of Change and Inequality,” 3.

109 Tribhuvan, “State of Social Inclusion in Nepal,” 15.

110 Ibid., 14; UNDP, “Nepal Human Development Report 2014,” 17.

111 Lawoti, “Exclusionary Democratization in Nepal, 1990–2002,” 374–6; Lawoti, “Introduction to Special Section,” 10.

112 Lawoti, “Introduction to Special Section”; Lawoti and Hangen, “Introduction,” 14, 18.

113 Hangen, The Rise of Ethnic Politics in Nepal; Lawoti and Hangen, “Introduction,” 15; Shakya, “Capitalism and Ethnicity Facing a Rising Wave of Communism in Nepal,” 53.

114 Hangen, The Rise of Ethnic Politics in Nepal, 30; Lawoti, “Exclusionary Democratization in Nepal, 1990–2002”; Lawoti and Hangen, “Introduction,” 13.

115 Lawoti, “Chapter 1,” 30.

116 Kantha, “Nepal's Protracted Democratization in Terms of Modes of Transition,” 371; Lawoti and Hangen, “Introduction,” 16; Shakya, “Capitalism and Ethnicity Facing a Rising Wave of Communism in Nepal,” 53.

117 Lawoti and Hangen, “Introduction,” 10–1; Pfaff-Czarnecka, “Debating the State of the Nation.”

118 Lawoti and Hangen, “Introduction,” 16–7.

119 Kantha, “Nepal's Protracted Democratization in Terms of Modes of Transition,” 371; Lawoti, “Informal Institutions and Exclusion in Democratic Nepal,” 13.

120 Lawoti and Hangen, “Introduction,” 17.

121 Ibid., 19.

122 Ibid., 18.

123 Brown, “Nepal,” 295.

124 Canelas and Gisselquist, “Human Capital, Labour Market Outcomes, and Horizontal Inequality in Guatemala.”

125 Brown and Caumartin, “Horizontal Inequalities in Post-conflict Reconstruction,” 104.

126 CEH, Guatemala, 82.

127 Brown and Caumartin, “Horizontal Inequalities in Post-conflict Reconstruction,” 105; Yashar, Contesting Citizenship in Latin America, 25, 78.

128 Thorp et al., “Inequality, Ethnicity, Political Mobilisation and Political Violence in Latin America,” 464.

129 Canelas and Gisselquist, “Human Capital, Labour Market Outcomes, and Horizontal Inequality in Guatemala,” 379; Yashar, Contesting Citizenship in Latin America, 25.

130 Madrid, The Rise of Ethnic Politics in Latin America; Yashar, Contesting Citizenship in Latin America: The Rise of Indigenous Movements and the Postliberal Challenge, 25.

131 Madrid, The Rise of Ethnic Politics in Latin America, 150.

132 Ibid., 159.

133 Canelas and Gisselquist, “Human Capital, Labour Market Outcomes, and Horizontal Inequality in Guatemala,” 390.

134 Sánchez, “Guatemala's Party Universe.”

135 Ibid., 139.

136 Sanchez, “Tax Reform Paralysis in Post-conflict Guatemala,” 114.

137 Brown and Caumartin, “Horizontal Inequalities in Post-conflict Reconstruction,” 106–7.

138 Ibid., 107.

139 Ibid., 109.

140 Ibid., 114.

141 Sanchez, “Tax Reform Paralysis in Post-conflict Guatemala.”

142 Brown and Caumartin, “Horizontal Inequalities in Post-conflict Reconstruction,” 107–8, 12.

143 Thorp et al., “Inequality, Ethnicity, Political Mobilisation and Political Violence in Latin America,” 465.

144 See Munck, “What is Democracy?” for a conceptual discussion of democratic quality.

145 Møller and Skaaning, Democratization and Autocratization in Comparative Perspective, 258; Gerring et al., “Democracy and Human Development.”

146 See Gerring et al., “Does Democracy Matter?”