Why do some indigenous communities experience assimilation while others obtain government protection for their longstanding institutions and cultures? I argue that historical experiences with labor extraction play a key role. In the nineteenth and twentieth centuries, Latin American governments enacted discriminatory policies that conscripted unpaid indigenous labor to build roads and railways. Communities exposed to this extraction faced a common enemy, increasing collective action capacity within and across communities. Over time, this ability to collectively mobilize endured and enabled native groups to obtain more substantial government protections for their institutions and cultures. I test this argument using an original natural experiment in which communities’ exposure to unpaid labor conscription on a 1920s Peruvian highway was as-if randomly assigned. I combine this design with historical data on indigenous rebellions and community-level collective action, as well as contemporary data on community conflict with outsiders, language survival, and the persistence of indigenous institutions. The theory and evidence shed new light on the long-term effects of discriminatory policies on indigenous groups’ political power.

A well-established political economy literature predicts a positive relationship between ethnic homogeneity and public goods provision. Yet, the initial phase of education provision in nineteenth-century Latin America benefited ethnically heterogeneous, rather than homogeneous, areas. This article explores this unexpected finding, highlighting the central role of longstanding norms around regressive, ethnic-based taxation, which redistributed wealth from marginalized Indigenous groups to a non-Indigenous political and economic elite. Employing a natural experiment from Peru and historical case studies, I show that non-Indigenous political elites provided education most broadly in heterogeneous areas where there existed both a coethnic demand for education and a large Indigenous tax base that could be exploited to meet this demand. Perhaps surprisingly, education provision in homogeneous non-Indigenous areas was relatively low---except where competition among non-Indigenous political elites induced politicians to incur the norms-based cost of taxing their coethnics. My findings add an important revenue dimension to the literature on coethnicity and public goods provision.

Existing studies of state-building often endeavor to explain incumbent decisions to project state authority under non-democratic regimes. Yet, we argue that the introduction of democratic competition often gave rise to a state-building strategy that followed a different logic than existing theories suggest. Specifically, elections provided information on where threats to an incumbent resided, leading to a more direct and efficient deployment of bureaucrats and strategic public goods (e.g., education and roads). These state-building efforts generally targeted areas governed by the incumbent's political opposition. We label this "politicized state-building." These investments had long-term implications not only for state-building but also basic welfare indicators, including literacy, road-building and future land reform, in affected areas. We test this theory using a regression-discontinuity design and data on local public goods provision and bureaucratic deployment in 1960s Peru under President Fernando Belaúnde (1963-1968). Our results suggest a new critical juncture in state-building that shaped both state presence and public goods provision in traditionally peripheral areas.

Why do some indigenous groups achieve coethnic political representation while others do not? In this paper, I highlight the primary role of communal property in shaping indigenous representation. While scholars often laud the developmental benefits of communal land titling, I argue that formalizing collectively held land can inhibit indigenous coordination to achieve political representation. Where communal land is informally held, indigenous groups are more likely to invest in traditional institutions that facilitate collective action to elect coethnic candidates to political office. Conversely, titling communal property secures indigenous land access but in the process erodes traditional institutions that would otherwise promote collective action during elections. I test my argument using a multi-method approach that includes interviews and experiments with three-hundred Peruvian indigenous leaders, historical land-title data, and information scraped from mayoral candidate CVs. The findings suggest that the oft-cited economic benefits of collective property may generate negative political effects.

Autonomy has long been regarded as a principal solution to indigenous groups' persistent marginalization from state-administered distributive politics. Despite its promise, however, autonomy also presents indigenous communities with a series of new challenges in obtaining access to needed public goods and services. The fundamental issue involves an autonomy-representation dilemma: autonomy provides indigenous groups with greater control over local governance, but that increased independence from the state may imply that indigenous leaders have fewer resources and administrative capacity to respond to local demands. As such, pursuing coethnic representation within the state might--under certain conditions--provide a preferable way for indigenous groups to obtain needed goods and services. Drawing on natural experimental evidence and an original survey of indigenous community presidents from Peru, I first demonstrate that achieving coethnic political representation within the state can expand indigenous groups' access to the public good they most need: water. I then illustrate how capacity constraints that arise from autonomy have prevented native groups in Bolivia's autonomous municipalities from achieving similar distributive gains. Ultimately, the findings provide insights for understanding the sources of--and potential institutional remedies for--persistent distributive inequalities between indigenous and non-indigenous groups in the Americas and beyond.

Weakly institutionalized party systems are a defining feature of third-wave democracies. Yet, in some countries, like Peru, party weakness is not a static equilibrium but rather part of a dynamic process of “party system erosion” in which weak parties become weaker over time as independents come to dominate subnational posts. As I argue, party system erosion is driven by a particular configuration of institutional factors—weak party brands, low barriers to ballot access, and limited partisan control over resource distribution during and after campaigns. These institutional features increase the likelihood that experienced candidates will run as independents. When these candidates are elected, experience enables them to obtain more intergovernmental discretionary transfers. The superior in-office performance of experienced, independent officials further weakens party brands, leading fewer and fewer experienced candidates to run with parties. I test this theory using a dataset of 80,000 subnational officials and a regression discontinuity design.

As increasingly large shares of the developing world’s population come to live in cities, it is important to examine the effects of political, fiscal, and administrative decentralization on urban governance and service delivery. Relevant academic scholarship and policy research, we show, suggests that clientelism, populism, and local capture often persist following the establishment of municipal elections. However, conditions such as political competition, independent fiscal resources, and strong civil societies can facilitate more democratic outcomes following decentralization. Meanwhile, our review of literature on decentralization’s impact on two quintessentially “urban” services—land market regulation and urban water and sanitation—suggests that decentralization involves important trade-offs. On the one hand, decentralization can help citizens to pressure more effectively for inclusion and access, particularly in the presence of political competition and a robust civil society. On the other hand, it can make it more difficult for policymakers to address metropolitan-level or long run concerns regarding investments in basic infrastructure that are often not at the forefront of voters’ minds. We also highlight the need for primary data collection, suggest research design strategies that would allow for more rigorous empirical analyses, and highlight important topics that have received very little attention.

A growing literature in political science and economics has addressed how the provision of information to citizens shapes accountability, corruption, and voter turnout in the developing world. Yet, only limited attention has been devoted to understanding the information environments of local-level elected officials, particularly whether they have adequate and accurate information about the effects of certain nationally funded programs and how different types of information affect their decisions to implement those programs in their municipalities. Two questions thus arise: 1) How do officials’ decisions to implement programs change once they are correctly informed about various program effects? And 2) are local officials more concerned with the policy outcomes of programs or their potential electoral effects? I use a field experiment with Peruvian mayors to answer this question.
Funded by: J-PAL Governance Initiative
Implementing partner: IPA

Instrumental-variables (IV) analysis bridges structural equation modeling and design-based methods for causal inference. In both frameworks, researchers invoke random or as-if random assignment of units to treatment conditions as a way of addressing concerns about confounding variables. Yet, despite areas of convergence, these different approaches to IV analysis rely on distinct underlying assumptions. In this chapter, we discuss areas of overlap and divergence in the modeling strategies. While structural equations embed several core assumptions in linear response schedules, design-based approaches disaggregate and clarify these assumptions. We then use a hypothetical empirical example—price elasticity of demand for coffee—to illustrate these similarities and differences. The chapter concludes with a note of caution on the use of instrumental variables for political science and international relations: whichever framework is invoked, IV analysis generates results that may not generalize beyond the specific intervention that gave rise to the instrument.

Platform companies disrupt not only the economic sectors they enter, but also the regulatory regimes that govern those sectors. We examine Uber in the United States as a case of regulating this disruption in different arenas: cities, state legislatures, and judicial venues. We find that the politics of Uber regulation does not conform to existing models of regulation. We describe instead a pattern of disrupted regulation, characterized by a consistent challenger-incumbent cleavage, in two steps. First, an existing regulatory regime is not deregulated but successfully disregarded by a new entrant. Second, the politics of subsequently regulating the challenger leads to a dual regulatory regime. In the case of Uber, disrupted regulation takes the form of challenger capture, an elite driven pattern, in which the challenger has largely prevailed. It is further characterized by the surrogate representation of dispersed actors—customers and drivers—who do not have autonomous power and who rely instead on alignment with the challenger and incumbent. In its surrogate capacity in city and state regulation, Uber has frequently mobilized large numbers of customers and drivers to lobby for policy outcomes that allow it to continue to provide service on terms it finds acceptable. Because drivers have reaped less advantage from these alignments, labor issues have been taken up in judicial venues, again primarily by surrogates (usually plaintiffs’ attorneys) but to date have not been successful.

The rise of labor platforms is a cause and consequence of an ongoing change in the nature of work toward a model where workers split time across multiple companies, without the rights and benefits of employment. Given that Uber exercises more control over workers than most other labor platforms, it provides a most likely case for regulation. Nevertheless, regulators in legislative and judicial arenas failed to regulate labor on ride-hailing apps in the first six years. We explain this failure in terms of 1) the problem of classifying a new type of firm, and 2) the relative distribution of power between labor and capital. The weakness of the former emerges from collective action problems and has given rise to a form of skewed surrogate representation. A handful of recent regulatory attempts introduces a note of uncertainty for the future.