publications

[book] Information, Accountability, and Cumulative Learning: Lessons from Metaketa I
Cambridge University Press (forthcoming, 2019)
co-edited w/ Thad Dunning, Guy Grossman, Susan Hyde, Macartan Humphreys, Craig McIntosh
abstract Throughout the world, voters lack access to information about politicians, government performance, and public services. Efforts to remedy these informational deficits are numerous. Yet do informational campaigns influence voter behavior and increase democratic accountability? Through the inaugural Metaketa Initiative, sponsored by the Evidence in Governance and Politics (EGAP) research network, this book aims to address this substantive question and at the same time introduce a new model for cumulative learning that increases coordination among otherwise independent researcher teams. It presents the cumulative results (meta-analysis) from six independently conducted but coordinated field experimental studies, the results from each individual study, the findings from a related evaluation of whether practitioners utilize this information as expected, and discusses lessons learned from EGAP’s efforts to coordinate field experiments, increase replication of theoretically important questions across contexts, and increase the external validity of field experimental research.

amazon | data | pre-analysis plans

[article] Does Electing Islamists Increase Religious Violence and Intolerance?
British Journal of Political Science (conditionally accepted)
w/ Nicholas Kuipers, Michael Weaver
abstract We estimate the effect of incumbency by Islamist parties on the incidence of religious violence and intolerance in Indonesia, exploiting discontinuities in the proportional representation system used to allocate seats in district legislative elections—the most local tier of parliamentary government. We find that the presence of additional Islamist (as opposed to secular nationalist) legislators exacerbates religious conflict according to certain measures. There is no evidence that Islamist rule affects average attitudes toward religious minorities among majority-group survey respondents, although it does increase expressions of extreme intolerance. Social emboldening may underlie these effects, as Islamist incumbency appears to boost the perceived acceptability of holding intolerant worldviews. The results shed light on the consequences of having extremist parties gain a share in local power.

paper & appendix | data | pre-analysis plan

[article] Voter Information Campaigns and Political Accountability: Cumulative Findings from a Preregistered Meta-analysis of Coordinated Trials w/ Thad Dunning, Guy Grossman, Susan Hyde, Macartan Humphreys, Craig McIntosh, Claire L. Adida, Eric Arias, Clara Bicalho, Taylor C. Boas, Mark T. Buntaine, Simon Chauchard, Anirvan Chowdhury, Jessica Gottlier, F. Daniel Hidalgo, Marcus Holmlund, Ryan Jablonski, Eric Kramon, Horacio Larreguy, Malte Lierl, John Marshall, Gwyneth McClendon, Marcus A. Melo, Daniel L. Nielson, Paula M. Pickering, Melina R. Platas, Pablo Querubin, Pia Raffler, Neelanjan Sircar
Science Advances (2019; v5/n7)
abstract Voters may be unable to hold politicians to account if they lack basic information about their representatives’ performance. Civil society groups and international donors therefore advocate using voter information campaigns to improve democratic accountability. Such interventions build on core assumptions of many theoretical models in political science. Yet, are these campaigns effective? Limited replication, measurement heterogeneity, and publication biases may undermine the generalizability and reliability of bodies of published research. We implemented a novel approach to cumulative learning, coordinating on the design of seven randomized controlled trials to be fielded in six countries by independent research teams. Uncommonly for multi-site trials in the social sciences, we jointly pre-registered a meta-analysis of results in advance of seeing the data. We find no evidence overall that typical, non-partisan voter information campaigns shape voter behavior, though exploratory and sub-group analyses suggest conditions under which informational campaigns could be more effective. Such null estimated effects are too seldom published, yet they can be critical for scientific progress and cumulative, policy-relevant learning.

paper | appendix | data | pre-analysis plan

[article] Secular Party Rule and Religious Violence in Pakistan
American Political Science Review (2018; v112/n1; p49–67)
w/ Niloufer Siddiqui
abstract Does secular party incumbency affect religious violence? Existing theory is ambiguous. On the one hand, religiously-motivated militants might target areas that vote secularists into office. On the other hand, secular party politicians, reliant on the support of violence-hit communities, may face powerful electoral incentives to quell attacks. Candidates bent on preventing bloodshed might also sort into such parties. To adjudicate these claims, we combine constituency-level election returns with events data on Islamist and sectarian violence in Pakistan (1988–2011). For identification, we compare districts where secular parties narrowly won or lost elections. We find that secularist rule causes a sizable reduction in local religious conflict. Additional analyses suggest that the result stems from electoral pressures to cater to core party supporters, and not from politician selection. The effect is concentrated in regions with denser police presence, highlighting the importance of state capacity for suppressing religious disorder.

paper | appendix | data

[article] Do Parties Matter for Ethnic Violence? Evidence from India
Quarterly Journal of Political Science (2016; v11/n3; p249–77)
w/ Michael Weaver, Steven Rosenzweig
abstract Ethnic group conflict is among the most serious threats facing young democracies. In this paper, we investigate whether the partisanship of incumbent politicians affects the incidence and severity of local ethnic violence. Using a novel application of the regression-discontinuity design, we show that as-if random victory by candidates representing India’s Congress party in close state assembly elections between 1962 and 2000 reduced Hindu–Muslim rioting. The effects are large. Simulations reveal that had Congress lost all close elections in this period, India would have experienced 11 percent more riots. Additional analyses suggest that Congress candidates’ dependence on local Muslim votes, as well as apprehensions about religious polarization of the electorate in the event of riots breaking out, are what drive the observed effect. Our findings shed new light on parties’ connection to ethnic conflict, the relevance of partisanship in developing states, and the puzzle of democratic consolidation in ethnically divided societies.

paper | appendix | data

[article] The Majority-Minority Divide in Attitudes Toward Internal Migration: Evidence from Mumbai
American Journal of Political Science (2017; v61/n2; p456–72)
w/ Nikhar Gaikwad
abstract Rapid urbanization is among the major processes affecting the developing world. The influx of migrants to cities frequently provokes antagonism on the part of long‐term residents, manifested in labor market discrimination, political nativism, and violence. We implemented a novel, face‐to‐face survey experiment on a representative sample of Mumbai’s population to elucidate the causes of anti‐migrant hostility. Our findings point to the centrality of material self‐interest in the formation of native attitudes. Dominant group members fail to heed migrants’ ethnic attributes, yet for minority group respondents, considerations of ethnicity and economic threat crosscut. We introduce a new political mechanism to explain this divergence. Minority communities facing persistent discrimination view in‐migration by coethnics as a means of enlarging their demographic and electoral base, thereby achieving “safety in numbers.” Our article sheds light on the drivers of preferences over internal migration. It also contributes insights to the international immigration literature and to policy debates over urban expansion.

paper | appendix | data

revisions requested

[article] The Fight Within: Intra-Party Factionalism and Incumbency Spillovers in India
American Journal of Political Science (R&R)
abstract Political parties are thought to play a key role in countering the centrifugal tendencies of federalism. Yet we possess little evidence about how nationalized parties work to shape electoral outcomes in developing states with federal systems. In such contexts, do powerful, resource-rich incumbents at the national level strategically mobilize votes for their parties in subnational elections? I theorize that incumbents’ propensity to help co-partisans running in lower-tier races hinges on internal party organization, and above all factionalism. Leveraging (quasi-)experimental methods, as well as new data on the linkages between India’s national and state elections for the period 1977–2008, I demonstrate the existence of large, positive electoral spillovers for unified parties, but negative spillovers for factionalized parties. This paper advances our understanding of political accountability in multi-level electoral systems. It also contributes to the literature on the incumbency advantage, and illuminates the principal-agent dilemma at the heart of internally divided parties.

paper & appendix

[article] Do Politicians Discriminate Against Internal Migrants? Evidence from Nationwide Field Experiments in India
w/ Nikhar Gaikwad
American Journal of Political Science (R&R)
abstract Rural-to-urban migration is reshaping the economic and social landscape of the global South. Yet migrants often struggle to integrate into cities. We conduct countrywide audit experiments in India to test whether urban politicians discriminate against internal migrants in providing constituency services. Signaling that a citizen is a city newcomer, as opposed to a long-term resident, causes incumbent politicians to be significantly less likely to respond to requests for help. Standard “nativist” concerns do not appear to explain this representational shortfall. Instead, we theorize that migrants are structurally disposed to participate in destination-area elections at lower rates than natives. Cognizant of this, reelection minded politicians decline to cater to migrant interests. Follow-up experiments support the hypothesis. We expect our findings to apply to a swath of fast-urbanizing democracies, with implications for international immigration too. Policywise, mitigating migrants’ de facto disenfranchisement should improve their welfare.

paper | appendix

working papers

[article] Dominant Parties and the Origins of India’s Weakly Institutionalized Party System
abstract Why are party systems well-institutionalized in some settings, and chronically weak in others? I argue that unstable party systems are more likely to arise in regions where nationally dominant parties monopolize political competition at the onset of mass-franchise democracy. Dominant parties crowd out political opposition. Hence the eventual breakdown of a dominant party entails the severing of party-voter linkages locally. In the resulting vacuum, politicians face uncertainty about the electoral prospects of newly emergent parties. This leads to a collective action dilemma wherein candidates defect from expanding parties and sort instead into smaller, fragmentary ones. Consequently, stable party systems fail to take hold. Subnational evidence from India buttresses the theoretical propositions. The success of the once-dominant Congress Party during the country’s inaugural elections (1951–2) robustly predicts greater electoral volatility in the decades following the decline of one-party dominance in the 1970s. Differential patterns of nationalist mobilization during the colonial period provide additional support for the paper’s core claims. Overall, the findings imply a striking paradox: dominant parties that help “bind the nation together” during democracy’s initial stages sow the seeds of long-run political instability.

paper

[article] Cumulative Evidence, Beliefs, and Out-of-Sample Prediction Accuracy: A Results Dissemination Experiment with Practitioners (submitted)
w/ Thad Dunning, Guy Grossman, Susan Hyde, Macartan Humphreys, Catlan Reardon
abstract Can social scientific evidence help change beliefs about the impacts of a political intervention? Recent years have seen a proliferation of credible empirical studies addressing policy-relevant topics. Yet little is known about the uptake of such evidence by frontline decisionmakers. We expose a sample of expert practitioners based in Washington D.C. to the results of six coordinated randomized control trials (RCTs). Our pre-registered dissemination experiment randomly varied the order in which subjects were presented with the findings of (i) a single RCT, (ii) a meta-analysis of multiple RCTs, and (iii) an observational study focused on the same theoretical question. We estimate how each condition affects the accuracy of out-of-sample predictions and attitudes about the intervention’s efficacy. Participants proved responsive to evidence. Meta-analytic results were especially impactful on beliefs, though less so on forecasting. The findings furnish evidence for external validity and are mostly sanguine about the persuasive potential of cumulative research.

paper | pre-analysis plan

in progress

[rct] Getting on the Grid: The Politics of Public Service Formalization in Urban India
w/ Anjali Bohlken, Nikhar Gaikwad
abstract Citizens in many cities of the developing world struggle to access state services. Instead, they rely on services that are provided informally or illegally. Such ‘off the grid’ services tend to be intermittent, expensive, and of low quality. Their existence undercuts the fiscal contract between citizens and states, and undermines governments’ ability to invest in infrastructure. This phenomenon raises three questions. First, what prevents poorer citizens from accessing public services to which they are legally entitled? Second, does transitioning into formal, state service provision affect political attitudes and behaviors? And third, does formalization improve welfare outcomes for marginalized groups? We develop a theoretical framework clarifying (a) the causes of low service delivery; (b) the impact of formalization on political attitudes and behaviors; and (c) its consequences for welfare outcomes. To test the theory, we conduct a cluster randomized control trial in Mumbai, India. A 2014 court ruling afforded all of the city’s slum residents the right to obtain municipal water connections. Yet even though Mumbai boasts a well-functioning chlorinated supply system, uptake of connects in the slums has been slow. Partnering with a local NGO, we assess how two types of interventions—bureaucratic assistance, and political pressure in the form of group voice—shape communities’ likelihood of of securing municipal water connections, and thus formalized water supply. We then leverage these encouragements to estimate downstream impacts. This draft document pre-registers our hypotheses offers a fully coded analysis using a simulated dataset. It was filed after the baseline and randomization had been completed but before the implementation of the intervention began.

pre-analysis plan | metaketa ii

[rct] A Field Experiment on Voter Registration Among Urban Migrants in India
w/ Nikhar Gaikwad

[rct] Can Right-Wing Nationalism Be Trained?
w/ Matt Lowe