Gulzar, Saad and Pasquale, Benjamin J. (2017) “Politicians, Bureaucrats, and Development: Evidence from India”,
          American Political Science Review, 111(1), pp. 162–183. (pre-print) (+)

– “Too many politicians spoiling the babu’s broth” Blogpost in Mint
– “When does politics work for development?” Blogpost on the paper in Ideas for India
– Inaugural Best Paper Award by the Quality of Government Institute at Gothenburg University, 2017
– Best Paper Award at UCLA Graduate Conference in Comparative Politics, 2015

Abstract: When do politicians prompt bureaucrats to provide effective services? Leveraging the uneven overlap of jurisdictions in India, we compare bureaucrats supervised by a single political principal with those supervised by multiple politicians. With an original dataset of nearly half a million villages, we find that implementation of India’s National Rural Employment Guarantee Scheme, the largest employment program in the world, is substantially better where bureaucrats answer to a single politician. Regression discontinuity estimates help increase confidence that this result is llenusal. Our findings suggest that politicians face strong incentives to motivate bureaucrats as long as they internalize the benefits from doing so. In contrast to a large literature on the deleterious effects of political interventions, our results show that political influence may be more favorable to development than is commonly assumed. 

Working Papers

Can Political Alignment be Costly?” Revised and Resubmitted, Journal of Politics
        with Michael Callen (UCSD), and Arman Rezaee (UC Davis) (+)

– Best Paper Award at George Mason University Graduate Conference, 2014clinics

Abstract: Research on the benefits of political alignment suggests that voters who elect governing party politicians are better off than those who elect other politicians. I examine this claim with a regression discontinuity design that isolates the effect of electing a governing party politician on an important publicly provided service in Pakistan: health. Consistent with existing research, governing party constituents receive a higher quantity of services: more experienced doctors are assigned to work in governing party areas. However, these doctors also report more connections with politicians and are significantly more absent from work. That is, a higher provision of healthcare is accompanied by impeded access and diminished quality in governing party areas. These findings contrast with the literature on political alignment by showing that alignment to the governing party affects voters’ welfare ambiguously: higher quantity of services may come at the cost of lower quality

Why Do Citizens Become Politicians? Experimental Evidence on the Social Dimensions of Candidacy
          with Muhammad Yasir Khan (UC Berkeley) (+)

saad_field– Funded by J-PAL Governance Initiative and International Growth Center
– Project Summary at JPAL
– AEA RCT Registry Entry
– EGAP RCT Registry Entry

Abstract: What motivates people to seek political office? What role does the social context play in the candidacy decision? In a field experiment, we increase the salience of personal reasons for running for political office — such as gaining respect and status — or social reasons for running — such as the ability to help others. We do this by making appeals to a random subset of the community  in one-on-one private meetings or in village-level public meetings. Our results show that,  first, making any appeal to run greatly increases the probability of candidacy. Second, compared to a neutral private or public meeting, where personal or social reasons are not mentioned, highlighting social reasons to run increases candidacy, while making personal reasons salient reduces candidacy. Significantly, these effects are only observed when appeals are made in a public meeting, leading to the conclusion that the social dimensions of the candidacy decision are particularly important. 

Data and Policy Decisions: Experimental Evidence from Pakistan” (under review)
          with Michael Callen (UCSD), Ali Hasanain (LUMS), Muhammad Yasir Khan (UC Berkeley), and Arman Rezaee (UC Davis) (+)

Coverage in The Economistdashboard
Featured Story on the World Bank website
Coverage on World Bank’s Governance for Development Blog
Project Summary at International Growth Center
Coverage in Herald Magazine
– Best Paper Award at NYU Graduate Political Economy Conference, 2013

We evaluate a program in Pakistan that equips government health inspectors with a smartphone app which channels data on rural clinics to senior policy makers. Data are transmitted in real-time, and submitted reports require timestamps, geostamps, and photos. The system led to rural clinics being inspected 74% more often. In addition, we test whether senior officials act on the information provided by the system. We find that highlighting poorly performing facilities raises doctor attendance by 18.1 percentage points from a dismal base of 24.5%. Our results indicate that technology can mobilize data to real effect, even in low capacity settings.

Combines two working papers:

The Political Economy of Public Sector Absence“, NBER Working Paper No. 22340

– Blog post on the paper at Vox EU
Paper Summary at JPAL

Abstract: Public sector absenteeism undermines service delivery in many developing countries. We report results from an at-scale randomized control evaluation in Punjab, Pakistan of a reform designed to address this problem. The reform affects healthcare for 100 million citizens across 297 political constituencies.  It equips government inspectors with a smartphone monitoring system and leads to a 7% increase in inspections. However, the surge in inspections does not always translate into increased doctor attendance. The scale of the experiment permits an investigation into the mechanisms underlying this result. We find that experimentally increasing the salience of doctor absence when communicating  inspection reports to senior policymakers improves subsequent doctor attendance. Next, we find that both the reform and the communication of information to senior officials are more impactful in politically competitive constituencies. Our results suggest that interactions between politicians and bureaucrats might play a critical role in shaping the success or failure of reforms.

Personalities and Public Sector Performance: Evidence from a Health Experiment in Pakistan“, NBER Working Paper No. 21180 | BREAD Working Paper No. 448

– Paper summary at CEGA, University of California
Project Brief at IGC
Coverage in The Guardian
Blog post at IDEAS

Abstract: This paper provides evidence that the personalities of policymakers matter for policy. Three results support the relevance of personalities for policy. First, doctors with higher Big Five and Perry Public Sector Motivation scores attend work more and falsify inspection reports less. Second, health inspectors who score higher on these measures exhibit larger treatment responses to increased monitoring. Last, senior health officials with higher personality scores respond more to data on staff absence by compelling better subsequent attendance. These results suggest that interpersonal differences matter are consequential for state performance.

Political Identity: Experimental Evidence on Anti-Americanism in Pakistan“, NBER Working Paper No. 20153  (under review)
          with Leonardo Bursztyn (UChicago), Michael Callen (UCSD), Bruno Ferman (FGV), Ali Hasanain (LUMS) & Noam Yuchtman (Berkeley) (+)

Abstract: peshWe identify Pakistani men’s willingness to pay to preserve their anti-American identity using an experiment imposing clearly-specified financial costs on anti-American expression, with minimal consequential or social considerations. Around one-quarter of subjects forgo payments from the U.S. government worth around one-fifth of a day’s wage to avoid an identity-threatening choice: anonymously checking a box indicating gratitude toward the U.S. government. When subjects anticipate that rejection will be observable, rejection falls, suggesting that pressure to conform outweighs the need to publicly signal one’s identity. A second experiment correlates rejection of the U.S. payment with membership in Pakistan’s major anti-American political party. 

Ballot Order Effects and Party Responses: Evidence from Lotteries in Colombiawith Nelson A. Ruiz (LSE) (under review)  (+)

Abstract: Ballot positions often affect the electoral performance of parties. Existing theories for this effect focus on voter-specific behavioral explanations. We present evidence on an additional unexplored mechanism: parties adjust their behavior to account for their position on the ballot. This adjustment matters for how we interpret existing results. First, we use a constituency level lottery of ballot positions across two elections and 1971 races in Colombia. We find evidence for a ballot position effect of 6.7 percent on vote share, and 8.8 percent on seat share. Second, we show that parties raise and spend more money on their campaigns if they are allocated a top spot on the ballot. Finally, we show that campaign spending is correlated with higher vote shares. Our results suggest that the existing literature may be overstating the contribution of the voter behavioral channel on ballot order effects. 

“Living in Ungoverned Space: Pakistan’s Frontier Crimes Regulation”
          with Michael Callen (UCSD), Arman Rezaee (UCSD) and Jacob Shapiro (Princeton) (+)

fcr_over_timeAbstract: Why do substantial swathes of territory within the boundaries of administratively competent sovereign states remain ungoverned for long periods of time? We explore this question in the context of a unique set of legal institutions in Pakistan that clearly demarcate spaces that are to be left ungoverned. During colonial rule, the British divided Pakistan into two distinct regions. The first was the Raj, where the British built modern political and bureaucratic institutions. In the second region, the British put a small number of political agents in charge of tribal areas and codified pre-colonial institutions in the Frontier Crimes Regulation (FCR). Legal decisions were left to customary law carried out by local tribal councils, or jirgas. Though the area under FCR has steadily decreased, FCR is still in place in the tribal areas of Pakistan today. Pakistan therefore offers a prime case in why governments leave certain territory ungoverned. Using primary legal documents we create a dataset of when and where FCR applied in Pakistan between 1901 and 2012 at the sub-district level. We then exploit the differential impact of the Green Revolution on potential land revenue at the sub-district level to empirically model the choice to leave territory ungoverned. We find that sub districts that we would see a disproportionate increase in potential land revenue as a result of the Green Revolution are disproportionately more likely to have FCR removed following the advent of the Green Revolution. 

“Official Attendance Data for Politically Connected Bureaucrats Are Less Accurate” 
          with Michael Callen (UCSD), Ali Hasanain (LUMS), and Muhammad Yasir Khan (UC Berkeley) (+)

Abstract: Research shows that official data can often deviate from the truth. This paper shows that the absence of bureaucrats is underreported when they areofficial_data politically connected. We compare absence reports in the universe of government inspections of public clinics in Punjab, Pakistan (N=79,318), with independent unannounced inspections of a representative sample of 850 clinics. We present robust evidence that government data underreport doctor and staff absence by up to 12.9 percentage points. Importantly, we show that doctors who personally know the local politician are less likely to be reported absent in official data. Our results signal caution in the use of official data as incentives to misrepresent data may be correlated with political objectives.

Manuscripts in Progress

“Marginal Politicians: Experimental Evidence on Candidacy” with Muhammad Yasir Khan (UC Berkeley)


– Funded by J-PAL Governance Initiative and International Growth Center
AEA RCT Registry Entry
EGAP RCT Registry Entry

Abstract: We design an experiment to understand what binds the choice of ‘marginal politicians’ to contest elections. This is a group whose decision is most likely to be affected by new policy reform. We ask a random subset of citizens in 192 villages, in Khyber Pakhtunkwa province of Pakistan, to identify people they think should run, but are undecided on the margin. This exercise yields a list of around 2000 people from 192 villages. We randomize, at the individual level, a combination of treatments to this group to study which aspect of their decision calculus prevents them from contesting elections. We make salient personal or pro-social benefits from office, offer the services of a lawyer to help file papers, and finally, inform subjects that they have been nominated by someone in the village to contest elections. Our results show that cost is a significant factor that prevents candidacy, however, benefits from office are more likely to improve the chances of election for those already contesting.

“The Political Economy of Electoral Quotas: Evidence from India” with Benjamin Pasquale (Independent Researcher and Owner Bar Uni)  (+)


Abstract: We study the impact of a large political quota on development consequences for the quota minority, the non-quota minority, as well as the general population. This quota is active in states representing half of India’s population and sets aside all local leadership positions for the historically disadvantaged Scheduled Tribe (ST) population. We build a new dataset, spanning nearly a quarter million villages, which combines census characteristics with outcome data on the National Rural Employment Guarantee Scheme, the world’s largest development program. We find that while the quota increases development program implementation by more than 60 percent among the quota minority (ST), the non-quota minority communities see no improvement. Regression discontinuity estimates increase confidence that our results are causal. While our analysis provides no evidence of positive spillovers of electoral quotas, we find that quota politicians are primarily motivated by electoral concerns.

Work in Progress

“A Field Experiment on Training First-Time Village Politicians in Pakistan’with Muhammad Yasir Khan (UC Berkeley)  (+)

– Funded by International Growth Center and JPAL Governance Initiative

Project Description: The Khyber-Pakhtunkwa province in Pakistan introduced a new local government system in 2015, electing village councils for the first time in Pakistan’s history. In this project we evaluate the impacts of training village-level politicians on leadership and negotiation skills – soft skills that are crucial for the success of these councils. First, the role of village councils is not defined perfectly, which leaves space for village councils to exert effort to have services and project money released to them by higher level governments. Second, first time politicians operating under a new legal framework are poorly informed about how to conduct the business of the state. Our training aims to release constraints on their legal requirements by making that process easier. Our outcomes will focus on the ability of these councils to deliver projects and services to villagers. Findings from this project will help understand how the capacity constraint at the local level can be bridged using in-office trainings.

“Monitoring Public Works with Smartphones: A Field Experiment with the Sindh Government, Pakistan” with Imran Rasul (UCL), and Dan Rogger (World Bank)  (+)

– Funded by World Bank’s i2i Fund

Project Description: Public works projects, such as roads, schools and health facilities, account for a majority of government development spending in Sindh, Pakistan. Poor construction quality in these projects leads to large amounts of resource leakage, both due to passive waste, such as unavailability of data for senior decision-makers, and active waste, such as collusion between government monitors and construction inspectors. This experiment comprises two parts: first, we randomly allocate smartphone facilitated visits to public works projects that reduce the ability of contractors and implementing agencies to collude (through time stamps, and geo-tagged pictures); and second, experimentally manipulate and release data on public works to bureaucrats, citizens and politicians. The objective is to test: 1) the effect of improved monitoring in improving efficiency of public spending; 2) the effect of information on accountability and efficiency.

“Improving Political Communication: A Field Experiment in Pakistan” with Miriam Golden (UCLA) and Luke Sonnet (UCLA) (+)

– Funded by JPAL Governance Initiative, International Growth Center and Empirical Studies of Conflict
– Pilot report for IGC

Project Description: Elections are a blunt tool for accountability. Can enhanced politician-voter communication in the periods between elections improve democratic outcomes? We partner with a politician in Pakistan and design an experiment with Interactive Voice Response (IVR) – a technology that enables him to robocall a large number of voters in his own voice to ask them questions and receive feedback. We randomize whether respondents receive a call soliciting preferences about upcoming decisions the politician must make. A follow-up call randomizes how responsive the politician is to voters’ preferences. We study the effect of this communication on voters’ evaluations of the incumbent, their views on government performance, and the likelihood that they will hold the politician accountable. 

“Decentralization, Candidate Selection, and Post-Disaster Stability in Nepal” with Michael Callen (UCSD), Rohini Pande (Harvard), and Soledad Prillaman (Oxford) (+)

– Funded by International Growth Center and Empirical Studies of Conflict, and Innovations for Poverty Action

Project Description: In May of 2017 Nepal started a massive process of political and fiscal decentralization, holding its first local elections in 20 years. It is widely hoped that this process will preserve the ongoing and delicate peace process following Nepal’s civil unrest over the last decade and result in more effective reconstruction from the 2015 earthquake. We study decentralization in Nepal,
focusing on the selection and training of newly elected representatives. We will combine a recent census from the fourteen districts most impacted by the earthquake with administrative data on election
candidates and winners and internal party lists on potential candidates to identify (1) who becomes a local politician and (2) the role of parties in the process of cultivating candidates. We will then conduct a
series of capacity-building interventions with newly elected representatives to examine whether this increases political stability and improves post-conflict reconstruction and service delivery.