Research

Research Interests

My primary research interests include development, education, labor, public policy, and health economics.

My research covers three important areas in development economics. The first area concerns evaluating government policies designed to uplift the impoverished and disenfranchised in developing countries. The second area revolves around the relationship between education, poverty and nutrition in developing and less developed countries. The third area involves understanding poverty, malnutrition, and ways to eradicate them. I use experimental and quasi-experimental econometric techniques to obtain estimates with causal interpretations to carry out my research. I also have experience in conducting surveys in developing countries and working with large scale observational and administrative data sets.

Publications in Refereed Journals

1. The Redistributive Effects of Political Reservation for Minorities: Evidence from India
(with Aimee Chin, University of Houston, NBER, IZA), Journal of Development Economics, vol 96 (2), Nov 2011, pp 265--277
        o Published Version
        o CReAM DP No. 03/10
        o IZA DP No. 4391

2. Do School Lunch Subsidies Change the Dietary Patterns of Children from Low-Income Households?
(with Larry Howard, California State University- Fullerton), Contemporary Economic Policy, vol 30 (3), Jul 2012, pp 362--381
        o Published Version
        o IZA DP No. 4427
        o Pre-Publication Version

3. Do Employment Quotas Explain the Occupational Choices of Disadvantaged Minorities in India?
(with Larry Howard, California State University- Fullerton), International Review of Applied Economics, vol 26 (4), Jul 2012, pp 489--513
        o IZA DP No. 5894
        o Pre-Publication Version

4. The Returns to English-Language Skills in India
(with Mehtabul Azam, World Bank & IZA and Aimee Chin, University of Houston, NBER & IZA), Economic Development and Cultural Change, vol. 61 (2), January 2013, pp 335--367
        o IZA DP No. 4802
        o CReAM DP No. 02/10
        o Pre-Publication Version

5.
Consumption and Social Identity: Evidence from India
(with Melanie Khamis, Wesleyan University & IZA and Zahra Siddique, IZA),
Journal of Economic Behavior & Organization, vol. 83 (3), August 2012, pp 353--371
        o IZA DP No. 5406
        o Pre-Publication Version

6. Skill Mapping and Skill Development for Employability: The Case of Cuttack
(with Satpathy, Anoop K.; Mishra, Jimuta), Indian Journal of Labor Economics, vol. 45 (4), 2002 *PDF*


Working Papers
  1. A Distributional Analysis of Public Private Wage Differential in India, (with Mehtabul Azam, World Bank & IZA), IZA DP No. 5132 *PDF*, Updated Version- May, 6, 2012
  2. Improving the Labor Market Outcomes of Minorities: The Role of Employment Quota *PDF* Dissertation Chapter
  3. Indirect Effects of Affirmative Action Policy: How Does Employment Reservation Affects Child Outcomes?, (with Tauhidur Rahman, University of Arizona)
  4. Education Policies and Practices: What have we Learnt so far and the Road Ahead?, (with Priya Ranjan, University of California - Irvine) [funded by International Growth Center Country Programme: India - Bihar] *PDF*
      • Presented at IGC Growth Week at London School of Economics, September 19, 2011 Slides
      • Presented at IGC Bihar Growth conference, Patna, India, December 14, 2011 Slides
  5. Women Reservation in Bihar and Children’s Health Outcomes Project, (with Santosh Kumar, University of Washington - Seattle) [funded by International Growth Center Country Programme: India - Bihar] *PDF*
  6. Cycling to School: Increasing Secondary School Enrollment for Girls in India, (with Karthik Muralidharan, UC- San Diego, NBER, BREAD, CEGA, J-PAL & IPA) [funded by International Growth Center Country Programme: India - Bihar]

Work in Progress

  1. The Behavioral Effects of Estate Tax: Evidence from Natural Experiment in United States, (with Edward Hoang, University of Colorado at Colorado Springs and Stephen Ross, University of Connecticut)
  2. The Role of Constitutions on Poverty: A Cross-National Investigation, (with Alanson Minkler, University of Connecticut) 
  3. Does Political Reservation for Minorities affect Child Labor? Evidence from India, (Elizabeth Kaletski, University of Connecticut)
  4. Livestock Insurance and Credit: Evidence from the Index Based Livestock Insurance Program in Northern Kenya (with Christopher B. Barrett, Cornell University; 
    Edward Hoang, University of Colorado at Colorado Springs and Sommarat Chantarat, The Australian National University)
  5. Do Criminal Representatives Hinder or Improve Public Good Outcomes for their Constituency? Evidence from Indian States, (with Yogesh Uppal, Youngstown State University and Marc Rockmore, Clark University)
  6. The Quantity-Quality Trade-off in Educational Outcomes: Evidence from The Right to Education Act in India, (with Elizabeth Kaletski, University of Connecticut and Abhiroop Mukhopadhyay, Indian Statistical Institute)
  7. Political Change and Crime Reduction in Bihar, (with Oliver Vanden Eynde, Paris School of Economics and Clement Imbert, Oxford University)
  8. Girls for Sale? Son Preference and Girls Trafficking in India, (with Krishna C. Vadlamannati, Norwegian University of Science and Technology)
      • Presented at Human Rights Institute at University of Connecticut, February 3, 2014 Slides

Current Fieldwork

Recent years have seen an increased interest in the economic and non-economic impacts of affirmative action policies across the social sciences, with a majority of the work focusing on developed economies.  Unresolved issues relating to the effective design of such policy are how to minimize stigma effects and stereotyping effects, at what ages interventions should take place, whether these policies carry disincentives for skill investment and if there are any effects on the majority. Affirmative action policies have been adopted in a larger scale in India than elsewhere, with a nation-wide programme of reservation(quotas)of new jobs, political seats and slots in higher educational institutions for the historically discriminated lowest-caste groups in Indian society.  Yet, surprisingly little evidence exists on the causal effects of reservation, in particular, whether such policies better the outcomes of the intended groups more than they would have in the absence of such programmes. The aim of the research projects of this group is to bring reliable empirical evidence on the economic and behavioral implications of anti-discrimination policies in employment and education, across both children and adults, to better understand the linkages of policies, caste and education and labor market outcomes across the life-cycle, and to find ways to erase persistent inter-group inequalities. The findings are also expected to shed light on current Danish and European research on increasing social segmentation, polarization and outgroup discrimination due to migration.

Over the past year, ILRI in collaboration with various partners has pursued a comprehensive research agenda aimed at designing, developing and implementing market mediated index-based insurance products to protect livestock keepers from drought related asset losses they face, particularly those in the drought prone Arid and Semi Arid Lands(ASAL). For pastoralists whose livelihoods rely solely or partly on livestock, the resulting high livestock mortality rate has devastating effects on asset levels, rendering them amongst vulnerable populations in Kenya.

Index-based insurance products represent a promising and exciting innovation that could allow the benefits of insurance to protect the climate-related risks that vulnerable rural smallholder farmers and livestock keepers face. Because index insurance is based on the realization of an outcome that cannot be influenced by insurers or policy holders (such as the amount and distribution of rainfall over a season), it has relatively simple and transparent structure. This makes such products easier to administer and consequently to more cost-effective to develop, and trade. Indeed the success of several pilot programs conducted in India, and various countries in Africa and Latin America, have proven the feasibility and affordability of such products.

Much of the initial phase of the project, which included an extensive program of field work and stakeholder consultation, is now complete. The research has generated useful insights that have been used in the design of index-based livestock insurance(IBLI) products that is better targeted to the various needs of the expected clientele. Currently, an IBLI contract has been modeled, priced, tested among the target clientele and is now ready for implementation. ILRI in collaboration with partners from the public, private and non-profit sectors now plans to pilot IBLI contracts for the long rain/long dry season scanning March 2010 to September 2010 in Marsabit district.

  • POLITICAL CHANGE AND CRIME REDUCTION IN BIHAR
  • Data collection in progress 2014 -- 2016

This project aims to understand how Bihar has improved its law and order situation after 2005. We aim to collect police station level data on crime outcomes to shed light on the mechanisms that explain Bihar’s remarkable drop in violent crime, which is credited with laying the foundation for the state’s remarkable economic growth.

Since 2005, Bihar has experienced a “remarkable recovery” (BBC, 2010). Bihar’s GDP growth, which was below the national average for decades, soared to 11% per year between 2005 and 2012, one of the fastest growth rates among Indian states. Strikingly, Bihar’s turnaround was marked by a dramatic and fast improvement in law and order: road robbery decreased by 15% between 2004 and 2008, murders dropped by 6% and kidnapping went down by 37% over the same period. Bihar’s unexpected recovery was extensively discussed in the international media. The international press and domestic policy makers alike have cited the return of law and order as a major cause of Bihar’s recently improved economic fortunes.

Our project aims to understand the mechanisms through which Bihar was able to leave its history of poor law an order behind, and to achieve high level of economic growth. While the reduction in violent crime after 2005 is clearly marked in the State level crime statistics, there has been no in-depth academic study of the mechanisms through which Bihar’s political change enabled this swift improvement. Interestingly, the drop in crime that Bihar experienced after 2005 was not evenly spread across Bihar (State Crime Records Bureau, 2012). This project aims to collect a unique data set of police station level crime data, to uncover sub-district level variation in crime outcomes. This will allow us to identify which policy changes can account for local differences in crime reduction. Finally, local variation in crime trends will also allow us to test whether starker drops in crime translated into better economic outcomes. The effect of crime on the local economy will be measured using the 2012 economic census and data on land prices, depending on availability.

Our study of Bihar’s improvement of law and order adds to a growing economic literature on the effectiveness of crime reduction strategies. We aim to explore channels that have remained understudied but could be very powerful. These channels relate to the efficiency of the judicial process, the management of the police force, the political pressures on police, and employment generation on public works. Our work on Bihar holds the promise of identifying mechanisms that are relevant for the many countries and territories that suffer from the double curse of low economic growth and poor law and order.


Concluded Fieldwork

The objective of this study is to measure the causal effect of Bihar Government’s move to reserve 50 per cent of the panchayat seats for women - thereby leading to women representatives occupying 54 per cent of the seats - on children’s health outcomes. Several studies demonstrate that men and women differ in their political and policy preferences (Edlund and Pande, 2002; Miller, 2008). Also a number of studies find that increased female representation in politics is associated with significant changes in policy-making (Chattopadhyay and Duflo, 2004; Munshi and Rosenzweig, 2008; Figueras, 2009; Rehavi, 2008; Powley, 2007). The "Women Reservation in Bihar" would directly contribute to the idea of "inclusive growth" if they have impact on policy outcomes that puts the economy on higher growth trajectory. A policy like this will contribute to economic growth of Bihar by several channels. For example, if this policy leads to a more investment in education and health then this will directly improve the human capital of population. Healthy and educated population are more likely to productive and efficient, and thereby contributing to the growth of the economy. Women Reservation Policy will also contribute to "inclusive growth" by allowing traditionally disadvantaged groups (women) to participate more directly in the growth of the Indian economy through increased empowerment and voice.

Women are underrepresented in political leadership positions throughout the world. In July 2008, women accounted for 18.4% of parliamentarians worldwide. Recognizing this gender gap in political participation and decision making, the Indian Government mandated 33% reservation of panchayat seats for women. The reservation bill was enacted in 1993 by bringing in 73rd amendment, paving the way for election of around 1 million elected women at village, block and district level. While many Indian states reserved 33% of panchayat seats for women, the state of Bihar took the bold step of reserving 50% of the panchayat seats for them. Currently elected women representative occupy 54% of seats in Bihar panchayat. By now Bihar have completed at least one five year term. What has been the impact of women leadership on women and children health-being is the objective of this study? This study examines the impact of female leadership on children and women health outcomes in Bihar, specifically examining the effect on survival probability of children by looking at its impact of neonatal, infant, and child mortality outcomes. Additionally, in order to pin point the pathways, the project will also investigate the impact of women leadership on antenatal care, institutional delivery, and immunization status of children.

In order to estimate the causal impact of female leadership, we will exploit the variation in program intensity (number of women leaders per 100,000 children) across districts. Our identification is based on the premise that children born in districts with more exposure to women leaders will benefit more compared to children born in districts with low exposure to women policymakers. In addition, we will also exploit the time-variation variation in the policy across Bihar and Jharkhand, an erstwhile part of Bihar, through using differences-in-differences estimation strategy. The policy data will be collected from election commission office in Bihar, and second and third waves of Reproductive and Child Health (RCH) survey will be used for the outcome variables.

Despite the importance of female leadership on a range of developmental outcomes, little is known about the relative performance of women as policy makers, about their impact on child development. Given the limited evidence on the impact of gender quota in political leadership, the results of this research will inform the policymakers to evaluate the benefits of women quota. Since, many other states are currently debating to emulate similar policy; the results of this study will be of tremendous policy interest to examine the broader impact of this policy.

The "Mukhyamantri Balika Cycle Yojna" (Chief Minister's Bicycle Program for secondary school girls) has been one of the flagship programs of the Govt. of Bihar in the past 5 years to improve secondary school enrollment among girls. The objective of this research work is to estimate the impact of this policy experiment on girl's educational outcomes.

Improving female education directly contributes to inclusive growth through both (a) the direct channel of improved human capital of female participants in the labour force, and (b) the indirect channel of improved human capital of the next generation (several studies have shown high-levels of inter-generational transmission of human capital from mothers to children). It would also contribute to "inclusive" growth by allowing traditionally disadvantaged groups (women) to participate more directly in the growth of the Indian economy through increased employment opportunities, by improving their intra-household bargaining position, and by providing them the 'capabilities' (in Amartya Sen's terminology) to live more meaningful lives in an increasingly complex world where the returns to education appear to be increasing.

Hence, a fundamental policy question in most developing countries (including India) is to identify cost-effective and scalable policies that improve school attainment of girls (an objective that addresses multiple Millennium Development Goals). While hundreds of schemes are launched as pilots and then discarded when the concerned officer (or government) changes, the bicycle program is one that has caught the imagination of voters as well as political leaders and its high visibility has led to heightened interest in replicating the policy across India.

So far the Government of Bihar has already spent Rs. 174.36 crores on this program in the past 3 years, but there is no reliable estimate of the impact of the program on key outcome variables or any attempt to estimate a social rate of return on this investment. A careful impact evaluation of the bicycle program is therefore a piece of analytical work that can directly feed into the policy discussions in Bihar and across India on whether programs like the "bicycle program" should be scaled up. Specifically, in this project we will measure the impact of the program on enrollment, drop-out, attendance, and school progression of adolescent girls.

The purpose of this project is to survey the literature on the effectiveness of education policies adopted in different parts of the world in improving the quantity and quality of education. We also propose to survey the policies adopted by the government of Bihar towards promoting educational outcomes in the state. By placing these policies appropriately in our broader survey framework, we aim to make this work a contextual survey.

The survey will cover various policies adopted by governments around the world to improve educational outcomes. It will start by discussing policies that increase enrollment in schools. In this context we will examine how policies that reduce the cost of schooling affect enrollment. Some of the policies that fall in this category are conditional cash transfers, merit scholarships, information provision regarding the returns to schooling, school based health programs. We will also study the policies that Bihar government has in place to reduce the cost of schooling. The recent program of the government providing bicycles to high school students will fall under this category. In addition, there are several scholarship programs run by the government that we will be looking at.


Next, we will look at policies aiming to improve school quality or learning outcomes. Programs that fall under this category include increasing schooling inputs (both teacher input and non-teacher input), pedagogical changes, incentive pay for teachers, and incentive for students to perform well. An example of incentivizing students is a recent program in Kenya that offered a small scholarship for the next year to girls who performed in the top 15% on an exam not only improved the performance of girls, but also of boys even though there were no scholarships for boys. Bihar government’s recent effort to hire contract teachers will fall under the category of increasing teacher input as well as the initiatives of Pratham (an educational NGO) to train volunteers to act as teacher’s assistants in government schools in Bihar and its remedial summer camps for school children where teachers from government schools were invited to teach. Currently, no state in India has adopted a policy of incentive pay for teachers; however, we are aware of some experimental studies done in India to see the effectiveness of teacher incentives. We will be surveying these studies as well as studies done on other countries. We will be studying the implementation architecture of the teacher incentive pay programs that have been adopted in some countries and some states in the United States with the aim of providing a guideline to the government of Bihar if it were to go down this road. Since many of the teacher incentive programs link incentives to students’ performance on some standardized tests, the first step will involve setting up a framework to measure student performance through standardized tests.


With the support of Government of Bihar and IGC-Bihar we aim to draw some policy lessons to benefit the policy makers in Bihar with regard to effective educational policy initiatives.