Nearly one billion people worldwide live in rural areas without access to national paved road networks. We estimate the impacts of India’s $40 construction program using a fuzzy regression discontinuity design and comprehensive household firm census microdata. Four years after construction, main effect new feeder roads is facilitate movement workers out agriculture. However, there are no major changes agricultural outcomes, income, or assets. Employment village firms expands only slightly....

Political favoritism affects the allocation of government resources, but is it consequential for growth? Using a close election regression discontinuity design and data from India, we measure local economic impact being represented by politician in ruling party. Favoritism leads to higher private sector employment, share prices firms, increased output as measured night lights; three effects are similar economically substantive. Finally, present evidence that politicians influence firms...

We study intergenerational mobility in India. propose a new measure of upward mobility: the expected education rank child born to parents bottom half distribution. This works well under data constraints common developing countries and historical contexts. Intergenerational India has been constant low since before liberalization. Among sons, we observe rising for Scheduled Castes declining among Muslims. Daughters’ is lower than sons’, with less cross-group variation over time. A natural...

The rural poor in developing countries, once economically isolated, are increasingly being connected to outside markets. Whether these new connections crowd out or encourage educational investment is a central question. We examine the effects on choices of 115,000 roads built under India’s flagship road construction program. find that children stay school longer and perform better standardized exams. Heterogeneity treatment supports standard human capital model: enrollment increases most...

Abstract The SHRUG is an open data platform describing multidimensional socioeconomic development across 600,000 villages and towns in India. This paper presents three illustrative analyses only possible with high-resolution data. First, it confirms that nighttime lights are highly significant proxies for population, employment, per capita consumption, electrification at very local levels. However, elasticities between night-lights these variables far lower time series than cross section,...

Abstract There is a long-standing debate over whether new roads unavoidably lead to environmental damage, especially forest loss, but causal identification has been elusive. Using multiple strategies, we study the construction of rural 100,000 villages and upgrading 10,000 kilometers national highways in India. The had precisely zero effect on local deforestation. In contrast, highway upgrades caused substantial which appears be driven by increased timber demand along transportation...

Abstract We study how natural resource rents affect the selection and behavior of holders public office. Using global price shocks to thirty-one minerals nationwide geological political data from India, we show that local mineral rent cause election politicians charged with serious crimes. also find a moral hazard effect: commit more crimes accumulate greater wealth when prices rise during their terms in These have direct influence over mining operations but no access fiscal windfalls...

Measurements of mortality change among less educated Americans can be biased because the least groups (e.g., dropouts) become smaller and more negatively selected over time. We show that changes at constant education percentiles bounded with minimal assumptions. Middle-age increases non-Hispanic Whites from 1992 to 2018 are driven almost entirely by bottom 10 percent distribution. Drivers differ substantially across groups. Deaths despair explain most young Whites, but older Blacks. Our...

Abstract We study judicial in-group bias in Indian criminal courts using newly collected data on over 5 million case records from 2010–2018. After classifying gender and religious identity with a neural network, we exploit quasi-random assignment of cases to judges determine whether favor defendants similar identities themselves. In the aggregate, estimate tight zero effects based shared or religion, including settings where may be especially salient, such as when victim defendant have...

To model how known COVID-19 comorbidities affect mortality rates and the age distribution of in a large lower-middle-income country (India), to identify which health conditions drive differences with high-income countries.Modelling study.England India.Individual data were obtained from fourth round District Level Household Survey Annual Health India, aggregate for England Global Burden Disease, Risk Factors Injuries Studies.The primary outcome was modelled age-specific each due risk factor...

Abstract There are very few estimates of the age-specific infection fatality rate (IFR) SARS-CoV-2 in low- and middle-income countries. India reports second highest number infections world. We estimate IFR using data from seroprevalence surveys Mumbai (population 12 million) Karnataka 61 million), a random sample economically distressed migrants Bihar with mortality followup. Among men aged 50–89, is 0.12% (95% C.I. 0.09%–0.15%), 0.53% (0.52%–0.54%), 5.64% among (0–11.16%). approximately...

This study investigates a potential mechanism to promote coordination. With theoretical guidance using belief-based learning model, we conduct multi-period, binary-choice, and weakest-link laboratory coordination experiment the effect of gradualism - increasing required levels (stakes) contributions slowly over time rather than requiring high level contribution immediately on group performance. We randomly assign subjects three treatments: starting continuing at stake, low stake but jumping...

Objectives To estimate age-specific and sex-specific mortality risk among all SARS-CoV-2 infections in four settings India, a major lower-middle-income country to compare age trends with similar estimates high-income countries. Design Cross-sectional study. Setting multiple regions representing combined population >150 million. Participants Aggregate infection counts were drawn from large population-representative prevalence/seroprevalence surveys. Data on corresponding number of deaths...

The rural poor in developing countries, once economically isolated, are increasingly being connected to regional markets. Whether these new connections crowd out or encourage educational investment is a central question. We examine the impacts on choices of 115,000 roads built under India's flagship road construction program. find that children stay school longer and perform better standardized exams. Treatment heterogeneity supports predictions standard human capital model: enrollment...

A basic problem in applied settings is that different parameters may apply to the same model populations.We address this by proposing a method using moment trees; leveraging intuition of classification tree, our partitions covariate space into disjoint subsets and fits set moments within each subspace.We prove consistency estimator show standard rates convergence post-model selection.Monte Carlo evidence demonstrates excellent small sample performance faster-than-parametric selection step...

The rural poor in developing countries, once economically isolated, are increasingly being connected to regional markets. Whether these new connections crowd out or encourage educational investment is a central question. We examine the impacts on choices of 115,000 roads built under India's flagship road construction program. find that children stay school longer and perform better standardized exams. Treatment heterogeneity supports predictions standard human capital model: enrollment...

We study how natural resource rents affect the selection and behavior of holders public office. Using global price shocks to thirty-one minerals nationwide geological political data from India, we show that local mineral rent cause election criminal politicians. also find a moral hazard effect: politicians commit more crimes accumulate greater wealth when prices rise during their term in These have direct influence over mining operations but no access fiscal windfalls mining; thus isolate...

Political favoritism affects the allocation of government resources, but is it consequential for growth? Using a close election regression discontinuity design and data from India, we measure local economic impact being represented by politician in ruling party. Favoritism leads to higher private sector employment, share prices firms, increased output as measured night lights; three effects are similar economically substantive. Finally, present evidence that politicians influence firms...