I estimate the effects of collaborative and adversarial intergroup contact. I randomly assigned Indian men from different castes to participate in cricket leagues or to serve as a control group. League players faced variation in collaborative contact, through random assignment to homogeneous-caste or mixed-caste teams, and adversarial contact, through random assignment of opponents. Collaborative contact increases cross-caste friendships and efficiency in trade, and reduces own-caste favoritism. In contrast, adversarial contact generally reduces cross-caste interaction and efficiency. League participation reduces intergroup differences, suggesting that the positive aspects of intergroup contact more than offset the negative aspects in this setting.

We document the impact of India’s COVID-19 lockdown on the food supply chain. Food arrivals in wholesale markets dropped by 69% in the three weeks following the lockdown and wholesale prices rose by 8%. Six weeks after the lockdown began, volumes and prices had fully recovered. The initial food supply shock was highly correlated with early incidence of COVID-19. We provide evidence that this correlation is due more to state-level lockdown policy variation than local responses of those in the food supply chain. Finally, during the recovery phase, the correlation between the food supply disruption and COVID-19 exposure disappeared, suggesting uniform recovery.

Why does capital not flow to developing countries as predicted by the neoclassical model? What are the direction and degree of capital misallocation across nations? We revisit these questions by removing public capital from total capital to achieve a more accurate estimate of the marginal productivity of private capital. We calculate marginal product of capital schedules in a large sample of advanced and developing countries. Our main result is that, in terms of the Lucas Paradox, private capital is allocated remarkably efficiently across nations. Tentative estimates of the marginal productivity of public capital suggest that the deadweight loss from public capital misallocation across countries can be much larger than that from private capital.

Non-family-based institutions for socializing young people may play a vital role in creating close-knit, inclusive communities. We study the potential for youth camps—integrating rituals, sports, and civics training—to strengthen intergroup cohesion. We randomly assigned Hindu and Muslim adolescent boys, from West Bengal, India, to two-week camps or to a pure control arm. To isolate mechanisms, we cross-randomized collective rituals (such as singing the national anthem, wearing uniforms, chanting support during matches, and synchronous dancing) and the intensity of intergroup contact. We find that camps reduce ingroup bias, increase willingness to interact with outgroup members, and enhance psychological well-being. Campers continue to have twice as many outgroup friends than control participants one year after the camps ended. Meanwhile, additional camp elements have heterogeneous effects: rituals have more positive impacts for the Hindu majority than the Muslim minority, while higher intergroup contact backfires among Hindus but not Muslims. Our findings demonstrate that inclusive youth camps may be a powerful tool for bridging deep social divides. Yet, we also highlight the conceptual challenges in crafting optimal integrative camps that help all groups.

Nearly all legislatures segregate politicians by party. We use seating lotteries in the Icelandic Parliament to estimate the effects of seating integration on bipartisanship. When two MPs from different parties are randomly assigned to sit together, they are roughly 0.5 to 1 percentage point more likely to vote alike. This limited peer influence is only robust in the case of voting on contested bills. In a survey of past and present MPs, most respondents doubt the possibility of peer influence. Exploring dynamics, we find that neighbor influence is temporary, disappearing the following year. These results support cue-taking and social pressure as likely mechanisms for the small effect of other-party proximity on voting. Finally, we find suggestive evidence that seating proximity builds weak ties, through co-sponsorship, despite the lack of persistent effects on voting.

We study household decision-making about female employment in India. We randomized which spouse was given a ticket enabling enrollment in a women’s weaving job, and cross-randomized the other to receive no information about the ticket, information, or information and discussion with their spouse. Consistent with a bargaining model with frictions, most experts predict information and discussion should raise enrollment. Instead, information had no effect, and discussion reduced enrollment by 50%. Negative effects are largest among couples in which the non-ticketed spouse was less supportive of female weavers, consistent with a model in which involving a spouse strengthens their veto power.

Intergroup contact is arguably the prejudice-reduction intervention with the most empirical support. However, recent meta-analyses of experimental contact interventions find signs of publication and reporting biases. In an effort to avoid such bias, I carry out a metaanalysis of 34 pre-registered contact experiments, considering only treatment effects on pre-registered primary outcomes. I find limited positive effects of intergroup contact of around one-twentieth of a standard deviation. Contact is more effective at changing behavior and attitudes towards people met than toward the outgroup as a whole. I conclude with suggestions for how contact researchers might make progress on this problem of generalization.

Social norms have been shown to facilitate anti-competitive behavior in decentralized markets. We demonstrate these norms can also reduce aggregate profits. First, we present descriptive evidence of competition-suppressing norms in Kolkata vegetable markets. We then report on a market-level experiment in which we induced a temporary relaxation of these norms by subsidizing some vendors to sell additional produce. Our intervention raised profits at the market level by over 60%, excluding the subsidy. Nevertheless, after the subsidy ended vendors largely stopped selling the additional produce. Our results suggest anti-competitive norms may partially explain the pervasiveness of small-scale firms in developing countries.

We study the impact of both the anticipation and receipt of mobile money transfers to a representative sample of low-income Ghanaians during the COVID-19 pandemic. In the short-run, the mere announcement of upcoming transfers affects neither income, consumption, labor supply, well-being, nor social distancing. Once disbursed, transfers increase contemporaneous food expenditure by 8%, income by 20%, and a social distancing index by 0.08 standard deviations. Over 40% of the transfers are spent on food. The positive effects on consumption and income do not persist to two years after the last transfer. Together, we learn that cash transfers can support households economically while also promoting adherence to public health protocols during a pandemic.

We hypothesize that broad contact, involving brief interactions with multiple outgroup members, and deep contact, meaning longer interactions with a single outgroup member, play distinct roles in shaping intergroup relations. We set up a factory in India and recruited Hindu and Muslim men to work in pairs on joint production tasks. We randomly assigned participants to work either with the same ingroup or outgroup partner daily (deep contact), a different outgroup partner each day (broad contact), or to a control group. While deep contact strengthens social and economic ties with the outgroup partner interacted with, only broad contact reduces misperceptions about outgroup strangers. These findings align with a model in which independent sampling (observing multiple outgroup members) promotes learning about outgroups more than prolonged interaction with a single individual does. Nevertheless, neither type of contact changes behavior toward the wider outgroup.

We diagnose gender gaps in a male-dominated, adversarial setting: the UK Parliament. Each week a lottery determines which politicians ask the Prime Minister a question in front of a packed and noisy chamber. We report four main findings. First, women are 12% less likely to submit questions than same-cohort men, and this gender gap has persisted since at least 1990. Second, this gender gap does not close with lottery-induced experience asking a question, or with years of parliamentary service. Third, the gender gap almost fully closes after a switch to a hybrid format in which questions are asked to a smaller, quieter, audience. Fourth, we find that the format change differentially draws in women with quieter speaking voices – consistent with the noisy environment deterring these women beforehand. Our findings support institutional change, rather than adaptation through experience, as a response to gender gaps in adversarial settings.

We study whether, and when, tweets about racial justice predict costly race-related behaviors. Individuals that tweet about racial justice are less likely to discriminate against Black individuals and more likely to donate to a civil rights organization. However, we find three pieces of evidence that higher signalling stakes reduce this informativeness. First, racial justice tweets became almost completely uninformative during the aftermath of the murder of George Floyd, when silence on racial justice became more scrutinized. Second, the informativeness of tweets is driven by low-visibility types of tweets, like retweets. Conditional on retweets, original tweets about racial justice are completely uninformative. Third, racial justice retweets are somewhat less informative of donation behavior than private statements of support for racial justice efforts. Collectively our results show that morally-charged statements on social media range from fully uninformative to highly informative, depending primarily on signalling stakes.

Cultural beliefs usually evolve slowly, but during times of religious revival, beliefs change rapidly. During the two-year Welsh Revival of 1904-5, roughly 6% of the adult population converted to Christianity, after decades of stable religiosity. This religious shock was temporary, with church membership returning to pre-Revival levels five years later. I report three main findings. First, church growth during the Revival was higher in areas with more crime and more mining industry. Second, comparing Wales with neighboring England, the Revival led to a reduction in aggregate crime by 15%. The crime reduction is concentrated among violent crime, and drunkenness, considered a major social ill at the time. Third, despite temporary effects of the Revival on church membership, effects on crime persist, suggesting an enduring shift in social norms. Collectively, these results provide support for Fogel’s theory of America’s Great Awakenings: social crisis predicts religious revival and revival brings social change.