We conduct an experiment to study whether individuals save more when information about the progress toward their self‐set savings goal is shared with another village member (a “monitor”). We develop a reputational framework to explore how a monitor's effectiveness depends on her network position. Savers who care about whether others perceive them as responsible should save more with central monitors, who more widely disseminate information, and proximate monitors, who pass information to individuals with whom the saver interacts frequently. We randomly assign monitors to savers and find that monitors on average increase savings by 36%. Consistent with the framework, more central and proximate monitors lead to larger increases in savings. Moreover, information flows through the network, with 63% of monitors telling others about the saver's progress. Fifteen months after the conclusion of the experiment, other villagers have updated their beliefs about the saver's responsibility in response to the intervention.

We develop a new class of random graph models for the statistical estimation of network formation—subgraph generated models (SUGMs). Various subgraphs—e.g. links, triangles, cliques, stars—are generated and their union results in a network. We show that SUGMs are identified and establish the consistency and asymptotic distribution of parameter estimators in empirically relevant cases. We show that a simple four-parameter SUGM matches basic patterns in empirical networks more closely than four standard models (with many more dimensions): (1) stochastic block models; (2) models with node-level unobserved heterogeneity; (3) latent space models; and (4) exponential random graphs. We illustrate the framework’s value via several applications using networks from rural India. We study whether network structure helps enforce risk-sharing and whether cross-caste interactions are more likely to be private. We also develop a new central limit theorem for correlated random variables, which is required to prove our results and is of independent interest.

We theoretically and empirically study an incomplete information model of social learning. Agents initially guess the binary state of the world after observing a private signal. In subsequent rounds, agents observe their network neighbors' previous guesses before guessing again. Agents are drawn from a mixture of learning types—Bayesian, who face incomplete information about others' types, and DeGroot, who average their neighbors' previous period guesses and follow the majority. We study (1) learning features of both types of agents in our incomplete information model; (2) what network structures lead to failures of asymptotic learning; (3) whether realistic networks exhibit such structures. We conducted lab experiments with 665 subjects in Indian villages and 350 students from ITAM in Mexico. We perform a reduced‐form analysis and then structurally estimate the mixing parameter, finding the share of Bayesian agents to be 10% and 50% in the Indian‐villager and Mexican‐student samples, respectively.

Can we identify highly central individuals in a network without collecting network data, simply by asking community members? Can seeding information via such nominated individuals lead to significantly wider diffusion than via randomly chosen people, or even respected ones? In two separate large field experiments in India, we answer both questions in the affirmative. In particular, in 521 villages in Haryana, we provided information on monthly immunization camps to either randomly selected individuals (in some villages) or to individuals nominated by villagers as people who would be good at transmitting information (in other villages). We find that the number of children vaccinated every month is 22% higher in villages in which nominees received the information. We show that people’s knowledge of who are highly central individuals and good seeds can be explained by a model in which community members simply track how often they hear gossip about others. Indeed, we find in a third data set that nominated seeds are central in a network sense, and are not just those with many friends or in powerful positions.

The DeGroot model has emerged as a credible alternative to the standard Bayesian model for studying learning on networks, offering a natural way to model naïve learning in a complex setting. One unattractive aspect of this model is the assumption that the process starts with every node in the network having a signal. We study a natural extension of the DeGroot model that can deal with sparse initial signals. We show that an agent’s social influence in this generalized DeGroot model is essentially proportional to the degree-weighted share of uninformed nodes who will hear about an event for the first time via this agent. This characterization result then allows us to relate network geometry to information aggregation. We show information aggregation preserves “wisdom” in the sense that initial signals are weighed approximately equally in a model of network formation that captures the sparsity, clustering, and small-world properties of real-world networks. We also identify an example of a network structure where essentially only the signal of a single agent is aggregated, which helps us pinpoint a condition on the network structure necessary for almost full aggregation. Simulating the modeled learning process on a set of real-world networks, we find that there is on average 22.4 percent information loss in these networks. We also explore how correlation in the location of seeds can exacerbate aggregation failure. Simulations with real-world network data show that with clustered seeding, information loss climbs to 34.4 percent.

In disseminating information, policymakers face a choice between broadcasting to everyone and informing a small number of “seeds” who then spread the message. While broadcasting maximises the initial reach of messages, we offer theoretical and experimental evidence that it need not be the best strategy. In a field experiment during the 2016 Indian demonetisation, we delivered policy information, varying three dimensions of the delivery method at the village level: initial reach (broadcasting versus seeding); whether or not we induced common knowledge of who was initially informed; and number of facts delivered. We measured three outcomes: the volume of conversations about demonetisation, knowledge of demonetisation rules, and choice quality in a strongly incentivised policy-dependent decision. On all three outcomes, under common knowledge, seeding dominates broadcasting; moreover, adding common knowledge makes seeding more effective but broadcasting less so. We interpret our results via a model of image concerns deterring engagement in social learning, and we support this interpretation with evidence on differential behaviour across ability categories.

We use unique data from over 600 Indonesian communities on what individuals know about the poverty status of others to study how network structure influences information aggregation. We develop a model of semi-Bayesian learning on networks, which we structurally estimate using within-village data. The model generates qualitative predictions about how cross-village patterns of learning relate to network structure, which we show are borne out in the data. We apply our findings to a community-based targeting program, where citizens chose households to receive aid, and show that the networks that the model predicts to be more diffusive differentially benefit from community targeting.

Social network data are often prohibitively expensive to collect, limiting empirical network research. We propose an inexpensive and feasible strategy for network elicitation using Aggregated Relational Data (ARD): responses to questions of the form "how many of your links have trait k ?" Our method uses ARD to recover parameters of a network formation model, which permits sampling from a distribution over node- or graph-level statistics. We replicate the results of two field experiments that used network data and draw similar conclusions with ARD alone.

We show that the entry of formal financial institutions can have far-reaching and long-lasting impacts on informal lending and social networks more generally. We first study the introduction of microfinance in 75 villages in Karnataka, India, 43 of which were exposed to microfinance. Using difference-in-differences, we show that networks shrank more in exposed villages. Moreover, links between households that were both unlikely to borrow from microfinance were at least as likely to disappear as links involving likely borrowers. We replicate these surprising findings in the context of a randomised controlled trial (RCT) in Hyderabad, where a microfinance institution randomly selected 52 of 104 neighbourhoods to enter first. Four years after all neighbourhoods were treated, households in early-entry neighbourhoods had credit access longer and had larger loans. We again find fewer social relationships between households in these neighbourhoods, even among those ex-ante unlikely to borrow. Because the results suggest global spillovers, atypical in usual models of network formation, we develop a new dynamic model of network formation that emphasizes chance meetings, where efforts to socialize generate a global network-level externality. Finally, we analyse informal borrowing and the sensitivity of consumption to income fluctuations. Households unlikely to take up microcredit suffer the greatest loss of informal borrowing and risk sharing, underscoring the global nature of the externality.

Can information from a credible messenger shift behavior in an information-saturated environment? In a randomized controlled trial involving twenty-eight million individuals in West Bengal, we find that SMS-delivered video messages containing information about COVID-19 symptoms and health-preserving behaviors recorded by a credible messenger increased adherence to targeted and non-targeted preventive behaviors, measured by two objective measures (symptoms reported to a health worker, and phone usage at home), as well as self-reported behaviors. We find large spillovers onto non-targeted recipients. Credible light-touch messaging can play an important role in crisis response, even when similar information is widely available.