Essays on Private Information in Village Economies

Download or read book Essays on Private Information in Village Economies written by Naoki Nishimura and published by . This book was released on 2019 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: This dissertation consists of three chapters related to informal risk sharing, social obligations, and the information asymmetries in village economies. In the first two chapters, I show that local communities do not assess the economic conditions of households solely based on per capita income or consumption, and that income-hiding seems to be a prevalent feature of Bangladeshi villages. In the third chapter, I develop a risk-sharing model with information asymmetry between the agents, and I test if the model better explains the characteristics of consumption path observed in the data from Indian villages. More specifically, the first chapter looks at how local communities assess the economic conditions of villagers. The motive for this work comes from an observation in rural Bangladesh that income measured in surveys and economic conditions as perceived by local communities do not correlate highly with each other. I consider two potential causes of this gap. The first possible cause is that local communities assess villagers' economic conditions solely based on income, but there is some noise-factor causing this mismatch. Such noise could be caused by local informants manipulating the rating procedure (sometimes referred to as elite capture), a disturbance that the local communities may not care about (such as measurement error of income or transitory income shock), or an information asymmetry between the econometrician who collected the data and the local informant. The second possible cause of the gap is that villagers think factors other than income are important. I discuss these possibilities using survey data from Bangladeshi villages, in which local informants rated and categorized the households in each village into five groups (very poor to rich). The advantage of this data set is that the data collected in these surveys was not directly used for a poverty-targeted program. Household ratings are collected for use in stratified sampling, and thus there was little incentive for the local informants to manipulate this rating (which is generally considered to be the major caveat of the community-based approaches). I find that when local communities assess villagers' economic conditions, they put different weights on each variable than those estimated by income regression. For example, even after controlling for per capita income, households with more than one family member, more highly-educated adults in their productive years (18-64 years old), household heads engaged in the service sector, more land ownership, more capital holdings, and migrant members are considered to be rich. This chapter is important because it provides an extension of the literature on the local concept of poverty. This study aims to add to that evidence by examining how local leaders and households subjectively evaluate the economic conditions of other households in the village using a data set that is free from the bias of elite capture. The second chapter looks at microfinance (MF) and income-hiding. Generally, wealthier households with alternative financial sources have been less likely to take MF loans due to the high monetary and administrative cost. Therefore, MF loans may be an appropriate tool for borrowers to signal to their neighbors that they are poor, thereby decreasing their expected contribution to public goods. I construct a costly falsification model in which MF borrowers enjoy lower income hiding costs. I use data from rural Bangladesh to test the proposition from this model that, ceteris paribus, MF borrowers hide more income than non-borrowers. I find that MF borrowers are judged to be poor by local leaders even after controlling for per capita income, land holdings, and other asset and demographic information. I also find evidence that MF borrowers increase their private consumption and disproportionally decrease their relative expenditure for social obligations like social events and religious purposes. This finding indicates that income-hiding within local communities is a prevalent phenomenon. In addition, this chapter studies the mechanism behind the impact of MF found in the literature. A series of impact evaluations of randomized controlled trials shows that MF borrowers substantially reduce their expenditure on social events (Banerjee et al., 2015a, b; Crþepon et al., 2015), and income hiding could be one possible explanation for this observed behavior. The third chapter looks at informal risk-sharing within villages. Although it is believed that villagers generally know their neighbors well, even very close relatives or friends do not know the household's exact income. I construct a dynamic risk-sharing model with two-sided private information (hereafter, called the "hidden-income model") in which two agents enter a risk-sharing contract, but each agent's own income is private information. First, I show the dynamic programming problem that derives the Pareto-efficient risk-sharing contract can be written in recursive and numerically solvable form. Using household data from rural India, I test which of the alternative model-predictions best fit the survey data: autarky, perfect risk-sharing, static and dynamic limited commitment, and the hidden income model. I find that the dynamic limited commitment model performs the best, while the private information model cannot solely explain the observed features of the consumption path. I also estimate the welfare effect of a counterfactual policy intervention to show the differences in consequences under different regimes. After the introduction of formal taxation, income-redistribution decreases welfare only in the hidden income model. This is because formal taxation increases incentives for households to hide their income. The inefficiency caused by this increased friction offsets the welfare effect of income-redistribution. This result shows that the policy effects could be quite different under different regimes, and it is important to understand the frictions within villages