MACMILLAN-CSAP WORKSHOP ON QUANTITATIVE RESEARCH METHODS
Abstract: Demand for many products is inter-linked across households in that the utility each household derives from consumption is a function of the number of friends and neighbors who also simultaneously consume the good. We develop a structural econometric model of demand under strategic complementarities, and apply it to the adoption of household latrines: a technology that holds large consequences for public health and development. We estimate the model using data from a large-scale experiment in which over 18,000 households in 380 communities in rural Bangladesh were randomly assigned incentives to purchase toilets. Complementarity effects and own price effects are identified in the model through a two-step randomization. First, we randomize the share of households receiving subsidy vouchers for latrines at the community level; then, within the community, vouchers are allocated randomly to specific households. We find that latrine purchases are strategic complements across households within a community: holding fixed the price faced by a household, the household is more likely to purchase when a greater share of other households in the community receive subsidies. Ignoring complementarities causes negative biases in estimates of both the level and the price elasticity of demand. We validate the structural model by comparing its predictions to the actual program evaluation outcomes observed in the randomized controlled trial (RCT). The model is able to replicate the RCT results reasonably accurately. We then use model estimates to conduct counter-factual simulations of the effects of subsidy programs on latrine investments when we vary the rates of baseline adoption, subsidy levels, and the dispersion of subsidies across households. The model then allows us to identify the design and allocation rule (concentrated to a few households or widely dispersed) that maximizes the spillovers and overall adoption rates, for a given subsidy budget.
Ahmed Mushfiq Mobarak, a native of Bangladesh, is a Professor of Economics at Yale University with concurrent appointments in the School of Management and in the Department of Economics. Mobarak has several ongoing research projects in Bangladesh, Brazil, Chile, India, Indonesia, Kenya and Malawi. He conducts field experiments exploring ways to induce people in developing countries to adopt technologies or behaviors that are likely to be welfare improving. He also examines the implications of scaling up development interventions that are proven effective in such trials.
This workshop series is being sponsored by the ISPS Center for the Study of American Politics and The Whitney and Betty MacMillan Center for International and Area Studies at Yale with support from the Edward J. and Dorothy Clarke Kempf Fund.