In many developing countries, the average firm is small, does not grow, and has low productivity. Lack of market integration and limited information on non-local products often leave consumers unaware of the prices and quality of non-local firms. They therefore mostly buy locally, limiting firms’ potential market size (and competition). We explore this hypothesis using a natural experiment in the Kerala boat-building industry. As consumers learn more about non-local builders, high-quality builders gain market share and grow, while low-quality firms exit. Aggregate quality increases, as does labor specialization, and average production costs decrease. Finally, quality-adjusted consumer prices decline.

This paper provides the first real-world evidence of Giffen behavior, i.e., upward sloping demand. Subsidizing the prices of dietary staples for extremely poor households in two provinces of China, we find strong evidence of Giffen behavior for rice in Hunan, and weaker evidence for wheat in Gansu. The data provide new insight into the consumption behavior of the poor, who act as though maximizing utility subject to subsistence concerns. We find that their elasticity of demand depends significantly, and nonlinearly, on the severity of their poverty. Understanding this heterogeneity is important for the effective design of welfare programs for the poor.

The Effects of Environmental Regulation on Technology Diffusion: The Case of Chlorine Manufacturing by Lori D. Snyder, Nolan H. Miller and Robert N. Stavins. Published in volume 93, issue 2, pages 431-435 of American Economic Review, May 2003

We estimate the causal effects of acute fine particulate matter exposure on mortality, health care use, and medical costs among the US elderly using Medicare data. We instrument for air pollution using changes in local wind direction and develop a new approach that uses machine learning to estimate the life-years lost due to pollution exposure. Finally, we characterize treatment effect heterogeneity using both life expectancy and generic machine learning inference. Both approaches find that mortality effects are concentrated in about 25 percent of the elderly population.