Working Papers
Working Papers
We study the impact of subsidies on the adoption decision of a more complex and less affordable energy technology: biodigesters for the production of biogas. In a clustered Randomized Control Trial where we randomly assign subsidy levels for the construction of a biodigester to village clusters,we find that subsidies have a large impact on contract signing for the construction of a biodigester, but a limited impact on actual construction. We identify household characteristics associated with higher likelihood of constructing a biodigester and highlight a critical disconnect between households’ intent to adopt and their eventual decision making that subsidies alone cannot bridge. Our results suggest that subsidies should be targeted to prospective adopters that attach higher value to the technology and for whom the technology is more appropriate.
I investigate the causal impact of cell network access on health outcomes such as infant mortality rates across all of Sub-Saharan Africa using widely available datasets. Coupled with a novel procedure to adjust for measurement error induced by DHS coordinate displacement, I utilize an Irregular Terrain Model to predict network receiving power at each DHS centroid. Using the cutoff of -100 dBm for basic cell phone use, a sharp Regression Discontinuity Design produces causal estimates for cell network impact without relying on dense spatial data, rollout timing, or a valid instrument. I find suggestive evidence that cell network access decreases infant mortality rates, but estimates are subject to significant attenuation bias.
We study spouses’ differences in risk preferences, the relative influence of spouses on household decisions under risk, and the implications on household education and health investment decisions in rural Cameroon. Using an Artefactual-Field-Experiment (AFE) in which husbands and wives choose between different pairs of risky lotteries individually and then jointly as a couple, we find high levels of risk aversion amongst all participants and husbands most often influence the couple’s decision-making. Using differences in choices between spouses and couples during the AFE, we construct a novel and improved proxy for female bargaining power. Our metric is uncorrelated with investment in children’s education as well as both preventative and reactive health. We explore the relation between preferences, individual bargaining power, and social norms, and we argue the lack of a correlation between female bargaining power and investments in human capital is driven by social norms: even the most empowered women believe that most household decisions are primarily a responsibility of their husband.
Information today is largely dispersed through social media. This project investigates how academic knowledge and findings are disseminated into policy debate. Utilizing data from Twitter, MIT Media Cloud, and Congressional Hearings, we can identify the path of information dissemination, evaluate which message types are more effective, as well as the role social media influencers play.