· How the Econometric Sausage is Made summarizes the Athey-Imbens changes-in-changes approach for doing difference-in-differences for quantiles.
· In Nature Communications, Yeh et al. report on using publicly available satellite data and deep learning models to predict levels and changes in asset wealth across African villages. They compare to asset indices from the DHS, and to LSMS data, using both night and daytime satellite imagery. The results highlight the challenges of measuring changes in wealth over time – using the daytime data helps pick up things like changing agricultural land use and urbanization, but explains only 15-17% of the variation in wealth – in part because asset wealth changes slowly. Stanford News has a discussion with some of the authors.
· The Economist profiles Leonard Wantchekon and discusses the African School of Economics in Benin that he has been building: “Its aim is to offer African students the highest standards of mathematics and economics teaching, ensuring they can compete with graduates overseas. It is refreshingly drab, with no splurging on a flashy campus or needless technology. The 100 or so students pay $2,400 per year, about the same as at a public university.”
· In the tradition of seeking creative ways to disseminate research findings, Dave Evans at CGD created a Youtube “reaction video” for a paper on refugee children’s education in Kenya. The source paper is Piper et al.’s “Are Refugee Children Learning? Early Grade Literacy in a Refugee Camp in Kenya.”
· Planet Money’s The Indicator podcast has an episode this week chatting to Melissa Dell about her work on the persistence of poverty, including her explaining her work on long-term impacts of the mita system in Peru and Bolivia.
· IDinsight reports the results of its handwashing nudge interventions in the Philippines – a 25 p.p. increase in handwashing 4 months later.