Syndicate content

Add new comment

Weekly links February 29: migration as development, Heckman on the extreme failure of early experiments, generalizing the family size instrument, and more…

David McKenzie's picture
  • Working paper of the week by Clemens and Pritchett on whether there is a new economic case for migration restrictions – I like this statement comparing the ultra-poor programs to migration “A two-year, six-component in situ intervention—guided by some of the top minds in development economics and backed by formidable financial and organizational resources—produced the equivalent annual consumption gain of the wage differentials of working in a rich versus poor country for one day.”
  • Those migration gains persist over time – as shown in this VoxEU post on the long-term impacts of migration, based on a 10-year follow-up of a visa lottery New Zealand runs for Tongans.
  • On Econtalk, Russ Roberts interviews Heckman. Interesting discussion about the early big experiments in the US: “I remember as a graduate student at Princeton University, it rolling some of the very first participants in the negative income tax experiment…actually one of the great legacies of the negative income tax studies was actually modern econometrics or micro-econometrics... Precisely because the experiments were so messed up and people did not understand when they were designing the experiments how much choice there was. How much attrition[?] there would be, how much individuals would respond to incentives in ways that weren't even thought about. So, the first round of experiments was generally viewed as a failure. I think John Cogan's testimony before Congress in the late 1970s or early 1980s was the capstone of that failure in the sense that he pointed out all of the variety of estimates and the need for using econometric estimates to adjust for the non-compliance, the self-selection, the nonresponse, and on and on and on…Meanwhile, the faithful continued....And so there's then a constant faith, despite this surge around [?] what I would call extreme failure. Nobody believed the New Jersey income tax experiments; and in the later Seattle experiments--nobody believed them because they were so heavily compromised by a whole set of other issues.”. (also worth reading for his views on how substitution bias compromises a lot of the more recent experiments on early childhood education).
  • From the Economist: the impact of paying higher wages to police in Ghana was more corruption