I was thrilled that Angus Deaton won the Nobel prize. Nobody in development economics had won in a very long time, and Deaton’s prize was very well-deserved. An embarrassing personal anecdote should help illustrate: back when I was applying to grad school, I was astonished to find that he was still at Princeton because I assumed that with the large amount of work he had put out, he must be a mythical figure long since deceased.
Deaton won for his work on poverty and welfare, as others have highlighted (see, for example, Chris Blattman or Tyler Cowen’s excellent summaries).
More recently, he has gotten a lot of attention for his skepticism of randomized controlled trials. I am very familiar with the debate, especially due to my controversial paper which found a lot of heterogeneity in treatment effects – essentially bringing a lot of data to the debate.
I think that the point that Deaton was making was more nuanced than how it is often described and applies to quasi-experimental studies as well as randomized controlled trials: that we should simply be more modest in our claims about what we know.
This is very honest, and I think in the long run we will look back and wonder why there was ever any controversy over such a claim.