Diane and I are in New England for some presentations and meetings While we were close to the Hebron town hall, we took the chance to turn in our absentee ballots for the fall elections. Naturally, this prompted me to link back to the three times that what Parfit calls “moral mathematics” has appeared on our blog in some form.
One thing I think Parfit’s discussion gets exactly right is the occasional importance of tiny probabilities, when attached to a very large cost or benefit. So, voting may be an obligation if the welfare consequences of one winner are very different from another (or if the vote total has a large signaling effect), even if the probabilities involved are small. By the same logic, it may be a very good idea to work hard to have a small chance of changing the rate of decline of open defecation in India (hint: do a latrine use survey).
On an unrelated note, today the New York Times has a well-deserved blog post on the excellent ASER survey of child learning, the same dataset that Sneha Lamba and I use in our paper on cognitive consequences of early life exposure to poor sanitation.