Papers using happiness, life satisfaction or “Where do you stand on the ladder of life?” data (“self-reported well-being data”) make strong assumptions about how that data relates to theoretical utility notions. In this paper, my coauthors Dan Benjamin, Mark Fleurbaey, Jakina Debnam Guzman, Ori Heffetz and I bend over backwards to make that OK, but it just isn’t. Different authors either explicitly or implicitly assume self-reported well-being data like this corresponds to different, mutually incompatible utility notions. For example some assume it represents flow utility while others assume the very same data represents lifetime utility. And data we collected about how survey respondents interpret these self-reported well-being questions indicates that the respondents include a little of everything in how they interpret the questions—something that isn’t consistent with any obvious theoretically attractive utility notion.

There is hope for dealing with these problems, but not with the usual approaches. We give suggestions in the paper for a way forward.

One overall lesson I have taken from an extensive research agenda I have been involved in using self-reported well-being data is that it takes a lot of careful, mathematically and econometrically intensive work—and careful data collection—to get things right. And by “getting things right” I mean to something that doesn’t have an obvious problem that I personally am quite worried about, not solving all the problems that someone else might worry about. I am optimistic about the ultimate potential of this research agenda for informing policy choices for governments and other institutions in the light of the effects of those choices on the full range of things people care about, including non-market goods that don’t generate transaction data. But it is a long journey from where we are to where we need to go. That is a journey my coauthors and I have set out on.

For more on some of the difficulties in the road ahead, see the paper I highlight in this post:

National Well-Being Indexes and Goodhart’s Law