In my previous post on Noah Smith, Brian Albrecht raises a question that I only mentioned briefly but considered writing a whole post on. He asks:
"How do you differentiate between "the economist" and "the citizen?" It is something I struggle with when reading others writing and my own."
What do you think?
It's difficult because of course we want to slip in normative/citizen claims whenever we're writing about something that matters for society at large (and not just as a scientific curiosity about how the world works). Economists aren't unique in this regard. All medical science and any biology or chemistry contributing to medical breakthroughs is going to have these little normative addenda too.
I think it's a spectrum, but we do science (in other words, we are "economists") at the end of the spectrum where the normative claims are minimized or included as a sort of wrap-up of the significance of the work and where claims of fact are evaluated on positive rather than normative bases.
Anything touching on welfare economics gets the trickiest in this regard. There are standard ways of talking about how one outcome unambiguously dominates another outcome regardless of actual preferences that people hold. There are also standard definitions of efficiency that we can use because they are standard without necessarily attaching ethical significance to them (of course we've defined them so that they might be useful in subsequent ethical conversations).
When we are acting as citizens I think we have much freer reign mixing positive and normative claims. After all, we're economists so it's pretty much incumbent on us, acting as citizens, to bring the positive claims into the conversation. I would just say you have to be very clear about which you are doing when, and of course you have to be honest. If you are considered to be an expert on a subject, you can't cherry-pick the evidence you find most suitable. If you downplay certain views you have to be able to justify it.
Introduction to Algorithms, Lecture Three
1 hour ago