When you cite a book, you're expected to list its city of publication. As in:
Sheffrin, Steven. (1996). Rational Expectations. Cambridge: Cambridge University Press.
Questions: Does anyone care what city a book was published in? Why do you need to know?
Citing the publisher gives you a signal of quality. Citing the city of publication seems like a total waste of labor to me. Maybe before the Internet was around, it had some use, though I doubt it.
Of course, I'm not going to stop citing cities all myself. There's a real path-dependence problem here - no one wants to deviate from accepted practice first, for fear of being seen as a dummy or a weirdo.
But the day I become a journal editor, the city of publication is history!