I was particularly struck with this bit:
Not all authors agree, of course, but in my view we are in the business of evaluating papers, not improving papers. If you want to improve your paper, ask your colleagues for advice. When you know what you want to say and how to say it, send it to a journal.I find it hard not to want to spend a lot of time trying to improve the papers I handle, whether as an editor or as a referee. This is not bad in and of itself but it has the side effect of greatly increasing my editorial and referee response time. Put differently, I think I am confused about the point of the business, which is indeed evaluation, not improvement.
I am also keen never to be caught up on the other end of something that once happened to me as an author. Back in the day, my friends Kermit and Dan and I wrote a great big paper on the effect of college quality on labor market outcomes - this paper was the distant ancestor of the paper eventually pubished, a decade and a half later, in the German Economic Review (in English, as are all the other papers in that journal). We decided it was too long to submit as written so we broke it up into what we called the "boy" and "girl" papers. We sent the boy paper to journal A and the girl paper to journal B. Each paper included copious citations of the other paper, as in at least one every page. After a reasonably long wait, we get back a rejection on the boy paper from journal A in which the editor states that "based on [his/her] careful readnig" he/she is rejecting the paper because it does not look at women, only men. Of course, if said editor had read even the first page of the paper (and not even very carefully), it would have been clear that there was a separate parallel paper entirely devoted to women. It was not so much the rejection that bothered me - the paper was okay but not great - but rather the straight out lying by the editor. As it turned out, I had the opportunity to present the boy paper at the NBER labor studies group with said editor sitting three spots down on the right at the main table. So I began my talk with the story of the one big paper and the two smaller papers, all the while looking straight at said editor. Said editor has been nice to me ever since.
Still, this experience looms large in my mind whenever I sit down to be an editor, and I think pushes me to spend more time than I should on the papers, especially the ones that are going to end up getting rejected.
Finally, I do know the solution to Preston's puzzle about why the referee reports he gets as an editor are better than the ones he gets as an author. It is because he is better at picking referees than most other editors!