Matt Spiegel nails it

for the most part in his analysis of the editorial process in finance journals (much of this also applies to economics). His experience as editor of RFS inspired him to write a little polemical piece with some sparkling insights. It seems that nowadays, every referee tells the editor how he or she would have written the paper, in a bid to seem smarter than the author... the final result is sometimes a big improvement, and often enough, no improvement at all. I can think of a paper of mine that went through N revisions in several years, before finally being published; the final version, bloated and cumbersome, isn't half as good as the original. True, another paper was so-so on submission, and then really improved through the editorial process. So it's a mixed bag overall.

What is Matt Spiegel's suggestion? Relax. The good articles, the ones that are truly important, will receive close scrutiny anyway, so don't demand robustness check no. 12345. His conclusion?

We can accomplish a lot more as a profession if we all do less as referees and editors. Referees should just read the article at hand and try to figure out if anybody else would. If the answer is yes, make at most a few suggestions. Limit them to ones that are editorial in nature. Help make the article more readable. Then sign off.

I think this is something to ponder...