FACTS AND MYTHS ABOUT REFEREEING

Referees' and editors' behavior is illustrated by data from a random sample of refereeing requests by seven economics journals. Referees tend to be higher-quality (better-cited, prime-age) than authors. Except for a few superstar authors, there is no matching of authors and referees by quality. Nearly 80 percent of those asked to referee do so, with a median completion time of less than two months. Except for a few very slow referees and another few who promise but fail to accomplish the task, the slow editorial process is not due to referees' behavior. Paying referees speeds the job, mainly by speeding up those who would barely not qualify for the fee.