On the lags between submission and acceptance: Are all referees created equal?

This short note links days and delays in refereeing with some basic characteristics of the reviewers. An altruistic approach is used, and 1994 data from Hamermesh is augmented and it is found that (i) long articles, (ii) referees outside the USA and Canada, (iii) lack of academic experience, (iv) relatively less requested referees, and (v) lower income, are linked with possible delays in response of the referees. There are no differences in gender, type of journal (general or specialized) or prestige of referees. Academic experience is not non-monotonic. 'Networking' effects are not present.