I would be willing to bet that there are at least a few people, (and I would guess more) who have decided to go to one school over another because of a ranking difference as slight as you are talking about. Especially if the change in rank is the difference between being able to say you went to a "top ten" school and a "top twenty" school.
That is a very good point about schools with more economists getting more votes as well. I bet this compounds the self selection bias since schools that USNWR ranks low probably discount its importance and therefore are less likely to respond, while schools with high rankings are more likely to believe the importance of the rankings and are therefore more likely to respond. This is pure conjecture on my part though.