Wednesday, January 10, 2007

Dating services & getting a job

One of the more onerous tasks faced by younger participants at the American Economic Association (AEA) meetings is to try to get a job.

The meetings provide the largest organized job market in economics. Each year around 1000 soon-to-be doctoral graduates attempt to get jobs in universities and public agencies around the world that send representatives to the meetings and interview. Most of the larger, and some of the smaller, Australian universities send representatives and there are now typically quite a few Australian-based job applicants at the meeting as well.

The task of getting a job is onerous for all concerned because of the sheer volume of interviews – talented students might secure up to 30 interviews over the 3 days. For the interviewing institution the cost of hiring a young Assistant Professor can range between $10,000-$15,000. It’s a lot and my guess is that it would cost Australian universities more than this.

Harvard’s Alvin Roth has sought to simplify the procedure and reduce everyone’s costs. To do this he borrowed a technique from online dating – each applicant is allowed to send to electronic pings to potential employers to ‘signal’ a serious desire to secure employment at a school and to reduce frivolous applications. This is a market design procedure that reduces the costs of both applicant and interviewing institution.

Candidates are advised not to waste a ping on a school who already knows they are interested and not to waste a ping on a school that is well away from their plausible range. Schools in turn are told not to take a failure to ping as a brush-off.

The approach follows the 2005 suggestion of Stanford’s Muriel Niederle and MIT economist Dan Ariely to the dating service Cupid.com. The difficulty in online dating services is that women get lots of approaches from men but men only get a few. Thus this dating service allocates two electronic roses to its male members which they can use to contact two women. The men tend therefore to be much more selective and the signals they do send are much more serious. This reduces the search costs of both men and women by weeding out the non-serious interest.

Each of these examples illustrate the idea that getting more information in a market does not always promote efficiency given positive search costs. Measures can often be devised which restrict the amount of information provided and which thereby deliver efficiency gains to all.

5 comments:

Anonymous said...

Al Roth's work on two-sided matching and its implications for market design is particularly interesting. He has coauthored some nice papers that apply two-sided matching concepts to sorrority rushes and the market for medical interns.

Anonymous said...

So in practical terms, how do people think the pings are working out? Any chatter in the sidelines of the meetings?

hc said...

Christine, I didn't get any feedback other than that the two Aussie schools I know who were recruiting there knew nothing about the pinging arrangement.

Anonymous said...

Ah well, no doubt there'll be a wrap up/analysis at some point on the JOE website. Interesting innovations though, the job market scramble and the pings. Hope they do improve the matching process in the longer run.

Anonymous said...

Interesting view.