Print this page  Print this article
  Email this article to a friend   Email this article
  Make the news text smaller  Make the news text larger  Text Size   

Monday, January 8, 2007
Job-hunting takes a line from dating


CHICAGO — In job-hunting, as in love, finding a match can be a harrowing experience that all too often ends in unhappiness. Some economists think they know how to make it less painful — and they are using their fellow dismal scientists as guinea pigs.

At this past weekend's annual meeting of the American Economic Association, which hosts a vast job market for aspiring professors, academics tested a technique — borrowed from online dating — to more efficiently match job candidates and potential employers. It is called "signaling," and it is designed to reduce the time and cost of hiring professors by weeding out those who aren't serious prospects and homing in on those who are.

Signaling depends on a centralized system through which each job seeker sends signals — essentially electronic pings — to two potential employers. With a limited number of signals to send, the logic goes, candidates will send them only to schools where they really want to work.

Job candidates were warned not to waste signals on schools that should already know they are interested or are out of their range, but instead aim at schools that wouldn't otherwise be aware of their special interest. Schools also were told not to take the absence of a signal as a brush-off.

"Think of it like a dating site," says Alvin Roth, a Harvard professor who chairs the AEA's Ad Hoc Committee on the Job Market. "We're trying to figure out if we can help people make matches."

The allusion to dating is apt. The AEA system shares an idea and an advisor — Stanford economist Muriel Niederle — with Cupid.com. On that online dating site, women face a problem akin to that of employers: Men signal their interest in women by sending electronic messages, but because it is easy to send hundreds of messages, it is difficult and time-consuming for women to separate spammers from good prospects.

In the summer of 2005, at the suggestion of Prof. Niederle and MIT economist Dan Ariely, Cupid began allotting each of its male members two electronic roses a month, which they could send along with messages to women whom they wanted to impress. The scarcity of roses motivates the suitors to be selective and serious.

"It's been a wonderful thing," says Eric Straus, CEO of Cupid.com, who estimates the roses have increased a suitor's chances of getting a reply 35 percent. "One of the problems in online dating is that men are ignored and women are inundated. Anything that allows a message to stand out is a great benefit."

Economists couldn't have found a better testing ground for their ideas than the grueling market for newly minted economics Ph.D.s. Every year, about 1,000 soon-to-be doctoral graduates converge on the site of the AEA meetings, creating traffic jams at elevators and squeezing as many as 30 job interviews into three or four days. Time and space constraints get so severe that candidates often find themselves sitting on hotel beds as they pitch themselves — sometimes to five or six interviewers, who typically are male and might include Nobel Prize winners and heads of economics departments.

"It's quite unpleasant, especially for women," says David Colander, a professor of economics at Middlebury College in Vermont. He estimates that, because of the arduous nature of the selection process, the hiring of one young professor can cost a school from $10,000 to $15,000.

This year, 2,300 new jobs were listed at the AEA. Amid the noise and rush, potential employers make a lot of mistakes in allotting their attention. Some lower-tier schools, for fear of being spurned, avoid interviewing graduates of top-tier programs, thus missing some who might have had a special reason to choose them due to location, family or hobby considerations. Others set their sights too high, inviting candidates who would never accept their advances, leaving better matches feeling unappreciated and unwanted. As a result, many qualified candidates fall through the cracks.

Profs. Roth and Niederle belong to a growing field of economics known as market design, which is rooted in the idea that markets, if left to develop on their own, often get into trouble and need to be fixed. Market design's most notable achievements include federal auctions for radio spectrum, which have earned the government billions of dollars in revenue, and the National Resident Matching Program, an automated clearinghouse that successfully matches hospitals to new doctors based on preferences each side feeds into the system.

If the signaling experiment works for economists, it could lead the way to meaningful improvements in a broad range of markets, from college admissions to placement of freshly minted business graduates and lawyers. "We want to make markets work more efficiently," Prof. Roth says. "That would result in more people being happy with their jobs, in more firms being happy with their employees and presumably society being more productive."

It is still early to say whether the endeavor will succeed. In all, Prof. Roth says, 969 job seekers sent signals, suggesting that most at least gave it a try. Participants offered mixed impressions.

Ethan Kaplan, a professor at Stockholm University who is interviewing candidates this year, said the signaling led him to focus on one prospect who otherwise might not have stood out.

Whatever the outcome, some have doubts that economists would ever willingly submit to a fully efficient market, such as the one that assigns doctors to hospitals. "The academic job market is a thing that could be enormously rationalized," says Prof. Colander, who has proposed an automated clearinghouse for economists. "But that doesn't mean that economists want to have their lives structured by a market."