3 ms·
I think this is from Freakonomics by Levitt and Dubner. After the initial publication of the book (and probably before it, since this is just a typical principa
by vitalique 14y ago
I think this is from Freakonomics by Levitt and Dubner. After the initial publication of the book (and probably before it, since this is just a typical principal-agent case) this point has been raised and discussed numerous times, though. National Bureau of Economical Research has a paper on this topic [0]. From the abstract: "It is not obvious that sellers benefit from those services".
[0] http://papers.nber.org/papers/w13796 http://papers.nber.org/papers/w13796
- mcguire 14y agoOff topic, but from that paper's abstract: "Our central finding is that, when listings are not tied to brokerage services, a seller’s use of a broker reduces the selling price of the typical home by 5.9 to 7.7 percent, which indicates that agency costs exceed the advantages of brokers’ knowledge and expertise by a wide margin." Whoa.
- fecak 14y agoThe main difference between houses and job candidates are that the houses can't refuse the sale and have no idea of their own worth. Recruiting is a sales job with a commission typically tied to salary of the hire (similar to commissions for selling houses that are related to sale price) - but where it gets interesting is that the brokered item in 99% of sales situations has no say in the matter. In recruiting, the candidate is to some degree the product of the transaction, and the product can reject the sale. It's an interesting dynamic that makes recruiting unique. This explanation is rudimentary and hopefully recruiters don't treat candidates like some common 'product' for sale, but it gives some insight into why analogies between recruiters and realtors or other sales/commission jobs are not entirely relevant.
- regerg 14y agoWhat? The job is what's for sale, not the employee. The hirer and the employee are the seller/buyer (or vice versa, doesn't really matter)
- fecak 14y agoThat is an interesting perspective, and again raises the question of contingency recruiting as a model. The company is paying for the service, so I guess you could equate the company as the seller paying a commission on the price. But the buyer (candidate in your model) and agent (recruiter) are the ones trying to drive price up (higher salary), which isn't how buyers act. I don't think that mentality is really prevalent in the industry, and the analogy seems to offend technologists. If there is a better analogy, I haven't found it yet.