Kathy G's discussion of monopsony models in low-wage labor markets is well worth a read, as is the standard book on this subject, Monopsony in Motion. The precise argument she's having with Megan McArdle, however, strikes me as slightly beside the point. Search friction -- the factors that make it harder for you to find and acquire another job -- is probably not the main reason to believe that the monopsony model is applicable among low wage workers. Rather, it's search prospects that hold folks back. What gives a worker bargaining power is his or her ability to leave and deprive the employer of their expertise. But unskilled, uneducated workers, don't really have that option. There are relatively few jobs for them where any expertise is developed, or where turnover isn't an accepted reality of the industry, and so jobs are constructed to require little more than a warm body. All this means that low wage workers rarely have much power to walk from the table. If you're totally replaceable, the employer won't care if you walk away from the table. And without that underlying threat,you can't bargain your pay into symmetry with your labor -- you're basically at the mercy of the employer and whatever laws have been passed to protect you. In fact, to understand this stuff you really don't need to read Monopsony in Motion. You just need to read The Grapes of Wrath. Update: This appears to be confusing folks, so it's worth saying that there's a difference between what your labor is "worth" and what it can command. So imagine a set of workers who are being paid $3 for their labor. Then a minimum wage law takes effect, and now they're being paid $5. But none are fired. We see this often, and the implication is this: The labor is actually "worth" more to the employer than they need to pay for it. In other words, they're willing to pay X, but are only having to spend Y. The question for anyone trying to better the low wage market is how to bring X into symmetry with Y.
