Countervailing Power
Tuesday, May 31st, 2005Thirty years ago, a popular notion among intellectuals was that of John Kenneth Galbraith’s “countervailing power”: the idea that if one side of a market was monopolized, it would be good to have the other side monopolized too, for balance. This is actually a respectable idea. I don’t know how Galbraith justified it in the 50’s, but the plausible basis for it is that whereas a monopoly (or monopsony) limits trade to improve its price, a bilateral monopoly would reach something closer to the perfectly competitive price, or perhaps would even bargain to exactly the efficient output. Thus, if we had just one employer in town, it might keep wages at $10/hour (and limit its hiring to do so), but if the workers unionized, creating a monopoly seller of labor, the result might be a wage of $15/hour and greater employment.
Looking at this idea from the point of view of the micro theory of 2005, though, it looks flawed. It is not bad as far as it goes, though since the loss from Harberger triangles is small, the inefficiency from monopoly or monopsony is not great (on the other hand, remember the Tullock rectangles). The problem is that bilateral monopoly creates a bargaining situation, and when information is asymmetric, that leads to a lot of inefficiency as parties try to bluff each other and trade breaks down entirely occasionally.
This could be looked at empirically in the labor context. How much do strikes cost, and how does this compare with the likely triangle loss from monopoly, using Harberger-like assumptions?
This drawing is by one of my favorite artists.