Often readers leave questions in the comment sections of individual posts that really deserve an answer. Just as often I am remiss in delivering to the deserved, and for that I apologize. In partial penance, I offer this post. Be, however, forwarned -- the econgeek rating on this one is pretty high.
Weeks back I linked to this from Mark Thoma:
Efficiency wage theory asserts that higher wages make workers more productive. In these models, wages cannot be lowered because the cost of falling productivity is larger than the savings from lower wages. Thus, lowering wages would increase costs of production...
... it is possible that increasing the minimum wage will increase effort. If the theory that higher wages cause higher effort has validity at all, it ought to affect workers at the lower end of the wage distribution who are most likely to be discouraged and shirk on the job. Alternatively, an increase in the minimum wage might attract more productive, discouraged workers back into the labor force.
My pal pgl asked:
What did you think of Dr. Thoma's efficiency wage defense for the minimum wage?
My answer: Well, I'm a neoclassical guy at heart, so I've never really thought that much about efficiency wage theory However, I am big fan of a (probably) not so well known paper by Mark Bils and Yongsun Chang that allows for varying labor effort when wages are pushed away from where unfettered market forces would take them. Bils and Chang are not concerned with the minimum wage, but changes in effective real wages that occur over the business cycle when nominal, or dollar, wages fail to adjust to equate demand and supply in the labor market.
The idea can be illustrated in a few pictures, adapted from Bils and Chang. Suppose that the labor market in its original state rests at a point like E in the following graph:
Now suppose that something happens, and the effective wage rises to (W/P) 1 . Bils and Chang are thinking about a situation in which the price level P falls, but the nominal wage (W) is "sticky", or slow to adjust. For the moment, however, feel free to think of this increase in the real wage as a mandated rise in the minimum wage.
There is obviously a bit of problem here. Given the demand and supply curves, the quantity of labor businesses want to hire at this new higher wage is less than the amount workers are willing to supply. In the usual case, the quantity of labor hired falls to the level businesses are willing to take, and workers end up wanting more hours, but getting less.
But now suppose that workers can, and will, adjust the amount of labor they expend per hour worked. In this case labor demand will increase, because businesses get more productivity for the wage they pay. On the other hand, the number of hours workers will supply falls (because they physically tire faster, for example). The supply and demand curves therefore might change as follows:
Two things are of note. First, the market is no longer in "disequilibrium": The adjustment of effort serves to bring the quantities supplied and demanded back into balance at the wage imposed on the market. Second, how the actual amount of labor employed changes depends on how much the demand and supply curves shift as effort changes.
In their study, Bils and Chang conclude:
... when we compare our model quantitatively to two models without an effort margin, one with flexible wages and one with sticky wages, we find output and consumption behave much like in the flexible-wage economy.
In other words, variable labor effort seems to undo a good measure of the (bad) effects that might otherwise follow from imposing a wage that is "too high".
Might this explain why the data do not speak so clearly on the overall employment effects of the minimum wage? Maybe. I suspect that the variability of worker effort is more plausible in a business cycle setting -- where the changes in effort are likely to be transitory responses to temporarily sticky wages -- than it is in the case of minimum wages, where the necessary adjustments in the intensity of work would have to last for a longer period of time. I also suspect that there are plenty of other margins that come into play when a binding minimum wage is imposed -- such as replacing labor with capital or, in the case of the legislation that instigated some of my earlier comments on this topic, moving the new Wal-Mart two feet outside of the Cook County border.
But the evidence is weak, so my hunch and a dollar still leaves you about three bucks short of a frappuccino. I'll stick with my original support of the position taken at Vox Baby, but it's good to remember that this is a value judgment, about which the economist in me has little standing.