The minimum wage reduces employment, especially among low-skilled workers for whom the minimum wage is most binding. That remains the consensus view but note that holding the consensus view does not preclude thinking that the decrease in employment is small relative to the increase in the wages of those who remain employed. If the employment effect is small, however, it is also important to understand why it is small – the policy implications of monopsony, which I think implausible, are quite different from the implications of the the idea that other aspects of the labor-contract adjust in response to enforced changes in wages (i.e. the converse of the hot water argument). See also Tyler on this.
When I discuss minimum wages in class I tell my students that one of the best ways to get a high-paying job is to get a low-paying job and work your way up. The minimum wage can put the least employable out of work and have permanent negative effects when training and work skills not acquired in youth are difficult to accumulate later on. I think the theory makes sense but until recently it had not been extensively investigated.
David Neumark and Olena Nizalova look at the how exposure to the minimum wage in the past impacts workers today. They find that teenagers who grow up in states with a minimum wage that is significantly and consistently higher than the federal minimum have lower earnings and work less a decade or more later when those workers are in their late twenties. The negative effects are larger for blacks, for whom the minimum wage tends to be more binding.
To generate variation, Neumark and Nizalova use data on minimum wages by state relative to the Federal minimum. The data is more aggregated than I would like and the variation by state only picks up in the late 1980s so there is less data than meets the eye. In theory, there is nothing special about the minimum wage as the driving factor that pushes people out of the work force, unemployment brought about by bad economic conditions should have similar effects. Thus, I wish that they had discussed the literature on hysteresis and unemployment. Welfare could also pull people out of the work force. I’m not fully convinced that they adequately control for economic conditions although they do use some clever techniques to try to address some of these issues. Nevertheless, my priors are supported so this must be a good paper! More seriously, Neumark and Nizalova are to be credited with opening the question of the long run effects of the minimum wage.
The bottom line? If you don’t work at McDonald’s when you are a teenager, don’t expect to manage a McDonald’s when you are middle-aged.
Addendum: Thanks to John Thacker and others who pointed out that one of my sentences, now fixed, was difficult to parse if you hadn’t read the paper – which sort of defeats the purpose of the blog, doesn’t it?