Friday, August 16, 2013

Evidence that increases in the minimum wage increase unemployment

A few days ago I published a post about how increasing the minimum wage leads to more unemployment among unskilled workers, especially urban youth who are disproportionately black. Rather than take my word for it I used BLS unemployment data and minimum wage data since 1972 to create two charts.

The first (click graphs to enlarge) shows that decreases in the real minimum wage (in this case the wage is in constant 1996 dollars to show the real purchasing power over time) coincides with decreases in unemployment for both white and black youth between the ages of 16 - 19. (Min. wage scale is on the right axis, unemployment is on the left axis.)

The data only goes back to 1972 so the big divergence between the two rates is not shown, as this had already happened. But you can see that both rates fell on average when the real minimum wage decreased. This is not a ceteris paribus analysis and you can see some increases in unemployment despite a falling min. wage and vice versa, especially in the recessions of the early 80's and early 2000's and the dot com bubble during the late Clinton presidency. But the overall trend is there.

A more convincing graph is below; this graph plots the real minimum wage and the difference in the unemployment rates of black and white youth.

Again the increases and decreases in the real minimum wage closely align with the increases and decreases in the gap between the unemployment rates of white and black youth. As the real min. wage increases the more unemployed black workers there are relative to white workers. This chart, while not completely ceteris paribus, is still more informative since both white workers and black workers were facing the same macroeconomic conditions i.e. recessions, expansions, etc. and yet on average the black unemployment rate increased more than the white unemployment rate when the real min. wage increased and vice versa.

Just to repeat, this is not a statistical, causal analysis. But it does provide evidence for a very sound theory that when wages are increased workers who cannot produce enough to justify the higher wage are left without employment. A better graph would plot unskilled workers between the ages of 16 - 19 vs. skilled or workers by education levels since race is only a proxy for these more informative measures.

That being said, in my opinion the burden of proof is on those who say that increasing the min. wage does not harm low skilled workers, and in particular urban, often black, workers. And if they have no evidence for this, they need to show that the benefits from a min. wage hike outweigh these costs. This does not even take into account the moral argument that people who are willing to work for a low wage should not be denied the opportunity to do so, but that is for another post.

1 comment:

  1. In your first chart, the correlation is not between min wage and unemployment (wages were increased in 1996-98 without changes in UE) but between RECESSIONS and unemployment. You would be served by greying in recessions where the real correlation would be shown.