This is a remark I made at the presentation of the PhD of François Ghesquière. I am not sure it is 100% correct and to be found in reality, but let me explain the paradox of increasing observed inequality when minimum wages increase.
We got used to thinking about minimum wages being fairly harmless to employment in monopsony markets. The minimum wage increases marginal costs above labour supply, so that the monopsonist limits demand for labour, but still pays the wage on the supply curve. Now suppose the supply curve is fairly flat (elastic). For a small amount of money offered, a great deal of extra workers will be available. Hence in this case, an increase in the minimum wage may increase the number of low-paid workers, and shift the tenth percentile more to the left than it does the median, thereby increasing lower tail (and overall) wage inequality, e.g. the gini coefficient.
Of course, the increase of the minimum wage appears to only have advantages in this case, as more workers entered the workforce and the market failure of the monopsonist is attenuated. The apparent increase is therefore due to the selection bias.
