The plight of the working poor

The graph below, from Greg Mankiw, shows income after taxes and transfers as a function of pre-tax and transfer income. At the high end of the income distribution the graph looks like you'd expect - as you get paid more you pay more in taxes, but your after tax and transfer income is still higher. In the "working poor" range, however, more effort to earn higher pre-tax and transfer income doesn't really pay off: various forms of assistance you get when you're poor (earned income tax credit, food stamps, Medicaid, etc.) phase out, leaving you no better off than you were with a lower income. That's bad.



The problem I have with this kind of analysis is, what would Mankiw do about it? Would he drop the green dot on the graph toward zero to improve incentives to work? Would he stretch out the transition period so as to reduce incentives to work for people in the $50,000-$70,000 range? Neither of those options is very palatable. If you're going to help the poor, you can either phase out incentives rather quickly as in the graph above, or phase them out really slowly, spreading the disincentives across the income distribution and greatly adding to the cost of your programs.

Disincentives to work are the unavoidable cost of a social safety net. Maybe it's ok that workers at Walmart don't have an incentive to take a second job at the 7-11; they still, after all, have an incentive to get an education and land one of them fancy college professor jobs on the upward-sloping portion of the graph.