Is Wage Insurance Optimal?

In academic economics, there is growing interest in a subfield called "dynamic public finance," which looks at optimal taxes and social insurance in situations where information unfolds over time. Here is one survey.

At the same time, in Washington policy circles, there is growing interest in social insurance schemes that have an important time-dimension. Some Democrats are proposing "wage insurance." According to the NY Times,

Under one proposal, by Representative Jim McDermott of Washington State, the government would pay as much as 50 percent of the difference between a person’s old and new wage — with a maximum benefit of $10,000 a year — for as long as two years. A welder who lost a job that paid $40,000 a year and who took a new job paying $30,000 would receive $5,000 a year in supplemental pay for two years.

Surprisingly, these two developments have been proceeding independently. As far as I can tell, the schemes being cooked up in Washington are not building on the academic literature, and the guys working on the academic literature do not pay much attention to what is happening in Washington. Perhaps it is time for a little intellectual arbitrage.

Personally, I am skeptical of the wage insurance plan. It seems that a worker who earns $40,000 in year one and then $30,000 in year two would end up with higher total income, after taxes and transfers, than a person who earns $35,000 in both years. Maybe there is a good argument for this idea, but I cannot immmediately see what it is.