The U.S. job market has tightened by many measures — more advertised job openings, fewer claims for initial unemployment insurance, substantial reduction in long-term unemployment and the number of discouraged workers. Yet the percentage of working-age population that is either working or looking for work (the labor force participation rate) remains extremely low. This is a big problem, since projections of future economic growth are constructed by adding expected growth of productivity to growth of the labor force.

Why have so many people dropped out of the labor force? Since they’re not working (at least in the formal economy), how do they pay for things like food, rent and health care?

One explanation answers both questions: More people are relying on a variety of means-tested cash and in-kind benefits that are made available only on the condition that recipients report little or no earned income. Since qualification for one benefit often results in qualification for others, the effect can be equivalent to a high marginal tax rate on extra work (such as switching from a 20 to 40 hour workweek, or a spouse taking a job). Added labor income can often result in loss of multiple benefits, such as disability benefits, supplemental security income, the earned income tax credit, food stamps and Medicaid. 

This graph compares annual labor force participation rates with Congressional Budget Office data on means-tested federal benefits as a percent of GDP. The data appear consistent with work disincentives in federal transfer payments, labor tax rates and refundable tax credits.