North Carolina’s Failed Experiment in Cutting Unemployment Benefits

I don’t usually associate the American Enterprise Institute with compassion for the unemployed or anything, really, other than pro-business, anti-government policy prescriptions and rhetoric. So I was surprised and heartened by a thoughtful post by AEI Money & Politics blogger James Pethokoukis, who skewers the notion that cutting unemployment benefits will spur job creation.

Pethokoukis analyzes the effect of reductions in weekly benefit levels and total weeks of unemployment compensation enacted in North Carolina this summer—cuts so draconian they led to the state being kicked out of the federal Emergency Unemployment Compensation program that provides weekly benefits to long-term jobless workers. North Carolina Republicans claimed the cuts would force lazy workers to find jobs, thereby solving the state’s unemployment crisis.

Instead, as Pethokoukis shows, tens of thousands of North Carolinians stopped looking for jobs that weren’t there once they were cut off from weekly benefits (which are only paid to people who are actively seeking paid employment). The labor force participation rate fell nearly a full percentage point, as 42,656 workers gave up looking and dropped out of the labor force. If they hadn’t, according to Pethokoukis, “the state’s jobless rate would have increased to 9.1% rather than sharply declining.” University of California at Berkley economist Jesse Rothstein predicted this dropout effect in a 2011 paper he presented at EPI, which disputed the notion that unemployment insurance causes significant unemployment.

Hopefully, the North Carolina experience will help persuade House Republicans like Dave Camp to stop arguing that killing the EUC program will boost employment. As EPI and the CBO have shown, paying out $25 billion in EUC in 2014 will help the economy, not hurt it. Killing the program won’t help a single unemployed person find work, but will instead depress aggregate consumer demand and cost the economy 310,000 jobs.