I knew that this paper was in the works, it was just a matter of time before it hit.
Those who’ve listened to President Obama’s speeches over the past couple months have heard him boast that 2014 has seen impressive improvements in the labor market — the best year in job creation since 1999, he points out, and he’s right. But there’s no obvious explanation for why 2014 has been, by a good margin, the best year of a weak jobs recovery. The president has naturally credited his policies (without any justification). But what if 2014’s jobs boom is mostly thanks to the expiration of a program that the Obama administration and Democrats fervently pushed to renew?
That’s the finding of a new NBER working paper from three economists — Marcus Hagedorn, Kurt Mitman, and Iourii Manovskii — who contend that the ending of federally extended unemployment benefits across the country at the end of 2013 explains much of the labor-market boom in 2014.
About 60 percent of the job creation in 2014, 1.8 million jobs, they find, can be attributed to the end of the extended-benefits program. That’s a huge amount, and suggests that long-term unemployment benefits, while there’s a good charitable case for them, could have played a big role in the ongoing lassitude of our labor market. (Indeed, an earlier working paper from a few of the same authors argued that extended benefits raised the unemployment rate during the Great Recession by three percentage points; see a summary of that paper here.)
So what was the program Democrats wanted to renew? States run their own unemployment-insurance programs, which provide around 26 weeks of benefits to people who’ve lost jobs and are looking for new ones. But during the recent recession, as they have in other downturns, Congress repeatedly authorized federal extensions that allowed people to draw benefits for much longer. At the end of 2013, the Senate narrowly passed a renewal of the program, but the House never took it up and the extensions, already much longer than any previous recession had seen, expired.
This created something of a “natural experiment.” States had unemployment-insurance programs of widely varying length — they ranged from 40 weeks up to 73, roughly — but after the end of the federal extensions at the start of 2014, the duration of benefits in almost all states went back to around 26 weeks.
The paper uses that shift to examine how expiring benefits might have affected the labor market, and they find that the expiration of extended benefits produced a big boost to job creation, labor-force participation, and hiring. It’s a dramatically different result than what the White House and Democrats were predicting at the end of 2013: The Obama administration was predicting that the drop-off in stimulative spending from the expiration would cost 240,000 jobs, while the NBER paper finds that it created 1.8 million jobs.
The authors don’t think this happened the way you think it might: It’s not so much that the cut-off drove individuals on benefits back to work, but more that less-generous benefits actually spurred job creation on a macro level, getting employers to hire and drawing into the labor force people who hadn’t been looking for a job. They don’t lay out how that worked, but in their October 2013 paper, argue that extended unemployment benefits artificially boosts wages — when they expire, employers then boost job openings and start hiring people.
Of course, the usual caveats apply: This is not a perfect experiment at all, and the paper, while very rigorous, can’t get past the fact that it’s just crunching numbers about macro trends. And there are some concerns with the authors’ county-level data, though they try to make up for that.
The simplest form of the analysis was just looking at states that had long benefit terms versus short ones. In 2013, job creation was worse in more generous states than the national average; in 2014, after those states dropped their much more generous programs, it was much better than the national average:
There’s a lot more analysis they did, which I won’t get into — but to untangle related effects, they look at neighboring counties in states with different unemployment regimes, etc.
Now, this is just one paper and it involves some fancy econometrics, but it answers an unresolved question — why 2014 saw the labor market perk up (there’s also a possible end-of-austerity explanation, but it’s the labor market, not the economy overall, that’s really improved noticeably).
It should prompt passionate supporters of the extended unemployment-insurance program to consider whether it made as much sense as they thought. Even conservative economists, such as Michael Strain, pushed for the extension of long-term benefits. The length and scale of benefits during the Great Recession was unprecedented, but advocates for the program argued that this was necessary so long as unemployment, and especially long-term unemployment, remained historically elevated. Besides the moral case for supporting the unemployed, the market-friendly case for extending benefits is that one has to be searching for a job to get them. Cut the benefits, and you’ll see the long-term unemployed drop out of the labor force for good, the argument went. (It’s extremely hard to tell what did happen with these people when benefits expired, and the NBER paper here doesn’t comment on that.)
Advocates for extended benefits also argued that it was just an effective form of stimulus for the economy, because recipients spend their benefits immediately. That was always a pretty lame case, since the program’s value to the economy in spending terms — in the Obama White House’s generous estimation, 240,000 jobs in 2014 – would probably be outweighed if either side’s arguments about the labor-market effects proved mostly true. Indeed, if the new NBER paper is right, letting benefits expire produced 7.5 times as many jobs as the White House said it would cost.
The general economic consensus has always been that unemployment insurance slightly boosts the unemployment rate. Even liberal economists accept this, although they lampoon the idea that people might prefer benefits to working (that isn’t the point, Paul — people act at the margin). But we still have unemployment insurance, of course, because we want a safety net for people in the event of job loss. That just has to be balanced against the costs that the program imposes on the labor market. The new NBER paper doesn’t find that those costs in general are much higher than economists generally assume; rather, it suggests that the benefits of reining in long-term programs can be quite substantial.
There was always good reason to think this is the case: One of the many differences between American and European labor markets is that most of the latter have unemployment benefits systems of effectively unlimited duration — and much higher levels of structural unemployment.
All of that is very nice, except they don’t take this into consideration:
The seasonally-adjusted SGS Alternate Unemployment Rate reflects current unemployment reporting methodology adjusted for SGS-estimated long-term discouraged workers, who were defined out of official existence in 1994. That estimate is added to the BLS estimate of U-6 unemployment, which includes short-term discouraged workers.The U-3 unemployment rate is the monthly headline number. The U-6 unemployment rate is the Bureau of Labor Statistics’ (BLS) broadest unemployment measure, including short-term discouraged and other marginally-attached workers as well as those forced to work part-time because they cannot find full-time employment.