CEPR - Center for Economic and Policy Research


En Español

Em Português

Other Languages

Home Publications Blogs Beat the Press Justin Wolfers False Symmetry on Unemployment Benefits

Justin Wolfers False Symmetry on Unemployment Benefits

Saturday, 26 July 2014 16:50

Justin Wolfers shows more of a commitment to coming down in the middle than to being true to the data in his pox on both your houses piece on the impact of North Carolina's cut in unemployment benefits. The basic story is that in July of last year North Carolina both cut the duration of unemployment benefits and vastly ramped up the job search requirements. The result was that the number of people getting benefits fell by 48,500, a decline of 53.4 percent between May of 2013 and May of 2014.

The conservative position on this cut was that ending benefits would give people the incentive they needed to get a job. The idea was that the government was needlessly coddling these people, when what they really needed was a little push to get them going. This story gives a very clear prediction.

The loss of benefits should lead to notably faster job growth in North Carolina than in the rest of the country. After all, the number of people losing benefits was more than 1.0 percent of the state's labor force. If the necessary kick story is right then we should have seen a notably faster increase in jobs in North Carolina than in neighboring states with similar economies.

By contrast, the liberal position is that the spending from benefits helps to boost the economy. However, this impact is likely to be relatively limited. First of all, we are not talking about that much money relative to the economy as a whole, and second all the spending will not boost employment in North Carolina. This is in part because North Carolians don't necessarily spend their money in North Carolina (some will spend it in neighboring states) and also because the spending itself will go to producers around the country and around the world.

Most of what someone pays for a car in North Carolina will go to a car company located in another state or country. The workers at the factories that assembled the car and produced the parts will almost all live outside of North Carolina. Similarly, the rent or mortgage payments made by a worker may go to a landlord or mortgage holder anywhere in the country. Therefore only a fraction of the impact of the spending will be seen in North Carolina.

If we back out the predicted impact of North Carolina's benefit cut on the state's economy, it is easy to see that it would be relatively modest. If we assume an average benefit level of $15,000 ($300 a week), the reduction in beneficiaries would lower payments by $720 million over the course of the year. Assuming a multiplier of 1.5 this would cut GDP for the country as a whole by roughly $1,080 million. If half of this loss accrues to North Carolina, it would lower demand in its economy by $540 million. This is roughly 0.12 percent of the state's $470 billion economy.

Okay, now lets get back to Wolfers' pox on both your houses story. If the people thrown off unemployment insurance are getting the kick they need to get a job then we should see some visible impact on job growth in the state. As Wolfers notes, we don't. North Carolina's job growth is no better than in neighboring states.

On the other hand, if we assume the liberal story is true, it would be very hard to detect a decline in growth of 0.12 percentage point, even if it did in fact occur. There are enough random factors and errors in the data that we could never expect to find such a modest impact even if the economy really did suffer from the cut in benefits.

Certainly some liberals did exaggerate the negative impact from the benefit cuts on the economy. But the evidence that Wolfers presents hardly disproves that the cuts did not have the negative impact that any reasonable analysis would have predicted.


Corrected Note: "billion" corrected to "million." Thanks to Michael Epton and Robert Salzberg.

Comments (9)Add Comment
written by Michael Epton, July 26, 2014 5:35
… it would lower demand in its economy by $540 MILLION.

But most of your regular readers will easily catch this because you have made them quantitatively literate.
Wolfers claimed Dems only worry that cutting UI cuts aggregate demand
written by jaaaaayceeeee, July 26, 2014 5:58

Robert Waldman at Angry Bear asked a better question: Does the data support Republican's claim that we should cut UI to increase incentive to find a job. He finds that the data disproves Republican claims. Justin Wolfers should have asked this question:
written by skeptonomist, July 26, 2014 6:00
One definite and immediate prediction from cutting benefits is that it should reduce wages, as the moochers are no longer supported and must undercut those who may already have jobs. In fact this is the real economic justification for cutting or eliminating benefits if there is one - it is assumed that downward wage adjustment is what the economy needs. In other words it is claimed that benefits put an artificial floor on wages. So have wages gone down in North Carolina?
written by skeptonomist, July 26, 2014 6:05
Data in FRED (NCWTOT) don't show any drop in wages and salaries in NC.
North Carolina Demand is Reliable Because It's Always There
written by Last Mover, July 26, 2014 7:24

As mentioned above the question isn't between creating more demand with UI spending versus finding a job. If jobs were available by searching more, filling the implied unfilled demand that conservatives claim is there is obviously superior to paying workers not to fill it.

The question is whether demand is actually there to justify elimination of UI. Wolfers confuses it by saying UI is a moral question, then plays the middle to conclude both sides are wrong since North Carolina is like other states - implying demand is not there overall, nor from UI spending.

As skepto says above the real test of whether (unfulfilled) demand is there is wages. If it's there then wages need not decline. If it's not there then wages must decline, presumably to result in price declines which then are claimed to stimulate demand. History and Keynes have shown this doesn't work in the short run to restore demand.

The extreme case would include desperate unemployed workers competing directly with employed workers, bidding for their jobs at lower wages. Since it doesn't increase aggregate demand it's just a zero sum game between the employed and unemployed (including those who leave the labor force because of it).

But North Carolina wouldn't have it any other way because they just know demand is already there or it will arrive shortly once UI is eliminated.
Primary Benefit
written by Jeffrey Stewart, July 27, 2014 6:05
"By contrast, the liberal position is that the spending from [unemployment] benefits helps to boost the economy." -D. Baker

Isn't the primary purpose of unemployment benefits to still allow the unemployed to purchase the necessaries of life?
Cut Wages More!
written by Ellis, July 27, 2014 10:19
Cutting unemployment benefits forces people either into abject poverty, or to take the first job offered, no matter how miserable. Companies love it. It reduces their premiums, and provides more fresh meat to feast on.

In plain English, it's class warfare.
Justin Wolfers prior defense of Reinhart-Rogoff
written by John Wright, July 27, 2014 12:32
A little over a year ago, Justin Wolfers and Betsey Stevenson wrote an op-ed excusing the sloppy work of Reinhart-Rogoff


I particularly like this statement, in which Wolfers-Stevenson maintain the newly computed, from the finally released Rogoff-Reinhart data, actual +2.2% growth of high debt countries vs the -0.1% growth figure of original Reinhart-Rogoff doesn't undercut the original findings:

"But let’s not get lost in the trees. In the end, all the corrections advocated by the critics shift the average GDP growth for very-high-debt nations to 2.2 percent, from a negative 0.1 percent in Reinhart and Rogoff’s original work. The finding remains that economic growth is lower in very-high-debt countries (see chart). It has been disappointing to watch those on the left seize on the embarrassing Excel errors but ignore this bigger picture."

Of course, the Wolfers-Stevenson Reinhart-Rogoff "refereeing" might have not been necessary if Reinhart-Rogoff had released their data earlier, something that Wolfers-Stevenson don't even approach as an concern.

I guess I'm still lost in the Reihhart-Rogoff trees and give less credence to Wolfers as a result.

written by howard, July 28, 2014 10:11
there are currently something like 4.6M job openings and 9.5M unemployed.

i have yet to understand how the unemployed are supposed to find jobs that don't exist, and i have yet to understand how eliminating long-term UI benefits increases the numbe of jobs available.

i do understand that wanting to work doesn't create jobs, so who cares if the incentives improve (or don't) to "find work" if there ain't any work to find?

Write comment

(Only one link allowed per comment)

This content has been locked. You can no longer post any comments.


Support this blog, donate
Combined Federal Campaign #79613

About Beat the Press

Dean Baker is co-director of the Center for Economic and Policy Research in Washington, D.C. He is the author of several books, his latest being The End of Loser Liberalism: Making Markets Progressive. Read more about Dean.