The effect of unemployment insurance on unemployment

By Felix Salmon
December 9, 2010

Last week, when I wrote my post on how to boost employment, the list started off unambiguously:

The first—and this can’t be stressed enough—is simply extending the federal unemployment extensions. As Menzie Chinn notes, the CEA has scored this, and the numbers are enormous: already, the program has increased the level of employment by 793,000 jobs. If the extensions are kept dead, there will be 593,000 fewer jobs in a year’s time than there would be if they were resuscitated, including more than 46,000 jobs in Florida and more than 26,000 jobs in Michigan.

This is not intuitive, especially to economist types who think that incentives matter and that at the margin, paying people to remain unemployed is not going to increase their chances of getting a job. But the fact is that those unemployment benefits are spent, and the extra economic activity naturally creates employment.

There is of course some effect by which paying people to stay unemployed will increase their chances of doing so. Rob Valletta and Katherine Kuang, of the San Francisco Fed, did the math back in April, concluding that the effect is “relatively modest”:

The question arises whether this extended availability of UI benefits has contributed to a lengthening of unemployment spells because jobless workers are staying in the labor force longer in order to continue collecting benefits. Such a dynamic could raise the unemployment rate. However, analysis of data on unemployed individuals decomposed by their reason for unemployment, which affects their eligibility for UI, suggests that extended UI benefits have had a relatively modest effect. We calculate that, in the absence of extended benefits, the unemployment rate would have been about 0.4 percentage point lower at the end of 2009, or about 9.6% rather than 10.0%.

Peter Coy has taken a detailed look at the interplay between the two effects:

Do the extra checks make unemployment higher than it would otherwise be by paying people to sit at home? Or do the checks sustain growth by supporting the spending power of households with out-of-work breadwinners?

In truth, unemployment benefit extensions do both—they raise the jobless rate a bit, and they make the economy grow faster. What’s clear is that extending jobless benefits makes more sense when the unemployment rate is exceptionally high, as it is now, at 9.8 percent in November… Because aid to the jobless is almost immediately spent (as opposed to tax refunds for the wealthy), it is the most effective means of stimulating demand.

Coy’s “bottom line” is clear: “Although the Obama-GOP tax deal extends unemployment benefits, it probably will not dissuade many jobless from seeking work.”

This all adds up to something reasonably clear. Unemployment insurance isn’t only just from a fairness perspective, it’s also extremely effective as stimulus. Any effect whereby it encourages people to stay unemployed is, in comparison, modest.

Which is why it’s very odd to find Kelly Evans, in the WSJ, writing the exact opposite.

More jobless benefits, more unemployment.

A likely rise in the U.S. jobless rate is the unfortunate reality of the government’s move to fund extended unemployment benefits for another 13 months.

The effect probably won’t be huge, but it will be significant. And it may well hamper any recovery in investor and business confidence.

Evans isn’t very good at math*:

Individuals not actively searching for work or willing to take available jobs may claim they are unemployed in order to receive benefits. That could artificially boost the size of the labor force, which is used to determine the unemployment rate.

Well yes, the labor force is indeed used to determine the unemployment rate, but it’s the denominator in that calculation. If the denominator goes up, the rate goes down. The problem is rather that in any ratio less than 100%, if you increase the numerator and the denominator by the same amount, then the ratio goes up.

Evans concludes:

Policy makers are hoping that extending benefits—along with other tax breaks—will generate enough short-term strength in spending and growth to overshadow any rise in the unemployment rate.

That may prove wishful thinking. The late rapper Notorious B.I.G. probably put it best: “mo’ money, mo’ problems.”

Evans’s piece elicited a smart smackdown from Zack Roth, who actually went to the trouble of phoning up the SF Fed’s Rob Valletta:

“These separate effects act in opposition to one another,” said Valletta. So the question becomes: Which effect is greater, in our current situation?

On this, Valletta was clear. In the current weak labor market, he said, the micro effect is relatively small. “I think the macro economic effects, in terms of reducing the unemployment rate, outweigh the micro effects that increase the unemployment rate,” he said.

This makes perfect intuitive sense, since the macro effects right now are huge, on the order of $60 billion being spent, and 600,000 extra jobs created. It really doesn’t seem plausible, in this economy, that more than 600,000 people will stay out of work and live on their unemployment checks rather than accept a job they would have taken in the absence of those checks; neither does it seem plausible that injecting $60 billion into the economy would send the unemployment rate up.

After Roth’s piece appeared, Evans responded, saying that “we’re all making the same point re: jobless benefits.” (It’s fantastic, by the way, that she’s happy and willing to join the public debate over her stories.) Roth was not convinced, replying that “your point was jobless benefits boost unemployment. everyone else’s is that they cut unemployment. seem like different points.” And so Evans clarified, here and here:

The disagreement is over net effect; will extending UI do enough for growth to overcome the rise in unemployment?

I’m not that optimistic about growth. Extending UI helps growth; but enough to overcome upward pressure on UR?

This misses the point: extending UI would be a good idea even — especially — if growth were sluggish. It’s when the economy isn’t growing that you need to apply stimulus, if only to prevent it backsliding into a double-dip recession. Growth doesn’t need to be high in order for UI to create employment; it just needs to be higher than it would have been absent the extra benefits. If you’re “not that optimistic about growth”, that’s all the more reason to want the fiscal stimulus of extending UI.

Evans might be right that the US unemployment rate is going to rise rather than fall. But if the unemployment rate does rise, the reasons for that rise will be found in the macroeconomy as a whole. Blaming any such rise on the extension of unemployment insurance will be silly.

*Update: This passage was ill-written, or ill-advised, or both. When I reposted this at CJR, I changed it to say that “Evans isn’t very good at explaining the math of why more unemployed people add to the unemployment rate”. I probably shouldn’t have included it at all, though, it’s not central to my point.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
Comments
12 comments so far

What’s missing from this discussion is the part about optimization of deployment of resources. If a skilled worker, say a skilled aluminum welder, can stay on unemployment long enough to find a skilled aluminum welder job at higher pay and productivity, that’s good for GDP and has a higher multiplier. But if that worker has to settle for a security guard job because he or she must pay the rent, then there is a net loss of productivity, GDP, and related multipliers.

Posted by Dollared | Report as abusive

Here’s how to reduce unemployment and allow the government to cut unemployment benefits: increase the corporate income tax rate to 60% or so, and then give 80% tax credits for hiring new workers. The credit would apply to the first $100,000 of salary of all new workers.

Businesses are sitting on a lot of cash, and it needs to be put to use. Corporations have adjusted to the lower levels of business that were thrust on them by the meltdown, so they are comfortably profitable. By combining a tax credit with higher taxes, the net tax increase for many companies could be a wash, while reducing unemployment and increasing revenues (assuming the businesses know what they are doing, which admittedly might be a stretch).

This would effectively force businesses to invest while reducing the amount of money the government has to spend on stimulus.

Posted by OnTheTimes | Report as abusive

Felix, to be counted as unemployed you need to be looking for work. To collect unemployment insurance you have to be looking for work. If you say that you are doing both when you are not you are boosting the unemployment rate. The denominator and numerator both increase but the denominator increases less proportionately.

Consider 1 more person who lies to collect benefits when unemployment is 10% of the workforce.
w u Real Unemployment Rate One Faker
100 10 0.1 0.108910891
1000 100 0.1 0.100899101
10000 1000 0.1 0.100089991
100000 10000 0.1 0.100009000
It always raises the rate. Kelly Evans’s second sentence is confusing but not wrong.

Think of it this way. To keep the unemployment rate unchanged at 10%, you’d have to add 10 people to the workforce for every person faking unemployment. But you don’t. One faker adds one person to the workforce and one to the unemployment rate. So that pushes the rate up.

Posted by OneEyedMan | Report as abusive

To be clear: yes, more people claiming unemployment does — obviously — raise the unemployment rate. But it does so entirely because the numerator — the number of people looking for work — has gone up. Yes, the labor force has gone up too. But that actually makes the rise in unemployment smaller.

Posted by FelixSalmon | Report as abusive

Hi Felix,

You can smack me down for this one but I have to ask why you think that the preservation of the status quo on UI amounts to stimulus. I can buy into new federal spending having a stimulative effect but I find the argument that maintaining a level of income for a certain segment of the population results in a boost to GDP and therefore creates jobs. I suppose their is a small multiplier effect but I question if it is measurable given the subsistence level of these payments. Clearly the elimination of the payments would have a negative effect.

But more to the point, I wonder if you caught this in the WSJ – http://online.wsj.com/article/SB10001424 052748703350104575653103884264496.html. The gist of it is that given the size and complexity of the US economy, estimates of employment, job creation, unemployment etc. are just that – rough estimates. With that in mind, would you agree that the employment estimates you quote might well be, um, sort of suspect?

I would be interested in your thoughts.

Posted by TomLindmark | Report as abusive

Does UI increase or decrease unemployment?

I think the truth is, it impossible to tell. You are right UI being real and effective monetary stimulus.

On the other hand, I do know of some who take UI rather than get the nearest low-paying job around. And that may actually be a good thing.

UI can be very beneficial to individuals and the whole economy for another reason: A skilled computer scientist or welder should have enough time to try to find work in line with their training, even if it takes a little while, rather than be forced to flip burgers.

That skilled worker could take much longer to find really appropriate work if they must spend their days making minimum wage in retail.

So UI could increase the unemployment rate, but that could be a good thing if the unemployed are using their time to find good work.

Posted by DanHess | Report as abusive

Does UI increase or decrease unemployment?

I think the truth is, it impossible to tell. You are right UI being real and effective monetary stimulus.

On the other hand, I do know of some who take UI rather than get the nearest low-paying job around. And that may actually be a good thing.

UI can be very beneficial to individuals and the whole economy for another reason: A skilled computer scientist or welder should have enough time to try to find work in line with their training, even if it takes a little while, rather than be forced to flip burgers.

That skilled worker could take much longer to find really appropriate work if they must spend their days making minimum wage in retail.

So UI could increase the unemployment rate, but that could be a good thing if the unemployed are using their time to find good work.

Posted by DanHess | Report as abusive

@TomLindmark — I think arguing about whether extending UI boosts GDP or just stops it from falling is silly. The point is the delta.

And sure, I’m happy to agree that any single employment estimate might be off — but that applies just as much to the effect of UI on unemployment than it does to the effect of stimulus on employment.

Posted by FelixSalmon | Report as abusive

It would be interesting if we tinkered with getting dissatisfied workers more active with regards to finding jobs.

Imagine if you could only buy alcohol if you were 21 AND you submitted a W-2 to the IRS (or were claimed as a dependent, social security, etc).

There are lots of little details with this proposal that need to be ironed out: I’m assuming that poor people who don’t pay taxes still file W-2s, and something would have to be done to account for stay-at-home moms, and it’s state territory, not federal – However, when people talk about workers leaving the workforce, it seems like an accepted fact as opposed to a problem looking for incentives.

Posted by djiddish98 | Report as abusive

Felix,

Re: TomLindmark’s comment, why is it silly? You are right that the delta matters, and there were plenty of measures in the proposal that were stimulative (i.e. likely to create spending above and beyond current levels), but while extending unemployment payments (which I am for) that already exist holds the line on aggregate spending, it doesn’t stimulate additional demand (which is what business are waiting for before adding jobs).

Certainly withdrawing the benefits would stifle growth (the delta you speak of), but that doesn’t mean we should call the extension of benefits a stimulus. I don’t bring this up to play gotcha on semantics, but because these debates rely so much on which terms are use and how they are defined/legitimized/understood.

Posted by BillPetti | Report as abusive

The question arises whether this extended availability of UI benefits has contributed to a lengthening of unemployment spells because jobless workers are staying in the labor force longer in order to continue collecting benefits.

Unemployed workers are in the labor force longer? What?
Brian Luke brianluke1@yahoo.com

Posted by BrianLuke | Report as abusive

It’s always interesting to hear the opinions of well-paid, employed critics debating what it is that motivates the unemployed. I know it is naive to expect only those without jobs should be allowed to pontificate about the issue of unemployment, but as someone who has been stuck down that road I find any suggestion that unemployed people prefer the barely livable pittance of unemployment benefits to actually having a job infuriating.

Jack
http://www.accidentinjurydirect.co.uk

Posted by jacktrip | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/