David R. Henderson  

The Effect of Unemployment Insurance on Unemployment

Working: The Graphic No... My Letter on California Studen...

Does standard microeconomics fail when there's a recession?

Last week, I addressed the issue of how much of the current unemployment is due to the many extensions of unemployment benefits. In some states you can now receive benefits for as much as 99 weeks, just shy of two years. To put that in perspective, we have never been there before. In the early 1980s, the longest anyone could receive such benefits was, I believe, 55 weeks. So this almost an additional year of benefits and it puts us in the same league as much of Western Europe.

There is much agreement among economists about the microeconomics of unemployment benefits, and it's worth reviewing why. I could lay it out fresh but Larry Summers said it quite clearly in his article on unemployment in my Encyclopedia. Here's Larry:

Each unemployed person has a "reservation wage"--the minimum wage he or she insists on getting before accepting a job. Unemployment insurance and other social assistance programs increase that reservation wage, causing an unemployed person to remain unemployed longer.

Consider, for example, an unemployed person who is accustomed to making $15.00 an hour. On unemployment insurance this person receives about 55 percent of normal earnings, or $8.25 per lost work hour. If that person is in a 15 percent federal tax bracket and a 3 percent state tax bracket, he or she pays $1.49 in taxes per hour not worked and nets $6.76 per hour after taxes as compensation for not working. If that person took a job that paid $15.00 per hour, governments would take 18 percent for income taxes and 7.65 percent for Social Security taxes [DRH note: this should be Social Security plus Medicare], netting him or her $11.15 per hour of work. Comparing the two payments, this person may decide that an hour of leisure is worth more than the extra $4.39 the job would pay. If so, this means that the unemployment insurance raises the person's reservation wage to above $15.00 per hour.

Moreover, as some people have pointed out, Paul Krugman and Robin Wells, in their textbook, Macroeconomics, 2nd ed., 2009, agree with this reasoning. They wrote:

Public policy designed to help workers who lose their jobs can lead to structural unemployment as an unintended side effect. . . . In other countries, particularly in Europe, benefits are more generous and last longer. The drawback to this generosity is that it reduces a worker's incentive to quickly find a new job. Generous unemployment benefits in some European countries are widely believed to be one of the main causes of "Eurosclerosis," the persistent high unemployment that affects a number of European countries.
Krugman argues, though, that this reasoning doesn't apply during a recession. He writes:
What's limiting employment now is lack of demand for the things workers produce. Their incentives to seek work are, for now, irrelevant. That's why comments by the likes of Sen. Kyl are so boneheaded -- anyone who thinks that high unemployment in the first quarter of 2010 has anything to do with workers getting excessively generous benefits must not get out much.
I think we're iterating. I think Krugman is right that lack of aggregate demand is the basic problem and I agree with Scott Sumner that the way to fix this is with monetary policy. But these explanations--lack of aggregate demand on the one hand and incentives for workers to take jobs on the other hand--are not mutually exclusive. Think through the microeconomics of what Larry Summers wrote above. For it not to apply during the recessions, it would have to be the case that workers literally can't find jobs so that whether the benefit is zero or $500 a week, the person remains unemployed either way. Is it really plausible that this applies to all workers?

Ask yourself this. You lose a job that paid, say, $40K a year ($800 a week) before tax and now you can get $25K a year ($500 a week) before tax--and, moreover, you don't pay the 7.65% employee portion of the payroll tax on that $500. You see a job that pays, say $30K a year. Do you think you might hold out for one that pays, say, $35K? There's nothing in this analysis that says you're lazy. What it says is that, in economists' usage of the language, "You're rational."

Here's the test: Can you find people getting unemployment benefits who have turned down jobs? If so, I'm right and Krugman's wrong. If not, Krugman's right and I'm wrong. For those who want to "get out much," this is the thing to get evidence on.

Comments and Sharing

CATEGORIES: Labor Market

COMMENTS (28 to date)
Snorri Godhi writes:

I am sure that you are right and Krugman is wrong, but this is not a good test:
Can you find people getting unemployment benefits who have turned down jobs?

If I found such people, I would not know whether they'd have accepted those jobs in the absence of unemployment benefits.

david writes:

What Snorri Godhi said.

On top of that - presuming you do find a way to measure such changes, then you'd also have to measure how many people lost jobs during to firms not finding enough customers to sell to at the margin. Aggregate demand and all that.

david writes:

*due, not during.

Of course, both effects could theoretically occur at the same time, with employment incentives undercutting the boost to aggregate demand. Then it just comes down to which dominates; presumably Krugman thinks the AD boost dominates whilst our kind host thinks the employment incentives dominate, given existing levels of incentives.

Lee Kelly writes:

The increased unemployment caused by extending and/or increasing unemployment benefits will be less during a recession than normal times. That seems fair to say.

wintercow20 writes:


My brother (let me call him Jack) fits your description exactly. He lost his job at a bond insurer and has had several offers for lower paying jobs from the rating agencies as well as some health care consulting companies. But he is holding out for his former $300k salary. He used up every last week of his unemployment insurance "benefits."

I also have a brother in law in quite the opposite position. He lives here in Western New York and works for a starter/alternator manufacturing company. He has taken both a wage cut and an hours reduction just to make sure he does not become unemployed. It is moderately amusing that this brother-in-law, who makes around $20,000 per year is paying into the UI system so that my brother, who lives in a nice house on Long Island and made $300,000 per year, can receive his benefits. So maybe my brother-in-law's position confirms the Krugman story.

Of course if that were true, then what was that about sticky wages?

Chris writes:

I know two people that have used the existence of unemployment benefits to turn down jobs that paid less than they wanted to make.

Ryan writes:

David H.,

Yes those individuals do exist. One lives right across the street from me.

Anecdotal: He has been in the airlines going on 10 years now and the company which is currently contracted by a major airline is leaving the airport thus he was forced, in order to maintain his current standard of living (income), to go for 1 of 4 supervisory positions. Not succeeding in getting that position, he was left with a choice between taking a full time position at a 40% pay cut (~$8.50 per hour) or go on unemployment insurance. He chose UI. Needless to say he's somewhat ashamed and looking forward to "a summer without work."

However, I'm not sure 1 or even a handful of stories similar or dissimilar to this one validate yours or Krugman's view. If conciliation matters, I tend to agree with your view. Thank you for and please continue with the insightful posts that continually corral my interest.

Braden writes:

In addition to the previously mentioned objections, labor market stickiness makes it very plausible that helping people hold out for better jobs is the efficient outcome.

David R. Henderson writes:

Baden writes:
"labor market stickiness makes it very plausible that helping people hold out for better jobs is the efficient outcome."

I'm not familiar with that term but I think Baden means sticky wages. But one reason wages are sticky is that people are holding out for a better or higher-paying job and one reason they're holding out is unemployment insurance.

Dan Weber writes:

Someone making $300,000 won't get anything close to even a quarter of that on unemployment. It might help keep the lights on, but it doesn't let one voluntarily stay unemployed unless holding out for $300,000 really is worth it. That or he just hates working.

Ano writes:

Here's the test: Can you find people getting unemployment benefits who have turned down jobs? If so, I'm right and Krugman's wrong. If not, Krugman's right and I'm wrong.

This does not settle the matter at all. Divide the currently-unemployed (limited to those getting benefits) into two groups:
(A) would be employed if there were no unemployment benefits.
(B) would be unemployed with or without the existence of unemployment benefits.

You are saying, roughly, "Group A exists." You are pointing out that Summers and Krugman have, in writing, stated that group A exists. You are correct.


Krugman, in his "don't get out much" post, is saying that group (B) spends a lot more money if they have unemployment benefits, while group (A)'s spending would be higher if they had jobs. So, if group (B) is sufficiently larger than group (A), aggregate demand is, on net, higher if unemployment benefits exist. Therefore, there is a group of people:
(C): currently has a job solely due to the extra spending by (B), and would not have a job if there were no unemployment benefits to juice B's spending.

Krugman is saying that (C)>(A). (Actually, he's saying that C exists, and you point out that he has written that A exists, so in order for him to be logically consistent, he would need to argue that C>A).

What say you Dr. Henderson? Do you think that A>C?

ray l love writes:

Krugman is the one who needs more out time. Find any black market activity and you will typically find a sizable percentage of those involved are receiving benefits of one kind or another.

Yesterday I met a man who retired from the military 38 years ago. He receives $35,000 a year from a pension [g-7], and SS, and partial disability. He also owns some rental properties which he makes improvements on by employing undocumented workers. He also buys and sells vehicles, equipment, tractors, horses. But I doubt there as many folks like him in Princetonland as there are here in Texas.

Martin K. writes:

Unemployment pay gives the illusion of freedom. People think they have greater freedom because they can stick to the same career field, they can accept wages close to what they had previously, they can stay within their state or metro area, and they can find a position that perfectly matches their skill set. By accepting this notion of freedom though, you are severely limiting your options.

David R. Henderson writes:

I may regret this because I'm answering quickly before going back to filling out my annual Faculty Activity Report, which must be submitted today.
But I think you're right. I think you've done a beautiful unpacking of all the statements. So here's what I'm saying.
I'm saying that I think A>C. I'm also saying, though, that Krugman is writing today as if he thinks that in a recession, A doesn't exist. If he admits the existence of A, then it becomes an empirical question whether A > C or C> A, and then he can't dismiss people as being boneheaded, not getting out much, etc. for thinking A > C.
Thanks for your very logical analysis.

Noah Yetter writes:

You have to pay taxes on unemployement benefits? Am I the only one who finds this to be unalloyed lunacy?

b-psycho writes:

Aren't unemployment benefits determined by previous employment anyway? That seems an odd type of arrangement to attempt to freeload off of...

Fiona writes:

[Comment removed for supplying false email address. Email the webmaster@econlib.org to request restoring this comment. A valid email address is required to post comments on EconLog.--Econlib Ed.]

Mike Y writes:

Your numbers are wrong. What state pays $500/wk in unemployment benefits? I agree that looking past $8.00/hour jobs for something better happens, but in my case, I was earning over $80k per year, and now with unemployment I'm getting $18,750. I'm willing to work for much less than I was currently earning, right now I'd jump at half, and even less than half.

If your theory is right, then a cheaper alternative would be for the government to simply eliminate payroll and income taxes for those going back to work after protracted unemployment. Say a two year period. That would give the employer a 7.5% tax break on SS, and the employee would not worry about cutting into their leisure.

Only flaw in this is that there are not many jobs out there. I can't get hired at Home Depot, Lowes, Costco, Golfsmith, Meijers, Wal-Mart, and others, believe me, I've tried.

I read a recent study that says small business isn't hiring because of low sales. Tax incentives, credits, etc. won't matter until people start buying again. That is the lynchpin in all the analysis.

David R. Henderson writes:

Dear Mike Y,
Your particular experience doesn't make my numbers wrong. The top weekly benefit is actually higher than $500. It's New Jersey's at $584.
Where my number is more questionable is $500 when your previous pay was $40K. If previous pay was $40K, then weekly benefit is likely to be more like $440 (55% of $800.)
Good luck in your job hunt.

Ben writes:

I am not sure if raising reservation wages is going to be a bad thing. When you account for job search frictions (don't assume perfect information and with it no transaction costs, that's stupid) we can understand how workers reservation wages are going to be LESS then their marginal revenue product of labor. This grants employers a dynamic monopsony as people are willing to accept less then their worth to avoid sitting at home. If we raise reservation wages we take that away (or lessen it).

Sean A writes:

I agree with most of the unemployment analysis. As for the "lack of demand":
Lack of "aggregate demand" is a symptom, not a cause. It is a problem of malinvestment stemming from loose credit. For example: the majority of those houses, strip malls, etc.. should have never been built or allocated loans. Are their failures the result of a "lack of aggregate demand?" I suppose then, I could start a company, Dog Feces Inc., selling dog feces and dog feces accessories; when it inevitably fails, I will cry out "If only I could have had sufficient demand."

jstaples writes:

Sean A, I am laughing my butt off at your analogy.

You are absolutely right of course. Ultimately, it all stems back to malinvestment due to government meddling. And the government is now doing everything in their power to prevent the correction. Let the banks fail, let home prices fall; let's take our lumps quick and get it over with.

jstaples writes:

David, I really appreciate your analysis here. I like where you broke things down to show that the unemployed may only be giving up ~$4.00/ hour for unlimited free time.

Did you also consider that in addition to unemployment compensation the former worker is now also eligible for a slew of other government subsidies (medicaid, food stamps, housing assistance, reduced fee utility programs, etc...)?

It is quite possible that for many people, unemployment offers INCREASED disposable income and more free time to enjoy it. Not a bad gig for 2 years if you can swing it.

Mitch writes:

I believe that the national average for unemployment benefits is $386, and that, on average, individuals on UI replace roughly 1/3 of their lost income. Seems like a pretty good incentive to get a job, on average, to me.

wintercow20 writes:

Where do UI payments come from? Are we trying to argue that the workers who are having these funds withdrawn from their checks would not have spent these funds otherwise? That they would have put them under a mattress? Then why not argue that workers don't need any of their salary? Let's increase UI premiums on every worker to 100%, and then give those payments to the unemployed. That will fix the aggregate demand problem ...

Julia writes:

I'm 55. I have a master's degree and over 20 years of experience...I was laid off from an administrative job over 1 1/2 years ago... I would be willing to accept less than what I made in order to receive a good benefit package (retirement, health care insurance, paid vacation). I'd probably consider any job offer--if I got one. Unemployment only gives me half of what I took home...so we've cut back on eating out, buying clothes, household items, vacations... When 300 people apply for a job, and six are highly qualified, even if you are in the top three, it's not good enough...there's only one job available.

Taking into account cost of sitting on the sidelines of productive society while your skills and self esteem erode, most people are not ENJOYING unemployment as though it were a paid vacation. There are bills, obligations, lifestyle changes, loss of self esteem, loss of routine, depression, stress...

Do not blame the victims.

HispanicPundit writes:

I know 5 people currently on unemployment insurance and I can honestly say that ALL of them are intentionally gaming the system. They are turning down jobs and instead preferring to live off of unemployment.

This is one of the main reasons why I lean libertarian on economic grounds: I've lived poverty, and the libertarian view of poverty is alot closer to reality than the romanticized liberal view.

Josh writes:

I know such a person. A friend of mine's wife lost her job about 6 months ago and she is currently collecting unemployment benefits. To help make ends meet, the husband got a 2nd job (at a grocery store, a job the wife is easily qualified for) because if the wife went back to work, it wouldn't pay for itself with the loss of her unemployment benefits.

Comments for this entry have been closed
Return to top