David R. Henderson  

Unemployment Insurance and Paul Krugman: Stating the Issue Correctly

PRINT
Imagine there's no economic in... Minimum Wage Increase Will Red...

Recently, there has been a brouhaha in the blogosphere over comments made by Paul Krugman and then responded to critically by Russ Roberts and Bob Murphy. Chris Dillow then defended Krugman.

The issue--and everyone on both sides agrees that this is the issue--is whether Krugman is being hypocritical in his discussion of unemployment insurance.

That's where the agreement ends. Russ Roberts and Bob Murphy thinks he is being hypocritical and Chris Dillow thinks he is not.

A quick background: In his textbooks, Krugman has laid out quite clearly how unemployment insurance gives people an incentive to be unemployed. This claim is not controversial among economists. See here, for example, for Lawrence Summers' statement of the perverse effect of unemployment insurance on unemployment. Incentives matter. Indeed, that's how Krugman discusses the issue--as one of incentives. Roberts quotes Krugman in his 2010 book, co-authored with Robin Wells and Kathryn Graddy, Essentials of Economics:

People respond to incentives. If unemployment becomes more attractive because of the unemployment benefit, some unemployed workers may no longer try to find a job, or may not try to find one as quickly as they would without the benefit. Ways to get around this problem are to provide unemployment benefits only for a limited time or to require recipients to prove they are actively looking for a new job.

Again, I remind you that this claim is not controversial. That's why Russ Roberts found it so shocking that Krugman recently wrote the following:
There's a sort of standard view on this issue, based on more or less Keynesian models. According to this view, enhanced UI actually creates jobs when the economy is depressed. Why? Because the economy suffers from an inadequate overall level of demand, and unemployment benefits put money in the hands of people likely to spend it, increasing demand.

You could, I suppose, muster various arguments against this proposition, or at least the wisdom of increasing UI. You might, for example, be worried about budget deficits. I'd argue against such concerns, but it would at least be a more or less comprehensible conversation.

But if you follow right-wing talk -- by which I mean not Rush Limbaugh but the Wall Street Journal and famous economists like Robert Barro -- you see the notion that aid to the unemployed can create jobs dismissed as self-evidently absurd. You think that you can reduce unemployment by paying people not to work? Hahahaha!


The shocking thing is NOT the first paragraph. The idea that UI can create jobs when the economy is depressed by increasing aggregate demand is an intellectually respectable one. What Roberts, and later Murphy, found shocking was paragraphs two and three above. In them, Krugman completely disowns the claim about incentives that he made in his textbook.

As Roberts put it:

There's nothing wrong with arguing that extending unemployment benefits is a good idea. There's nothing wrong with arguing that extending unemployment benefits might reduce unemployment [Roberts added the word "benefits" here but that's clearly a typo] by increasing aggregate demand. But how do you argue that your opponents are ideologues because they believe the opposite-that paying people to be unemployed increase[s] unemployment when you yourself have conceded that that idea is true?

Murphy writes:
Before continuing, let's be clear at the rhetorical devices Krugman uses here. First, he sets it up as the "standard view" that extending unemployment benefits (in a depressed economy) will boost job growth, through Keynesian demand-side effects.

Then, Krugman racks his brains trying to figure out how somehow could possibly disagree with this "standard view." He says "I suppose" you might worry about the larger budget deficit that this would cause. He doesn't offer any other possible mechanism by which someone might oppose it.

Finally, Krugman says that that's not the argument that opponents are using. Instead, they are relying on a supply-side argument, claiming that it would reduce the incentive to work if you paid people not to work. In the context, it is clear that Krugman thinks this is NOT a good objection to the "standard" Keynesian view.


Murphy then adds:
To be sure, a Keynesian like Krugman could argue that in the middle of a big economic slump that such supply-side issues are of only minor importance, and get trumped by demand-side factors. But that's not at all the argument Krugman is making in this latest blog post. Instead, he is making it sound like Barro et al. are grasping at straws, and not even relying on a coherent argument (such as fear of bigger deficits) when trying to oppose extension of unemployment benefits.

So here's where we are so far. Both Roberts and Murphy have made it clear that they are not accusing Krugman of hypocrisy for advocating extended unemployment insurance benefits during an economic slump or slow recovery. They are astounded, and certainly implicitly accusing him of hypocrisy. for disowning an argument that he himself has made.

Now to Chris Dillow's defense of Krugman. Here's Dillow's first sentence:

Paul Krugman is being accused of hypocrisy for calling for an extension of unemployment benefits when one of his textbooks says "Generous unemployment benefits can increase both structural and frictional unemployment." I think he can be rescued from this charge, if we recognize that economics is not like (some conceptions of) the natural sciences, in that its theories are not universally applicable but rather of only local and temporal validity.

Do you see what Dillow did? He misstated the Roberts/Murphy charge. They had both granted that a Keynesian could argue, completely consistent with his own views, that the government should extend insurance benefits. That's not why they're accusing Krugman of hypocrisy, as they are carefully to make clear. So Dillow starts off by setting up a straw man. I've noticed in the various comments on his post that no one seems to have caught his straw man. Is it just possible that none of the commenters bothered to check what Roberts and Murphy actually wrote?


Comments and Sharing





COMMENTS (33 to date)
Jon Murphy writes:
Is it just possible that none of the commenters bothered to check what Roberts and Murphy actually wrote?

Before I begin, I fully realize the irony of what I am about to say.

With that out of the way, now this:

Prof. Henderson, I think you're right in your question. I would not be surprised if many (I will refrain from saying "all") of the commenters did not read Dr. Robert's and Dr. Murphy's articles. In my anecdotal experience commenting frequently at Cafe Hayek and Carpe Diem (so take this for what it's worth), I find that a decent number of commenters, whether they support or oppose the post in question, do not actually read the links to which the blogger is talking about. Some just look for validation of their own personal bias. Some just want to hijack the conversation for their own ends. Some seek to create strawmen of their own. In same cases, I have been guilty of these crimes myself.

I comment primarily on Cafe Hayek and Carpe Diem, and (less frequently) here. Primarily on those two blogs (no so much here), I find the dissenters are the kind of people whom I am not writing to respond to (although my own giant ego often compels me to respond to them). Without knowing the quality of commenters on Dillow's blog but only acting off my own experience (which I know is illogical), I would not be surprised if they did not read the links, and thus would not see the strawman.

Dale writes:

"In them, Krugman completely disowns the claim about incentives that he made in his textbook"

No. In his comment Krugman disowns the claim that incentives in this manner are the primary driver when the economy is depressed(a claim he did not make). The entire section is under the condition "when the economy is depressed", and if you add that to every sentence of the paragraphs discussing the issue the intent becomes obviously clear.

Krugman should not have to restate the premise he is working under if that premise does not change.

Additionally, the what Krugman is saying in the third paragraph is not making a positive argument for no incentive structure, but making a negative argument against the WSJ and Robert Barro position. Unless he had previously made an argument for "unimployment insurance cannot possibly increase employment ever" then this again is not some sort of hypocrisy.

It should be pretty easy to read this intent out of those last two paragraph. How did you get the idea that he disowned his textbook which contained the non-depressed conditional, when he wrote that, clearly talking about depressed conditions?

Asking such a question on the differences between a depressed and non-depressed economy might be a reasonable inquiry for an undergraduate, graduate student, or well anyone "not a macro-economist". But surely Russ Roberts and Robert Murphy, who are supposedly well versed enough to cover macro economics for their respective organizations shouldn't have a problem with it. And certainly they shouldn't get so confused as to recognize that the two situations (depressed vs non) can be seen as different even if they don't agree that they are.

Jon Murphy writes:

Dale, the incentives do not change in a depressed vs. non-depressed economy.

TycheSD writes:

I think Krugman could be in the same boat as Bruce Bartlett, in the belief that different times call for different economic prescriptions. Bruce Bartlett, once a famous supply sider, now sounds more Keynesian.

Krugman, in the past, obviously believed, as most people thought, that extending unemployment benefits for long periods of time would produce an incentive for people not to look for a job. But, as folks even on the opposite end of the political/economic spectrum concede (See AEI), these are not normal times. There are dangerously high numbers of the long-term unemployed - still. Therefore, it would make sense to continue paying the benefits until the job prospects for these people improve. Most see structural problems in the job market as a major cause of long-term joblessness.

TycheSD writes:

I think Krugman could be in the same boat as Bruce Bartlett, in the belief that different times call for different economic prescriptions. Bruce Bartlett, once a famous supply sider, now sounds more Keynesian.

Krugman, in the past, obviously believed, as most people thought, that extending unemployment benefits for long periods of time would produce an incentive for people not to look for a job. But, as folks even on the opposite end of the political/economic spectrum concede (See AEI), these are not normal times. There are dangerously high numbers of the long-term unemployed - still. Therefore, it would make sense to continue paying the benefits until the job prospects for these people improve. Most see structural problems in the job market as a major cause of long-term joblessness.

Daniel Kuehn writes:

I agree with Dale. I'm a little shocked at how this is taken off, given the flimsiness of the case. This talk of hypocrisy (here) and a lack of intellectual credibility (Russ) is not appropriate IMO.

Daniel Kuehn writes:

What he's pushing back against is this idea that labor supply effects must equal or are clearly equal to aggregate employment effects. He should push back against that. Nowhere in the piece does he say "there are no labor supply effects".

Rob Rawlings writes:

Dillow said:

"Paul Krugman is being accused of hypocrisy for calling for an extension of unemployment benefits when one of his textbooks says "Generous unemployment benefits can increase both structural and frictional unemployment."

David Henderson says:

"Both Roberts and Murphy have made it clear that they are not accusing Krugman of hypocrisy for advocating extended unemployment insurance benefits during an economic slump or slow recovery. They are astounded, and certainly implicitly accusing him of hypocrisy. for disowning an argument that he himself has made."

and then "Do you see what Dillow did? He misstated the Roberts/Murphy charge"

I'm just not seeing that. Dillow's statement seems to capture Murphy's and Robert's charges quite fairly in my view. (If anything it is Henderson's statement of the disagreement that is misleading because I don't think Krugman is "disowning" his earlier text book view - he's just stating that it doesn't apply in a deep recession).


Daniel Kuehn writes:

Another way of putting my point is that I think people are confusing "Krugman shouldn't laugh at labor supply effects" with "Krguman shouldn't laugh at labor supply effects determining the unemployment rate response".

The former is defensible, the latter is not, and the latter is all Krugman's done as far as I can tell.

Not a cause for a tirade of name-calling.

Art Deco writes:

[Comment removed for supplying false email address and for rudeness. Email the webmaster@econlib.org to request restoring your comment privileges. This is your final notice. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

BZ writes:

I find Dillows statement far more shocking than Krugmans.

The Law Of Gravity says that things with mass are drawn together. When you see a helium balloon rising into the air, you are not seeing a Gravity Law "not universally applicable but rather of only local and temporal validity". We all understand that. And yet, that's what Dillow seems to be saying about labor market incentives.

@DK - correct, Nowhere in the [Krugman] piece does he say "there are no labor supply effects". However, he does seem to say that mentioning those effects as part of UI reasoning is somewhere south of nutty.

Mike Rulle writes:

I have mentioned the following on this website before. I think we obsess over what PK has to say. I do not why so much time is spent on him. He has chosen in his columns to be primarily a proselytizer and activist. Whatever his intellectual capabilities as an economist may be, he has chosen to bait those on the right with political rhetoric. By doing so he brings more attention to his writings. He does not care if he is a hypocrite. Why should we care? If he writes something, respond as one would to any political writer posing as an economic thinker. He is winning the war of words by forcing us to argue the obvious, thus making the obvious appear less so---or why would we be arguing so hard?

Thought experiment. Imagine he is a new political opinion writer, named, say, Leonard Skinner. How would we respond to his comments?

Daniel Kuehn writes:

BZ -
Not seeing that. He refers to "the notion that aid to the unemployed can create jobs dismissed as self-evidently absurd" as south of nutty, not your claim about "mentioning those effects as part of UI reasoning".

If you're going to attribute that to him I think you need to back it up with something.

He may be reading Barro wrong, but he's definitely not being hypocritical.

Dale writes:

John. No one said that incentives changed in a depressed vs non-depressed economy. Though they might, tbh i don't have much experience in looking at this situation so i won't pretend to make that call.

It was stated that the aggregate effect might change. What Krugman is arguing against Barro is that the incentive structure does not have to dominate the demand effects and specifically that its ridiculous to claim that it must

Such a position does not renounce his prior work and if you're going to attack it then you need to attack it on the grounds of showing the demand effects are necessarily lower than the supply effects or some sort of empirics to show that the effect is dominating. My understanding is that the empirics tend to support Krugman's position(but again, not an expert on this) as should a reasonable folk theory for depressed economies.

Certainly some would argue otherwise but that does not mean Krugman has renounced anything.

David R. Henderson writes:

@Mike Rulle,
I have mentioned the following on this website before. I think we obsess over what PK has to say.
Yes, you have. Have you also noticed that you have had virtually no effect on the number of PK posts I comment on? You might want to think about why. Although I’ve already said why. If you missed it, you might want to reread it.
Thought experiment. Imagine he is a new political opinion writer, named, say, Leonard Skinner. How would we respond to his comments?
Not a good thought experiment. If he were not read by a lot of people, and if his ideas weren’t widely held, I wouldn’t comment. Now let’s get back to the fact that he is widely read.

Josh Wexler writes:

Seems to me Krugman's critics could read him more charitably in this instance (even if he doesn't afford the courtesy in kind). Krugman's words are consistent with following (cogent) belief set:

1. One effect of UI is to increase incentives to remain unemployed which tends to increase i employment.

2. A countervailing effect of UI is to increase demand, which may outweigh the perverse incentives enough to make extending UI a net gain in employment.

3. These right wing ideologues only talk about effect #1 and never even consider effect #2, so they are too quick to dismiss the utility of UI. Idiots!

That may be unfair, but there's nothing hypocritical in it.

Jon Murphy writes:

Dale,

If incentives don't change, then I don't understand your objection.

Bob Murphy writes:

Daniel Kuehn, the only name-calling I see in this episode is Krugman calling Russ and me "idiots" with a strikethrough.

Emil writes:

[Comment removed pending confirmation of email address and for being ad hominem. Email the webmaster@econlib.org to validate your email address. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Rich Berger writes:

Imagine if Dr. Krugman's second paragraph had read:

You could, I suppose, muster various arguments against this proposition, or at least the wisdom of increasing UI. You might, for example, be worried that people respond to incentives. If unemployment becomes more attractive because of the unemployment benefit, some unemployed workers may no longer try to find a job, or may not try to find one as quickly as they would without the benefit. Ways to get around this problem are to provide unemployment benefits only for a limited time or to require recipients to prove they are actively looking for a new job.

This might have confused or upset his faithful readers, who cling to his every pronouncement. I think he was just showing concern for them.

Gene Callahan writes:

I'm sorry, but he does not "disown" his previous argument AT ALL: zero, zilch, nada. You have not even shown him mentioning it!

Gene Callahan writes:

@Jon Murphy: "Dale, the incentives do not change in a depressed vs. non-depressed economy."

"If incentives don't change, then I don't understand your objection."

Yes, it is celar that you don't understand. But it is really not hard. If you are tied to to a large helium balloon, you ought to worry about floating away. But, if at the same time you have a much larger lead ball tethered to your ankle, the effect of the balloon can be ignored.

That does not mean that the behavior of helium suddenly has changed!

Gene Callahan writes:

Murphy: "Finally, Krugman says that that's not the argument that opponents are using. Instead, they are relying on a supply-side argument, claiming that it would reduce the incentive to work if you paid people not to work. In the context, it is clear that Krugman thinks this is NOT a good objection to the "standard" Keynesian view."

Note what Murphy does here: he *completely* re-writes what Krugman actually said, to shove the "hypocritical" words he wants Krugman to say into K's mouth!

Eric Falkenstein writes:

Since when has a second-order effect in a poor predicting science like economics 'obviously' trump a first order effect? This is the same guy who said the $800 stimulus in the crash didn't work because it was mainly transfers, not 'spending.'

Scott Sumner writes:

On the question of whether extended UI reduces unemployment I obviously agree with Barro. But I actually don't think Krugman was being unfair to Barro, for reasons I explain here:

http://www.themoneyillusion.com/?p=25996

Jon Murphy writes:

@ Gene Callahan: "Yes, it is celar [sic] that you don't understand. "

I should hope it's clear as I admitted it :-P

But on a serious note, I really do not understand your objections. What Krugman seems to be saying is that the arguments put forth by Barro et al, that the incentive system created by UI may result in longer unemployment or fewer-people-than-seeking is absurd. He seems dismissive of them.

Michael Byrnes writes:

Jon Murphy wrote:

"If incentives don't change, then I don't understand your objection."

The opportunity cost of collecting unemployment is extremely dependent on the state of the economy, isn't it?

As far as I can tell, the incentives argument is that people who could get a job choose not to because they are able to collect unemployment.

If you assume that all unemployed people are the same and all potential jobs and compensation are the same, and the state of the economy has no effect on any of the above, then I suppose all that matters is the incentive effects. But those are not realistic assumptions.


Jon Murphy writes:

@ Michael Byrnes:

As far as I can tell, the incentives argument is that people who could get a job choose not to because they are able to collect unemployment.

That is incorrect. The argument actually is, since UI makes unemployment less painful, it reduces the need for unemployed to find a new job quickly. Thus, people stay unemployed longer as they search for a job they want (to the extent they can).

Glen Smith writes:

I can construct an argument for extending unemployment benefits using a more Austrian or a monetarist view but not one against extending unemployment in a Keynesian influenced view.

TallDave writes:

"Hypocrisy" is the compliment vice pays to virtue, the failure of one's actions to echo one's exhortations.

In that here he merely frames his argument to fit his ideology in contradiction of his own prior claims in order to traduce his opponents, more correct descriptions for the method employed by Krugman here would be "prevarication," "tendentious," or simply "NYT columnist."

Aaron Zierman writes:

I believe Russ Roberts clears some of his thoughts up more thoroughly in his most recent post.

I agree completely with Roberts. Krugman is misleading as usual and dismissive of the argument rather than being willing to explain and truly break things down.

David R. Henderson writes:

@Scott Sumner,
On the question of whether extended UI reduces unemployment I obviously agree with Barro. But I actually don't think Krugman was being unfair to Barro, for reasons I explain here
I’m not sure if you’re addressing me or commenters on this post, but I wasn’t claiming Krugman was being unfair to Barro. I was claiming--and still do claim--that Krugman disowned his own views on the incentive effects of extending UI.

Michael Creswell writes:

@ Gene Callahan:

"I'm sorry, but he does not "disown" his previous argument AT ALL: zero, zilch, nada. You have not even shown him mentioning it!"

Gene, I think that is the point: Krugman fails to disclose his own stated views on the subject.

Comments for this entry have been closed
Return to top