Casey Mulligan on Redistribution, Unemployment, and the Labor Market
Dec 3 2012

Casey Mulligan of the University of Chicago and the author of The Redistribution Recession, talks with EconTalk host Russ Roberts about the ideas in the book. Mulligan argues that increases in the benefits available to unemployed workers explains the depth of the Great Recession that began in 2007 and the slowness of the recovery particularly in the labor market. Mulligan argues that other macroeconomic explanations ignore the microeconomic incentives facing workers and employers.

Casey Mulligan on Cuba
Casey Mulligan of the University of Chicago talks with EconTalk host Russ Roberts about life in Cuba. Mulligan, who recently returned from a trip to Cuba, discusses the economy, the standard of living and some of the peculiarities of communist...
David Autor on Trade, China, and U.S. Labor Markets
David Autor of MIT talks with EconTalk host Russ Roberts about the fundamentals of trade and his research on the impact on workers and communities from trade with China. Autor's research finds large and persistent effects on manufacturing jobs and...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Greg G
Dec 3 2012 at 10:28am

So the story here is that our failure to make the unemployed suffer enough has been the main factor in making the Great Recession and recovery from it as bad as it has been. Since the social safety net for the unemployed is greater than it has ever been, and since we know that will cause some people on the margin to decline to work, it follows that this explains “the depth of the Great Recession.” Wow.

I was, at least, looking forward to a discussion of some of the more obvious and stronger arguments against such a theory. That never happened.

Reinhart and Rogoff make a really good case that recoveries from major financial crises are always brutal. What does Mulligan think of this research? You won’t find out from this podcast.

The nineteenth century had some long and brutal depressions without a government safety net for the unemployed. Wages were free to fall as far as the market would take them. How does this theory explain that? You won’t find out from this podcast.

Most unemployment is in low skill jobs and a great deal of that is due to technology and globalization. At least Russ did ask about the possibility that those jobs just aren’t out there anymore. Mulligan replies that, since some people got jobs we can’t say there were “no jobs.” Well the counter argument isn’t that there were “no” jobs, it is that there were clearly not nearly enough jobs.

Mulligan concludes from the fact that married people have lower unemployment rates, that effect is caused by their ineligibility for government benefits due to the income of a spouse. Maybe the the same personal qualities that make for a successful marriage make for a successful employment situation. But why look any further once you find a story that fits your ideology.

Russ concludes by complimenting Mulligan on his courage. Is it really that courageous to make this kind of argument at The University of Chicago?

And Russ invokes Hayek in suport of Mulligan. I would be more inclined to think of Hayek as warning us against thinking we could single out one factor such as benefits to the unemployed and speak with this much confidence about its effect on the larger economy.

László Sándor
Dec 3 2012 at 11:31am

You are not doing justice to our field, our science. Economists who worried less about Congress expanding UI were not (only?) sketchy macro-Keynesians. It is not about the multiplier.

Congess listened to Till von Wachter, for instance, multiple times. You can find his testimonies at:

And there is a reason why Congress asked him:

Really, economists don’t talk about both sides of the trade-offs?

I guess Mulligan wants to see huge average tax rates, to allow for slowly rising MTRs and still give something to the low earners. Guess what, other people did think about the tradeoffs!

You know, this is why this literature has *optimal* in its name. It is not new that there are constraints in the constrained optimization problems…

By the way, you could perhaps appreciate that the CEA chair also happens to have worked on the long-term unemployed and incentives they face.

[comment revised by commenter–Econlib Ed.]

László Sándor
Dec 3 2012 at 7:31am

You are not doing justice to our field, our science. Economists who worried less about Congress expanding UI were not (only?) sketchy macro-Keynesians. It is not about the multiplier.

Congess listened to Till von Wachter, for instance, multiple times. You can find his testimonies at:

And there is a reason why Congress asked him:

Really, economists don’t talk about both sides of the trade-offs?

I guess Mulligan wants to see huge average tax rates, to allow for slowly rising MTRs and still give something to the low earners. Guess what, other people did think about the tradeoffs!

You know, this is why this literature has *optimal* in its name. It is not new that there are constraints in the constrained optimization problems⬦

By the way, you could perhaps appreciate that the CEA chair also happens to have worked on the long-term unemployed and incentives they face.

[comment revised by commenter–Econlib Ed.]

Grieve Chelwa
Dec 3 2012 at 1:18pm

@Greg G:

“Reinhart and Rogoff [R and R] make a really good case that recoveries from major financial crises are always brutal.” It is true that R and R show that recoveries from financial crisis-induced recessions are always sluggish. The trouble’s that R and R don’t discuss the mechanism through which this happens. John Cochrane at the Booth School of Business has argued convincly that perhaps financial crisis-induced recessions last longer because of govt’s misguided policy interventions following the financial crisis. Interventions meant to fix the evil bankers: Frand-Dodd, Tobin Tax over in the EU, higher tax rates for the wealthy and corporations, higher capital requirements for banks, QE 1,2 and infinity, etc…(see here, here and here).

Then you go on to claim that “the nineteenth century had some long and brutal depressions without a government safety net for the unemployed. Wages were free to fall as far as the market would take them.” Can you kindly provide empirical evidence to support your claim that the nineteenth century had “brutal” reccessions and that wages were in free-fall. For one thing, work by Christina Romer shows that: The declines in industrial production in the recessions of 1920, 1929, and 1937 were larger than in any recessions in the pre– World War I and post–World
War II periods.”

I agree with you that the lower skilled are more likely to be unemployed, it is not true that this is due to technological progress. The US economy over the last century achieved incredible feats interms of technological advancement and innovation but the unemployment rate over this period hardly budged (as a matter of fact it trended downwards for most of the 20th Century). And in regards to your claim that globalization causes unemployment, the liberal and nobel prize winning trade economist Paul Krugman has thoroughly debunked this line of logic in his book from the 1990s “Pop Internationalism”. .

Dec 3 2012 at 2:23pm

Its very interesting that Mulligan did not apply his theory to other recessions, that would seem like a very important part of this research. What percentage of the increase in MTR’s go to unemployment insurance? It would very difficult to convince politicians and the public to reduce benefits during times of recessions. I would be interested in Russ’s and Mulligan’s thoughts on a comparison of our current system vs a negative income tax?

Sonic Charmer
Dec 3 2012 at 2:59pm

I think it’s hilarious that it’s controversial in the field of economics to suggest that paying people to be unemployed might lessen economic growth.

Greg G
Dec 3 2012 at 3:07pm


Even though R & R don’t develop a theory of why recoveries after financial crashes are sluggish, their research shows that this pattern of slower recovery precedes by centuries the kind of social safety net that is said to cause that sluggishness in this podcast. We don’t need a special theory to expect causes to precede their effects.

There was a six year depression starting in 1837 and a very serious depression from 1873-1879. I am no expert on industrial production but I am skeptical that citing declines in industrial production in the 20th century is the best way to do comparisons to the hardships of the depressions in the more agricultural economy of the nineteenth century.

I did not say that wages were “in free-fall” in those 19th century depressions. I said they were free “to fall” without being impeded by the kind of government interventions that we are talking about here.

I agree that technological progress and globalization are good things. But I think they come with trade offs and one of those trade offs is the loss of low skilled jobs in this country.

Dec 3 2012 at 3:08pm


Reinhart and Rogoff’s finding is that recovery after financial crises tends to be a lot slower than after typical downturns. It could be that this slow recovery is due to the way government reacts to the crisis, rather than to the crisis itself. But it would have to be a government response that was specific to financial crises, rather than to downturns generally. I don’t think unemployment insurance, food stamps, etc. fit that bill.

Russ Roberts
Dec 3 2012 at 5:00pm

Greg G,

You ask why I complimented Mulligan on this courage. One answer is that it isn’t a lot of fun to read paragraphs like the one you wrote:

So the story here is that our failure to make the unemployed suffer enough has been the main factor in making the Great Recession and recovery from it as bad as it has been. Since the social safety net for the unemployed is greater than it has ever been, and since we know that will cause some people on the margin to decline to work, it follows that this explains “the depth of the Great Recession.” Wow.

That’s not just a distortion of what Mulligan argued. That doesn’t just ignore the amount of work that Mulligan has done to try to quantify his claim. It implies that Mulligan is a bad person. Publishing work allows people to take a cheap shot at your work–that takes a little courage. Maybe that’s too strong a word, but that’s all I meant.

I read a number of criticisms of Mulligan before the podcast and raised some of the objections raised by the critics. Sorry I couldn’t get to all of them. As for the argument about financial crises, I do hope to have a podcast on that soon.

Greg G
Dec 3 2012 at 5:25pm


I did not mean to suggest that Mulligan is a bad person, just that I thought he made a bad argument. I don’t doubt he sincerely believes these policies are bad for the unemployed in the end. I apologize for coming on too strong and leaving the impression that I did. My point about courage was just that this is pretty orthodox stuff for the University of Chicago. No doubt it would take a lot of courage at many universities and it still does whenever you leave that campus as you point out.

Jan D
Dec 3 2012 at 5:30pm

The host’s argumentation seemed extremely unconvincing to the point of rediculousness, it seemed to me that instead of honestly arguing a point he was basically just creating doubt and shooting little holes (if at all) into the “conventional wisdom” (which I think is still basically correct).

As for the rediculousness, that really reached a peak when he talked about sales tax (VAT) demotivating people from working (seriously?!). SERIOUSLY?!?

Are you saying that if stuff gets more expensive because of a VAT increase I’m going to want to work less? And how does stuff being more expensive make me want/need it less?
Is this guy one of those “economists” who believe that if someone can’t afford something that’s because he doesn’t really want it? The sort of person who thinks that being poor is a voluntary choice? Or that unemployment is voluntary? I’m amazed that people in the 21st century still believe that and get away with it so easily.

I think *some* economists think too much of their field and of it’s applicability to the real world. In most areas of life there are more important things than economics and economists trying to rationalize and answer everything through economics make rediculous conclusions just like when literary critics try to talk about science. And we all know what that looked like.

I’d like to see what economic theory would explain a violent uprising/mass looting because of a government’s anti-social (but “economical”) policy.
I’d like to see this guy walk up to unemployed people and tell them straight in their face what he’s saying here. I know a lot of unemployed people (in a different country) and and they’re all trying to find a job hard. I’ď like to see what economic theory explains where all the unemployed people who don’t want a job are hiding, because I haven’t met any.

Dec 3 2012 at 7:35pm

Very interesting and a lot of food for thought. I never really took these arguments seriously, in part because I spent some time unemployed a few years ago and I was miserable and worried throughout the duration of my unemployment.

There is one issue I wish you had been able to address during discussion of the JOLTS data and the decline in the number of “quits”.

My understanding is that, most of the time, there is a fairly high rate of people quitting their jobs – usually, far more people are quitting their jobs than are laid off. And, the main reason why people quit their jobs is to move to a better job. In that sense, a high quit rate is actually a sign of a healthy, dynamic economy.

From late 2008, the number of quits dropped considerably, falling below the number of layoffs (which increased) and not exceeding layoffs until early 2010. Even today, the quit rate is lower than it had ever been from 2002 through 2008.

My interpretation of the low quit rate is that it is a sign of a poor job market. It’s generally easier to find a job if you are already employed, and more generous unemployment benefits should not have much of an incentive effect on people moving from one job directly to another.

Let’s suggest two hypotheses for the high unemployment rate:

1. It is driven by the increased benefits associated with unemployment.

2. It is driven by a terrible job market.

To me, the low quit rate is highly suggestive of #2, as #1 should not have much of an effect on the incentives of those who quit their job to take a different one. Given this, I found Mulligan’s suggestion that the low quit rate was simply a matter of employers letting people go who would otherwise quit to be less than convincing, and I would have liked to hear some discussion of the reduced rate of job to job moves.

That said, I’m sure his scenario (unemployment rate is higher than it would be if unemployment benefits, etc. had not been extended) is correct to some extent. How much, and whether or not we would be better off without extended benefits is another question (and one that is not easy to answer!)

Mike Germaine
Dec 3 2012 at 8:33pm

I find Mulligan’s arguments plausible but not quite convincing. I think it is really hard to prove causality here and especially the degree of causality.

With that said, the harsh negative reaction to these ideas is annoying. People just seem to be downright offended that someone would suggest the unemployed might respond to incentives that keep them unemployed. The visceral reactions by many haughty left wingers leads me to believe Prof. Mulligan is on to something.

Josiah Neeley
Dec 3 2012 at 9:53pm

It’s plausible that increasing unemployment benefits etc. would increase the unemployment rate. But we would expect this effect to be attenuated during a downturn.

For example, suppose that there are 3 million people who decide not to work because they’d rather just live off the UI benefits. If the number of unemployed workers is roughly equal to the number of job openings, then the UI benefit will increase unemployment by 3 million. On the other hand, in a situation like the one we have today where you have 3-4 unemployed workers per job opening, then any openings that the 3 million don’t take will likely be filled by some other unemployed worker, and the total number of unemployed won’t be much higher than it would be without the UI benefit (note: this effect is separate and independent of any stimulative effect UI benefits may or may not have).

To the extent that Prof. Mulligan’s estimates are based on the disemployment effect that things like UI have when we are close to full employment, then, they are likely to greatly overstate the effect they are having today.

Dec 3 2012 at 10:04pm

I have to agree with Greg G on many points, although not the original tone, as he later recanted. I do appreciate Mr. Mulligan’s effort to bring this discussion out, even in unpopular times.

However, I find that the kernel of truth at the base of his argument may not extend as far as he would like. Stating that incentives extending the unemployment benefits may make some people stay on unemployment longer than they otherwise would (sometimes by choice, although not always) does not extend to mean that an alternate of falling wages to regain higher employment levels would result in increased consumption (essentially his argument about the auto market near 8:00). i.e. – lower wages for work does not equate to increased consumption sales in my mind. Especially when he mentions that the “take home pay” might be nearly equivalent in the two instances.

In this example, what does increase is the public debt, which is not a desirable end either, so public policy should weigh the problem of debt against the problem of extreme poverty in the absence of a safety net for some of the population – admittedly a decision based on a judgement call between the lesser of two evils.

Finally, in reference to the R&R research about financial crisis-induced recessions, my take (thanks to the many discussions on Hayek, and some limited reading of his work on follow-up) would be that these recessions are by definition recessions caused specifically by large-scale over-indebtedness. Rather than one industry having a bubble, and the rest of the economy absorbing the shock, when overall debt is too high, everyone needs to replenish their rainy day funds simultaneously. Coincidentally, it looks a lot rainier, so folks are likely more conservative about how much they’ll need in that fund. And all this occurs just at the time when producers and employers (and tax collectors) would like everyone to bounce back to get sales back up. Seems like a formula for a longer recovery to me. Perhaps Keynes’ animal spirits are just the logical reaction of people to Hayek’s over-expanded credit market.

At any rate, the nature of the recession, and not the nature of the response to it still seems more convincing to me. Not that adding debt from the private ledger to the public one is a great idea, but I don’t see the way this ties as directly to the labor market as Mr. Mulligan has argued.

Dec 4 2012 at 10:25am

How many jobs were unfilled during the Great Recession? Why did the unemployed choose not to apply? Did the unemployed physically move to locales where employment was more plentiful? As we now approach permanent , lifetime unemployment benefits (aka Disability) for the masses (as in nearly 9,000,000, listen to Autor and Roberts), what would ever cause them to seek employment again?

Dec 4 2012 at 10:37am

“Most unemployment is in low skill jobs and a great deal of that is due to technology and globalization.” -Greg G

Do you believe that minimum wage has any effect?

Dec 4 2012 at 12:21pm

Fascinating, provocative, and overall excellent podcast. Thanks.

A couple of thoughts/reactions. First, thanks for linking to the Ohanian podcast (8/20/12) above — very appropriate. Although it wasn’t discussed with Mulligan, Ohanian posits that government interventions during the 1930s depression help account for the duration of the depression — focussing largely on government collaboration with unions to keep wages high. In his podcast, however, if I recall correctly, Ohanian didn’t seem to extend that analysis to the sluggish recovery from the 2008-09 great recession. Mulligan’s thesis, however, applies a similar explanation to the current recovery, though he attributes the downward wage rigidity to expanded safety-net policies rather than union/government interventions (though we could cite a few of those as well over the past several years).

Second, I was glad the discussion touched on extending Mulligan’s work with international data. The basic idea, if I understood correctly, would be two compare recession/recovery experiences across countries accounting for differences in changes in safety-net interventions. Cross-section difference analysis of this type is certainly sensible. But I also wonder about straight cross-sectional state analysis — examining differences in employment/unemployment across countries against various explanatory variables including the magnitude of safety-net programs and other sources of implicit/explicit marginal taxation on the decision to work and the income work generates.

In this regard, I can’t help thinking of Korea, a country which the OECD indicates lags other developed economies in the extent of its welfare programs, and where one also finds high employment and examples of types of employment that essentially don’t exist in the United States. For instance, the self-employed street cart fruit vendor visible in any Korean city might be regarded as functionally unemployed in a U.S. context, and the same individual here would do far better with safety-net income than undertaking the same type of job. Such jobs in Korea (a country with an average per capita income of $30-35K) are typically, or at least often, transitional, between better employment opportunities. Here, UI and the other programs that Mulligan documents are transitional.

Mort Dubois
Dec 4 2012 at 12:56pm

It’s worth noting, on the subject of quits/layoffs, that the company bears a significant part of the cost of the unemployment benefits paid to the laid off worker. At least that’s how it works in Pennsylvania. I laid off half of my workers in the fall of 2008, and my unemployment insurance tax rate went up by 10 times as a result. Lots of skin in the game for me as an employer – I don’t lay people off just for chuckles. On the other hand, if you ask a group of 8 workers whether they would rather I laid off 2 of them or cut everyone’s wages by 25%, the vote would be 6 to 2 for the layoffs. They already know who will go. And would much rather keep their own wages constant, possibly due to debt overhang but definitely due to basic human nature. After all, the laid off people are gone and out of sight. The others feel validated as workers and are much more cheerful than they would be in the wage cut scenario, with commensurate increase in productivity. I, the boss, would much prefer cheerful workers, as well as the opportunity to get rid of the lowest performers. It’s worth the extra unemployment costs. That could well be an explanation for the increase in productivity seen throughout the economy. “To encourage the others….”

Justin P
Dec 4 2012 at 2:11pm

I do have to agree with Mike Germiane; “The visceral reactions by many haughty left wingers leads me to believe Prof. Mulligan is on to something.”
I don’t think there is any doubt that providing more goodies to the unemployed will decrease their willingness to stay unemployed, on the margin. The real question is, how big is the margin. I think Russ and Dr Mulligan pointed out more than a few times, it’s an empirical question.

I think Russ is right, Dr. Mulligan does indeed show some guff by pointing it out. Most Academics are swing left, and usualy (not always) they have a defensive reaction to anyone that questions the welfare state. I think GregG’s initial comment displayed some of that, but to Greg’s credit he retracted. But talk to most Leftists and any question of unemployment insurance usually comes back as “wanting people to starve to death.” Like charity is completely gone from society, that everyone is so into themselves that they wouldn’t help their fellow man. I think that speaks volumes about the people making that charge than anything else.

Bravo Russ for correcting your pronunciation of Nevada! It’s not like I’d boycott your program for saying it wrong though.

Dec 4 2012 at 4:03pm

Today Florida’s “Jobs Czar” stepped down because he took unemployment benefits during the recession. His sin wasn’t necessarily the unemployment benefits, but he had enough money to go to Europe multiple times and he owns two homes worth a combined $1.1 million.

Just one more example of bad incentives created by these programs.

Jonathan Bechtel
Dec 4 2012 at 6:00pm

My favorite part of the interview was Russ’s comment about how ashamed labor economists would be to make Casey’s argument.

Which got me thinking…..what would people have said if someone like Casey used the same methodologies to create a case for supply side effects as the cause of increases in economic activity?

My guess is that no one would care all that much and he’d have a lot more people publicly agreeing with him.

Maybe we’re naturally capitalists on the way up and socialists on the way down.

Dec 4 2012 at 10:31pm

I also found Prof Mulligan unconvincing, though maybe the book does a better job. It’s not that he’s entirely wrong here, there’s no question that social programs influence peoples’ decisions to work and I’m sure he’s done so good work quantifying some of the ways it does, but he pushes that explanation too far here. There was a contraction that happened simultaneously across the entire world. It’s beyond implausible that it somehow is largely explained by redistribution policies everywhere it happened, including in the US.

The University of Chicago Econ department in general has not gotten through this crisis well. Much of their recent work really looks like nothing more than trying to save a model of the economy that events have shown to be lacking.

I think the Professor would be better served framing his work as a better way to think about designing public policy to reduce frictions and encourage labor force participation.

Dec 5 2012 at 4:03am

“There was a contraction that happened simultaneously across the entire world. It’s beyond implausible that it somehow is largely explained by redistribution policies everywhere it happened, including in the US.”

Did he say that, or did he say that the reason the recession has been extended without much growth and low employment is due to the high level of relief available?

I think people are reading way too much into this as attacks on their sacred goats and probably they have ZERO real world experience with the kinds of people who take these deals.

John Thurow
Dec 5 2012 at 7:01am

OK, I will be cynical:

Economic Benefit = 1/(K * Government Programs)

Political Benefit = K * Government Programs

i.e inversely and directly proportional to, respectively

I don’t think that economics had anything to do with the increases in these programs… I think FDR and the current president figured this out.

Becky Washington
Dec 5 2012 at 10:41am

People of good conscience, whether left- or right-leaning can be the voice of reason in looking at incentives.

Incentives matter. If benefits programs result in a situation where “people find themselves in a trap, the more they work the less they have”, that’s bad.

If employers want to drop the wage and increase productivity, workers may be people who “find themselves in a trap, the more they work the less they have.” That’s bad.

I do not assume that Mr. Roberts or Mr. Mulligan are in this to blame the victim but, rather, to get us to a better understanding of our best approaches to supporting a strong economy.

I respect and appreciate the effort.

Greg G
Dec 5 2012 at 12:00pm

I want to do a little more to explain why I failed to anticipate the reaction that Russ and many others had in reading my first comment as taking a cheap shot and suggesting that Mulligan was “a bad person.” I hope to do this while still taking full responsibility for the fact that I left a comment that many people found offensive and that was less effective than it could have been in advancing my own ideas. I appreciate the fact that these podcasts and the comments that go with them are one of the few places that you can reliably find hot controversies debated politely and constructively.

I take it as a given that most people in this debate (and all who I am interested in responding to) really believe they are right. Those on both sides believe the policies they advocate would best help the economy. Both sides understand that a bad economy hurts everybody but it hurts the poor and the unemployed the most.

I don’t think anyone wants the unemployed to suffer just for the sake of seeing them suffer. But I did hear the following quote in the podcast:

“Thanks to that help, people don’t suffer as much when they are not employed as they would have if they would have experienced this ten years ago or five years ago or fifteen years ago. And when it’s less painful to be without a job there will be more people who are without a job.”

That was the basis of my reference to the idea that ” that our failure to make the unemployed suffer enough has been the main factor in making the Great Recession and recovery from it as bad as it has been.” I think there really is something of a no pain, no gain philosophy here but I recognize (even though I failed to show it earlier) that there is a sincere belief that the policies advocated here will be best for everyone in the long run.

It is true, as Russ said, that I ignored the work that Mulligan did to quantify his claims. The reason for that is that I have no quarrel with the data quantification process. My beef is with the assumptions and the logic used in deciding what the numbers mean. For example, as I explained, I accept his finding that married people are less likely to be able to access certain benefits and are less likely to be unemployed. I just don’t think it follows from that that this should be regarded as good evidence for his thesis.

I think that the reason debates like this are so sensitive is that everyone correctly realizes the wrong policies will do real harm to real people. That does not mean that either side intends real harm.

Dec 5 2012 at 2:35pm

“Instances where people would actually have to pay to go to work. In tax rate terms, their tax rate was above 100%.”

What specifically does he mean by this? Is he referring to opportunity costs related to unemployment benefits, or literally that their effective tax rates are over 100%? How is the latter possible?

Dec 5 2012 at 3:59pm

Russ, this sounds like Nobel Prize work. This is what macroeconomics is supposed to look like. Great find. Thank you for this interview.

@Greg G
You wrote,”Maybe the the same personal qualities that make for a successful marriage make for a successful employment situation.”

Thoughtful point!

@Sonic Charmer
You wrote “…it’s hilarious that it’s controversial in the field of economics to suggest that paying people to be unemployed might lessen economic growth.”

I share your feeling. I was confused when Russ started talking about the courage required to publish work which essentially makes that very statement.

@Jan D
You wrote “[Is Casey Mulligan] the sort of person who thinks that being poor is a voluntary choice? Or that unemployment is voluntary?”

You were right to pick up on this idea. It is a big concept. That is precisely what he was saying. Unemployment is voluntary. And he is correct. Drop your price low enough, and someone will hire you.

Whether being “poor” is voluntary is far more complicated because its definition varies so widely. Perhaps we can explore some of its nuances in a different post.

You wrote “My interpretation of the low quit rate is that it is a sign of a poor job market.”

Good point!

@Joseiah Neeley
You wrote “in a situation like the one we have today where you have 3-4 unemployed workers per job opening…”

I think you left out the necessary qualifier “at the present wage rates” from that sentence. What I mean is, if you say that there are more job openings than there are jobs then you are describing a surplus of labor and a shortage of jobs. If prices were allowed to adjust then neither is possible over a long period of time. If you are correct and that situation does exist–and you are correct–then something is preventing the price of labor from adjusting. That “something” is a price-support. Unemployment insurance, as it happens, is a price-support for wages.

You go on to suggest that “Prof. Mulligan’s estimates…greatly overstate the effect [Unemployment insurance] are having today.” On the contrary, unemployment insurance is precisely the kind of government tool that CAUSES a surplus of labor because it prevents the price for labor from dropping lower than the value of the unemployment insurance payments. I don’t think it is possible to “overstate” the magnitude of such an enormous price support. Casey Mulligan may even have underestimated its impact.

Mort Dubois
Dec 5 2012 at 7:47pm

More things that puzzled me: I was struck that an unemployed person who took advantage of every benefit would be replacing a high percentage of a modest income. What percentage of the unemployed population actually took advantage of these programs? How easy was it to learn about and gain access to a variety of benefits? And the bulk of unemployment happened early in the crisis. Was there a significant lag between the rise in unemployment and the appearance of the programs? If the programs existed but very few people used all of them then the argument is much weaker.

Dec 5 2012 at 10:07pm

@Mort I think you raise a good question, but if you go to the podcast on Soc Security Disability and look at the massive expansion of that program, or the massive expansion of SNAP, then I think it’s pretty clear that people did take advantage of these programs.

As to the issue of causing “pain” or not, I think that the main effect of the 99 weeks of unemployment benefits is that people keep trying to rationalize how they will recover their former job, income, and lifestyle, if only they can just wait for the right opportunity. Procrastination and hope are two important forces.

Personally, I have been downsized twice in the past 3 years. My current salary is 40% less than I made previously.

In my last job search, I immediately listed my house for sale (to sell at a huge loss) and expanded my job search to a national, rather than local search. While I was fortunate enough to find work without moving or selling my house, I have been close enough to the flame twice to see that drastic measures are sometimes necessary.

If they take away the safety blanket and people (metaphorically) sleep on the cold hard ground for a few nights, they will be increasingly willing to take a job, any job, just to start rebuilding their lives. It is unfortunate and painful on a personal level, but that is simply part of life, and the economy. Skills get re-priced occasionally. We don’t make typewriters anymore…

Dec 5 2012 at 10:14pm

The problem with the idea that wages can just drop to some market clearing level and then we’re back to growth has one major problem. Most people’s absolute debt level won’t fall in conjunction with their wage drop. No matter the economy, they still have their monthly nut!

If wages go down 10%, then the percentage of income devoted to debt service goes up and disposable income drops dramatically. That means aggregate demand falls, meaning more unemployment and so on- basically the paradox of thrift story played out involuntarily across the economy. How is that not a more compelling explanation both of why wages don’t fall as we’d really expect and also why, after a balance sheet recession, it takes a long time for real recovery to take hold?

Prof. Mulligan mentioned debt briefly, but never balance sheets. Isn’t that the central story of the crisis? Again if it’s based on incentives, then why hasn’t anybody in the world, beyond the hyper fiscal stimulators in China, managed to recover more quickly.

There’s no denying that incentives matter with employment, but the argument here is that incentives aren’t the main cause of the slow, worldwide recovery.

Dec 6 2012 at 5:59am

‘As to the issue of causing “pain” or not, I think that the main effect of the 99 weeks of unemployment benefits is that people keep trying to rationalize how they will recover their former job, income, and lifestyle, if only they can just wait for the right opportunity. Procrastination and hope are two important forces.’

This will be the major grouping. I am self-employed and for a while my company was touch and go and I probably should have cut bait, but the job market looked bad, and the money was just enough to keep going and keep key employees. I spent almost all my savings. I got the business back to limping along. Now I make much less than a California state employee, but it may work out. One more recession though…God forbid.

Then there will be those who are not really work motivated and who find that they enjoy leisure time more than they thought. I suspect if I did not have a family, I could fall into this category easily. Nowadays, you can live pretty well without much money, depending on your hobbies. If you let yourself go, then you could find yourself very hard to employ indeed even if you really could do the job. Professors can take sabbaticals to Italy. Average joes who take a sabbatical to the World of Warcraft don’t come back with the same cachet.

Also, regarding taxes above 100%, there was a graph for Pennsylvania that said if you make below 29,000 it only pays to take a higher paying job, net benefits, if its a $70,000 job. Amazing.

Greg G
Dec 6 2012 at 8:37am


While your response to Jan D on the voluntary nature of unemployment is technically true in some sense, I think it confuses more issues than it clears up. Certainly no one is unemployed in a hunter gatherer or subsistence farming economy. Certainly most unemployed people today could find work if they were willing to move to California and help pick the strawberry crop. Even in the absence of any government safety net there are many costs that would prevent most people from doing so. And even if they did, emphasizing the resulting drop in the unemployment rate might not be the best way to think about the arrangement.

I often wonder how many of the economists who blame the length of the recession on stubborn, greedy or lazy workers have ever run a business. Don’t get me wrong. I know such workers exist. They always have, in good times and bad. When I was hiring entry level sales clerks I thought that most of the people I interviewed were in need of some kind of reality check. Most of them I wouldn’t have let near my customers if they had been willing to work for free. We always found good people but it took a little effort.

A very large part of the stickiness of wages is due to the fact that employers understand psychology a whole lot better than economists. Most employers are very reluctant to lower the wage scale because they correctly understand that would cause serious morale problems that could easily come back to bite them. Most employers do not want to hire overqualified employees because they correctly anticipate those employees will be the first to leave and the cost of training all employees is high.

During most of the recessions that we have lived through inflation has had the effect of lowering real wages without lowering nominal wages. That is less of a factor in this recession. If all employers imposed a 10% wage reduction on everyone there would be a revolution. If there was 10% inflation no one would be mad at their employer.

As an employer, if I have excess capacity, there is no price low enough for me to want to hire more workers. The Nineteenth Century saw some nasty and prolonged depressions without any government social safety net to impede recovery.

When we are near full employment (whatever that is) workers declining to work increase unemployment in a straightforward way. When there are a lot of excess workers compared to jobs available, if a slacker refuses a job someone else usually fills it.

Patrick R. Sullivan
Dec 6 2012 at 11:23am

Obviously the ‘safety net’ has an effect–if you believe that incentives matter, i.e. you’re an economist. It isn’t a question of ‘we’ making ‘them’ suffer. Suffering is the natural condition of mankind (ever hear the term ‘vale of tears’?).

The question (I borrow from Deirdre) is, ‘How big?’

Which Mulligan seems to be engaged in a good faith attempt to quantify.

Russ Roberts
Dec 6 2012 at 11:45am

Greg G,

You wrote:

I often wonder how many of the economists who blame the length of the recession on stubborn, greedy or lazy workers have ever run a business.

Replace “blame” with “explain.”

Replace “the opportunities available to.”

Artful Monk
Dec 6 2012 at 12:46pm

I am highly skeptical of the claims and model of Mulligan. He indicated that benefits amounted to 80% of wages. Looking at his model, at its peak in 2009 the avg unemployment benefit was $16K/yr (Fig1.1). So I am assuming that he is talking about someone who makes 20K/yr ($16K/.8). If we were talking about wages being the median of $48K or let’s say a high earner who makes $75K, then that rate drops to 33% & 21%, respectively. In general, I believe that unemployment cash benefits represent 20% of income. So for >50%, unemployment benefits have little impact on their employment because the incentive is for them to work.

Mulligan uses the avg weekly benefits of $290 to get to his adjusted avg monthly benefits of $1169. However, if we replace the avg benefits with the median monthly benefits of $692 (from “AvgMargWorker” tab), the benefits decrease to $12K annually & the benefits ratio declines to 60%. However, Mulligan is talking about low income earners (the bottom 20% making $20K of less) & they get benefits on a sliding scale. So assuming they get 20% (sanity check of median benefits of 692*12/45K), the bottom 20% of wage earners would get ~$333 in monthly benefits (.2*20K/12). When using that calculation the benefits ratios decreases to 45%. Also note that even his model shows that the $20 “bonus” payment & payroll tax exemption disappear after 2009.

Other perceived padding in his model reflects him including COBRA insurance and Medicaid. I don’t see why someone would pay for both. Even with a govt. subsidy, you still pay b/n 400-600/month for COBRA & I would expect people at higher ends of the distribution to partake in the program not low income (same goes for mortgage modifications). Also he factors Medicaid costs and applies them to the unemployed population. Granted he discounts a little but I would expect a majority of the costs to be related to long term end of life care. Once adjust for those, you get back to the historical norms of 30% benefits ratio & it is difficult to see how that changes the employment incentive from historical norms.

Here is the link to his model

Dec 6 2012 at 3:33pm

Jan D:

Are you saying that if stuff gets more expensive because of a VAT increase I’m going to want to work less? And how does stuff being more expensive make me want/need it less?

Certainly we can agree that for many wives working for in home consumption, which is untaxed, verses working in the taxed economy depends somewhat on how high taxes are.

To others commenters, people often take big pay cuts to reenter the taxed work force and often delay taking a low paying job in hopes of getting a better offer and the more money from unemployment insurance may encourage/allow them to hold out longer for hoping for a better offer. The rate of unemployment depends on the length of time people are out of work.

Now on the other side some benefits, though costly to the government, like medicaid are not as highly valued by the recipient. Casey did not mention that but he should adjust for that.

Dec 6 2012 at 5:21pm

BLS numbers cited today on NPR’s Planet Money blog seem to contradict Mr. Mulligan’s assertion (below) that job losses in the construction and manufacturing industries during the recession didn’t contribute significantly to the overall decrease in labor utilization. Am I missing something here or is Mr. Mulligan too quick to dismiss factors that diminish his conclusions?

Kudos to Russ for doing a fine job (as usual) at playing the devil’s advocate but I was disappointed that he didn’t press harder on this point.

Guest: I’m not sure I agree with your description. Measuring in percentage terms, the percentage shrinkage of the construction industry was huge. The percentage shrinkage in a lot of manufacturing industries was huge. But the fraction of unemployed people who came from those industries is much smaller. Because those industries were not that large to begin with. Russ: Correct.

Michael Byrnes
Dec 6 2012 at 7:39pm

I think another issue that needs to be considered is that not all incentives are financial ones.

In today’s world, we do not have a lot of secrets from our future employers. For example, I had to pass a drug test and undergo a background check when I was hired at my current job, and even had to provide old W-2s and tax returns to the background check company because they had touble verifying some of my old employment info.

This has consequences for the unemployed. A lot of people argue that someone on unemployment should be ready to take a huge paycut, and should be ready to take a job, any job, rather than collecting unemployment. But taking the wrong sort of job can have a negative signalling value. Taking a job at a huge paycut can have long lasting repercussions, because your future employers will, in all likelihood, have access to this information and they will factor your current salary into their offers and even into whether you are qualified for the position.

Of course, a long gap between jobs can also have similar negative consequences. So it becomes a tradeoff between taking a job that may look bad on your record versus not taking one which also looks bad.

I think Mulligan is correct that unemployment compensation and other safety net programs create incentives not to work and thus lead to higher unemployment than would be the case otherwise. But I think also one needs to consider the non-financial incentives at work.

Bradley Calder
Dec 7 2012 at 10:38am

If Mulligan’s estimates that the cumulative effect of the benefits from being unemployed exceed the benefits of working for several million unskilled people are accurate, then any argument that these particular unskilled people would choose to work implies that these people are stupid/uninformed of the existence of the benefits/prefer work to leisure.

I would appreciate some evidence that either supports the view that they are uninformed, or that Mulligans estimates are inaccurate. Any other discussion on this topic is irrelevant.

Jim Bryan
Dec 8 2012 at 8:53am

Interesting podcast, I agree with the essence of the argument. Most people agree there must be a safety net, but it doesn’t need to be gold plated. I have a small business and work about 60-70 hours a week, I resent that SNAP can be used for fast food. If I have to buy your food, eat rice! We have the richest poor in the world!

I have a customer and his son’s friends have been enjoying the above average benefits. They did a lot of snow machining last winter and fishing this summer. They weighed a working wage (minus gas, time, clothing) versus unemployment, SNAP, free medical. They figured it would be fun to take some extended time off.

I don’t see either party taking the country’s finances seriously. If you haven’t read it, it might be a good time to read Dying of Money: Jens O. Parsson.

[url removed–copyrighted material–Econlib Ed.]

Dec 8 2012 at 1:04pm

@William’s Planet Money link:

It’s almost as if the economy lost 4 million construction & manufacturing jobs, and 4 million jobs in sectors of the economy providing services to those individuals.

I guess all those 50-something factory workers need to stop being such lazy losers and get trained for the booming nursing sector. By the time they have the necessary diplomas, they’ll be ready to be patients in a nursing home.

Dec 9 2012 at 6:39pm

Casey Mulligan in the interview said, “…when you are talking about helping people who are poor or unemployed or both, there is a tradeoff that economists have always recognized, politicians have recognized it even–that you can help people a lot; that will make for some bad incentives, some inefficiencies, but you are helping people. Or you can hold back the help, and then people will suffer, but you’ll have more efficiency and more incentives. It’s a tradeoff.”

This statement caught my attention when he said it. It seemed plausible so I tried to let it slide. But I couldn’t let it go. It kept nagging at me. And today I realized my unease is due to the fact that his statement is false.

Mr. Mulligan was saying that economists and politicians agree that “you can help people a lot” through government redistribution schemes and assistance programs. Several facts have to be ignored in order for that sentence to seem true.

1) The loss is ignored. Only one side of the equation is accounted for—the benefit side. But the benefit did not materialize from thin air. It was taken forcefully from someone by the government. When the other person’s loss is included in the equation, then there is no gain from the transaction. Some people take issue with this mathematical fact and say there really is a net gain to society because the gain feels larger than the pain thanks to an invisible multiplier called the law of diminishing marginal returns. But that law is descriptive, not quantitative. The values it describes are not quantifiable and are therefore not useful outside of an individual considering his options. Thus, any handicapping of one person’s values over another’s is not the law of diminishing marginal returns but is instead a contemptuous and deliberate subordination of ALL values held by the injured to ANY values held by the beneficiary. It is a calculated valuation of the victim’s losses as zero. It is the arbitrary and thoughtless dismissal of one citizen’s property right for the material gain of some other citizen.

2) The movement of Time is ignored. Granted, for the single moment in time a person receives unearned treasure they are, by definition, better off. Their standard of living is higher than the moment just prior to the gift. But the momentary gift is an anomaly—by definition. It confers no guarantee of improved standard of living going forward. The moment passes. The gift is quickly spent and time washes away the benefit, rendering it meaningless in the long run.

3) The benefits to the government are ignored. The benefits to the government are costs to the other parties in the transaction. Thus the gift received is smaller than the treasure taken because the government’s cut is removed prior to the completion of the transaction. So not only are the parties to the exchange not “helped” in aggregate, but they are worse off because more is lost than gained.

4) Justice is ignored. Adam Smith defined Justice as the negative precepts that prevent individuals from harming strangers. He said Justice was a minimum standard required for society to exist. Taking treasure involuntarily from a stranger absolutely harms that stranger. That momentary violation of Justice creates a precedent for similar acts of injustice in the future. The incentive structure for all three parties going forward changes in the following ways: The government has an incentive to continue to violate Justice through compulsive exchanges so that it continues to receive a percentage of the spoils. The beneficiary has an incentive to continue to violate Justice in order to maintain the standard of living that he could not otherwise sustain without considerable effort. The robbed individual has an incentive at first to protest; then to passively resist; then to actively avoid further injury through hiding, suberterfuge, and misdirection; and ultimately to either leave or retaliate in kind once he realizes that he is the perpetual victim of ongoing Injustice and, as such, is no longer a member of the society. He is outside its protection and he is its slave.

Mr. Mulligan’s statement must ignore a great deal in order to seem true. Once the total picture is revealed, however, the statement is false. Knowing this, he may instead have said:

“…when you are talking about helping people who are poor or unemployed or both, there is a tradeoff that some economists have always recognized, but politicians have not—that you cannot help people by stealing someone’s treasure and giving it to another unearned; that will make for incentives so bad, inefficiencies so egregious, that society is made WORSE OFF immediately; made CONFLICTED in the short term; made INEFFICIENT in the medium term; and utterly DEBILITATED in the long term. It’s a tradeoff: Justice and Prosperity for Injustice and Misery. They move together. You cannot have one without the other.”

Dec 9 2012 at 8:44pm

Do you know what research from the fishing industry he is talking about at around 33:35?

Dec 10 2012 at 12:54pm

While I don’t dispute the general principle that the level and duration of unemployment benefits are providing incentives to the unemployed not to get a job, I think that Mulligan’s case is unsubstantiated (Note that I haven’t really read the book, I am simply referring to his general argument). He can’t explain the magnitude and the duration of the current crisis with labour regulations. It is so for two main reasons, both touched in one way or another by other commenters or in the podcast.

First, these incentives are there regardless of the economy’s position in the business cycle. That is, we should expect higher levels of unemployment also when the economy is in a boom (relative to previous booms). Labour market regulations have an effect on the level of frictional unemployment – it is econ101. Since the duration of the benefits in USA was increased after the beginning of the crisis, this comparison can’t be made right now.

Second, international comparison is a must and before it is done it’s all jumping to conclusions. The IMF have a panel database for labour market regulations, however it does not include the period of the crisis and any possible changes. It can be found here, and the graph on page 22 of the companion paper shows no increase in the unemployment benefits (shown as gross replacement rate) level in Western Europe (also for North America), and the general increase in this period comes from Europe and Central Asia mostly (starting at 1990, which is most probably the result of these being former socialist countries in which no such benefits existed). The Center for Economic Studies has data on the duration of unemployment benefits in Europe (see graph for 2011). Approximately half the countries have benefits for an year only and the EU average is a percentage point higher than that of USA (Belguim is the outlier in being both with unlimited period and with lower than average duration for the EU). Of course, there are other factors, such as the level of the benefits, conditions to qualify for them, etc. – some of them can be found here but until all the pieces of the puzzle are there, I think Mulligan’s argument is weak.

Dec 10 2012 at 1:10pm

Just wanted to add that there is a recent paper (2012) with more up-to-date data for Europe which includes the first years of the crisis (until 2010).

C S Ewerse
Dec 11 2012 at 5:09pm

I very disappointment with this podcasts because of the lack of an international perspective. I live in Germany where they have a very generous welfare system (public insurance based) compared to my home country UK which is nearer to the Anglo-saxon model i.e. lower benefits to provide incentives. Germany provides up to 70% of an average of your last 3 years earnings for one year and thereafter still provides a lot higher benefits than the UK or the USA.

In Germany, unemployment has been traditional lower than the UK and in the recent recession they entered the recession with an unemployment rate of approx. 8% and left with 8% compared to 5% for the UK in and 8% out. This includes Germany dealing with incorporating the old GDR (Communist East Germany).

Russ Robert at the end claimed that he would expect that other countries would have increased their benefits in a recession like the US which shows how clearly he does not understand the European welfare model. The European has constant model of open ended situational benefits – there are no real time limits. Therefore, there is no need to have special benefit increased in a recession. There was a famous quote by a German Finance Minister (on the right of the political debate here) criticising the UK for “crass Keynesian” of extra stimulus during the recession.

This is because in Germany they see their generous unemployment insurance system as good automatic stabiliser in recessions. So overall the debate was very US centric to the point that I had hard time taking it serious.

Before moving to Germany, I was more in the Anglo-saxon camp on benefits they have to be low to provide an incentive to work (otherwise it creates unemployment) – what I have seen in Germany is that a public insurance based model can also work by ensuring that people are not afraid of taking jobs nor starting a business because there is a safety net that works for the low and high earner all with low unemployment rates.

C S Ewerse
Dec 12 2012 at 3:22am

Just to add to the simplistic line that incentives matters. Yes they do but how incentives are interpreted and acted upon is culturally bound, among other context/constraints. If that was not the case one would not see different cultural behaviour in communities and countries facing the same incentives. A great example of that is in Germany people are reluctant to jay walk even late at night with no cars, famously so. Go to the UK and you will see people jay-walking even with Children. The incentives are the same but German culture has developed the belief that one it is a bad example to Children and two they are more willing to following rules (because they believe this can and does bring greater societal good even when the individual is inconvenience).

Whilst Economics is a greater subject, it will not move forward unless it realises as science it only tells one part of the human story. This post support my earlier post which I hope gets published.

Dec 12 2012 at 7:57am

It was hard for me to decide whether “disgust” or “repulsion” was the appropriate word to characterize my feelings when I listened to this.
This is in line with the (cozy and tenured) economists who suggested during the great depression that the crisis must be “allowed to wreak its destruction”. Unbelievable that this line of reasoning can be part of 21st century economics

Chris the Keynsian from Austria

Dec 12 2012 at 1:10pm

Having grown up in my father’s factory, which went backrupt because the bank canceled it’s revolving line of credit, which resulting in the customers finding other tool and die factories, being unemployed, and working in collections.

I find the shock by some to be similiar to the shock by Harvard professors that men and women are different. To ignore individuals poor choices or choices to do the opposite of what others would like, does not seem to be best course in trying to make economics a science.

When I got laid off after the dot com bubble, I took several months off to sit by the pool, play video games, jog, etc. in my condo in Florida. It wasn’t until the benefits started running out that I got panicky and moved back to Western New York to find work. I have a degree from Columbia and a masters degree, so it’s not like I’m lazy, but it’s nice to have someone pay you not to work.

When I worked in collections, there were so many people who did not have any savings. These government programs reduce the incentive to save. The idea solution for the economy as a whole is to have people save their own money. When you have to use your own money, you will be more careful and energized to use it wisely. If you take my savings this time to pay someone who didn’t save, I will not save money next time. You can find someone who was stupid enough to save money so you can take from them and give it to me.

Anyway, Paul Krugman wrote about this whole issue in his textbook. You know a lot of Germans thought England was the aggrevise country.

Lastly, the unemployment benefits made the employment figures worse while they were provided; however, it is only half the issue. The second half is where did the money come from. The problem with the recession is the misallocation of resources. Through banking regulation, Basil II & III & increased capital requirement and pressure on the financial institutions, the country is not investing in infrastructure and technology. The government is taking money that was being lent to small business like my father’s factory and providing those same workers with checks every week. The amount of money lent to business and especially small business has decreased while money lent to the government has increased. Are economists really saying that having a factory worker buy a television for doing nothing is a better investment then that worker buying a television for building a casing for an industrial air conditioner?

Jan D
Dec 12 2012 at 4:17pm


So what you’re basically saying is that if one cannot find a job one can always be someone’s unpaid slave. So why have a welfare system, right? Isn’t that what your argument comes down to?

[fixed per commenter–Econlib Ed.]

Dec 13 2012 at 2:06am

Mulligan made some weak arguments here and didn’t support the idea that benefits worsened the recession. In some cases I disagree completely, but even if he is right, I don’t see how the effects he suggests amount to much. The magnitude is overstated.

E.g. the boyfriend girlfriend analogy is wrong, because companies fire employees with cause, resulting in no UI. And, he can’t backup a claim that unemployed workers are avoiding employment, while suggesting they would spend more if they had a lower paying job instead of UI.

Much of the lower class is paycheck to paycheck, and UI, if they receive it, isn’t enough to cover expenses. Plus, you’re required to apply for jobs, and accept offers, as a condition of the UI.

Evidence on underwater homes suggests most stay, against their own interest. This irrational behavior shows us that there aren’t enough people interested in slacking, and they can’t get enough to hold the recovery back materially.

What is in a Name?
Dec 13 2012 at 9:08pm

[Comment removed pending confirmation of email address. Email the to request restoring this comment. A valid email address is required to post comments on EconLog and EconTalk.–Econlib Ed.]

Dec 13 2012 at 11:34pm

@Jan D
You wrote “So what you’re basically saying is that if one cannot find a job one can always be someone’s unpaid slave. So why have a welfare system, right? Isn’t that what your argument comes down to?”

I am not quite sure what you mean in your first sentence. Slavery is when human behavior is performed under threat of violence or actual violence. A job is a voluntary activity. No one would perform a job voluntarily if they received no compensation, although compensation is not always monetary. There are many people at my local hospital who volunteer and are paid nothing. I do not think they consider themselves slaves. My previous comment to you was simply describing how the price of labor affects the allocation of labor. If there are a lot of people seeking jobs but not many people wanting to hire them, it means the people seeking jobs are asking too much for the skills they offer. They must lower their price to get a job or acquire skills that are more valued by the other market participants.

Regarding your second question: Since we already have a welfare system, a more relevant question is why NOT have a government-run-welfare system? My previous post details four reasons why the welfare system is harmful to society. In brief, stealing from one person to make another person wealthier is a violation of Justice. Justice is a set of negative precepts that collectively mean “don’t harm your neighbor.” Stealing from a rich person to make the poor, middle class, and government employees wealthier is a violation of Justice. Every single time we allow or cause harm to our neighbors—rich or not—, society is made poorer overall. And that is the silver lining. In the long run repeated violations of Justice lead to massive poverty, brutality, discrimination, corruption, and war. The most amazing aspect of this predictable tragedy is that the poor people—for whom the breach of Justice was done to help—are hurt more than anyone else.

The Holocaust is the most extreme—but by no means unusual—example of what happens when the concept of Justice is systematically ignored by a government. The Jews were the wealthy minority of their day. The German majority—middle class and poor—and the government they controlled were all too happy to dismiss the property rights—including the right to life—of the Jews for their own enrichment. Did it work for them? Were they made better off as a society? Think WAR followed by horrendous poverty. Their poor were made poorer. Their middle class was made poor. Their previously productive Jewish minority was killed off or fled. Violating Justice made the Germans feel wealthy only for a moment,—like any form of borrowing from the future does —but in the long run led them to ruin. This is what Adam Smith meant when he said “Justice…is the main pillar that upholds the whole edifice. If it is removed, the great, the immense fabric of human society…must in a moment crumble into atoms.” (The Theory of Moral Sentiments II.II.18)

Dec 17 2012 at 11:16pm

I was disappointed Mulligan didn’t talk about how the bias against stale labor might deter workers from staying unemployed. 99 weeks lazing about with 70 percent of your former income may have a disastrous effect on your lifetime earnings that 25 or even 50 weeks does not, as it makes reentering the job market more difficult. So you might find that increasing how long people can spend on unemployment has sharply diminishing returns after a year or so, when you get in stale labor territory. Treating a 2 year gap in the middle of your working life as equivalent to retirement makes Mulligan look insensitive to that. Does he try to quantify this effect in his book?

Comments are closed.


About this week's guest:

About ideas and people mentioned in this podcast:Articles:

Web Pages:

Podcasts and Blogs:



Podcast Episode Highlights
0:33Intro. [Recording date: November 21, 2012.] Russ: Now, there are many different explanations for the recession that began in 2007, and the unusually poor recovery that followed that. Many of these explanations are macroeconomic in nature, as one would expect, meaning they look at how various aggregates such as the money supply or aggregate demand or economy-wide interest rates, or the level of indebtedness: how did those macroeconomic variables affect the overall health of the economy? But your book, which is a very technical work of applied statistics, takes a very different approach. You argue that government policy changed the incentives for people to work and for people to hire workers. It's a much more microeconomic approach about incentives and the impact those incentives had on the outcomes that we then measure in the macro-economy. Lay out the main argument of the book. Guest: The main idea is, I start by looking at changes in public policy that affect regular families. So, I'm not looking at banking policy and things like that. Things that affect regular families: the Unemployment Insurance (UI) program, Medicaid program, food stamp program, other safety net programs that can help people when their incomes fall or they are out of a job. I looked at those; I looked at how they were run before the recession; I looked at how they were run during the recession. And there were changes, significant changes. And then I tried to quantify that. Some of the changes were pretty important. Others were more minor. I tried to quantify those things and come up with kind of a total impact of the new government rules on how much help you could get when your income was low or you were out of a job or both. And then I said: Okay, once I have an understanding of how much help and incentives change, then I ask: When that much help comes along, how much does that affect people's behavior? Economists have studied that second question a long time, and all I do is bring it into the 21st century, and say: we have 21st century rules changes here; what kind of changes in the labor market should we expect? And the answer was: Well, labor markets should be depressed as a result of those rule changes. And I work out how much, and it turns out about half, probably a little more of the depression of the labor market you can link directly to these new rules for helping people when they are unemployed or have low incomes. Russ: Now, it's not unusual that during a recession, labor markets are not very happy; that there's a high unemployment rate, that the employment rate goes down. What's startling I think about this recession is that the employment-population ratio, meaning the number of people working as a proportion of the entire population, plummeted--it was a big recession, it went down a lot--but it has not budged a bit since the recovery officially began in July of 2009. So for three years there has been no recovery in that ratio. That's extremely unusual. If you look at the post-war period, after almost every recession that number bounces back dramatically. Very quickly. In fact, the deeper the recession, the quicker it bounces back. There's a little bit of a hesitation in the early 1960s that is something like we have here. But this is very unusual. So, you are suggesting that that failure of that employment number to bounce back, and possibly you are suggesting that the recession itself, is a result of government policy specifically related to the labor market. So, comment on that. Guest: I might quibble a little about your description. It's true that the employment-population ratio hasn't really bounced back. But absent recession, wherever it came from, there were forces pushing that ratio down anyway. Especially the baby boom coming into retirement age. Russ: Sure. Guest: So, I would say that that ratio has bounced back partly to where we'd expect it to be, but still it's down significantly. Continues to be. And we are surprised that it didn't bounce back more. I think we can agree with that. Russ: Yeah. It's pretty flat. Guest: And thanks to a lot of help that the government has offered. Not just the government. Also in the book I look at some private sector help that's there. Thanks to that help, people don't suffer as much when they are not employed as they would have if they would have experienced this ten years ago or five years ago or fifteen years ago. And when it's less painful to be without a job there will be more people who are without a job.
6:01Russ: So, you have been criticized for explicitly or implicitly saying that unemployment is voluntary. You are suggesting that, well, if there is not a big enough return to work, because if you go back to work you lose a lot of these benefits, so you choose not to. And you are suggesting then that the really painful aspects of this recession, which saw unemployment reach a little over 10% and stay stubbornly high, now just dropping a little bit under 8% in the last numbers, that that's not due to some stickiness in the labor market as the Keynesians would suggest; it's not due to a failure of aggregate demand to be stimulated sufficiently. You are suggesting that's due to the choices of workers, rational choices that people have given what it costs them to go back to work. Is that fair? Guest: Yeah; it's a little oversimplified. But a primary factor is incentives for the group of the population whose incentives have been severely eroded. As a result of what's happening to that group of the population, that leads to some stickiness; and I discuss this in the book. Why should people accept low wage jobs when they were getting a better deal when they were not working? So, in normal situations maybe wages would have fallen; they are not going to fall when there's more generosity on the unemployment side of things. So, I don't disagree with the stickiness. I just don't view it as something that came from heaven. It came from public policy. Same with the aggregate demand. When you have a group of people who are making the best of a bunch of bad choices they are presented with, which is to be on public assistance, they are going to trim their belts. They are going to spend less, maybe get by on one car instead of two. And so if you are a guy who makes cars, even if you have no access to the safety net programs, you are going to find your labor market is not too good either, because your customers have decided to spend less. But why are they spending less? That has to do with the expansion of safety net programs. Russ: So the irony here is that the Keynesian view says that these safety net programs help repair the economy, because it injects spending into the economy; it puts money into the hands of otherwise unemployed people. You are suggesting it's exactly the opposite: that it not only doesn't make things better, it actually makes things worse, if you measure it by the labor market. Guest: Right. It does put money in a group of people's hands; it takes it out of another group of people's hands. And the net reduction in the economy is actually less spending. Because, you know, you have less work going on. So there's less total income to be spent. And so the people who are going to suffer from that, depending on the industry they work in, they are going to see the drop in demand for what they make. And they may not appreciate my story; but they don't understand--they need to appreciate: Why aren't their customers spending? If you drill down to the bottom of that you are going to see that the safety net expansions are a big part of it.
9:17Russ: So, one thing that's great about this book and the work you've done around it is that there are a lot of explanations that have the general flavor of yours--I'll mention a couple, maybe--where someone says: Look, we know why such and such happens; it's because there's this effect, and this effect has the right sign. So, for example, I've talked a lot on the program and had guests who've talked about the uncertainty that was created by the increase in all kinds of activities of the government. And many people have blamed the poor recovery on the uncertainty, the riskiness then of investing, etc. But it's generally very difficult to quantify that. It's hard to show that the riskiness is higher and uncertainty is higher; there's always a lot of uncertainty. So, is it really plausible that it causes this problem? It's a nice story. It was sometimes is called a Just-So Story; it's what sometimes is called an ex-post narrative. One could accuse you of that except that you don't just do that. You go out and try to measure the actual size and magnitudes of these changes to see if it's plausible. Doing that is quite challenging econometrically, and we probably won't get into the details of that. People who are interested can look at the book. But just to give the reader some of the flavor of that, why don't you talk about what some of these changes in policy are in the safety net and some of the magnitudes, so that we can get a feel for whether it really is, just back of the envelope-wise, a reasonable claim. Guest: Okay. Well, the major--one thing that's kind of interesting about what's happened is there's not a single program or single rule you can point to and give a lot of the blame. There were a lot of relatively small changes happening at the same time, all in the same direction. So, if I had to name one, it would still be a minority of what was happening. Russ: Name more than one. Guest: The number one expansion would be the infamous 99 weeks of unemployment insurance. Normally it's 26 weeks, a half a year. And it went up to almost 2 years. And that puts a whole bunch of people in a position to receive benefits. They wouldn't have received them otherwise. If they had been unemployed 27 weeks, under the old rules they would have gotten off of that program. Russ: And before we lived in this highly charged political atmosphere we seem to be living in, there was a consensus among economists that unemployment compensation increased the unemployment rate, and when those 26 weeks ended, you could see that the rate would go down. Is that a correct summary of the literature in this recession? Guest: I think so. The magnitudes were debated there, for sure. But if 5 years ago an economist were to stand up and say, I don't think unemployment insurance has any effect on the unemployment rate, that wouldn't have been a credible position. Russ: And again, I think now a lot of people argue the opposite. Nancy Pelosi being one of them. But actual academic economists as well said: Well, unemployment compensation, that puts money in the hands of the poor; that increases spending; that, through the Keynesian multiplier leads to greater output; that leads to more hiring; and that way the unemployment actually goes down. You actually have people arguing that now. Guest: They said that. They don't have models to that effect. I'm not aware of a Keynesian model that shows that when you put money into the hands of the poor, that expands demand in any nontrivial way. I mean, maybe at a fourth decimal place or something like that. Russ: But they have a higher marginal propensity to consume. Guest: Right, but all the modern Keynesian models also recognize that consumption is not the only activity. There's investment. All the modern Keynesian models recognize you've got to get the money from somebody, and that person is going to spend less. So, the modern models don't have that. Even though economists talk about it; you put them on TV and they talk about it. But it's not part of their research. Really the closest thing that gets to the research is some of this empirical multiplier stuff where they'll look at the relationship between government spending and employment or output. And then--and mostly that's an analysis of wars or peacetime defense buildups. And they'll say when you pay guys to build tanks or you pay guys to march in the jungle, that makes the economy bigger. They're probably right. But then they take this huge leap and say: Well, instead of paying guys to build tanks, we pay them not to build anything; that will also make the economy bigger. And that part--it's not the economists talking when they say that. Russ: But they do say it. Guest: They do say it. Russ: I hear it all the time. And some of them have Nobel Prizes. So, it's a claim they'd like to make. Guest: Yeah. And when I push them on this, they say: Well, everyone else says it. When I say, show me the research, they say: Everyone else says it. Like the Director of the Congressional Budget Office (CBO) told me that. I said: Why are you using the multiplier for unemployment insurance? He said: Well, that's Keynesian economics. That's what he told me. It's not that they did a study of some country that expanded its unemployment insurance and saw what a boom they enjoyed. There's been no such study. Russ: The good news is, I think the CBO has now increased their range for the multiplier to a factor of 6, I think--their bottom end to the top end. It used to be 4-fold. Now I think it's 6-fold. They've used Valerie Ramey's lower range estimates now for their bottom range. Which makes the range even bigger. Interested listeners can listen to that episode we did.
15:23Russ: But to go back to your list. So, you said unemployment compensation, number one. List some of the others. And then I want to raise the obvious question, which is: Unemployment compensation is not that generous. So, if you are suggesting that by extending it from 26 to 99 and then adding in the other things I'd like you to list, that then people just said: I'd like to not work. So, first list some of the other programs that expanded, not just in terms of the number of people--obviously, as you mention many times in the book, we understand that there's an endogeneity, a feedback loop, there that affects total spending. But it's not just that more people are eligible. It's that what they got when they were on there got more generous. So, talk about what other programs are relevant here. Guest: Yeah. I'm not--all these things I mention, I call them 'rule changes.' Russ: Yeah. Guest: Changes in what the statutes say about how people should be treated. Not a different economy being filtered through a set of rules that have been there for a long time. So, one of the rule changes was unemployment insurance going to 99 weeks. Another change, set of changes, was in the food stamp program. There was a new set of rules about who could be on the program, a more expansive set of rules. Now, in most states, they don't check whether you are genuinely poor. You could have lots of money in the bank or you could own a yacht and you can be on this program. The only thing they do is they check your income at the time that you apply, whether it be a quarter or a half a year or something like that. And so if you happen to be out of a job, even if you were doing quite well only 9 months before, you can be on food stamps. They expanded the food stamp benefit, the amount that each family on food stamps gets was expanded twice, once toward the end of 2008 and again 6 months later. Russ: And roughly how big were those increases? Guest: They were, in percentage terms, about 15% or so each, for a combination on top of each other maybe around a 40% increase. Russ: That's a huge increase. Guest: In percentage terms. Now, the dollar amounts in the programs are smaller. The typical food stamp benefit is less than the typical unemployment benefit. And so this would be one of those expansions that's more relevant for people who don't earn that much normally. It would be less relevant for someone who--someone who is earning $75,000 a year wouldn't find the food stamp benefits to be all that significant. But at the low end they are significant. They expanded a lot. They added on--another thing they did was with the unemployment program. Not regarding the number of weeks. But they added on a bonus, really. I think they called it a bonus. The official name is Federal Additional Compensation. They gave people $25 a week. Everyone who was on UI got their usual benefit plus a $25 bonus. Another thing they did, they expanded the number of people who could get UI, especially people who were weakly attached to the labor market. Under the old rules, they'd say: You didn't even work that much before the recession came along; we can't give you benefits. They changed that. As long as you had even a little bit of work history, they'll let you on the program. The federal government is involved with this; lenders in general are involved with this: They gave breaks to people whose homes are under water. Gave them mortgage breaks. But only if their incomes were low. So, that's another kind of set of help that you get. If you spent part of the year or even all the year unemployed, you'd get some help on your mortgage, but if you continued to work like you were before the recession, forget about getting any of that help. Another major one is they paid for the health insurance, 65% of the health insurance, for people who were laid off from their jobs. So, that's a big type of relief. It never gets cited when they talk about the average unemployment benefit. But you could have 65% of your health insurance paid for. You'd have to pay the other 35%. And 2-2.5 billion people took advantage of that expansion. Russ: Just to make it clear: You are somewhat, or maybe totally, agnostic about whether this was a good idea or bad idea, these changes. That's not the point of your book. Your point is to say that when we assess the effectiveness of the labor market as we come out of this recession and try to think about what we need to do to deal with it, you might want to take these things into account. Guest: Yeah. I think Milton Friedman called it 'positive' analysis. The book is about what happened in the labor market and why. Let's figure that out first. And I'll contribute what I can. You never solve every puzzle fully, but I'll contribute what I can in terms of the causal factors. And after that you can ask: How would it be different? In this case, when you are talking about helping people who are poor or unemployed or both, there is a tradeoff that economists have always recognized, politicians have recognized it even--that you can help people a lot; that will make for some bad incentives, some inefficiencies, but you are helping people. Or you can hold back the help, and then people will suffer, but you'll have more efficiency and more incentives. It's a tradeoff. You've got to pick one or the other, or degrees or balance the two. And from that point of view it's hard for me to sit here and say: Well, okay, we rebalance in the favor of more help and less efficiency, that was wrong. Well, no; maybe the balance before wasn't proper. Or, maybe there were good reasons to rebalance even if it were proper before. So, the main theme in the book is we rebalanced in the direction of less efficiency and more help, and not judging it. Now, with that said--and that is the main message of the book--but in going through and studying the help, I do find some instances where the balancing was done very poorly. Instances where people would actually have to pay to go to work. In tax rate terms, their tax rate was above 100%. I mean, that's a poor--economists all recognize that to force somebody to pay to go to work rather than earn from working is no way to manage that balance. It's terrible for incentives. It gives very little help given the amount of incentives it takes away. It's demoralizing: people find themselves in a trap, the more they work the less they have. It's a terrible way to do it. I uncovered plenty of instances of that in my research. But still, that's for a later book, at a later time, to say: Okay, how could we have done the help better? The first approximation: We helped more, and maybe it was worth the inefficiencies we got. Russ: So, you lay out in the book in much more detail, obviously, than we have time for here, the changes in the rules that you just summarized. A couple you did not mention, I just want to mention. I'm not sure this is in the book. We had an episode with David Autor, who points out that Disability rolls have expanded dramatically in recent years. And that's not, again, necessarily because a lot more people are disabled in the way we usually interpret the word, but the legal system has made it easier to establish disability. Again, that might be a good thing, might be a bad thing. But certainly the rules of that game have changed. And for me--I don't like the word 'efficiency' in economics; I think it's a bad term that is misunderstood and sometimes misused. The real issue for me is: I'm not so worried about if it's inefficient. I'm worried about what are the policy implications of the current level of unemployment? What should we do about it? We have people saying that because unemployment is still stubbornly high, we need to spend a trillion dollars more in public spending to somehow stimulate aggregate demand. So, I'm more worried about the implications for--I think we should draw the right lessons, and that's where I think your book is provocative and interesting.
24:04Russ: Let me get to something I raised at the end of the last question which was: Are these magnitudes plausible given how relatively ungenerous unemployment programs are? So, it's gotten more generous. Okay. That's made it easier for people to stay unemployed. Are you suggesting that--you claim that those magnitudes are big enough to explain a big chunk of the changes in the amount of work people are doing. Make the case for that. Guest: Okay. But I think we've got to break it down a little bit. Russ: Sure. Take your time. Guest: First of all, let's look before the recession. Can we describe the programs as ungenerous? I'm not sure I'd jump to that conclusion. It's not a crucial part of my book, but I don't want to adopt an incorrect description even if it's not relevant. Typically before the recession, unless you are well into the upper half of earnings, when you lost a job you got half of your earnings replaced. So if you used to earn $600 a week, you'd get an unemployment check for $300 a week. And I guess you are referring to kind of that $300 dollar number--it seems $300 isn't very big. Well, if you earn $600--I earn more than $600 myself--but if you earn $600, then $300 is not all that trivial. Number one. Russ: But it's only half. Guest: Okay, but let's get to number two: There's all kinds of taxes you don't have to pay when you are unemployed. Russ: True. Guest: Payroll taxes--forget about it. You don't pay it when you are unemployed. A big chunk of income taxes you are not going to pay when you are unemployed. So when you put all of that together, without even getting to other help you might get from food stamps or Medicaid, put it all together, before the recession about 70%, maybe a little more, of your earnings would be replaced. Not half. And that's without getting into, like I said, other types of programs. And when you start with 70% as your baseline--so you are going to get 70% on the old rule, and you are going to put a bunch of new rules in there--it pushes the 70% up to 85 or 90%. I don't think we can call that trivial any more. Russ: I agree with that. Guest: When people plan to retire, which by all accounts is a voluntary transaction, they embrace and accept that their income may go down by 20%. But it's okay. They are not working. They planned for that. The most studious, forward-looking planners say: I don't need the kind of income that I had when I was working, when I am not working. A 10 or 20% pay cut is what I need. And with these expanded programs we had lots of people, millions of them, who got less than 20% pay cuts from being laid off. Russ: Meaning they got 80% or more of what they'd earned before. Guest: Right. And there were people--I point out in the book and I'm doing further research on this--there were people who actually had more to spend after they were laid off. That was not a typical situation, but still it numbers in the millions--a million or two. So, that's the first point. The second point is, even if it were true that these were small amounts, there are always people on the margin of working or not. We know that's a fact because before the recession there were like 140 million people working, and about 140 million adults now are not working. So there were a lot of people working before and a lot of people who were working and going in between. There were people on that margin. And people who are on that margin and you change what you pay them on one side of the fence, they are going to fall off the fence. So what happened is benefits got more generous; you had 130 million people stay on the working side of the fence; and depending on how you treat the discouraged-worker side of the issue you had on the order of 10 million people fall onto the not-working side of the fence. So, it was a small minority of the population that fell onto the not-working part of the fence, on that point of view. You don't need a huge change in benefits to get 10% of the population or 8% of the population to change their behavior. Russ: But a counterargument to that would be: They didn't have a choice. It's true that it was more attractive now to be unemployed than employed, but even if it weren't true, the jobs just weren't out there. What's your response to that? Guest: First of all, I think during the recession--they call the recession 2008 and 2009. There were 240 million people who started new jobs during that period. So, there were lots of people starting new jobs. Now, it would have been better if there were 280 million. But I don't like this description of it that says there were no new jobs to be had. Tell that to the 240 million people who did start new jobs during that period. Russ: Some of those are more than once, presumably. Guest: Some are more than once. So there are guys who, during the period when there were supposed to be no jobs, found a job twice. So, the 'no jobs' is a terrible description of what happened. Russ: Let's just stop and talk about that for a minute. Because I think that's something that's often forgotten. And I assume you are drawing that from the Job Openings and Labor Turnover Survey (JOLTS) data. Guest: Right. Russ: So JOLTS is a Bureau of Labor Statistics (BLS) survey that's fairly new. It goes back before the recession. What it used to show--when times are good--there are millions of jobs being destroyed, even in low-unemployment times. But there are just more than that number being created, and the total employment number is increasing. What a recession is, is there are still millions of jobs being created; there's millions of jobs being destroyed; but during a recession, it's usually the case that the number being destroyed outweighs the number being created. But there's still millions being created. That's your point, right? Guest: There were millions being created; that's right. Russ: It's just surprising to most people. Guest: There are also millions normally being destroyed. And there was an increase above that normal. But again, unemployment benefits have a lot to do with that. There's somehow a treatment that says: Well, I'll consider the idea that people don't go back to work because of these benefits. But they never also consider that maybe the efforts that employers and employees have to hang onto those [?] Russ: Say that again? Guest: I lost you there, Casey. Say that again? Guest: There are workers and employers and employees all the time considering should they continue this operation. We know that because as you said, before the recession there were lots of jobs that dissolved. That were destroyed. And those considerations are always happening out there in the economy. And one of the things they consider is: Well, how painful is it going to be to put that guy out on the street? And the more painful it is, the more they do to try to take steps to avoid the layoff. Maybe taking a wage cut, maybe working harder, whatever it may be. Cutting the price that the product sold for. They are looking at the painfulness of unemployment, and by making unemployment less painful, the government gave employers and employees less reason to retain the jobs that we already have. So, I don't think it's an accident that we had a spike in layoff in 2008. In fact, I think it would have been foolish for those people to try to retain those jobs and somehow the unemployment benefits and other programs got less generous. Then you're going to put a guy out on the street when there's less help. The layoffs I think made a lot of sense from individual employer and employee points of view. It doesn't mean it made it a happy time. You don't like to be in a spot where you even have to consider trying to resuscitate your job or your company. But when you are in that position, people are not thinking about the costs and the benefits. Russ: So, you are suggesting that employers are compassionate, not just profit seeking? That they are going to look at what alternatives there are when they let workers go? Guest: The market relationship forces them to be compassionate. Take a simple example. We've got a factory and people are buying less. They think about: Maybe we should charge less for this product so we can keep the product moving. But we are going to have to have a wage cut in order to do that. And employees in the place, ask them: You want to go for that? You want to go for a wage cut so we can keep the factory going? And if they can already do that well when not working, they are going to say no, I don't want you to cut my wage. Russ: This gets back to stickiness being an endogenous response to the policy changes, not some inherent given from heaven, given the reality of economic life. It's a very good point. It's a deep point. Guest: It's not something I've invented. Again, in this book I draw on prior research that economists have recognized this effect before, they've studied it. I show some dramatic examples, like the fishing industry, where the layoffs are very much a function of what's going on with unemployment insurance. And the whole economy, too. They've studied it; they know there's that effect. Of course, they debate the magnitude of it, but the direction of the effect is clear. Help people more when they are not working, and you are going to have more layoffs. And slower returns to work.
34:05Russ: Now what about--the BLS also collects data on quitting. And during the recession there was a huge drop in the quit rate. The data only goes back to 2000, that I could find in the BLS, so in the 2001 Recession there was a drop in the quit rate; people were less likely to quit their jobs. But there was a big or much bigger decrease, as you'd expect, in this recession. Is that inconsistent with your claim? People are trying to keep their jobs. They are not so eager to quit. Guest: Well, no; in fact it's part of the evidence in favor of my claim, really. First way of thinking about quits--I think of it as a marriage type of situation, or boyfriend and girlfriend. When they split up, sometimes it's not all that interesting to ask: Did the boy break up with the girl or did the girl break up with the boy? 'Quit' is somehow implying an answer to that, somehow that the employee broke up with the employer. And a layoff is the opposite. And I'm not sure there's a lot of deep meaning in that. However, there is public policy meaning in it. If the employee broke up with the employer, then he can forget a lot of benefits. The unemployment insurance program will not help him if he says he's the one who initiated the breakup. On the other hand, if the employer steps forward and says, I initiated this breakup, that turns on a whole spigot of benefits for people. So it's no surprise to me that when we increase the spigot of benefits, we open it wider, that more of these breakups were classified as layoffs--the employer initiating the breakup--than being classified as quits. Russ: Now, how does productivity play into your story? Because you have a very interesting finding about utilization of other inputs during this recession, which I think is very provocative. Explain what's going on with capital and productivity over the course of this recession and whether it's similar or different from those in the past. Guest: There were a number of recessions historically, not all of them--the Great Depression would be one and there were a number of others--where productivity, output per worker, fell. That is, as the labor force was shrinking, the workforce was shrinking due to layoffs, for example, output, what was produced, was shrinking even more. And so as a ratio, output per worker fell. That's happened a lot of times. I can't say it's happened in every prior recession; that wouldn't be correct. But it's happened in a lot of recessions. It happened big time in the Great Depression. That didn't happen in this recession. Output fell somewhat, but it fell less than employment did. Output per worker went up during this recession and as of the beginning of 2012, output per worker was still higher than it was before the recession, even above kind of a prior trend for output per worker. Russ: What do you think is going on there? Guest: And that fits with the idea that employers perceive employees to be more expensive. When they perceive employees to be more expensive, they are going to try to produce in ways that don't need employees. Using more capital, using it better, using it more wisely, using the management more intensively. Whatever they can do to get by with less workers but still satisfy the customer. So, in one of the early chapters of the book I document these productivity changes and explain how on the employer's side of things, they are acting like employees are more expensive. And then the rest of the book is: Well, why are they more expensive? And redistribution is one answer. Redistribution increases the cost of employees in a couple of ways. Some of the redistribution falls on employers, as with the Obamacare 2012 levy per employee. Some of the redistribution goes directly to employees, but as I mentioned earlier, that makes them less willing to take wage cuts when the time comes that wage cuts are needed. So, employees are really more expensive for a variety of reasons. And the result is less labor and more productivity.
38:58Russ: Now, I don't remember in the book if you drilled down to this level of microeconomics, but one of the challenges is trying to measure these impacts given the diversity of the labor force, of the sectors of the economy. So let's turn to two issues that I think are relevant for that. One is that you would think that these effects would matter very dramatically depending on whether you are married or single, and also on whether you have high income or low income. If you look at unemployment rates by education, there are much higher unemployment rates for low-skilled workers--that's before the recession. They seem to increase more dramatically, but it's not obvious that that's--it's hard to say. But I think your claim would be that the recovery would be seen more in the higher skilled jobs than in the lower skilled jobs. That is, the bounce-back would be bigger for higher educated, higher skilled workers than for lower educated, lower skilled workers. And I'm not sure that's there in the data. Have you looked at that? Am I representing it correctly? Guest: I like how you described the theory, that these program expansions, the safety net, does not impact everybody uniformly. In particular, I mentioned the $25/week bonus. If you are low skill, $25 a week is worth paying attention to. If you a pretty high skilled person who earns $2000 a week, what's the big deal about $25 a week? So there's that effect. Oh--a lot of benefits are kind of proportional. Unemployment insurance is proportional over a range. A lot of these debt issues, that higher income people have more debt measured in dollars and therefore more to be forgiven-- Russ: That's true. Guest: So some of these things are proportion. Russ: I'm thinking of food stamps, though. Food stamps wouldn't be true. Guest: Right, food stamps wouldn't be proportional. So there's a mixture of it. Chapter 6 in the book starts to look at that. It's mostly a macro book so I don't get into a lot of cross-section comparisons, but Chapter 6 is about that. And then I had an expansion on Chapter 6 in a paper that's coming out on tax policy and the economy in a couple of months. And there I broke down the population into 10 groups based on 5 skill groups, from low to high skill, and marital status. And I do kind of the same steps in the book: first I ask what happened to their incentives due to these rules? And sure enough, as you said, low skill people for the most part had bigger changes in their incentives. Although not totally, because there is a tendency for very low skill people to be able to be on the programs even when they are working. So it doesn't change their incentive to work. But for the most part, low skill people have bigger changes in incentives. And then unmarried people have significantly bigger changes in incentives. Because married people don't have access to a lot of these programs, because the spouse is probably on their own putting the family out of poverty moneywise. Russ: Right. Guest: And then the last chart in that paper compares changes in labor market outcomes for the entire 10 groups and changes in their incentives for the 10 groups. And they line up very well. Unmarried people--the unmarried part is dealt with in Chapter 6 quite a bit--had employment per capita or hours worked per capita fall a lot more, about twice as much, than for married people. And I don't interpret that as difficulty in finding a job. I just don't. I don't think that employers look at the wedding finger and say: You don't have a ring, I'm not going to hire you. If they did it would be a good business to rent people rings to take on job interviews. But I think the real story is that unmarried people are getting more help from the government than married people are, and married people are in a little bigger of a hurry to get back to work. Russ: The other micro and disaggregation issue I would think would be sectoral differences. So, if you look at this recession it's particularly hammered manufacturing and construction, which has led to some geographic differences that are very dramatic. The recession was much harder in so-called 'sand states,' where construction business had been part of the boom. So we're talking about places like Nevada--let me correct that: Nev-[Midwestern-flat-a sound]-da, for my Nevadan friends. Arizona, Florida, Michigan--a non-sand state. So, unemployment was very concentrated in those geographical regions, in those states. And if you look at the changes in employment in manufacturing and construction, it's a huge portion of the job losses from the recession. How does your story fit in with that, if at all? Is that relevant? A lot of people would say that's the source of the problem; we just need to get those sectors back and everything will be fine. Guest: I'm not sure I agree with your description. Measuring in percentage terms, the percentage shrinkage of the construction industry was huge. The percentage shrinkage in a lot of manufacturing industries was huge. But the fraction of unemployed people who came from those industries is much smaller. Because those industries were not that large to begin with. Russ: Correct. Guest: So I think, obviously unemployment rates would be higher for construction workers, but I think, especially the low recovery, you can't link it back to those industries so easily. Same thing on the regional side. There are differences by regions, for sure; but there's an awful lot of unemployment that's not in Nevada. Moreover, the differences by region--I think Chapter 6 shows this as well--there are some differences between the sand states and the non-sand states in the labor market experiences. But it's nothing like the difference between married and unmarried people, or nothing like the difference between elderly people who didn't get much help from these program expansions and people who aren't elderly. So, I think really the geographical differences fit my story pretty well in terms of how small they are compared to the safety net differences that you expect and you find. And then the third part of it would be that the safety net changes weren't the same everywhere. Russ: That's true. Guest: The debt overhang is a much bigger issue in a place like Nevada. Nevada is full of people who, whether they realize it or not--I expect they do realize it--if they work, they are working for their banker. Their family is not going to get any of the proceeds of that. It's just going to go toward digging their loan a little further out from the bottom of the ocean. So incentives, I'm afraid, are pretty bad in Nevada, California, Arizona. Hopefully getting better, but these incentive changes weren't uniform geographically. Russ: And of course there are people who went and worked for their banker because they think that's the honorable thing to do. They borrowed the money; even though they are under water they want to pay it back; they don't want to have that hit to their credit rating. But again, you are not claiming, and I wouldn't claim that that's everyone. It's just a question as you say at the margin, meaning people who are sort of close to indifferent might now be pushed one direction or the other. Guest: Right.
47:04Russ: One criticism your work has received is: what about overseas? What about other countries that have had this horrible recession, that had slow recoveries? Did they experience the same changes in their safety net? The critics who point that out say there's something else is going on, you are missing the boat. How do you respond to that? Guest: Actually, I'm working on an international study. I'm only one guy; I can only work on one country at a time. Russ: You're so lazy, Casey. You're a slacker. Guest: But it's definitely something that needs to be researched. I want to understand all the countries. And the first thing I would do, looking at any other country would be what I did looking at America: what's changed with incentives? I don't believe the claim that incentives were constant in other countries. Russ: I don't either. Guest: The sales tax went up in the United Kingdom and Ireland. And sales tax is another thing that reduces your [?] to work. The reason you work is so you can buy stuff. But if when you go to the store, the government is going to take it, what's the point of working? That's got to be factored in. And somebody needs to do it. If somebody else doesn't do it, I'll get to it. I'd want to look at that. The other thing I have looked at already, I looked at early on in the recession, was: Are the countries the same in terms of what happened to their labor market? And they are not. The U.S. labor market contracted a lot more. A lot of these commenters point to U.S. GDP and compare it to European GDP--they are cherry picking now. My theory is a theory about the labor market. My model is about what happened to employment. And then they don't want to look at the employment data. The employment situation in the United States was a lot different; you might say a lot worse. But employment went down a lot more in the United States in 2008 and 2009 than it did--I was not sure I was able to find any country where it went down--any major country--where it went down that much. I never said that the entire situation that we have right now is due to the safety net. I said it's most of it, it's a big part of it. So what I'd expect to see in the country data would be the U.S. situation to be different to the degree that its safety net situation was different. Russ: Just to make a comment about this kind of criticism: I think Fannie Mae and other housing policies had something to do with the housing boom and bust. Maybe a lot to do with it. Which in turn might have had something, maybe a lot, to do with the recession. It's a complicated question. You have monetary policy, you have all kinds of other things going on. You have the financial sector having trouble because of all of this. But one of the--when you push the argument that it's caused by Fannie Mae, other people go: Well, what about the rest of the world; they don't have Fannie Mae. Well, that's true; they don't have something called Fannie Mae. But I am sure that a lot of other countries have increased their role in encouraging home ownership. So it's an empirical question. Just to say, the other countries don't have Fannie Mae, therefore Fannie Mae has nothing to do with it, is not a good argument. And similarly, people who reject your argument by saying: Well, other countries don't have a Supplemental Nutrition Assistance Program (SNAP) program, or 99 weeks. Well, no; they had a very generous safety net before; I suspect in the face of a crisis they probably had the same political forces that caused the changes here, had similar changes there. And as you say, then it's an empirical question, whether the magnitudes are plausible or not. But the idea that your idea is clearly wrong because other countries have problems is kind of a bad argument.
51:03Russ: What about past recessions? I know you don't have--this book, by the way, has a huge amount of empirical work, some of which involves a lot of aggregation. That's not my favorite kind of empirical work, but a lot of just very careful measuring of what these changes were, which is my kind of empirical work. And I salute you for it. Since you don't have time to look at every country, you probably don't have time to look at every recession. But I would think one of the best tests of your claims would be to look at past recessions and see how changes in the generosity of safety net or unemployment benefits were changed during those recessions. Have you looked at that? Have other people looked at it? Guest: Yes and no. It's something I've thought about a lot and something I've wanted to do. There's a big area in macro that before this research area came along came out and said marginal tax rates do not change over the business cycle. And they kind of guided their research based on that claim whether it was right or wrong. It's just wrong. We know that--well we didn't go to 99 weeks in other recessions. We went to 52 weeks, 73 weeks. So incentives did change over the business cycle, and you want to quantify that. That's something that needs to be done. I think it would be great. I form a marginal tax rate series monthly for the years 2007-2011. I think it would be great to take that back 10, 20, 30 more years. Russ: Well, there is a puzzle in macroeconomics which is that these last few recessions, 1991, 2001, and this one would seem to be different. Are they different because the safety net got more generous? Or are they different because policy was different, other kinds of policy, monetary policy? Are they different because of structural changes underlying the economy? That remains to be seen. And we may never settle it. But it would be very useful to get some more evidence on it. Guest: Yeah, I think it would be useful. I do have some hints, some guesses I am going to pursue on that line. The changes in the safety net are relevant, and we talked about those in the latest recession. But it also matters, the backdrop onto which those changes occur. If you go from a 70% marginal tax rate to 90%, I think that's a lot bigger deal than going from 10 to 30%. And I think the safety net has been getting bigger over the last 10, 15 years. The food stamp program would be an example. It's been growing over about a 10-year time frame. The disability program that you mentioned has been expanding over maybe a 20 year time frame. So I think our business cycle has been increasingly superimposed on a larger welfare state. And I think that may lead to different business cycle dynamics. That's kind of an educated guess right now, but that's something that should be looked at. Russ: Well, one of the things I particularly like in your book--you do some sensitivity analysis, which I think is always a good idea. You look at possible ranges of labor market response to these changes in the rules. So, you look at more elastic, less elastic--meaning more responsive, less responsive--and you have a nice chart where you show, could debate whether it's a reasonable range, but you take what seems to be a reasonable range taken from the literature on responsiveness and that range, bottom to top, is still far away from what actually happened. Meaning: you could attribute a lot of the changes to these rule changes. I didn't say that very well, but I hope listeners know what I meant. I mention the sensitivity because we are taping this in November of 2012; the so-called 'fiscal cliff' is looming over us, and there are potentially fairly large marginal tax rate increases coming. Either for everybody, if they can't resolve it, if Congress and the President can't resolve it, or at least for the wealthy Americans if that's the compromise. And a lot of people argue that labor supply is not very elastic, it's not very responsive for high income people; nobody's going to stop working because their marginal tax rates are going to go up; in the 1950s we had marginal tax rates of 70% and 90% and people still worked hard and created stuff and the economy was pretty good; so, all these issues about responsiveness that are really what your book is about, those are overblown; they are not important. What's your reaction to that? Guest: Well, there's a few pieces to it I think. One piece, in my book and a lot in general I've been looking at employment statistics. And employment statistics are, by construction, democratic. Everybody counts once. We don't count one person three times and other people half a time. Russ: One worker, one vote. Guest: So, if you want to understand unemployment statistics you kind of have to understand what's happening to kind of everybody. Large groups of people. That's why I looked at a lot of these programs. And, something that's happening just to rich people--well, rich people are a small group. They are the 1% or 2% or a half percent or whatever they are. Their direct effect on the employment statistics can't be that big. Now maybe there will be indirect impacts you want to get to. But it can't be that big. Now, the impacts on GDP of rich people could be very large. And the impacts on spending could be very large. Because spending and GDP are not democratic measures. People contribute to those in different proportions. So, I find that discussion kind of a little bit of switching gears. Just because you have a policy that doesn't affect the employment rate very much because it hammers rich people doesn't mean it's not going to affect GDP and overall efficiency. Different measure of damage. Russ: Fair enough. But let's talk about this elasticity issue. Do you believe that if marginal tax rates went from, say, 35 to 40%, which is roughly the range that's being talked about now, which is a 15% increase in marginal tax rates. It's a 5 percentage point increase, but it's roughly a 15% increase in tax rates for the last dollar earned, for high income people. Would you expect a significant labor supply response from that group? Especially given this claim that when it was 71% or 90% people still worked hard; they worked full time. Guest: I'd expect a significant response in the group of people impacted by such rate changes. I'm not sure we're going to see a 5 point change across the board. It'll be in specific groups. I'd expect to see a significant response. Now my book, first of all--I don't take my personal beliefs on this responsiveness. I rely on the literature. I've done prior work and I think labor is pretty responsive, whatever. But I don't bring that to bear here. I draw on the literature and I include in my range some, I think, ridiculously small--ridiculous is too extreme--pretty darn small responsiveness to incentives. The main driver in my book and the reason why I get such big labor market changes is because the incentives change so much. Not that the behavioral change per unit incentive change, per marginal tax rate point, is all that big. So, my opinion really doesn't come into the book that way. And the other thing I do in the book--I'm focused on the aggregates. You mentioned you don't like aggregation. But we're all interested in these aggregate numbers, so we need to have aggregate incentives. And I cannot make a calculation that only applies to a small group. That's going to be at best a piece of the total. And so I've bent over backwards to try to tell you how much marginal tax rates change on average. Including everyone: including the people who had no change or people whose marginal tax rates went down. And I find that on average marginal tax rates went up--or at the median would be more accurate. At the median, I find, marginal tax rates went up by 8 points, from 40 to 48. That's a lot. Russ: It is. Guest: For an average change. Russ: But as you point out, it's zero for some people and it's even larger than that for a lot of others. My skepticism isn't over the underlying mechanism here. It's that trying to measure that across the whole economy without disaggregating seems to be a little bit of a heroic leap of faith.
1:00:06Russ: But I salute you for it. Particularly because of the cultural aspects of your work. We're out of time; I just want to close on this. What you are doing here is very ambitious. It might be true. Evidence is always uncertain, and truth is elusive in economics. But it's a provocative idea that's very different from the standard views that you hear coming out of the macroeconomics profession. And most labor economists I think are ashamed to make these kinds of arguments. They weren't ashamed when unemployment was 5%. They were happy to make them. But when it became 10%, everybody became a Keynesian. All of a sudden people were very eager to get behind increases in government spending--almost everyone. And they were equally eager to throw away all the incentive-based standard microeconomic, labor supply, labor economics literature that you are relying. And you are not ashamed; and I think what you've done here is pretty courageous. And so I salute you for that. Even though I'm not sure it's true, it's certainly provocative. And why don't you close by talking about some of the reactions you've gotten in the profession. Maybe I'm exaggerating, but I think this is a hard argument to put forward now. Even though I think 10 years ago it would have been relatively easy. What are your thoughts on that? Guest: Well, I should say: The book is about the economy, not about economists. But your last question is asking me about economists. And so I'll answer it. I'm not expert on what they feel and everything. But I do sense what you said, that it doesn't feel good to analyze labor supply and think about incentives for the unemployed, think about incentives for laying people off. It doesn't feel go, so maybe that's why they don't do it. I don't know. One reaction I've gotten from economists is, when I show them this stuff, they say: Boy, I didn't realize the sum total of all this. I had heard about this program or that program. But now that you tell me, it makes sense. But I didn't realize that there were so many expansions happening that the sum total could be so big. I'm sure they may dispute details of my numbers. Maybe I got it too small. But they say, yeah, somebody needs to do this. Somebody needs to think of the sum total of the government policy changes and how they affect people. Russ: You going to keep doing it? Guest: Yeah, I think so. I mention that I think we want to look at prior recessions from this perspective; we want to look at other countries. The thing I've been working on when you call is the Affordable Care Act. That has another flavor, another image that you get looking at that is a lot of apparently little changes that add up to big changes in incentives. I don't have any numbers to give you now, but there are big changes in incentives. On the order of the changes that we saw, that you see in my book, in 2008, 2009. And so, if I'm right about that and I'm right that behavior depends on these incentives, we may see the labor market contracting significantly in the next, say, 4 year time frame. Russ: Well, it's always good to quote F. A. Hayek: The curious task of economics is to demonstrate to men how little they know about what they imagine they can design. And when you talk about all the different small programs adding up and apparently unintended consequences that you've uncovered, I think that's a very relevant thing to think about. Guest: Well, thanks, Russ. One thing I forgot to mention--I have the web page for the book and one thing that's available on that page is spread sheets with all my numbers. So, if you think I made a wrong judgment here or there, in doing the numbers, people can go in and tweak it. And come up with their own answer. I'm very much in favor of somebody doing that. I want to make it easy for the next guy to do it better. If I made a mistake, do it better. But I want to subsidize people considering incentives.

More EconTalk Episodes