Are the Long-Term Unemployed on the Margins of the Labor Market?

There is new Brookings research by Alan B. Krueger, Judd Cramer, and David Cho:

The short-term unemployment rate is a much stronger predictor of inflation and real wage growth than the overall unemployment rate in the U.S. Even in good times, the long-term unemployed are on the margins of the labor market, with diminished job prospects and high labor force withdrawal rates, and as a result they exert little pressure on wage growth or inflation.

Consistent with my earlier views, this work is suggesting that many of the long-term unemployed are/have become an economically segmented group.  This is noteworthy too, as it implies the problem is not merely initial discrimination:

…even after finding another job, reemployment does not fully reset the clock for the long-term unemployed, who are frequently jobless again soon after they gain reemployment: only 11 percent of those who were long-term unemployed in a given month returned to steady, full-time employment a year later.

I would consider that evidence for a notion of zero marginal product workers.  Furthermore, in my view (I am not speaking for the authors here), right now further inflation is as likely to harm as to help these individuals.  To ask whether the Fed “should give up” on the long-term unemployed is a biased framing which is more likely to mislead us than anything else.

There is a good piece up at 538:

Krueger and his coauthors, Princeton economists Judd Cramer and David Cho, find evidence that the long-term unemployed aren’t getting jobs even in parts of the country where the job market is comparatively healthy, suggesting that a stronger economic rebound won’t be enough to put them back to work.


Shall we make no effort at all to help? What is the worst that could happen, the interest rate stays lower for longer than it would otherwise? That would only be a good thing, a la soltas overshoot.

So, to make up for workers not getting paid enough to consume all of GDP, we make interest rates so low banks must loan to the unemployed and working poor so they will buy all of GDP and stop more job losses?

Exactly right.

Last time I checked, those ZMP workers sure were able to build astounding numbers of houses in the deserts between LA and Albuquerque. But I guess that counts for nothing?

Tyler is usually fair-minded and interesting, but he swings and misses huge with his offensive tabbing of millions of people as ZMP workers. It's possible he's right. But as someone who boosted Scott Sumner, and knows his Milton Friedman, he should take pause and intellectual caution from the fact that he and his fellow ZMP/inflationistas/pain caucus confreres may collectively be negative multi-trillion dollar marginal product workers in present value terms.

I'll take ZMP over that any day. Those were some pretty nice houses.

Any research on the psychological make up of the long term unemployed? Is this a persistant problem that is exacerbated by the severity of recent unemployment levels? Causes that come to mind may be unique personality profiles or limited skill sets. Does this play out along various job categories, or is it more pronounced in certain industries?

I am not an economist or a student. Just find this reading much more informative than mainstream media outlets.

Why not ask about the emotional state of employers?

Maybe the attitude of employers is "I want all the money for myself and don't want to pay workers - other employers and government should pay my consumers to buy more from me, but I won't pay workers to buy from other employers because the money is mine, mine, the rings are mine mine...."

Surprisingly enough, employers need employees in order to make money.

Labor intensity is a thing, not constant, and often an independent variable.

So what if it is?

It's really the ZMP, zero marginal product, that matters. The workers just end up being the ones producing it, similar to last hired first fired.

And yet, these long-term unemployed are eating, being housed and otherwise supported. I would be interested in knowing how they accomplish this: are they living off savings, being supported by a spouse or other family member, getting government benefits, working off the books, etc.? I do not know how this would be determined other than through interviewing a representative sample.
The long-term unemployed are not working because they do not have to.

And how many of the long term unemployed became homeless or turned to crime?

Do we count our homeless, and know how the populations are changing? Or is that something America really doesn't want to know?

@Rich Berger,
And they did have to before?
Yes, we have longer unemployment benefits. But other than that (and the new blips we might see post ACA), none of your above have changed much. The gal with a working husband was just as likely to have one before this last depression -- more likely, right? Off the books existed before. Moving back in with the parents was an option before if it's one now. If you can live off welfare now, you could have lived off it then, if you can get disability with X condition now, you could have taken it then. As a population, if people choose to retire earlier now, they could have chosen to require earlier then.

There is a point where people who wish to work and are willing to work, even though they don't "have to", work. And there's a point where people who wish to work and are willing to work, and don't "have to", stop working. The question is what has made so many people move from one point to the other.

Maybe they did not realize they could live without a job until they lost theirs

I would suspect that was a big part of it.

Here's something relevant that is interesting -

No one will ever hire me again -- I didn't know there were any generations after Y. If we're up to Z already, do we have to start again with AA or is it straight to the apocalypse?

Some interesting stuff there, particularly her links.

The long term unemployed are the price of high profits.

In an efficient economy, profits will be zero.

Speaking as an economist, of course. Profits are not ROIC, but a sign of restraint of trade. Like monopoly, for example.

In a free market, profit would result in doing whatever it takes to increase production. Lowering prices and profits to gain market share, or entering the market at a lower price and profit to take market from the monopolist.

To increase production in response to your high profit, or your competitive target, you must hire workers, either for your production, or to buy capital that must require labor to produce.

If no labor is involved, then machines can manufacture machines for zero and the logical result in all production costs zero. Robots don't get paid wages. If a small number of robot owners are getting 100% of revenue as profit because zero labor is required, then individuals will be required to establish a parallel economy that is based on labor.

Who is a capitalist going to sell to if he keeps 100% of revenue as profit. Only the other robot owners? He won't sell to any individuals who are by definition unemployed.

Profits since Obama has taken office have been increasing to record levels, but the profit can't be spent because the people who get that profit have too much to put the profit back into the economy buying GDP because they have too much consumption and too much capital assets for the existing GDP.

But without the profits buying GDP, GDP must shrink relative to growth, and that means factors of production must be idled. And every factor of production ends up ultimately being labor.

High profits accumulating cash means decreased employment.

If decreased employment does not reduce profits, then decreased employment leads to more decreased employment.

The US has two forces driving growth against the high profits killing growth:
Government borrowing money to buy GDP
Increased population which gets to borrow thanks to government to buy GDP

Of course, we are in an era of free lunch pillage and plunder capital destroying economics.

So called capitalists find no profit in buying capital assets. In fact, they find buying capital too distasteful because they must pay labor, and they see labor costs as a total loss. Never mind that they want consumers to buy what they produce, but they want high profits and someone else needs to pay their consumers to buy their production.

Thus in free lunch economics, the government pays the consumers, the capitalists pay as little as possible to labor, ideally getting government to pay for labor to have housing, clean cloths, be fed, and healthy, and provide subsidized transportation, so the capitalist can make high profits from their labor.

What free lunch economist want to erase is economics as zero sum.

The money to labor plus the money to capital MUST equal labor's buying of GDP plus capital's buying of GDP.

Labor can give money to capital to buy GDP, or capital can loan money to labor to buy GDP, but that is not sustainable.

Buying shares of Apple and Google is not buying GDP. Since 1980, the solution to getting all production bought is lending to labor, directly or by way of government debt so labor can buy more than they earn as share of GDP.

And the Tea Party has forced pillage and plunder which kills jobs because they want to fix things so there will be more jobs. Burn natural capital to kill jobs building energy harvesters. Pound roads and bridges to destruction to kill jobs. Let water pipes decay and leak to kill jobs. The Tea Party is very opposed to paying for labor out of their income, with many on Social Security getting it from government.

+ 1; well stated

We're becoming a drop-out nation; people dropping out the the workforce, dropping out of college saddled in debt they have no hope of paying off. Nowadays, economic value is created through passive consumption such as engaging in social media and streaming with netflix. Other are becoming wealthy through real estate, buying stocks, speculation, and web 2.0. If capitalism is defined as an economic system where those who possess capital have precedent of those those who don't then without equivocation capitalism is thriving in America, and capital has triumphed over labor. The chasm between the haves and the have-nots will keep widening, but this doesn't pose a threat to the economy, contrary to popular belief.

This reads like a press release from the Occupy movement. All the prescriptions I read from them to "fix" the labor/capital dichotomy would increase labor's disadvantage in the long run

This is one of the best things I've ever read.

'I would consider that evidence for a notion of zero marginal product workers.'

And that sentence is evidence of why for some, eugenics seems so attractive as a perspective, especially if 'ultimately the Nazi connection will be seen as a bump in the road.' ( )

Especially when one considers who likely belongs to that bloodlessly designated group of 'zero marginal product workers.'

From zero to Godwin's Law in 70 minutes. Most impressive.

I really would like to understand better the long term unemployment not being reset even after a spell of employment.

For example, I read somewhere that very qualified people are at a more than normal risk of LT unemployment. But it seems hard to believe they,re ZMP.

BTW, the link in the above post in my name is wrong. Posted from a new machine, with no auto-fill...

Anyhow, I had commented on ZMP before and I wanted to share those thoughts again:

i.e. to which degree the fact that output is 'demand-constrained' makes some workers irrelevant/useless to the employers and to which degree the discrimination against LT unemployed is pure discrimination?

I really would like to see more on this: "even after finding another job, reemployment does not fully reset the clock for the long-term unemployed, who are frequently jobless again soon after they gain reemployment: only 11 percent of those who were long-term unemployed in a given month returned to steady, full-time employment a year later".

The 11% isn't logically connected to the sentence before despite the ":" . The question or claim wasn't about whether LT unemployed can find full-time employment a year later. We'd be interested in the subset that DID find a job and whether they manage to keep it. I also would like to link this to 'underemployment' i.e. the reality that, these days, many jobs and thus a fair amount of people are just done on a short term/contract basis with high rotation among the contractor being implicit. And this affects not just barely qualified people. In my sector, finance, a lot of back-office or middle office jobs are done like that. By contractors, who rotate a lot and for whom stints of unemployment are unavoidable. More so in bad times, less so in good times but there you are. These days, some of these guys might definitely be LT unemployed and it'd be hard to argue they're ZMP/useless.


I think you've hit the nail on the head--some long-term unemployed aren't ZMP all the time, but just some of the time when demand is constrained. When demand is greater, less people become ZMP. In other words, the ZMP status is elastic and temporary, although Tyler seems to treat it as a permanent thing.

I wonder why...perhaps because it feeds into the libertarian ideology of the leeching masses and the brave John Galt. Perhaps because it makes office workers and bureaucrats surfing on feel better about themselves (nothing provides greater value than a senior quality assurance analyst, right?). Maybe it's just more ideological fodder to consolidate power in the hands of capital and against labor--aren't Cowen's and Tabarrok's salaries paid by the Koch brothers anyhow?


I don't think Pr. Cowen or Tabarrok are really beholden to the Koch brothers but you might be on to something when you mention libertarianism. If we just accept that they are genuine-if-practical libertarians then they have a inbuilt bias into finding narratives that "blame the victim".

And, btw, LTU easily threatens well/over-qualified people. People likely to read Marginal Revolution...

So, basically, I wouldn't be so negative on their motivations. That's why I like this blog. I don't always agree, far from it but both Pr. Cowen and Tabarrok seem like genuine and nice people.

Pffff, that was emotional!

We need a Godwin's law for political/economics forums for the Koch brothers. A good first step to not be taken seriously. I also think you may have an overly simplistic view of libertarians if you think they take pride in "blaming the victim", though I will give you that they may disagree on your definition of victim.

At the risk of sounding callous, we need to look at the phenomenon of NMP workers (you figure out the acronym). ZMP workers may be useless but they don't put you out of business by destroying machinery, offending customers, and ruining your reputation.

I would rather the Fed obsess about robust economic growth...inflation and employment less so...

Here's another opinion (from Fed Chair Yellen) on research like this that suggest the long-term unemployed are on the margins of the labor market and thus have little effect on wages or prices:

“I think it would be tremendously premature to adopt any notion that says that that is an accurate read on either how inflation is determined or what constitutes slack in the labor market,” she responded. “I wouldn’t endorse — I certainly don’t think our committee would endorse the judgment of the research that you cited.”

"tremendously premature" .... There are studies and analyses that contradict or at least point to other reasons why inflation and wages did not fall more in the recession. And the massive influx to LTU might suggest that comparisons to other recent cycles is appropriate. This is a good paper but it is not definitive. I liked Appelbaum's NYT write up of the paper (which included the above press conference quote) better than 538's.

One more comment ... and this one is purely my opinions in reaction to this part of the post:

"I would consider that evidence for a notion of zero marginal product workers. Furthermore, in my view (I am not speaking for the authors here), right now further inflation is as likely to harm as to help these individuals. To ask whether the Fed “should give up” on the long-term unemployed is a biased framing which is more likely to mislead us than anything else."

The "notion of zero marginal product workers" still sounds a lot like an individual type to me. It seems outrageous that right before the recession we had an usual build up of workers with nothing to contribute to economic activity. I will grant you for health or personal reasons there are always some workers who struggle with or have little interest in working continuously ... but this was a massive influx to long term unemployment. And a ZMP worker pins a label to the worker ... why not a worker in a ZMP economy? We 'made' a lot of people ZMP by allowing massive, highly persistent shocks to hit the economy with little buffer to the workers swept off their jobs as a result. I do think long-term unemployment is very harmful to workers: it erodes their skills and heightens (statistical) discrimination against them. But it is not a type as your label suggest. If the Great Recession had not happened there would be many fewer long-term unemployed and given their even larger share now, why wouldn't they be pulled into a stronger economy?

As for your "inflation harms" more than helps the long-term unemployed. Don't you think some of them might be carrying variable-interest debt to smooth through the lost of income? Don't you think they have more time to search and adjust their consumption bundle and buffer a rise in prices? You still might be right but a) we remain at low levels of inflation and b) you are stretching beyond any evidence to say that.

"To ask whether the Fed “should give up” on the long-term unemployed is a biased framing which is more likely to mislead us than anything else." I dunno maybe they are just having a conversation on what "maximum employment" in the Fed's dual mandate means. I could re-write the conversation in terms of slack, markups, Okun's Law errors aka less emotional words (to non-economists) but it does not change the fact that this is an important issue. And we are talking about people, recall.


Especially "a ZMP worker pins a label to the worker … why not a worker in a ZMP economy". To still be looking for supply side issues in the USA post-2007/8 is... a bit weird, no matter how much I like Pr. Cowen's blog (see my take on "The Great Stagnation" here: )

IMHO, if one really must look for supply side issues, the EU is a far more fertile ground than the USA...

Yes, what was there about this last depression that made these workers suddenly obsolete? Seems to me that if the problem is people without the ability to produce in a new economy, a huge jump in their numbers should be accompanied by some huge jump in the "progress" of this new economy. Sure, there will be fits and starts, but this was quite a fit and start.

Seems to me much more likely that the idea of the ZMP American is just another version of the old story -- we have a problem, and we need to tag it on to some "other". It's not that our economy is dysfunctional and more and more distant from anything free market and that's a bad thing. No, we're fine. In fact, we're just so darned fine that we're leaving those poor schmucks behind. Goofy slobs don't even know how to write code for a trendy app, produce a surprising sociological study on how children might trend to being younger than adults, or formulate a feasibility study. What good are they?

I look forward to the day the ZMP crowd decides to give up on the Fed.

From the November 2007 peak to the January 2010 trough, the private sector lost 8.69 million jobs.Since then, the private sector has added 8.62 million jobs. A long painful climb just to get back to where we were more than 6 years ago.

Government jobs (mostly state and local) rose by 146,000 between November 2007 and January 2010 (stimulus), and has since fallen by 630,000 jobs.

Government is broke, and broken.

As I keep saying, the resilience of the US economy is something I quite marvel about. Yeah, yeah, yeah, return to the mean. But, frankly, returning to mean instead of collapsing into a vicious circle seems pretty amazing to me.

That said, having the same amount of private sector jobs isn't good enough when pop. has been growing...

The population is growing and not all jobs are equal. Most of the jobs we are creating are low level McJobs. The ratio of unemployed to job openings is well over three (we get hundreds of applicants per opening) so it seems unfair to call people ZMPs when there are 10mil or more people chasing fewer than 4mil jobs.

If I recall correctly, Tyler and a number of followers reacted with horror to the idea that Zero Marginal Product jobs might be a problem. (It is a marketeer's just-so story. Every job is perfect, and every non-job is perfect as well.)

This is what I was going to say. How strong can the evidence here be? Don't you need an explanation for why there was a sudden increase of such workers? Even the participation rate is at its lowest point in decades. What changed in the population of workers so suddenly?

The highest executives in corporations don' hire and fire people, human resource employees do. These people follow guidelines that they've learned in the human resource departments of business schools. Rather than sort potential employees as individuals, they use arbitrary factoids like length of time unemployed to separate applicants. Employees are hired on the basis of how well they fit bureaucratic profiles rather than ability, which is more difficult to determine than the date a person was made redundant. This is the same sort of arbitrary thinking used by the state. A person is incapable of handling alcohol until the day of their twenty-first birthday, they're not qualified for some jobs unless they've had 10 hours of training, they can't see well enough to drive unless they can identify certain letters on a chart. The human resource departments of companies are the spawn of academia, who have, in large part, determined that the long-term unemployed aren't qualified to be hired. Academia is to be blamed for this issue.

you really tend to see this imbalance in technical fields, where the HR staff literally does not know how to handle the resumes they receive. Can a B.A. Humanities major really reliably judge who is better for technical engineering jobs? So they fall back on easier-to-them rubrics like "when was the last time you were working?"

I think you guys vastly oversimplify the hiring process. The HR department is just the first screening, they don't make the hire/fire decisions. They decide who gets an interview but beyond that it is a technical manager who decides. Its absurd to think a BA in Humanities is picking software engineers at Google. When you receive hundreds of resumes and can only give a dozen interviews, you need certain markers to sort on and UE length is one of them. As Tyler's statistics show that increased UE length is an indicator for how short a person lasts at a new job, I'd say it isn't such a bad screening metric.

Correlation vs. causation.

And we still don't have much details on this 'LTU do not last long in a new job'... There are plenty of explanations possibles besides 'he/she just sucks'.

Do they discount part of it being they don't last long in the first job after the period of unemployment, because once HR no longer looks upon them as lepers they move on to a better job?

Obviously, then, what the long-term unemployed need is to raise the minimum wage by about 30 percent.

I would consider that evidence for a notion of zero marginal product workers.

Where I live there is the "narrow pyramid" effect ie. an experienced (ie. over age 45) but not-so-senior engineer can still find work (even if it's with a consulting company).

But someone who is a domain expert or has lots of experience managing complex projects can be out of the job market for months or years until there is some opening that is sufficiently senior and where the manager does not feel threatened,

While this seems like a very real concern to me, I can't imagine it's common enough to move the needle on unemployment. Something like 40 percent of the unemployed are long-term unemployed. There aren't that many senior and highly specialized people.

As some people upthread mentioned, it seems worthwhile to better understand the long-term unemployed. I know a couple of ZMP types who were basically "found out" in the recession. But I think there are other types also, like the ones you mention. So I don't think Tyler's crazy, but neither do I think he's capturing the whole picture.

I think we can go to the well on this. It is automation and outsourcing, again. Companies built up large staffs, with seniority (even when not highly specialized), to realize that with automation and outsourcing they could do with much less. In this story, the recessions provide "socially acceptable periods" to re-balance, away from high priced senior US labor.

If Tyler's cruelty is just at LegalZoom makes a good lawyer into a ZMP one ... I think that might be a result of his discomfort about markets and their solutions.

This is not hard. You get the behavior you subsidize, every time. When we have made it risky to hire permanent employees, and possible to live on unemployment for extended periods, why in the world are we surprised that there are long-term unemployed?

Indeed, in a world where anyone unemployed for more than 60 days is sumarrily shot and left to rot in the street there would be no long term unemployed.

Don't be obtuse. I'm not saying long term unemployed don't want to work. I'm saying we have incentivized them not to, by subsidizing unemployment. Do you disagree?

I'm consistently surprised by economists' refusal to see populations as individual rational actors, and not well-behaved data points in a model.

If that was the case why the divergent behavior for short term versus long term unemployed? In fact, the opposite would be the case since it's easier for someone who was just laid off to opt to stay on unemployment than it is for a long term person who has already exhausted or will soon exhaust his unemployment. Likewise it doesn't make sense that long term unemployed people who do become fully employed should have a harder time remaining so than short term unemployed.

long-term unemployed here. Huge MR fan, thought I'd write in.
Our gracious host points out that LTU, even when they get jobs, don't tend to keep them a year later. This is evidence that LTU's are ZMP workers. And indeed, many are.
However, from the front-line trenches of LTU I can tell you it's not that easy. What you tend to see are very distinct HR preferences away from hiring the LTU, so any place that will hire the LTU is..."odd". In the last 3 years, a solid half of which I've been unemployed, I have been hired at
-a company that went bankrupt 3 months later (X2)
-a company that was acquired by another and then "rightsized" all the former employees

In my experience, the calibre of company that hires the LTU is very low, and they are hiring you as a desperate, last-ditch hail mary. Which, of course, makes your resume even more toxic.

+1. And not the only post mentioning HR criteria as an issue. It certainly jigs with my own experiences in Europe...

I really really would like Pr. Cowen to reply to this kind of stuff. Is it just anecdotes? Or is there a way to investigate this at a more data-driven level so that we can put this ZMP stuff to the test?

Nick Corcodilos wrote a good piece a while ago on how online job boards and linkedin are actually making it harder for employers to find suitable employees (and vice versa), and how this is good for the job boards and linkedin but not for anyone else

1. I'm coming around to Tyler's ZMP Worker framework, however this requires us to find whatever the full employment point is in this new economy. Hence stimulus should be pushed to the point that we actually see signs of inflation.

2. Not considered here is those not in the labor force who do not fall into the 'long term unemployed' bucket. Recent college grads chilling as slackers in mom & dad's basement, hangers on, non-working spouses. It's quite possible if employment picks up the labor force will see an influx of people but rather than coming from the long term unemployed from the previous crises they will come from those who otherwise aren't participating fully in the labor force.

3. "Furthermore, in my view (I am not speaking for the authors here), right now further inflation is as likely to harm as to help these individuals"

These individuals are probably being supported by working individuals (examples, working spouses, children supporting parents etc.) or gov't programs indexed to inflation (social security for those who opt for earl(ier) retirement). Pushing to the point that inflation begins will boost income for those employed which will probably do more to indirectly help these people than keeping the labor market soft by depressing the economy.

"...suggesting that a stronger economic rebound won’t be enough to put them back to work." I don't think we'll ever know enough to get past "suggesting", because we're saying that there are people whose personal particular situation lessens their chances of being employed. The problem is that the list of Specifics of "Personal Particular Situations" could be very large, and not subject to targeting a national program to some of these Specifics, let alone most of them. In that case, it's very hard to figure out what to do besides just targeting higher growth which we tend to want to do in any case. But should we then give up?

I think we should give up trying to solve such a complex problem with a Targeted Plan, but not give up on the people in these situations by casting their fates to the wind ( Vince Guaraldi Reference ). A Guaranteed Income of some sort would alleviate the need for us to solve this problem specifically, while ensuring that people caught up in these diverse situations still have enough to get by until something comes up.

The labor force participation rate in modern times has never been 100%. It has never gotten greater than the high sixties or so from a low of the high fifties, both the high and low periods being mostly accounted for by women once being mainly homemakers to now mostly part of the laborforce.

It's not a given, though, that it's optimal for society to be all in the labor force. In the 50's society thought it was best if half of the adult population (namely women) were not in the labor force. It may well be that we are simply entering an age where the number of people who are optimally in the labor force is starting to decline.

If that's the case then it still makes sense to maximize employment, but that takes on a different meaning. It would mean to find the point of greatest possible employment without touching off serious inflation. Greatest possible employment, though, may not mean everyone has a job but may mean more like those who are 'job inclined' all have jobs and their income is as high as it possibly can be.

The question can then be addressed as what should be done about those who aren't optimally in the labor force? In the 50's the solution was strick social codes that required those in the labor force to provide for those who weren't. The modern day version of this may look radically different (i.e. house husbands rather than wives, 20 something 'kids' in extended live at home periods, housholds headed by grandparents who are still working strong) but may end up relying on the same principle. Perhaps guaranteed incomes will make a policy comback as a solution. To me that seems a better solution than the alternative, a Cato Cailen economy dominated by a few super successful where everyone else plays the role of 'live in bum' in their mansions where they are kept around perhaps because they are entertaining or to avoid boredom and loniliness.

The economy is no longer capable of providing a job to everyone who wants one. This is obvious every time you read how many thousands of people applied for a couple of dozen jobs at a factory or big box store. Sometimes the economists are incredibly dense. National basic income and communal living to stretch the NBI will be the answer. Giving people more money if they don't procreate will also happen.

Fixing the economy not being an option?

Comments for this post are closed