ZMP workers and morale externalities

by on April 18, 2013 at 9:52 am in Economics | Permalink

On Twitter, Bryan Caplan asks me to clarify why zero marginal product workers do not clash with the notion of comparative advantage.  The point is simple: some workers destroy a lot of morale in the workplace and so the employer doesn’t want them around at any price.

Most of us buy into “morale costs” as a key reason behind sticky nominal wages.  If your wage is too low, your morale falls, you produce less and so the wage cut isn’t worth it.  Well, what else besides low wages makes people unhappy in their workplace?  Very often the quality of co-workers is a major source of unhappiness; just listen to people complain about their jobs and write down how many times they are mentioning co-workers and bosses.  (I do not exempt academics here.)  A “rotten apple” can make many people less productive, and you can think of that as a simple extension of sticky nominal wage theory, namely that installing or tolerating a “pain in the ass” is another way of cutting wages for the good workers, they don’t like it, it lowers their productivity, and thus it is not worth tolerating the rotten apple if said apple can be identified and dismissed.

There is no particular reason to think that ZMP workers are especially stupid or in some way “disabled.”  If anything it may require some special “skills” to get under people’s skins so much.  (Of course there are some individuals who, say for health reasons, cannot produce anything at all but they are not usually in or “near” the active labor force.)  To draw a simple analogy, the lowest-publishing members of academic departments are rarely those who make the most trouble.

To the extent production becomes more complex and more profitable, ZMP workers are more of a problem because there is more value they can destroy.  The relevance of these morale costs also varies cyclically, in standard fashion.  A company is more likely to tolerate a “pain” in boom times when the labor itself has a higher return.

Note also the “expected ZMP worker.”  Let’s say that some ZMPers destroy a lot of value (that makes them NMPers).  You pay 40k a year and you end up with a worker who destroys 80k a year, so the firm is out 120k net.  Bosses really want to avoid these employees.  Furthermore let’s say that a plague of these destructive workers hangs out in the pool of the long-term unemployed, but they constitute only 1/3 of that pool, though they cannot easily be distinguished at the interview stage.  1/3 a chance of getting a minus 120k return will scare a lot of employers away from the entire pool.  The employers are behaving rationally, yet it can be said that “there is nothing wrong with most of the long-term unemployed.”  And still they can’t get jobs and still nominally eroding the level of wages won’t help them.

In the perceived, statistical, expected value sense, the lot of these workers is that of ZMPers.

One policy implication is that it should become legally easier to offer a very negative recommendation for a former employee.  That makes it easier to break the pooling equilibrium.  There also are equilibria where it makes sense to “buy the NMPers out” of workforce participation altogether, pay them to emigrate, etc., although such policies may be difficult to implement.  Oddly, if work disincentives target just the right group of people — the NMPers — (again, hard to do, but worth considering the logic of the argument) those disincentives can raise the employment/population ratio, at least in theory.

Addendum: Garett Jones offers yet a differing option for understanding ZMP theories.

rifaat April 18, 2013 at 10:03 am

I hadn’t realized that ZMP workers were just assholes.

Frederic Mari April 18, 2013 at 10:19 am

To bounce on that, I too find this a very dubious idea. Sure, it makes intuitive sense (‘Nobody likes assholes and employers aren’t looking for trouble-makers, from union leaders to disgruntled pessimists’) but it ignores that employers can be assholes too, bosses certainly can be (Tyler takes a few swings at academia. Ever tried finance?) and somehow THAT did not stop them from reaching their dizzy professional heights.

So yeah recruiting nay-sayers and trouble makers probably is something employers are afraid of but discriminating against ALL long term unemployed on the concept that ‘something must be wrong with at least a meaningful percentage of them’ was also an argument used to justify racial and sexist discrimination. “Not all Blacks are… but enough of them are…” with the ‘best’ of the discriminators arguing “it’s not even race/sex genetics, it’s cultural”…

RPLong April 18, 2013 at 11:18 am

You and rifaat have said everything I was thinking about this.

john personna April 18, 2013 at 11:46 am

Tyler takes the ZMP concept a bridge too far. It is entirely appropriate to think that organizations have a structure/opportunity situation that makes +1 worker, even a good worker, unproductive. I’m afraid it becomes a just-so story though, to say that since we cannot believe most companies are in such situations (because that would be a Krugman), we must assert something about the general quality of +1 workers. They must all be bad, because that makes markets correct, and efficient, and firing on all available cylinders, etc.

Chris April 18, 2013 at 3:02 pm

To be fair, the point isn’t that all marginal workers are bad, but we cannot a priori tell the difference, and there are enough out there to make getting one a real risk.

That said, I think the “they’re assholes who lower productivity of others” is a stretch too far. They don’t have to contribute -$80k on a $40k salary to be a risky hire, even contributing $30k on a $40k salary is a bad hire, particularly for a small company.

john personna April 18, 2013 at 4:58 pm

I’m fine with those who say we should lower the hurtles for trial periods – though I think they should be at full pay, to prevent a “spin cycle.”

John April 18, 2013 at 6:21 pm

@ john personna

If I understand your statement you’re misinterpreting Tyler. His claim is not that organizations are at some well-balanced maximum capital-labor, or even labor-labor, point such at adding any new worker will reduce total organizational productivity.

He’s talking about interpersonal work dynamics where the wrong person added to the team can change the dynamics from positive to negative and so reduce the teams productivity.

sp6r=underrated April 18, 2013 at 11:38 am

Great reply but I think readers need to be realistic. Tyler isn’t giving his honest opinion of these workers because the description of why ZMP are ZMP changes month to month based on the data. What Tyler is doing is trying to provide an intellectual justification for the government not involving itself in helping ZMP workers.

Tyler’s view is that government intervention into the labor market is a drain on long-term growth. The benefits of growth compound so the costs of intervention into the labor market on behalf of ZMPs outweigh the benefits. He isn’t willing to make this argument honestly because he feels it would be unpersuasive so he results to deception.

If the ZMP are just assholes, Tyler Cowen is just a bullshitter.

Christian Gormsen April 18, 2013 at 10:40 am

Sounds like a great argument for relaxing firing regulations: “We cannot fire the assholes.”

John B. April 18, 2013 at 10:50 am

Yes! The obvious answers are first, to ditch legal wage restrictions so that people can be offered extra-low wages during a check-out period and second to ditch the “employment protection” laws/regulations so that an employee can be fired without the threat of an expensive wrongful termination suit (i.e. permit “fired because we say so; we don’t need a reason”). Getting rid of disparate impact regulations would be a big help as well for minority candidates.

Unfortunately , most people don’t understand that if firing is expensive, hiring will be discouraged, so I don’t think attempts to switch to a more sensible employment policy regime can ever get political support.

bjssp April 18, 2013 at 11:34 am

How much of the cost of firing is due to regulations versus training someone new for the job, to the extent it’s required?

Frederic Mari April 19, 2013 at 4:35 am

And, in France, this is an issue I would certainly agree with you upon. It’s not just the assholes you cannot fire but also the low quality employees in general.

OTOH, I’ve worked in the UK where the barriers to firing are pretty light and you STILL have companies complaining about how difficult it all is. I am not very experienced with the US situation but I somehow doubt that it is radically more difficult than in the UK (I am open to being wrong on that).

One would start to suspect that the UK employers find it difficult because it takes a bit of ruthlessness to deprive people of their means of existence… i.e. inertia and some remains of decency makes it hard for at least a fair number of managers to just fire people, regardless of how easy it might be.

Greg April 20, 2013 at 2:11 pm

My relatively uninformed view is that it’s even easier to fire people in the US than in the UK. Much employment here is theoretically “at will” – either party can end it for essentially no reason. But once you get HR and the legal department involved, the whole process becomes much more risk-averse. And perhaps given how litigious the US is, HR simply knows something I don’t. So the actual regulatory framework isn’t oppressive, but the risk of a lawsuit may be.

Frederic Mari April 22, 2013 at 3:06 am

Thanks for the precision. And, for the UK, I don’t think suing is a major major issue. British may sue more than French (imhe, open to being wrong) but I am pretty sure they sue less than Americans.

Still, in the US, how often do courts find in favour of employees in wrongfully dismissed cases? ‘seems to be above 50%

Thelonious_Nick April 18, 2013 at 11:48 am

Tyler never said they were assholes and I’m not sure why everyone is assuming what he means. I’ve certainly worked with ZMP workers–one who comes to mind was a great guy, but hardly did any work and spent all his time moving cubicle to cubicle chatting up co-workers. He certainly lowered productivity, and once he was talking with you it was impossible to get rid of him. But I wouldn’t call him an asshole.

mavery April 18, 2013 at 12:52 pm

But was this the guy who got fired first? Would he have been less likely than his peers to be rehired initially but then become indistinguishable from his more hard-working peers who also lost their jobs and hadn’t found one after six months? That’s the mechanism that’s weird about this explanation, not whether or not they’re assholes.

Part of me thinks this is Tyler ridiculing the idea of ZMP workers by pointing out the ridiculous implications that result from ZMPs being the cause of long-term unemployment.

Michal April 18, 2013 at 1:59 pm

Working in a software industry, I can attest that there are people who are a big drain on your resources. One person day of poor quality coding will cost you several person days of debugging by your brightest developers. The downside is not only the opportunity cost but also that your brightest developers understandably hate cleaning others people mess. If you use them too much for this work (what is what you have to do to have a working product) you’ll eventually lose them. To me, the ZMP theory presented by TC makes perfect sense for “O-Ring” industries, see

From my experience, this is what the employers intuitively feel and while there is a vicious fight for accomplished developers, new grads without any prior experience have pretty hard time even to get to an in-house interview (I am talking about smart kids with great GPA’s from the best universities in technical fields but little hands on programming experience).

Michal April 18, 2013 at 2:13 pm

Another important point is that it hard fire a bad apple. During the interview process it is the candidates part to prove that they will be contributing and the default “do nothing” decision is to not hire the candidate. On the other hand, firing a ZMP employee means that somebody needs actively to take the action and defend it (“do nothing” decision is to keep the person on the team). I have seen many bad developers that weren’t fired even though everybody in the team suspected they are somewhere between “negative MP” to “very little MP”. Hope this sheds some light on the question why hiring managers might be ueber-conservative in some industries.

mavery April 18, 2013 at 2:14 pm

Are there very many programmers among the long-term unemployed?

While I don’t doubt everything you said is true, I’m not sure this is the group we’re talking about.

Dan Weber April 18, 2013 at 2:48 pm

No, programmers have very low unemployment right now.

Computers can expose other ZMP workers, though. Maybe some goofball opens the attachment everyone knows not to open and now you have to spend 3 weeks cleaning out a malware infestation.

Michal April 18, 2013 at 11:04 pm

These programmers will not end long-term unemployed but they may be pushed out of some more lucrative sub-industries. Programmers who are considered NMP for Google might still be employable by less sophisticated companies.

But the basic principle holds and I can easily imagine that in some industries the “less sophisticated companies” are not hiring or they do not want to hire “overqualified staff” (yes, I have seen people rejected for that reason) or they already outsourced all the “low paying” jobs (thus hiring the employee in US would mean rebuilding a lot of infrastructure that was already moved somewhere else).

I am afraid that the current economy is pushing more and more companies to the “high productivity regime” where they have to compete for a limited number of high skilled workers while they have no use for the low skilled ones.

mavery April 18, 2013 at 10:16 am

So the basic theory here is that a bunch of jerks lost their jobs along with non-jerks, and because of inefficient signaling during the hiring process, employers are unable to distinguish between them?

But you also presuppose there is a higher incidence of “being a jerk to the point of negative marginal product” among the long-term unemployed. This seems to indicate that either “being a jerk” makes you less likely to search hard for a job early on (possible, I suppose) or that employers were able to sort these folks out at some point when they were recently fired, leading to them being unemployed long-term with higher incidence, but then lost that ability to filter once they had been out of work for six months. I guess what it comes down to is that this group of folks who are such jerks that they have large, negative marginal products are in some fundamental way distinguishable from subset of the workforce (those able to get jobs within ~six months of losing their previous one) and indistinguishable from a bunch of workers who were not able to get new jobs quickly but are otherwise perfectly employable, positive marginal product folks.

This seems unlikely to me. At the least, less likely than the idea that employers get lots of applications when there are lots of people who are unemployed and one of the quickest ways to filter is how long someone has been unemployed.

Marcos April 18, 2013 at 10:29 am

“At the least, less likely than the idea that employers get lots of applications when there are lots of people who are unemployed and one of the quickest ways to filter is how long someone has been unemployed.”

Let me see if I understood correctly. You are saying that the most likely explanation is that employers think that Tyler is right, but it is unlikely that he’s right?

Makes a lot of sense.

mavery April 18, 2013 at 12:48 pm

What an obtuse reading. I was saying that if Tyler is right, the most likely explanation seems to be the one I gave. That being a very complicated and counterintuitive theory, I pointed out a much simpler and satisfactory one.

dead serious April 18, 2013 at 8:26 pm

Your post made sense. Marcos is a ZMP reader.

Marcos April 19, 2013 at 10:57 am

Why would an employer even look at the unemployment time if he didn’t belive it correlated with some undesiirable characteristic?

Frederic Mari April 19, 2013 at 11:18 am

Because he is just victim of groupthink and discriminating baselessly?

Thomas DeMeo April 18, 2013 at 10:21 am

I really don’t know why economists don’t see this, but creating a scenario where wages weren’t sticky would never work on a wide scale. Most employees don’t want to take on so much of the business risk, and most employers don’t want to share so much profit. Keeping these roles clear is a feature, not a bug.

wiki April 18, 2013 at 10:49 am

Economists do discuss this issue in microeconomics of organization classes. It’s why paying janitors in stock options doesn’t work even if average compensation would be the same or slightly higher. It’s even why few top executives are purely compensated in terms of varying performance (usually it’s salary plus bonus). But if that’s part of the story, then that supports all the theorists of nominal rigidity. It leads to macroeconomic externalities requiring monetary easing in bad times. What is optimal at the individual firm level may have consequences for the entire macroeconomy. And this ignores the psychological aspects of wage flexibility and worker morale.

Thomas DeMeo April 18, 2013 at 11:53 am

I think you are ignoring the fact If wages are adjusted to absorb risk that employers used to bear, we will get riskier behavior by the employers.

Willitts April 18, 2013 at 10:34 am

Too often it is the managers who are the ZMP employees. If they aren’t control freaks who are destroying the individual contributions to the team, they are an empty chair collecting information that doesn’t contribute to output.

Finch April 18, 2013 at 10:53 am

Yeah, but this isn’t about some sort of balance of badness. The relative badness of current employees and managers is irrelevant. You still want to hire to keep out costly mistakes. The probabilistic explanation makes a lot of sense.

dead serious April 18, 2013 at 8:33 pm

Yes but one manager salary equals 1.5 – 3 “ZMP” workers. Possibly more.

Get your management hiring wrong and it doesn’t matter if you have all non-ZMP staffers. They’ll leave. I’ve left companies for that reason.

superflat April 18, 2013 at 10:41 am

i love reading the posts that make clear that, even if tyler is a supergenius, like the other supergeniuses, he gets stuff wrong. this just doesn’t seem plausible at all, while g. jones has a good point. basically, the employer/employee relationship is quantized, based on all sorts of fixed costs, expenses, etc., so an employee who might be able to add something to a business at $5/hour and/or no health benefits is a ZMP employee given minimum wage, or health care law, etc. this also seems to make ZMP case-specific — it’s not that these folk have nothing to add to any business, but perhaps nothing to add to any business that’s hiring (e.g., they aren’t hiring more garbage men or whatever (no disrespect to garbage men, i’ve hauled trash, but you obviously don’t need to know excel or whatever for a job like that)).

whatsthat April 18, 2013 at 11:23 am

As we have learnt, knowledge of excel is not really a requirement for economists.

wiki April 18, 2013 at 10:44 am

I understand, it’s like trying to parse comments while skipping those by prior_approval.

Brian April 18, 2013 at 10:51 am

“One policy implication is that it should become legally easier to offer a very negative recommendation for a former employee.”

If it really takes an elaborate exposition of ZMP to justify this sensible measure, then we have much more to worry about, as we already do, for labor policy.

We have a major case study already, in the form of K-12 Teacher Unions, of where it is both difficult to fire & difficult to give a poor recommendation to a bad employee. The unfortunate results of this are now more widely (?) known.

If it can be identified that a more general labor policy is closely similar to anything inside the labor rules guiding Teacher Unions, it should be adopted with caution, and retained only if the resulting impediment on growth and innovation is acceptable.

Mark April 18, 2013 at 11:04 am

In discussions of K-12 education, unions, teacher performance pay, etc, there often seems to be an assumption that the school administration (principals, school board) are themselves competent and well-meaning. As in, if it weren’t for the teachers unions, the school administrators could be running things so much more efficiently. I’m not so sure about this.

Brian April 18, 2013 at 12:47 pm

I accept this point fully in so much that we would acknowledge who is the ultimate “consumer” of the education – which seems reasonable to assume would be the families of the students. The way the current system is set up, the administrations, as you point out, just might have too much arbitrary power and little accountability. They are subject to the school board, which is often an elected body – but this bring us back to my premise here. The families should be – already are – the consumers, and allowing them to exercise choice, would create forces and pressures that do the hiring and firing anyway.

Having an elusive “democracy” in the form of school boards and or dysfunctional school administrators be the final say is a problem, but the appropriate solution is not to make teachers less accountable. The argument you present (which i often hear) is that (good) teachers need these overreaching protections from the horrible administrator.

If vouchers are too extreme, then charters serve some of the function of allowing families to choose the kinds of schools (and teachers) that they find best. Teacher Unions fight tooth and nail against vouchers. Teacher Unions often really don’t like charters. But parents do know who the good teachers and bad teachers are – but even at the site level, it is often not allowed for students to pick the section with the teacher that they want – and this comes from both the administrators and the union.

bjssp April 18, 2013 at 11:46 am
Jeff April 18, 2013 at 10:53 am

It isn’t just people with bad social skills. I do IT consulting with federal agencies and once encountered a long-time federal software developer whose code was so badly-written that nobody else could use it. This became apparent when he went on vacation and nobody could figure out wtf his code was doing. Later during a large conversion project none of the code he had written could be ported over, everything had to be started from scratch.

Within a year he was not getting any project work worth much.

NPW April 18, 2013 at 10:57 am

Tyler is right on ZMP workers.

Pointing out that there are good people who are long term unemployed does not eliminate that fact that if I have a thousand resumes on my desk there is a higher likelihood of finding a good contributor who is currently or recently employed. This is the same type of rationale for hiring a college grad when a 6th grade education would work.

Just because something isn’t “nice” doesn’t mean it isn’t part of an adult reality. Some of these comments sound like a child whining, “That’s not fair”.

sp6r=underrated April 18, 2013 at 11:42 am

As another poster pointed out your argument for discrimination against the long-term unemployed on the grounds that they are less skilled on average is the exact same argument that employers made for discriminating against African Americans who are less skilled on average.

Cliff April 18, 2013 at 12:06 pm

Except that argument is completely specious. Discriminating against young men on insurance premiums on the grounds they are more aggressive drivers on average is directly analogous to the argument made by employers for discriminating against African Americans. See how that works? It was only bad because discrimination on the basis of race is illegal. Discriminating on the basis of long unemployment is not illegal, nor should it be. There is no long history of irrational prejudice against unemployed people.

sp6r=underrated April 18, 2013 at 12:36 pm

Actually there is a long, document history of irrational prejudice against unemployed people in the labor market.

MikeDC April 19, 2013 at 11:50 am

The problem is that it’s completely rational prejudice.

The hiring agent has limited time and lots of resumes. If you have 1 position and 100 resumes, you have to prioritize them via some legally acceptable means. You don’t get any certainty of getting a good employee in any case, but you want a cost effective means of maximizing your chances. In my experience, this process tends to work along the lines of:

1. Anyone with an insider recommendation and remotely close to qualified gets the first shot.
2. Exact matches for domain/skill experience. (which is very hard in many fields if you’ve been unemployed for a long period)
3. General matches (is this guy trainable?) based on education and apparent worker characteristics (does he look like a solid employee, or does he bounce around a lot).
4. A complete shot in the dark… everyone else.

Point is…. you’re very often going to find multiple candidates before getting to the long-term unemployed.

sp6r=underrated April 19, 2013 at 3:21 pm

The logic you used below justifies discrimination against certain racial minorities such as African Americans and Native Americans who much like the long-term unemployed are under-skilled in comparison to the general population.

NPW April 18, 2013 at 12:06 pm


Hitler liked his dogs, therefore, all dog owners support concentration camps. The previous poster’s arguement was just as flawed.

Frederic Mari April 19, 2013 at 4:53 am

Assuming I am indeed the previous poster you’re referring to, this was not my argument.

You and Cliff are saying there are observable proofs of LT unemployed lower contributions if they were to be recruited. As sp6r points out ” there is a long, documented history of irrational prejudice against unemployed people in the labor market”. As there was/is against Afro-Americans, including by Afro-American managers/business owners.

And against women. One reason women aren’t promoted to true positions of power is that women are deemed to have less “strategic vision” than men. Personally, I would question the value of ‘strategic vision’ but what the hey, it’s deemed important at the top levels. Even women are prone to stereotype other women in this way. However, as it happened, inasmuch as something like ‘strategic vision’ can be assessed, women do not have any less of it (or more) than men i.e. it’s pure, shared, prejudice.

NB: I am not against the concept of ZMP employees per se. I think it’s just a way to describe a situation – ‘this employee is not contributing in this position at this company’ and, a bit more generally, a way to think about technology – ‘technological evolution is making a lot of people redundant’ as well as employers’ practice – ‘squeezing the life out of our remaining employees is an awesome way to reduce labor costs while not losing anything’…

mavery April 18, 2013 at 12:55 pm

You just said Tyler was right and then used a completely different argument about why long-term employment is a problem. Your explanation is far more reasonable than his. I’m not sure where your agreement comes from.

zbicyclist April 18, 2013 at 10:57 am

This seems labored to me.

ZMP workers are those whose skills can’t be productively used (e.g. steelworkers after the mill closes).

NMP people whom I’ve known aren’t so much assholes (many assholes have areas of great competence — that’s why they still have jobs) as Peter Principled, people who have been promoted beyond their competence and therefore make bad, very expensive decisions.

Yet in my experience NMP people don’t really get their comeuppance — even if fired, they get jobs other places. These may be smaller places but still decent jobs.

An example might be Gerry Faust, the successful high school football coach who utterly failed as coach of Notre Dame, but was then hired as head coach at Akron. When fired there after a 1-10 season, he became a fundraiser for the university. He now works as a motivational speaker.

Ricardo April 19, 2013 at 4:46 am

Heh, I was reading the Wikipedia page of Nick Leeson — a former trader based in Singapore who caused the collapse of Barings Bank — a while ago and he, too, also became a motivational speaker after serving time in prison. It seems like the ideal profession for failures and frauds who can draw on some personal charisma.

Cyrus April 18, 2013 at 11:03 am

Obnoxiousness is one of the few traits that seems plausibly discoverable at the interview stage.

Go Kings Go! April 18, 2013 at 11:03 am

So late-career Antoine Walker and Stephon Marbury?

ohwilleke April 18, 2013 at 11:16 am

Essentially the same line of reasoning explains why nepotism can be an attractive hiring practice.

bjssp April 18, 2013 at 11:42 am

Funny, I was just thinking the same thing, although in my experience, it can backfire perhaps even more easily than it can work out.

James April 18, 2013 at 11:20 am

“To draw a simple analogy, the lowest-publishing members of academic departments are rarely those who make the most trouble.”

I suspect that this might not be true for more teaching oriented schools.

sp6r=underrated April 18, 2013 at 11:33 am

ZMP workers as another poster pointed out are just assholes according to Tyler Cowen. What I want to know is why the percentage of workers who are assholes jumped dramatically in the U.S and Europe at the exact same time?

Finch April 18, 2013 at 11:54 am

Not that I totally buy into this, but the idea is that a new way of distinguishing assholes appeared (inability to convince someone to hire you over a long period of time), not that the number of assholes changed. In previous tight job markets, you just had to take your chances and hire more risky people.

Finch April 18, 2013 at 12:00 pm

I don’t know why the focus is on people being jerks. There’s any number of reasons someone could stay long-term unemployed that would make them a bad potential hire, such as a lack of industriousness, problems that show up in interviews, bad references, etc.

sp6r=underrated April 18, 2013 at 12:35 pm

Why not make the labor market tight again so people can get jobs?

Finch April 18, 2013 at 12:53 pm


Furthermore, there might not be a path from here to there. For purposes of argument accepting this thesis, maybe long-term unemployment has outed a bunch of bad employees. So even if the economy grew a lot, employers would be really reluctant to add the folks that spent a long time out of work in this period because they know they’re bottom-of-the-barrel.

What would help the fraction of good employees stuck in long-term unemployment is some sort of reliable way of distinguishing them from the bad employees, given that employers appear to be assuming the long-term population is biased towards bad employees. The bad employees need to seek employment and wages more appropriate to their abilities; good and bad being relative to the requirements of a job.

bjssp April 18, 2013 at 11:41 am

This is intuitively appealing in some senses, but it seems rather limited as an explanation, because it seems pretty damn difficult to distinguish the person who is dead weight and the person who didn’t do anything wrong but got the short of the stick (because they worked for an unlucky/unprofitable employer, because they worked in an industry that was contracting, because they were the last hired, and so on). Not impossible, mind you, but certainly difficult.

Is it too much to think that process of not hiring the unemployed is equally logical, in that their skills really have diminished, and arbitrary, because those hiring really do need some way to shorten the pile, and this, while kind of heartless, makes the job get done much quicker?

Or maybe, just maybe, cyclical unemployment really is becoming structural–when it never had to be in the first place, were we to have better fiscal and monetary policy?

Chris April 18, 2013 at 3:41 pm

You are in violent agreement with the premise.

To the degree the explanation is effective, it is precisely because it is difficult to distinguish between the deadweight and the unlucky. They (presumably) know which they are but employers don’t.

Its just like the used car market problem. Sellers know whether or not their car is a lemon, but buyers do not, so they assume everything is a lemon. Increasing information symmetry, such as through services like Carfax or “Dealer Certified” used cars mitigates this problem – just like reliable and trusted employee references would.

Max April 18, 2013 at 11:44 am

I think the idea that a pool can appear ZMP due to some who are NMP makes sense, but I don’t think the “negative morale cost” is the reason.

In my own office where we are trying to hire someone but keep turning people down, the concern seems to be of hiring someone who will take up more of managers’ and coworkers’ time in managing them than they will produce in output. If my manager makes 2x what I make, and he spends 2 hours a day thinking of tasks for me to do, providing guidance, and checking my work, I have to produce 4 hours’ worth of “analyst” quality work for the firm just to break even.

I think our concern is that many people would produce, not negative “gross output,” but less than 4 hours’ equivalent of gross output, or require more than 2 hours of guidance/checking from one or more managers/coworkers. Either of these scenarios would be ZMP for the firm.

Bob April 18, 2013 at 12:41 pm

In some industries, most new hires are NMPs for months. And if you don’t pay attention to them, trying to turn them into just ZMPs, they’ll quit, because they are not learning anything. This makes hiring in those industries very difficult: You want to hire extremely senior people that will be useful relatively quickly, but once that pool runs out, nobody wants to train the next rung of people, because then you eat the training costs, and once they reach the senior status, you can lose them before you break even.

Not great for the economy, but wonderful for those that are senior already, as they are so in demand that they can change employers every year with a significant raise to come with it.

All of this could be mitigated by more complex employment contracts, instead of at-will agreements.

Max April 18, 2013 at 3:08 pm

I agree most people are NMP when they start, but I think there’s a concern not just that some willl be NMP, get up to speed, and then choose to leave. I’ve heard a concern that some may be NMP, not get up to speed, and choose to stick around for a few years. Since no one enjoys firing people, there’s a strong aversion to accidentally hiring someone like this.

Brian Donohue April 18, 2013 at 12:28 pm

I dunno. If what you say is true, it shouldn’t be too hard for the 2/3 ‘deserving poor’ to distinguish themselves from the bad apples. It’s likely that such a sorting process has already been going on for a while.

sp6r=underrated April 18, 2013 at 12:42 pm

If Tyler Cowen is correct about ZMP and that discrimination against the long-term unemployed by employers is rational and economically beneficial the logical result is that the government will have to either create jobs specifically for these people or we have to return to the old welfare for life system.

Ray Lopez April 18, 2013 at 12:54 pm

Here are my thoughts:

1) Doubtful if there’s an information dislocation (“Market for lemons” problem): Negative recommendations are rare outside of government former employers, who are immune from lawsuits. But my experience is that even negative recommendations (from govt or even from private employers) mean little. The market determines whether you will get an offer. In a hot field, or with limited candidates, you’ll get a job even if you’re an ass.

2) Yglesias’ hypothesis seems more plausible, citing ‘geographic dislocations’ (e.g., why doesn’t everybody move to DC): TC’s hypothesis seems a bit ‘just so’. E.g. color discrimination as rational–well, it could be explained with TC’s theory, or, it could be an example of ‘groupthink’ that lasts decades. Take Jackie Robinson–why did baseball employers discriminate vs blacks? You could say that white ballplayers would be less productive if blacks were on the team, or that fans would not pay to see black ballplayers, etc, or you could say it was just groupthink that discrimination happened for so long (in fact, I read a piece by Sol White [sic], “The Color Line” (1907) bemoaning color discrimination in baseball–from back then. It took 40 years to overcome this groupthink). Same logic for ‘openly gay people in the military’ (would hurt morale in combat–or not)

3) I just read the most cool, most awesome, most controversial, ‘alternative economics’ article by an economics person, here: (arguing prices are not only sticky based on empirical surveys–citing Alan Binder–but also marginal costs actually are falling with more output, contrary to classic theory, and perhaps consistent with non-perfect competition)

Conclusion: prices are sticky due to: geographic dislocations, groupthink (goes in cycles), and non-perfect competition, but ‘market for lemons’ signaling does not appear to me to be a problem.

Ray Lopez April 18, 2013 at 1:04 pm

Note: in my definition, ‘market for lemons’ information asymmetry is rational, whereas ‘groupthink’ is not. I am saying the former is absent in the actual labor force but the latter is present. Another way of presenting this problem: if ‘market for lemons’ is the case now, why not back in 2006 when employment was higher? Or 1999? Back then employers would take a chance with anybody with a pulse (no experience necessary, or out of the labor force for years, etc). So it’s not market for lemons, which is rational, it’s more like HR departments will, when demand is down, not take a chance with long term unemployed since they ascribe to the theory ‘don’t take chances with long-term unemployed’–and this flawed logic is pervasive. When AD is up, HR has less sway since managers will override their HR departments.

collin April 18, 2013 at 1:01 pm

I have come to accept ZMP in the modern workplace because

1) It easier than fire 5% of employees than negotiate with 100%. Realize the if the High Marginal workers think their salaries are cut will leave the firm.

2) A lot of ZMP workers are ZMP projects. They are worth the money but most new projects do come out as planned so you have ZMP workers. Think of Microsoft various new markets here being the most obvious example.

3) In a lot positions, one can be ZMP one year and the hero the next. Also most workers are ZMP during their first 3 – 6 months. For example Matt Y. blog post diminished a lot when he moved from ThinkProgress to Slate and after a year they have improved.

allan harris April 18, 2013 at 1:06 pm

Cowen identifies the ZMP worker and advocates firing him or her. Cowen, however, after parenthetically mentioning the ZMP boss, does not explain how to eliminate the ZMP boss problem. Since Cowen is not simply an ass kissing apologist for capitalist bosses, I’m sure he just forgot to explain how to fix the ZMP boss problem.

Here is the best solution: the workers take a vote and the ZMP boss is told to clear out his/her desk and be out of the building by 5:00 p.m. Result: huge increase in worker morale as they cheer the ZMP SOB out the door.

dirk April 18, 2013 at 1:07 pm

That makes sense to me. It explains precisely why I’m a ZMP worker.

My guess is that the boom-busts of the past two decades *created* a lot of ZMP workers. It became fashionable to job hop because one could easily give themselves a 50% raise every time they changed jobs (I job hopped a lot). It was tolerated during a boom, and one even felt like one of the smart ones for doing it. But the bond of trust between employer and employee has been irredeemably broken. No longer can an employer view an employee as a long-term investment. Therefore, the skills mismatch is a bigger problem, because firms don’t trust employees to still around long enough to pick up the relevant skills for their positions. Employees are hired to “do a job” not to build organizational capital, because the odds are low they will stick around long enough to build organizational capital. This would explain the extreme prejudice against the long term unemployed. The odds are better that a recent graduate will stick around for ten years than that someone who has been in and out of the job market over the years — and experienced the luxury of a boom — will.

bjssp April 18, 2013 at 1:15 pm

But isn’t the sort of person who can job hop so easily also the sort of person who is likely to require the least amount of investment by a firm in the first place?

dirk April 18, 2013 at 3:28 pm

Yes, if the skills are a perfect match. But if there is any sort of skill-mismatch, which can come from long-term unemployment, such a person will not be perceived as a good investment.

Andrew' April 18, 2013 at 1:32 pm

I’ve never seen an employer actually do anything for morale.

Andrew' April 18, 2013 at 1:35 pm

We had this super-shitty truck that we had to use to get stuff. Once the department head had to drive it. They made a speech afterward about how that shitty truck was bad for morale. So, they got rid of it. What they were really saying was that shitty truck was bad for image. Having to walk, which we did after they nixed the truck, was bad for morale. That’s not their concern.

RPLong April 18, 2013 at 4:26 pm

Allowing me to add emoticons to my entries in the time-management system would improve my morale at virtually no cost to the company. Where’s MY little slice of heaven???

Andrew' April 18, 2013 at 1:35 pm


Chris April 18, 2013 at 3:47 pm

Andrew’ you are a nut. I mean that as a compliment.

Paul April 18, 2013 at 1:55 pm

Perhaps it might be better to say that ZMP workers are those who impose negative externalities upon the development of organizational capital within the firm, particularly affecting their fellow workers?

Arthur April 18, 2013 at 1:57 pm

Employers have the impression that all long term unemployed are assholes? I don’t think you are on to something. Caplan’s theory seems better.

Chuck Ross April 18, 2013 at 2:38 pm

Long-term unemployed aren’t necessarily assholes as some people here are suggesting was the point of Cowen’s piece. Long-term unemployed are just more likely to be the type of people that melt the glue that holds a workplace together. That’s one type of capital that isn’t mentioned.

They can be assholes or they can bring drama to work or they can sleep with too many people or sell drugs to other employees. These don’t act through traditional forms of capital – they don’t destroy value that can be measuring on a balance sheet, but that’s because, as Tyler says, those employees aren’t given the opportunity to do that. This falls on labor capital.

Yog Sothoth April 18, 2013 at 2:40 pm

Temporary hiring would seem like a solution to this problem. Give somebody 3 months. If it doesn’t work out they only cost you one quarter of negative productivity. Otherwise, you found a useful employee and can retain them longer term. It’s not a frictionless solution but it seems like one channel American labor markets could use to sort out the problem over time. My company did this. We hired four temporary people for a short term project. I’d say one was NMP, one was ZMP and two were PMP and were offered permanent positions.

Donald Pretari April 18, 2013 at 2:50 pm

Here on Planet Earth, Assholes rise to the Top and hire Acolytes. ZMP Workers are Highly Competent Workers who can’t Stab People in the Back or Kiss Ass. They’re the so-called Salt of the Earth. They don’t do well here.

Chris April 18, 2013 at 3:48 pm

I can’t decide if you are Trying to Write in Titles or are German.

RPLong April 18, 2013 at 4:28 pm

Thus Spake Donald Pretari.

Donald Pretari April 19, 2013 at 2:22 pm

My Point was that it’s probably a Cornucopia of Causes, but I was having a Ripe Day yesterday. I Apologize.

Donald Pretari April 18, 2013 at 3:57 pm

I can.

Philo April 18, 2013 at 4:32 pm

Tyler should have stipulated not just that NMP workers constitute 1/3 of the long-term unemployed, but that they constitute *much less than* 1/3 of other potential workers. By the way, if Tyler’s “statistical discrimination” scenario were accurate there would be a huge market opportunity for the rare HR specialist who *could* readily distinguish, among long-term unemployed workers, those with PMP.

Claudia April 18, 2013 at 4:59 pm

Wow, I totally misunderstood what ZMP workers were. Thanks for clarifying, but you might consider a better name, something with externalities or morale. I also stand corrected ZMP is probably the most neutral, inoffensive term you could come up with. I still wish you would convey the non-intrinsic, state-dependent nature of the negative externality more. I have had jobs and definitely projects where I am a NMP worker … I showed early on I was not well suited to, even kind of dangerous at, farm work, and even if my current job if there was nothing to do (almost seems like a dream world) I would not be among the last ones kept. I think you’ve got a lot of work to do to flesh this mechanism out, but hey join the club of macro story telling.

Claudia April 18, 2013 at 7:17 pm

Oh and need to make sure no one messes up the ZMP formula in Excel:

John April 18, 2013 at 6:11 pm

While I agree that hiring new employs requires consideration for how team production is affected I’m having a bit of an problem connecting some dots. My recollection is that the ZMP theory came out as something of an explanation of why unemployment was not falling as rapidly as some expected after the recovery started. Perhaps that’s an error on my part but is only slightly important to my question.

If these ZMP workers who are unemployable now at as described why did the have and keep jobs prior to the crash? Is still find the explanations a bit at odds with what we observed. I think the story is much more to do with being in a bubble economy and even if these people were ZMP it simply didn’t matter — managers could brag about the size of teams they managed and other non-corporate profit related rewards.

Alvin April 18, 2013 at 6:52 pm

One of the reasons I left my previous job (small office part of large university) was because of an asshole co-worker (who wasn’t even productive) that nobody liked but management didn’t have the balls to discipline or fire. So I took matters into my own hands and let the office bully have it a couple of times (after getting the green light from the boss). Of course, like every bully I’ve encountered, he ran to the boss and complained. We had the most surreal meeting after that and the boss took the side of the bully. I got my new job offer about two weeks after that meeting and walked right out after I signed with my new employer. No two week notice, nothing.

Floccina April 18, 2013 at 8:07 pm

If that were the problem why would hiring temps and then moving the best to permanent not solve it.

Bryan Willman April 18, 2013 at 9:41 pm

Apparently TC has never had to deal with someone who is flat out incompetent. The most awesome negative output people I ever met were (a) wonderful human beings, (b) diligent! and “hard working”, and (c) made such awful messes that it cost something like 4x their salary to clean up after them.

The whole “signaling” theory of education has gotten too much play, there ARE INCOMPETENT PEOPLE IN THE WORLD. Really.

Avoiding these people can be THE number one goal of a hiring manager.

Douglas Levene April 19, 2013 at 1:50 am

I concur with the comments above that making it easier to fire employees would significantly increase the likelihood that risky employees will be hired. Under current law, a terminated employee can bring a suit for age discrimination, disability discrimination, racial or gender discrimination, whistle-blower retaliation, and so on. The costs of defending such cases, especially where the employer has done nothing wrong, can be quite burdensome. It’s so much easier just to forgo hiring the high risk employee. If you really wanted to help the long-term unemployed, you’d repeal all the anti-discrimination and whistle-blower laws and restore the right to hire and fire for any or no reason.

Minority Bolshevism April 19, 2013 at 2:04 am

“One policy implication is that it should become legally easier to offer a very negative recommendation for a former employee.”
That makes a lot of sense. Maybe too much.
Therefore, it is more likely to have a bipartisan ‘No ZMP Worker Left Behind Act’ introduced in Congress.
Any bets?

NL7 April 21, 2013 at 1:23 am

Maybe allowing employers to pay probationary employees less, or even zero, for some period could also help. This potentially runs afoul of various laws, including minimum wage, because internships must be for the benefit of interns and not the employer. The limit on this practice is that it can spoil the new relationships with employees. Maybe the wages can be held in escrow pending successful probation. There is the question of exploitative employers getting a few weeks or months of free labor by falsely claiming unsatisfactory performance; not sure the best way to solve this but some still might think it better than unemployment despite this risk.

best way to make money at school April 27, 2013 at 3:14 pm

As with lots of MLM opportunities, the turn key, automated home business
uses an automated system that works around the clock selling the ideas or products
available, whilst you sleep. Several skills and expertise
that men and women pay well for include home lessons, writing music or commercial jingles, ads, ghost writing, and affiliate campaigns.
The way to gain that knowledge would be to educate
yourself on every aspect of building a successful business while you can.

Comments on this entry are closed.

Previous post:

Next post: