Heterogeneous Labor

by Jerry O’Driscoll  

In the September 4th issue of the Wall Street Journal, Jon Hilsenrath chronicles the debate over the reasons for persistently high unemployment.  What is being described is the problem of heterogeneous labor.

Labor, like capital goods, is specialized and specific to certain occupations. When those occupations disappear in recession, the next best alternative immediately available locally may pay considerably lower wages. Workers may “know” they have better alternatives, but their knowledge capital has also depreciated with the crisis and downturn. They must search for employment opportunities.

The example in the article is construction and health care. Construction suffers disproportionately in the recession (as it had benefitted disproportionately in the boom). Health care is expanding, but that industry has little need for construction workers.  Nurses are needed, but the period of production is lengthy.  So there is an excess supply of construction workers and excess demand for skilled health-care workers. Those two markets don’t clear each other, except with the most drastic of wage movements or the passage of time allowing for retraining.

Heterogeneity (of labor and capital), structural shocks and time are all grist for Austrian mills.  More Austrians and classical liberal political economists should be jumping into the fray.  The issues are mainly price theoretic; no detailed knowledge of capital theory or monetary economics is needed.

The structural aspect of the recession is now obvious to all but the invincibly ignorant.  The distinguishing feature of the Mises/Hayek analysis of economic fluctuations is its structural character, or what Hayek called the disproportional nature of both the expansion and contraction phases.

“Spending” won’t cure any of these problems, which reflect coordination failures.  Wage and price changes still need to occur. Fears of deflation may lead to policies impeding the needed changes in relative prices.  Long-run adjustments will involve retraining, occupational changes and geographical reallocation. “Compassionate” policies to cushion the adjustments, like lengthening the period of unemployment payments, prolong the adjustment and deepen the downturn.

37 thoughts on “Heterogeneous Labor

  1. For the past 25 years, the US economy has been creating jobs that have not been essential to its long term well-being, neither they were strategically focused.

    Except for the need to fill up retail positions that came into existence as a consequence of the vast expansion in consumption due to cheap and easy credit which was the result of China’s funneling its steady rising surpluses into treasuries driving down yields, no deeply structural changes were made and this period, specially the last decade could be considered a lost opportunity. In a way, one can say much of it was fat now being trimmed.

    It might take a good part of the next 10 years to reform the basis of the US economy into something that is more sound. And this would be like steering the Titanic.

    In the meantime, many of us are going to be suffering as a result.

  2. Subsidizing anything creates a larger supply of it. Subsidizing a good creates more of the good. Subsidizing child care for those on welfare creates more demand for child care by those on welfare, driving up prices for everyone else. Subsidizing unemployment creates — unemployment! People could and should be retraining and readjusting, but are convinced that things will “get back to normal”. The subsidization of unemployment and the government “doing everything it can” to get the economy “back on track” send false information to the public that that is even possible. We could have been out of this by now.

    Now, more and more, I am hearing buzz about an education bubble. As an unemployed Ph.D. in the humanities, that scares the hell out of me. I’m already unemployed as a Ph.D., and it looks more and more like there is going to be a major readjustment soon. My only hope is that the “deadwood” that needs to be cleaned out of education (and it does) will be the postmodernists who think the humanities provides no value to students. Then there might be a place for me in a university, since my approach to the humanities is quite the opposite. Still, pretty disturbing. And when it happens, is the government going to provide extended unemployment to the most educated among us, who are often quite highly paid? Much will have to change, because those with higher educations don’t qualify for most government programs. That will either be good insofar as the system will then be allowed to adjust — or the government will change the law, and all hell will break lose.

  3. Jerry,

    I agree that “heterogeneous labor” (= heterogeneous human capital) is an important reason for unemployment in the US today. Right on. I think you would agree that regime uncertainty is compounding the problem, right?

    As an old commie, I wonder whether it’s such a bad thing to cushion the adjustment process even if it does imply greater persistence of unemployment. I guess I’m asking how one would calculate that tradeoff. I freely admit that I don’t have a snappy answer of my own to defend our current relatively generous policies, though my soft heart favors such generosity.

  4. Rather than cushioning the reajustment process with unemployment payments, while not remove the State’s jackboot from the economy’s neck? That would put more workers back to work faster.
    If I recall correctly, Rothbard referred to labor as a non-specific facor of production in M,E,&S. A humanities PhD could presumably work as a short-order cook, taxi driver (which Bruce Kovner was before he started his hedge fund, on the road to becoming a billionaire), or in similar occupations, whereas a back hoe isn’t good for anything other than digging dirt out of the ground or moving it around above ground.

  5. You’d be surprised. Nobody will hire someone with my level of education. And in order to get hired in a corporation, I’d have to lie about my education, meaning that if they found out, they could fire me for it. So I would always be in danger of losing my job, even if I could move up in the company. I work night audit at a hotel. A good return on my investment. Can’t pay my loans back. So I”m underemployed and stupidly employed, which means I’m a malinvestment even being employed.

  6. I don’t think you have to put your PhD on your resume if you’re applying for a job that doesn’t require one. In fact you don’t have to put your year of graduation on it. Resume experts also say you don’t have to put irrelevant work experience on it if it was more than 15 or 20 years ago.
    Say your BA was 1984, your PhD was 1989, and you had your first job in 1990, but it was a low level job with nothing to do with anything. Then you got a sales job at a local courier firm in 1993, and now you’re applying to FedEx. Start your experience with the 1993 sales job, and list your BA under Education without the year. You want to maximize your chance of getting an interview; they’ll eventually find out the details on the job application.

  7. Ph.D. in 2004, all my experience between then and now have been adjunct teaching positions, except the hotel night audit position. Without the schooling, I end up with insanely long gaps in employment. Don’t see how to get out of it.

  8. I’m overwhelmed by the responses, which are all on target. Thanks to all who commented on a holiday.

    Roger, you’re not a commie, just a classical liberal with a heart. That is true liberalism.

    I don’t know how to calculate the tradeoff on unemployment, but I think we should all be thinking of a better way of dealing with the social consequences of malinvestment.

    I’d like to understand how the Germans came through this so well. No ideology, just facts.

    Regime uncertainty is a huge problem and I hope we all agree about that. Let’s solve problems.

  9. Roger Koppl:

    I agree that most people may be hit hard by the recession without having a financial cushion. However, a structural problem is a structural problem: the policy analyst should ask whether cushioning malinvested human capital against bad outcomes is compatible with the elimination of malinvestment.

    And I’m not talking about efficiency vs fairness: malinvestment is a future liability. I’m talking about welfare now against a higher welfare tomorrow.

    I think that some small help may not be that dangerous. Welfare-to-work policies pushing for human capital retooling and active job-seeking may reduce the problem somewhat. *In theory*

    Most social policies are either of the type “less woes today for some extra woes on the future generation” (like subsidizing unemployment), or of the type “more woes today for all, but gains for my political friends” (like minimum wages).

    In Italy we have a social system which keeps people from losing their job by perpetuating submarginal firms. These corporate zombies are often a drain on economic growth, and the people who have leaked out of the system (the young and the southerners) have troubles finding a job, or finding a good salary.

    From a political economy point of view, it is never wise to assume that politicians care about social gains (they usually prove they don’t, otherwise we wouldn’t be talking about this mess): they usually care about social control, and economic dependency on politicians’ good will is not a social policy, but a political yoke. A free society can’t be a society of beggars depending on the government: but that’s what destroying market and offering a political “social” alternatives achieves.

    This slippery slope has only one stable outcome: an almighty political class, lots of welfare-destroying policies, the atrophy of social cooperative abilities (social capital), economic dependence on politicians, the ruin of the rule-of-law, economic instability, economic stagnation.

    First learn to tame the leviathan, and then ask it for help, eventually. Doing the other way round begets Bushes and Obamas, and welfare losses in the long run.

    It’s not only about designing a good “welfare-to-work” system which doesn’t destroy incentives for recovery. I think this may exist, or at least some marginal help can be given. It’s also about being independent on political whims: this aim is always incompatible with giving politicians free rein and discretion, especially after they caused all this mess.

  10. Jerry, great post,

    Recessions are reorganisations. There is search and mismatch unemployment in recessions because finding the next best job opportunity is entrepreneurial knowledge that must be discovered and tested. This information is not manna from heaven.

    Rothbard acknowledges and then retreats from search unemployment in MES. Firstly:

    “”It might be objected that workers often do not know what job opportunities await them. This, however, applies to the owner of any goods up for sale. The very function of marketing is the acquisition and dissemination of information about the goods or services available for sale. Except to those writers who posit a fantastic world where everyone has “perfect knowledge” of all relevant data, the marketing function is a vital aspect of the production structure. The marketing function can be performed in the labor market, as well as in any other, through agencies or other means for the discovery of who or where the potential buyers and sellers of a particular service may be.”

    This is a good discussion of search unemployment.

    But next, when discussing mismatch unemployment as identified by Hayek after a shortening of the production structure on the market where there might be temporary unemployment of workmen in the higher stages, lasting until the workers can be reabsorbed in the shorter processes of the later stages, Rothbard’s repost to this possible case of involuntary unemployment is:

    “… the shortening of the structure means that there is a transition period when, at final wage rates, there will be unemployment of the men displaced from the longer processes. However, during this transition period there is no reason why these workers cannot bid down wage rates until they are low enough to enable the employment of all the workers during the transition. This transition wage rate will be lower than the new equilibrium wage rate. But at no time is there a necessity for unemployment.”

    Jerry, the nub of your debate by proxy with Rothbard is whether the Austrian natural rate of unemployment is zero? Is any amount of unemployment above zero voluntary unemployment, including for new entrants on their first day of seeking work?!

    Labour is more heterogeneous than capital.

    The notion that buyers and sellers in the labour market can pair up instantly contradicts the Austrian traditions that markets only tends to equilibrium and entrepreneurs are needed to move things along.

    Morgan O. Reynolds makes a good point in his labour economics textbook about how labour markets are different from other markets because there are no speculators and no forward markets in labour to quickly clear the market and allow entrepreneurs to drive the market towards equilibrium through arbitrage as quickly as elsewhere.

  11. Jim Rose: if workers were immediately productive and had no non-monetary costs, labor demand would be infinite at 0$/h, so there would be 0% employment at any time in a positive neighborhood of 0$/h.

    However, even in the freest of the markets, there would be search delays, non-monetary costs (such as office space, training), and a time-structure of productivity (newbies are costly but not productive).

    Besides there may be externalities problems: people accepting low wages during the bust will move to better jobs as soon as they are available. Training costs will be bore by old employees. The time mismatch between training costs and productive labor services is apt to cause opportunistic behavior. I.e., in order to be employed, workers should PAY to (learn how to) work, in these conditions. I wouldn’t call it voluntary unemployment.

    As it was too often the case, Rothbard is making it all too easy.

    Anyway, I do believe that most unemployment is either voluntary or government-caused. People do choose to look for jobs only above a certain reservation wage, and only in a specific sector, or to wait for better opportunities. And often short-term and low-wage hiring is illegal, and there’s lots of additional non-monetary costs involved in hiring due to legal prescriptions.

  12. Pietro M: Robert Lucas in his famous 1978 paper argued that all unemployment was voluntary because involuntary unemployment was a meaningless concept. He said:

    “The worker who loses a good job in prosperous time does not volunteer to be in this situation: he has suffered a capital loss. Similarly, the firm which loses an experienced employee in depressed times suffers an undesirable capital loss. Nevertheless the unemployed worker at any time can always find some job at once, and a firm can always fill a vacancy instantaneously. That neither typically does so by choice is not difficult to understand given the quality of the jobs and the employees which are easiest to find. Thus there is an involuntary element in all unemployment, in the sense that no one chooses bad luck over good; there is also a voluntary element in all unemployment, in the sense that however miserable one’s current work options, one can always choose to accept them.”

    I agree that we all make choices subject to constraints. To say that a choice is involuntary because it is constrained by a scarcity of infomation on job-opportunities is to say that choices are involuntary because there is scarcity. Alchian said there are always plenty of jobs because to suppose the contrary suggests that scarcity has been abolished.

    Lucas, Alchian and Rothbard do under-rate time to build, time to plan and time to train delays in establishing new job capital. Job capital is the physical, human can organisational capital underpinning a job. Job capital is neither free nor freely available.

    Understanding the reallocation of labour at the end of the recession requires careful attention to the 1980s writing of Alchian on the theory of the firm. He argues that you cannot understand the labour market without a working theory of the firm, which is still a work in progress.

    Alchian wrote specifically on how job layoffs would be preferred to wage cuts to protect specific human capital and preclude post-contractual opportunism over dedicated assets. Alchian did not write on contracting for new hires.

    My interpretation is the majority of employment relationships are capital intensive long-term contracts.

    Employers spend a lot of time searching and screening job applicants to find those that will stay longer to recoup the fixed costs of employment and reduce any disruption to team production. In less skilled jobs and spot market jobs, employers will hire the best applicant faster because job turnover costs are low.

  13. I think you are probably right. I’m just puzzled that Rothbard sometimes reasons like Lucas.

    I didn’t know about Alchian’s argument, to me it’s definitely right, and it’s complementary to “my” point. I too think that firm theory would help develop a better understanding of the issue.

  14. Today’s New York Times (Sept.7th) has an interesting article about the slowdown in hiring in the tech sector. I think this is consistent with Jerry’s analysis.

  15. Talking about ‘regime uncertainty’…
    How do we know that it is having significant and quantitatively important effects on the recovery? (I am completely persuaded by Higgs’ argument, but mainstream economists who are looking empirical/econometric evidence all the time may not be so persuaded, and argue (as a professor from Penn argued to me) that ‘regime uncertainty’ may not be so important.

    Also, I would like to ask, as a Spanish citizen who doesn’t have a real contact with the American reality and politics: is being Obama too aggresive towards business in his rhetoric, as FDR? Is the comparison between Obama and FDR valid, in this respect?


  16. amartimoro: relevance should be a concern also for Austrians.

    Mises was quite clear in Human Action: relevance cannot be judged apriori, we can say if an argument is logically correct, we can’t know if it is also relevant.

    That’s should be where data analysis becomes important, although hardly ever conclusive.

    I think the most important source of regime uncertainty is the deficit. With over a trillion dollar deficits, and still no convincing signs of recovery (i.e., others trillions of debt in the future) everybody must expect higher taxes.

    If these taxes are known in advance, it’s crowding out. People won’t invest.

    If they are “random” and unpredictable, it’s regime uncertainty.

    Both problems affect investments. I too think that crowding out – rectius, Ricardian equivalence – is more on the spot than regime uncertainty, both problems do exist.

    There are models of investment choice with uncertainty. When it comes to make irreversible choices – and fixed capital is an irreversible choice – an increase in uncertainty will lead to an increase in the option value of not committing to anyone. Making choices becomes uneconomical.

    It’s funny that this model was proposed by Ben Bernanke in 1983.

  17. @ Pietro:

    I always enjoy your comments here and at CP. So please forgive me for saying so, but it seemed your above comment to me was kind of all over the place and yet failed to address my rather humble point.

    Jerry’s right, I’m no commie as much as I enjoy pretending otherwise. I am indeed a classical liberal. So I share your fear of Leviathan and I agree with Hume’s maxim that when we are designing political institutions, we need to assume everyone is a knave. Absolutely.

    I don’t quite see how any of that lets me figure out how to balance the immediate suffering of unemployment against a more speedy adjustment in the labor market. You might argue that if we had a better approximation to the ideal, then you would not have much a problem about cyclical unemployment. Okay, but we do have a problem with cyclical unemployment and we have to figure out how to tradeoff unemployment relief vs. retarding market adjustment. How do you comments help us with that tradeoff?

    At the risk of being too wordy (moi?), I will make two more points. First, I think theory tells us to avoid state program meant to retool workers. The knowledge problem and all that. Just put ’em on the dole so they can find their own new path. Second, I don’t think the slippery slope is so slippery that you must inevitably land in the bowels of hell if you have a little social insurance or whatever. Consider your very case of Italy. I think we would probably have pretty similar opinions about the problems there, or at least regarding economic policy. But it’s not like Italians are living in some sort of gray Soviet disaster zone. It’s a pretty livable place.

  18. My post mixed Hayekian elements with Alchian reasoning on search. All resource owners are searching for profit maximizing prices. Sometimes they withhold an item from sale in anticipation of a better price. Labor is no different in that regard.

  19. Koppl:

    I probably didn’t get your point, but of course I wasn’t thinking you were a commie. I had the impression after writing that I said nothing of importance, anyway.

    When I talked about the slippery slope, I really was thinking about Italy, not the USSR: a declining country with a reinforcing feedback of social unrest and social interventions. Italy is probably the first country in the developed world which is experiencing a process in which sons are poorer than their parents.

    Maybe my political optimism is calibrated to Italian politics and US monetary policy, so I may overestimate the steepness of slippery slopes.

  20. The NYT article linked by Mario is, indeed, on point. It focuses on the hitech industry. Lots of other labor-market issues raised in the piece, too.

  21. Roger: “First, I think theory tells us to avoid state program meant to retool workers. The knowledge problem and all that. Just put ‘em on the dole so they can find their own new path.”

    I appreciate the knowledge-problem point; how can the planners know which fields people should be trained in? Subsidizing retraining could easily move people into the wrong professions. However, “putting ’em on the dole” also creates unintended consequences, including encouraging people to *stay* in the wrong professions. It’s not obvious to me which policy would result in the greater misallocation.

    Perhaps the best policy to balance humanity with efficiency (if that’s your goal) might be something like this: make the dole contingent upon enrolling in some kind of privately provided retraining program. Of course, even this would be subject to public-choice shenanigans. So don’t put me down as a definite advocate of this position. But I wonder if knowledge-problem insights could be applied to design a less damaging form of the dole.

  22. Glen,

    You’ve got my gestalt right: Okun’s leaky bucket.

    I just don’t see where any sort of contingent assistance program could be as nearly neutral as plain old unemployment relief. The contingency necessarily induces a direction to the reallocation process that was not market driven. You may get inertia with the dole, but I would have thought it is sort of clear that any such inertia will likely be less distorting than contingent programs. What am I missing?

  23. Jerry O’Driscoll writes: “I’d like to understand how the Germans came through this so well. No ideology, just facts.”

    Lawrence H. White has an excellent op ed in today’s Wall Street Journal, “The Wirtschaftswunder: Another Look” (Sept. 8, 2010, p. A19). While it doesn’t answer the O’Driscoll query beyond alluding to the salutary effect of recent German government spending cuts, it does usefully review the history of the German “Wirtschaftswunder” (economic miracle) of 1948-1963 and its inspiration frin the free-market ideas of Walter Eucken and Ordoliberalism, centered at the University of Freiburg. (Friedrich Hayek, be it noted, also taught there from 1962-1968). This made me curious as to what relationship might have existed between the work of the Ordoliberals and Wilhelm Roepke.

  24. Thanks to Richard Schulman for the heads up on Larry White’s op ed. I second his recommendation that all should read it.

  25. Roger: “I would have thought it is sort of clear that any such inertia will likely be less distorting than contingent programs.”

    I guess that’s just not clear to me. I take the baseline to be no intervention at all; that would generate a good deal of pain, but the least amount of distortion. In that situation, I presume a lot of people would try very hard to find some kind of work, even if it meant exiting their preferred profession. So if that’s the baseline, which policy will come closest to mimicking the process (while minimizing the pain involved) — one that encourages people to wait for jobs to arrive in their preferred professions, or one that encourages people to seek work in other fields if need be?

    If you’re making a public-choice argument that the retraining requirements might be gamed to such a degree that the added distortions overwhelm the reduced distortion from job-seeking, then I agree that’s possible.

  26. Hi Glen,

    I guess you are thinking that the unemployed have been working in the “wrong” types of jobs and need to move on if we are to have our required sectoral adjustments. The dole discourages people from moving on and may thus easily be just as distorting as contingent forms of relief. Is that it? I guess my reaction would be that the narrowing effect of the contingency would restrict search by the unemployed. If the state puts construction workers on the dole, they may be, in some sense, too slow to move on to their new industries. But they will eventually make many such moves. If, on the other hand, you have to train as a medical technician to get paid, then we will end up with lots of ex-construction-worker-med-techs who can’t get jobs. The odds are low that the state program would correctly spot the next big thing. The odds are high that the training or whatever required would be aimed at too small an array of options. Overall, the odds seem to me to favor relief programs without contingencies vs. contingencies such as “jobs programs.” Do you think those comments address our different intutions?

    Even if I am right, such programs would likely slow market adjustment. I’m fine with that, because the gain is that we cushion the adjustment process for those thrown out of work by the downturn. Now we’re back my earlier exchange with Jerry. Neither of us had an algorithm for calculating that tradeoff.

  27. Roger — yes, you have my model correct. I am concerned about people already in the wrong fields being encouraged (by unemployment benefits) to stay there.

    As soon as the government has to choose which specific fields to target for retraining, I completely agree that the knowledge/incentive problems would be insurmountable. What I was suggesting (without actually advocating) was a more open-ended requirement, e.g., to get extended unemployment benefits you must be engaged in some form of training program, with only minimal restrictions on what field the training is in. The devil is in the details, of course.

  28. Glen,

    Ah! Okay, I see that. Yeah, I don’t see how a “knowledge argument” would really shoot that one down. As you said already, there’s the public choice angle, but details matter as you also said.

  29. Wait, maybe I want to flip back to my original skepticism. I said earlier “The odds are high that the training or whatever required would be aimed at too small an array of options.” I think I’m gonna stand by that and say it’s a reason to reject even the sort of open-ended requirement you have been considering. (You never really proposed it.) We can, perhaps, imagine something a bit better than just unemployment checks, but our theory should caution us that the actually existing program we would likely get would quickly move away from that point even if we could get there for a minute or two. Mario has written some good pieces on slippery slopes with some guy in California and I think that work my apply to this question as well.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s