November 2017

Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30    
Blog powered by Typepad

« Austrian Economics Daffynitions | Main | Policy is the Problem »


Feed You can follow this conversation by subscribing to the comment feed for this post.

I'm inclined to agree, but what evidence is there of greater labor specificity today? I only ask because you also hear about the growing importance of broader cognitive skills, non-specific human capital, and reduced job tenure (which of course is associated with non-specific human capital). Jobs may be more specialized today - I could believe that. But I don't know if workers are less homogenous today.

I imagine it largely depends on what industry or region the worker has been in.

This sort of argument is certainly not new, even before this jobless recovery:

You allude to the prospect of some empirical analysis of all this. Do you know of any empirical work on Austrian economics that has been done specifically with these heterogeneity concerns in mind? I found a few that just (ironically) looked at aggregates but nothing specifically on capital heterogeneity.

There's an idea for another paper here that you and I might want to consider: labor specificity being ignored in the Post Keynesian call for the government to serve as Employer of Last Resort.

Note in the link that there are quite standard AD/Keynesian explanations as well that also don't in any way conflict with this labor specificity argument, contrary to Cowen's assertion that the nature of the jobless recovery gives reason to discount an AD story (granted, many people pointed this out to Cowen a couple weeks ago when he made the same assertion). This isn't really a test case to arbitrate between these approaches. In all likelihood, both processes are occuring. Which makes the empirical work untangling the two that much more important.

Earlier this week, the Wall Street Journal devoted a substantial part of its main section to the employment issue in all its complexity. One striking fact is how many firms are trying to expand and can't find workers with the skills they need (specificity).

One firm needs machinists and decided to train candidates themselves. I think 16 out of 24 candidates completed training. A couple older ones dropped out because they didn't want to invest in learning new skills at their age. (A separate story chronicles how many older workers are dropping out of the labor force and taking early social security.)

There are excess supplies of some labor and excess demands for others. The story fits Steve and Roger's narrative quite well.

A spearate but related story I've heard from employers. It is that many college graduates are unemployable when they graduate. They have no useful skills. One employer remarked it takes a year of training and he doesn't have the time. They seek out the schools that do a good job and only hire form them.

Jerry - there's a big push for career and technical education and apprenticeships to fill precisely this gap, as well as greater reliance on community colleges. We have something of a bimodal skills distribution in this country, and it's those middle technical skills that we're short on.

I think you're more likely to see these issues in specific industries, though - particularly advanced manufacturing and technician type jobs that do require asset-specific training. Generally speaking, the problem we're facing doesn't seem to be on the labor supply side, does it?

This is very good Steve.
Lachmann argued that the more complex the economy the vulnerable it is to macro-disturbances - like a house with many specialized servants, what happens when the cook gets sick?; versus one with a few jacks-of-all trades, hence redundancy.
But the hypothesis that increasing heterogeneity-complexity has lead to a higher amplitude in business fluctuations depends (as I have argued) not only on the extent of specialization (heterogeneity) of the productive structure; it depends also on the capacity of the economy to adjust to change. And I believe the latter has increased significantly (owing to improved information processing, education and training opportunities, increased mobility of human capital, labor-saving aids to learning and doing, etc.). In fact, it may be more accurate to view the extent of specialization-heterogeneity as endogenous dependent to some large degree of the capacity of the economy to adjust to circumstance. After all, the incentive to acquire specialized human capital depends on the risks involved.
What does the data say? Does it seems that the amplitude of fluctuations has increased. My casual impression is that the opposite is true.
That being said, fluctuations are, in essence, not really macro-phenomena. They are aggregations of micro-events and I do believe that more specialized human-capital experiences greater gains and losses - hence the higher returns.

One comment on Peter's point: I wasn't arguing that fluctuations have greater amplitude per se. I was suggesting that the response of an economy to the same inflationary stimulus will be greater, hence the recovery will be longer. I'm not suggesting the inflations themselves have been larger. In other words, for a GIVEN rate of inflation, we should expect larger booms and busts, but observed cyclical variation is due to inflations of different amounts, so just looking at those variations doesn't address my argument, which would require an experiment we could never implement.


According to the NYTimes story, unemployment is concentrated among the least educated workers. Are you saying that they have the more firm and industry specific human capital and are thus most at risk of relatively lengthy bouts of unemployment?

Two reactions:

1. To your question - "What exactly government is supposed to do to "directly employ" unemployed folks with these highly specific skills? "

What if government took the stimulus $ and invited the unemployed into jobs centers where they would TELL government what their previous jobs and experience had been (fill out a form provide a resume), then take unemployment benefits for a few weeks until an appropriate job was available. Government would collect forms and create jobs to order, essentially.

Although this is easiest to do directly when taking housing construction workers and using them for school renovation or road building, it would also be possible to take finance workers and get them jobs by creating a new macroeconomic forecasting arm of the CBO or something.

2. "it may well be the case that an economy with a greater division of labor and specialization, and therefore more heterogeneous human and physical capital will actually suffer larger booms and longer and deeper busts "

This has a distinct flavor... now what is it? ... Oh yeah, Marx.

Hi Steve, good article. Would you say that the threat of discoordination (sustained idleness) is higher in a complex economy – assume here a free market characterized by monetary equilibrium - than a simple one (also free and in monetary equilibrium) because the process of recalculation (given some real shock) is more protracted in the former?

@Roger: there's two facts at work here. One is that the least educated are most likely to be unemployed. I suspect those are folks in construction and manufacturing type jobs. The former took the hit from housing and the other from the resulting downturn across the board.

The other is that *among the unemployed* older and more educated workers are most likely to be unemployed for the longest.

It may well be that the least educated will have the most difficulty shifting to those areas of the economy where the jobs are still around: services and technology. The housing boom made jobs requiring relatively less formal education look more common and remunerative than was justified by long-run fundamentals. It's a mismatch of human capital with the underlying variables. The manufacturing problem has been ongoing. These folks have fairly general human capital in a world where more specific human capital is now required.

At the other end, there were the well-educated folks who got caught in the undertow, perhaps in the financial sector. They might well have somewhat overly specific HC and are finding it difficult to adjust to where the jobs are now.

As others have suggested, more detail on the empirics of the unemployed would be really helpful here. If I were writing a real article, that's where I'd be going. The blog post was meant to be suggestive.

@JP: I think that's what I'm saying, yes. The more complex the economy, the more scope there is for mischief in malinvestment in human and physical capital due to increased specificity (Peter L's house example) and therefore the more protracted is the adjustment after the boom turns to bust.

@Steve: Given your clarification I understand you to say that the hypothesis is - given the size of the (credit-induced, mal-investment-producing) boom, the larger will be the bust, the more complexly specialized the economy. Worth considering and investigating.

Yup, that's it Peter.

Some Points:

(1) Labor specificty should not be equated with technical skills. On-the-job experience makes every worker have a higher value in that specific job in that specific firm.

(2) Unskilled and skilled workers may be complementary. How many unskilled or semi-skilled workers lose their job for every loss of a skilled workers' job? (Construction is a good example.) By numbers, the job losses may be concentrated among unskilled labor. But those losses are a function of the loss of capital values and skilled workers.

(3) We need more micro data.

The problem I have with this explanation is that employment has jumped in the UK recently. That may be a statistical anomaly certainly.

My view though is that welfare and employment costs are more important. There have been no rises in unemployment benefit in the UK or Ireland. Rates in Ireland have been cut slightly from a high starting point.

Thanks Steve.

I may be repeating what Peter Lewin said earlier but if a complex free market economy has developed (ie. one with greater division of labor and therefore more heterogeneous human capital), then somewhere along the way private institutions will have emerged to cope with this complexity, specifically those specializing in the mechanisms for reallocating capital and speeding up recalculation. Here I am thinking head hunters, internet job sites, work training through private schools, private unions providing temporary funds, NGOs specializing in unemployment etc.

Perhaps a complex free market will be better suited to deal with real shocks than a complex mixed economy because the above institutions (those specialized in aiding recalculation) operate in a competitive environment and therefore have incentives to do their job well, whereas in a mixed economy many of those institutions are government-run or -sanctioned monopolies without the same incentives.

1. The CCC was in part a skills-training program for young men.

2. Of course many more people were farmers during the GD. Farming requires skills in everything from finance to ditch-digging, so perhaps people really were more willing and able to adjust to changing demand for labor.

3. Training and education take longer and cost more. To move from web designer to biologist, for example, would take ~10 years and hundreds of thousands of dollars. To learn machining at my local community college takes two years. Who can predict conditions then?


You had both those points in your post and yet I wasn't quite glomming onto the one. Thanks for the repetition; I'm afraid I needed it!

Overall, I confess, I don't see a clear and clean story here. My apologies if I"m just being obtuse. It makes sense that older people would take longer to get re-situated. But I don't see why the same would be true of the more educated unless it is because they are richer and can afford a longer search. Age and education are both correlated to wealth, so it might just be that you look longer when you're rich.

In some sense complexity is a prerequisite for a trade cycle. Robinson Crusoe cannot suffer cyclical unemployment. But it kind of goes against the grain to say that growth makes the trade cycle worse. Facts are facts and I'll take it if that what the evidence says. But it violates my intuition to think "the market" would grow worse at self repair the richer a society is. If I were working on the topic I might be more inclined to look at the net equity of the unemployed, transfer payments, number of two-earner households, and so on.

I think that there is something to this basic story over time in the US. However, there are problems when we compare internationally. In particular, Germany had as sharp a decline in GDP as the US did, but had barely any decline in employment, thus tending to keep workers in those jobs for which they are well trained by experience. They had very different approaches to dealing with that than we did. Maybe theirs will hurt them in the long run (and they have had higher unemployment rates than the US in recent decades), but they have certainly done far better on the employment front than has the US.

Some points, and I apologize if these have already been brought up (I will scour the comments once I return from work):

1. Structural unemployment issues regarding the ability of government to employ the unemployed seems a little simplistic to me. Even assuming that most of the unemployed were employed in a type of white-collar labor, this doesn't mean that they lose their ability to clean streets, dig holes, or operate machinery.

Furthermore, even if it was true, this would only provide greater incentive to the government. Indeed, think of all the stimulus it could do by simply training these unemployed! This would, in effect, be no different from the idea of subsidizing education.

In fact, I seriously doubt that the government, so far, has put much effort into directly employing the unemployed (apart from programs such as the census). There is also the question of whether or not government jobs are desirable for the unemployed at this point. An ex-teacher may not be willing to turn into a highway worker, just yet (the situation is not that dire, especially with extensive social safety nets in place).

2. I would say that the labor market is having trouble recovering because of wage disequilibrium and stagnant productivity. Again, I also question to what extent the government has directly employed the unemployed (or tried to).

3. It's true that in a bust in a more specialized labor market it may be more complex to employ the unemployed. Nevertheless, in a more extensive division of labor the level of entrepreneurship is also more complex. The possibilities of investors are much wider. Furthermore, I argue that (even if at a lower wage) in a growing economy employers are willing to train future employees. The problem, again, is a lack of productivity.

4. I think it's worth looking at the labor markets in two cases. First, the great reallocation of labor from agriculture to manufacturing during the so-called Long Depression (1873 to 1894). Second, the post-war years, where according to Higgs around 40% of the population was essentially unemployed until 1946 (these include those in the armed services and those working for the military indirectly, including munitions manufacturers - these skills were not directly transferable to the post-war private sector).

The evidence seems to support the thesis that the problem is a lack of productivity, not any inflexibility in the structure of the market.

The structural complexity of employment is important, no doubt.

The other issue the article quoted by Prof. O'Driscoll brought up is high indebtedness, especially connected to housing, and how that prevents people from moving for work.

So, if jobs with your skills are far away geographically, it is a lengthier process to match up work and people since the avenue of moving for work is closed.

So, if I understand, from the employers point-of-view labor is just capital (organizational capital in one formulation) which suggest a certain 'liquidity', but from the employee's point-of-view specificity of skills makes recalculation hard. In other words, when saying that 'labor is capital' one might have to switch between 'capital' as 'liquid asset' and 'capital' as 'specific machinery or technology not easily reconfigured'.

You might want to check out this Beveridge Curve, which shows that jobs are in fact being created, but the unemployment rates remains at ~9.5%, suggesting a structural change in the economy. If this level of unemployment remains, what we are in fact looking at is the emergence of the economy (as complex adaptive system) into a new level of complexity. This new level isn't necessarily more complex, however. This kind of structural change is what one would expect in a complex adaptive system if simplifying rules are added to it: the system would simplify until it reached a critical point, where it leaped into a less complex level. This recession may have been precisely that discontiunous leap. I hope that's not the case, but I fear it may be. If it's not this one, I predict it will nevertheless come soon enough.

Didn't like the way I tried to make the link for the Beveridge Curve:

Barkley makes a good point. Notwithstanding one's view of government's role in these matters, the US approach to jobs is counter-productive. It alters the calculation of an unemployed worker in favor of delaying taking a job.

The German approach is to the alter the firm's calculation in favor of retaining workers. An alternative approach was articulated by Bob Litan years ago, and it tilts an unemployed worker's decision in favor of accepting a job earlier rather than later.

No jobs policy can avoid unintended consequences. On the surface, however, both alternatives make more sense to me than the current US policy.

Again, I leave open the issue of the federal government's proper role.

i thought this was funny. Natalie Dee's comic shows what happens between egg friends when one them becomes a chicken. As Serious Eats put it.

Glad I drop by! This is a great site and contains fantastic contents. Thanks for sharing this to us.

The comments to this entry are closed.

Our Books