Arnold Kling  

Jobless Recoveries and Structural Change

PRINT
Murphy on Global Warming Polic... Robert McNamara ...

The Economist's blogger (the magazine has a policy against signed articles, and they have stupidly decided that the same policy has to apply to their blog) writes,


there is every indication that recovery from this recession, when it arrives, will do little for many of the unemployed for years to come. Time to start thinking about why this is the case, and what can be done about it.

Thanks to Mark Thoma for the pointer.

I want to emphasize once again that the structure of the economy has changed over the last fifty years. We have gone from over two-thirds of the labor force having no more than a high-school education to over two-thirds having at least some college education.

Up through the early 1980's, many of the unemployed in a recession were low-skilled workers on temporary layoff, in many cases concentrated among large industrial firms. Once demand recovered, they were called back--end of story.

The Dotcom recession was different. The people who were laid off by Webvan and Pets.com and Worldcom were not going to be called back. The companies were not viable. From late 2000 onward, a lot of the labor force simply disappeared. There was a recovery in productivity, as firms slashed marginal workers. The unemployment rate did not get terribly high, in part because fewer people looked for work. The employment-to-population ratio never recovered. Maybe it's a Tyler Cowen story--leisure got cheaper because of the Internet, and so people consumed more of it. (Just kidding)

My thesis is that unemployment is more persistent when the layoffs come from structural change rather than from excess inventories. With excess inventories, once the excess has been absorbed you can go back to work at your exact same job. On the other hand, when firms and industries permanently shrink, you have to find a new job, and possibly even an entirely new occupation. It is rare for people to have the capacity to do that, and it takes quite a bit of time when they do.

Part of the problem is credentialism. The demand for health care workers is high, but if state regulations say that you need a doctorate to be a physical therapist (a real example, in my state of Maryland), then shifting into health care is not going to be easy.

Most of the structural adjustments in the labor force consist of a cohort of young workers joining and a cohort of older workers leaving. This changes the mix of skills and occupations.

No economist knows how to deal with structural unemployment on the scale that we are seeing nowadays. Fiscal stimulus is not a proven remedy, and it may not be a remedy at all. Worker retraining programs are a logical idea and an empirical failure.


Comments and Sharing


CATEGORIES:



COMMENTS (13 to date)
fundamentalist writes:

Along similar lines, Hayek wrote that most jobs today require a large amount of capital. Very few jobs, except entry level ones, can be done without a lot of capital equipment mixed with the labor. So if the capital doesn't exist, then the jobs can't exist either. It's the curse of modernization.

Alex J. writes:

In the late 90s, Wake County NC had an unemployment rate of 2%. If, as seems reasonable, employment was in a bubble, it seems unlikely that we will get to the same employment-to-population level without also being in a bubble at that point too.

If the structural changes you mention are to blame for persistent unemployment, then we should be looking to lowering the cost of retraining by raising the productivity of education The trend seems to be just the opposite: spending more and more on education for less and less benefit. Contrast language learning DVDs with single occupant dorm rooms.

TomB writes:

I have many friends who worked for dot com companies right before and after the bust. I also have a number of friends who currently work in finance.

In a booming industry, salaries are higher and money flows. Many people in boom industries are able to accumulate financial resources. When a bust follows, the people most affected by the collapse are those that are in the best position to weather the storm.

One factor contributing to the early 2000s jobless recovery may be that many people in the dot com industry had an entrepreneurial mindset. Many people worked for start-ups before the bust, and start-up companies are an inherently risky venture. Start-up employees are more likely to have periods of unemployment between ventures. After the bust there were fewer start-ups so people spent longer periods of time between jobs. Many people also made large sums of money by selling their companies to Microsoft, Yahoo, or whoever. With large surpluses, they did not need to worry about having an income. Thus people could pick and choose what work they did during the jobless recovery.

With finance, much the same might happen. Many people in the finance industry have savings in the 100s of thousands or millions of dollars. Not only did they have big salaries, but they knew how to make money investing. Many laid off finance workers do not need to work. They can wait until they find an opportunity that suits them or they can live off investment income. Examples: One of my friends bought an apartment in Paris and is now going to cooking school there. Another plans on coming to DC to work in politics.

Granite26 writes:

Do you think it would be better to stop reporting an unemployment rate and rather report an average length of unemployment? Saying that 3% or 11% or 1 in 10 people don't have jobs is very different than saying the mean length of unemployment is currently 3 months.

It's not as scary when it shouldn't be, and it is significantly more meaningful. The REAL mean length would be slow to respond, but using an average employment length, you could turn the unemployment rate into the length.

Just a thought

El Presidente writes:

Interesting observations.

What do we do with extra people? The British refer to layoffs as eliminating "redundancies". Are people redundant or is our effort to utilize and value each other simply insufficient?

Phillip Huggan writes:

Care to post links or name google key words of empirically failed worker retraining programmes, or any analysis of such? If so, thx.

TomB writes:

Here is some anecdotal evidence of failed worker retraining programs:

http://meganmcardle.theatlantic.com/archives/2009/07/retraining_isnt_the_answer_1.php

She also links to this article, which provides some links to empirical evidence:

http://www.nytimes.com/2009/07/06/us/06retrain.html?_r=1&hp


I think that length of unemployment is more meaningful from a personal standpoint - I want to know how long I will have to wait to get a job. I do not know if it is more meaningful that total unemployment - which essentially measures the percentage of labor resources currently idle.

Lance writes:

Unfortunately, conventional output gap estimators have not taken into account the possibility that potential output has fallen (which may occur due to a structural change recession), which produces these scary charts where we have a output gap of -6%.

We'd probably still have an output gap in excess of -2% even if output gap models were more able to account for structural vs. cyclical/crisis induced recessions. But it would certainly take the wind out of the sails of those agitating for a second stimulus premised on an output gap larger than was imagined.

El Presidente writes:

Lance,

Interesting approach.

Monte writes:

No economist knows how to deal with structural unemployment on the scale that we are seeing nowadays. Fiscal stimulus is not a proven remedy, and it may not be a remedy at all. Worker retraining programs are a logical idea and an empirical failure.

The hopelessness of these concluding remarks implies further intervention may be warranted. Can we assume a hysterisis effect inconsistent with theories in which the natural rate of unemployment is independent of aggregate demand? If so, what remains, other than to attack the problem directly with a series of labor market reforms reinforced by some level of demand management?

Aren't there any libertarians out there who are even cautiously optimistic about our long-term job prospects without the assistance of a visible hand?


SC writes:

Most of the structural adjustments in the labor force consist of a cohort of young workers joining and a cohort of older workers leaving. This changes the mix of skills and occupations.

Are the older cohorts leaving? It seems to me that the trend has been to work longer and later in life, and now more recently with the value of 401K and pensions lost in the downturn, more older workers are putting off retirement or coming out of retirement to work again. Seems like there are fewer openings for younger workers.

Phillip Huggan writes:

I'll read what few co-op University studies are available and retraining case studies. The meganmcardle thread respondants suggest employers want job experience. *If* this is a hurdle, would employers recognize participation in co-op programmes (1/2 schooling 1/2 on jobsite) as job experience?

Sandwichman writes:

No economist knows how to deal with structural unemployment on the scale that we are seeing nowadays.

Hold on a minute there, Arnold! Dean Baker is an economist and he knows. John Maynard Keynes was an economist and he knew. What would Keynes do? Surprisingly, not what today's Keynesians insist.

The "jobless recovery" is an oxymoron. It also stands the hoary lump-of-labor fallacy claim on its head. Instead of a "fixed amount of work to be done" there is a growing amount of work to be done by a fixed number of workers. And it's not an assumption -- it's a fact.

Well, the remedy is obvious -- or would be obvious but for the precision-engineered blinkers worn by almost all economists these days: "So long as there is one man who seeks unemployment and cannot find it, the hours of work are too long." Who said that? Why the great Keynesian himself, John Maynard!

Wait! No! That was Samuel Gompers. What Keynes said was:

"The full employment policy by means of investment is only one particular application of an intellectual theorem. You can produce the result as well by consuming more or working less. Personally I regard the investment policy as first aid... Less work is the ultimate solution."

So, the first Obama stimulus was first aid. Let's assume the patient is stabilized. The economy is no longer "falling off a cliff". So, what would Keynes do?

Jobless or work less?

Comments for this entry have been closed
Return to top