CEPR - Center for Economic and Policy Research


En Español

Em Português

Other Languages

Home Publications Blogs Beat the Press Doctors Remove Bullet From Victim's Head: Seek to Determine Cause of Death, Dylan Matthews Edition

Doctors Remove Bullet From Victim's Head: Seek to Determine Cause of Death, Dylan Matthews Edition

Wednesday, 05 September 2012 19:40

Dylan Matthews usually writes insightful pieces, but he really strays from the mark today in buying a story that unemployment in today's economy is to any significant extent an issue of skills. He picks up on a new paper by two economists (Nir Jaimovich from Duke and Henry E. Siu at the University of British Columbia) that argues that structural changes rather than weak demand is the main factor behind weak job growth in the last three recoveries.

Tying structural change to the last three recoveries should immediately raise some hackles. The largest shift upward redistribution in wages from the middle and bottom to the top, occurred in the early 80s. In the folklore of skills biased technical change that dominates the economics profession, the early 80s really hold pride of place as the period that best fits the story. But Jaimovich and Siu want to put the early 80s back in the post-war golden age where we all prospered together. If they are right, then a lot of other economists are seriously wrong.

But getting to the substance, the real issue here is if there was a ton of new demand in the economy -- the government had a huge make work program, the dollar plunged and net exports soared, and job creators doubled their investment inspired by the Romney-Ryan vision for America (just kidding) -- whether or not we would not see a huge surge in employment. It's hard to see anything in Jaimovich and Siu's paper that would suggest otherwise.

We know that less educated workers bear the brunt of unemployment any time the economy falls below full employment. This was the theme of my book with Jared Bernstein a decade ago (The Benefits of Full Employment). There's a simple logic to this. The marginal workers that firms hire when demand picks up tend to be less educated on average than the core workers who stay employed no matter what.

Think of a factory where output has been cut by 50 percent or a store where sales fall by the same amount. The managers likely stay on in their jobs, as well as the most skilled maintenance crew in the case of the factory. The people who get laid off will be the line workers in the case of the factory or the retail clerks in the case of the store. However this change in skills mix is not an indication of a change in technology, it's an indication of weak demand.

If we actually had a serious issue with technology changing the mix of skills needed in the economy then we should be able to find large sectors of the economy where there are large numbers of job openings, where wages are rising rapidly and the average weekly hours are surging. We don't see this. (Yes, employers complain about not being able to find qualified workers, but if this is true then our job creators have forgotten how to raise wages.)

The real story of this downturn seems very simple. We had a huge housing bubble that was driving the economy through its effect on consumption and residential construction. When it collapsed, it created a gap in demand on the order of $1.2-$1.5 trillion. There is no easy way for private sector demand to fill this gap. (Anyone got a story of how that would work?)

This means that we have to rely on government deficits to fill the demand gap in the short-term. In the longer term we should be looking to reducing the trade deficit. Residential construction should also help as it gradually returns to its trend levels from the extraordinarily depressed levels that resulted from the overbuilding of the bubble years.

Anyhow, the housing bubble story provides a simple answer as to why we are not creating jobs. I suppose we can keep a lot of economists employed looking for other explanations, but aren't there more productive things they would be doing?

Comments (4)Add Comment
written by JSeydl, September 06, 2012 8:38
Dylan Matthews clearly has not seen Ed Lazear's new paper, which argues against the skill-mismatch claim. Lazear obviously has made many other silly claims, but at least he's starting to recognize the demand slump.
What Exactly Are the Structuralists Saying?
written by Last Mover, September 06, 2012 9:16
The largest shift upward redistribution in wages from the middle and bottom to the top, occurred in the early 80s.

The marginal workers that firms hire when demand picks up tend to be less educated on average than the core workers who stay employed no matter what.

This implies if unemployment was structural then core and marginal workers could be affected equally over time as turnover of skills would occur on the demand and supply side.

If the housing bust created structural, not cyclical unemployment, that implies aggregate demand never declined below full employment levels that existed before the bust. Instead there was more demand for different skills in conjunction with less demand for others.

If structuralists claim the unemployment existed before the housing bust then it somehow existed as part of cyclical full employment without causing wage inflation.

As for the structural unemployment and correction in the early '80s that resulted in upwards wage redistribution,
it didn't occur during full employment either given Volcker's policy of high interest rates to grind down high inflation concurrent with less cyclical employment.

It seems even where structural unemployment is legitimate, most of the employed workers no longer needed (especially core workers) are let go opportunistically during a cyclical demand slump, then workers with different skills are hired in the subsequent recovery.

Then again, look what the car industry did in this recovery using two tier wages that gave new workers lower income than existing ones, clearly not a structural difference, rather a cyclical opportunity to lock in lower wages and benefits for the long term.
written by skeptonomist, September 06, 2012 9:20
Actually the growth of inequality began in the 60's and the crash of real wages at the lower end of the income scale was in the 70's:


By the time Reagan administration policies began to take effect growth of real wages had already come to an end. Evidently people who think what happened in the 80's was critical are just not looking at the data for before then.

What caused the crash of real wages? It wasn't China and it wasn't Reagan or Bush. Think about it.
written by AlanInAZ, September 06, 2012 10:21
I think Dean Baker and others similar views on structural employment are too dismissive of the fact that all of the employment growth over the last 2 decades has gone to people with more than a high school education. Since 1989 employment for workers with HS or less has declined 14% whereas employment for workers with college degrees has increased 82%. This trend has been in place during the bubble years as well as during the recession periods. I think one important element is that educated workers have a broader skill base and are more apt to be hired even if their skills do not exactly match the job since employers can provide job specific training. Increasing economic activity through whatever means will almost always preferentially advantage the more skilled and educated.

Write comment

(Only one link allowed per comment)

This content has been locked. You can no longer post any comments.


Support this blog, donate
Combined Federal Campaign #79613

About Beat the Press

Dean Baker is co-director of the Center for Economic and Policy Research in Washington, D.C. He is the author of several books, his latest being The End of Loser Liberalism: Making Markets Progressive. Read more about Dean.