The Challenge of Long Term Job Growth: Two Big Hints

June 5th, 2011 at 11:26 pm

Busy day tomorrow, so unlikely to post much.  Starting the day out with some early squawking—guest-hosting on CNBCs Squawk Box, 7-9am.

Also giving a talk later in the day on longer term concerns re job growth, as in how can we generate enough jobs, not just in the cyclical sense, like now, but over the next few decades?

I’ll have more to say on this when I’ve got some time, but these two charts have influenced my thinking a lot.

The first one shows the long term trends in productivity and employment growth.  People sometimes worry that we’re getting too productive, able to satisfy the demands of our economy with “too few” workers.  That’s an age-old worry, and those who want to downplay it cite the fact that, as the graph shows, there is a positive, not a negative, correlation between productivity and job growth overtime.

But look at the end of the graph.  Productivity accelerates while employment growth decelerates.  And that ain’t no blip either…it suggests the possibility of a structural change in this relationship.

As usual, lots of other stuff was going on there too—financialization, for one: the ramp-up of the financial market craziness that helped inflate the housing bubble.  Increasing inequality, bad tax policy…so a lot to sort out here.  But it’s a big hint as to how to start thinking about the problem of not enough job creation.

Here’s the second hint.  I basically copied this from a neat graph in the Washington Post; it’s just all the jobs sectors—construction, manufacturing, services, etc.—indexed to 100 in January 2007 and plotted through last month.

I didn’t bother labeling every sector because I just wanted to make one simple point.  Look at health care/education (which is driven by health care).  Even throughout the worst recession since the Great Depression, with payrolls tanking worse than I’d ever seen, HEALTH CARE ADDED JOBS EVERY SINGLE MONTH.

The graph makes the point…um…graphically.  It’s a straight line sloping steadily up amidst all that carnage!

So there’s the other hint.  This is a sector with great demand, tough to outsource, strong gov’t involvement (and yes, unsustainable spending—and that’s not solely a public sector issue; it’s just as bad on the private side), and limits to productivity growth, relative to say, widget production.

I’ll explain the hypothesis this leaves me with in later posts, but for now it’s got me wondering: is Baumol’s disease part of the cure?

(I apologize if this sounds cryptic–just saying that the fact that service sector productivity growth naturally lags that of manufacturing may be more of a friend than an enemy.  Like I said, more to come on this.)

Print Friendly, PDF & Email

21 comments in reply to "The Challenge of Long Term Job Growth: Two Big Hints"

  1. John says:

    “Technology devalues labor”…

    I was commenting to that effect to your posts on TPM Cafe’, about the time those two lines began to diverge.

    I have both of Google’s reference model smartphones. Both have 1GHz processors and GB’s of storage. The world’s fastest supercomputer of 20 years ago was slower and had less memory.

    Apple projects iPads sales of ~40 million units just for 2011. The iPad 2 has two CPU cores running at 1GHz; 4-core smartphones will be common by this time next year.

    And of course, the Internet is now ubiquitous. That’s happened in the past 20 years.

    Things have changed. Dramatically.

    What’s the basic, fundamental purpose of robotics? Simple answer: to replace human labor.

    In terms of “creative destruction,” information technology is not like light bulbs replacing candles, or cars replacing horses. Those kinds of things made human labor more productive.

    The factor of production that information technology is replacing – and is intended to replace – is human labor itself.

    The following is not the future – it’s present-day Japan:

    I tell folks that automation is just the last stage of technology devaluing labor – earlier stages are incremental. That’s been happening for a while now. To see it a bit more clearly, compare productivity and wages over the past 30 years.

  2. Virgil Bierschwale says:

    we’ve all seen that one coming.
    problem is health care jobs pay substantially less than manufacturing or information technology jobs so we’re not comparing apples to apples via gdp, tax revenue, etc..

  3. D Furlano says:

    I was thinking about the productivity/employment issue the other day and I have an inclination it has to do more with consolidation. I have yet to find any data that would indicate this.

    Don’t know if you can link images but her is the URL to the NA M&A activity 1985-2010

    More can be found on the stat page.


  4. Neildsmith says:

    I guess this means that controlling health care costs will wreck the last engine of GDP growth in our economy. Ironic.

    Let’s pretend for a moment that we do all the infrastructure projects we can to put the able-bodied unemployed back to work. With a mountain of debt, they will take a few years to pay it off. But isn’t this job program just another bubble? We’ve progressed past the time when labor-intensive work will stimulate the economy. The building materials will come from China (remember that smelly dry wall?) and to the extent the newly employed buy things, they will also be largely made overseas. Then what happens when all the bridges are repaired? It doesn’t solve the problem, it just delays it.

    • Barry says:

      “With a mountain of debt, they will take a few years to pay it off. But isn’t this job program just another bubble?”

      Please note that the Federal government can borrow long-term at ~3%. Paying that off is trivial.

      • Neildsmith says:

        I was referring to household, not government, debt. Once the newly employed get a job, they have to pay off their own debts.

  5. Phil says:

    Does the health/ed line include the growth in private insurance companies as well? If so, that would be bad.

  6. Lee A. Arnold says:

    Is Baumol’s disease part of the cure? Yes! Let’s treat the healthcare sector as a growth industry, and stop trying to avoid the obvious.

  7. Sandwichman says:

    There is a third, very important, hint about the challenge and this is clearly illustrated in table one of “The compensation-productivity gap: a visual essay” on page 3 of the BLS article.

    Although the productivity/employment gap seems to only begin in the late 1990s, the productivity/compensation gap opened up in the 1979 to 1990 period.

    • Sandwichman says:

      Actually, the gap begins to open up soon after the 1969-70 recession. On August 15, 1971, President Nixon imposed wage and price controls under the authority of the Economic Stabilization Act of 1970, passed by a Democrat-controlled Congress. Donald Rumsfeld was the chairman of Nixon’s Cost of Living Council. Dick Cheney was his deputy.

    • Sandwichman says:

      If wage controls worked (in, say, tipping the balance between wages and rents) it shouldn’t be a surprise to see a gap open up between productivity and compensation and if wages stagnated it shouldn’t be a surprise to see families supplying more hours to the labor force, as analyzed in a sharp little paper from a few years ago, “Running Faster to Stay in Place,” by some guy named Jared Bernstein and Karen Kornbluh.

  8. Ken Houghton says:

    Sandwichman got it. Baumol abides, but enabling excessive rents to be taken from capital is self-perpetuating and ultimately deleterious to growth, unless you’re delusion enough to believe in “benevolent autocrats,” as Bill Easterly doesn’t but your ex-boss still does.

  9. Artie Gold says:

    Now, wait a minute — that line looks suspiciously like the punditocracy….

  10. John says:

    1981: IBM PC and MS-DOS.
    1983: Microsoft Windows.
    1984: Apple Macintosh.

    2007: iPhone.

    Mene. Mene. Tekel. Uparsin.

    • John says:

      How could I forget?

      1990: Tim Berners-Lee prototypes the _World Wide Web_.
      1993: Mosaic browser.
      1998: WWW substantially commercialized.

      1999-2001: Dot-com bubble and bust.

      2008: “Great Recession.”

      How big are the major players in this history?


      Lack of demand is a problem. But as I used to try to tell you, it’s not cyclic.

      Business profits have so far tracked with productivity. Wages and unemployment have not, not since the advent of the personal computer, and increasingly less so since the commercialization of the Internet.

  11. Fed Up says:

    “But look at the end of the graph. Productivity accelerates while employment growth decelerates. And that ain’t no blip either…it suggests the possibility of a structural change in this relationship.”

    The economy has gone from mostly real supply constrained to real demand constrained?

    And, “But it’s a big hint as to how to start thinking about the problem of not enough job creation.”

    Nope. Not enough retirement creation.

  12. robert says:

    See this post, which expands on what Jared wrote:

    Is the Historical Relatonship between Productivity and Employment Breaking Down?

    Labor-saving technology is pushing us toward a tipping point. Historically, technolgical progress has resulted in increased productivity, increased employment, and increasing wages. That relationship is breaking down. Wages for average workers have been stagnant for DECADES–even as productivity continues to increase. The gains from rising productivity are all going to the TOP 1% of the income distribution. Now Jared shows us that the relationship between employment and productivity is also breaking down.

    This WILL GET WORSE. Technology is accelerating. More and more workers at all levels will be displayed.

    See this book, “The Lights in the Tunnel” for a detailed look at this problem:


  13. D Furlano says:

    To me there are three major contributors;

    1) outsourcing to foreign countries

    2) Technology

    3) M&A and consolidation

    I think M&A and consolidation is overlooked and least understood.

  14. BT says:

    What happened?

    The divergence in productivity and employment co-incides with the start of the housing bubble, the biggest credit bubble in human history. The US economy was running on an asset bubble instead of real investment. Jobs were exported to Asia and products were imported, leaving corporate profits intact but opening up a large current account deficit and fewer productive jobs.

    Now the housing bubble is deflating, the USA has become Japan – except worse because the US is not a net exporter, so unemployment is very high.

  15. Lawrence J. Kramer says:

    This is my first comment here – stumbled by on a link from Seeking Alpha.

    Consider the possibility that employment, especially current employment, has been made obsolete by automation as the technology of choice for distributing goods and services. We need to use some other basis. (Note that manna from heaven was distributed per capita without regard to wealth. There is some wisdom there.)

    Globalization may be a distraction, an anomaly that results from a transient constellation of comparative advantages. Specifically, in order of cost, the “work” required to make things for Americans can be done by (i) Emerging market humans, (ii) American machines, (iii) Foreign machines, and (iv) American workers. Our only winning strategy is for (ii) to overtake (i) and stay ahead of (iii). The American worker is not in the contest.

    The existentially necessary strategies – the ones we will adopt or suffer for not adopting – are to shorten careers and reduce the number of workers per household. That means more entitlements and, probably, more constraining gender roles. So it’s not likely to come about without a serious crisis. But it is where we are going, if we are to keep going at all.

  16. E Shackcloth says:

    The system in general is outdated and needs replacing honestly. And I mean right down to the overly fake concept of money.

    I am not qualified to go into detail of what needs to be done or how, but it’s painfully obvious to me as a young adult, that this trend is only going to continue exponentially. Trying to “create new jobs” to fill the gap is a fools errand and a stop-gap measure that’s doomed to fail. Society needs to learn to embrace this change, and how to become stronger for it. A modern society doesn’t need it’s people wasting time flipping burgers and scanning bar codes, any more than we need people carrying buckets from the rivers or hand washing our clothes in tubs.

    Even our technological growth is being painfully stunted by the drip feeding of technologies to consumers, and the artificial expense that is placed on progress.

    Technology has outgrown nature, and Capitalism is swallowing it’s own tail. Humanity flounders in confusion, and governments blindly struggle to maintain a gradually failing system. People need to come to understand that a future resembling today’s society and economic model is impossible.