Why Is Productivity Declining?

I found this post by David Goldman AKA “Spengler” from Asia Times via MENAFN thought-provoking. His basic point is that the only time we have seen such large declines in productivity as we are presently seeing was during periods of high inflation. He makes some points with which I disagree. For example

In addition, the Federal government in the late 1970s and early 1980s spent 1.2% of GDP on federal R&D, funding the invention of the Digital Age. Tax cuts on income and capital gains provided the incentive for entrepreneurs to bet on these new technologies, creating the longest peacetime expansion in American history.

There are several things in that with which I disagree. First, I don’t think it’s was the federal R&D spending of the late 1970s and early 1980s that produced the “Digital Age”. IMO the federal contribution was earlier than that in the 1960s. DARPANet, the progenitor of the Internet, was first operational in 1969. The PC boards and semiconductor research that led to the “Digital Age” was in the 1960s, too. What happened in the 1980s was the enormous private investment in personal computers and networking. The reason for that was obscure then and it remains obscure. I can show you contemporaneous articles in business and technical magazines musing about it.

The part of his post that I really wanted to highlight was this:

The worst labor productivity decline on record during the second quarter of 2022, meanwhile, points to long-term and persistent inflation. Output per manhour in the nonfarm business sector declined by 2.2% at an annual rate. Think of it as the economic equivalent of long Covid.

We haven’t seen labor productivity declines of this magnitude since the 1970s, during the last great wave of inflation.

Why? He doesn’t really answer that. My observation from that chart is that the trend has been down for 80 years.

I think that like most things that’s multi-factorial with explanations including:

  • Increasing reliance on the service sectors of the economy for growth. Services don’t increase in productivity very quickly if at all.
  • Public and private overinvestment in services, see above.
  • Declining returns on investment
  • Shortfall in private investment, cf. above
  • Treating the absolute number of jobs as a mark of success. That inclines policy to encourage minimum wage jobs and the productivity of minimum wage jobs is low, practically by definition.

My opinion is that if we want to stop the decline in productivity it will take changes in a lot of things including trade policy, immigration policy, and tax policy. Said another way it will be hard but that’s what we need to do if we want wages to rise.

6 comments… add one
  • bob sykes Link

    That is a very ragged chart. I don’t think you make any interpretation of the short-term fluctuations themselves. They are likely random noise. The only information I get out of it is a small downward trend since the beginning in 1948, which you already observed.

    I think your suggestions are correct. That trend is 74 years old, and it traverses several very different US economies, from the industry-heavy, labor-intensive, unionized post-war economy to today’s services-heavy, un-unionized economy.

    So, as much as I like and read Goldman, I think his post is essentially nonsense. He’s trying to interpret noise.

    I am less confident than you that there is a cure for the decline. I think low productivity is just baked into services. How do you increase the output of teachers or nurses or auto mechanics or police and firemen?

  • CuriousOnlooker Link

    The rise in 2020 and the dip now has a clear explaination.

    In 2020; when the pandemic occurred, layoffs due to forced shutdowns were concentrated in the service industry (hospitality, gyms, etc) that had low worker productivity — therefore the average productivity of the remaining workforce went up. Now its going in reverse, job growth from reopening is concentrated in the service industry which is pulling the productivity average down.

    As to the long term trend; Lyn Alden commented that the generational trends in productivity usually reflects energy consumption or price. Its downstream of the fact increasing productivity is the result of humans exploiting ever denser forms of energy. The last real big advancement on that front was nuclear power in the 1960’s.

  • Andy Link

    I’ve been saying for a couple of years now that the economy is weird. Unexpected things keep happening – or at least things that don’t fit the historical patterns we rely on.

    In my view, that is not comforting. It suggests that we don’t know as much as we thought we did, and/or that our data collection and analysis are bad or focused on the wrong things, and/or that a unique combination of conditions are taking us into “there be dragons” territory that are inherently unpredictable.

  • CuriousOnlooker Link

    “that our data collection and analysis are bad”

    That’s get into a growing beef of mine — the way the government collects data is byzantine.

    As an example, look at the jobs report each month (household survey and establishment survey). The household survey is a poll of 60K households and the establishment is a poll of 131K buinesses and government agencies. Why are we using polls or surveys instead of using payroll tax / income tax withholding that employers / self-employment do every week / 2 weeks / bi-monthly / monthly? That should give accurate numbers job numbers for every person that pays taxes. Given the data has tax id’s — its also easy to correct nuances like multiple job holders.

    Another example is anything to do with COVID. The government never did collect basic epidermological data.

  • Andy Link

    Those are two excellent examples.

    It’s another example of how the federal bureaucracy has become sclerotic and soul-crushing. It’s one of the main reasons I left the federal service and also stayed away from contracting, where I could have made a lot more money than I’m making now.

  • steve Link

    If I stay on as a 1099 employee I believe I only need to file quarterly estimated taxes. The more you impose the more difficult it it for truly small businesses. Big companies can do this stuff easily. While it is always nice to have better data I think you need multiple sources and multiple methods. Not sure we have a good waytocapture off the grid, under the table stuff.

    Increasing productivity of nurses is possible though it varies by what they do.

    Steve

Leave a Comment