Dean Baker On Economists’ Contradictions On Productivity

Dean Baker has been writing some of the best articles about the U.S. economy in the past few weeks. I have already linked to his articles on this blog. If you have missed them, follow the tag with his name at the end of this post.

His analysis has been about economists’ incorrect narrative about productivity and U.S. trade. This narrative simply is that workers are losing jobs because of automation and not because of U.S. trade and globalization. Dean Baker shows that it’s quite the opposite.

In his latest post, he points out the inconsistency of economists stands. One on the one hand, economists seem to be saying that productivity rises will be large because of automation and on the other hand saying that productivity won’t rise much.

This is the stand of Paul Krugman whom Baker cites. In his column On Economic Arrogance for The New York Times, Krugman argues that it’s impossible for the U.S. to grow at 3-3.5% in the next decade. Krugman says:

The only way we could have a growth miracle now would be a huge takeoff in productivity — output per worker-hour. This could, of course, happen: maybe driverless flying cars will arrive en masse. But it’s hardly something one should assume for a baseline projection.

Now, it’s quite possible that Donald Trump, being the erratic person he is, can mess thinks up. You can’t be sure what he is up to. He may come up with a large fiscal expansion or not. But the more important question is about the possibility. Would Hillary Clinton or Bernie Sanders been able to achieve the growth rate of 3-3.5% had they been the President? According to Krugman it’s simply unlikely because he thinks productivity growth will be low.

The same kind of argument was given by Krugman when Bernie Sanders released his economic plan. See John Cassidy’s article Bernie Sanders And The Case For A New Economic Stimulus Package in The New Yorker from for a good analysis, written February 2016. In the article Cassidy reminds us of the Kaldor-Verdoorn Law according to which faster GDP growth leads to faster productivity rises.

To break out of this low-growth trap, the economy needs policies designed to boost demand and push it onto a higher growth path: one in which rising investment, higher levels of productivity, rising rates of participation in the labor force, and higher wages all reinforce each other. With these conditions in place, companies would have more of an incentive to make capital investments, and as the price of labor rises they would also have an incentive to innovate and move up the value chain. Realistically, we can’t expect 5.3-per-cent G.D.P. growth and 3.3-per-cent productivity growth to persist for a decade. But we don’t necessarily have to settle for the 2.1-per-cent G.D.P. growth we’ve become accustomed to, or even the 2.3-per-cent rate that the Council of Economic Advisers has identified as its long-term potential. The U.S. economy has the resources and the ingenuity to do better than that.

As [Gerald] Friedman points out, the idea that faster G.D.P. growth generates higher productivity growth (and higher wages) has historical support. Back when I was an undergraduate, it was associated with Nicholas Kaldor, the British Keynesian, and with P. J. Verdoorn, a Dutch economist. (In a footnote, Friedman cites both of them.) More recently, in the late nineteen-nineties we saw rapid rates of G.D.P. growth and productivity growth appear in tandem. Some analysts would claim that the latter generated the former, rather than vice versa, but that argument isn’t convincing. In a Kaldorian virtuous cycle, G.D.P. growth spurs productivity growth, which, in turn, spurs G.D.P. growth. Causation goes both ways.

So the U.S. economy has a lot of scope for growth in the next decade. There are ways:

  • The U.S. GDP can grow without productivity rising simply because there’s huge unemployment. The U-6 unemployment rate is currently 10.1%. Because of Okun’s law which says that a growth of 1.65% will reduce unemployment by 1% (see estimation of this parameter for the U.S. by Matias Vernengo here.)
  • Rise in the workforce. Over time, work force also rises, (depending on demographics), this adds to the capacity of the economy. A notable thing during the crisis is the fall in the workforce in the U.S. Am not sure about details around this. Seems this is not captured in U-6. One would have expected the labour force to have not dropped and unemployed being captured in U-6. In other words, it’s possible that there’s more potential labour available.
  • As above, because of Kaldor-Verdoorn law, higher production leads to higher productivity. The Verdoon coefficient is 0.63 as per the paper cited above. So a growth of 1% will lead to a rise in productivity by 0.65%. Higher productivity implies that it takes less labour to produce the same amount of things. This in turn implies that the labour can be employed elsewhere or to produce more of the same stuff. So rise in productivity increases production capacity.

So these three points are enough to prevent the U.S. economy from hitting full capacity for a long time. Scarcity is a thought which needs to go from economics.

At any rate, the main purpose of this post was to point out the contradictory things economists say. Economists seem to be saying that automation will rise, which translates to rise in productivity. At the same time they are also saying that productivity will be low. Further Dean Baker has also pointed out how the Federal Reserve is on the path to raise interest rates. If they are worried about automation killing jobs, why aren’t they worried about the Federal Reserve’s rate hikes destroying jobs?

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Note: If not opted-in, comments in moderation can be previewed only for a short while.

This site uses Akismet to reduce spam. Learn how your comment data is processed.