Manufacturing dissent

United States manufacturers, like their Australian counterparts, are indulging some hyperbolic alarm about their future, but for different reasons. U.S. economic growth seems too anaemic to support demand in the sector; Australia’s economic growth seem to be bypassing it. Still, this self-interested plea in the NYT from a director of GE for public subsidies (“innovation support”) for U.S. manufacturing industry has a familiar ring.

Rebuilding our manufacturing capacity requires the demolition of the idea that the United States can thrive on its service sector alone. We need to create at least 20 million jobs in the next decade to offset the effects of the recession and to address our $500 billion trade deficit in manufactured goods. These problems are related, given that the service sector accounts for only 20 percent of world trade.

Extract from Manufacturing a Recovery – NYTimes.com

Three points on which I dissent from this opinion.

  1. When the U.S. economy does create net jobs, you can bet that (competitive) manufacturing won’t be at the top of the list (or it would’t be competitive). It’s much more likely that U.S. job growth will remain dominated by services industries.
  2. The manufacturing trade “deficit” is the correlative of consumer benefit from trade; not a sign of recession.
  3. The size of the services sector in world trade is no indicator of its contribution to a manufacturing economy (services trade is, in any case, hugely under-counted by current methods).

But I agree with one of Ms Hockfield’s sentiments; Australia, like the United States, certainly needs to stay in the business of real—not just invisibles—output to sustain growth.

Manufacturing has “shrunk” in terms of employment and measured-output (GDP) share in all OECD countries over the past half-century because its productivity has improved, due in large measure to improvements in production and non-production services inputs. But those services industries—which have absorbed labor at a rapid rate over the same half-century—may (probably do) exhibit lower rates of own-productivity improvement. The overall outcome for the growth of the economy is the sum of the output growth of all industries; so it’s probably essential for both (high productivity) manufacturing—or farming, or mining—and for services industries that they are joint producers. [Geek note: I’m debasing a line of argument from Baumol, to Solow to Acemogelu and others on the role of services in modern economies. See this survey paper by Joe Francois and Bernard Hoeckman or this one by Robert Inklaar et al.]

In Australia, services most valuable contribution to measured-output may be linked to their continued collaboration with merchandise industries. We’re not like Singapore, for example, or Belgium, endowed to be predominantly a services economy based on entrepôt activities. The geographical extent of our country and it’s location means we’re endowed for other kinds of production where land plays a role. Certainly, we’ve become an economy dominated by services output, but I see zero chance that real production is going away. It’s safe to predict that when the price of capital assets not specialised for services production (farms, mines, steel-mills) falls relative to the price of other productive assets there will be investors: whether from Australia or elsewhere is irrelevant.

Although it may, in some circumstances, make sense to subsidise services industries (“innovation”) in the guise of supporting merchandise output industries, what most of the evangelists for manufacturing or agriculture or even resources seem to miss is that such spending will very likely increase the size of the services input industry (research and development, logistics, process design, product design) and shrink the level of employment and possibly even the physical scale of the output industry (as productivity rises).

No Comments

Leave a Reply

Your email is never shared.Required fields are marked *