Remember the story of that once-great nation that sacrificed its well-paying manufacturing jobs for low-wage, burger-flipping jobs at the altar of free trade? At one time, that story was a popular rejoinder of manufacturing unions and their apologists to the inconvenient facts that, despite manufacturing employment attrition, the economy was producing an average of 1.84 million net new jobs per year every year between 1983 and 2007, a quarter century during which the real value of U.S. trade increased five-fold and real GDP more than doubled.
The claim that service-sector jobs are uniformly inferior to manufacturing jobs lost credibility, as average wages in the two broad sectors converged in 2005 and have been consistently higher in services ever since. In 2011, the average service sector wage stood at $19.18 per hour, as compared to $18.94 in manufacturing. (But I don’t recall buying any $25-$30 hamburgers last year.)
One reason for U.S. manufacturing wages being higher than services wages in the past is that manufacturing labor unions "succeeded" at winning concessions from management that turned out to be unsustainable. The value of manufacturing labor didn’t justify its exorbitant costs, which encouraged producers to substitute other inputs for labor and to adopt more efficient techniques and technologies.
With the superiority-of-manufacturing-wages argument discredited, new arguments have emerged attempting to make the case that there is something special – even sacred – about the manufacturing sector that should afford it special policy consideration. Many of those arguments, however, conflate the meanings of manufacturing sector employment and manufacturing sector health or they rely on statistics that don’t support their arguments or they become irrelevant by losing sight of the fact that resources are scarce and must be used efficiently. And too often the prescriptions offered would place the economy on the slippery slope that descends into industrial policy.
I recently submitted this rebuttal to this essay by an environmental sciences professor by the name of Vaclav Smil, who commits those errors. (Judging from the tone of his mostly evasive response to my rebuttal, Smil doesn’t seem to have much tolerance for views that differ from his own.) Perhaps most noteworthy among Smil’s slew of questionable arguments is his claim that manufacturing companies, like Boeing, valued at $50 billion, are better for the economy than service companies like Facebook, which is also valued at $50 billion because
[i]n terms of job creation there is no comparison... Boeing employs some 160,000 people, whereas Facebook only employs 2,000.
Granted, Boeing’s operations support more jobs. But is that better for the economy than a company that provides the same value using 1/80th the amount of labor resources? Of course not. We need economic growth in the United States to create wealth and increase living standards. Economic growth and employment are not one and the same thing. In fact, the essence of growth is creating more value with fewer inputs (or at lower input cost). Creating jobs is easy. Instead of bulldozers, mandate shovels; instead of shovels, require spoons. Inefficient production techniques can create more jobs than efficient ones, but they don't create value, which is the economic goal.
With 2,000 workers producing the same value as 160,000 – one producing the same value as 80 – Facebook is 80 times more productive than Boeing, freeing up 158,000 workers for other more productive endeavors (perhaps 79 more Facebook-type operations). If those companies were individual countries, the per capita GDP in Facebookland would be $25 million, but only $3.125 million in Boeingia. Where would you rather live?
Smil calls my assessment a cruel joke, presumably for its failure to empathize with unemployed and underemployed Americans, by considering value before job creation. But policies designed to encourage more Boeing’s, as Smil supports (or, in fairness, any businesses that employ at least X number of people or meet this requirement or that) would likely retard the establishment of firms, like Facebook, that produce the goods and services that people want to consume. The provision of goods and services that people want to buy – rather than those that policymakers in Washington think people want to buy (or are happy to force them to buy) – is the essence of value creation.
Thus, policies should incentivize (or, at least not discourage) the kind of innovation and entrepreneurship needed to create more Facebooks? This kind of business formation occurs in environments where the rule of law is clear and abided; where there is greater certainty to the business and political climate; where the specter of asset expropriation is negligible; where physical and administrative infrastructure is in good shape; where the local work force is productive; where skilled foreigners aren’t chased back to their own shores; where there are limited physical, political, and administrative frictions; and so on. In other words, restraining the role of government to its proper functions and nothing more would create the environment most likely to produce more Facebooks in both the manufacturing and services sectors.