Suzanne Berger’s discussion of manufacturing in the United States raises many good issues, but it asks more questions than it answers. On its face it gives short shrift to the main factors determining manufacturing employment and output, specifically trade, the value of the dollar, and the lingering effects of the Great Recession. At a deeper level it lacks a clear sense of the tradeoffs involved in moving toward more high-road manufacturing.
On the first point, it seems difficult to contest that trade has been a major factor in declining manufacturing employment in the United States. The country first faced large trade deficits in the early 1980s, as the soaring dollar made U.S. goods less competitive. Manufacturing employment fell from 22 percent of total employment in 1980 to 18 percent by 1985. We saw a similar drop when the Clinton administration’s botched bailout from the East Asian financial crisis led to another surge in the dollar. The share of manufacturing employment fell from 15.2 percent in 1997 to 11.8 percent in 2004. The Great Recession led to a further decline of roughly one percentage point, most of which has not yet been recovered. (The drop off in manufacturing employment would appear smaller if we looked at hours, since the average manufacturing workweek has lengthened by 5 percent since 1980, while it has been cut back by 5 percent in the rest of the economy.)
If we want to talk about manufacturing, we must keep our eyes on the trade deficit.
The loss from trade comes against a backdrop of worldwide declining manufacturing employment. Even in Germany, the share of employment in manufacturing fell from 34.8 percent in 1980 to 19.8 percent in 2012. This drop reflects the fact that productivity tends to grow more rapidly in manufacturing than in other sectors, and demand for goods is fairly inelastic. As the relative price of manufactured products declines, we spend less money on them.
Berger’s reference to Houseman on mis-measured productivity matters little in the context of this debate. If we overestimated manufacturing productivity because we underestimated the value of imports, then we underestimated our trade deficit in manufacturing. We should expect to see a decline in the manufacturing share of employment, and in the United States the decline is more severe than elsewhere in large part because of a trade deficit that is still close to $500 billion per year (3 percent of GDP).
This points to a remedy that is economically simple, even if the politics are not: bring the dollar down to make U.S. goods more competitive internationally. This will boost manufacturing and, if we actually balanced trade, get us most of the way back to full employment.
The Bureau of Labor Statistics reports that average hourly compensation in German manufacturing was $47.40 in 2011, one-third higher than the $35.50 reported for the United States. Clearly Germany, along with several other European countries, is doing something right compared to the United States if its manufacturers can afford to pay employees so well while remaining in business. Part of the story likely involves patient, long-term capital investment, as Berger suggests. But we need more evidence and some sense of the tradeoffs. If German firms provide lower returns to investors, how much lower is it?
The other important part of the story is the difference in labor relations in the two countries. The United States is alone among the wealthy countries in allowing employment at will. Unless workers have a union contract, their employer can fire them at any time. Germany, like the rest of Europe, has strict rules governing dismissal, requiring substantial severance pay for long-term employees. This changes the equation for German firms: they are largely stuck with their workers and have a strong incentive to make them as productive as possible.
Of course Germany also has a far higher unionization rate than does the United States. And its laws guarantee workers a voice in works councils at all medium and large firms, whether or not they are unionized. My guess is that the difference in labor relations has at least as much to do with Germany’s success in pursuing high valued-added manufacturing as do the differences in the financial structure.
In sum, Berger has raised some important questions, but if we want to talk about manufacturing, we must keep our eyes on the trade deficit, which will be the low-hanging fruit in efforts to revive American manufacturing. Her models for more long-term planning are promising too, but a consideration of labor relations would fill out the story.