Don’t Step in Trump’s Trade and Manufacturing Indiscretions

Trump’s eulogy for U.S. manufacturing is dead wrong. He writes:

Trump is not the first to assert that trade killed manufacturing or to call for trade restrictions or industrial policy to bring it back. But the fact is that manufacturing is not only alive, it’s thriving. By any relevant measure — output, value‐​added, revenues, exports, imports, investment, R&D expenditures — U.S. manufacturing remains a global “powerhouse.“With respect to most of those metrics, year‐​after‐​year (with the exception of during recessions), the sector sets new records.

U.S. manufacturing attracts more foreign direct investment than any other country’s manufacturing sector. After 12 straight years of net growth, the stock of FDI in U.S. manufacturing surpassed $1 trillion in 2014. By comparison, the stock of FDI in China’s manufacturing sector — the world’s second largest manufacturing investment destination, which is now famously rife with overcapacity — is less than half the value of the U.S. stock. And Mexico’s is less than half of China’s. So, if low wages and lax environmental standards were really an investment magnet or an impetus to “ship jobs overseas” — if there really were this “race to the bottom” — we wouldn’t see foreign companies choosing to own $1 trillion worth (and growing) of U.S. manufacturing assets.

Sure, output in manufacturing contracts and expands from month to month (as it does in other sectors), but the trend has always been upward. If by “rapid deindustrialization,“Trump means that manufactured goods account for a smaller share of U.S. output than in the past, I grant him the point, but not the characterization. Manufacturing’s share of the U.S. economy peaked in 1953 at 28.1 percent, whereas today manufacturing accounts for only 12.1 percent of GDP. But in 1953, U.S. manufacturing value added amounted to $110 billion, as compared to a record $2.1 trillion in 2015. A sector that produces, today, more the six times the value in real terms what it produced when it actually was the engine of the U.S. economy, can hardly be described as having deindustrialized.

For a nation whose consumers spend twice as much on services than on goods, and where 90% of the workforce is employed outside the manufacturing sector, the obsession with manufacturing is misplaced.

Trump continues:

U.S. manufacturing employment reached its peak of 19.4 million jobs in 1979 — fourteen years before the implementation of the North American Free Trade Agreements and 22 years before China joined the World Trade Organization. So the downward trend in manufacturing employment, along roughly the same trajectory for 37 years, began long before the common scapegoats for manufacturing job loss even existed. Manufacturing employment stands at about 12.6 million jobs today.

Trump and others like to say that trade — specifically the North American Free Trade Agreement — caused those job losses. But the data don’t support that conclusion. In the 14 years between the 1979 peak and 1993 (the last year before NAFTA implementation), the manufacturing sector shed 2.7 million jobs. In the 14 years between 1993 and 2007, manufacturing shed 2.9 million jobs. In other words, the pace of job decline in manufacturing was virtually unchanged between the periods. Something else (hint: productivity gains) explains much of the reduction in manufacturing jobs. It’s worth mentioning that manufacturing jobs actually increased by 800,000 in the first 5 years following NAFTA’s implementation, which is not to suggest that NAFTA created those jobs, but to reinforce the fact that the data do not even remotely support the assertion that NAFTA destroyed U.S. manufacturing jobs.

Manufacturing job decline was most pronounced and continuous between the post‐​NAFTA peak of 17.6 million jobs in 1998 and the post‐​recession trough of 11.6 million jobs in 2010. The causes of that decline are many, but Trump and others blame China. Trade with China increased rapidly during that period, but alongside revolutions in information technology, consumer technology, and communications. It wasn’t trade that caused U.S. film and camera manufacturing jobs to disappear, but digital cameras. It wasn’t trade that reduced the demand for jobs manufacturing digital cameras, paper maps, day planners, metal compasses, wrist watches, calculators, parking meters, CD players, garage door openers, and numerous other products. The iPhone and the app industry spawned by the ubiquity of smart phones had a lot to do with waning demand for employment in those industries.

There is no doubt that shedding 6 million manufacturing jobs in 12 years — 500,000 jobs per year — constitutes wrenching change. But it doesn’t mean that U.S. manufacturing is in decline. To the contrary, record levels of output with a declining number of workers means that U.S. manufacturing has become more efficient, capable of producing more with less — in fact, much more with much less.

For those worried about whether U.S. factories can continue to compete globally, this should be reassuring. Trump’s suggestion that our economic future and national security are threatened by the alleged frailty of U.S. manufacturing is total bunk. U.S. manufacturing is strong and thriving.

But that doesn’t mean we should be cavalier about jobs. If producing one ton of widgets in a day required 10 workers on the production line last year, but only 5 workers and a new piece of capital equipment this year, we should call that progress. Widget workers are now twice as productive — output per capita has doubled. But what about the 5 workers who are no longer needed in widget production? For the economy to actually grow, for wealth to be created and standards of living to be improved, those 5 workers must be redeployed elsewhere in the economy, where they can create value.

The U.S. economy used to be quite adept at this process — at facilitating and accommodating job churn. Historically, there has been a strong relationship between economic growth, trade, and job creation in the United States. In the quarter century between 1983 and 2007 (the last year before the Great Recession), as real GDP more than doubled and the real value of U.S. trade increased five‐​fold, the U.S. economy created 46 million net new jobs, or 1.84 million net new jobs per year. That engine seems to have blown a cylinder or two in the past decade.

Trade, technology, adoption of more efficient production methods, changing consumer demands, and business failures all contribute to job churn. Policymakers should be less concerned about which of these factors is most responsible for job loss. After all, each makes society better off.

Instead, policymakers should aim to reduce the impediments to labor mobility and job creation. The collective residue of decades of piling bad policies on top of bad policies has gummed up the works. Fixing a tax code that deters the repatriation of an estimated $2 trillion of U.S. corporate profits abroad is one idea. A regulatory audit and overhaul designed to eliminate superfluous regulations and reduce the costs of regulatory compliance could induce massive amounts of investment. Eliminating import duties on industrial inputs and intermediate goods to reduce U.S. production costs would make a big difference. Reining in the protectionist practices of occupational licensing would facilitate cross‐​state labor mobility. Potential reforms abound.

But Trump’s ideas would make matter much worse. Taxing imports from China at 45 percent or to taxing imports of Fords assembled in Mexico would do nothing to fix the U.S. engine. It would spur massive job loss, reduce labor mobility, and make all Americans vastly poorer. Hey, but at least we’d get our revenge. Montezuma’s Revenge!