One of the most fashionable COVID-19 proposals being circulated by Washington politicians and pundits these days is the “repatriating” of global supply chains for pharmaceuticals, medical devices, and “personal protective equipment” that have been “outsourced” over the past three decades, leaving America utterly dependent on foreign countries, especially China, for these essential products. Bringing that manufacturing back home, so the story goes, is the only way to ensure future “resiliency” in the face of the next global pandemic. Otherwise, we’ll once again be left empty‐handed and helpless — and, even worse, at the mercy of China and other untrustworthy foreigners.
It’s a dire prediction. Fortunately, it is also mostly wrong. While COVID-19 has revealed certain products or policies that need a second look, full‐scale repatriation — via protectionism, industrial policy, and other state micromanagement of private behavior — would leave us less, not more, prepared for the next national emergency.
The first way repatriation plans go wrong is by relying on bogus data. The terrifying and oft‐repeated claim that “80 percent of America’s pharmaceutical drug supply comes from China” is, as Reason’s Eric Boehm details, essentially baseless. A Washington Free Beacon piece provides an excellent case study in how easy it is to misunderstand the data:
China … makes many American generic drug imports, including 95 percent of ibuprofen, 70 percent of acetaminophen, and 40 to 45 percent of penicillin. The slowdown attributable to the coronavirus’s effect on Chinese manufacturing has already created the possibility of a shortage of over 150 drugs, according to an FDA report.
Those percentages sound very bad. But isolated import‐share figures tell us very little about actual “vulnerabilities,” because they omit domestic production and local inventories. According to a new study from the St. Louis Federal Reserve, China supplied almost 30 percent of all imported “essential medical equipment” (hand sanitizer, masks, personal protective equipment, ventilators, etc.) in 2018 but accounted for only 9 percent of total domestic consumption because American producers supplied the vast majority (more than 70 percent) of these products. The U.S. does indeed need better data on pharmaceutical production (something the CARES Act seeks to remedy), but it’s clear that the United States still makes a lot of generic drugs — and even plays host to one of the largest ibuprofen plants in the world (in Texas). So China’s share of imports, without more context, is pretty meaningless.
At the same time, we have massive stockpiles of other critical drugs to prepare for crisis‐related spikes in demand. When India temporarily banned hydroxychloroquine exports, for example, everyone freaked out because India makes a lot of it. The freakout quickly dissipated, however, when it was revealed that we already had 31 million doses (donated by multinational drug companies) in the Strategic National Stockpile. State and local governments have since stockpiled 30 million more pills, and local pharmacies have their own inventories. Should hydroxychloroquine turn out to be a miracle drug (please consult your doctor!), it looks like we’ll be just fine, regardless of what India does next. (Though it certainly wouldn’t hurt to stop picking needless trade fights with India, the EU, and other allies.)
Days after that scary‐sounding FDA story about the “possibility” of drug “shortages” was leaked in late February, the agency published a formal “Coronavirus (COVID-19) Supply Chain Update,” finding no serious threats at that time. In fact, according to the FDA, there were 109 drugs “currently in shortage” on December 19, 2019 — weeks before the coronavirus hit; on February 28, 2020, in the middle of China’s crisis, that number was 103; last weekend, in the middle of our crisis, it was . . . 108. Maybe major vulnerabilities in the drug‐supply chain will materialize in the future, but so far, so good.
While the data on drug production are far from perfect, the numbers we do have aren’t nearly as bleak as the politicians and pundits claim. As already noted, more than 70 percent of essential medical supplies consumed in the United States in 2018 were made here (so much for that “deindustrialization” we hear so much about). Meanwhile, according to the Food and Drug Administration, of the roughly 2,000 global manufacturing facilities that produce active pharmaceutical ingredients (APIs), 13 percent are in China; 28 percent are in the USA, 26 percent in the EU, and 18 percent in India. For the APIs of World Health Organization “essential medicines” on the U.S. market, 21 percent of manufacturing facilities are located in the United States, 15 percent in China; and the rest in the EU, India, and Canada.
A new report from the Pacific Research Institute provides more good news: China accounted for just 1 percent ($1.5 billion) and 18 percent ($1.2 billion) of finished pharmaceutical imports and APIs, respectively, into the United States in 2019, and “combined, the United States had a $1.6 billion trade surplus with China for these pharmaceutical products and ingredients.” Within the United States, moreover, there are about 1,300 plants making drugs (often using imported inputs) employing about 120,000 American workers. According to the World Trade Organization, the United States is not just a top global producer and importer of medical goods but also a top exporter (No. 2 overall, right behind Germany).
Yet even if the data were as scary as claimed, widescale repatriation and “self‐sufficiency” policies defy basic economic sense. For starters, there is the problem of maintaining pandemic‐level capacity in non‐pandemic times. According to Politico, for example, the United States could use 20 times the number of N-95 masks (500 million) that it used last year, and U.S. hospitals expect demand to multiply “8.6 times for face shields, 6 times for swabs, 5 times for isolation gowns and 3.3 times for surgical masks.” Maintaining that much excess capacity in times of normal demand is extremely costly (for many industries, profitability usually kicks in around 80 percent capacity utilization), and running at that level when there’s no pandemic would produce a global glut — ironically, quite similar to the ones we complain about when China subsidizes “global excess capacity,” and certain to cause new trade tensions.
At the same time, there is burgeoning evidence of American manufacturers (and small‐business owners) adapting their operations to meet the explosion in U.S. demand, but this often entails moving from higher‐value products (e.g., cars or designer jeans) to products that are, in normal times, much lower‐value products (e.g., ventilators or face masks) that can be made much more cheaply elsewhere. Redirecting this productive capacity to lower‐value goods reduces these companies’ and their workers’ welfare and also raises prices for American consumers. That’s acceptable during a crisis, but not permanently. Long‐term public health in the United States will not be advanced by forcing American doctors and their patients to pay five times as much for “Made in Texas” face masks as for the same thing made in China. Indeed, it’s precisely for this reason that dozens of health and medical associations recently wrote to the president opposing new “Buy American” restrictions on medical goods.
Moreover, there is a thick historical record showing that protectionism, in whatever form, fails to produce a lean, thriving, and innovative domestic industry. Instead, past cases such as steel tariffs and Jones Act shipping restrictions show that government efforts to protect industries deemed “essential” to national security result not only in foreign retaliation and higher consumer costs but also in bankruptcies, layoffs, lower domestic output (e.g., fewer ships), and a small cadre of politically connected zombie companies whose overcompensated executives divert corporate resources from innovation and efficiency to lobbying and executive (not worker) pay. That’s precisely what you don’t want from an industry making life‐saving pharmaceuticals or medical devices.
On the other hand, there is ample evidence that “globalization” (i.e., the free flow of goods, services, capital, and information) has been an absolute blessing for the medical field and, thus, humanity — well beyond simply cheap PPE and generic drugs. Today, doctors and researchers from around the world work together to cure diseases that once killed hundreds of thousands of people each year. And it’s just this type of collaboration that might help to beat the current pandemic: as recently documented by the Cato Institute’s Chris Edwards, “globally, dozens of biotech and pharmaceutical companies are rushing to develop vaccines and treatments for covid‐19 using a diversity of approaches.” Some of this might even involve (gasp!) China, as the Martin Sandbu of the Financial Times notes: “Within weeks of the new coronavirus’s emergence, Chinese scientists had sequenced its genome and shared their knowledge with the entire world.” Sandbu rightly adds that such a move “was made possible by our unprecedented degree of globalisation of technology, knowledge and communication, which in turn had piggybacked on expanding economic exchange.” Indeed.
Of course, none of this happy talk changes the fact that a global pandemic raises unique problems that a normal market economy can’t (indeed, that no kind of economy can) immediately address. Fortunately, there are far easier solutions than fundamentally changing the U.S. and global economies via broad‐based protectionism and industrial policy. The most obvious place to start is better government stockpiling of essential medical goods — something Switzerland, for example, has used to great effect during the current crisis (and something the United States is now quietly trying to fix in the CARES Act). Stockpiling is precisely the type of government action that even libertarians can support, and it’s also relatively cheap: For example, keeping pandemic‐level quantities (500 million) of two‐cent N-95 masks costs a whopping $10 million (a rounding error in D.C. these days).
Other policies include requiring domestic manufacturers to maintain larger inventory levels for critical inputs; following Korea’s lead and implementing “pandemic preparedness” regulations that allow for the swift approval and production of testing and other essential medical equipment; entering into new trade agreements with Vietnam, India, and other countries that can serve as alternatives to China; or, as already noted, collecting and publishing better data on medical‐goods production and trade. All of these policies — along with private companies’ post‐COVID efforts to fortify and diversify their supply chains — are infinitely better than forcefully “repatriating” them via something like the “Jones Act for Medical Goods” (shudder to think).
Maybe these moves would still leave small holes in the system, requiring heavier‐handed government planning, but maybe they won’t. Right now, we just don’t know. We know enough, however, to be skeptical of big new programs, especially when they just so happen to be the same ones that these same politicians have been pushing for years, often in response to imagined crises such as trade deficits or sugar — yes, sugar — imports.
COVID-19 is a real crisis and has raised real concerns. It demands real solutions — not protectionism.