On May 14, U.S. President Donald Trump threatened to break the United States’ economic relationship with China. “There are many things we could do,” he told Fox Business host Maria Bartiromo. “We could cut off the whole relationship. Now if you did, what would happen? You’d save $500 billion.” It was Trump’s most extreme anti-China rhetoric to date, but it wasn’t out of step with the mood in Washington. Both Republicans and Democrats agree that China has transformed from a competitor into an adversary, and perhaps even an enemy.

As tensions have mounted as a result of the coronavirus pandemic, which originated in China and which Beijing initially sought to conceal, the Trump administration has taken steps to curtail economic relations. Last month, it directed the Federal Retirement Thrift Investment Board, which manages hundreds of billions of dollars in government retirement savings, to halt investments in Chinese companies. It also prevented the Chinese telecommunications manufacturer Huawei from using U.S. technology to design or produce semiconductor chips. And after China imposed new national security laws on Hong Kong, Trump announced that he would revoke the territory’s special trade privileges.

Washington has become obsessed with “decoupling”—the notion that the United States and China should sever the complex supply chains that bind them together. If the United States doesn’t “re-shore” these supply chains in the wake of the pandemic, Trump’s trade adviser Peter Navarro, who wrote a book called Death by China, recently warned, it “will sink into the abyss.” But for all the official enthusiasm for decoupling, there is little agreement on what it would actually entail. Does decoupling mean reducing U.S. economic vulnerabilities? Making the United States less dependent on China? Exploiting China’s dependence on U.S. technology? Withdrawing wholesale from the World Trade Organization? Turning any of these proposals into effective policies would require a level of technical knowledge that neither the U.S. government nor the private sector has right now. Flying blindly ahead, moreover, risks hurting the United States as well as China.

Slapdash efforts to sever risky dependencies on China could end up lopping off healthy and important economic relations with not only that country but the rest of the world. Blocking Huawei’s access to U.S. technology, for instance, may just encourage foreign companies to redesign their supply chains around non-U.S. technologies. Instead of decoupling, the United States should therefore think about “recoupling” the world’s supply relationships in ways that will make them less vulnerable to accident and attack. Doing this right, however, will require U.S. policymakers to develop a whole new body of expertise.


Decoupling countries with complex economic relations is like carrying out a difficult surgical operation. One had better have a good idea of where the patient’s vital organs are before one starts cutting. The Trump administration learned this the hard way in April 2018, when it imposed sanctions on the Russian aluminum giant United Company Rusal in retaliation for Moscow’s interference in the 2016 U.S. election. Rusal turned out to make a specialized aluminum product that was vital to European car manufacturers, which suddenly had to worry about paying vast fines if they continued doing business with the Russian company. Backlash from the automotive industry eventually pressured the Trump administration to backtrack and lift the sanctions.

U.S. policymakers are now making a similar discovery about China: its economy is not a discrete organism that can easily be separated from the global economy but rather a Siamese twin, connected by nervous tissue, common organs, and a shared circulatory system. The coronavirus pandemic has revealed the hidden vulnerabilities of this interconnected system of supply chains, especially for medical equipment. Growing tensions in the U.S.-Chinese relationship have revealed similar hidden vulnerabilities in the technology sector, among other sectors. But cutting the wrong supply chain can have unexpected effects, as it did in the case of Rusal. And just cutting is insufficient: it is also necessary to reconnect supply chains in ways that make them more resilient.

China's economy is not a discrete organism that can easily be separated from the global economy but rather a Siamese twin, connected by nervous tissue, common organs, and a shared circulatory system.

The global economy has become vastly more complex and vastly more interconnected in recent decades, but foreign policy expertise has lagged behind. During the Cold War, policymakers who worked on nuclear strategy regularly spoke to nuclear scientists who understood the workings of the weapons systems being deployed. Supply chains are nearly as complex as nuclear physics, but those who study them rarely engage with policymakers. The result is that policymakers now face a dilemma similar to the one surgeons faced at the dawn of the age of modern medicine: pressing demand to fix problems but limited knowledge of how to do so. Today’s policymakers can vaguely grasp that some healthy-seeming economic relationships have become dangerous and some even gangrenous. But they don’t know which relationships should be saved, which should be severed, and which should be rearranged—and they are working with little more than prayers and blood-speckled hacksaws.


To determine where and how U.S. supply chains need to be reengineered, the U.S. government will have to vastly improve its understanding of the physical economy—the supply relationships that hold the economy together but that most economists ignore. In doing so, it can build on a few pockets of excellence. The Department of Defense has people who understand supply chains, albeit in the specialized context of military spending. The Department of the Treasury’s Office of Foreign Assets Control has also built up specialized knowledge of supply chains in order to administer U.S. sanctions policy. It still gets things wrong on occasion—it didn’t foresee that sanctions on Russian aluminum would devastate European car manufacturing—but it is developing an intimate understanding of how the world economy really works. The International Trade Centre has also started to map out supply chains using publicly available data. All of these agencies will need to coordinate their efforts to chart supply relationships more closely, perhaps through a new institutional arrangement overseen by a revitalized National Security Council.

But even a better-coordinated U.S. government will need to draw on outside expertise. As they did during the Cold War, policymakers will have to start talking to people with deep technological knowledge—of supply chains in this case, rather than chain reactions. Some of these people are in industry: supply chain and logistics managers. Others are in the academy, studying the relationships as well as the markets that make up the real economy.

As they did during the Cold War, policymakers will have to start talking to people with deep technological knowledge—of supply chains in this case, rather than chain reactions.

Only once policymakers have assembled the requisite knowledge and expertise will they be able to identify vulnerabilities—the firms and relationships that create chokepoints and may therefore pose risks to U.S. security if they pull out of supply chains. The Cambridge University economists Vasco Carvalho, Matthew Elliott, and John Spray have begun to map these supply relationships. They argue that it is possible to use a combination of machine learning and clustering algorithms to identify the supply relationships and “bottleneck firms” that can’t be disrupted without wider economic consequences. When such relationships or firms are found to be in potentially hostile jurisdictions, or vulnerable to hostile actors, policymakers may decide that supply chain adjustments are necessary. These adjustments would amount not to wholesale “decoupling” but rather to targeted “recoupling” efforts aimed at embedding supply chains within sustainable frameworks that balance the need for efficiency against newly visible security risks. In some situations, that may mean reducing reliance on key firms. In others, it may involve fostering new supply relationships, including firms in other jurisdictions. The crucial point is that not all firms, suppliers, or countries are equally risky and that it may be better to build in redundancy for possible bottlenecks than to cut them away entirely.

Remaking global supply chains in ways that minimize risk will not be easy. Complicated trade relationships are difficult and expensive to replicate. As a result, manufacturers are unlikely to diversify their supply chains unless they are compelled to do so, and they will add the cost of redundancy to their final products. The best approach may thus be what Susan Helper, an economist at Case Western Reserve University, calls the “high road” approach to supply chains. Companies often try to cut costs by using suppliers in lax jurisdictions, but in doing so, Helper argues, they weaken their supply chains. Improving regulation can prevent such outcomes. Helper offers an example from the pharmaceuticals industry. If the United States were to require better labeling (explaining which ingredients are made where), and carry out surprise inspections of overseas facilities, drug makers might be less inclined to produce vital drugs in low-cost jurisdictions for fear that customers would stop buying them and regulators would take action. As a result, the United States would reduce its dependency on foreign pharmaceutical manufacturers.


Regardless of what Trump says, it will be impossible to fully separate the U.S. and Chinese economies—and still more so to cut the U.S. economy adrift from the world. Every U.S. action toward China—offensive or defensive—will therefore continue to produce a Chinese reaction that is felt by the United States. The benefits of blocking Huawei’s access to U.S. technology, for example, are already clearly inseparable from the risks. China may retaliate. It will certainly redouble its efforts to develop sophisticated semiconductors of its own. Foreign businesses and countries may decide that they can best minimize risk by limiting their contacts with the U.S. economy. Cutting China out of the U.S. innovation system, in other words, will likely prompt China to cut the United States out of its innovation system—and could cause the United States to lose access to other innovation systems as well.

As the U.S.-Chinese rivalry heats up, it will need to be managed, just as the U.S.-Soviet rivalry was, through expertise, risk management, and careful efforts to forge mutual understanding. But U.S. policymakers will have to enlist new kinds of experts to develop the doctrines, stratagems, and tactics to manage a conjoined adversary. At the same time, they will have to create new kinds of knowledge and expertise, fostering a community of experts and practitioners who understand logistics as well as national security. Finally, they will have to align the United States’ industrial strategy with those of U.S. allies in Europe and Asia that are working on their own re-shoring proposals. All of this means that the United States should be focused on recoupling rather than decoupling, rewiring rather than retreat.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • HENRY FARRELL is Professor of Political Science and International Affairs at George Washington University.
  • ABRAHAM NEWMAN is a Professor at the Edmund A. Walsh School of Foreign Service and in the Department of Government at Georgetown University.
  • More By Henry Farrell
  • More By Abraham Newman