Skip to content

Reshoring Supply Chains: A Practical Policy Agenda

Why can’t the greatest economy in the history of the world produce swabs, face masks and ventilators in adequate supply?
                    —Lawrence Summers, Twitter, March 21, 2020

The Covid-19 crisis exposed the stark fact that almost everything we use—and need—is no longer produced in the United States. For example, 95 percent of surgical masks and 70 percent of respirators used in the United States are produced abroad. Indeed, according to some accounts, in a single year (2004), surgical masks went from being 90 percent domestically produced to 95 percent foreign produced when Kimberly-Clark shut down its U.S. factories and offshored its production.1

Though medical equipment and pharmaceutical production might be the most visible offshored sectors right now—the last domestic penicillin plant closed in 2004—the supply networks2 of other stra-tegically important industries such as telecommunications equipment have similarly left the United States. Further, most of that production was moved to one region of the world—Greater China.

One might think that the current crisis and its sharp exposition of the real risks and vulnerabilities, to both corporations and our society, would lead us naturally to reshore production. What sane CEO would risk seeing all global operations shut down because of a natural or political crisis in a small part of the world? Hedging one’s global production networks, by ensuring multiple sources and sites of production across the world, seems to be the rational choice.

Nevertheless, reshoring production networks to manufacture more domestically is no easy matter. While there have been calls to invoke the Defense Production Act to produce medical equipment, they ignore the fact that the United States has lost manufacturing skills and capabilities, as well as entire sub-supplier networks. The current reshoring discussion similarly avoids the underlying financial reasons why the United States has offshored so much production and taken the path of deindustrialization. If these reasons are not adequately addressed with new policies, we should not expect widespread or permanent reshoring of production to the United States once the crisis is over. This applies to medical supplies as well as to other critical industries. As a result, America will become even more vulnerable when the next global crisis—natural, medical, political, or military—hits.

Offshoring took place in the context of extreme U.S. financialization, and was pursued without regard to America’s long-term health or wealth. This article will identify the key drivers of offshoring and briefly describe the principal domains in which we need to develop new policy tools for onshoring supply networks.

Offshoring Since 2000

Between 2000 and 2007, the United States lost 3.4 million manufacturing jobs, about 20 percent of its total. It lost a further 1.5 million manufacturing jobs between 2007 and 2016. These rapid losses were unique in American history, and without parallel in other major Western countries. A widespread explanation, championed even today, is that the employment collapse occurred because of automation. This argument holds that U.S. manufacturing became so technologically advanced, replete with robots, that it no longer required a large workforce. But in reality, U.S. manufacturing labor productivity was stagnant or even declining during this period. And U.S. manufacturing output was stagnating or declining as well, making up a smaller and smaller portion of the total American economy. Further, research on other economies has consistently shown that production facilities that employed more robots also employed more, not fewer, people.

In a sort of bad dialectic, too many American economists simultaneously held the contradictory points of view that America retained advanced manufacturing, but also that manufacturing was never coming back, without empirically investigating either idea. In their minds, they didn’t need to: in a circular logic rooted in beliefs about a post-industrial future, the more the United States deindustrialized and turned to services, the more advanced it was seen to be. Greater China held very different beliefs about the importance of manufacturing, and pursued mercantilist policies to achieve the goal of attracting offshored American jobs and developing its own export industries.

The sharp decline in domestic manufacturing employment coincided with China’s accession to the WTO. This meant that U.S. firms could offshore production to China without fear of future tariffs on reimports, and it gave Chinese firms reliable access to U.S. markets. The biggest American corporations, in particular, accounted for a sizable portion of the offshoring of supply networks and the drop in U.S. manufacturing employment.3

Production of medical goods was offshored, as is now clear, but this same pattern occurred across advanced manufacturing sectors. For instance, a 2015 study by economists Susan Houseman, Timothy Bartik, and Timothy Sturgeon found that computer and semiconductor manufacturing had shifted from the United States to Asia, “both overall and within the most important and technologically demanding product types from a manufacturing perspective.”4 For instance, components of the new server farms supporting the internet in the United States were largely made abroad, even if they were still purchased from American brands. The study found that “the shift of server manufacturing to outside the United States does not mean that American-branded server companies are losing global market share, only that the United States is losing ground as a location for server manufacturing.” In other words, these companies were offshoring.

Offshoring was sold to the American public as a win-win. In fact, this was the title of a McKinsey Global Institute 2003 report which concluded, “Offshoring is as beneficial to the United States as it is to the destination country.”5

Economics provided the “scientific” rationale and intellectual cover for this view. Larry Summers, who has since moved to a more moderate approach, was a leading promoter of this consensus point of view, explaining to PBS in 2001:

Globalization is not about helping rich people at the expense of poor people. . . . I think there will be more jobs for Americans, better prices for American consumers, and a safer and more secure world because China is being admitted to the WTO. . . . There’s a basic cost of open markets, such as somebody losing a job, which is particular or very obvious, [but] the benefits are much less clear. Who said on Christmas Day, “Gosh, thanks, without open markets I would have been only able to buy half as many toys for my kid?” I think economic internationalism is the best possible forward defense of our interests, [such as] combating infectious diseases in the countries where they first start to spread.6

As the crisis has made clear, however, these views about the largely cost-free benefits of free trade are really little more than religious enthusiasms—although expressed through the aesthetics of mathematical modeling rather than hymns and encyclicals.

For some Americans—namely those working in finance, consulting, and the “new economy”—there was no trade-off at all. Their earnings and company profits rose as manufacturing jobs were offshored. U.S. firms engaged in global labor market arbitrage to find the cheapest labor inputs. In contrast, former U.S. factory workers were shunted off to poorly paid service jobs, or left with no jobs at all. In the end, it turned out that this specific version of globalization actually did mean helping rich Americans at the expense of poor Americans.

Offshoring, therefore, clearly has a strong class-conflict component, even if this has never been acknowledged by economists. Michael Lind sees it as part of a “new class war,” where Western managerial elites were “able to escape from their mid-twentieth-century social contract with national workers by offshoring production.”7

Beyond its distributional impact, offshoring caused the United States to lose several critical capacities that are needed if America ever wants to reshore production: (1) production and production engineering skills and capabilities, (2) networks of suppliers and sub-suppliers of components and sub-systems needed to ensure the production of final goods, and (3) production management skills.

Since 2000, we have trained a whole generation of our best managers and CEOs in how to destroy production capacity. We lack business leaders who know how to build it. We replaced our business heroes like Thomas Edison, Thomas J. Watson, and George Eastman with the likes of Jack Welch, Jeff Bezos, and James McNerney. America now has the world’s best management talent only if your aim is to offshore and financially engineer. It has very little management talent for building, producing, and physically engineering.

Financial Incentives for Offshoring

Financial market and management incentives drove the great offshoring forward during the last few decades, as the United States engaged in an orgy of financial engineering. Stock price, or more politely, “shareholder value,” became the idol to which everything had to be sacrificed. The extreme financialization of the U.S. economy was accompanied by short-termism, lax antitrust enforcement leading to market concentration, and the preference for stock buybacks and dividends over productive investment—all of which accelerated the flight from domestic production. Markets and shareholders rewarded corporations for offshoring and hollowing out domestic productive capacities, with no concern about the externalities affecting the United States as a whole or even the company’s long-term ability to prosper. Meanwhile, the decline of unions meant that there was no countervailing labor power to slow these trends down. These were the push factors for offshoring.

Mercantilist practices of other countries—such as currency manipulation, production subsidies (and outright bribes), market access conditioned on technology transfer, and cheap nonunionized labor—provided the pull factors.

Accounting measures, and the way bonuses are calculated, provided a further rationale to end physical production and offshore. Managers receive bonuses based on short-term stock performance. This can be manipulated more easily through stock buybacks and dividends than through long-term productive investments in manufacturing. In addition, moving assets such as capital equipment and whole factories off the balance sheet improves the company’s return on assets (ROA). The same logic is true of “lean” manufacturing. Holding very low inventories, and offloading the risk and cost to suppliers, makes a company seem leaner and more efficient—until a small disruption in supply turns out to be devastating.

There are subtler costs and disadvantages to offshoring production that standard accounting metrics don’t capture. Offshoring means giving up innovation that arises from improvements in manufacturing processes over the long term, including entirely new production technologies, which, in turn, enable the creation of new products and categories.

Offshoring and outsourcing reduce the corporation’s long-term ability to innovate even in its core competencies. Apple and Intel, working together, are already two years behind Huawei on 5G phones. The United States is even further behind in other critical telecommunications infrastructure which no American company knows how to develop, innovate, or produce.

Under the new economy that has come to dominate America, old-school manufacturing conglomerates that led innovation in the past, such as DuPont, were broken up, or in GE’s case under Jack Welch, turned primarily into a finance company. IBM, once one of the finest engineering and technology companies the world has ever seen, is now celebrated as a success because it managed to transform itself into a pale imitation of Tata Consultancy Services (employing the same Indian workforce instead of American engineers). Domestic supply networks withered and died. Sub-suppliers went out of business. Skills, knowledge, and capabilities were lost, perhaps forever.

Boeing epitomizes the current American way of doing business. In 1997, Boeing was allowed to merge with McDonnell Douglas, creating a domestic monopoly, one focused on finance over engineering. It outsourced 70 percent of construction of the 787, primarily to companies in other countries. One of the most technologically advanced components of the plane, the wings, went to Mitsubishi, with Boeing sharing its most prized technology in exchange for a few billion dollars in hypothetical savings. (It is certainly possible that Japanese industrial policymakers helped to financially incentivize this move to Japan.)

Since 2013, Boeing has spent $43 billion on stock buybacks. Yet it shut down its first attempt to develop an entirely new aircraft to replace the 737, in order to avoid spending a comparatively paltry $7 billion. The company decided instead to foist new engine technologies on the fifty-year-old 737 frame. This unwieldy combination was repackaged as the 737 MAX, a plane that has trouble flying without software corrections (the development of which software was at least partially outsourced to India).

Even before the two crashes of the 737 MAX, Boeing’s production problems were becoming well known. Nonetheless, the stock market loved Boeing’s corporate direction away from aerospace engineering and into financial engineering, seemingly oblivious to where it was all heading. Matt Stoller, who has closely tracked Boeing’s problems, writes, “one of the weirder aspects of the Boeing scandals is that the corporation has had significant operational difficulties for 20 years, and yet the stock has done quite well.”8 This disconnect between stock price and operational skills finally came to an end with a crash, actually two crashes, costing hundreds of lives and tens of billions of dollars to fix (if the 737 MAX can be fixed at all).

Failed Attempts at Reshoring

In 2012, Apple attempted to move production, or at least assembly, of the Mac Pro back to Texas. It found it couldn’t domestically source a tiny screw needed for the Mac, indicative of the challenges of trying to produce things in the United States. According to the New York Times account of the fiasco, in China Apple could turn to any number of factories that could produce custom screws.9 Texas had almost none. Further, as said by none other than CEO Tim Cook himself: “In the U.S., you could have a meeting of tooling engineers and I’m not sure we could fill the room. In China, you could fill multiple football fields.” Eventually, Apple had to import the screws from China, and this was merely one production bottleneck among many. In June 2019, Apple announced that the next version of the Mac Pro will be produced in China.10

If even the mighty Apple can’t source simple things such as screws in the United States, or find enough tooling engineers, we shouldn’t expect one company to solve what is now a systemic problem. It is not rational to expect individual CEOs to expose their companies to the higher costs and bottlenecks that come with reshoring. We should not expect the market to fix this suboptimal Nash equilibrium problem, either, nor can a one-time injection of taxpayer’s money, however large. These coordination, demand, and financial problems are just too staggering for even our most powerful corporations to solve by themselves.

These are the obstacles we have to acknowledge if we want to re-shore production: We lack the necessary skills from basic shop floor production all the way to management of a productive corporation. We lack the thick supplier and sub-supplier networks that are needed to make domestic production feasible. And we have created a whole set of incentives that will severely punish any CEO or CFO who tries to invest in long-term production capacity in the United States.

In order to change paths, we now need focused and sustained efforts in several interlinked domains: demand, skills, and incentives (financial and managerial). These all need to be coordinated to bring about a long-term solution. Our proposals go from the simple, which can be readily added to already proposed legislation, to the complex, which require new thinking altogether.

Ensuring Demand For Domestic Products

To encourage reshoring, it is critical to ensure demand for domestic production. Without demand for domestically produced products and components, no company would invest in building up production capacities, and no workers, managers, and investors will invest in acquiring and honing the necessary skills. This demand cannot be a one-time spike. For long-term reshoring and the rebuilding of our production facilities and innovation capabilities, all economic actors need to know that an elevated level of demand is here to stay.

Government purchasing. Government purchasing—at the federal, state, and local level—should have a “Made in America” component, and not just for medical goods and pharmaceuticals. There was some discussion, in late March 2020, of a “buy American” executive order focused on the health care supply network, which apparently withered in the face of industry opposition. Not only should this executive order be revisited, it should be extended to other sectors and industries. Companies that are given federal aid as part of the Covid-19 response should also be required to move some (significant) percentage of their sourcing and production back into the United States. This can be done either in outright procurement programs at all levels (local, state, and federal), as well as in authorizing reimbursement programs, such as Medicaid.

Disclosure of supply network sources. A key challenge is defining what “Made in America” means. For example, one of us just bought a DeWalt drill, believing the company slogan that it is “Proudly American.” We soon found out, however, that our new Proudly American drill was “Made in China.” It is clear that better measurement and disclosure is necessary. Industries should disclose how much of critical production actually takes place in the United States. With the right branding, many consumers would be willing to pay a premium on such products, both for quality and moral reasons (e.g., fair trade). Pending Senate proposals calling for this disclosure in pharma production and sourcing should be extended to other industries. Disclosure and better information can by itself be impactful, as has been widely documented.

Aggregate and map demand. Sometimes the issue is not a lack of absolute demand, but of matching demand with supply, especially over a large geography with multiple local markets like the United States. Government can solve this problem relatively easily by mapping out and aggregating demand and making this information public. Indeed, just highlighting that there is still significant demand to be tapped in the United States will change the investment rationale of all economic actors.

Infant industry protection. The United States currently runs a large trade deficit in even its most advanced technologies, which leave for Greater China almost as soon as they are invented at U.S. universities. Infant industry protection can create the demand required to grow these industries.

Reskilling the Domestic Workforce

U.S. workers are often criticized by U.S. corporations for lacking the necessary production skills. But given how much of production has been offshored, how can workers be expected to retain these skills, and why would it be rational for them to do so? Apprenticeship programs could be an answer, but despite the hype, they are pitifully small. Instead, we propose:

Diffusion of skills. Thanks to the Department of Defense, the Manufacturing Extension Partnership programs, and initiatives such as the Advanced Manufacturing Partnership, the United States has actually saved pockets of manufacturing skills and resources. Nevertheless, they exist in isolation, and it is difficult to widely diffuse their knowledge and skills. The country needs to find a way to diffuse those skills quickly.

Innovation intermediaries. These are institutions that can translate academic research into practical advice for manufacturing. They are widespread in other countries, such as the Fraunhofer Institutes in Germany, but not in America. The U.S. attempt to create innovation intermediaries here, “Manufacturing USA,” offers some lessons, both of success and failure, and a possible way forward.

Financial and Managerial Incentives

The economy needs broader, systemic changes in terms of investor and managerial incentives. Currently prevailing incentives tend to favor short-termism, nonproductive investments, and offshore over domestic supply networks. Return on assets and other accounting metrics have proven to be inadequate, even harmful. The theme of these broader reforms is that the market should become a means to an end—prosperity for all Americans—rather than an end in itself.

Managerial incentives. Executive contracts should be tied to long-term returns, rather than short-term fluctuations in share prices. In addition to eliminating one rationale for buybacks, this long-term perspective would internalize some hidden costs of offshoring, namely the lack of future innovation in manufacturing. It would also help internalize some of the benefits of producing in-house that only come with time, such as the downward-sloping cost curve that comes with learning. For instance, during World War II, Boeing experienced a 28 percent decline in the cost of producing Flying Fortresses each time output doubled.11

Investor incentives. The financialization of the U.S. economy since the 1980s has seen finance turn away from its purported role of funding the real economy to one of exploiting the real economy in order to generate outsize returns. This has led to many negative externalities for America. How to rechannel investment to productive growth is a vast topic. It includes stronger antitrust enforcement, and the need to prevent “war profiteering” by Wall Street. But the American economy will require more explicit industrial policy considerations, beyond just ensuring markets are competitive, so that the country isn’t pushed out of entire industries, as has happened with telecommunications hardware.

New sources of manufacturing financing. How can loan issuers be incentivized to channel loans to manufacturing, the way they have been for mortgages and student loans? There are many promising proposals, such as the pending SBA Reauthorization Act. But we still need ways to improve the skills and knowledge of loan officers when it comes to financing manufacturing.

Avoiding moral hazard in financing. The crisis may lead to new sources of funding for reshoring production and manufacturing, which is a good thing. The risk here is one of moral hazard. The funding could be used to create monopolies, favor incumbents, and lead to general stagnation because of the implicit “soft budget constraint.” Other countries—such as Korea, Germany, Taiwan, and Israel—offer models for how regulation and competition can be built into new funding provisions to avoid moral hazard. Funding needs to be made into a repeat game, not just because it gives firms certainty, but also because it gives governments leverage to ensure that the funding will be used for previously agreed-upon goals.

Beyond the Covid-19 Crisis

With U.S. policymakers trying to figure out which proposals to prioritize in rebuilding the disrupted medical supply network, we urge them to look at the larger issues surrounding offshoring, and to find ways to permanently reshore more production to the United States. America remains vulnerable across its supply network in the many industries that have been offshored, not just pharma and medical goods.

It is unclear if the Covid-19 crisis will reshape American production and lead to more onshoring of supply networks. The economic factors that gave rise to offshoring remain in place, meaning that once the crisis is over, without new policy approaches, the United States is unlikely to reshore production, aside from face masks and ventilators. Even these may just be stockpiled rather than produced here going forward. But there are many other strategically important “hardware” sectors besides medical goods in which the United States has lost manufacturing capabilities.

The rationale for onshoring is not just strategic; it is also economic. Reshoring is essential to ensuring the possibility of shared prosperity for all Americans. Offshoring has benefited a handful of corporations, consultants, and asset owners. Redistribution efforts such as a tax on billionaires or universal basic income can only partially offset the vast inequality that has resulted, and these measures won’t create fulfilling careers for those who are afflicted. America’s widespread abandonment of manufacturing and related innovation in terms of production has had dire economic consequences. The United States risks losing out on entire new industries and products which stem from breakthroughs in manufacturing.

The Covid-19 crisis has exposed these insidious long-term economic trends that have impacted the health and wealth of Americans. We don’t know how many Americans will die from Covid-19, but we do know that “deaths of despair” (suicide, drug overdose, and alcoholism) claim the lives of hundreds of thousands of formerly working-class Americans’ lives every year.12 This curve of “deaths of despair” has been steepening since the turn of this century, coinciding with the collapse of manufacturing employment. Reshoring supply networks can help flatten this curve of despair, as well as the coronavirus curve, positioning America to be better prepared for pandemics, wars, and the unforeseen challenges of the future.

This article originally appeared in American Affairs Volume IV, Number 2 (Summer 2020): 6–17.


1 Yuki Noguchi, “Not Enough Face Masks Are Made in America to Deal with Coronavirus,” NPR, March 5, 2020. Source for the Kimberly-Clark fact is Mike Bowen, EVP of Prestige Ameritech, a remaining U.S. producer of facemasks.

2 We use “supply networks” rather than “supply chains” because (1) it is the correct description of how production really happens, and (2) the term implies coordination, while “chains” imply command and control.

3 Christoph E. Boehm, Aaron Flaaen, and Nitya Pandalai-Nayar, “Multinationals Offshoring, and the Decline of U.S. Manufacturing,” Working Paper 17-22, Center for Economic Studies, U.S. Census Bureau, 2017.

4 Susan N. Houseman, Timothy J. Bartik, and Timothy Sturgeon, “Measuring Manufacturing: How the Computer and Semiconductor Industries Affect the Numbers and Perceptions,” Measuring Globalization: Better Trade Statistics for Better Policy, vol. 1, Biases to Price, Output, and Productivity Statistics from Trade, ed. Susan N. Houseman and Michael Mandel (Kalamazoo, Mich.: W. E. Upjohn Institute for Employment Research, 2015), 151–93.

5 Diana Farrell et al., “Offshoring: Is It a Win-Win Game?,” McKinsey Global Institute, August, 2003.

6 Lawrence Summers, “Commanding Heights” (transcript), PBS, April 24, 2001.

7 Michael Lind, “The New Class War,American Affairs 1, no. 2 (Summer 2017): 19–44.

8 Matt Stoller, “Airplanes and Accounting Games: The Coming Boeing Collapse?,” BIG, December 19, 2019.

9 Jack Nicas, “A Tiny Screw Shows Why iPhones Won’t Be ‘Assembled in U.S.A.,’” New York Times, January 28, 2019.

10 Cade Metz, “Apple to Manufacture New Mac Pro Computer in China instead of U.S.,” New York Times, June 28, 2019.

11 Kazuhiro Mishina, “Learning by New Experiences: Revisiting the Flying Fortress,” Learning by Doing in Markets, Firms, and Countries, ed. Naomi R. Lamoreaux, Daniel M. G. Raff, and Peter Temin (Chicago: University of Chicago Press, 1999), 145–84.

12 Angus Deaton and Anne Case, Deaths of Despair and the Future of Capitalism (Princeton: Princeton University Press, 2020).

Sorry, PDF downloads are available
to subscribers only.


Already subscribed?
Sign In With Your AAJ Account | Sign In with Blink