On Domestic Sourcing

There is growing recognition across the American political spectrum that neoliberal globalization has been a disaster for the economy and people of the United States. During the COVID-19 pandemic, Americans discovered the extent to which the U.S. is dependent on the industry of China, a hostile great power, for basic medicines and medical equipment, among many other manufactured goods.¹For U.S. medical imports from China, see Karen M. Sutter, Andres B. Schwarzenberg and Michael D. Sutherland, “COVID-19:  Chinese Medical Supply Chains and Broader Trade Issues,” CRS Report, Congressional Research Service, April 6, 2020. For reasons of public health and national security, the reshoring of essential supply chains to American soil is imperative. Combined with other tools of national industrial policy, local content requirements (LCRs) provide a direct, market-friendly, and effective method to achieve the goal of national independence in strategic industries.

LCRs are among the numerous tools in the toolkit of industrial policy that can be used in the service of strategies of economic development. LCRs require designated final goods to use a specified percentage of domestic value-added or domestically sourced intermediate products. LCRs can apply to all firms or be limited to government suppliers. The term “localization measures” has been used as a broader category that includes, in addition to classic LCRs, other measures such as requiring that data be stored and analyzed locally, conditioning bailouts, government contracts, export financing, as well as tax, tariff, and price concessions on local sourcing, and mandating that products be tested locally.²Karan Bhatia et al., “Why General Electric is localizing production,” Vox CEPR Policy Portal (June 21, 2016).

LCRs have been widely used by developing and developed countries alike, including Brazil (the oil and gas industry), Russia (automobiles), South Africa (wind energy), India (solar power) and Taiwan (refrigerators and televisions). They have been deployed both by nation-states and multinational trade blocs. In addition to state and local laws, the U.S. has a number of federal laws mandating local content for defense procurement and other purposes, often with waivers at the discretion of officials.³Kate M. Manuel et al., “Domestic Content Restrictions:  The Buy American Act and Complementary Provisions of Federal Law,” Congressional Research Service (September 12, 2016). The U.S.-Mexico-Canada Agreement (USMCA) negotiated by the Trump administration to replace the North American Free Trade Agreement (NAFTA) raises the North American local content of automobiles eligible for special treatment under the treaty from 62.5% under NAFTA to 75%.⁴David Lawder and David Shepardson, “Automakers to pay $3 billion in new U.S. tariffs under USMCA: budget estimate,” Reuters (December 18, 2019).

The Prohibitionist Campaign against LCRs

As one student of the subject has noted, “Despite their importance, economic evaluations of these policy measures are surprisingly limited.”⁵Francisco M. Veloso, “Understanding local content decisions: economic analysis and an application to the automotive industry,” Journal of Regional Science 46, no. 4 (2006), 747. The most influential early study was G.M. Grossman, “The Theory of Domestic Content Protection and Content Preference,” The Quarterly Journal of Economics 96, no. 4 (1981). Many of the studies of LCRs are polemical in nature, employing the tone of temperance movement pamphlets denouncing the evils of alcohol. For instance, a 2016 paper from the Peterson Institute for International Economics (PIIE) speaks of “the United States’ notorious Buy American statutes.”⁶Cathleen Cimino-Isaacs and John Zilinsky, “Local Content Requirements: Backdoor Protectionism Spreading Under the Radar,” Peterson Institute for International Economics (PIIE) (July 22, 2016). A 2009 PIIE analysis is entitled “Buy America: Bad for Jobs, Worse for Reputation.”⁷Gary Clyde Hufbauer and Jeffrey J. Schott, “Buy America: Bad for Jobs, Worse for Reputation,” Peterson Institute for International Economics (PIIE) Policy Brief Number PB09-2 (2009). Notorious… reputation… The sinners must be shamed into sobriety.

Typical of this mentality was an announcement from the Obama administration’s office of the U.S. Trade Representative in 2013: “Last week’s CTI meeting marked the first time the economies of APEC collectively addressed LCRs to gain a better understanding of how they distort trade and investment flows [emphasis added].” The claim that LCRs distort “natural” markets is also found in a 2015 OECD study of LCRs, which claims: “Overall, LCRs distort input markets and potentially inhibit innovation by removing access to technologically advanced inputs, undermining efficiency gains from lower value chains [emphasis added].”⁸Susan Stone et al., “Emerging Policy Issues: Localisation Barriers to Trade,” Organisation for Economic Co-operation and Development (OECD) Trade Policy Paper 180 (January 5, 2015). The underlying assumption is that the normal and sound condition of the world economy is a single market, with no “distortions” in the flows of goods, services, workers, or money across borders. But of course, such an economy never has existed and never will.

This dogmatic assumption vitiates most studies of the alleged harmful effects of LCRs. If LCRs raise any costs to consumers or producers previously reliant on imports as inputs, this is taken to prove the existence of a harmful “distortion.” But the very purpose of LCRs, along with other import substitution measures, is to “distort” trade and production by replacing an existing pattern with a new pattern. Whether the price of localizing a particular industry in the interest of national security, economic diversification, or economic growth is worth paying is a values-based policy judgment which economists qua economists are not qualified to make.

Some critics of LCRs will argue that, even if it were legitimate for governments to promote particular industries, they should do so by means of “horizontal” measures, such as creating a business-friendly environment, encouraging corporate social responsibility, expanding training, and investing in infrastructure.⁹Gary Hufbauer et al., “Local Content Requirements:  Report on a Global Problem,” Peterson Institute for International Economics (PIIE) (October 7, 2013). See also Gary Clyde Hufbauer and Jeffrey J. Schott, Local Content Requirements: A Global Problem (Washington, D.C.: Peterson Institute for International Economics (PIIE), 2012). But if targeted policies to promote the localization of specific industries within a country are “distortions” of the global market, then broad policies to encourage localization of industry in general must be “distortions” as well.

It is true that protectionist policies can be promoted by self-interested firms and industries. But it is also true that policies that encourage offshoring of industry or indifference to foreign mercantilism are promoted by selfish interests that benefit from them. Special interests and national interests are aligned in the case of policies that improve and diversify the domestic industrial base, but they are divergent in the case of policies that allow multinationals to profit by dismantling the nation’s manufacturing ecosystem or allow foreign mercantilist regimes to rob national producers of domestic and global market share. In the words of Alexander Hamilton: “To preserve the balance of trade in favor of a nation ought to be a leading aim of its policy. The avarice of individuals may frequently find its account in pursuing channels of traffic prejudicial to that balance, to which the government may be able to oppose effectual impediments.”¹⁰Alexander Hamilton, The Continentalist No. V, in Henry Cabot Lodge, ed., The Works of Alexander Hamilton, Federal Edition No. 1 (New York:  G.P. Putnam & Sons, 1904), 267-9.

Debating Alternatives

Because neoliberals and libertarians are opposed on principle to national industrial policy, they have as little to say about the appropriate mix of LCRs, tariffs, or subsidies as crusading prohibitionists have to say about the relative merits of beer, wine, and vodka. But there is already an interesting debate underway among adherents of the school of political economy that Robert D. Atkinson and I have dubbed “national developmentalism.”¹¹Robert D. Atkinson and Michael Lind, “National Developmentalism: From Forgotten Tradition to New Consensus,” American Affairs (Summer 2019). Members of this school would agree that countries, as well as firms and individuals, are actors with their own particular interests in the global economy, and that national industrial policies are therefore legitimate. This debate—a pragmatic one, over the strengths and weaknesses of LCRS in particular cases and circumstances compared to the merits and disadvantages of other industrial policy measures—is the debate in which policymakers ought to be engaged: not over whether to have a policy at all, but what the policy ought to be.

Tariffs and quotas are one possible alternative to LCRs. They come in two kinds: temporary measures, used as bargaining chips to pressure trading partners into opening up their own markets, and permanent measures, intended to localize certain kinds of industrial production in the national interest, regardless of the behavior of trading partners. If the goal is to retaliate against particular instances of foreign mercantilism or to pressure trading partners into reciprocal trade liberalization, then temporary tariffs or quotas may be more effective than LCRs, which tend to be permanent.

If, however, the goal is to nurture vital infant industries, to preserve mature industries that might otherwise be lost to import competition, or to repatriate supply chains lost to private firms’ offshoring strategies, then LCRs are arguably superior to tariffs and quotas. In this case, the goal of tariffs, quotas, and LCRs are the same—the permanent localization of certain industries or supply chains. But the tariffs and quotas are indirect measures that merely incentivize localization without guaranteeing it. It is simpler to mandate localization directly by means of LCRs.

If LCRs are justified by a long-term national or regional bloc development strategy—not short-term tit-for-tat trade diplomacy—they should apply equally to all trading partners, including military allies, neutrals, and rivals alike. In the absence of national security concerns, the national identity of the owners and managers of multinational firms compelled by LCRs to produce in national or bloc markets would be a matter of indifference, and foreign firms would be free to repatriate their profits from in-market production.

Compared to tariffs and quotas, then, as a policy to create or preserve permanent domestic supply chains, LCRs are more direct, simpler to administer, and potentially less diplomatically contentious.

Subsidies are another alternative approach worth considering.  But they should be used with caution. Government subsidies to favored industries—like tax breaks for domestic production of certain goods—might achieve the same goal as LCRs, but at the price of greater misallocation of investment and labor. (For the purposes of this discussion, I will define government procurement policies as a kind of LCR, rather than a subsidy).

If countries are allowed to subsidize their national champion firms to compete in global markets, but forbidden to encourage or require in-market production for goods sold in their domestic markets, the result may be escalating subsidy wars that end in global gluts in the subsidized sectors. Farm subsidies have produced “mountains of butter” and “lakes of milk” in the U.S. and EU, which are then sometimes dumped on world markets. Because of subsidization, domestic production of targeted goods would be ensured, but at the cost of oversupply, overinvestment, and wasteful misallocation of resources and labor.

In contrast, widespread reliance by countries and blocs on LCRs, rather than subsidies, to promote strategic industries would not encourage gluts. If most major countries or blocs were to use LCRs, no country could hope to use a dumping strategy to drive foreign firms out of their home markets, and in-market production, undistorted by subsidies, would not exceed in-market demand.

Another problem with subsidies is that conventional approaches—tax breaks for targeted industries, for example—reduce tax revenue, which may need to be made up by other taxes, a price that may be worth paying in the interest of national security, job creation, or productivity growth. Of greater concern should be subsidies in the form of preferential credit policies by public or private banks nudged by national governments. If used in excess, as in Japan and China, these can result in nonperforming loans in zombie financial institutions. Here, too, compared to subsidies to favored industries, LCRs may be a superior tool of industrial policy.

It is not enough for policies to be effective; they must also be politically sustainable. It is much easier for opponents to denounce tariffs as taxes than to campaign against “Buy American” policies—although consumers and firms reliant on imported inputs may suffer similar adverse effects. Unlike subsidies, LCRs require no legislative appropriations of funds, alterations of the tax code, or special credit policies. Once they are enacted, LCRs are relatively invisible to consumers and taxpayers, but they are likely to be defended against repeal by the industries they benefit. This is a feature, not a bug, if the goal is to permanently localize part or all of those industries within national borders.

Consistency requires that any policies that incentivize firms to invest in one country rather than another—including the provision of law and order, public education, and infrastructure—should be defined as “distortionary” and prohibited by international trade treaties. As members of the national developmental school will point out, “distortionary” policy is unavoidable in policymaking. Rather than neglect some policies for being “distortionary,” we should assess them on the merits of their “distortion.”

Selective, strategic reshoring, not complete national autarky, should be the goal of LCRs. As noted above, local content requirements can be imposed by trading blocs like NAFTA/USMCA as well as by nation-states like the U.S. The most critical supply chains, however, should be located within national borders, where they are safe from interdiction or disruption. Replacing American dependence on Chinese producers with dependence on producers in India, Vietnam, or other Asian countries might reinforce an anti-Chinese diplomatic coalition led by the U.S., but it would do nothing to increase America’s own economic independence and resilience. It is also a mistake to allow U.S. industry to become too dependent on factories in nearby Mexico, an unstable, crime-ridden country with a corrupt and unreliable government.

Of the potential tools that American policymakers can use to reshore strategic supply chains, LCRs have fewer drawbacks than other policies—including tariffs, quotas, and subsidies. Their virtue lies in their simplicity.  They do not require exercises of vast discretionary authority by the executive branch. They do not require targeted allocations of public money. Nor do they require technocrats to craft detailed policies to offset specific alleged “distortions” or “externalities.” The federal government simply formulates a list of which goods must be manufactured in the U.S. and allows private firms and private investors, both American and foreign, to figure out how to meet the requirements.

The LCR is a policy tool that should not be neglected. Beginning with pharmaceuticals, personal protective equipment, and critical inputs used in defense manufacturing, Congress should authorize the executive branch to designate and enforce U.S. local content requirements and to withdraw from any existing international treaties that prevent the U.S. from exercising this essential right of economic self-defense.