What AMLO’s Endorsement of ‘Import Substitution’ Reveals About Future North American Trade

Introduction

Following the North American Leaders’ Summit (NALS) on January 10, 2023, in Mexico City, Mexican President Andrés Manuel López Obrador delivered a media statement on the commitments made with his U.S. and Canadian counterparts. He declared:

“We agreed to strengthen our economic and trade relations, for which a joint committee will be created for planning and import substitution in North America; to aim to be increasingly self-sufficient; and to make the development, cooperation and the well-being of all the countries on our continent a reality.”[1]

By characterizing widely touted plans between the U.S., Mexico, and Canada for cooperation in chip, electric vehicle (EV), and other areas of high-tech production as “import substitution” — a pejorative term that fell out of vogue several decades ago, particularly in Latin America — López Obrador surprised his tripartite meeting partners. Yet he may have been right in using the term, defined as “the idea that blocking imports of manufactured goods can help an economy by increasing the demand for domestically produced goods,” to describe their plans for enhancing regional competitiveness.[2]

This brief examines the concept of import substitution in light of the 2023 North American Leaders’ Summit and its outcomes, as outlined in a joint statement titled “Declaration of North America” (DNA). Ultimately, the document’s provisions on competitiveness appear promising if faithfully implemented — but the risks of regionalist policies are many, and officials of all three countries would do well to remember the lessons of the past as they work to implement them.

The Past Practice of Import Substitution

López Obrador’s use of the term “import substitution” likely did not reflect the Mexican government’s position on the most effective means of strengthening the competitiveness of North America or even the Western Hemisphere, according to Kenneth Smith Ramos, Mexico’s chief trade negotiator for the U.S.-Mexico-Canada Agreement (USMCA).[3] López Obrador’s economic training, however, dates back to when import substitution was Latin America’s main development strategy. Thus, the confusion generated probably results from the negative reputation of import substitution as practiced in most of Latin America in the 1950s, 1960s, and 1970s. Those practices, which were based on the recommendations of Argentine economist Raúl Prebisch,[4] were viewed as highly protectionist and economically unwise; they involved creating tariff and quota barriers to foster the growth of domestic manufacturing industries and create jobs, reduce imports (as well as trade deficits), and facilitate technology transfer. That apparently was López Obrador’s own understanding of what the North American leaders were proposing. However, as Smith Ramos accurately observed, the practice of import substitution over time “did not yield the results that an increasingly globalized world economy demanded” and had been generally abandoned by the 1980s. Nor did Latin American import substitution ever progress much beyond national boundaries, despite multiple regional groupings such as the Central American Common Market[5] and the Latin American Integration Association (ALADI).[6

Mexico, for example, which had practiced import substitution for decades — resulting inter alia in closed markets, slow growth, expensive and poor-quality consumer goods, and uncompetitive industrial exports — largely abandoned it in the mid-1980s. In doing so, it joined the General Agreement on Tariffs and Trade (GATT) in 1986 and began to institute neoliberal economic policies, including significant tariff and nontariff barrier reduction and promotion of the private sector. This ultimately resulted in its joining with Canada and the United States in the North American Free Trade Agreement (NAFTA) in 1994.[7] López Obrador himself became a major opponent of this liberalization and built his career on resisting it. It should not be surprising then that this framework still appeals to him ideologically, even if the context is completely different and the proposals do not exactly reflect just such a strategy.

The emergence of the term during the 2023 NALS should not be surprising. Despite criticism of import substitution policies in Latin America by the United States and other developed countries that were supporting trade liberalization at the same time,[8] the practice in one form or another has been followed by virtually every major government during its period of industrial development. In the United States, for example, during the late colonial period through the 19th century (with a few exceptions), the federal government imposed high tariffs or quotas on imports of most manufactured goods in order to grow and protect the domestic manufacturing sector. (The practice primarily benefited the mid-Atlantic and New England states to the detriment of the agricultural South, which would have enjoyed lower foreign tariffs on agricultural exports instead of the higher tariffs imposed by such major markets as Great Britain and France in partial retaliation.)[9]

More recently, during the last quarter of the 20th century, Hong Kong, Singapore, South Korea, and Taiwan — Asia’s “four tigers” — used some import substitution techniques to boost employment and build domestic manufacturing industries in both industrial goods, such as steel and higher-tech semiconductors, and consumer electronics. (South Korea and Taiwan did this for a longer period than Hong Kong and Singapore.) They employed these techniques with the implicit consent of the United States, Japan, and the European Union. For a variety of reasons, the process worked better in Asia than in Latin America but ultimately proved impractical there as well.

Import Substitution in the Declaration of North America

The DNA, issued jointly by U.S. President Joe Biden, Canadian Prime Minister Justin Trudeau, and Mexico’s López Obrador after the NALS on January 10, did not mention import substitution explicitly. But as López Obrador’s media statement indicates, its section titled “Competitiveness” is, in substance, not so far removed from classic import substitution practices as some critics have suggested. It states:

We seek to deepen our regional capacity to attract high quality investment, spur innovation, and strengthen the resilience of our economies, recognizing the benefits brought by the United States-Mexico-Canada Agreement. To boost regional competitiveness, the three countries will seek to forge stronger regional supply chains, as well as promote targeted investment, in key industries of the future such as semiconductors and electric vehicle batteries, which will be critical to advance electric vehicle development and infrastructure. We will convene public-private dialogues and map out supply chains to address common challenges and opportunities.[10]

This focus on the USMCA region with regard to investment, innovation, and employment is welcome news for those who have feared that the Biden and López Obrador administrations have been tacitly endorsing the concept of import substitution at the national level, albeit in different forms.

Practically speaking, virtually all of the hundreds of existing free trade areas and customs unions, including that created by the USMCA parties, reflect a regional form of import substitution explicitly authorized with certain restrictions under Article XXIV of the GATT.[11] Under that article, trade among parties to an agreement is conducted with lower, preferential tariffs compared to the “most favored nation” tariffs levied on goods from nonparty countries (hence the term “preferential trade agreements”). Economically and legally, there is thus nothing to prevent the practice of import substitution among two or more nations that are parties to a regional trade agreement, even though the term “import substitution” is seldom, if ever, used in that context. Arguably, the practice of import substitution on a regional level can distort trade to a lesser degree, particularly when the regional trade agreement includes both low- and high-wage participants and promotes efficiencies among complementary economies. Examples of such regional partnerships are the USMCA, the EU, and the Comprehensive and Progressive Agreement for Trans-Pacific Partnership.

Policy Impacts in the United States

The cornerstone of the Biden administration’s economic policy has been “Buy American, Invest American, Employ Americans.” The administration has sought (as did the Donald Trump administration) to reshore the production of semiconductors, autos, and some other manufacturing goods. Thus, both the Creating Helpful Incentives to Produce Semiconductors and Science (CHIPS) Act[12] and the inaccurately named Inflation Reduction Act (IRA)[13] focus primarily on substituting national (or in some welcome cases, North American) production of chips, EVs, and EV batteries, often to offset the current broad reliance on China for finished goods, intermediate products, and raw materials.[14]

This is not to say that such policies that aim to reduce reliance on the United States’ major adversary, China, are unwise or unjustified; it is just to note that they are a more refined form of traditional import substitution. (Whether the U.S.’ import substitution through discriminatory subsidies of domestic EV and EV battery production for imports from the EU, South Korea, and Japan is wise economic and national security policy is more problematic, given the retaliation that seems likely to result.)[15] It is unclear whether the full economic costs of the desired shifts in sourcing have been fully examined. These costs will depend on, in part: how rapidly private industry is required to implement the shifts; the extent of the subsidies considered essential by the U.S. government to bring the shifts about; and the changes in labor costs that will result from substituting American workers for workers in China and other Asian developing countries (hence the importance for the United States and Canada of including Mexican co-production in the equation).

Still, this likely extensive retreat from globalization (or reevaluation of it, in the words of U.S. Trade Representative Katherine Tai) will almost certainly continue in the United States and elsewhere. It is driven by many factors, including but not limited to: 1) the ongoing U.S.-China trade war with its penalty tariffs, 2) reshoring or nearshoring pressures encouraged by long Asian supply lines, and a desire for “just-in-case” rather than “just-in-time” supply lines, 3) recent disruptions caused by COVID-19 and extreme weather events, 4) growing pressures against long transport distances in response to climate change and the carbon neutrality goals embraced by multinational enterprises (MNEs), and 5) national security concerns in the U.S. related to goods such as semiconductors, EV batteries, personal protective equipment, and pharmaceutical products.[16]

At the World Economic Forum in Davos, Switzerland, in mid-January, Tai characterized the Biden’s administration evolving approach to trade by emphasizing:

“Enormous amounts of prosperity without an inclusiveness that comes with it [and] rising inequality in many different economies is driving a desire on our part to lead the conversation and to lead the thinking around what a new version of globalization might be, what a new economic world order might look like that builds on what we’ve just experienced, but more importantly, learns [a] lesson.”[17]

Even those who would prefer the current approach to globalization, including many nations and most MNEs, should be willing to admit that the U.S. and most other countries that might wish to reduce income inequality and improve access to the middle class will find it politically impossible to do so using tax and related government policies. One alternative that may work at least to some extent for the Biden administration, which it has already committed to attempt to do, is to bring back at least some of the manufacturing positions that were exported to China (and to a lesser extent, Mexico) over the past three decades. Implementing this change in the regional context of North America, rather than simply in the U.S., may thus be the most practical route.

Policy Impacts in Mexico

In Mexico’s case, the policies seem more pernicious. López Obrador’s reversion to the import substitution/statist policies of the 1960s and 1970s — especially in the energy sector, favoring state monopolies Petróleos Mexicanos (Pemex) and the Comisión Federal de Electricidad (CFE) over both domestic and foreign private industry — will make it much more difficult for Mexico to participate materially in the cooperative strengthening of North American industry and competitiveness as envisioned by the DNA.[18] López Obrador’s policies have seriously degraded the rule of law and worsened crime in Mexico, and his unpredictable actions, guided by personalistic and arbitrary decision-making, have alarmed the private sector. This is not likely to encourage new investment, especially in higher-value-added sectors such as those contemplated. (Mexico’s medium- and long-term development is not significantly strengthened by new toy factories in the border regions.) Needless to say, drug and gang violence, weak and ineffective police protection, and increased highjacking risks for truck shipments (which can be worth millions of dollars per container) do not build confidence in Mexico as a locus of high-tech manufacturing or co-production.

The indirect negative effects of López Obrador’s statist form of import substitution — along with the squandering of some $15 billion to $18 billion on a white-elephant oil refinery[19] and additional billions spent on an environmentally disastrous trans-isthmus railroad[20] — are likely to be counterproductive as well. The expansion of the high-tech sector in the U.S. and, hopefully, in Canada and Mexico depends on adequate sources of clean energy, particularly electricity and, at least for a long interim period, natural gas. These are, or will be, more widely available in many parts of the U.S. and Canada than Mexico. According to knowledgeable observers, new investment in Mexico depends on a supply of electricity that is reliable 24/7, reasonably priced, and clean. The CFE’s production meets none of these criteria now and is unlikely to in the foreseeable future, as privately financed solar and wind investors are squeezed out and new clean independent power plants (IPPs) may soon no longer be authorized by government authorities.[21] The CFE’s increasing reliance on Pemex’s dirty heavy fuel oil (HFO) for power generation will add to the concerns of potential new electricity consumers. Why would an MNE with a public commitment to achieving worldwide carbon neutrality by 2045 or 2050 invest in substantial manufacturing or service facilities or rare earth mining anywhere in Mexico, if it would act against its clean objectives for at least the next several decades? Meanwhile, the Mexican government’s efforts to nationalize the lithium-mining industry (a sector directly involved in the manufacturing of EV batteries) are also generating uncertainty among investors.[22]

Interestingly, López Obrador’s policies go against the emerging narrative that Mexico is poised to capture much of the industry that will be reshoring to North America, primarily due to its lower wage advantages and proximity to the United States. This narrative is now widely reported in the media and even touted by the López Obrador administration itself. It is not consistent with López Obrador’s actual policies, however.

Concluding Observations

Overall, the DNA’s language on competitiveness is to many observers, including this one, a welcome breath of fresh air, particularly if one can assume that Canada’s Trudeau will continue to be a strong advocate for the agreement — and he does not appear to need convincing. Trudeau is the only new left progressive among the three national leaders (balanced by an old left U.S. president and a Mexican doctrinaire populist).[23] The DNA appears to constitute a reaffirmation, at least in words, of the importance of the North American economy — rather than just the individual U.S., Mexican, and Canadian economies — and a commitment to the USMCA. (Whether it has any major significance for the rest of Latin America is a topic for another paper.) However, the substitution of regionalism for globalism may not be without its perils for the three North American nations, whether it is characterized as “import substitution” or “boosting regional competitiveness” or something else. It will be as challenging today as it was in the 1970s to achieve the intended results with such policies, including but not limited to the follow-up work necessary to create the promised Cabinet-level summit on semiconductors,[24] a reward to the entreaties of the Business Council of Canada, the U.S. Chamber of Commerce, and Mexico’s Consejo Coordinador Empresarial.[25]

Now as before, the risks of such policies are many. As the economist Douglas Irwin opined, “The lessons of past experience might temper one’s enthusiasm for the resurrection of such anti-import policies,” given their reputation for promoting inefficiencies in local industries, among other factors.[26] One of these lessons learned is that U.S. “Buy American” practices could ultimately take precedence over regionalist objectives, as they did in the country’s recent dispute with Mexico and Canada over the USMCA’s automotive rules of origin.[27] In the United States’ understandable desire to decouple key sectors from China for national security or competitiveness reasons, it may overreach, as it has with bogus “national security” tariffs and quotas on imported steel and aluminum. Ideally, the current focus on chips and EVs, as well as related areas such as telecommunications, will not be unduly broadened. The fact that the IRA contemplates imports from countries with which the United States has free trade agreements — more than 20 including Canada and Mexico — gives some basis for hoping that the country will take a rational approach to reshoring or nearshoring, although there is no guarantee.

That being said, the governments, businesses, and workers of the U.S., Mexico, and Canada have been closely tied economically for 28 years, particularly in the automotive, agricultural, and metals sectors. In 2019, combined U.S. trade with Canada and Mexico produced nearly $1.4 trillion.[28] The economies are highly interdependent; for example, some 40% of the value of Mexican exports to the United States comprises materials and components originating in the U.S.,[29] and trade between Canada and Mexico totaled an approximate $21 billion in 2020.[30]

For its part, Mexico would do well to revert to predictable government action and shift its focus back on improving its investment climate, producing clean, reliable energy, and more effectively counteracting domestic crime (This may be not likely before October 2024, when López Obrador leaves office, but hopefully soon after.) Without such a shift, Mexico will not be able to play a meaningful role in “boosting regional competitiveness” beyond increasing low-value investment in maquiladora-type production along its northern border, to the detriment of its U.S. and Canadian partners as well as Mexican citizens and businesses.

What about Canada? The country is the only USMCA party with an unequivocal commitment to North American regionalism, and existing Canadian policies are well reflected in the DNA (with the exception of milk imports). Moreover, one can argue that of the three partners, only Canada today stands ready, willing, and able to implement the DNA’s competitiveness provisions. One can hope that Canada’s influence on the United States and Mexico will continue, and that its government will avoid backsliding (as it did with its COVID-19-era border closings, which many both in Canada and abroad regarded as excessive and counterproductive because they exceeded those being imposed by most other jurisdictions).[31]

For most observers (and the three governments), “boosting regional competitiveness” is a much less pejorative and more positive term than “import substitution,” as the implementation risks for regional trade liberalization are less significant both politically and economically. Many administrative officials within the three USMCA governments are intimately familiar with the process as implemented in Latin America and Asia and even historically in the United States, and they will hopefully put this knowledge to good use as North American regional efforts move forward. Moreover, the “boosting regional competitiveness” approach is hardly confined to North America, as the process of globalization, which began well before NAFTA, continues to evolve.[32]

Finally, as former Canadian diplomat Colin Robertson recently observed:

“With abundant natural resources, a strong industrial base with pools of self-generated capital, and a market of half a billion people who, by global standards, are young and well-educated, the North American advantages are real. The business community is there. But just as with the original vision for NAFTA, political leadership is essential.”[33]

Let us hope that officials and politicians are up to the task of implementing the lofty and worthwhile undertakings embodied in the DNA. In my view, the evolution of globalization to a form that emphasizes regionalism would be a largely positive step for North America. But in any event, the “import substitution” approach as applied in the second half of the 20th century is best left for the history books.