A Case for a North American Common Tariff

Trade Creation and Regional Economic Convergence

In the mid-20th century, Jacob Viner proposed grading a trade agreement based on how much trade it "creates" versus how much it "diverts." An agreement is said to create trade if it augments efficient intraregional supply and, conversely, divert trade if it augments inefficient intraregional supply that displaces more efficient extra-regional options. In 1997, we published a report on the emergence of regional trade agreements² and pointed out that international trade theory and empirical evidence are ambiguous regarding their impact on trade creation or diversion. Research suggests that only the net outcome depends on the rules governing the agreements. We warned that, to prioritize trade creation, an agreement must adhere to three fundamental rules: first, comply with the letter and spirit of Article XXIV of the GATT, which essentially states that it should not increase trade protection among its members relative to the rest of the world; second, grant non-discriminatory treatment to investment from outside the region; and third, maintain effective dispute-resolution institutions. 

NAFTA and the USMCA contributed to more significant trade and investment, and although slight trade diversion effects were observed, the sustained growth of extra-regional exports indicates trade has also been created.³ Since 1994, annual intraregional trade growth averaged 6%, while investment grew at an annual average of 8%. The USMCA member countries (the Parties) registered positive extra-regional trade growth during the same period. Moreover, supply chain integration has generated synergistic growth across the region; for example, by 2012, each dollar of Mexican exports to the United States contained 40 cents of U.S. exports to Mexico⁴. The overall strong performance of the trade arrangement explains its widespread acceptance in Mexico despite early internal opposition. Many of its critics in 1990 now consider the USMCA a cornerstone of the country’s economic policy.

Dissipation of Trade Asymmetries in Mexico

Thirty-five years ago, it would have been difficult, if not impossible, to establish a common tariff without disrupting entire production chains in Mexico or generating substantial trade diversion. NAFTA negotiators recognized that Mexico’s economic liberalization was still incomplete and that the country needed time to eliminate four key trade asymmetries concerning the other two Parties: first, tariffs and non-tariff barriers; second, contingent protection and competition policy; third, foreign investment and intellectual property rights; and fourth, international trade and investment dispute resolution mechanisms.

When NAFTA became law, Mexico had elevated and distorted tariffs. Its average tariff was about 9%, more than three times higher than the average U.S. tariff. Additionally, high tariff dispersion caused abundant cases of negative adequate protection. Non-tariff barriers also engendered distortions; while import permits had been eliminated, a complicated system of quotas and regulatory barriers remained in place, administered in a discretionary and opaque manner. 

The prevailing system to prevent and penalize unfair trade practices was weak regarding regulation and administration. For example, dumping margins and injury determinations were calculated using procedures and regulations adopted from multilateral codes, but applied with little rigor and transparency. The country lacked a legal framework and an administrative body to promote competition, penalize collusion, and prevent the exercise of injurious market power.

Foreign investment was frequently subject to arbitrary and discriminatory treatment. The guiding principle of the foreign investment law could be summarized as follows: "Except what is permitted, foreign investment is prohibited." Also, intellectual property rights were not protected per international protocols, exposing investors to restrictions and quasi-expropriatory mandates, such as the forced transfer of intellectual property.

Generally, foreign investors had to litigate complaints against federal or local authorities in Mexican courts and could not access alternative dispute resolution mechanisms, including mediation and arbitration.

With NAFTA, Mexico reduced those trade asymmetries. The average tariff was reduced to less than 5%, and effective protection distortions were largely eliminated. Tariff convergence followed; by 2023, minimum tariffs in Mexico and the United States were identical for 40% of the tariff subheadings, which accounted for over 60% of bilateral trade.

Image Credit

Pedro Noyola and Jaime Serra

 

 

A new foreign trade law aligned the system against unfair international trade practices with multilateral disciplines. A new foreign investment law turned the guiding principle of the previous law on its head, so now: "except what is prohibited, foreign investment is deregulated." Additionally, industrial property and copyrights were afforded protection on par with those offered by the United States and Canada, and arbitration was strengthened as an extrajudicial dispute resolutionmechanism.

Economic Convergence

The reduction of trade asymmetries and NAFTA’s tariff elimination calendar also catalyzed the co-intgration of key macroeconomic variables, as attested by applying the Johansen cointegrationtest to evauate their relationship over time.⁵ When applied to industrial production in Mexico and the United States, between 1980 and 1993, the null hypothesis (no cointegration) was accepted; however, from 1994 to 2023, the null hypothesis was rejected (cointegration was statistically confirmed). Similarly, starting in 1994, the null hypothesis was rejected for inflation, short-term interest rates, and the volatility of the Mexican and Canadian currencies relative to the US dollar. The cointegration tests confirm that macroeconomic symbiosis began to emerge with the implementation of NAFTA.

When applied to industrial production in Mexico and the United States, between 1980 and 1993, the null hypothesis (no cointegration) was accepted; however, from 1994 to 2023, the null hypothesis was rejected (cointegration was statistically confirmed). Similarly, starting in 1994, the nullhypothesis was rejected for inflation, short-term interest rates, and the volatility of the Mexican and Canadian currencies relative to the US dollar. The cointegration tests confirm that macroeconomic symbiosis began to emerge with the implementation of NAFTA.

---

From NAFTA to USMCA

NAFTA promoted the convergence of trade policies and economic variables, but the agreement also accumulated distortions. For example, it contained restrictions on the trade of services, export, and import quotas, and, crucially, included rules of origin that reflected the average regional content of intraregional exports of 35 years ago. The negotiation of the USMCA allowed the parties to address the deficiencies of the original agreement and propose an improved integration model to create more trade and avoid diversion.

However, the original impetus of the USMCA was political, not economic. For starters, the United States government threatened to abandon the agreement if the "substantial" regional production criterion in the rules of origin was not made more restrictive so the United States could "export more to its partners and import less from them." Also, and most unfortunately, the USMCA contains a"poison pill" that forces the parties to explicitly agree to extend the agreement in 2026 or begin a negotiation that could last up to 12 years to avoid its permanent termination. Furthermore, the USMCA negotiation accommodated political interests in a nontransparent way. For example, Mexico requested the elimination of NAFTA’s energy chapter, even though it agreed to keep its obligations; some carried over from the TPP, and others added in ad-hoc footnotes. The overlap of the negotiation with the U.S. withdrawal from the TPP also introduced uncertainty and made the negotiation process more challenging.

Rules of Origin

NAFTA’s negotiators wanted the rules of origin to reflect ex-ante intraregional content. They recognized that criterion would generate some trade diversion, but it would avoid abrupt disruptions in the production chains. Over time, innovation and increasing integration modified the parameters of production functions, but the rules of origin remained fixed and as source trade diversion.

The USMCA negotiations opened a natural forum to reduce or eliminate the distortions accumulated by the existing rules of origin. However, the talks on the subject had a different tone and purpose; they aimed to do the opposite: namely, arbitrarily raise the thresholds of certain goods to qualify as “substantially” produced in the region. As a result, for example, the USMCA rule of origin for motor vehicles is approximately 20% more restrictive than it was under NAFTA, to the detriment of regional social welfare.

It is safe to assume that agents benefiting from higher origin thresholds will continue to advocate formaintaining and increasing the implicit effective protection of existing rules of origin, grandfathering the baseline in each renegotiation. Thus, the rules of origin threaten to become a protectionist vortex, a permanent source of trade diversion and net social loss. Damage would be even greater if the origin were conditioned to the nationality of beneficial owners. Such actions could trigger retaliatory measures by the countries affected and, ironically, disrupt the supply of critical inputs for regional manufacturers leading to restrictions on North American exports and worldwide investments.

Delinkage of Mexican Supply Chains

Mexico's preferential treatment to a subset of exporters (including maquiladoras) increases import propensity and undermines the export multiplier. The existing regime de facto subsidizes the consumption of extra regional intermediate goods and inhibits the domestic value-added of exports (backward linkage). This partially explains why annual Mexican export growth since 1994 has been five times greater than GDP growth.

Exporters operating under the preferential regime benefit on various fronts. On one hand, they require less working capital than companies subject to the general system because their imports enter the country tariff-free and are exempt from value-added taxes and customs processing fees. They can also reduce their income tax bill by operating without a permanent establishment in Mexico, benefitting from special deductions (for example, payroll benefits are deductible), and applying transfer prices that allow them to move taxable income out of Mexico. Additionally, they are allowed to document exchanges of goods amongst each other that never take place through "virtual" purchase orders. By allowing maquiladoras to choose from several operating "models," the legal framework lends itself to evasive practices that, in addition to having a high tax cost, indirectly inhibit backward linkages.

Customs Operations

Customs procedures add inefficiency to intraregional trade. For instance, the average cost of a Mexican customs agency on the U.S. border ranges between 1% and 3% of invoice value. Customs agents adhere to the same rates by exercising a form of tacit collusion. The partial deregulation allowing exporters to designate their employees as customs representatives introduced some competitive forces, but this mechanism helps only a few large companies.

Border rules require customs agents to invest in dual storage on both sides of the border and in transport equipment to move goods between warehouses. The costs of loading-unloading-reloading, shrinkage, and the time elapsed from when goods arrive at the export customs warehouse to when they leave the import customs warehouse are passed on to destination (DDP) prices.

---

Anxieties, Dualities and Setbacks

Notwithstanding its generally positive results, the treaty faces short-term structural challenges that threaten its effectiveness and support. In the United States, for example, a significant proportion of the population blames the trade agreement for dismantling industrial supply chains, destroying industrial jobs, and promoting the employment of undocumented migrants. However, these phenomena are more likely explained by innovation, technological change, and the entrenchment (and political exploitation) of economic, cultural, and even racial anxieties within the U.S. collective psyche. Moreover, recent protectionist actions weaken trade creation and subject the region to greater trade diversion going forward.⁶ For example, U.S. tariff increases in 2018, and Mexico's responses, have distorted effective protection. Also, the assault on the multilateral trading system and its institutions threatens to render dispute resolution mechanisms irrelevant, thereby increasing investment risks. In the short term, this protectionist bias will have a more pronounced impact on Mexico’s economy, but its negative effects will gradually spread across the rest of North America, ultimately undermining the competitiveness of the region’s collective supply.

In Mexico, the treaty’s benefits have not cascaded over the entire economy, and a significant portion of the population operates within a system plagued with structural weaknesses that hinder total productivity and per capita income.⁷ These issues are not new; they afflict the Mexican economy since beforeNAFTA. When the original treaty’s negotiations concluded in 1992, the negotiators pointed out that treaty was not a panacea, a wand that would magically transform society. Complementary reforms were needed, eliminating barriers to market entry (and exit), boosting supply across businesses of all sizes in every region of the country, and opening secure institutional avenues to accommodate the legitimate aspirations of all Mexicans. In other words, Mexico’s economic and political reality demanded a comprehensive overhaul, not just of the economic model but of the entire social contract. NAFTA alone was not enough. Important progress followed, such as deregulating certain services and agricultural supply chains, but the effort was incomplete. A large portion of the Mexican economy was left behind, exposed to chronic informality, corruption, violence, and stagnant productivity. The partial and patchy implementation of reforms created two subeconomies: one, modern and productive; the other, laggard and non-productive. This economic duality left Mexico with fragile internal growth engines and weak institutions that undermine total productivity, stagnate per capita, income and deepen social divides that foster division and embed a rancorous "Us and Them" public discourse.⁸

Many Mexican public policies perpetuate incentives that favor informality, undermine economic competition, and destroy the rule of law. This combined set of self-undermining policies rewards participation in activities with very low productivity and fosters an “anti-Schumpeterian” dynamic, which Santiago Levy and Luis Felipe López-Calva coin as "destructive creation".⁹ Recent policy decisions have promoted trade diversion and more "destructive creation", including tariff increases that distort effective protection, executive defiance of arbitral rulings that violate property rights, and judicial reforms that threat en constitutional balances and the rule of law. Ominously, many  recent actions are doppelgangers of similar ones in the United States, such as the tariffs and countervailing duties the U.S. applied in 2018 on about 13% of imports, including up to $200 billion from China.

The U.S. government’s lukewarm commitment to intraregional free trade weakens the institutional buttress needed to properly enforce the USMCA, which is leading, at best, to misunderstandings and, unilateral violations of the treaty. Mexico, for example, has accumulated multiple lawsuits for breaching its obligations under the USMCA in critical sectors for intraregional trade, such as motor vehicles, energy, and agriculture. It is currently the country facing the most trade lawsuits in North America and the fourth most sued globally. Mexico’s contingent trade liability sums up to 2% of GDP.

The dissolution of independent agencies, including sector-specific regulators and the federal antitrust commission, will complicate the dispute resolution, because the government will act as judge and party inunilateral actions affecting commercial and property rights. This complication is compounded by theuncertainty generated by recent constitutional reforms of the judicial system and the introduction of a constitutional "supremacy clause" that will be a potential breeding ground for more investor-State and State-to-State conflicts.

The USMCA review will take place in this light-and-shadow environment. The process may take up to ten years, looming inevitably like a sword of Damocles over the entire regional economy. At first glance,the recent policy setbacks in the region might suggest that a comprehensive renegotiation of the agreement would be too challenging to execute and that the path of least resistance would be for the Parties to extend the USMCA with minor adjustments, leaving its core features intact. However, extending the current agreement would preserve the region as a free trade area, where each Party maintains independent extra regional trade policies and its exporters continue to "certify" their goods as "substantially" produced in North America  to move them within the region tariff-free. This approach would also entrench distortions inherited from the older NAFTA and deepen in the current USMCA. It would also weaken the region's competitive position in the world due to international trade shifts toward models that favor sub globalized supply chain

---

Towards a Common Tariff

The USMCA review faces two potentially contradictory objectives: on the one hand, limiting intraregionalfree trade to goods that are "substantially" produced in North America, with an increasingly restrictive definition of "substantial" regional production; on the other hand, creating a more competitive regional supply to compete in the world markets with both individual countries,  such as China and India, and clustered nations, such as the members of the European Union. The contradiction will materialize if the inherent and accumulated distortions of the USMCA survive, particularly the rules of origin and, in Mexico, other policies that continue to stagnate total factor productivity. However, the contradiction could be resolved by transforming the free trade area into a customs union that unifies extra regional trade policy and promotes the relocation of intermediate input production and final goods to North America. Adopting a common tariff would be a natural progression for an increasingly integrated region whose trading regime has evolved from sector-specific bilateral agreements, such as the Auto Pact between the U.S. and Canada and the bilateral export quotas available to Mexico under the Generalized System of Preferences, to the USMCA.¹⁰

Reorganizing North America as a customs union with a common tariff would boost the competitiveness of regional supply for several reasons. First, a common tariff would eliminate origin certification and customs operations for intraregional trade. Also, scale effects would  lower long-term marginal costs across many sectors. The arrangement would also enhance the demographic and resource complementarity of North America. For instance, Mexico could attract production of intermediate goods and manufactured goods currently produced in Southeast Asia; the United States could focus on manufacturing capital-intensive industrial goods currently sourced from Europe and Northeast Asia; and Canada could benefit from a boost of demand for its ample energy supply.

Common Tariff

With the adoption of a common tariff, goods would circulate freely within North America. The tariff could be designed with two main objectives: minimizing trade diversion and leveraging the region's purchasing power.

Generally, the “optimal” tariff for a good whose demand does not affect international prices is zero since any favorable tax reduces welfare. However, if demand affects the international price, the optimal tariff can be positive and, reducing import prices, promote greater social surplus.¹¹ Using this fundamental economic concept, an optimal tariff vector could be calibrated with tariffs on various pharmaceutical goods, medical equipment, furniture, plastics, and organic chemicals, as well as other products wheredemand from the region is so high that it affects international prices. Two natural candidates for the imposition of optimal tariffs are machinery and equipment (including computers) and automotive(vehicles). North American imports (500 billion dollars annually and 400 billion dollars annually, respectively) represent a high percentage of the exports of the countries supplying these goods (30% to 50%).

Items with mixed tariff regimes would unify; for example, agricultural imports would be subject to the same ad valorem and/or specific tariff, and quotas would be aggregated and auctioned among importers of all the region.

The common tariff would take advantage of gaps between the current (MFN) and consolidated tariffs each country is bound to under the World Trade Organization (WTO), so the new agreement "on the whole" becomes more open than the current free trade area, as required by Article XXIV of the GATT. Initially, a common tariff could apply to goods with an identical tariff regime, with a transitional schedule toward the common tariff.

Tariff Revenues

Tariff sharing in existing customs unions is governed in various ways. One formula used by the European Union allocates a percentage of the revenue to the country where the goods arrive and the rest to the European Commission. Another formula used by the Southern African Customs Union distributes the revenue based on a negotiated formula that considers the size of each member country's economy and per capita income. In North America, a plausible revenue-sharing formula would have two components: one, allocating a percentage of revenue to the country of entry (arrival criterion), and second, distributing remaining revenue based on each party's contribution to intraregional trade (trade creation criterion).

Extraregional Trade Relations

The principal challenge for WTO approval would be to align the common tariff with currently consolidated national tariffs. Mexico’s consolidated are generally greater than the "most favored nation" (MFN) tariffs it currently applies. This gap is smaller in the United States and Canada, so the reduction of consolidated tariffs in Mexico's tariff vector could offset selective increases in the tariff vectors of its partners to abide by the “on the whole” criterion.

The convergence of bilateral agreements would likely be more complicated since the parties have negotiated a "spaghetti bowl" of trade agreements, as U.S. negotiator Julius Katz anticipated 30 years ago. Each bilateral agreement has its rules, exceptions, unique mechanisms, and institutional framework. For example, under the agreements with Europe and Japan, certain Mexican imports are duty-free but subject to tariffs in the U.S. Therefore, the convergence of bilateral commitments would require distinguishing the obligations that a Party would keep with other countries or regions ("grandfathering") from those that would gradually be unified ("phaseouts"). Save exceptions, the former would be limited to obligations with no impact on border revenue, while the latter would converge after some time.

Contingent Protection and Competition Policy

The new arrangement would require establishing a common contingent protection policy (anti-dumping quotas, countervailing duties, and temporary safeguards) and a common competition policy. To avoid the creation of supranational bodies, the parties would individually operate mirror legal systems.¹²

Mexico would bifurcate its system so it mirrors the U.S. and Canadadian systems by, for example, assigning the calculation of dumping margins and countervailing duties to the Secretary of Economy (SE) and injury determinations to an independent body.¹³ Additionally, the Parties could use two methods to "accumulate" evidence of injury or threat of injury. Though unusual, both are allowed by the WTO (third-party and concurrent investigations). For example, Mexico could adopt the U.S. Department of Commerce memorandum dated October 26, 2017, which determined that China was still not a market economy and, consequently, applied third-country prices to determine normal value to all Chinese imports under investigation. The parties would base the institutional framework of this coordinated system, on Articles 316 and 317 of the USMCA: Article 316 to agree on a coordinated compliance system, and Article 317 to create a coordination mechanism for investigations involving a third country.¹⁴

Intraregional anti-dumping investigations would be replaced by a single competition policy administered by national bodies. In Mexico, an independent entity from the SE, which could be the same one that determines injury in unfair practices cases, would investigate illegal anticompetitive practices. Across the region, investigators would use uniform criteria to determine the scope of a relevant market, market power, and remedial measures.¹⁵

Backward Linkage in Mexico

With the new arrangement, the distinction between temporary and permanent imports would cease to make sense, so Mexico could eliminate the exceptional regime for exporters, if necessary, with a period that allows them to process inventories, fulfill existing contractual obligations, and transition to the general taxation regime. Nearshoring dynamics are a centrifugal force that would counteract centripetal pressures opposing the elimination of the preferential regime. To stimulate backward linkage, the transition would have to be accompanied by  infrastructure, education, security, and energy supply improvements.

Implementation

The transition to a common tariff would alter negotiation logic. "Give-and-take" would replace a collaborative process focused on articulating a common trade policy. This would transform negotiators into "implementers," whose task would be to create the institutional framework upon which the new arrangement would operate.

The arrangement could be implemented in stages, prioritizing the adoption of the common tariff. One possible sequence would be the following: first, creation of a new tariff vector and inclusion of goods classified in subheadings with equal tariffs in all three countries today; second, homogenization of non-tariff barriers in all subheadings of the new tariff vector; third, addition of common tariff in all subheadings not included in stage one; fourth, homogenization of national economic competition policies and institutions; and fifth, homogenization of the policies and institutions that govern national contingent protection systems. A staged implementation of the arrangement would allow the parties to gradually adapt their operations without creatingsupranational institutions.¹⁶

Thus, the transition to a customs union would draw up separate convergence schedules: one for the common tariff, another one to unify external trade relations, and a third one to adopt common contingent protection and competition rules.

---

Final Remarks

A new customs union agreement anchored on a common North American tariff could trigger other complementary regional agreements to bolster regional competitiveness. For instance, to take advantage of the region’s combined demographic pyramid, the parties could negotiate a trilateral circular mobility agreement to match labor supply and demand throughout the year. To help prevent the illegal influx of extra-regional labor, circular mobility could be scaffolded with other region-specific security agreements; for example, one focused on dismantling the illegal supply of fentanyl in Northwestern Mexico and another one designed to combat human trafficking in Southern Mexico.

Unlike 35 years ago, today, no consensus exists on how to rank the models of international production. Unexpected supply shocks and the rejection in many countries to subject their trade to rules and bureaucracies distant from their national institutions of popular representation have discredited the global model of the late 20th century. In its place, sub-global production chains emerge, probably leading to less efficient trade but also likely to be more resilient to unforeseen events.17 In this sub-globalized world, anchoring a common tariff with a customs union agreement would contribute more effectively than the USMCA to help keep North America as the most productive region of the world.

---

^{1 We thank Ximena Ángeles for data processing and José Antonio González, Luis Alberto Ibarra, Santiago Levy, BeatrizLéycegui, Luz Elena Reyes and Luis Rubio for feedback and suggestions.}

^{2 Serra, Jaime, et.al., Reflections on Regionalism: Report of the Study Group on International Trade, Carnegie Endowment for International Trade.}

^{3 Serra Jaime, Pedro Noyola, et.al., 1997.}

^{4 Koopman, Robert, et. al., “Tracing Value-added and Double Counting in Gross Exports”,}

^{Working Paper 18579, National Bureau of Economic Research, 2012.}

^{5 The Johansen co-integration test is a statistical method used to determine whether the values of certain variables movein a similar manner over a period of time. It employs a binary criterion based on the "null hypothesis": Certain statistics of the time series are computed, and if their values do not exceed a "critical" level, the null hypothesis is confirmed (there isno cointegration); otherwise, the null hypothesis is rejected (there is cointegration).}

^{6 Bernard Hoekman and Ernesto Zedillo criticize the logic of recent US protectionism by highlighting what they call "fivemysteries" (or perhaps more accurately, "five self-inflicted wounds") of US trade policy: (i) withdrawing from the Trans-Pacific Partnership (TPP), (ii) denying essential economic principles, (iii) renegotiating NAFTA, (iv) trade warring with China, and (v) undermining the multilateral trading system. Hoekman, Bernard and Ernesto Zedillo, “The International Trading System in Prostration, Courtesy of the United States”, Trade in the 21st Century - Back to the Past?, The Brookings Institution, 2021}

^{7 According to Santiago Levy and Luis Felipe López-Calva, Mexico experienced an average annual reduction of 0.5% in total factor productivity and an average annual increase of only 1.1% in per capita output, half of the average of the 30years prior to NAFTA. They argue that this phenomenon is explained by three persistent weaknesses in economic policy: a plethora of economic and institutional incentives that favor informality; poor regulation and oversight of economiccompetition, particularly in strategic and non-tradable sectors; and a broad veil of legal insecurity over the property rights of individuals. Levy, Santiago and Luis Felipe López-Calva, “¿Qué falló? ¿Qué sigue? México 1990-2023”, Nexos, 2023.}

^{8 Claudio Lomnitz discerns the anthropology of this duality noting that the State itself promotes societal division into twogroups: one living in islands of productive legality (where the export sector operates); the other existing in an ocean of illegality (where informal markets operate). Lomnitz, Claudio, “El Neo Estado. La ínsula de los derechos y el mar de la extorsión”, Nexos, 2022.}

^{9 Levy, Santiago and Luis Felipe López-Calva, 2023.}

^{10 Trade agreements line up in a continuum, so to speak, that incrementally broaden, from sectoral agreements withlimited tariff preferences (such as in Latin America under the auspices of ALADI), to broad free trade agreements (such asin North America), to customs unions with common tariffs (such as in Southern Africa), to common markets with factormobility (such as in the Southern Cone of America), to economic unions with single currencies (such as in Europe). Withinthe continuum, each type of agreement adds one or more elements of integration to the immediately lower type.}

^{11 The literature on optimal tariffs is ripe. Following Francis Edgeworth’s contribution in the late 19th century, the theory became well-established, but the debate over its empirical importance lasted throughout the 20th century. ChristianBroda, Numo Limao, and David Weinstein provide evidence that countries do exercise market power through their tariffpolicy. Broda, Christian, et. al., “Optimal Tariffs: The Evidence”, Working Paper 120333, National Bureau of Economic Research, 2006.}

^{12 The Canadian system against unfair international trade practices resembles that of the United States; it is bifurcated: the Canada Border Services Agency (formerly Revenue Canada) determines dumping margins (or countervailablesubsidies) and the Canadian International Trade Tribunal, an institution similar to the International Trade Commission inthe U.S., determines injury.}

^{13 A few years ago, the government considered separating the International Trade Practices Unit from SE, but the initiative was dropped.}

^{14 There appears to be an error in Article 317 of the USMCA. Unless the numbering of GATT articles has been revised, the correct reference should be to Article 14 of the "Agreement on Implementation of Article VI of the General Agreement onTariffs and Trade." Beatriz Léycegui and Luz Elena Reyes identified this oversight.}

^{15 A unified competition policy without a supranational regulatory authority would require adequate coordination between national regulators.}

^{16 In Europe, the European Commission claims supranational authority, facilitating operational coordination but limitingnational participation. Other customs unions in Africa, Latin America, and the Middle East grant greater authority to their national institutions.}

^{17 One of us argues elsewhere that given the complexity of the current period of modernity, it is of utmost importance toinvestigate apparent truths and conventional wisdom with analytical rigor and an appreciation for the changing contexts of society. They argue that both excluding reason and mythologizing it are intellectual eccentricities that result in an incomplete understanding of a complex, random, and often contradictory world. They conclude the most promising economic models will emerge from a tempered concept of optimality, based on a "commitment to reason" that may lead to "suboptimal" equilibria but, in exchange, provide the social scaffolding modern societies require to anchor stable andmore durable social contracts. Noyola, Pedro, and Armando González, Modernidad y economía, Reflexiones sobre el Cambio, 1999.}