VoxEU Column

Renegotiating NAFTA: The role of global supply chains

The Trump administration has been outspoken in its criticism of NAFTA, which the president has called “the worst deal ever made”. This column, taken from a recent Vox eBook, argues that reversing the current NAFTA policy environment would not simply wind back the clock to the pre-agreement economy from 20-plus years ago. Instead, it would throw spanners and blockages into today’s very different and deeply integrated North American economy.

Editor's note: This column first appeared as a chapter in the Vox eBook, Economics and policy in the Age of Trump, available to download here.​

The Trump administration has been outspoken in its criticism of the North American Free Trade Agreement (NAFTA), which the president has called “the worst deal ever made”. This disparagement is not just campaign season hyperbole. Closing in on his hundredth day in office, Trump reportedly drafted – though ultimately nixed – an Executive Order withdrawing the US from the agreement (Financial Times 2017). He has also repeatedly issued public promises to renegotiate or withdraw from the pact: “If they don’t treat [us] fairly, I am terminating NAFTA.”1 At the same time, supporters of the deal predict calamitous effects of raising barriers between the US and its two closest trading partners, Mexico and Canada.

Here’s the thing: while NAFTA may have done little to boost or harm overall growth and prosperity on the continent, it has had a powerful role in redefining how and where products are made.2 And so even if NAFTA had been a raw deal, abandoning the agreement could have devastating consequences, especially in the near term.

Like it or not, the fortunes of North American firms, workers and consumers are now deeply intertwined through a dense network of regional and global supply chains. This interconnectedness makes the North American economy more competitive with the rest of the world, but also leaves it vulnerable to policy changes.

Pulling out of NAFTA would send widespread and long-lasting shock waves throughout the North American economy. To understand why, it helps to first appreciate the extent to which the deal has shaped the current economic landscape of the US, Canada, and Mexico

Ask not how much a country makes, but how it makes it

In aggregate terms, NAFTA has had a decidedly modest impact on the size and growth of the North American economy. According to one study (Caliendo and Parro 2015), the overall welfare gains from the tariff reductions under the agreement have been largest for Mexico, at roughly 1.3%, while the US has seen much smaller welfare gains of roughly 0.08%; Canada’s welfare is estimated to have fallen by 0.06% as the US shifted commercial attention toward its southern border (see also Hufbauer and Schott 2005, among others).

In contrast, the evolution in the composition and pattern of economic activity since NAFTA has been profound. Over the past 20 years, the North American economy has grown up and around and through the policy scaffolding afforded by the provisions of the agreement (Hanson 2001, Bair and Gereffi 2002). According to Caliendo and Parro (2015), the tariff reductions alone under NAFTA caused the volume of intra-North American trade to rise by 41% for the US, 11% for Canada, and more than 118% for Mexico.

While NAFTA was neither great nor terrible for the size of the overall economy, it was a game-changer for how the North American economy works.

To be clear, not all of the increase in North American trade is due to NAFTA alone. Since (especially) the 1990s, the world has seen a revolution in the nature of global commerce. Technological and logistical innovations (together with increased economic openness) have spurred on the phenomenon known as production fragmentation: the ability to design, source, assemble, and refine products through increasingly complex domestic and global supply networks (Johnson and Noguera 2017, Fort 2017, Bernard et al. 2017). NAFTA did not create these global supply chains, but its rules governing commerce at and behind North American borders allowed them to flourish.

The products that Americans consume – everything from a toaster to an iPhone or Audi Q5 – are produced by combining and recombining constituent parts through often- complex supply networks. These stretch from design to mining and farming of raw materials to construction and marketing of the final goods that ultimately shape our lives. Supply networks weave together the economic fortunes of firms and workers from the headquarters of multinationals and refineries of heavy industry to independent assembly plants, cottage industries, and small farmers.

This is especially true in North America, where NAFTA’s tariff reductions and ‘deep provisions’ – like regulatory reforms and investment protections – have created one of the world’s most integrated regional economies. In turn, greater specialisation and fluidity within the production process has helped to keep North American products competitive with the rest of the world (Hufbauer and Schott 2005).

At the same time, production fragmentation has afforded firms and workers the chance to specialise in increasingly narrow slivers of the global production process, carving out a competitive niche in the global marketplace. As a result, more workers and more firms now take part in regional and global trade than ever before.3

Production fragmentation rewrites the book on how to think about trade policy

Most importantly from a trade policy perspective, production fragmentation knits together the economic interests of firms (and workers) up and down the supply chain. This 21st century trade also redefines the ‘winners’ and ‘losers’ from increased trade: in the era before foreign direct investment and global supply chains, trade liberalisation often benefitted local consumers at the expense of local producers. But with these linkages, the producer-side gains from trade that used to accrue only to foreign exporters are shared – and often divided differently – on both sides of the border (Blanchard 2010).

Consider two scenarios to illustrate the point. In scenario A, a traditional producer in Mexico makes a product (say, camping tents), from start to finish, in its local manufacturing facility, which it then exports to consumers in the US. If the US lowers its tariffs on tents imported from Mexico, more tents are sold and at a lower price, and the gains are split between US consumers and the Mexican producer: end of story. Contrast this with Scenario B, in which the Mexican producer conducts the final assembly of camping tents, using parts (fabric, thread, plastic coatings, metal fittings, etc.) imported from the US and design services developed in Canada. Now, if the US lowers its tariff against tents from Mexico, the producer-side gains will be shared among the downstream Mexican assembly plant, the US suppliers of intermediate inputs and the Canadian design firm.4

These supply chain linkages mean that some – potentially even all, depending on the nature of supplier contracts – of the production-side gains from trade liberalisation are passed back up the supply chain to upstream firms and workers, including those in the country that is lowering its tariffs. This changes the fundamental calculus of trade protection.

By lowering US tariffs on goods imported from Mexico and Canada, the NAFTA directly benefits US-based suppliers of inputs used to produce its neighbours’ exports.5 The more interwoven are North American supply chains, the more broadly shared are the gains from NAFTA’s open borders.

The vulnerability of interdependence

The flip side of the new opportunities afforded by open borders and production fragmentation is that some workers (often in the US) have suffered job losses as firms have moved in-house operations abroad (often to Mexico) and away from more expensive existing factories. In the US, these job losses have been highly concentrated in a handful of regions and worker-groups, to devastating local and personal effect (Hakobyan and McLaren 2016).

The subsequent populist reaction against globalisation in general, and NAFTA in particular, should not be surprising. And indeed, in 2016, President Trump was elected in part based on his sharp opposition to existing free trade deals.

While the deep integration of North American supply chains has increased overall efficiency, it has also sharpened political and economic divisions, and left the economic system more vulnerable to potential disruptions in the freedom to move goods and services across borders. The sitting president has vowed to disrupt the existing NAFTA structure: what is at stake if he does?

What will happen if NAFTA is reversed?

It cannot be emphasised enough: reversing the current NAFTA policy environment will not simply wind back the clock to the pre-agreement economy from 20-plus years ago. Instead, it would throw spanners and blockages into today’s very different and deeply integrated North American economy.

Today, much of every of dollar that the US spends on imports from Mexico consists of US ‘value added’, the benchmark measure of upstream, supply chain inputs. Due to NAFTA’s supply chains, a considerable share of Mexican production consists of Canadian value-added as well. And vice versa.

If NAFTA were abandoned, the short-run consequences for firms and consumers could be devastating until – or unless – global supply chains adjust to a new (or no) NAFTA world (and we do not know how long that would take).

Abandoning the key tenets of NAFTA – especially vis-à-vis trade with Mexico – could have a profound negative impact on the economies of all three signatory nations. According to recent research on the auto industry (Head and Mayer 2016), withdrawing from NAFTA would reduce the US’s share of world auto production, not least because it would force an expensive reversal of North American automotive supply chains. One economic simulation predicts that all three NAFTA signatories would suffer losses from a return to MFN tariffs, with the most acute consequences predicted for Canada and Mexico; in contrast, the same simulation predicts that the rest of the world would see a relative gain in market share as North American car makers become less competitive.

Under a separate worst-case ‘Trumpit’ scenario (in which NAFTA is replaced with Trump’s threatened 35% tariff against Mexico), the same study predicts that Mexico’s share of world auto output would decline by a startling 41%.

There is, thus, not only an enormous potential internal cost of withdrawing from NAFTA, but also a potential external cost: retreating from open borders would almost certainly damage North America’s ability to compete with the rest of the world, perhaps dramatically. This relative disadvantage would be compounded by the potential efficiency gains in ‘Factory Asia’, already a fierce competitor of the North American economy, if Asia Pacific nations implement RCEP (China’s proposed regional free trade agreement). Initially an answer to the proposed Trans Pacific Partnership, RCEP is now the only game in town and possibly all the more potent, as a result.

Uncertainty isn’t helping

Even without renegotiation, uncertainty around even the potential for a NAFTA withdrawal is likely to damage the North American economy. Tough talk on trade has a chilling effect on firms’ willingness to make new investments or supply contracts on either side of the borders in question. Research demonstrates that even in the absence of actual changes in trade policy, this induced uncertainty can be every bit as costly as tariffs themselves (Handley and Limão 2017).

Given the stakes, what can we expect?

It is hard to know how the NAFTA shake up will play out. Is the President’s tough talk just a high-stakes gambit calculated to improve US bargaining positions on the Mexican border wall or the long-standing dispute over Canadian softwood lumber? Or is it possible that core tenets of NAFTA – tariffs and other ‘deep’ provisions (e.g. rules of origin, bilateral safeguard provisions, etc.) are truly on the table?

The outcome presumably will hinge on domestic politics, where the competing influences of Trump’s populist supporters are pitted against powerful multinational firms who vie for the president’s attention. For much of the 20th century, US trade policy has seemed a better reflection of the latter (e.g. Gawande and Bandyopadhyay 2000, Blanchard and Matschke 2015), but this has proven to be a year of surprises.

That said, rhetoric aside, recently leaked documents suggest not an across-the-board increase in tariffs against our trading partners, but a reopening and renegotiating of deep agreement provisions on labour and environmental standards, intellectual property and digital trade protections, state owned enterprises, and rules of origin (Wall Street Journal 2017).

Notably, these provisions appear to be close parallels to the proposed building blocks of the now-abandoned Trans-Pacific Partnership. Updating NAFTA’s outdated rules would be to everyone’s benefit.

Trade is not a zero-sum game, and if we play our cards right on NAFTA, everyone could gain. But a negotiating misstep could trigger a wholesale collapse of the agreement. Given the extent of deep supply chain connections, there is every reason to expect that severing ties would cause hardship on all sides.


Bair, J. and G. Gereffi (2002), “NAFTA and the Apparel Commodity Chain: Corporate Strategies, Interfirm Networks, and Industrial Upgrading”, in The North American Apparel Industry After NAFTA, G. Gereffi, D. Spener and J. Bair (eds.),Temple University Press).

Baldwin, R. (2016), The Great Convergence: Information Technology and the New Globalization, Harvard University Press, Cambridge, MA.

Bernard, A., J.B. Jensen, S. Redding, and P. Schott (2017), “Global Firms”, Journal of Economic Perspectives, (forthcoming).

Blanchard, E. (2010), “Reevaluating the Role of Trade Agreements: Does Investment Globalization make the WTO Obsolete?”, Journal of International Economics 82(1): 63-72.

Blanchard E. and X. Matschke (2015), “U.S. Multinationals and Preferential Market Access”, Review of Economics and Statistics 97(4): 839-854 .

Blanchard, E., C.P. Bown and R. Johnson (2016), “Global Supply Chains and Trade Policy”, NBER Working Paper 21883, January 2016.

Caliendo, L. and F. Parro (2015), “Estimates of the Trade and Welfare Effects of NAFTA”, Review of Economic Studies 82 (1): 1-44.

Financial Times (2017), “Draft Nafta order signals tougher US stance on trade partners”, 28 April.

Fort, T. (2017), “Technology and Production Fragmentation: Domestic versus Foreign Sourcing”, Review of Economic Studies (forthcoming).

Gawande, K. and U. Bandyopadhyay (2000), “Is Protection for Sale: Evidence on the Grossman-Helpman Theory of Protection”, Review of Economics and Statistics 82(1): 139-152.

Hakobyan, S. and J. McLaren (2016), “Looking for Local Labor Market Effects of NAFTA”, Review of Economics and Statistics 98(4): 728-741.

Handley, K. and N. Limão (2017), “Policy Uncertainty, Trade and Welfare: Theory and Evidence for China and the U.S.”, American Economic Review (forthcoming).

Hanson, G. (2001), “US-Mexico Integration and Regional Economies: Evidence from Border City Pairs”, Journal of Urban Economics 50(2): pp 259-287.

Head, K. and T. Mayer (2016), “Brands In Motion: How Frictions Shape Multinational Production”, CEPR Discussion Paper DP10797, (summary and extensions available from as “Reversals of Regional Trade Agreements: Consequences of Brexit and Trumpit for the Multinational Car Industry”, November 12, 2016).

Hufbauer, G.C. and J. Schott (2005), NAFTA Revisited: Achievements and Challenges, Peterson Institute for International Economics, Washington, D.C.

Johnson, R. and G. Noguera (2017), “A Portrait of Trade in Value Added over Four Decades”, Review of Economics and Statistics (forthcoming).

Wall Street Journal (2017), “Trump Administration Signals It Would Seek Mostly Modest Changes to Nafta”, 30 March.


1 AP News, April 21, 2017 interview transcript at

2 NAFTA had sharp distributional consequences for certain individuals and regions (Hakobyan and McLaren 2016), but the overall impact of the agreement on aggregate US growth and income was both small and positive (e.g. Hufbauer and Schott 2005).

3 Baldwin (2016) makes a compelling case that GSCs have democratised trade.

4 See Blanchard et al. (2016) for formal treatment.

5 The trade-liberalising influence of multinational firms and GSCs on trade policy in practice is demonstrated in Blanchard et al. (2016) and Blanchard and Matschke (2015)

4,409 Reads