The economic impact of a (partial) NAFTA breakdown

From 21 until 29 January, the US, Canada and Mexico will continue their sixth round of NAFTA negotiations in Montreal. So far, the negotiations have been far from smooth and during the fourth round, the US put a number of hard to swallow proposals on the table. In anticipation of the sixth round of renegotiations, there are more and more indications that the US maybe planning to pull out of NAFTA (Reuters, 2017), but it is difficult to separate the rhetoric from reality when assessing negotiations.[1]
In this Special we make an assessment of the potential macroeconomic fallout in two NAFTA breakdown scenarios: a partial dissolution of the NAFTA (soft breakup scenario) and a total breakup of the trade agreement (hard breakup). These scenarios are compared to our benchmark scenario in which NAFTA is continued as it currently stands.. It is important to stress that our aim in this Special is not to determine the likelihood of each scenario occurring, nor is it to assess the advantages and disadvantages the current NAFTA has created thus far.

[1] This is particularly true when President Trump has a well (self) documented history of a negotiating style that starts at the extreme in order to shift the entire referencing frame for future negotiations.

There have already been five official rounds of NAFTA negotiations out of a (currently) scheduled total of seven. Negotiations are currently set to conclude before the end of 2018Q1. Indeed, all three parties have acknowledged the importance of completing negotiations before the July 1^st Presidential elections in Mexico and the US mid-term elections scheduled to be held in November. The original plan was to finish negotiations in 2017, but during the (extended) fourth round of negotiations held from October 11-17^th, 2017, it was decided that talks would continue into the new year. The fifth round of negotiations was held from November 17-21^st in Mexico City, Mexico and little progress was made relative to the fourth round. Attention now turns to the sixth and seventh rounds. The sixth round was initially scheduled to be held from 23-28^th January, 2018 in Montreal, Canada but the start of negotiations were brought forward to 21^st and they are now scheduled to end on the 29^th. The extension from six days to nine days makes this the longest round of NAFTA negotiations so far. Talks began on Sunday the 21^st, with energy, investment, financial services, agriculture and other issues on the agenda. Rules of origin (see below) – one of the thorniest subjects, related to manufactured goods such as cars – will be reserved for the final days of talks, as in previous rounds (Bloomberg, 2018).

Poison pill proposals

During the fourth round of negotiations in Washington D.C. during October 2017, various US proposals highlighted the extent of differences of opinions, or as the official Trilateral Statement on the Conclusion of the Fourth Round of NAFTA Negotation put it: “New proposals have created challenges and Ministers discussed the significant conceptual gaps among the Parties. Ministers have called upon all negotiators to explore creative ways to bridge these gaps. To that end, the Parties plan on having a longer intersessional period before the next negotiating round to assess all proposals.” Three hard to swallow proposals that the US put on the table in October are the sunset clause, rules of origin and dispute panels (see below). To exemplify how far-reaching these proposals are: the CEO of the US Chamber of Commerce of the US called the US demands ‘poison pill proposals’, which could jeopardize the whole deal. Below, we elaborate on the US demands separately.

Sunset clause

In essence, a sunset clause would require that any renegotiated NAFTA would need to be reviewed and ratified by all three parties every five years and there would be a so-called 'sudden death' mechanism if all three countries do not agree. Initially, both Canada and Mexico rejected such an idea suggesting it would introduce a significant amount of uncertainty into the relationship and could discourage investment. Since that juncture, Ildefonso Guajardo, Economy Minister, and José Antonio Meade, the then Secretary of Finance and Public Credit (now the PRI Presidential candidate in the Mexican general elections) suggested that a watered down 'sunset clause' could be acceptable as long as it doesn't contain the 'sudden death' element.

Rules of origin

In terms of rules of origin, the Canadian government notes that in the current agreement each NAFTA country retains its external tariffs vis-à-vis non-members' goods and levies a lower tariff on the goods ‘originating’ from the other NAFTA members. Rules of origin provide the basis for customs officials to make determinations about which goods are entitled preferential tariff treatment under NAFTA. Negotiators of the agreement sought to make the NAFTA's rules of origin very clear so as to provide certainty and predictability to producers, exporters and importers. They also sought to ensure that NAFTA's benefits are not extended to goods exported from non-NAFTA countries which have undergone only minimal processing in North America.

The new rules of origin proposal from the US comes in two flavors. Firstly, the idea of an increased NAFTA-region 'rules of origin' requirement and secondly, a proposal for the addition of a US specific 'rules of origin' requirement. The details on current product-specific 'rules of origin' requirements can be found in Annex 401 of the current agreement but media focus has been on how the US' proposals will impact the automotive industry. As it stands, a minimum of 62.5% of the material in a car or light truck made in the NAFTA region must come from North America. The US is suggesting raising this to 85% and adding a rule that requires 50% of materials must come specifically from the US. Unsurprisingly this was not welcomed by either Canada or Mexico but even the US industry has pushed back with Cindy Sebrell, a spokeswoman for the Motor Equipment Manufacturers Association, suggesting that: "Forcing unrealistic rules of origin on businesses would leave the U.S. unable to compete by increasing the cost of manufacturing and raising prices for consumers."

Dispute panels

'Dispute panels' or 'Investor-State Dispute Panels' (ISDS) refer to the mechanisms under which trade and investment disagreements between parties are settled. The Office of United States Trade Representative (USTR) defines ISDS as: a neutral, international arbitration procedure. The US' top trade representative, Robert Lighthizer, has suggested that changes to ISDS are necessary to respect our national sovereignty. Indeed, Lighthizer has also been vocal about his criticisms of dispute panels within the World Trade Organisation (WTO), suggesting that: “the WTO becoming a litigation-centered organization” and “too often members seem to believe they can gain concessions through lawsuits that they could never get at the negotiating table.”

The details of ISDS within the NAFTA framework are outlined in Chapters 11, 19 and 20 of the current agreement and the US is looking to significantly scale back Chapters 11 and 20, while effectively abolishing Chapter 19 altogether. Essentially Chapter 11 is aimed at encouraging foreign investment by ensuring that disagreements with governments will be heard by an impartial tribunal. The US wants to change the rules so that countries must decide whether they want to opt-in or not. In other words, participation is voluntary. Chapter 20 effectively allows governments to engage in legal proceedings with other governments and the US wants to make any verdicts non-binding.

Chapter 20 lays out steps "intended to resolve disputes by agreement, if at all possible. The process begins with government-to-government (the Parties) consultations. If the dispute is not resolved, a Party may request a meeting of the NAFTA Free-Trade Commission (comprised of the Trade Ministers of the Parties). If the Commission is unable to resolve the dispute, a consulting Party may call for the establishment of a five-member arbitral panel. Chapter 20 also provides for scientific review boards which may be selected by a panel, in consultation with the disputing Party, to provide a written report on any factual issue concerning environmental, health, safety or other scientific matters to assist panels in rendering their decisions."

Meanwhile, Chapter 19 is arguably the most controversial given that the US wants to scrap it entirely, but Canada has suggested since talks began that such a move would be a deal-breaker. Indeed, Canadian Trade Minister Chrystia Freeland, has threatened to walk away from negotiations if the US insists on abolishing Chapter 19. There are a number of components to Chapter 19 but essentially it ensures that the installment of binational panels[2] as an alternative to judicial reviews by domestic courts in antidumping and countervailing duty cases.

[2] A request for Panel Review is filed with the NAFTA Secretariat by an industry asking for a review of an investigating authority's decision involving imports from a NAFTA country.

What is NAFTA?

The North American Free Trade Agreement is a trade agreement signed by Canada, Mexico and the United States that forms a trilateral trade bloc across North America. NAFTA was signed on December 17th, 1992 and came into effect on January 1st, 1994 but its history goes back significantly further than that. Ronald Reagan first touted the idea of a North American free trade-zone when campaigning in 1979 but it was the signing of the Canada-United States Free Trade Agreement (CUSFTA) on January 2nd, 1988 that led the way for the formation of a North American free-trade bloc. Mexican President Carlos Salinas de Gortari approached US President George Bush to propose a similar bilateral trade agreement. Canadian Prime Minister Brian Mulroney then joined the talks and the three parties set about creating a trilateral agreement. On December 17th, 1992, all three leaders signed the NAFTA. The agreement then needed to be ratified by each country. During that time there were numerous leadership changes but US President Bill Clinton signed the Bill into US law on December 8th, 1993 and in so doing, NAFTA replaced the Canada-US FTA. Two additional agreements were also made between the three countries: the North American Agreement on Labor Cooperation (NAALC) and the North American Agreement on Environmental Coorperation (NAAEC).

The aim of NAFTA is laid out in Article 102 of the agreement which states the objectives:

a) eliminate barriers to trade in, and facilitate the cross-border movement of, goods and services between the territories of the Parties;
b) promote conditions of fair competition in the free trade area;
c) increase substantially investment opportunities in the territories of the Parties;
d) provide adequate and effective protection and enforcement of intellectual property rights in each Party's territory;
e) create effective procedures for the implementation and application of this Agreement, for its joint administration and for the resolution of disputes; and
f) establish a framework for further trilateral, regional and multilateral cooperation to expand and enhance the benefits of this Agreement.

It is important that NAFTA is not a common market, such as the EU. This means that customs administrations exist and goods entering Canada, Mexico or the United States must still comply with each country's laws and regulations. Hence, unchecked movement of people or goods among Canada, Mexico and the US is NAFTA is not allowed. Given the pre-existing FTA agreement between Canada and the US, the implementation of NAFTA saw more of a shift in terms of the relationship with Mexico. Most tariffs were to be eliminated within 10yrs although certain sensitive sectors, such as agriculture, had a 15 year time frame.

What has been the economic impact?

There is no doubt that NAFTA has liberalized trade between the US, Mexican and Canadian to a substantial degree. US trade with Canada and Mexico tripled ever since the agreement came into effect (see figure 1, 2 and 3). Moreover, under NAFTA, investment barriers between the partners were reduced significantly, ensuring basic protection and installing a dispute settlement regime between (Villareal and Fergusson, 2017). NAFTA provided non-discriminatory treatment of foreign investment by NAFTA partners in certain sectors. Exceptions are the national security sectors, the Mexican energy sector and the Canadian cultural industries. Data from BEA shows that under NAFTA, foreign investment from the US to Mexico rose from USD 15bn in 1993 to USD 88bn in 2016. US investment in Canada surged from USD 70bn in 1993 to USD 360bn in 2016. It is very difficult to isolate the effect of NAFTA on trade and investment flows from other effects, such the increasing globalization of trade or technological progress. We do not have a counterfactual situation what the world would look like without NAFTA (USITC, 2003). Looking at, for instance, FDI flows, it is striking that FDI shares of Canada in total outward US FDI flows have been remarkably stable and even decreased over time in Mexico (figure 4). This raises suspicion that upward trending variables in levels, such as trade and FDI, were also subject to other exogenous developments.

In the literature, the economic impact of NAFTA is also disputed. There is a multitude of studies examining NAFTA across different sectors and from different perspectives but opinions remain divided. When looking from the US angle, there are several studies underlining the economic merits (albeit small) of NAFTA (e.g. Romalis, 2007; USICT, 2003) for the US, but at the same time, others blame NAFTA for US job losses and subdued wage developments (e.g. Scott, 2011). It is beyond the scope of this Special to examine the advantages and disadvantages of NAFTA across all three countries since its inception, but one of the main arguments against NAFTA is in respect to its impact on US manufacturing. Some argue that NAFTA eroded the strength of the US manufacturing industry and led to a mass exodus of jobs. Indeed, this is the argument against NAFTA that we hear regularly from President Trump. That said, on the topic of manufacturing job losses it is worth highlighting the chart below which suggests that the introduction of China into the WTO had a far greater impact on the US manufacturing industry than the start of NAFTA (figure 5). We can also argue that without NAFTA, US manufacturing might would have suffered even more and become less competitive in comparison to other major manufacturers such as China. Furthermore, it is important to note the impact of technological advancements seen in recent decades; automation has had a notable impact on manufacturing jobs globally, regardless of trade agreements.

Trump and NAFTA

Donald Trump has been a vocal critic of NAFTA for many years. The critique is based on large trade deficits that the US runs with its main trading partners (figure 6). In Trumps view, this large deficit reflect dishonest trade practices, which have been at the expense of the US manufacturing base. By raising protectionist barriers, Trump hopes to bring back jobs to the US.[3]

[3] This is, however debatable, as the US trade deficit is no so much the result of trade policy, but of savings and consumption patterns of US households.

Before he won the election and became President of the United States, Trump had already discussed his plans to renegotiate the agreement. In fact, during the first Presidential debate against Hilary Clinton, Trump said: “your husband signed NAFTA which was one of the worst things that ever happened to the manufacturing industry”, before suggesting that “NAFTA is the worst trade deal maybe ever signed anywhere but certainly ever signed in this country.”

There have been numerous threats to withdraw or terminate the agreement coming from the Oval Office, but perhaps the clearest message from Trump is also the most ambiguous. In October 2017 he stated that: “It’s possible we won’t be able to make a deal, and it’s possible that we will”. Complicating matters further is another of Trump's election campaign promises: the building of a physical wall on the US’ border with Mexico. President Trump made this one of his main promises heading into the election and this was accompanied by his stance that said the wall must be paid for by Mexico. He has not changed his stance since taking office and on January 11th, 2018 he suggested in a Wall Street Journal interview that: “They can pay for it indirectly through NAFTA” and “We make a good deal on NAFTA, and, say, I'm going to take a small percentage of that money and it's going toward the wall. Guess what? Mexico's paying.”

In a historic overview, Ciuriak et al. (2017b) argues that Trump’s policy direction to tackle US bilateral trade imbalances is not new. Moreover, the US has a long history in stepping up efforts against these deficits by renegotiating agreements or forcing policy measures on its trading partners.

Although the economic benefits of NAFTA are disputed, there is no doubt that the US and Mexico and the US and Canada have continued their economic integration under NAFTA. This implies that a sudden disruption of the current trade regime will have consequences for all NAFTA partners. Below, we will elaborate on the mechanisms that are relevant.

Trade and market share

In case of a NAFTA breakdown, we expect trade barriers to rise between the US on the one hand and Canada and Mexico on the other. Despite a scale down on protectionist trade barriers after NAFTA, in recent years, the US increasingly imposed discriminatory trade measures against its NAFTA partners. These measures have outnumber the liberalization measures significantly (figure 7, Global Trade Alert). We expect this trend to speed up considerably and lead to a significant rise in trade barriers. There are two types of trade barriers that are important. First, there are tariffs on the trade of goods. In the case of a hard NAFTA breakdown, trade will be subject to the regime of the World Trade Organisation (WTO). This means that the trade in goods is subject to a so-called Most Favored Nation (MNF) tariff. The current MFN tariff s for the NAFTA partners are: US: 3.5%, Canada: 4.1% and Mexico: 7%.

Empirical studies show that the bigger chunk of welfare losses due to protectionism is not caused by tariff barriers, but by non-tariff barriers (see, e.g., Felbermayr et al. (2017), Egger et al. (2015)). A NAFTA withdrawal by the US could result in the emergence or intensification of non-tariff barriers, such as sanitary and phytosanitary (SPS) measures, customs charges, permits, quotas, trade regulations, import licenses, and border restrictions.

Both tariffs and non-tariffs trade barriers negatively affect export market prices and this impacts the export market shares of all NAFTA partners versus each other (Canadian-Mexican trade being the exception). The negative impact on export market share is mitigated in Canada and Mexico by the depreciation of their respective currencies which makes their goods cheaper in USD terms. That said, the downside of the shock depreciation of the Mexican peso and Canadian dollar is that imports of intermediates and cost-push inflation props up domestic prices and weighs on private consumption in these countries. Trade share declines are also mitigated by export substitution to other locations. Indeed, the recently signed Comprehensive Economic and Trade Agreement (CETA) between Canada and the EU and the soon to be renewed bilateral trade agreement between Mexico and the EU can cushion some of the expected trade fallout in case of a NAFTA collapse, although the total mitigating effect will be limited.

Foreign direct investment and dynamic productivity effects

Besides short-term trade and investment effects, if NAFTA collapses it will also have implications for a country’s long-term growth potential. First, foreign investment in Canada and Mexico is usually from firms that are more productive than the average, so the marginal efficiency of capital of US FDI is higher. This means that when US firms decide to invest less in Canada and Mexico, more Canadian and Mexican domestic investment is needed to restore output to pre-NAFTA breakup levels. A decline of US penetration in Canadian and Mexican markets could also have adverse productivity effects on firms as higher competition stimulates firms to reduce X-inefficiencies and increase efforts to innovate.[4]Edwards (1998) and Alcalá and Ciccone (2004) find that international trade has a robust positive effect on productivity.

Another channel that will probably have a large impact on Canada is the substantial R&D investment that comes from US firms. In fact, US R&D investment in Canada accounts for close to 23% of total Canadian intramural business R&D investment (see OECD databases: Activities of Multinationals and Main Science & Technology Indicators). If NAFTA leads to a decline in US firms’ inward R&D investment in Canada or even a reshoring of innovative activities to their home base, this might have large adverse effects on the Canadian economy. It is well-known that R&D investment has major implications on domestic productivity development.[5] Conversely, the stakes of Canadian firms in US R&D activities is a meagre 0.1, so any decline in Canadian R&D investment in the US will probably have very limited effects on US productivity. Finally, much economic literature shows that international knowledge developed has a larger impact on domestic productivity if a country is more open to either foreign trade (Coe and Helpman, 1995; Grossman and Helpman, 1991) or foreign direct investments (Branstetter, 2006). Hence, both FDI and trade are important conduits to benefit from international technological development, and a decline in either of these variables will adversely affect productivity growth.

In order to assess dynamic productivity effects of a potential NAFTA breakdown, one would have to estimate country-specific total factor productivity models (see Erken et al. (2017)). This is something that we will look at separately in a follow-up study.

Supply-chain integration

As a result of NAFTA, some segments of the US are highly integrated with the Mexican and Canadian economy. Any disruptions to supply chains would have serious negative consequences. Last year, Rabobank published a report looking into supply chain integration between the US on the one hand and a number of trading partners with which it runs large trade deficits (Erken and Tulen, 2017). In this report, we concluded that the global value chain integration of US firms is most prominent in three Mexican industries: motor vehicles (18.1%), electronic equipment (17.2%) and electrical machinery (16.7%). For example, for the Mexican motor vehicles industry, this means that 18.1% of final exports to the US consists of US value added. Conversely, Canada, Mexico, China, Germany and Japan are the most important suppliers of intermediate goods to US firms (Table 1). US exports of motor vehicles, petroleum and fuel products and basic metals are the sectors that depend the most on foreign intermediates. In case of a disruption to global value chains, US firms can decide to substitute intermediates from countries being hit by protectionist measures. However, these changes tend to be time consuming and involve transaction costs as other suppliers often do not instantly meet proper requirements and know-how. As is the case with dynamic productivity effects, we are not able to capture the negative impact of supply-chain disruptions in case of a NAFTA breakdown.

[4] The term X-inefficiencies refers to slack in the production process and higher production costs than necessary, which are the result of a lack of competitiveness pressure in the market.

[5] Erken et al. (2016), Griffith et al. (2004), Engelbrecht (1997), Bassanini and Scarpetta (2002) and Cameron et al. (2005) all find that there is a positive relationship between domestic technological capital on the one hand and productivity on the other.

In this study we examine two NAFTA breakdown scenarios:

1. Hard NAFTA breakdown: Negotiations fail and NAFTA is terminated in March. After the failed negotiations, the US immediately announces it is to withdraw unilaterally from NAFTA (Q1 of 2018). Canada and Mexico will continue to apply the NAFTA framework for their bilateral trade.
2. Soft NAFTA breakdown: Mexico, Canada and the US disagree on the conditions of NAFTA 2.0, but do agree to partially continue a slimmed-down free trade agreement. This new trade agreement won’t result in tariffs being raised, but non-tariff barriers will emerge and investment will be affected as well.

These scenarios are compared to our benchmark scenario. In the benchmark scenario, Canada, Mexico and the US reach an agreement on continuing NAFTA. NAFTA is adjusted with only limited impact on trade and investment flows. This baseline is the scenario that relates to our forecasts published in Rabobank’s most recent economic quarterly outlook (Van Es and De Vreede, 2017). There are two exceptions. First, given the approved US tax bill (Tax Cuts and Jobs Act), we have raised GDP growth in the US in by 0.5 percentage points. Second, the USD/MXN forecasts are based on a continuation of the status quo of the current Mexican government with respect to the July 1st election. This does not reflect Rabobank's view on the election, but has been assumed in order to focus on the NAFTA impact and not any potential distortions to USD/MXN stemming from heightened volatility in the lead up till the Mexican election.

The timeline of the assumed economic shocks in both NAFTA breakdown scenarios is discussed extensively in the Annex of this report and illustrated in Table 2. The infographic below also illustrates three assumptions in our scenario analyses.

In order to assess the economic impact of our two NAFTA breakdown scenarios, we use National Institute Global Econometric Model (NiGEM), which is developed by NIESR in London. NiGEM is a macro-econometric world trade model, estimated in a 'New-Keynesian' framework. This means agents are forward looking, but rigidities results in a slow adjustment process in case of external events or shocks. Rabobank has been using this econometric model for over a decade now and other institutions such as the Europe Central Bank and the Bank of England use the model as well.

Using NiGEM has three main benefits. First, the model allows us to assess the impact of several key variables in the short to medium term, such as exchange rate fluctuations, trade flows, foreign direct investment and the labour market. Second, all NAFTA partners have a separate model in the framework, but countries are linked to each other through trade and competition, interaction of financial markets and international asset stocks. This NiGEM ensures that all economic variables are viewed within a closed accounting setting and economic shocks, such as a NAFTA breakdown, are accounted for via these interdependencies. Third, NiGEM is an error-correction model, which ensures that short-term deviations of GDP from a country’s growth potential are made up eventually (see figure 9). So in the long-run, growth is driven by structural factors, such as capital formation, structural employment and labour-augmented technological change.

Tariff version of NiGEM

For trade-related topics, such as a NAFTA breakdown or Brexit, we use the expanded tariff version of NiGEM, which was released in 2017. This version has additional benefits over the conventional model in the sense that tariffs are embedded in the original version of NiGEM and as such, import inflation is properly taken into account. Higher tariffs are able to affect real wages, real disposable income and to lowers private consumption. At the same time, relative competitiveness is hurt by cost-push inflation as intermediates become relatively more expensive, which raises production costs, export prices and lowers competitiveness. We have added dummies to the equations for the export market size of Canada, Mexico and the US to be able to take into account trade substitution effects, which somewhat mitigates the negative effects of a NAFTA breakdown.

Limitations

Using NiGEM has its upsides, but there are also some limitations to our methodological approach. We merely assess the impact of a NAFTA breakdown from a macro economic standpoint, which means we are unable to reflect on the impact in different sectors, firms or products classes. This requires a completely different approach, for instance by using a model such as Worldscan (see the CPB study on Brexit: CPB (2016)) or GTAP (see Ciuriak et al., 2017a). Moreover, even a sectoral approach does not capture all disrupting impact on integrated supply chains between the NAFTA partners (see Erken and Tulen (2017) for more information on supply-chain integration between the US and Mexico). An empirical approach also fails to capture any geopolitical ramifications of changes to a trade agreement that could increase tensions between countries. A final limitation of our approach is that we disregard dynamic productivity effects in this study. In NiGEM, these productivity effects are more or less fixed, as labour-augmented technological change in the supply side of the model is exogenous. We know, however, that the dynamic productivity effects can potentially be very large, and even double the negative economic impact in case of a sudden trade shock (e.g. HM Treasury (2016), CPB (2016) and Erken et al. (2017)). Dynamic productivity models for the US, Canada and Mexico are necessary to include dynamic productivity effects in the tariff model of NiGEM. In a follow-up study to this one, we plan to cover dynamic productivity effects in the US and Canada of a NAFTA breakdown using separate productivity models for these counties. Unfortunately, we lack sufficient data to develop such a model for Mexico.

We will now discuss the most important results of our scenario analyses. We will discuss the impact on each NAFTA partner separately, as this enables us to discuss all interlinkages per country integrally.

The US

The US will face relatively moderate losses to economic growth in either the hard or soft NAFTA breakdown scenario. In both scenarios, the largest GDP growth losses of roughly 0.5ppts will materialize in 2019 (figure 10). Cumulative GDP losses up till 2025 are 1ppts compared to how the US economy would perform without a NAFTA collapse. Despite these relatively mild losses, the bill expressed in terms of costs per person employed would be as high as USD 1,180 in case of a soft NAFTA breakup and USD 1,250 in case of a hard breakup.

The collapse of NAFTA has major implications for export volumes in the US (figure 11). In our soft scenario, export volume growth in the US drops to 0% and even declines in a hard NAFTA breakup scenario by -2.5% in 2019.

The fact that the total economic impact in the US in both scenarios is roughly equal can be traced back to the different inflationary paths in both scenarios. In both scenarios, US inflation spikes in 2019 due to higher trade barriers, and consequently, higher trade prices (see figure 12).

However, higher inflation in our soft scenario is more pronounced than in our hard scenario due to a US economy that is experiencing a higher positive output gap (i.e. production vis-à-vis potential production) and more stringent labour market conditions which raises nominal wages and results in overall higher production costs. Furthermore, in a hard NAFTA breakdown scenario the US faces lower import prices due to the sharp depreciation of the Mexican peso and Canadian dollar, which has a downward impact on inflation. Consequently, higher inflation weighs on private consumption to a larger extent in our soft scenario compared to the hard NAFTA breakdown scenario. Lower private consumption growth in the soft scenario makes up for the larger negative trade impact on the US economy in the hard scenario. This makes sense as trade only constitutes 13% of total US GDP, whereas private consumption is responsible for 70%.

In case of a hard NAFTA breakup, the Canadian economy would be hit the most in the short run. The reason is that of the three NAFTA partners, Canada’s economy is most dependent on trade. Canadian GDP growth would slow down to 0.2% in 2019, whereas without a breakup we expect Canadian growth to arrive at 2% in that year (figure 13). Ultimately, cumulative GDP losses up till 2025 would be as high as -1¼% in a soft NAFTA breakup and -2% in a hard breakup. Although this impact seems limited at first glance, keep in mind that a hard breakup would imply that Canada misses out on the equivalent of one full year of economic growth. Or interpreted differently: total GDP losses up till 2025 are USD 1,800 per Canadian employee in case of a hard breakup against the situation where NAFTA remains in place. In case of a soft scenario, the bill per person employed would end up at roughly USD 950.

The negative impact caused by a NAFTA withdrawal by the US has its roots in imploding Canadian exports (figure 14), which is the result of higher export prices due to trade barriers, and lower growth of Canadian private consumption (figure 15). Exports in the hard scenario plunge by -5.5%, which can be explained by the very open character of the Canadian economy and its large dependency on the US as a destination for its exports.

With respect to private consumption, we already forecast declining growth rates due to very high private debt levels. Indeed, Canadian households are amongst the most highly leveraged in the OECD and in 2017Q3, household debt to disposable income reached a record high of 171%. This is an ongoing concern for Canadian officials and the Bank of Canada explicitly highlighted this in its first Monetary Policy Report of 2018, stating that: Elevated levels of household debt are likely to amplify the impact of higher interest rates on consumption, since increased debt-service costs are more likely to constrain some borrowers, forcing them to moderate their expenditures. Here lies the Bank of Canada's dilemma in that consumption is likely highly sensitive to rising interest rates. Furthermore, what does not bode well for private consumption is the very high cost-push inflation that Canada would be confronted with due to the trade barriers imposed by the US (figure 16).

A NAFTA breakdown would have negative consequences for the Canadian labour market as well (figure 17). Unemployment levels have only recently been coming down from a surge right after the global economic meltdown. A hard NAFTA breakdown would result in rising unemployment to levels above 7%. In case of a soft breakdown, the downward trajectory that unemployment was on will be stopped and unemployment levels will rise to 6.4% in 2019.

Mexico

The impact on Mexican economic losses are more distributed over time compared to the effects in Canada and the US (figure 18). This is due to the fact that a strong depreciation of the Mexico peso against the USD would cushion a lot of the short-term pain that higher trade barriers might inflict on the Mexican economy. The depreciation of the peso makes Mexican exports to the US more competitive. The annual losses in economic growth in both scenarios does not exceed 0.6ppts vis-à-vis our baseline anywhere between 2018 and 2025. However, the negative effects on the Mexican economy tend to persist much longer than in Canada and the US. This is due to slow recovery of the export growth, as Mexico’s export market is hit hardest by the NAFTA breakup compared to Canada and the US. This can be explained by the composition of Mexico’s export market: Mexico has the highest share of its exports heading to NAFTA partners, and the lowest share of exports to the EU. The cumulative GDP loss in 2025 is the highest of all three NAFTA partners: -2.2ppts in a soft NAFTA breakup scenario and -2.6ppts in a hard NAFTA breakup scenario. In terms of costs per employee, we are talking about USD 500 to USD 575 per Mexican worker. These costs are lower than in the US and Canada, but given the much lower level of wealth per capita in Mexico, a NAFTA breakdown would come with a relatively much steeper price tag for the average Mexican compared to the average US American or Canadian.

In both scenarios the decline in export volumes is roughly equal (figure 19), with the larger decline measured in 2019: -0.2% in the soft NAFTA breakup scenario and -1.2% in the hard scenario. Export prices in a hard NAFTA breakup are subject to a larger export price increase in Mexico as a result of higher imposed trade barriers compared to the soft scenario (17.3% in hard versus 6.9% in soft). However, this effect is counterbalanced by a much steeper depreciation of the Mexican peso against the USD (17.5% in hard against 10% in soft), which result in lower Mexican export prices in USD terms, and consequently stimulates a larger market share. All in all, the negative effects on export volumes even out in both scenarios.

The large depreciation of the peso mitigates the impact on exports, but at the same time results in surging cost-push inflation (figure 20). We expect inflation to rise temporarily to a staggering 12% in 2019, which would have a stark negative impact on private consumption.

Impact on the Netherlands

As a NAFTA breakdown would lead to small export substitution effects to the European Union, an open economy, such as the Dutch one, might benefit from trade restrictions being implemented across the Atlantic (figure 21). Given the relative limited trade that the Netherlands has with Mexico and Canada, the additional trade that might rise from a NAFTA breakdown is limited as well. In 2019, export growth in the Netherlands will be 0.5ppts higher in a soft NAFTA breakup (3.8% versus 4.3%) compared to the baseline and even 1ppts in a hard scenario (3.8% versus 4.8%). In a hard NAFTA breakdown, this would lead to 0.4ppts higher growth of the Dutch economy in 2019, which is equal to EUR 3bn additional wealth in the Netherlands.

Differences in methodology, assumptions and reported output variables complicate a comparison of our results to other recent studies that have looked into the economic impact of a NAFTA collapse. Two recent studies are from Ciuriak et al. (2017b) and Yalcin et al. (2017).

Ciuriak et al. (2017b) use a sectoral trade model (Global Trade Analysis Project or GTAP). Their reported effects illustrate the long-term impact, after all the dust has settled and do not take into account adjustment costs. We, in contrast, do report the cumulative impact on GDP towards the end-point. Ciuriak et al. (2017b) find that a NAFTA lapse would generate long-term GDP losses in Canada of -0.5% GDP (permanent effect in 2023). However, in scenarios where CUSFTA holds or Canada and Mexico continue to adopt NAFTA as their bilateral trade regime, the impact on Canada GDP would even be slightly positive. The US would face losses ranging between -0.09% and -0.04%. Mexico takes the largest GDP losses in 2023 of approximately -1.1%.

The upside of the methodology used by Ciuriak et al. (2017b) is that they can map the negative impact in different sectors. Their results show that in Canada, business services would take the biggest hit in case of a NAFTA collapse. But also, chemicals, rubber and plastics complex and automotive sector would face a decline in Canadian exports to NAFTA partners. Of all the F&A sectors, the Canadian beef sector would face the largest losses (exports are expected to decline by USD 500mln). For the US, automotive sector imports would decline by USD 22bn. Due to this lower foreign competition, the US automotive sector would be able to seize a larger domestic market share of 7.7bn, but at the same time would also face lower exports to NAFTA partners. The overall net effect is that the US auto industry experiences a decline in total sales. Moreover, US agricultural exports would take a hit by billion-dollar export declines in pork and poultry, beef and dairy. Unsurprisingly, Mexico’s most affected area in case of a NAFTA breakdown is the automotive sector, especially light truck assembly. Production in that segment would be ceased in Mexico immediately and relocated to the US in order to avoid a 25% tariff. Across food and agriculture, Mexico is able to seize market share in a number of sectors, such as pork and poultry, dairy, beef, and cereal grains.

Yalcin et al. (2017) use the ifo Trade Model, which is a static, general equilibrium model of international trade. The authors assume that the US will install MNF tariffs to its NAFTA partners and introduce the full set of non-tariff barriers that correspond to the costs that were eliminated due to NAFTA. Ultimately, the authors report an effect on real annual household income in the long term of roughly -1.5 percent for Canada, Mexico -1.0 percent and the US -0.2 percent.

Some studies have also looked at effects of a NAFTA breakdown on specific sectors, such as Johnson (2017) for the F&A sector and Head and Mayer (2016) for the automotive sector. The latter study concludes that withdrawing from NAFTA would reduce the US’ share of auto production because it would force US car manufacturing an expensive reverse North American automotive supply chains. Rabobank’s Food and Agriculture Research team have also looked specifically at the impact of changes to NAFTA on the dairy sector (Milking NAFTA, Baily and Perry (2017)).

The negotiations of NAFTA so far have been far from smooth. We are currently in the sixth round of negotiations with the seventh (and currently) final round scheduled to be held next month and the target is for negotiations to be completed before the end of 2018Q1. Given the nature of negotiations, it is difficult to distinguish rhetoric from reality but the threat of a NAFTA break-up is very real. In this Special we studied the macroeconomic effects of two potential scenarios: a partial dissolution of the NAFTA (soft breakup scenario) and a total breakup of the trade agreement (hard breakup scenario).

In a hard NAFTA breakup scenario, Canada and Mexico would take the heaviest losses in terms of gross domestic product (GDP) up till 2025. The two countries would lose 2% and 2.6% respectively when compared to a situation where NAFTA remains in place (our benchmark scenario). For Canada, economic growth in 2019 would fall to 0.2% compared to an expected growth rate of 2.0% in our benchmark scenario. The US would face total GDP losses of 1%. US economic growth in 2019 would drop to 1.6% compared to an expected growth rate of 2.2% in our benchmark scenario. This implies that any positive impact from the recent tax cuts could be wiped out completely by a trade conflict with the two neighboring countries that have become integral parts of US supply chains. In absolute terms, the economic fallout in a hard NAFTA breakup scenario is roughly USD 30bn to USD 33bn in Canada and Mexico respectively. In the much larger US economy, the absolute losses would add up till USD 195bn by 2025. These absolute losses imply that the average Canadian worker would miss out on USD 1,800 of additional wealth up till 2025 and the average US worker would miss out on USD 1,250 USD. The price tag for each Mexican worker could be as high as USD 575, which is less than in its Northern neighbors, but in relative terms it is much higher given the much lower level of wealth per capita in Mexico

In our soft NAFTA breakup scenario, the economic fallout in the US and Mexico are roughly as high as in a hard breakup scenario. In the US, the less severe negative trade effects in the soft scenario are counterbalanced by higher inflation and, consequently, lower private consumption. Higher in the soft scenario is caused first of all by higher import inflation, as the Canadian dollar and Mexican peso depreciate to a lesser extent than in the hard scenario. Moreover, there is more upward price pressure in the soft scenario due to a more positive output gap and stringent labour market conditions, which raises nominal wages and production costs.In Mexico, the negative trade effects in both a hard and soft NAFTA breakup scenario are roughly equal as the impact on export prices of higher trade barriers in the hard scenario is counterbalanced by a much steeper depreciation of the Mexican peso against the USD (17.5% in hard against 10% in soft) which consequently lowers Mexican export prices. Canadian GDP losses up till 2025 in a soft NAFTA breakup scenario are lower than in our hard breakup scenario: 1.3%

It is important to note that the negative fallout of a NAFTA breakup could be higher as adverse effects on labour productivity have not been taken into account, nor have disruptions of internationally integrated supply chains, or potential geopolitical ramifications that could emerge if trade wars ensue.

Trade barriers

Under the NAFTA regime, most goods traded between Canada, Mexico and the US are currently traded subject to no tariffs. Before NAFTA came into force in January 1994, Mexico applied substantial tariffs on imports from the US and Canada. Average tariffs on US imports to Mexico were levied at 10% in 1993, whereas the US only targeted Mexican import at 2,07% in 1993 (see Villareal and Fergusson, 2017). The trade relation between the US and Canada was already largely liberalized by the Canada-US Free Trade Agreement (CUSFTA).

All three NAFTA partners are members of the World Trade Organisation (WTO). In case of a hard NAFTA breakdown, trade will be subject to the trade regime of the World Trade Organisation (WTO). This means that goods trade is subject to a so-called Most Favored Nation (MNF) tariff. The current MFN tariff for the NAFTA partners are: US: 3.5%, Canada: 4.1% and Mexico: 7%. Canada and the US could revert to the zero tariffs during 1989 and 1993 from NAFTA’s predecessor CUSFTA. However, in case of a hard breakup scenario, we assume that the US does not want to have any bilateral trade agreements in place with any of its NAFTA partners, including Canada. Hence, in a hard breakup both countries will not revert to CUSFTA, and will impose MNF tariffs on each other as well (Canada: 4.1% and the US: 3.5%). Three recent examples show that the relationship between the US and Canada has soured over trade since Trump has been sworn in as president of the US last year. First, the Trump administration has introduced levies on Canadian softwood lumber which average 20.2%. The move was heavily criticized by the Canadian government which vowed to retaliate (Bloomberg, 2017). Second, Canada put the trade relationship with the US under pressure by filling a complaint to the World Trade Organisation, listing 200 cases of trade wrongdoings of the US against Canada and other countries (The Guardian, 2018). Third, the US has escalated its dispute with Canada over aircraft maker Bombardier, by introducing a levy of 80% to sales by the Canadian aircraft maker to the US, after reports that these aircrafts were sold below cost price to Delta Air Lines in 2016 (see The Guardian, 2017). And the trade woes are not over yet as the US Department of Commerce is planning to step up the tariffs to a staggering 220%. These examples illustrate that the US is not reluctant to make far-reaching decisions as far as trade with Canada is concerned, at least when it comes to specific sectors or cases.

We assume that Canada and Mexico maintain a status quo, as the 91 NAFTA article 2205 provides that the agreement remains in place for the remaining countries in the trade agreement. In a soft NAFTA breakup, we assume that the US, Mexico and Canada will continue goods trade at no tariffs, but trade will be subject to non-tariff barriers.

Non-tariff barriers

A withdrawal from NAFTA will have consequences for non-tariff barriers (NTB) to trade, such as customs charges, permits, quotas, trade regulations, import licenses, border restrictions, etc. Before NAFTA was effectuated, US-Mexican trade was subject to severe non-trade barriers such as sanitary and phytosanitary (SPS) rules, Mexican import licensing requirement and US marketing orders (Villareal and Fergusson, 2017). Marketing orders are standards for, e.g., product quality and market promotion. Food and agricultural products from Mexico such as oranges and avocados were subject to these orders.

It is hard to quantify by how much NAFTA has brought down non-tariff barriers and it is even harder to assume what will happen with NTB’s if NAFTA were to collapse. Economic literature shows that the ad valorem tariff equivalent (AVE) of non-tariff barriers have dropped markedly in the US and Mexico (see Table 1 for an overview of two important studies). Our latest readings of NTB levels are an average for the period 2002-2011 from Ghodsy et al. (2016). Abbyad and Herman (2017) provides a survey of different studies looking into the trade costs of non-tariff barriers. Egger et al. (2015) have calculated that the cost reductions of NTB’s in case of a deep and comprehensive free-trade agreement, such as NAFTA, could be as large as 13.7%. The exiting country, i.e. the US, would lose out on these costs reductions, and the remaining trading partners are also confronted with barriers from the exiting country.

We don’t expect NTB cost reduction by NAFTA to diminish completely in case of a US withdrawal. Based on the progress made over time, we assume that in case of a hard NAFTA breakup, Mexican exports to the US will be subject to 50% of the NTB cost reduction under NAFTA of 13.7% (i.e. 6.9ppts), which is roughly equal to the NTB difference for the US of 6.4ppts calculated by Kee et al (2009) and Ghodsy et al. (2016), see Table A.1. For Mexico, we assume that imports shipped from the US will face an additional levy of to 75% of the NTB cost reduction of 13.7% (i.e. 10.3ppts), which is very close to the difference of 11ppts-12.3ppts shown in Table A.1. The Canadian NTB’s appear to be very stable over time (Table 1). Moreover, the US and Canadian economy are more aligned in terms of institutions[6], so we assume that trade between Canada and the US will be subject to only 25% additional NTB costs pf 13.7% (i.e. 3.4ppts). In case of a soft breakup, we assume that tariffs will not apply in the newly-negotiated FTA, but NTB’s will still apply. In the soft scenario, Mexico won’t reverse the progress in bringing down NTB’s under NAFTA seen since the turn of the century but will raise NTB’s by the same amount as the US.

Total trade barriers in two scenarios

Ultimately, we arrive at a change in total ad valorem trade barriers illustrated in Table A.2. In case of a hard breakup, Mexican exports to the US will face additional trade costs of 10.4ppts. US exports to Mexico will be subject 17.3ppts additional trading costs and will face an additional ad valorem tariff equivalent of 7.5ppts when shipped to Canada. Finally, Canadian exports to the US will face a total duty of 6.9ppts. In case of a soft breakup, we assume tariffs on goods trade are not raised, but NTB will inevitable emerge under the slimmed-down free-trade agreement.

[6] For instance, the Canada-United States Regulatory Cooperation Council (RCC) has an important task to align the regulatory framework and reduce trade costs in areas such as health, safety, and environmental protection.

Timelines

It is also important to reflect on the actual timing of changes in the trade regime between the three NAFTA partners after, as notification of a withdrawal does not mean that trade barriers kick in immediately. Economists and legal specialists disagree on the timelines and mandate that the US president has in withdrawing from NAFTA (e.g. Bloomberg, 2017; Politico, 2017). Article 2205 of 91 NAFTA allows any party to withdraw from the trade agreement six months after notifying the other two countries in writing of its plan. The timing of the implementation of trade barriers is included in Section 125 of the Trade Act of 1974 which contains a subsection e) Continuation of duties or other import restrictions after termination of or withdrawal from agreements. The section states that NAFTA rates would be continued for one year, unless within 60 days after the withdrawal the President proclaims restoration of tariff rates to what they would be without the FTA (Murrill, 2016).[7] In the case of Mexico, this would be the MNF tariff, but the US and Canada would probably fall back on the trade regime CUSFTA prior to NAFTA. It is unclear, however, if the proclamation of prior rates would remain in effect for only one year or longer (Villareal and Fergusson, 2017). We assume that non-tariff barriers will emerge six months after the US formally withdraws from NAFTA (written notice). See Table 2 for an overview on the timeline of the assumed economic shocks.

Approval of Congress

There is no consensus among legal experts as to whether or not the President of the United States can withdraw from NAFTA without approval from Congress. According to the US Constitution, authority over international trade is apportioned between the President and Congress. However, the Supreme Court has suggested that the President possesses exclusive constitutional authority to communicate with foreign powers. So if the President’s unilateral withdrawal from NAFTA can be characterized as an exercise of the President’s power to communicate with foreign sovereigns, the court might be more likely to uphold such an action. However, legal experts disagree whether this characterization is valid.

According to international law, the President’s delivery of a notice of withdrawal to other NAFTA partners appears sufficient to release the US from its NAFTA obligations from the effective date of withdrawal onward. What’s more, Section 201 of the NAFTA Act enables the President to proclaim additional duties following consultation with Congress. ‘Consultation’ is not the same as approval; a meeting between the US Trade Representative and other cabinet members with the Senate Finance Committee and the House Ways and Means Committee should suffice. However, the US Constitution gives Congress the authority to impose duties and to regulate commerce with foreign nations. Therefore, federal statutes implementing NAFTA cannot be repealed by the President.

Finally, Congress may also try to avert a withdrawal from NAFTA by blocking unrelated policy initiatives by the President. After all, Congress is responsible for government spending. What’s more, the Senate can also refuse to confirm the President’s nominees for various positions, such as the Supreme Court and the Federal Board of Governors.

Exchange rates, prices and trade

The export market share between the NAFTA partners is assumed to change after a NAFTA breakdown. Trade shares are subject to changes in relative export prices, which in turn are affected by rising trade barriers and exchange rate effects. Below, we will separately discuss these elements, starting with the exchange rate paths.

Exchange rates

The exchange rate forecasts in our two NAFTA scenarios for the Canadian dollar (CAD), Mexican peso (MXN), United States (USD) dollar and euro (EUR) were calculated by our FX specialists. The paths of USD/MXN and USD/CAD are illustrated in figure A.1. In a hard NAFTA breakdown, we expect the CAD to depreciate by 10.4% against the USD immediately after the announcement of a US withdrawal in March. The decision would likely see the Bank of Canada take its foot off the monetary tightening pedal and the market would price out future implied rate hikes thus leading to a depreciation of the currency.

[7] More in general, the President must recommend within 60 days the appropriate rates of duty for all articles affected by the withdrawal.

The MXN faces a shock depreciation of 17.5% of its value against the USD. The Mexican peso is the only LatAm currency that is fully deliverable, fully convertible and tradeable 24 hours a day and this characteristic, combined with the fact it is the 10th most liquid currency in the world, results in MXN being a high beta currency that is sensitive to changes in risk sentiment and prone to overshooting in times of increased stress. Given this, we expect the MXN would recoup some of its initial losses in the months following the US’ decision to leave NAFTA but it would remain at levels weaker than during its time as a NAFTA currency.

In a soft scenario, the CAD still faces a depreciation of 6.4% in March 2018 and MXN by 10%. The euro gradually appreciates against the euro in a soft NAFTA scenario. In a hard scenario we expect a stable EUR/USD.

Exchange rate pass-through on export prices and trade shares

The depreciation of the CAD and MXN partly mitigates the negative impact of trade barriers on export prices. In dollar terms, Canadian and Mexican companies exporting to US become cheaper in US dollar terms, but are also forced to raise prices of their goods in dollar terms, as intermediates from the US (or which are denominated in USD) become more expensive. There is only a limited pass-through on export prices, however, for two reasons. First, exporters mainly assess the long to medium term when determining their export strategy and thus will also incorporate a partial reversal of the CAD or MXN. Second, exporters use the increased demand for their products to increase their profit margins instead of increasing market share. Economic literature finds a general exchange rate pass-through on export prices of 55% (IMF, 2015). However, exchange rate pass-through to export prices varies substantially between countries, depending on the inelasticity of demand of their export basket and market power.

Bussière et al. (2014) have calculated for a set of countries that the exchange rate pass-through on export prices is quite substantial.[8] For Canada, an exchange rate pass-through (ERPT) is reported of 0.387. This means that a depreciation of 1% will feed into lower export prices in USD terms by 0.713%. This is in line with recent research by Devereux et al. (2017), who generally find a pass-through of 0.4 to 0.6, depending on the exporters’ market share of companies. For Canada, therefore, we adopt an ERPT of roughly 50%. For Mexico, Bussière et al. (2014) find a pass-through of 0.557 (which is equal to a pass-through of 44.3% in USD terms), which we will use in our calculations.

[8] Canada: 0.387, Mexico: 0.557 and US: 0.35.

Trade shares are reduced by multiplying our estimated increase in export prices (due to tariffs and non-tariff barriers, and at the same time taking the depreciation of currencies into account) to export elasticities found in the literature. These elasticities are taken from a study by the European Commission (Imbs and Mejean, 2010). For Mexico, we use an elasticity of -2.9, which means that an increase in export prices by 1% results in a decline in export volumes by 2.9ppts. The US has an export price to trade elasticity of -2.2 and Canada an elasticity of -3.8. Ultimately, we expect in a hard scenario that export volume from Mexico to the US will decrease by -11.6%, from Canada to the US by -17.8%, from the US to Canada by -15.7% and from the US to Mexico by -34.7%. In a soft scenario the decline in export volume is as follows: Mexico ® US: -11.8%, Canada ® US: -7.4%, US ® Canada: -8.8% and US ® Mexico: -23.8%.

Exchange rate pass-through on CPI

In contrast to the very scarce amount of studies looking into the ERPT on export prices, there is a vast literature examining EPRT on domestic prices. For Mexico, an ERPT on domestic prices appear to have dropped markedly after adopting inflation targeting 2001 (e.g. Mihaljek and Blau (2008), Dilla et al. (2017) and Baharumshah et al. (2017)). Moreover, covering the period 2011 and 2016, the Mexican Central Bank (Banxico) shows an exchange rate pass-through of 0.074 on CPI (Kochen and Samáno, 2016). This means that a 1% depreciation of the Mexican peso leads to a rise in consumer prices by 0.074ppts. We use this factor to adjust the impact of the MXN depreciation on import prices.[9] For Canada, Savoie-Chabot and Khan (2015) report a pass-through on consumer prices of roughly 0.06.

[9] The elasticity in NiGEM by which FX affects inflation is 0.33. In order to lower this elasticity to 0.073, we multiply the impact of exchange rates on inflation by a factor 0.22 (=0.073/0.33). For Canada, the elasticity is 0.2, which leads to a multiplier of 0.3 (=0.06/0.2).

Export substitution effects

Despite some disagreements over brand names, Mexico and the EU are on the brink of renewing their bilateral trade deal, which will upgrade the current one that came into force in 2000. Negotiations on a new deal were launched in May 2016. The question is if a new deal could become a mitigating factors in case NAFTA is torn apart by the US and Mexico could possibly substitute. Erken and Heijmerikx (2018) evaluate the impact of the FTA between Mexico and the EU dating back from 2000 on bilateral trade. This effect is subsequently used to proxy a potential substitution effect of Mexican exports to the EU in the potential event of a NAFTA breakdown. They find a robust and statistically significant impact of the FTA between Mexico and the EU on exports, raising total bilateral trade between 4 and 5 percent. In terms of 2016 trade figures, this comes down to approximately USD 750mln and USD 1bn additional Mexican exports due to the FTA.

We use the elasticity from Erken and Heijmerikx (2018) to mitigate part of the loss in Canadian and Mexican export due to a NAFTA breakdown. For Canada, we assume that the Comprehensive Economic and Trade Agreement (CETA) between Canada and the EU will make up for 5% of the decline in export volume to the US due to the NAFTA breakdown. For Mexico, we assume an effect of 7,5%, as the newly expected bilateral trade agreement is a considerable upgrade compared the one that came into force in 2000. In order to calculate the substitution trade effect, we link the percentage decline of Canadian and Mexican exports to the US to the trade flows to the US of approximately USD 300bn in 2016 for both countries (in USD). In a hard NAFTA breakup scenario, we expect Mexico to lose out on USD 35bn of export to the US and Canada will lose even more: USD 53bn. For Mexico, this implies that USD 3bn of the export losses to the US will be made up for by the renewed FTA between Mexico and the EU, which is equal to an increase of Mexican export to the EU of 13.6% (=3bn/19bn). In a similar fashion, the trade substitution effect of Canadian export to the EU is 8.8% of total export.[10]

[10] We phase in the trade substitution effects, as we assume that the FTA with the EU will only be able to reach its fully mitigating potential after one year into force. If we assume the new trade deal between Mexico and the EU will be in place in 2018Q1, the trade substitution effects in a hard NAFTA scenario will be: 2018Q3: 3.4%, 2018Q4: 6.8% and 2019Q: 13.6%. For Canada, the substitution effect will kick in directly, as CETA came into force in Q3 of 2017. Hence, the Canadian substitution effects we use in a hard NAFTA scenario are: 2018Q1: 2.2%, 2018Q2: 4.4% and 2019Q3: 8.8%.

Foreign direct investment

A NAFTA breakup will probably also have a negative impact on foreign investment between NAFTA partners. The reduction of barriers between the US and Mexico because of NAFTA has led to an increase of FDI flows between both countries. Consequently, a NAFTA breakdown would reverse that trend and lead to a decline in FDI flows. The US has always been the largest foreign direct investor in Mexico, but the US share of Mexican inward FDI has increased dramatically since NAFTA, at the expense of for example the UK and Germany. Geographical proximity, low labor costs and less strict environmental regulations – in addition to reduced barriers – have been incentives for US firms to invest in Mexico (Pacheco-Lopez, 2005).

The impact of NAFTA on Canadian inward FDI has been counterintuitive. Intuitively, the reduction of barriers between Canada and the US should increase the incentive for firms to locate in Canada because of easier access to the much larger North American market. However, at the same time firms can locate in the US closer to larger markets and have easy access to the Canadian market. An empirical analysis by Hejazi and Safarian (2002) shows that – controlling for a range of other determinants of FDI – the second effect has been dominant: NAFTA has had a negative impact on Canadian inward FDI. Consequently, a NAFTA breakdown and the possible threat of increased trade barriers between Canada and the US is likely to have – although counterintuitive – a positive impact on Canadian inward FDI. However, the absolute size of the impact of NAFTA on Canadian inward FDI is about half the size of the impact on Canadian outward FDI (Hejazi and Safarian, 2002). This is not surprising given the opposing effects on Canadian inward FDI.

Based on the empirical evidence (Ciuriak et al. (2017b); Hejazi and Safarian (2002); Pacheco-Lopez, 2005), we make the following assumptions regarding the impact of a NAFTA breakdown on FDI flows. US inward FDI flows fall by 5%, while Mexican inward FDI plunges by 20%. In contrast, Canadian FDI inflows increase by 2.5% (table A.3).

Impact on business investment

The expected impact of a NAFTA breakdown on FDI patterns have to be translated to business investment to get a general idea of the magnitude of the shocks. It appears that the shocks are very small. In the US, FDI consists roughly 20% of all business investment. BEA data shows that the contribution of NAFTA partners to total US inward FDI position of USD 3,725bn is limited (i.e. the Mexican share is actually in the graph). These low shares combined with expected investment decline due to a NAFTA breakdown (Table A.2) would lead to lower total business investment of -1.2% due to lower investment from Canada and -0.1% due to lower investment from Mexico. Hence, this would hardly result in significant negative effects for the US economy.

For Canada, the share of FDI in total business investment on average is 33% and the US makes up for a large chuck of these investments (roughly 50%). However, as the literature shows that we can even expect a positive effect on Canadian inward investment from the US (+2.5%), we hardly register any impact on the Canadian economy from an foreign investment point of perspective.

US FDI in Mexico are important as well: about 55% of the FDI position in Mexico is due to investment from the US. What mitigates the impact of a potential large decline of 20% in US FDI in Mexico, is that the share of FDI in total business investment is on average only 18%, so this mitigates the economic impact substantially. All in all, the Mexican gross fixed capital formation would on average lose roughly USD 20bn of investment over up till 2025, which is just over 10% of total expected average annual private investment of USD 175bn in that period. This affects Mexican domestic demand by roughly -0.5%.

Abbyad, N. and P. Herman (2017). How much do non-tariff measures cost? A survey of quantification methods. U.S. International Trade Commission.

Alcalá, F. and A. Ciccone (2004). Trade and productivity. The Quarterly journal of economics, 119(2), 613-646.

Baharumshah, A. Z., Sirag, A., & Soon, S. V. (2017). Asymmetric exchange rate pass-through in an emerging market economy: The case of Mexico. Research in International Business and Finance, 41, 247-259.

Bailey, T. and Perry, A. D. (2017). Milking NAFTA – Changes to NAFTA and Scenarios for Dairy. Rabobank.

Bassanini, A. and S. Scarpetta, (2002). Does human capital matter for growth in OECD countries? A pooled mean-group approach. Economics Letters, 74(3), 399-407.

Branstetter, L. (2006). Is foreign direct investment a channel of knowledge spillovers? Evidence from Japan’s FDI in the United States. Journal of International Economics, 68(2), 325-344.

Bussière, M., S. Delle Chiaie and T.A. Peltonen (2014). Exchange rate pass-through in the global economy: the role of emerging market economies. IMF Economic Review, 62(1), 146-178.

Cameron, G., Proudman, J., and S. Redding (2005). Technological convergence, R&D, trade and productivity growth. European Economic Review, 49(3), 775-807.

Ciuriak, D., Ciuriak, L., Dadkhah, A., & Xiao, J. (2017a). The NAFTA renegotiation: what if the US walks away?. C.D. Howe Institute Working Paper.

Ciuriak, D., Ciuriak, L., Dadkhah, A., & Xiao, J. (2017b). Quantifying the termination of NAFTA. C.D. Howe Institute Working Paper.

Coe, D.T. and E. Helpman (1995). International R&D spillovers. European Economic Review, 39(5), 859-887.

Devereux, M. B., Tomlin, B., & Dong, W. (2015). Exchange rate pass-through, currency of invoicing and market share (No. w21413). National Bureau of Economic Research.

Dilla, S., Achsani, N. A., & Anggraeni, L. (2017). Do Inflation Targeting Really Reduced Exchange Rate Pass-through?. International Journal of Economics and Financial Issues, 7(3), 444-452.

Edwards, S. (1998). Openness, productivity and growth: what do we really know? The Economic Journal, 108(447), pp. 383-398.

Engelbrecht, H.J. (1997). International R&D spillovers, human capital and productivity in OECD economies: An empirical investigation. European Economic Review, 41(8), 1479–1488.

Egger, P., Francois, J., Manchin, M., & Nelson, D. (2015). Non-tariff barriers, integration and the transatlantic economy. Economic Policy, 30(83), 539-584.

Erken, H.P.G., P. Donselaar and A.R. Thurik (2016). Total factor productivity and the role of entrepreneurship. Journal of Technology Transfer, 1-29.

Erken, H.P.G., R. Hayat, M. Heijmerikx, C. Prins en I. de Vreede (2017). The permanent damage of Brexit. Rabobank.

Erken, H.P.G. and M. Tulen (2017). US global value chain integration: a major impediment for Trump’s protectionist trade agenda. Rabobank.

Es, F. and I. de Vreede (2017), Visie op 2018: groei wereldeconomie blijft in dezelfde krachtige versnelling, Rabobank.

Felbermayr, G., M. Steininger & E. Yalcin (2018). Quantifying Trump: The Costs of a Protectionist US. Ifo Institute-Leibniz Institute for Economic Research at the University of Munich.

Ghodsi, M., Grübler, J., & Stehrer, R. (2016). Estimating importer-specific ad valorem equivalents of non-tariff measures. Wiener Institut für Internationale Wirtschaftsvergleiche.

Griffith, R., Redding, S. and J. Van Reenen (2004). Mapping the two faces of R&D. Productivity growth in a panel of OECD industries. Review of Economics and Statistics, 86(4), 883-895.

Grossman, G.M. and E. Helpman (1991). Innovation and growth in the global economy. Cambridge, University Press.

Head, K. and T. Mayer (2016). Reversal of regional trade agreements: Consequences of Brexit and Trumpit for the multinational car industry. VoxEU.

Hejazi, Wahid and A.E. Safarian (2002). Explaining Canada’s changing FDI patterns, Rotman School of Management, University of Toronto.

Imbs, J., & Mejean, I. (2010). Trade elasticities: a final report for the European Commission. Directorate General Economic and Financial Affairs (DG ECFIN), European Commission, no.432.

IMF (2015). Adjusting to lower commodity prices, Washington.

Kochen, F., & Sámano, D. (2016). Price-Setting and Exchange Rate Pass-Through in the Mexican Economy: Evidence from CPI Micro Data. Banxico.

Kee, L.H., A. Nicita and M. Olarreaga (2009). Estimating trade restrictiveness indices. The Economic Journal, 119(534), 172-199.

Mihaljek, D., & Klau, M. (2008). Exchange rate pass-through in emerging market economies: what has changed and why?. BIS Papers, 35, 103-130.

Murrill, B.J. (2016). U.S. withdrawal from free trade agreements: frequently asked legal questions. Congressional Research Service.

Noland, M., G.C. Hufbauer, S. Robinson, and T. Moran (2016). Assessing Trade Agendas in the US Presidential Campaign. Peterson Institute for International Economics.

Pacheco‐López, P. (2005). Foreign direct investment, exports and imports in Mexico. The World Economy, 28(8), 1157-1172.

Savoie-Chabot, L., & Khan, M. (2015). Exchange rate pass-through to consumer prices: Theory and recent evidence. Bank of Canada, Bank of Canada Discussion Paper, No. 2015-9.

Scott, R.E. (2011). Heading South: US-Mexico trade and job displacement after NAFTA. Economic Policy Institute Briefing Paper, 308.

USITC (2003). The impact of trade agreements: effect of the Tokyo Round, U.S.-Israel FTA, U.S.-Canada FTA, NAFTA, and the Uruguay Round on the U.S. Economy. Publication 3621.

Villareal, M., & Fergusson, I.F. (2017). The North American Free Trade Agreement (NAFTA), Congressional Research Service.