The trade tensions between Canada and the United States since 2025 provide a
particularly telling example of the vulnerability that high trade concentration can create when a political shock suddenly alters the conditions of access to the main export market. For the GVC and CIRANO, this situation also highlights the analytical value of the digital twin (DT): rather than simply observing the consequences of a tariff policy after the fact, the simulator allows for the exploration of multiple scenarios in advance and the identification of the sectors and trade corridors most at risk. The developments observed in 2025 and 2026 are consistent with several of the mechanisms illustrated by the DT, though they do not constitute a predictive validation of the model.
From Simulated Scenarios to Actual Impacts
The simulations presented in the GVC’s digital twin are based, in particular, on three scenarios: no additional tariff, a 10% additional tariff, and a 25% additional tariff. They reveal a clear relationship: the higher the U.S. tariff barrier, the more the projected trajectory of Canadian exports to the United States deteriorates. This sensitivity is particularly evident for vehicles, aluminum, wood products, and petroleum and mineral fuels (GVCdtLab, tariff simulator, 2026 screenshots).
In the case of vehicles, for example, the DT reports approximately $69 billion in Canadian exports to the United States in 2024 and projects approximately $57 billion for 2025 without additional tariffs, $53 billion under a 10% scenario, and $47 billion under a 25% scenario (GVCdtLab, tariff simulator, 2026). The observed reality is consistent with the contraction mechanism illustrated by the simulator. After the
U.S. tariff measures took effect in the spring of 2025, Canadian exports to the United States fell sharply, particularly in automobiles and light trucks (Statistics Canada,
2025). More broadly, Canadian merchandise exports to the United States declined by 5.8% for the full year 2025 (Statistics Canada, 2026).
However, the comparison should not be made on a dollar-for-dollar basis. The categories used in the simulator and those used in the official trade statistics are not necessarily identical, and, more importantly, th DT primarily highlights the effect of a tariff shock, whereas the actual trade figures observed result simultaneously from tariffs, demand, prices, exchange rates, production capacity, and other economic factors. The simulator’s value therefore lies more in its ability to identify the direction and potential magnitude of the shock than in the exact prediction of a future value.
The case of aluminum is particularly instructive. The DT showed a decline in the export trajectory as the tariff rose from 0% to 10%, then to 25% (GVCdtLab, tariff simulator, 2026). However, the shock actually implemented exceeded even the DT’s most severe scenario. The United States imposed a 25% tariff on Canadian aluminum on March 12, 2025, before raising it to 50% on June 4, 2025, with no USMCA-related exemptions for these products (Statistics Canada, 2025).
This exposure was particularly significant: in 2024, Canadian aluminum exports totaled $13.4 billion, and more than 94% were destined for the United States. After the tariffs were imposed, monthly aluminum exports fell to an average of about $848 million between April and August 2025, a decline of more than 24% compared to the monthly average for 2024 (Statistics Canada, 2025).
For 2025 as a whole, exports of raw aluminum and aluminum alloys declined by 7.7 percent, but a particularly interesting trend emerged: shipments to countries other than the United States rose from approximately $670 million to $2 billion (Statistics Canada, 2026). This trend is consistent with one of the mechanisms that the digital twin seeks to highlight: a tariff shock cannot only reduce bilateral trade but also trigger a geographic reallocation of trade flows.
This shift is most clearly evident in Statistics Canada’s data. When comparing exports recorded through August 2025 with those for the entire year of 2024, Canadian aluminum exports to the Netherlands had already increased by 74.3 percent, those to Italy had nearly doubled, while those to Poland had quadrupled (Statistics Canada, 2025).
Forest products and energy products also highlight the need to distinguish between tariff simulation and comprehensive macroeconomic forecasting. The DT shows how higher tariffs reduce projected export trajectories, but the actual observed results are also influenced by changes in world prices, demand, and transport capacity. This difference is therefore not a weakness of the DT; rather, it defines its role as a tool for counterfactual simulation, allowing for the isolation of a specific economic mechanism.
A Concentration in the U.S. Market That is Beginning to Decline
At the level of the Canadian economy, a shift is now measurable. In 2024, 75.9% of
Canadian merchandise exports were destined for the United States. This proportion
fell to 71.7% in 2025. During the same period, exports to the United States declined by
5.8%, while exports to countries other than the United States increased by 17.2% (Statistics Canada,2026).
Total merchandise trade with countries other than the United States thus rose from 484 billion dollars in 2024 to 553 billion in 2025, an increase of 14.3 (Statistics Canada, 2026).
Data directly related to companies also show that this shift is not limited to a few high value trade flows. In 2025, the number of Canadian firms exporting to the United States decreased by 542, or -1.3%, while the number of firms exporting to destinations other than the United States increased by 292, or +1.8%. The proportion of Canadian exporting companies serving markets outside the United States thus rose from 34.1% to 34.8% (Statistics Canada, 2026). It would nevertheless be premature to interpret these results as an economic decoupling between Canada and the United States. Geographical proximity, existing infrastructure, the USMCA, and several decades of value chain integration will continue to strongly favor north-south trade.
Diversification should therefore be understood not as a replacement for the U.S. market, but as a two-pronged strategy: reducing the risk of concentration and, when costs justify it, taking advantage of better market opportunities. This distinction is important: the geography, size, and historical integration of the U.S. market explain
much of the concentration of trade, but they do not prove that every observed trade flow is economically optimal. CIRANO’s research on cost-based comparative advantage reveals significant heterogeneity: certain North American supply chains, particularly in the automotive and energy sectors, are already highly efficient, while other Canadian products offer untapped cost advantages in third-party markets. The challenge, therefore, is not to trade less with the United States, but to distinguish between trade flows that are already efficient and those for which an alternative destination could create greater value (Warin, 2025).
Carney-Tump: From Negotiations the Suspension on August 21
The events of August 2026 show that the tariff dispute has entered a new phase.
After several weeks of intensive negotiations, the United States agreed on August 18 to postpone the implementation of new 50% tariffs on a range of Canadian products until the end of August 21. Ottawa stated at the time that progress had been made, but that significant issues remained to be resolved (Office of the Prime Minister, August 18, 2026).
This sequence is important: it shows that access to the U.S. market no longer depends solely on existing trade rules, but also on political negotiations whose parameters can change rapidly. On August 21, Canada suspended negotiations following last-minute changes to the U.S. terms, which were deemed unfair, unprofitable, and likely to undermine the reliability of an agreement. The central issue is therefore no longer just the level of a tariff, but the predictability of the market access regime.
The Canadian government’s position remains explicitly twofold: to preserve the best possible access o the U.S. market for Canadian companies and, at the same time, to diversify partnerships abroad. These objectives are not contradictory. The first safeguards the benefits of deep North American integration; the second expands companies’ options when market access conditions deteriorate (Office of the Prime Minister, August 21, 2026). Diversification should therefore be viewed as a complement to integration, not as its opposite. It can also create value when another market offers better economic conditions.
The resulting political climate can be described as pragmatic, but highly transactional. Political channels remain open, but the conditions for access can now be changed rapidly depending on the outcome of negotiations. This instability increases the value of credible commercial options for companies. It does not mean that the U.S. market is becoming secondary; rather, it increases the cost of concentration. The problem, therefore, is not proximity to the United States, but the lack of options when access deteriorates.
The uncertainty does not stem solely from the tariff levels themselves. It also stems from the use of tariffs as a bargaining tool in several areas, including automobiles, steel, aluminum, and the rules governing preferential access. The upcoming revision of the USMCA adds another layer to this uncertainty. Trade resilience thus becomes a matter of choosing among options, not just the level of tariffs.
The ACEUM remains a central pillar of this relationship, but its revision could alter the incentives that shape value chains. In the automotive sector in particular, any change in rules of origin or conditions for preferential access could affect the location of investments, sourcing, and competitiveness. This possibility underscores the value of tools capable of quickly identifying, on a product-by-product basis, economically viable alternative markets, without assuming that the relationship with the United States is either always optimal or easily replaceable.
The Canada–U.S. relationship therefore does not represent a break, but rather a relationship in which market access remains essential while having become more contingent. For businesses, the strategic value of diversification lies precisely in the ability to avoid dependence on a single option.
From Commercial Risk to a Resilience Strategy
August 21, 2026 marks a turning point in this dynamic. The suspension of negotiations and the announcement of new U.S. tariffs of 50 percent on approximately $28 billion worth of Canadian goods confirm that the relevant risk is no longer just the level of a tariff, but the possibility of a rapid change in market access conditions. For businesses, this uncertainty has an economic cost of its own: it affects decisions regarding investment, capacity, pricing, and location even before trade volumes have a chance to adjust.
This trend does not lead to the conclusion that integration with the United States was a mistake. Rather, it prompts us to distinguish between two issues. The first is the concentration of risk: a company whose sales depend on a single market is more vulnerable to a shock specific to that market. The second is efficiency: a large volume of trade is not, by itself, proof that this destination is the best economic outlet.
At the corporate level, diversification can therefore play a dual role. By analogy with portfolio theory, spreading sales across markets whose risks are not perfectly correlated reduces exposure to a specific shock. But the argument goes beyond risk mitigation: certain alternative markets may also offer better economic conditions. Diversification can thus simultaneously improve both resilience and business performance.
It is precisely in this context that the CCA takes on its full significance. The goal is not to seek a substitute for the U.S. market at any cost, but to identify, product by product, the markets where the cost to Canada is competitive compared to existing suppliers. Diversification thus becomes selective: North American supply chains are maintained when they are efficient, and other markets are developed when they offer a credible economic advantage. It is companies—not the country as an abstract entity—that actually make these trade-offs.
From Tariffs to Trade Redistribution: The Strategic Value of CCA
In this context, the Cost-Based Comparative Advantage (CCA) module developed by the GVC directly complements the tariff simulator.
The simulator essentially seeks to answer the question:
What might happen to Canadian exports if the cost of accessing the U.S. market increases?
The CCA addresses the following question:
If this market becomes less attractive, to which other markets might certain Canadian companies redirect a product when their landed cost there is competitive compared to that of suppliers already operating there?
This second question is crucial because diversification cannot simply consist of shifting the same products to just any market. Competitiveness depends, in particular, on the value of the product, its cost per kilogram, transportation costs, the size of the import market, and the international competitors already present.
The GVC CCA thus compares the costs associated with Canadian exports to those of international competitors and identifies, at the highly detailed HS 6 level, the markets in which Canada might have a comparative economic advantage. Screenshots from the digital twin show, for example, this analysis for unwrought aluminum and alloys (HS 760120), certain stainless steel products (HS 722211 and HS 721810), automotive suspensions and shock absorbers (HS 870880), certain wood products (HS 440791), and refined petroleum products (HS 271012) (GVCdtLab, CCA module, 2026).
The maps thus distinguish between markets with relatively favorable and unfavorable positions for potential trade diversion. The purpose of the CCA is therefore not to predict that a given country will necessarily purchase a specific quantity of Canadian goods, but to provide an economic ranking of potential markets.
The data collected since the introduction of U.S. tariffs provide a particularly useful empirical illustration of this logic. The shift in aluminum exports to the Netherlands,
Italy, and Poland, alongside the relative decline of the U.S. market, is a concrete example of the reorientation of trade flows (Statistics Canada, 2025).
At the aggregate level, this shift is also evident: Canadian merchandise exports
to destinations outside the United States increased by 17.2% in 2025, while exports to the United States declined by 5.8% (Statistics Canada, 2026).
However, an important caveat must be noted. Statistics Canada estimates that a substantial portion of the increase in exports to destinations outside the United States in 2025 came from gold, silver, and platinum group metals, particularly against a backdrop of strong demand for safe-haven assets. Even excluding these products, however, Canadian exports to markets outside the United States increased by approximately $14 billion, while those to the United States decreased by nearly $31 billion (Statistics Canada, Spring 2026).
This nuance precisely underscores the value of the CCA: an aggregate increase in exports outside the United States is not sufficient to conclude that successful structural diversification is underway. It is necessary to determine which products, to which markets, with what cost advantage, and with what degree of sustainability can
actually be redirected—without assuming that the historical destination was necessarily the economic optimum.
What this implies for the GVC and CIRANO
The trends observed since 2025 are consistent with a three-step analytical approach: measuring exposure, simulating the shock, and identifying opportunities for reallocation.
First, the digital twin shows that deep trade integration with the United States represents both a considerable economic advantage and a concentration risk. The tariffs introduced since 2025 have transformed this risk, which was previously largely hypothetical, into a concrete economic constraint.
Second, the observed results are consistent with the central mechanism illustrated by the simulator: when the cost of accessing the U.S. market increases, exports to that market tend to decline. Statistics Canada reports, in particular, a 15.8% drop in Canadian exports to the United States in April 2025, as companies began to adapt to the new U.S. measures (Statistics Canada, 2025).
Third, the data show that this contraction is already accompanied by some geographic diversification. The U.S. share of Canadian exports fell from 75.9% to 71.7% between 2024 and 2025, and exports to other markets increased by 17.2% (Statistics Canada, 2026).
Fourth, this diversification should not be interpreted as a strategy aimed at replacing
the United States. It must be evaluated on a company-by-company and product-by-product basis. In some cases, North American supply chains are already efficient; in others, third-party markets may offer a better cost position combined with a reduction in concentration risk. Diversification can therefore improve both resilience and business performance.
Finally, the suspension of negotiations on August 21 confirms that even a potential resumption of talks would not automatically restore the pre-2025 trade environment. The Canadian government is now explicitly pursuing two objectives: preserving the best possible access to the U.S. market and accelerating the diversification of partnerships abroad. These objectives are complementary: they combine trade
effectiveness, strategic autonomy, and risk management (Office of the Prime Minister, August 21, 2026).
The main risk, therefore, is no longer just the current tariff level, but the uncertainty surrounding its future trajectory.
In this context, the main strategic value of the GVC’s digital twin may not be to determine whether the next U.S. tariff will be exactly 10%, 25%, or 50%. Rather, it lies in the ability to quickly answer two key questions for decision-makers:
What portion of Canadian value chains becomes vulnerable when a tariff is imposed? And what trade alternatives are economically viable if this risk materializes?
The combination of the Tariff Impact Simulator and the Cost-Based Comparative Advantage provides precisely this complementary analysis. The former allows us to assess the vulnerability of a sector or product to a tariff shock; the latter allows us to examine opportunities for trade diversion based on Canadian competitiveness.
The events of 2025–2026 thus give these tools very concrete significance. The data show both a contraction in trade flows to the United States, growth in exports to other markets, and, for certain products such as aluminum, the emergence of new trade destinations. This development suggests that diversification is no longer merely a forward-looking issue: it has already become one of the mechanisms through which Canadian companies are adapting to the new North American trade environment.
The relevant conclusion for the GVC and CIRANO is therefore not so much that Canada is moving away from the U.S. market as that it is engaging in selective diversification—both defensive and offensive. Canada will likely maintain an extremely important trade relationship with the United States. But we must avoid two symmetrical errors: assuming that any decline in trade with the United States
constitutes a loss, or that any diversification constitutes a gain. The challenge is to identify, company by company and product by product, the markets that enhance resilience, economic efficiency, or both.