Trading with Bullies

Economics for a Hard-Power World

Watching the US impose 25% tariffs on Canada and Mexico yesterday, I was stunned by how our trade landscape has changed. Two countries that bet everything on trade integration with the United States — sending it over 80% of their exports — find themselves treated as economic adversaries. This isn't an isolated incident but part of a transformation where trade increasingly becomes weaponized as another tool in the geopolitical arsenal.

This shift has led many economists and policymakers to rethink how countries should approach trade policy. Some call for a retreat from free trade principles, arguing that the traditional economic rules only work in peaceful times. They turn towards mercantilism, embracing generalized industrial policy, worrying about trade deficits, and choosing to restrict trade — all in the name of security.

I believe this approach is wrong. Abandoning free market principles would be throwing the baby out with the bathwater. Rather than discarding sound economic thinking, we should adapt it to account for national security concerns.

Economics in a Hard Power World

One of the first principles students of economics learn is that markets do a remarkably good job allocating resources—except when market failures arise. Government intervention makes sense when it targets these specific failures with the least intrusive tools available. This framework isn’t outdated—we just need to recognize an additional market failure: the ‘national security externality’ (Kooi, 2025).

While the outbreak of conflict has drawn all our attention to the importance of national security, war is the exception not the rule. Countries want to achieve their geopolitical ambitions, but the destructive nature of conflict makes it in their mutual interest to avoid it. The problem of national security is therefore often not about the application of force but the exploitation of potential force (Schelling, 1960). The core problem is strengthening one's bargaining position — one of the benefits of having a large military is that you may never need to use it.

Bargaining power derives not only from the ability to win a conflict (military power) but also from one's resilience in withstanding its costs. Crucially, while governments control military power, many decisions that affect resilience are made by the private sector. Private companies decide whether to source chips from Taiwan or produce them at home, choices that have strategic implications beyond their commercial calculations.

The national security externality exists because private actors do not account for how their decisions affect their government's bargaining power through resilience to conflict. When a US company imports cheap Chinese chips and builds infrastructure dependent on them, it creates a strategic vulnerability — a cost not reflected in market prices.

Once we account for this security externality, standard economic thinking gives us clear guidance. Markets underinvest in resilience because they fail to price the strategic value of certain productive capabilities. Society overall would accept lower economic returns for investments that only pay off in crisis scenarios, but businesses do not (Murphy and Topel, 2013). Excess capacity in healthcare or defense appears inefficient in normal times but is invaluable during emergencies.

The US CHIPS Act shows this thinking in action. Despite higher costs compared to Asian production, domestic chip factories address a strategic weakness that became obvious when chip shortages halted manufacturing. The US Defense Department has been doing the same for critical minerals, creating stockpiles against future disruptions. Similarly, the EU's Economic Security Strategy mitigates critical bottlenecks through diversification or strategic reserves, providing insurance against coercion without dismantling the broader free trade architecture.

Building resilience

How can governments intervene effectively without risking ‘anything goes’? The basic rule is that policy interventions should target sectors where production capacity would become scarce during the conflict, as measured by potential price increases relative to peacetime (Kooi, 2025). The optimal subsidy is proportional to how much the price of a sector's capital goods would increase during the specific conflict being prepared for.¹

Prices increase when the demand curve shifts outwards, and the elasticity of demand is low. As Alfred Marshall showed, demand for an input (such as chips or energy) is inelastic when the demand for the final output is itself inelastic or when few substitutes are available. For instance, semiconductor capacity during a Taiwan crisis would see dramatic price increases due to their critical role as inputs, limited substitutability, and concentrated production. The same is not true of corn.

Kooi's analysis reveals a powerful insight: no sector is inherently ‘strategic’ in an absolute sense. Instead, what makes a sector strategically important depends entirely on the specific conflict scenario being considered. Suppose conflict takes the form of a trade dispute. The European Union faces radically different vulnerabilities in a trade dispute with China than in one with the USA.

The optimal subsidy for any sector depends on how much prices would rise in that sector in a particular conflict. This ratio may vary dramatically depending on which adversary and which type of conflict we're preparing for. For instance, in a trade dispute with the USA, the vulnerabilities would be in high technology sectors. Using the criteria of high concentration, few substitutes, and low final demand elasticity, the prices that could be expected to increase most would be critical semiconductors and computing hardware, where Europe fully depends on Nvidia, AMD, Intel, and Qualcomm, and US-dominated cloud services from AWS, Microsoft Azure, and Google Cloud for essential business infrastructure. The aerospace industry shows similar dependencies, with Airbus requiring US suppliers like RTX, GE Aviation, and Honeywell for aircraft engines and avionics systems. In healthcare, the US leads biotech innovation through companies like Pfizer, Moderna, and Amgen, making the EU dependent on American firms for advanced biologics and mRNA technology.

This is why blanket industrial policies that designate certain sectors as universally ‘strategic’ often miss the mark. The economic approach instead requires policymakers to think clearly about specific conflict scenarios, model the resulting price effects across different sectors, and target resilience-building investments accordingly.

Once we accept that trade flows can become instruments of political influence, it is clear a dominant state can deliberately concentrate trade ties with a smaller country to create dependency and then exploit that dependency for strategic gains. But because power concentrations are highly nonlinear, even modest diversification of critical dependencies can yield ‘much economic security with little overall fragmentation’ (Clayton, Maggiori, and Schreger, 2024).

For instance, if one country controls almost all of a vital input, losing access to it is extremely disruptive. But introducing just one or two alternative suppliers dramatically decreases vulnerability because it prevents the dominant supplier from dictating terms or coercing others by threatening supply cutoffs. Hence, rather than reshoring everything, the goal should be avoiding extreme dependency on any single foreign supplier for essentials. This approach preserves most of the efficiency gains from global trade, while effectively addressing strategic vulnerabilities.

When integration enhances security

Contrary to what the new mercantilists would claim, deeper economic integration can sometimes be the best strategic response. A ‘conditional free trade strategy’ suggests that if a country can credibly threaten punitive trade sanctions in a future conflict, it should promote integration in peacetime (Becko and O'Connor, 2025). Deeper trade ties make potential adversaries more dependent on you, increasing your leverage.

The U.S.-EU energy relationship demonstrates this dynamic. After Russia's gas cut-off, Europe turned heavily to American LNG, with the U.S. supplying nearly half of EU imports by mid-2024. This new dependency gives the US significant leverage.

A Europe that faces increasingly transactional American trade policy, rather than decoupling, needs to look at which economic ties create leverage versus vulnerability. In sectors where European products are essential to American industries, deeper integration actually enhances strategic autonomy. The most obvious case here is industrial machinery, where German, Dutch, and Italian firms (Siemens, ASML, KUKA, Trumpf) dominate in high-precision industrial machinery, used in semiconductor production, aerospace, and advanced manufacturing. These tools are difficult to replace with U.S. or Asian alternatives.

The important caveat here is that this approach only works when you have credible options for sanctions. If a country lacks such leverage—perhaps because it's too small or its exports aren't essential to the rival—then diversification policies are more warranted.

The risk of overreaction

The risk of security policies expanding into broad protectionism is significant. If every country independently pursues ‘economic security’ policies, the result is inefficient duplication that wastes the gains from trade. Individual responses fail to account for how coordinated action would benefit the entire group of rules-respecting, market-oriented economies (Clayton, Maggiori, and Schreger, 2025).

What the mercantilists — so-called ‘post-neoliberals’ — want to do is use the existence of some legitimately strategic industries as justification for large-scale intervention across many sectors. The risk is that aiming to obtain votes, politicians are liable to find almost any investment — from agriculture to any high technology — as deserving protection on national security grounds.

Using economic analysis in a hard-power world helps in two ways. First, it explains the problem better. Those who think all trade with rivals is dangerous can't explain why Russia worked so hard to create energy dependencies through Nord Stream 2. Second, only economic models allow us to state the cost of these things accurately. Yes, the new mercantilists might tell you that you're justified in giving money to your chip industry, but soon, you'll find yourself billions of dollars deep trying to save a state-backed champion and wondering how you ended up there.

Preventing this risk requires serious work. Objective criteria linking subsidies to expected price appreciations during conflicts provide a concrete, measurable standard that makes it harder to justify interventions based on vague security claims.

The existence of a national security externality does not mean every other economic cost ceases to exist. In fact, the story is the opposite — precisely our fragile moment means that the stakes of growth and prosperity are much higher. Now is the worst moment to embrace the boondoggle.

References

Becko, John Sturm, and Daniel O'Connor. 2025. "Strategic (Dis)Integrationy." Working Paper. January 2025.

Clayton, Christopher, Maggiori, Matteo, and Schreger, Jesse. 2024. "A Theory of Economic Coercion and Fragmentation." NBER Working Paper No. 33309.

Clayton, Christopher, Maggiori, Matteo, and Schreger, Jesse. 2025. "The Political Economy of Geoeconomic Power." NBER Working Paper No. 33353.

Kooi, Olivier. 2025. "Power and Resilience: An Economic Approach to National Security Policy." Working Paper, University of Chicago.

Marshall, Alfred. 1890. Principles of Economics. London: Macmillan and Co.

Murphy, Kevin M., and Topel, Robert H. 2013. "Some Basic Economics of National Security." American Economic Review 103 (3): 508-511.

Schelling, Thomas C. 1960. The Strategy of Conflict. Cambridge, MA: Harvard University Press.

1. According to Kooi (2025) optimal subsidy for capital in sector i is proportional to

\([(r_i^C / r_i^P) - 1]\)

Here the r_i^C and r_i^P are the prices of sector i capital during conflict and peace respectively. If semiconductor manufacturing capacity would become 10 times more valuable during a Taiwan crisis, the ratio equals 10.