Tariffs as National Security Tools
Rethinking Comparative Advantage and Framing National Security
Higgins: It is simple economics. Today it is oil, right? In 10 or 15 years, food, plutonium, and maybe even sooner. Now what do you think the people are going to want us to do then?
The Condor: Ask them.
Higgins: Not now, then. Ask them when they’re running out. Ask them when there’s no heat in their homes and they’re cold. Ask them when their engines stop. Ask them when people who’ve never known hunger start going hungry. You want to know something? They won’t want us to ask them, they’ll just want us to get it for them.
This post includes a brief quotation and clip from Three Days of the Condor (1975), directed by Sydney Pollack and starring Robert Redford and Cliff Robertson, used for analysis and critical commentary in line with fair dealing / fair use principles. All rights belong to the original copyright holders.
Source: © 1975 Paramount Pictures – Three Days of the Condor, directed by Sydney Pollack. Excerpt extracted from YouTube. (This excerpt is used for commentary/educational purposes only. All rights belong to the original copyright holders.)
Last month, at the European Winter Meeting of the Econometric Society, I delivered a keynote speech on a topic that is new for me: “Tariffs for National Security”. A few months earlier, when I was invited to give the European Central Bank plenary lecture, the suggested theme was tariffs. At first, this seemed straightforward. I thought I could talk about tariffs by drawing on some of my existing work and adapting it to discuss their logic and implications.
However, the more I thought about it, the less comfortable I became with that approach. I felt like there were several interesting angles of the current reality that were not captured in the traditional way in which macroeconomists have framed, and continue to, frame tariffs.
Around the same time, I was working with my brilliant colleague Aurélien Eyquem on a project examining the role of essential goods in macroeconomic policy. That work highlighted that some goods matter not because of their weight in the consumption basket, but because of their importance for economic viability. Once this distinction is taken seriously, the standard treatment of tariffs feels incomplete. Comparative advantage remains a useful benchmark, but it is less informative when disruptions affect the trade of goods that are difficult to substitute (e.g., energy inputs, semiconductors, etc.) and are central to the functioning of the economy.
That realization led us to step back and rethink the problem of trade from first principles. We decided to begin with the simplest and most familiar trade model – the one that provides the intellectual foundation for trade policy discussions and, more broadly, for the globalization consensus: the Ricardian framework. Starting from this Ricardian comparative-advantage benchmark allows us to be precise about what breaks, why it breaks, and what follows once essential goods and risk enter the picture.
In what follows, we retrace that research journey and lay out some preliminary insights.
What is the Ricardian model?
For more than two centuries, trade theory has conveyed a simple and powerful message: countries should specialize according to their respective comparative advantages, trade freely, and “everyone” gains. In the Ricardian world, tariffs are distortions, specialization is efficient, and competitive market outcomes are optimal.
The intuition is straightforward: countries differ in productivity across goods, and by specializing in what they do best, they expand the global production frontier. Prices then reflect relative marginal productivities and labor (the only factor of production in the model), flows to the country with the most productive use, thus raising welfare everywhere. In this benchmark framework, there is no meaningful role for trade policy beyond facilitating exchange. Free trade is not just desirable; it is efficient.
This logic, however, rests on a set of key assumptions whose inveterate legitimacy is quickly being corroded by empirical realities. As a quick refresher: goods are perfectly tradable across countries, access to markets is guaranteed, and trade relationships are reliable. Furthermore, goods are treated symmetrically, differing only in productivity, and, crucially, the model abstracts from the possibility that some inputs or goods might be indispensable for the functioning of the economy, or that access to them might fail.
Once we step outside these assumptions, the picture becomes more complicated. Not all goods are the same. Not all components are substitutable. And trade is not always safe or reliable, as recent global supply-chain and trade policy disruptions have made painfully clear.
To be fair, trade economists have long incorporated risk into their analysis. When trade is uncertain and markets are incomplete, diversification can serve as a form of insurance. How strong this motive is depends on the substitution possibilities and on the technology – deep structural features of the economy that interact with risk and shape optimal specialization.
What is missing?
What is largely missing from this literature, however, is a systematic treatment of essential goods: goods that are necessary for economic viability, and an explanation of what happens when access to those goods can fail.
Think of critical medicines, energy inputs, semiconductors, or defense-related components. The exact list evolves as economies undergo industrial, technological, and digital revolutions – but the underlying issue remains the same: losing access to such goods is not a marginal country welfare loss; it can compromise the functioning of the economy altogether.
When trade can break, and when certain goods must be available no matter what, the classical principle of comparative advantage can no longer be taken at face value.
To understand why, we need to reframe the problem in economic terms.
National Security as an Economic Concept
How do we capture all this in a simple way? In pure economic terms, national security is not about flags or borders; it is about guaranteed access to essential goods in bad states of the world. Framed this way, two elements are needed.
First: essentiality.
Some goods have a minimum required level of consumption. Below that threshold, welfare does not merely decline, it collapses. Hospitals need medicines, armies need equipment, and power grids need energy. These are not goods that can be easily postponed or substituted away from. In economic language, this is naturally captured by subsistence requirements in preferences: there is a floor below which consumption cannot fall without severe consequences.
Second: disruption risk.
Access to these goods may fail. Trade routes can be blocked, sanctions imposed, or key nodes in global supply chains turned into chokepoints. In a bad state of the world, imports of an essential good may simply not arrive. This is not price risk, but quantity risk – the physical unavailability of the good.
Put differently, national security is about meeting subsistence needs under disruption risk.
Why do we need both elements? Because neither, on its own, is sufficient to cause serious problems to the economy. Essentiality without disruption risk is not a problem for trade: if markets always function, countries can safely rely on imports to meet even very rigid needs and in bad states of the world. Disruption risk without essentiality, in turn, generates volatility and insurance motives, but not existential concerns: consumption can adjust, welfare falls smoothly, and standard diversification arguments apply.
It is only when essential goods are subject to disruption risk that a fundamentally new issue emerges. In that case, failure to secure access in bad states leads not to marginal welfare losses, but to failures in economic viability. This is the sense in which national security enters the trade problem, and why it forces us to rethink the logic of specialization and free trade.
Why Comparative Advantage is modified
Once we introduce essential goods and disruption risk, the logic behind comparative advantage needs to be modified. If imports of an essential good can disappear in a bad state, then specializing away from it becomes dangerous. In that case, the economy faces a hard domestic capacity constraint: in the worst state of the world, it must be able to produce enough of the essential good at home to meet its minimum needs.
This leads to a stark implication. Even if a country has a comparative disadvantage in an essential good, it may still need to maintain domestic capacity to produce it..
This is not simply insurance in the usual sense of smoothing consumption across states. Insurance addresses fluctuations around a baseline. What is at stake here is economic viability. Below the minimum threshold, adjustment is no longer marginal, and welfare does not decline smoothly – the economy stops functioning altogether. Domestic production of essential goods is therefore not about hedging risks but about avoiding economic failure.
Comparative advantage continues to matter, but it no longer dictates complete specialization. Instead, it determines the cost of security: how expensive it is for a country to maintain the operational capacity needed to survive a bad state.
The One-Country Logic: A Capacity Externality
In our new framework, private incentives naturally push the economy toward specialization: firms follow the logic of comparative advantage – more precisely, they make production decisions based on prices in the good state when trade is open and essential goods can be imported cheaply. Because markets for state-contingent trade are missing, the possibility of a disruption is not fully priced into private decisions. As a result, the bad state is simply not part of their decision problem.
From the perspective of the functioning of the economy as a whole, however, this logic is incomplete. Specialization in the presence of essential goods creates dependence. If the economy fully specializes away from an essential good, it becomes exposed to a chokepoint: a single failure of access is enough to disrupt economic viability.
This is where the social planner’s perspective becomes crucial: the planner recognizes that access to the essential goods must be guaranteed even in the worst state of the world. This requirement translates into a domestic capacity constraint: the economy must maintain sufficient productive capacity at home to meet minimum needs when imports are unavailable.
Producing the essential good domestically, therefore, has a value (and thus a cost) that private firms do not internalize: domestic goods production ensures feasibility in the bad state. The planner internalizes the gap between private and social returns to domestic capacity, a national-security externality that arises from the risk of economic disruption.
Left to themselves, markets underinvest in essential capacity, not because firms behave irrationally, but because the value of maintaining that capacity only materializes in states of the world that markets do not take naturally into account (low probability of realization of the bad state of the world). This is where policy enters.
Tariffs as National Security Policy
The logic of the argument points to the need for national-security policies. Tariffs are one possible tool.
In normal times, a tariff applied to the imports of essential goods raises the import price. That, in turn, encourages firms to keep producing them at home and preserves productive capacity. The key point is that this is not about protecting inefficient firms or favouring special interests; it is about correcting a problem that markets, on their own, cannot solve.
Firms make decisions based on what is profitable when trade is working smoothly. They do not—and cannot— consider the possibility that access to essential goods might suddenly fail. As a result, the economy becomes too specialized and too dependent on foreign suppliers for goods it cannot do without.
A national-security tariff addresses this problem directly. It nudges firms to maintain domestic production of essential goods, even if importing them would be cheaper most of the time. In doing so, it aligns private decisions with what society as a whole needs: the ability to function in bad states of the world.
Seen in this light, tariffs are not a rejection of free-market principles; they are a practical way to ensure economic viability when trade is risky, and some goods are simply too important to leave entirely to global markets.
Conclusions
These are our main findings, but there is clearly more to understand. In our preliminary analysis of a two-country setting, for example, the framework opens the door to studying strategic interactions: how the risk of supply disruptions can be deliberately increased, how tariffs respond to such “weaponization,” and how these forces interact in equilibrium. This is a natural next step, and one that brings the analysis even closer to real-world policy debates.
The framework also speaks directly to current discussions about economic security, supply-chain resilience, and trade fragmentation. Much of the recent debate has been shaped by the insights of Farrell and Newman and by Fishman’s work on chokepoints and economic warfare. Our contribution is to provide a simple economic logic that formalizes these ideas. By focusing on essential goods and the risk of physical disruption, the framework clarifies why dependence itself can become a strategic vulnerability, and why policy responses aimed at preserving domestic capacity are not ad hoc departures from economic reasoning.
At the same time, the analysis connects naturally to the emerging geoeconomics literature, pioneered by Clayton, Maggiori, and Schreger. That literature emphasizes contractual frictions and enforcement as sources of strategic dependence and fragmentation – our mechanism is complementary. We focus instead on physical disruption and viability constraints yet arrive at similar conclusions: when interdependence can be weaponized, free trade no longer delivers robust outcomes, and fragmentation can emerge as a rational response.
The broader message is simple: adapting to a challenging and uncertain world requires understanding its defining features and being willing to revise the intellectual frameworks we use to analyze trade.
References:
[1] Benigno, G. and A. Eyquem (2025) “Tariffs for National Security”.
[2] Clayton, C., Maggiori, M., & Schreger, J. (2024). A framework for geoeconomics. NBER Working Paper No. 31989. See here.
[3] Farrell, H., & Newman, A. (2019). Weaponized interdependence: How global economic networks shape state coercion. International Security, 44(1), 42–79. See here.
[4] Farrell, H., & Newman, A. (2023). Underground empire: How America weaponized the world economy. New York: St. Martin’s Press. See here.
[5] Fishman, N. (2025). Chokepoints: American power in the age of economic warfare. Oxford: Oxford University Press. See here.
This is very interesting, Gianluca. A clear treatment of a very important subject.
I have one question: wouldn’t production subsidies, and not tariffs, be the first-best way to maintain subsistence capacity in critical sectors in your model?
I also have one comment: access to critical goods are exactly the main source of the gains from trade. I have a paper in the JIE called “Why Trade Matters After All” that tries to make this point. So, one has to be careful not to commit economic suicide because one fears economic death.
An alternative path would be to prop up comparative advantage sectors (with subsidies), which produce goods that are critical for other countries. Like this, one could build leverage to deter any weaponization of trade policy. And it would solve the problem by investing in strengths rather than weaknesses.