Geoeconomics

Cathrin Mohr (University of Bonn) and Christoph Trebesch (Kiel Institute for the World Economy). Annual Review of Economics 17 (2025), pp. 563–587; open access. This is the published version, and it supersedes Kiel Working Paper 2279 of January 2025, from which an earlier version of these notes was written. Twenty-five journal pages, body text pp. 564–579, Literature Cited pp. 579–587. No talk recording exists. The published version has no tables, figures or equations, so nothing is cropped from it. It is the first survey of the field and the row above Block 1 on this list.

A field is a definition plus a reading list

A survey of a three-year-old field cannot summarise results, because there are not many yet. What it can do is draw a border and furnish the rooms inside it, and that is what this paper is: a definition, five subfields, and roughly 230 references sorted into them, a page or two of annotated bibliography each with a sentence on what is known and one on what is not. The authors concede the problem in the act of solving it — the field “is only beginning to emerge and consolidate” (p. 564).

The definition is the load-bearing part. Geoeconomics, for Mohr and Trebesch, is “the field of study that examines the links between geopolitics and economics,” with geopolitics borrowed from the French geographer Yves Lacoste as “the study of rivalries for power or influence over territories and the people who inhabit them” (Lacoste 2006, p. 9, quoted at p. 565). So: economics where the other party is a rival, and the rivalry is about power. This is deliberately wide, and the authors say so by measuring it against the narrower definitions in circulation — there are now three rungs, because the published version added one. Their definition “overlaps with” Thoenig’s “study of the interaction between trade, diplomacy, and geopolitics,” but “ours is broader in that it also includes international finance and other arenas beyond trade” (pp. 565–566); the closest match is Chatham House’s, geoeconomics “as the interplay of international economics, geopolitics and strategy” (p. 566). Anyway,

Our definition is broader than that of Clayton et al. (2023), who define geoeconomics as a strategy in which “governments use their countries’ economic strength from financial and trade relationships to achieve geopolitical and economic goals.” Their framework focuses primarily on economic power and economic warfare, while ours deliberately extends to issues of actual warfare as well—including military finance, arms production, or the economic drivers and consequences of external wars. Our definition is also broader than that of Blackwill & Harris (2016, p. 20), who define geoeconomics as “the use of economic instruments to promote and defend national interests, and to produce beneficial geopolitical results.” (p. 566)

Blackwill and Harris, the authors note, are restating what political scientists used to call economic statecraft — routed here through the Encyclopedia Britannica’s gloss on Baldwin, “the use of economic means to pursue foreign policy goals” (p. 566) — and Mohr and Trebesch want the new word to mean more than the old one: not just tools but the whole two-way traffic between power rivalry and the economy, including the part where the rivalry becomes a war. Two auxiliary concepts are fixed alongside: geopolitical risk in Caldara and Iacoviello’s sense, “the threat, realization, and escalation of adverse events associated with wars, terrorism, and any tensions among states and political actors that affect the peaceful course of international relations,” and geoeconomic fragmentation in the IMF’s, “a policy-driven reversal of global economic integration often guided by strategic considerations” (both p. 566).

How you survey a field that barely exists

The Kiel working paper answered that question in a sentence. It said it picked recent papers that were highly cited “and/or that were presented at newly established conferences on the topic,” and it named them: the Kiel-CEPR Annual Geoeconomics Conference, launched in 2022, and the NBER Summer Institute session on international economics and geopolitics, launched in 2024. The published version keeps both conferences and deletes the rule. They now appear as evidence rather than as a filter — “recent conferences give indication of where the field is heading” (p. 564) — which is a claim about the field instead of a claim about the bibliography. The survey no longer says how its bibliography was assembled.

The accent it was admitting to is still audible. The Kiel-CEPR conference is Trebesch’s institute’s, it is named on the first page of text, and eight of the works in the Literature Cited have Trebesch as a coauthor out of roughly 228 entries, clustered in the finance and war-finance rooms. That is not a complaint, since the alternative was no map, but a reader should know whose hand drew it, and the reader now has to work that out from the reference list rather than from a sentence. The published version also adds something that reads like an answer to a referee asking whether any of this is new: a paragraph placing the survey inside existing literatures on trade and power (Hirschman, Mansfield, Martin and coauthors, Jackson and Nei) and on hegemony and reserve currencies (Nurkse through Gourinchas and Rey, Farhi and Maggiori, Eichengreen on exorbitant privilege), at p. 565. The honest reading of that paragraph is that a field is a definition plus a reading list, and this one has been assembled out of two reading lists that already existed.

The paper’s own account of why the field revived has two layers. The surface layer is events: weakening American hegemony, China’s rise, Russia’s invasion of Ukraine, Trump’s transactional trade policy, the 2022 sanctions and the German gas debate that followed (pp. 564, 568, 570). The deeper layer is historical. Between the 1920s and the 1970s Hicks, Hirschman, Keynes, Leontief, Schelling, von Neumann and others worked on coercion, deterrence, war finance and bottlenecks, and that work was then “overshadowed by the more famous work on peacetime issues” (pp. 564–565). The authors expect today’s turbulence to turn a cohort of economists back toward security, and they are plain about the starting point: “Most economists today (including ourselves) have only limited training or expertise on questions of security, war, or international political tensions” (p. 565). The intended reader is a graduate student, and the paper is written as a syllabus.

Tools: the best-understood weapon, and a verdict that changed

Sanctions are called “the best-known geoeconomic policy tool and also the best-understood” (p. 567), and here the published version does something worth stopping on. The Kiel draft summarised the empirical literature in a subordinate clause hung off a citation string — trade sanctions, financial sanctions, central-bank asset sanctions, smart sanctions, “with overall mixed results.” The published version deletes that clause and puts a full paragraph in its place:

Overall, the empirical literature on sanctions shows that sanctions, by and large, tend to achieve their stated economic objective, in particular by hampering trade flows and access to the international financial system. However, there is disagreement on how large the effects are, and more work needs to be done with regard to financial sanctions, which are only starting to be explored in depth. (p. 568)

The list of papers cited immediately above that sentence is the same list in both versions. Nothing was added to the evidence base; the summary sentence changed. “Mixed results” became “by and large achieve their stated economic objective,” with the disagreement demoted from the finding to its magnitude. You can tell a charitable story about this — a referee pushed back, the authors reread their own citation string, and the pooled reading of it really is that sanctions cut trade and financial access, which is all “stated economic objective” claims — and that story is probably right. It is still a survey changing its verdict on a literature without the literature changing, which is a thing that happens more often than surveys admit, and a reader who quotes the published version on sanctions should know that the earlier draft, with the same footnotes, said the opposite.

The paragraph then gives back some of what it took. Targeted governments “can effectively take measures to shield certain industries or firms from the effects of sanctions, but this is costly and increases the overall cost of sanctions within the country,” and what determines who can shield, and with what long-run effects, is open (p. 568). Smart sanctions — the ones aimed at the elite — need more work outright. The frontier is firm- and account-level data, which “facilitates the identification of causal effects,” a phrase that quietly concedes something about the older literature. The crisp theoretical result is Itskhoki and Mukhin’s: asset freezes and export sanctions depreciate the target’s currency, import sanctions appreciate it (p. 568). And the de-dollarisation claim now carries a hedge that anyone quoting it must carry too: sanctions may undermine the dollar as a trade invoicing currency and as a reserve currency in targeted countries, “although the aggregate effect of this seems to be limited” (p. 568).

Embargoes now arrive with their general finding stated first, which the draft did not do. Drastic embargoes, especially the world-war ones, “can have a major economic impact and cause much suffering to the civilian population in the targeted countries, sometimes causing more deaths than the actual war” — and yet substitution shows up fast, so that “the adoption process in response to embargoes can be surprisingly quick and effective, with high elasticities of substitution” (p. 568). The German gas debate is then the instance rather than the rule: forecasts of a GDP fall of up to 12 percent against Bachmann and coauthors’ mild recession, and the verdict that they “were proven right, as German GDP barely shrunk after Russia (not Germany) imposed a unilateral gas embargo” (pp. 568–569). Tariffs sit in a separate drawer from sanctions in the standard databases because they are aimed at audiences at home, though the line is blurring. Export controls are “little research” and “many open questions,” with the two available results pointing in opposite directions: American controls on Chinese firms worked but cost American firms revenue and market value with “little evidence for reshoring,” while Chinese rare-earth controls induced innovation downstream abroad (p. 569). Sabotage is “not understood yet” beyond the finding that cyberattacks cost firms value and propagate through supply chains (pp. 569–570); aid is geopolitical in direction and timing whether or not anyone says so.

Trade: Hirschman, then three modern questions

The trade subfield opens with a history of thought from mercantilism through Smith and Hume to Hirschman’s 1945 book, and the claim that the mercantilist idea, that stronger countries extract gains from trade from weaker ones by threatening to cut them off, “resurfaced again and again, especially in times of geopolitical upheaval” (p. 570). The modern papers are sorted into three questions.

Coercion. Clayton, Maggiori and Schreger’s hegemon solves the “notorious problem of limited enforceability of international contracts” by coordinating joint threats, amplified through input-output networks, and thereby plays a dual role: it extracts rents from smaller countries and supplies a public good by making cross-border contracts enforceable (p. 571). Their follow-up adds the target’s response: decoupling is individually rational and, done by many at once, produces a “fragmentation doom loop” (p. 571). Becko and O’Connor and Kooi have countries subsidise at home to be ready for conflict abroad, which is how the survey rationalises support for domestic industries “even though such support may seem inefficient during peacetime” (p. 571).

Hegemony. Kindleberger’s hegemonic stability, with Gilpin and Krasner’s point that hegemons favour integration out of self-interest, is formalised by Broner, Martin, Meyer and Trebesch: trade rises with alignment, a large hegemon induces alignment around itself, a second large power can unravel it, and 200 years of treaties show that countries aligned with a hegemon trade more with each other, not only with it (pp. 571–572).

Fragmentation. The authors side with Antràs and with Goldberg and Reed against the deglobalisation thesis (the data show “slowbalization”) while accepting that Russia’s attack on Ukraine opened a new era, and quote Goldberg and Reed’s line that “national security is the most powerful argument against unconstrained, market-driven globalization to date,” alongside Adam Smith’s “defence, however, is of much more importance than opulence,” which the published version is careful to mark as cited in Irwin 2023, p. 412 (all p. 572). The reported bottom line is that “trade flows are increasingly being redirected along geopolitical lines, although there is disagreement on how large this divergence really is,” and that fragmentation is costly (p. 572). Supply chains get the great reallocation away from China toward Mexico and Vietnam (p. 573).

Finance: the section that says “underexplored”

Currency dominance is the crowded branch, and the published version leads it with economists rather than with political scientists — dollar dominance and exorbitant privilege first, Kennedy and Strange and Frieden after (p. 573). Liquidity explains four centuries of florin, pound and dollar, and Kennedy’s observation that currency dominance is often the last of the great power privileges to fall is illustrated by sterling still dominating in the early 1920s (p. 573). The end of dollar hegemony “has been wrongly predicted many times before,” and Eichengreen, Gourinchas and Mukhin nonetheless all expect a multipolar order of regional currencies eventually (pp. 573–574). The Chinese numbers are the house specialty: more than $150 billion in bailout lending over a decade, about a fifth of IMF lending in the same period, and a Belt and Road loan book larger than all Western governments’ combined (p. 574).

Capital flows open with “the link between geopolitics and the international allocation of capital is underexplored, both theoretically and empirically” (p. 574). What exists is state-directed lending: Chinese overseas lending is almost entirely state-controlled, larger than thought and priced like private credit; and back to 1790 the rising hegemon of the day has been the dominant official lender, with state flows rising exactly when private flows recede (p. 574). The prediction is that states will again dominate global finance as tensions rise, possibly behind capital controls. The section ends with “More work is needed on the role of geopolitics and hegemony in the international financial system” (p. 575).

Risk and war

Geopolitical risk gets a page, unchanged from the draft but for its page number: the Caldara–Iacoviello newspaper index, its finding that risk shocks lower investment, employment and equity prices, and the literature that regresses inflation and bank lending on it (p. 575). There is no strategic actor anywhere in the section, which matters below.

War is in scope by construction and confined to external wars. The published version adds a sentence the draft lacked, pointing at the game-theoretic literature on war and nuclear deterrence — Schelling, Fearon, Powell — as “highly influential during the Cold War and… seeing a revival today” (p. 575), which is the paper’s own forgotten-ancestry thesis showing up as a citation. The core empirical result is Martin, Mayer and Thoenig’s: bilateral trade lowers the probability of war between two countries, multilateral trade raises it by making any single partner replaceable, so globalisation has made large wars rarer and small ones more common, with the average distance between war parties halving between 1950 and 2000 (p. 576). Thoenig’s extension adds the security dilemma, since the diversification that cures dependence on a rival lowers the cost of conflict with everyone, and yields the policy result that raising US–China trade costs brings “geoeconomic welfare gains” at first and a collapse in welfare if costs grow too large (p. 576). The rest is costs, war finance (most military booms in 150 years were debt-financed, with cuts to social spending “the rare exception,” pp. 577–578) and military production.

What the survey says it does not know

The closing section is a list of gaps, and the gaps are the useful part. Measurement: better measures of alliances and spheres of influence, and “a conceptualization of economic coercion” and “the quantification of geoeconomic rents and their extraction” (p. 578), which admits that the central object of the narrow definition has not been conceptualised, let alone measured. History: the Cold War and 1880–1945, now cheap to mine with machine learning and image recognition. Weapons: the sanctions theory “has been lagging behind” the empirics, and nobody has studied the whole “poison cabinet” of instruments jointly to ask which work and how they interact (pp. 578–579). Finance: “theory and empirical work linking geoeconomics and international financial markets remain scarce” (p. 579). Military economics: China’s and Russia’s buildups, arms supply chains and the “choke points” of industrial production (p. 579). Technology gets one named exception, Konrad 2024, and otherwise a shrug.

Where it sits

The list uses this paper twice. Its broad definition decides what is in scope, which is why a block on the geopolitics of finance and the macro transmission of geopolitical shocks can sit beside the coercion theory at all; the narrower Clayton–Maggiori–Schreger definition, power exercised through threats under limited enforceability, decides what is central. The five subfields map onto the three blocks almost one to one: policy tools and the geopolitics of trade are blocks 1 and 2, the geopolitics of finance and geopolitical risk are block 3, and the economics of war is left out except where it touches trade, through Martin–Mayer–Thoenig and the Fragmentation Paradox. The list’s Hirschman genealogy also comes from here, and the theory papers that follow assume it.

The caveat it imposes is that its breadth makes “geoeconomics” cover two kinds of paper that do not cite each other. One has a strategic agent, a hegemon or a target, choosing threats and dependence; the other treats geopolitics as a shock process and estimates transmission. Both are here, on adjacent pages, under one word, and peer review did nothing to draw the line between them. A reader moving on to the theory papers should keep the distinction the survey does not, and should read the finance and risk sections as inventories of what has not been done.

It is a syllabus written by people who host the conference that decides what is on it. The draft said so; the published version does not, having kept the conference and cut the confession.