The US institutions are becoming weaker in governing, enforcing the law, and engaging internationally. The so-called "middle powers" should not wait for the US to recover or improve. Instead, they should start building domain-specific networks among themselves through supply-chain diplomacy, writes Dr Aleksei Turobov.
At this year’s Munich Security Conference, the United States (US) presented itself as two countries. Secretary of State Marco Rubio told allies that “the old world is gone“. Down the corridor, on the other hand, California Governor Gavin Newsom assured Europeans that Donald Trump is “temporary” – gone in three years. The conference’s own report, titled Under Destruction, concluded that the post-1945 international order is being dismantled by the very power that built it.
While there is agreement that the international architecture is cracking, there is no visible blueprint for what will replace it. The instinct to wait for a new US president, for institutional self-correction, for the pendulum to swing, rests on a misdiagnosis. The problem is a structural failure of institutional transmission: the mechanisms that once converted public concern, legal evidence, and democratic dissent into government accountability are degrading simultaneously – across domestic governance in the US, in the rule of law (specifically in international law), and international engagement. Middle powers cannot afford to wait for a self-correction that the American and therefore international system may no longer be able to deliver. They need an alternative operating framework: supply-chain diplomacy – organising cooperation around mapped dependencies and domain-specific leadership distributed across major powers, rather than deals or declarations.
When the institutional machinery stalls
The United States has fallen to its lowest-ever ranking on Transparency International’s Corruption Perceptions Index – 29th, tied with the Bahamas, behind Lithuania and Barbados. Enforcement of the Foreign Corrupt Practices Act has been frozen. Some examples: a firm controlled by the United Arab Emirates’ (UAE) national security adviser purchased a $500 million stake in a crypto venture co-owned by the US President’s family, with $187 million flowing to Trump entities. Months later, the administration reversed a longstanding policy to approve the sale of 500,000 advanced AI chips to the UAE. The scale of corruption is unprecedented, but Congress has held no hearings on the matter, and an ethics review has not been initiated.
Another systemic issue – the Epstein story that has not emerged in 2026. For nearly two decades, one of the most extensive elite criminal networks in modern American history was shielded by US institutions and elites. The original 2008 plea deal, negotiated under a Republican administration, allowed Jeffrey Epstein to avoid federal charges. No deeper investigation followed under the Democratic administration that succeeded it. Epstein died in federal custody in 2019, under circumstances that remain contested. This is a systemic failure – a core institutional foundation sustained by structures that consistently prioritise the protection of the powerful over accountability.
The Epstein files, released by the US Justice Department in January 2026, triggered sharply divergent responses across three democracies. In Norway, the ambassador to Jordan resigned, a former prime minister was charged with aggravated corruption, and parliament launched a rare external inquiry into the foreign ministry. In the United Kingdom (UK), Peter Mandelson resigned from the House of Lords and the Labour Party, the Metropolitan Police opened a criminal investigation, the Prime Minister’s Chief of Staff stepped down, and Andrew Mountbatten-Windsor was arrested.
In the United States, by contrast, the Deputy Attorney General stated there would be no additional prosecutions. Congressional oversight committees have not compelled testimony from those named in the files. Same evidence – three systems – two activated their institutional machinery, and one country where the crimes occurred declared the matter closed.
The same institutional system that cannot convert evidence into accountability at home is being actively dismantled in the international architecture through which it once projected those standards abroad. Washington has withdrawn from the World Health Organization (WHO), the Paris Agreement, and, by presidential memorandum in January 2026, sixty-six additional international organisations. USAID has been gutted. At COP 30 and G20 2025, the US sent no official delegation. This is the structural dismantlement of the architecture through which the US exercised influence for 80 years.
The pattern is the same: the transmission mechanisms that historically converted evidence, dissent, and public concern into institutional accountability are degrading simultaneously. Gavin Newsom’s “dormancy” metaphor assumes the capacity to reawaken, but the evidence suggests that the institutional system of checks and balances is impaired.
From rupture to rewiring
Carney was right to name the “rupture” and to call on middle powers to act collectively. But his “variable geometry” concept – differing coalitions on different issues – is a direction without design and mechanisms. The next steps should start with an honest audit of where structural power actually sits. Middle powers are conventionally grouped by size, GDP, or diplomatic disposition, hence coalitions likeMIKTA (Mexico, Indonesia, South Korea, Turkey, and Australia) or IBSA (India, Brazil, South Africa). These groupings have consistently underdelivered because they are coalitions of the similarly positioned, not the mutually complementary. A more productive architecture would organise cooperation around domain-specific leadership: identifying, within each critical capability chain, which state holds the structural advantage – the chokepoint, the standard-setting authority, the irreplaceable know-how – and assigning it what might be called a domain anchor role.
For example, the Netherlands, a modest power by conventional metrics, manufactures through ASML (largest supplier for the semiconductor industry), dominating in making lithography systems and driving the AI boom – a chokepoint that gives it leverage far exceeding its GDP ranking. The UK, handling roughly 38 per cent of global foreign exchange trading, anchors international financial services regulation through the Financial Conduct Authority (FCA) and the Bank of England in ways no other middle power can replicate. Canada’s critical minerals reserves (lithium, cobalt, nickel, and rare earths) position it as the natural anchor for economic security in any network seeking to diversify away from Chinese supply chains. Japan’s leadership in advanced materials,robotics, and precision manufacturing makes it an anchor for technological innovation across multiple domains. Brazil’s agricultural technology and its stewardship of critical biomes give it a structural role in food security and climate resilience that no coalition can afford to overlook.
None of these states is a ‘superpower’. Each holds a position that the others cannot substitute – that mutual dependency is itself the enforcement mechanism: in a network built on complementary strengths, free-riding carries a real cost – exclusion from capabilities no member can replicate alone. Existing forums and bilateral partnerships (UK-Japan, Canada-European Union, the Minerals Security Partnership) provide the political wiring. What is missing is the circuit design: an organising logic that matches capabilities to gaps.
How should this network operate? The Munich Security Report (2026) warns that the world is shifting toward a “deals-based order” – transactional, episodic, governed by leverage rather than rules. This is the wrong operating logic – deals are bilateral by nature, asymmetric by design, and unstable by definition. A middle-power network built on deal-making will simply reproduce, at a smaller scale, the power dynamics it is trying to escape.
The alternative is supply-chain diplomacy, applying the analytical method of supply-chain mapping – identifying nodes, dependencies, chokepoints, single points of failure, and redundancies – to the architecture of international cooperation. The perspective draws on political economy and the geopolitics of technology, where supply-chain thinking already structures policy on semiconductors, critical minerals, and AI governance. The difference from commercial supply-chain management lies in the operating logic: the relationships it produces are political commitments, not commercial transactions, with different enforcement mechanisms, legitimacy requirements, and time horizons.
This is distinct, too, from what Farrell & Newman have termed “weaponised interdependence” – the use of network chokepoints to coerce. Supply-chain diplomacy uses their map of chokepoints for the opposite purpose: building complementary resilience rather than exercising coercive leverage. The European Union’s internal debate over digital sovereignty, overwhelmingly prioritises France and Germany, based on economic and historical weight, but a supply-chain map of European chip production reveals a different picture: the Netherlands, through ASML, occupies the most critical node. The framework corrects for political bias and surfaces where domain anchor power actually sits. Applied across a middle-power network, it would do the same: match capabilities to gaps, assign leadership to demonstrate strength rather than diplomatic seniority, and distribute resilience across the system.
It might be objected that middle-power coalitions have been tried before and always underperform. But these unsuccessful groupings were not assembled around functional complementarity. A network structured around domain anchors creates different incentives: each member holds something the others cannot replicate. The India-EU trade deal signed at Davos and Canada’s recent moves to diversify trade toward China suggest the political will now exists to take this path.
The European Union’s internal coordination would similarly benefit from the privileged function – who anchors what – over the deference to economic and historical weight that currently distorts its strategic priorities.
Governments in middle-power clusters should now commission cross-domain dependency maps to identify structural strengths and single points of failure. Existing forums should then restructure their work programmes to focus on domain-specific coordination and develop resilience metrics, exploring cross-dependencies in their supply chains.
The post-American order will not arrive by default. It must be engineered – node by node, chain by chain.
The views and opinions expressed in this post are those of the author(s) and not necessarily those of the Bennett Institute for Public Policy.
