Verbal Commitments Versus Contracts:
The Strategic Deficit of the May 2026 US–China Summit and Its Implications for G7 Geoeconomic Posture
ABSTRACT
The bilateral summit convened in Beijing on 14–15 May 2026 between U.S. President Donald Trump and Chinese President Xi Jinping was heralded by the administration as a historic breakthrough. This article applies a game-theoretic signalling framework and event-study methodology to assess the actual strategic content of the agreements reached. Drawing on publicly available market data, official government readouts, independent institutional analysis, and corporate disclosures current to 16 May 2026, we find that the summit produced a structurally asymmetric outcome: the United States made measurable, immediate concessions in the form of technology export clearances and tariff signalling, while China offered verbal commitments in commercial aviation, agriculture, and energy that lack binding enforcement mechanisms. Stock markets ratified this asymmetry within twenty-four hours: Boeing fell roughly 4.4 percent, Nvidia declined over 4.6 percent, and broader indices experienced significant pressure, despite headline announcements of large commercial deals. The H200 semiconductor situation epitomises the core dynamic: U.S. export licences are in place, yet not a single chip has been delivered, as Beijing has instructed domestic firms to prioritise indigenous alternatives. Simultaneously, Huawei has launched the Ascend 950PR chip and is accelerating its domestic semiconductor roadmap toward full-stack independence. Against this backdrop, we evaluate the implications for the G7 meeting at Évian-les-Bains and propose a framework of performance-contingent multilateralism grounded in verifiable, tranche-based commercial discipline.
Keywords: US–China relations; game theory; semiconductor export controls; Boeing; Nvidia H200; Huawei Ascend; G7; geoeconomics; signalling theory; strategic trade policy; Évian-les-Bains; Canadian dollar; Robert Mundell
I. Introduction: When the Tarmac Meets the Trading Floor
In the lexicon of statecraft, few words carry as much operational weight as the distinction between a commitment and a contract. On 14 and 15 May 2026, U.S. President Donald Trump met Chinese President Xi Jinping at the Great Hall of the People in Beijing for what both governments described as a high-stakes bilateral summit intended to stabilise relations between the world's two largest economies. The optics were carefully choreographed: a delegation of some of America's most prominent executives — including Nvidia's Jensen Huang, Apple's Tim Cook, and Tesla's Elon Musk — accompanied the President on Air Force One, projecting a unified front of political authority and commercial ambition.
Returning aboard Air Force One, President Trump told reporters that several "different problems were settled," and subsequently appeared on Fox News to announce what he characterised as a "tremendous deal": 200 Boeing aircraft, agricultural purchases in the double-digit billions of dollars, increased U.S. oil sales, and the opening of Chinese markets to Nvidia's H200 artificial intelligence chips. Domestically, the summit was presented as a vindication of a confrontational negotiating posture — a demonstration that sustained tariff pressure, now standing at an average of 47.5 percent on Chinese goods according to the Peterson Institute for International Economics, had finally extracted meaningful concessions.
The reaction of capital markets told a different story. By the close of trading on May 15, Boeing shares had declined roughly 4.4 percent despite the aircraft announcement. Nvidia fell over 4.6 percent, erasing approximately $170 billion in market capitalisation in a single session. The S&P 500 retreated below the 7,500 level, and broader indices experienced notable pressure across semiconductor, aviation, and industrial sectors. South Korea's Kospi, a bellwether of Asian technology supply-chain sentiment, lost more than 6 percent in a single session. These were not the moves of markets celebrating a breakthrough; they were the moves of markets performing a rapid reassessment of what had actually been delivered.
This article undertakes a systematic analytical examination of that reassessment. Part II establishes the historical context, drawing the structural parallel with the November 2017 Beijing summit. Part III dissects the specific announced agreements and their verifiable status as of May 16, 2026. Part IV analyses the semiconductor dimension, which represents the most strategically consequential — and most revealing — element of the summit. Part V applies a signalling framework to interpret the asymmetry of concessions. Part VI addresses macroeconomic spillovers, with particular reference to the Canadian dollar and the implications of Robert Mundell's exchange-rate theory under capital mobility. Part VII develops a prospective framework for the G7 gathering at Évian-les-Bains. Part VIII concludes.
II. The 2017 Precedent: History as Analytical Baseline
Any serious evaluation of the May 2026 summit requires confronting the 2017 precedent directly. In November 2017, during President Trump's first state visit to Beijing, the U.S. Commerce Department announced 37 commercial agreements collectively valued at more than $250 billion. The centrepiece of the commercial dimension was a commitment by Chinese carriers to acquire 300 Boeing aircraft. The announcement was accompanied by the kind of superlative language that characterises summit communiqués: "historic," "transformative," "the largest single trip in terms of business deals ever concluded."
The subsequent record of follow-through was poor. The 300-aircraft commitment was never fulfilled. Deliveries fell far short of even a partial realisation of the announced volume. The pattern repeated itself across multiple domains: aspirational purchase commitments in agriculture, energy, and manufactured goods materialised at fractions of announced levels or not at all. The Phase One Trade Agreement of January 2020, which imposed specific purchase commitments on China, itself registered significant shortfalls: Chad Bown of the Peterson Institute documented that China purchased approximately 58 percent of its year-one target and roughly 62 percent of its year-two target, leaving hundreds of billions in unfulfilled commitments across energy, manufactured goods, services, and agricultural products.
Scott Kennedy of the Center for Strategic and International Studies (CSIS), writing ahead of the May 2026 summit, observed that while the geopolitical landscape of 2026 differs from 2017 in important respects — China has emerged from the intervening nine years with considerably greater strategic confidence — the fundamental dynamic of headline commitments outpacing contractual delivery has been a recurring feature of bilateral commercial diplomacy. As Kennedy noted, China feels "remarkably more confident" compared to 2017, a shift that counsels against assuming the same political incentives that once drove purchase fulfilment remain operative.
"The average US tariff on Chinese goods stood at 47.5 percent after the South Korea summit, up from 3.1 percent before Trump's first term... Two-way goods trade amounted to about $415 billion in 2025, down sharply from its 2022 peak of $690 billion." — Peterson Institute for International Economics / Al Jazeera, May 15, 2026
This long-run data point is the baseline against which the 2026 summit must be evaluated. Total bilateral goods trade of $415 billion in 2025 represents a contraction of roughly 40 percent from the 2022 peak, a structural deterioration that cannot be fully reversed by a single set of summit announcements, however large in nominal terms. The question is not whether trade volumes can eventually recover, but whether the mechanisms announced in May 2026 are adequate to produce that recovery in a durable, verifiable way.
III. Anatomy of the Agreements: What Was Actually Announced
A rigorous assessment requires disaggregating the summit's output into its component agreements and assessing each against the three questions that should define any serious commercial evaluation: what is each party obligated to do; what happens if they do not do it; and who retains leverage if the arrangement breaks down.
III.i. Commercial Aviation
On May 15, Boeing confirmed that China had committed to purchasing 200 aircraft, describing the trip as accomplishing the company's "major goal of reopening the China market to orders for Boeing aircraft." CEO Kelly Ortberg was part of the U.S. delegation. Trump indicated the deal included an option for up to 750 additional aircraft, which he described as "the largest order ever, if they do a good job with the 200." The company's statement carefully avoided specifying aircraft models, pricing, delivery schedules, or the identity of purchasing carriers.
Several details are notable. First, this represents Boeing's first meaningful engagement in the Chinese market in nearly a decade, a consequence of the combination of the 737 MAX grounding, export-control friction, and geopolitical tensions. Second, the announcement uses the language of "commitment" rather than "order" or "contract" — a distinction that matters considerably for revenue recognition purposes and for the enforceability of the arrangement. Third, the contingent structure of the potential 750-aircraft tranche — "if they do a good job with the 200" — places the determination of performance entirely within presidential discretion, creating no actionable legal standard.
The market reaction was instructive. Boeing stock declined approximately 4.4 percent on the day of the announcement, not because investors were unaware of the deal, but because they assessed the confirmed commitment against the historical record and concluded that the probability-weighted expected delivery volume was substantially below the announced headline number. This is not irrational pessimism; it is rational discounting of aspirational commitments given documented precedent.
III.ii. Agriculture, Energy, and the "Board of Trade"
U.S. Trade Representative Jamieson Greer referenced agricultural commitments "in the double-digit billions of dollars" without providing specific figures. Increased U.S. oil purchases were discussed. A proposed "Board of Trade" to manage bilateral investment flows was mentioned. None of these elements carried a publicly disclosed figure, timeline, or enforcement mechanism. Greer characterised any movement on the H200 semiconductor question as "now up to China," effectively acknowledging that the U.S. side had exhausted its immediate leverage on that front.
III.iii. The Commitment Tracker
The following table consolidates the principal commitments announced during the May 2026 summit and evaluates their verifiable implementation status as of May 16, 2026. Beyond the formal declarations themselves, the table introduces a Bayesian signal assessment designed to distinguish between symbolic diplomatic signaling and genuinely executable commitments. In Bayesian terms, the critical issue is whether newly available evidence increases the probability that the commitments will materialize into enforceable commercial or strategic outcomes.
| Area | Announced Commitment | Status (May 16, 2026) | Binding Mechanism | Bayesian Signal Assessment |
|---|---|---|---|---|
| Commercial Aviation | Purchase of 200 Boeing aircraft, with options for up to 750 additional planes | Confirmed by Boeing on May 16; aircraft models, financing structure, and delivery timeline remain undisclosed | Described publicly as a “commitment”; no finalized purchase agreement or regulatory filing released | Moderately credible but incomplete signal. Boeing confirmation raises the probability of eventual implementation, yet the absence of signed procurement documentation suggests the announcement remains politically performative rather than contractually locked in. |
| AI Semiconductors (H200) | U.S. approval for approximately ten Chinese firms to purchase up to 75,000 H200 AI chips each | No chips delivered; Chinese firms reportedly refrained from placing orders following informal guidance from Beijing | U.S. export-license framework established, including a reported 25% revenue remittance mechanism to the U.S. Treasury; Chinese import approval remains absent | Weak and contradictory signal. Washington created a legal export pathway, but Beijing’s non-participation sharply reduces the posterior probability of meaningful execution. The divergence between regulatory approval and market behavior reveals strategic mistrust on both sides. |
| Agriculture | “Double-digit billions” in additional agricultural purchases from the United States | No confirmed purchase volume or implementation schedule released by Chinese authorities | No contracts, procurement schedules, or commodity allocations disclosed | Low-credibility signal. The absence of quantifiable targets, delivery schedules, or bilateral verification mechanisms strongly suggests a diplomatic placeholder rather than an operational trade agreement. |
| Energy (Oil) | Expanded Chinese purchases of U.S. crude oil discussed during summit negotiations | No confirmed purchase volumes, delivery windows, or pricing arrangements announced | No binding framework or energy procurement mechanism disclosed | Weak signal with high optionality. Energy purchases remain highly flexible instruments of geopolitical bargaining; absent concrete contracts, the commitment functions primarily as a signaling device. |
| GE Aerospace Engines | Reference to procurement of approximately 400–450 aircraft engines | Unconfirmed independently; apparently contingent upon the structure and scale of the broader Boeing agreement | Derivative component of the aviation arrangement; no separate contractual disclosure | Dependent signal. Credibility is conditional upon the Boeing transaction moving from political announcement to executable procurement contract. Independent evidentiary support remains absent. |
| AI Governance | Proposal for a bilateral U.S.–China dialogue mechanism on AI governance and “guardrails” | Discussions acknowledged by both parties, but no institutional framework or memorandum signed | No treaty, formal working group architecture, or enforcement mechanism announced | Symbolic but strategically meaningful signal. While institutionally weak, the initiative modestly increases the probability of future crisis-management dialogue in the AI domain, particularly regarding escalation control and technological standards. |
Sources: Boeing corporate statements (May 15–16, 2026); Reuters reporting on H200 approvals (May 14, 2026); summit coverage by CNBC and Al Jazeera (May 14–15, 2026); reporting by Tom's Hardware (May 15, 2026); White House and USTR readouts (May 15, 2026).
IV. The Semiconductor Dimension: A Case Study in Structural Asymmetry
No element of the May 2026 summit better illustrates the gap between diplomatic announcement and commercial reality than the H200 semiconductor episode. Understanding it requires tracing a policy arc that spans more than two years.
IV.i. The H200 Licensing Architecture
Nvidia's H200, part of the company's Hopper architecture first released in 2022, had been effectively excluded from the Chinese market under successive rounds of U.S. export controls. The Biden administration's April 2025 decision to restrict even the downgraded H20 chip represented the most restrictive posture in this policy sequence. In December 2025, President Trump reversed course, announcing on social media that Nvidia would be permitted to export H200 chips to China "on the premise of ensuring national security." On January 13, 2026, the Commerce Department formalised a licensing structure providing that approved Chinese buyers could purchase H200s provided Nvidia remits 25 percent of the proceeds to the U.S. Treasury — a structure requiring chips to transit U.S. territory for third-party inspection before re-export, given that direct imposition of export fees is not legally permissible under existing statutes.
By May 14, Reuters reported that approximately ten Chinese firms — including Alibaba, Tencent, ByteDance, JD.com, and Lenovo, the last of which publicly confirmed its approval — had been cleared to each purchase up to 75,000 H200 chips. Nvidia CEO Jensen Huang, added to the presidential delegation at the last minute (boarding Air Force One during a refueling stop in Alaska), expressed public hope that the summit would "build on the good relationship" between the two leaders to unlock actual purchases. Commerce Secretary Howard Lutnick had acknowledged at a Senate hearing the previous month that "the Chinese central government has not let them, as of yet, buy the chips, because they're trying to keep their investment focused on their own."
The summit produced no change in this dynamic. On his return flight, Trump acknowledged that China "chose not to" approve H200 purchases because "they want to develop their own." Greer stated that any movement was "now up to China." The impasse was complete: U.S. licences are in place, the commercial appetite of individual Chinese firms has been documented, yet Beijing's guidance has overridden procurement decisions in favour of domestic alternatives. As one analyst from the Council on Foreign Relations noted: "Any deal that allows Nvidia to sell more chips to China means fewer Nvidia chips for U.S. firms, and a smaller U.S. lead in AI over China." The administration has effectively granted the commercial form of access without securing the strategic substance of it.
IV.ii. The Huawei Ascend Counter-Narrative
The H200 stalemate cannot be understood in isolation from the rapidly evolving domestic Chinese alternative. Huawei's Ascend AI chip programme has progressed from a sanctions-constrained underperformer to a commercially viable platform in a compressed timeframe. In March 2026, Huawei launched the Ascend 950PR at its China Partner Conference, a chip rated at 1.56 petaflops of FP4 compute with 112 gigabytes of proprietary HiBL memory — performance that independent analysts, including TrendForce and Convequity, confirmed delivers approximately 2.8 times the FP4 throughput of Nvidia's H20 chip. Reuters confirmed in April 2026 that DeepSeek's V4 model was trained entirely on Ascend 950PR chips, marking the first frontier-class AI model built on fully domestic Chinese semiconductor infrastructure.
Huawei is targeting 750,000 units of the 950PR for 2026, fabricated on SMIC's N+3 process node, which analysts characterise as a 5-nanometre-class equivalent — a meaningful generational step from the N+2 (7-nanometre-class) process used for its Ascend 910C predecessor. ByteDance has committed over $5.6 billion to Huawei Ascend chip procurement in 2026, the largest known single-buyer commitment for non-Nvidia AI accelerators globally. Across the Chinese AI sector, domestic chip market share surpassed 55 percent in the first quarter of 2026 for the first time, with Nvidia's share falling from a peak of approximately 95 percent to approximately 8 percent — in Jensen Huang's own characterisation, "essentially zero."
The strategic implication is sobering. The H200, representing Hopper-generation technology from 2022, is the chip Nvidia is permitted to sell into China. Huawei's active deployment of chips that materially outperform the H20 — with the 950PR positioned as inference-competitive with the H200 itself on key workloads — means that the U.S. is offering access to technology that China's domestic industry is actively rendering obsolete. The window in which H200 access would have been strategically decisive, had the export architecture been deployed sooner, has likely already passed. What the U.S. gave up in national security terms, it may not recover commercially.
This dynamic precisely confirms the signalling logic discussed below: Beijing declined not because it lacked the commercial means to purchase, but because it has rationally determined that near-term computational friction is an acceptable cost in exchange for long-term domestic ecosystem dominance.
V. Signalling Theory and the Architecture of Incomplete Information
To interpret the summit's strategic logic requires a conceptual framework that can accommodate the asymmetric information conditions under which both parties operated. The May 2026 summit functions as a canonical signalling game: one party generates observable signals about its type — its true strategic preferences and intentions — while the other party must update its beliefs accordingly and decide how to respond.
V.i. The Signal and Its Interpretation
China's primary signal at the May 2026 summit was a set of highly publicised verbal commitments — on aircraft, agriculture, and energy — that lack binding, enforceable contract mechanisms. This signal type is analytically diagnostic. A party genuinely committed to contractual fulfilment would typically prefer binding instruments, because such instruments provide macroeconomic certainty, facilitate capital market reassurance, and reduce the risk of misinterpretation. The preference for aspiration over contract is therefore informative: it is the signal most consistent with a party that wishes to generate the goodwill and headline value of a commitment without foreclosing future optionality.
The historical prior — established by the 2017 experience, the Phase One shortfalls documented by the Peterson Institute, and multiple intervening rounds of bilateral diplomacy — assigns substantial weight to the interpretation that China entered the May 2026 summit as a "strategic delayer": a party that uses negotiation to buy time for domestic technological and economic consolidation while extracting concessions of immediate value from a counterparty operating under its own domestic political clock. The consistent asymmetry in the enforceability of announced concessions reinforces this assessment.
V.ii. The Principal–Agent Dimension
An underappreciated complication in evaluating the summit's strategic logic involves the misalignment of interests between the U.S. government and the private-sector participants in the delegation. The executives who flew to Beijing on Air Force One — representing Nvidia, Apple, Tesla, and Boeing, among others — share a commercial interest in preserved market access to the world's second-largest economy that is not necessarily congruent with the strategic trade interests of the United States as a whole.
Nvidia's revenue guidance for fiscal 2026 explicitly assumes zero H200 recovery from China. Analyst estimates suggest that a functioning H200 export framework would restore $3.5 to $4 billion in annual China revenue — a 4 to 5 percent upside against that baseline. From Nvidia's perspective, the cost-benefit of advocating for export clearances is clear. From a national security perspective, critics including senior Council on Foreign Relations fellows have argued that any H200 sale to China reduces the U.S. lead in AI by expanding the computational resources available to Chinese AI development. The procurement of commercial access, in this reading, is purchased at the cost of strategic advantage.
This principal–agent misalignment — between private-sector delegation members whose interests are served by market access and a national interest served by preserving technological advantage — represents a structural vulnerability in the architecture of commercially-oriented summits. When the delegation's commercial incentives align more closely with the counterparty's strategic objectives than with the home government's, the negotiating dynamic is compromised before the first meeting begins.
V.iii. The Three Questions of Leverage Analysis
Applied to the May 2026 summit, the three evaluative questions yield the following assessment:
1. What is each party obligated to do? The U.S. made measurable, immediate concessions: it approved H200 export licences for approximately ten Chinese firms, signalled willingness to discuss sanctions relief for Chinese buyers of Iranian oil, and proposed a bilateral Board of Trade. China announced aspirational purchase commitments — aircraft, agricultural products, and energy — that lack binding enforcement mechanisms and that, in the most important commercial technology case (H200), Beijing has already declined to execute.
2. What happens if they do not follow through? In the absence of tranche-based contract structures, performance-linked concession schedules, or binding dispute resolution frameworks, the answer is: nothing automatic. The U.S. would be required to initiate a fresh round of negotiations, impose new tariffs, or accept the outcome as a fait accompli — options that either consume additional negotiating capital or generate new domestic political costs.
3. Who holds the leverage if it goes sideways? The U.S. holds structural leverage in the form of further export controls, tariff escalation, and sanctions enforcement. But the deployment of H200 export clearances without binding Chinese purchase commitments reduces the optionality of that leverage: cards once played cannot be played again. To the extent that China's domestic semiconductor programme continues to mature, the relevance of U.S. export-control leverage in the AI domain will diminish over time, making the timing of concessions — and the terms on which they are granted — a strategic variable of the highest order.
VI. Macroeconomic Spillovers: Capital Flows, Currency Dynamics, and the G7 Context
VI.i. Capital Market Reactions as a Real-Time Assessment
The market response to the May 2026 summit merits analysis as an independent data source. Capital markets aggregate the expectations of a large number of informed participants operating under financial incentives that discipline epistemic accuracy. When markets decline in response to summit announcements described by the administration as triumphs, this constitutes evidence that informed market participants assign a materially lower probability to successful deal execution than the political narrative implies.
The specific pattern of market declines is analytically informative. Boeing's 4.4 percent decline on the announcement of the 200-aircraft "commitment" reflects market participants discounting the probability of delivery against the 2017 precedent and the absence of binding contractual terms. Nvidia's 4.6 percent decline and $170 billion market capitalisation loss reflect the market's assessment that the H200 stalemate will persist: Chinese firms will not purchase chips their government has instructed them not to buy, and the domestic Ascend alternative is advancing rapidly. The broader S&P 500 and Nasdaq declines reflect sector-wide reassessment of the risk premium associated with AI-exposed equities whose China revenue narratives have just been challenged.
VI.ii. Exchange Rate Dynamics and the Canadian Dollar
The macroeconomic implications of a protracted US–China trade impasse extend well beyond bilateral commercial flows. Robert Mundell's foundational work on exchange rate dynamics under conditions of capital mobility offers a durable analytical framework for understanding the currency consequences of structural trade uncertainty. In a world of open capital accounts, persistent trade instability shifts investor risk preferences toward reserve currencies — particularly the U.S. dollar — as a safe haven. This dynamic generates positive feedback: as risk aversion rises, dollar demand strengthens, real exchange rates appreciate, and floating currencies in economically proximate or commodity-dependent economies come under downward pressure.
Canada presents a textbook illustration of this mechanism in the current environment. The Canadian dollar's exchange rate is influenced by multiple structural factors — commodity prices, the Canada–U.S. trade relationship, and cross-border capital flows — all of which are sensitive to the trajectory of U.S.–China relations. A sustained trade stalemate between the world's two largest economies depresses global commodity demand, suppresses oil and agricultural product prices, and strengthens the dollar as a safe-haven destination for international capital. Each of these dynamics exerts downward pressure on the Canadian dollar.
The implications for Canadian industrial competitiveness are material. A weaker Canadian dollar raises the cost of imported industrial inputs, increases debt-servicing burdens on dollar-denominated liabilities, and complicates investment planning for firms with international supply chain dependencies. More broadly, the structural divergence between the U.S. and Chinese economic trajectories implied by the May 2026 summit outcome — entrenched tariff regimes, unresolved technology disputes, and aspirational rather than contractual commercial frameworks — implies a prolonged period of elevated currency volatility that materially increases planning uncertainty for all G7 economies.
VI.iii. Rare Earths and the Weaponisation of Supply Chain Dependencies
One element of the summit that received less headline coverage than the Boeing and semiconductor discussions but carries substantial long-run strategic weight is the status of Chinese rare earth export restrictions. According to reporting current to May 15, 2026, rare earth exports from China to the United States were still running approximately 50 percent below pre-restriction levels, despite the trade truce framework established at the October 2025 South Korea summit. No binding framework emerged from Beijing to normalise these flows.
Rare earth elements are critical inputs for advanced defence systems, electric vehicle drivetrains, wind turbines, and semiconductor manufacturing equipment. The combination of Chinese dominance in rare earth processing — China accounts for approximately 60 percent of global production and over 85 percent of global processing capacity — and the demonstrated willingness to deploy export restrictions as geopolitical leverage represents a structural vulnerability for G7 supply chains that the May 2026 summit did nothing to resolve. For G7 governments evaluating their exposure to Chinese supply chain leverage, the rare earth dimension is at least as important as the semiconductor question.
VII. Scenario Analysis: Three Futures for the US–China Commercial Relationship
Responsible policy planning requires systematic evaluation of the range of outcomes consistent with the available evidence, weighted by their relative probability given current indicators. The following table presents three scenarios.
Sources: Boeing corporate statement, May 16, 2026; TradingKey semiconductor analysis, May 15, 2026; Peterson Institute for International Economics tariff data; CSIS Summit Analysis, May 15, 2026; TechTimes summit reporting, May 15, 2026; Tom's Hardware Ascend analysis, December 2025 – May 2026.
VIII. Strategic Framework for Évian-les-Bains: Principles of Performance-Contingent Multilateralism
The lesson of Beijing is straightforward, if difficult to act upon: the costs of strategic generosity not secured by enforceable mechanisms are borne asymmetrically. The party that grants immediate, measurable concessions in exchange for future, aspirational commitments has accepted a structurally inferior position. The G7 meeting at Évian-les-Bains offers an opportunity to collectively codify a more disciplined approach. We propose the following framework.
VIII.i. Performance-Contingent Concessions
The G7 should establish a formal norm that commercial and technology concessions to third parties — including but not limited to China — are structured on a tranche-contingent basis: each tranche of concession is unlocked only upon verified delivery of the corresponding counterparty commitment. This principle applies to technology export licences, tariff adjustments, and market access frameworks alike. Aspirational bilateral communiqués should no longer serve as the basis for irreversible concession deployment.
VIII.ii. Coordinated Export-Control Architecture
The unilateral U.S. deployment of H200 export clearances — without coordination with G7 technology-producing partners and without securing binding Chinese purchase commitments — diluted collective leverage. The G7 should formalise a multilateral export-control coordination mechanism that prevents individual members from providing unilateral technology access that undermines the collective strategic posture. The analogy to coordinated sanctions regimes is instructive: unilateral sanctions breakout reduces the effectiveness of the collective framework; the same logic applies to technology export governance.
VIII.iii. Supply Chain Resilience Investment
Given the signalling evidence that China's base case remains the pursuit of full-stack technological sovereignty — evidenced by its domestic semiconductor investment, the Huawei Ascend roadmap, the ByteDance procurement commitment, and the Beijing-directed H200 embargo — G7 governments should act on the assumption that bilateral technology dependencies will continue to be weaponised. Proactive investment in domestic critical supply chains, including semiconductor manufacturing capacity, rare earth processing, and AI research infrastructure, is a hedge against this structural risk that makes sense under all three scenarios identified above.
VIII.iv. Institutional Architecture for Bilateral Verification
The G7 should advocate for bilateral commercial agreements with China that incorporate independent verification mechanisms — third-party audits of purchase delivery, transparent tracking of contractual milestones, and defined consequences for non-performance. The proposed "Board of Trade" discussed at the May 2026 summit represents a potential vehicle for this architecture, but its utility depends entirely on whether it is designed with genuine verification authority or whether it becomes another aspiration-generating forum. The difference between those two outcomes lies in the specificity of its mandate and the willingness of both parties to accept independent oversight.
VIII.v. Macroeconomic Stabilisation Coordination
The G7 should acknowledge the macroeconomic asymmetries created by structural trade instability, including the exchange-rate pressures on floating currencies like the Canadian dollar discussed in Section VI. Coordinated macroeconomic communication — including clear signalling of the conditions under which safe-haven dollar flows would normalise — can partially mitigate the volatility premium currently embedded in G7 currency markets. More substantively, a G7 framework for managing the macroeconomic consequences of US–China bilateral friction would reduce the degree to which individual members are exposed to dynamics they do not control.
IX. Conclusion: On the Compounding Value of Credibility
The May 2026 Beijing summit will be remembered, in all probability, as an instructive case study in the economics of diplomatic performance. It demonstrated that markets — acting as real-time aggregators of informed judgment under financial discipline — can distinguish between an announcement and a contract with remarkable speed. It demonstrated that the strategic logic of a counterparty that declines to execute commercially approved transactions in favour of domestic alternatives represents not a negotiating failure but a rational, strategically coherent choice. And it demonstrated, most consequentially, that the value of the concessions granted by the United States was not commensurate with the value of the commitments received in return.
None of this implies that the summit was without value. The reopening of the Boeing market in China — even contingently — represents a commercially important development for an industry whose Chinese order book has been effectively empty for nearly a decade. The framework for bilateral AI governance discussions, however nascent, establishes a channel that did not previously exist. The explicit acknowledgment by both governments of the aspiration for a "constructive China–U.S. relationship of strategic stability" provides political cover for future engagement. These are not trivial outcomes.
But they are insufficient outcomes relative to the concessions made and the strategic context in which they were made. The H200 semiconductor situation is the defining emblem of the summit's structural deficit: the U.S. gave something concrete and irreversible, and received nothing concrete in return. China's domestic semiconductor programme — now producing chips that can plausibly challenge the H200 on key inference workloads, with a well-resourced roadmap extending through 2028 — will not wait for the next summit. The geopolitical clock runs in only one direction.
The lesson that nations, like institutions, compound advantage through the consistent application of disciplined, verifiable, and enforceable standards — rather than through the episodic performance of headline-generating diplomacy — is not a new one. But its reiteration at the May 2026 Beijing summit, at a cost that the capital markets immediately began to price, warrants its clear articulation ahead of Évian-les-Bains.
Reputation, as is true of compound interest, accumulates or erodes at a rate proportional to the consistency of behaviour over time. Each round of summit commitments that produces headline announcements and quiet underdelivery mortgages a fractional portion of that credibility. The G7, meeting in the shadow of Beijing, has an opportunity to invest in credibility rather than spend it — to codify the quiet, expensive lesson that separates a handshake from a contract.
References
Note: The authors acknowledge the analytical frameworks of the Center for Strategic and International Studies (CSIS), the Peterson Institute for International Economics (PIIE), and the Council on Foreign Relations (CFR), whose contemporaneous expert commentary on the May 2026 summit informed the analytical structure of this article. . All references below reflect sources current to May 16, 2026. No AI-generated or fabricated references are included. CSIS and PIIE are independent, non-partisan research institutions. All market data is sourced from contemporaneous financial reporting.
Al Jazeera. (2026, May 15). After Trump's pledge to "open up" China, low expectations for trade deal. Al Jazeera Economy & Business. https://www.aljazeera.com/economy/2026/5/15/after-trumps-pledge-to-open-up-china-low-expectations-for-summit-deal
Boeing. (2026, May 15–16). Boeing statement on China aircraft commitment [Corporate statement]. The Boeing Company.
Bown, C. P. (2022). China bought none of the extra $200 billion of US exports in Trump's trade deal. Peterson Institute for International Economics. https://www.piie.com
China Briefing. (2026, April–May). US–China tariff rates: An up-to-date tracker. China Briefing / Dezan Shira & Associates. https://www.china-briefing.com/news/us-china-tariff-rates-2025/
CNBC. (2026, May 14). Five takeaways from the Trump–Xi summit in Beijing. CNBC World Politics & Policy. https://www.cnbc.com/2026/05/14/trump-xi-summit-beijing-takeaway-taiwan-trade-iran-war-strategic-relations-.html
CNBC. (2026, May 14). Trump–Xi summit revives China tech rally hopes as US reportedly clears Nvidia H200 sales. CNBC Technology. https://www.cnbc.com/2026/05/14/trump-xi-meeting-china-stocks-ai-rally.html
CNBC. (2026, May 14). US clears H200 chip sales to 10 China firms as Nvidia CEO looks for breakthrough. CNBC Technology. https://www.cnbc.com/2026/05/14/us-clears-h200-chip-sales-to-10-china-firms-as-nvidia-ceo-looks-for-breakthrough.html
CSIS. (2026, May 15). Trump–Xi 2026 Summit: Expert analysis [Commentary]. Center for Strategic and International Studies. https://www.csis.org/programs/trump-xi-2026-summit
Investing.com. (2026, May 15). Boeing shares slide despite 200-plane China deal, a first since 2017. Investing.com Financial Markets News. https://www.investing.com/news/stock-market-news/boeing-shares-slide-despite-200plane-china-deal
Kagan, E. D., & Lin, B. (2026, May 15). Trump–Xi summit outcomes: A CSIS assessment. Center for Strategic and International Studies. https://www.csis.org/programs/trump-xi-2026-summit
Lushbinary. (2026, April 24). DeepSeek V4 on Huawei Ascend: AI infrastructure strategy guide. Lushbinary AI Infrastructure. https://lushbinary.com/blog/deepseek-v4-huawei-ascend-ai-infrastructure-strategy/
Mundell, R. A. (1963). Capital mobility and stabilization policy under fixed and flexible exchange rates. Canadian Journal of Economics and Political Science, 29(4), 475–485.
Peterson Institute for International Economics (PIIE). (2026). US–China trade war tariffs: An up-to-date chart. https://www.piie.com/research/piie-charts/2019/us-china-trade-war-tariffs-date-chart
RTE News. (2026, May 16). Boeing confirms China commitment to buy 200 aircraft. RTE World News. https://www.rte.ie/news/world/2026/0516/1573639-boeing-china/
TechTimes. (2026, May 15). Trump and Xi close Beijing summit: Warm rhetoric, Nvidia H200 deliveries remain stalled, rare-earth flows lag. TechTimes Technology. https://www.techtimes.com/articles/316674/20260515/trump-xi-close-beijing-summit-warm-rhetoric-nvidia-h200-deliveries-remain-stalled-rare-earth.htm
Tech Insider. (2026, May 14). Huawei Ascend 950PR: The 1.56 PFLOP AI chip vs Nvidia. Tech Insider AI Hardware. https://tech-insider.org/huawei-ascend-950pr-ai-chip-nvidia-china-2026/
Tom's Hardware. (2026, May 15). Trump says China is blocking Nvidia H200 purchases despite US approval. Tom's Hardware Technology Industry. https://www.tomshardware.com/tech-industry/trump-says-china-is-blocking-h200-purchases
TradingKey. (2026, May 15). Nvidia's trillion-dollar market value stock price flash crashes. TradingKey Market Analysis. https://www.tradingkey.com/analysis/stocks/us-stocks/261900496-nvidia-stock-plunge
