Tea, Taiwan, and Tariffs: Who Really Won the Trump-Xi White House Summit?

Executive Summary

Chinese President Xi Jinping arrived at Joint Base Andrews on September 23, 2026, for his first state visit to Washington in eleven years. He left two days later with a two-month extension of the Busan trade truce, two giant pandas earmarked for the Atlanta Zoo, an invitation to 100,000 American students to visit China, and a continued silence from Washington on the pending $14 billion Taiwan arms package. Trump said American farmers would be "very happy." Xi called the visit a "milestone." The White House described it as "very productive" without specifying how.

Framed correctly, the September 2026 Trump-Xi summit was not a summit that produced outcomes - it was a summit that deferred them. On trade, the truce was extended rather than resolved. On Taiwan, the $14 billion package remained in limbo. On artificial intelligence, neither leader showed appetite for meaningful safety guardrails despite months of diplomatic preparation. What the three-day visit accomplished was something more specific and arguably more durable: it demonstrated that Xi and Trump have built the kind of personal relationship that allows both to claim public success while leaving every hard question for a later meeting. The next two come quickly - APEC in Shenzhen in November, and the G20 in Miami in December.

The strategic implications of this outcome extend well beyond Washington and Beijing. For Taiwan, a $14 billion arms package still in American hands represents prolonged strategic uncertainty. For India, a US-China accommodation - even a managed, fragile one - reshapes the calculus of the Quad, the China-plus-one trade opportunity, and New Delhi's relationship with Washington simultaneously. For the global economy, a two-month tariff truce resolved nothing about rare earth export controls, semiconductor restrictions, or the structural imbalances that produced the trade war in the first place.

The Road to Washington: Three Summits in Eleven Months

The September 2026 summit was the third Trump-Xi meeting in roughly eleven months. The sequence matters for understanding what Washington and Beijing were each trying to accomplish.

The first meeting, at Busan, South Korea on October 30, 2025, on the sidelines of APEC, was the reset. Trump and Xi had not met face to face since 2019. Trump described Xi as a "very tough negotiating opponent." Xi said the two sides had reached a "basic consensus" on trade. The Busan agreement reduced tariffs from triple-digit peaks - lowering US tariffs on Chinese goods toward an average of around 47 percent and pausing China's planned export controls on rare earth minerals - while committing Beijing to resume large-scale agricultural purchases. The deal was framed as a one-year truce, subject to annual renegotiation.

The second meeting, Trump's state visit to Beijing in May 2026, was the relationship-deepening exercise. Trump received a full ceremonial welcome at Zhongnanhai, toured the Temple of Heaven, and attended a state banquet at the Great Hall of the People. The substantive outcome was a White House announcement that China had pledged to purchase at least $17 billion per year in US agricultural products through 2028 - in addition to existing soybean commitments. Xi raised Taiwan as "the most important issue" in bilateral relations. Trump left with warm language about the relationship and no decision on a pending $14 billion arms package for Taipei that had been in the pipeline since the December $11.1 billion package was approved.

The third meeting - September 23-25 at the White House - was, in theory, where the momentum from Busan and Beijing would translate into something lasting. In practice, it produced the narrowest set of confirmed outcomes of the three summits, against the highest backdrop of ceremonial ambition.

What Actually Happened: A Summit by Category

Trade and the Tariff Truce

The most concrete outcome of the White House summit arrived before Xi's plane landed. On September 23, US Treasury Secretary Scott Bessent announced on Fox News that the US and China had agreed to extend the Busan truce by two months, to January 10, 2027. The original truce was due to expire in mid-November. Most analysts had expected a six-month extension. Two months is materially shorter - a signal from both sides that the longer-term trade architecture remains genuinely unresolved rather than simply deferred as a formality.

The truce extension kept in place the key provisions of Busan: reduced tariffs from their peak trade-war levels, continued suspension of China's rare earth export controls, and Beijing's commitments on agricultural purchases. US tariffs on Chinese goods were running at approximately 36.5 percent at the time of the summit, with Chinese tariffs on American exports near 31 percent - both well above pre-trade-war baselines on both sides.

China's performance on agricultural commitments, however, had been conspicuously weak. The May Beijing summit produced a White House pledge that China would buy at least $17 billion per year in US farm products. Agriculture Department trade data showed China had bought approximately $3.9 billion in non-soybean US farm products in the first seven months of 2026 - barely above levels from the peak tariff standoff of the prior year. Soybean purchases were more robust and consistent with the separate commitment from Busan, but the broader agricultural pledge was significantly underdelivered. Trump told reporters that farmers would be "very happy" with the summit outcomes without providing specifics on what additional commitments, if any, had been made.

What did not happen: no comprehensive trade deal, no announced tariff reductions on specific sectors, no formal resolution of the rare earth licensing problem - which the European Chamber of Commerce in China noted remains practically unworkable for companies that lack a standardized process for applying for rare earth export licenses even when the export controls are nominally suspended.

Taiwan: The $14 Billion Question

Taiwan was the gravitational center of the summit that never appeared directly in any joint statement or American public readout. Beijing's official Xinhua readout quoted Xi telling Trump that the United States should "adhere to the correct position of opposing Taiwan independence" and handle the Taiwan question "with prudence." The State Department confirmed that Secretary Rubio had affirmed that US policy on Taiwan had not changed. No halt to arms sales was announced.

Trump had approved a record $11.1 billion arms package for Taiwan in December 2025 - the largest ever at that point - and a second package worth approximately $14 billion had been in bureaucratic limbo since before the May Beijing summit. Trump has referred publicly to Taiwan arms sales as "a very good negotiating chip" with China. At Beijing in May, he declined to approve the $14 billion package before the summit; the same outcome repeated at Washington. A senior administration official told reporters that Trump would make a determination on the package "in a fairly short time" - a phrase that has now been used on at least two separate occasions without producing a decision.

Xi apparently raised Taiwan concerns directly during the White House Oval Office meeting, and Trump emerged saying "I'll be making decisions" while noting his priority was avoiding a war. Reporting from Axios and NBC indicates Beijing had warned Washington before the summit that approving the $14 billion package before the meeting could jeopardize Xi's attendance. Xi attended. The package remains unresolved. From Beijing's perspective, the tactic of using the summit as leverage to delay an arms approval that Taipei's parliament had already appropriated $25 billion to fund is working - at least temporarily. Brookings analyst Ryan Hass has assessed that Beijing's approach is specifically designed to drag out the review process to allow China's military modernization to outpace Taiwan's force acquisition timeline.

Taiwan's government issued a careful post-summit statement saying that continued US arms sales are "in Washington's national interest" and that Taipei remains in "continuous communication" with Washington. The political pressure on Taiwan's President Lai Ching-te from the prolonged delay is real but not yet structurally destabilizing.

Artificial Intelligence: Dialogue Without Guardrails

Artificial intelligence was perhaps the most substantive area of potential cooperation at the summit - and also the area of greatest rhetorical distance between what was said and what was accomplished. Both leaders acknowledged AI's importance. Xi told reporters that the two countries should develop and manage AI "for good" and keep it under human control. Trump, who had declared at the UN General Assembly just days earlier that the US was outcompeting China in AI and intended to keep it that way, compared an agreed AI dialogue to the Cold War "red phone between the Kremlin and the White House."

What was confirmed: both sides agreed to continue the bilateral AI dialogue that China's foreign ministry had confirmed had begun. Neither leader expressed interest in the kind of binding AI safety agreements that many experts and arms control scholars had urged, analogous to nuclear risk reduction mechanisms. Foreign Policy reported that "neither Trump nor Xi appeared interested in such safeguards," despite the UN Security Council's calls for benchmarks measuring AI capabilities. The practical outcome is a sustained dialogue structure without governance teeth - arguably better than nothing, and far short of what the moment technically requires given the pace of AI capability development on both sides.

Symbolism as Strategic Communication

The ceremonial elements of the summit were not incidental. They were a deliberate communication strategy from both sides, worth reading as carefully as any joint statement. Trump greeted Xi on the airport tarmac at Joint Base Andrews - an unusual gesture for a president who typically receives foreign leaders at the White House. There was a full military review, low flyovers by US aircraft, a state dinner, a White House tea ceremony with both couples, and a tour of the National Archives where Xi and Trump walked alongside the Declaration of Independence and Constitution. Trump gave Xi a large bald eagle sculpture. Xi offered Trump rose seeds. Two pandas were announced for Atlanta.

Xi's comment at the White House arrival - "The Thucydides Trap can be overcome" - was the single most strategically loaded public statement of the summit. The phrase is a reference to Harvard political scientist Graham Allison's argument that a rising power and an established power are historically prone to catastrophic conflict. Xi deploying it at the White House, in the presence of the American president, is a carefully calibrated message: China does not want conflict with the United States, believes conflict is not inevitable, and is prepared to invest in managing the relationship to prevent it. Whether that reflects China's actual strategic calculation or its preferred public framing is a question Xi did not answer.

Power Center Analysis: Who Needed This Summit More?

Both leaders came to Washington with genuine needs, but the asymmetry of those needs tells a more interesting story than the pageantry suggested.

Trump needed a domestic political deliverable. American farmers had been promised $17 billion per year in Chinese agricultural purchases and had not seen meaningful follow-through on non-soybean commitments. The trade war's supply-chain disruptions continued to create pressure on American manufacturers. The Iran war in early 2026 had consumed significant political bandwidth and Patriot missile stocks; Trump needed a stable Pacific front while managing Middle Eastern aftereffects. A successful summit - or one that appeared successful - served his political narrative of deal-making competence.

Xi needed stability at a moment of structural vulnerability. China's economic growth has been running below government targets. Rare earth export controls, when threatened, created significant domestic industrial complications - the minerals need to be exported to generate revenue, even as their restriction generates diplomatic leverage. China's access to American technology remains constrained by semiconductor export controls that the truce did not touch. Xi's strategic calculation appears to be that managed competition is preferable to uncontrolled confrontation at a moment when China has not yet achieved the industrial and technological self-sufficiency to comfortably absorb a full decoupling. The summit, on this reading, bought China time.

A third view, articulated by CSIS analysts Scott Kennedy and Ilaria Mazzocco and implicitly reflected in the CFR's pre-summit assessment, is that both leaders came to Washington primarily for the optics of the relationship rather than its substance. The summit confirmed what Busan and Beijing had already established: Trump and Xi have a functional personal relationship that both find useful, and maintaining that relationship has become a goal in itself rather than merely a means to specific policy outcomes. That is not nothing - but it is categorically different from a strategic agreement.

The Hidden Drivers: What Neither Side Said Publicly

Several factors shaped the summit's limited outcomes that neither government discussed openly.

The Iran war context was central. The US military strike campaign against Iran in early 2026 - conducted without Chinese endorsement and against China's stated preference for diplomatic resolution - created genuine bilateral tension that persisted into the September summit. Beijing wanted the Iran issue on the agenda. Washington preferred to keep it managed below the headline. The result was a diplomatic agreement to maintain the AI dialogue partly to prevent any future AI-enabled miscalculation from triggering conflict, though even this was framed in the tamest possible language.

Rare earth market dynamics gave China structural leverage that the summit did not resolve. Washington remains deeply dependent on Chinese rare earth processing capacity for the production of advanced semiconductors, defense systems, and electric vehicles. The truce suspended China's export control threats, but it did not build American processing capacity or supply chain alternatives. A US law giving the president authority to impose tariffs of up to 100 percent on major buyers of Russian oil and gas - including China and India - creates a potential pressure point but also limits Washington's freedom of maneuver with Beijing.

The upcoming APEC and G20 summits function as a structural incentive for both sides to leave issues unresolved at Washington. Every unresolved question - the $14 billion Taiwan package, a comprehensive trade architecture, AI governance - becomes a deliverable at a future summit. Both leaderships have learned, through three meetings in eleven months, that the summit format itself generates political value for both sides. The perpetuation of that format requires the perpetuation of unresolved questions.

India's Strategic Reading

New Delhi watched the Washington summit with more complex calculations than most outside observers recognized. The Trump-Xi accommodation creates both risks and opportunities for India, and the balance between them depends heavily on how durable the US-China managed competition model proves to be.

The risk side is significant. India-US relations had been under considerable strain heading into the summit, with Trump imposing among the highest tariffs on Indian goods of any trading partner, trade negotiations stalled over farm and dairy product market access, and the Quad summit that India was scheduled to host remaining uncertain. If China receives tariff relief and maintained market access while India faces trade penalties, New Delhi may draw the conclusion that Washington rewards leverage rather than partnership - a conclusion that has already informed Modi's strategic hedging, including his attendance at the SCO summit in Tianjin alongside Xi and Putin in September 2025 and at the World War II parade in Beijing.

The opportunity side is also real. If US companies continue diversifying supply chains away from China - even in the context of a managed trade truce - India's manufacturing base, which has seen electronics exports grow sharply in 2026, stands to benefit. India's merchandise exports reached $215.91 billion during April-August of the current fiscal year, up over 17 percent year on year, with electronics exports rising by nearly 90 percent in August 2026 alone. This suggests the China-plus-one dynamic has real momentum regardless of what happens at any specific Trump-Xi summit. The question is whether that momentum continues even if Washington's urgency about diversifying away from China diminishes under a more managed bilateral relationship.

A US-China detente does not resolve India's concerns about China. Beijing's relationship with Islamabad, competition across the Indian Ocean, and China's military modernization program continue regardless of how the tariff truce is structured. But a Washington that is managing its China relationship for stability is also a Washington less inclined to invest political capital in the Quad, in the Indo-Pacific strategy, or in the kind of technology transfer arrangements with India that the Quad framework has been developing. The competitive urgency that made India strategically valuable to Washington is partially reduced when Washington and Beijing are having tea together at the White House.

India's optimal posture - maintaining its relationship with both Washington and Beijing while leveraging strategic autonomy - becomes more complex in a world of managed US-China competition. The binary between a confrontational US-China relationship (which maximized India's strategic value) and a fully accommodated one (which would diminish it) is being replaced by something more ambiguous: a managed rivalry that neither resolves the strategic contest nor fully escalates it. India must calibrate its responses to that ambiguity without the luxury of waiting for clarity that may not arrive.

Implications for Taiwan and the Indo-Pacific

Taiwan emerged from the Washington summit in a position of managed but genuine uncertainty. The $14 billion arms package is not cancelled - it remains in the pipeline, appropriated by Taiwan's parliament and pending American approval. But every month of delay is a month in which China's People's Liberation Army continues its modernization program, its near-daily air and naval activities in Taiwan's air defense identification zone, and the gradual normalization of military pressure as a baseline condition rather than a crisis signal.

Taiwanese government officials, displaying careful diplomatic restraint, characterized the summit outcome as better than feared. The worst-case scenario for Taipei was a public American statement opposing Taiwan independence - a concession that would have fundamentally reframed the US commitment. That did not happen. Secretary Rubio affirmed US policy was unchanged. But what Taiwan needed was the $14 billion package approved, the arms delivery timeline accelerated, and a clear American signal that using Taiwan as a "negotiating chip" - Trump's own framing - was off the table. None of those things happened.

The Brookings-attributed assessment that Beijing's strategy is designed to allow China's military modernization to outpace Taiwan's force acquisition timeline is the most strategically consequential claim in the summit's aftermath. If accurate, the indefinite deferral of the arms package is not a neutral outcome - it is an incremental strategic shift in the cross-strait balance that accumulates across every successive summit where the package remains undecided.

Risk Matrix

Tariff re-escalation risk: The two-month truce extension, rather than six or twelve months, signals that both sides view the broader trade architecture as genuinely unresolved. If agricultural purchase commitments continue to underperform and domestic political pressure in Washington mounts ahead of November APEC talks, re-escalation to higher tariffs becomes a plausible scenario on a short timeline.

Taiwan arms decision risk: A decision by Trump to approve the $14 billion package before November APEC could trigger Chinese retaliation - including reimposition of rare earth export controls, a threat to boycott the Miami G20, or a significant increase in PLA activity around Taiwan. A continued decision to withhold the package strengthens Beijing's assessment that the summit-as-leverage model works.

AI incident risk: The agreed US-China AI dialogue exists at the level of bureaucratic process rather than binding commitment. Neither side has agreed on what constitutes an AI-related incident requiring notification, who is responsible for communicating it, or what the response timeline should be. A serious AI-related military or intelligence incident - a misidentification, an autonomous system failure - has no agreed management protocol beyond the general dialogue framework.

China agricultural non-compliance risk: If China continues to underperform on the $17 billion non-soybean agricultural commitment through the end of 2026, Trump faces domestic political pressure from farm-state constituencies that his China diplomacy has repeatedly used as a success metric. Sustained non-compliance without consequence undermines the credibility of future Chinese commitments as a negotiating currency.

India-US relationship deterioration risk: The combination of US-China managed accommodation and stalled India-US trade negotiations creates a strategic window for China to deepen its economic relationship with India at a moment when New Delhi is recalibrating. This is not an automatic outcome but a risk that is elevated precisely because the summit demonstrated Washington's willingness to accommodate Beijing's priorities under sufficient personal diplomatic pressure.

Scenario Analysis

Base Scenario - High Probability: The two-month truce is extended again at APEC in Shenzhen in November, with a longer horizon attached. Agricultural purchase commitments are partially clarified with additional detail on non-soybean categories. The $14 billion Taiwan arms package is approved between November and January, with Beijing registering a formal protest but not retaliating through rare earth controls. The AI dialogue continues at working-group level without producing binding governance frameworks. The G20 in Miami in December is a fourth Trump-Xi meeting, again ceremonially substantial and structurally limited. The pattern of managed competition continues with no definitive resolution of any major structural issue.

Key assumptions: Trump prioritizes domestic political stability over confrontational postures with China; Xi calculates that continued engagement preserves China's access to American markets and technology better than confrontation; neither a Taiwan military incident nor a major AI-related incident disrupts the diplomatic calendar.

Bull Scenario - Low-to-Moderate Probability: The APEC and G20 meetings produce a comprehensive trade framework covering tariffs on a wider range of goods, a standardized and functional rare earth export licensing mechanism, and formal AI risk notification protocols. Trump approves the $14 billion Taiwan package, which Beijing accepts without retaliating after a period of formal protest. India-US trade negotiations reach a partial agreement before the end of 2026. The managed competition model becomes genuinely institutionalized rather than episodically improvised, creating greater predictability for global markets and for US allies in the Indo-Pacific.

Key assumptions: Both governments face sufficient domestic pressure to produce substantive deals rather than symbolic extensions; Xi has sufficient political authority to make concessions on trade that reverse China's agricultural diversification strategy; Trump is willing to absorb Chinese diplomatic displeasure over Taiwan arms sales without treating it as a threat to the broader bilateral relationship.

Bear Scenario - Low-to-Moderate Probability: China's non-compliance on agricultural purchases accumulates to the point where Trump, facing a domestic political calculation before the December G20, reimplements punitive tariffs. China responds by reimposing rare earth export controls. The $14 billion Taiwan arms package is approved in this period of elevated tension, and China responds with a significant PLA military exercise around Taiwan that is larger and more provocative than the 2022 exercises that followed Nancy Pelosi's visit. The AI dialogue breaks down. The Miami G20 becomes a stage for public bilateral confrontation rather than continued accommodation. Global equity markets, supply chains, and semiconductor prices react severely. The managed competition model collapses, and both sides return to a more explicitly confrontational posture without the guardrails that the summit diplomacy of 2025-2026 had tentatively established.

Key assumptions: Agricultural non-compliance triggers Trump's domestic political red lines; China miscalculates the American response to a Taiwan military exercise; a third-party incident - an AI system failure, a maritime collision, an escalation in the South China Sea - occurs during a period of elevated bilateral tension.

Intelligence Forecast

Over the next 6 to 24 months, the following developments are assessed as most likely based on available evidence:

The APEC summit in Shenzhen in November 2026 will be the next major test of the managed competition model. Xi is the host. The home-field advantage and the ceremonial optics of an American president in China for a multilateral summit create strong incentives for a managed outcome. A six-month or twelve-month truce extension at APEC is assessed as the most probable tariff outcome, with some additional detail on agricultural purchasing commitments.

The $14 billion Taiwan arms package is assessed as likely to be approved before or at APEC, in a form that Beijing will protest diplomatically but not retaliate against militarily. The domestic political cost of continued delay - particularly given Taiwan's parliamentary appropriation and Taipei's increasingly public communications about the urgency of the package - is rising for the Trump administration faster than the diplomatic cost of approval.

The US-China AI dialogue will continue to meet at working-group level but is unlikely to produce binding governance frameworks within the 12-to-24-month forecast window. Both governments are in the middle of competitive AI development cycles and have no incentive to constrain capabilities during the period when their respective AI programs are advancing most rapidly. The comparison to nuclear arms control is historically resonant but structurally inapt: nuclear weapons had a forty-year development history before arms control became seriously negotiated; AI capability is advancing on a timeline measured in months.

India's strategic position will remain genuinely ambiguous. If India-US trade negotiations remain stalled while the US-China trade truce deepens, New Delhi will continue its multi-directional hedging - maintaining the appearance of Quad alignment while deepening economic ties with China and avoiding clear alignment with Washington on issues where US-China managed competition requires both powers to avoid embarrassing the other. The Quad summit remains uncertain.

China's rare earth export licensing problem is assessed as the most likely near-term flashpoint in the trade relationship. Companies operating in the rare earth supply chain have consistently identified the lack of a standardized licensing process as operationally crippling even when export controls are nominally suspended. If the functional blockage on rare earth supply is not resolved by the January 10, 2027 truce expiry, it will become a major issue at the Miami G20 regardless of what the truce's formal terms say.

Global Chanakya Assessment

The framing question that every major news outlet used to cover this summit - "what did they agree to?" - is the wrong question. The right question is: "what kind of relationship are they building, and at whose expense?"

The answer to the second question is more interesting than any of the summit's announced outcomes. Trump and Xi have now met three times in eleven months and constructed a personal relationship that functions as a buffer against structural rivalry. That buffer has real value - the world is materially better off with a US-China hot line than without one, and a US president who personally enjoys his interactions with the Chinese leader is less likely to allow bureaucratic or congressional pressures to escalate a trade dispute into something more dangerous. The buffer is not an illusion.

But the buffer has a cost, and that cost is being distributed unevenly among third parties who were not in the room. Taiwan's arms procurement is being used as diplomatic currency. India's strategic value to Washington is reduced every time Washington signals that the US-China relationship is stable enough to not require urgent counterbalancing. The Quad has quietly been deprioritized as US-China managed competition reduced the urgency of building multilateral frameworks for the Indo-Pacific. Countries in Southeast Asia, the Pacific, and the Global South that had calibrated their own foreign policy calculations based on a US-China confrontational dynamic are now navigating a more ambiguous environment where the two powers manage their rivalry without consulting anyone else about the terms.

The most important overlooked variable is China's agricultural non-compliance. The $17 billion non-soybean purchase commitment from May 2026 has barely been met. Every summit has produced a new Chinese purchase commitment. Every post-summit review has shown those commitments partially met at best. The pattern from Phase One of Trump's first term repeats almost exactly - a headline number, ceremonial signing, underperformance, extension, repeat. If Trump is willing to accept performative compliance as genuine compliance, China has found the cheapest possible form of diplomatic accommodation. If he is not, the tariff peace is more fragile than the summit's optics suggested.

The indicator that would most clearly signal whether the managed competition model is genuinely durable: the rare earth export licensing problem being resolved with a standardized, functional, company-level process before the January 10 truce expiry. If that happens, something structurally changed at this summit beyond what was publicly announced. If it does not, the summit will have been exactly what most analysts assessed it to be: spectacle in service of stability, purchased on credit that comes due in two months.

Indicators to Monitor

Decision on the $14 billion Taiwan arms package - timing and Chinese response
Rare earth export license approvals at company level - standardization and processing speed
Chinese agricultural purchases of non-soybean US farm products - monthly trade data
APEC Shenzhen outcomes - truce extension length and new commitments
G20 Miami summit format and agenda - whether Taiwan or AI governance appears in the communiqué
PLA military activity around Taiwan - exercise frequency, scale, and geographic scope
US semiconductor export control policy updates - further tightening or relaxation toward China
India-US trade negotiation status - whether a partial deal is reached before year-end
Quad summit scheduling - whether India sets a date for the delayed leaders' meeting
China's GDP growth rate and domestic consumption data - as indicators of Beijing's negotiating flexibility
US-China AI working-group meeting frequency and public readouts
China rare earth processing capacity expansion - as a signal of long-term leverage maintenance
American company earnings calls referencing China supply chain conditions - as a real-economy indicator of truce effectiveness
Beijing's treatment of the Iranian issue in US-China diplomatic communications

FAQ

Was the September 2026 Trump-Xi summit the first time Xi visited Washington? No. Xi last visited Washington in September 2015 during the Obama administration. The September 2026 visit was his first to Washington in eleven years and his first as a guest of Trump.

What is the Busan trade truce? The Busan Summit on October 30, 2025, in South Korea produced a one-year trade truce under which the US reduced tariffs from triple-digit peaks toward an average of about 47 percent on Chinese goods, China paused planned rare earth export controls, and Beijing committed to large-scale agricultural purchases from American farmers. The truce was due to expire in November 2026 and was extended at the September White House summit by two months, to January 10, 2027.

Why hasn't China been fully meeting its agricultural purchase commitments? China has been in the process of diversifying its agricultural import sources away from the United States - toward Canada, Argentina, Brazil, and Australia - for strategic reasons predating the trade war. Reversing that diversification requires purchasing from the US at the expense of countries China has cultivated as alternative suppliers, which has geopolitical costs beyond the bilateral US-China relationship. Agricultural trade data shows Beijing has been meeting soybean commitments more reliably than non-soybean commitments.

What is the pending $14 billion Taiwan arms package? It is a proposed US arms sale to Taiwan that includes air defense missiles, anti-drone systems, and other defensive capabilities. It follows a separate $11.1 billion package approved in December 2025. Taiwan's parliament appropriated $25 billion to fund both packages, but American approval of the $14 billion tranche has been repeatedly deferred, with Trump describing Taiwan arms sales as a "negotiating chip" in his conversations with Xi.

What is China's Thucydides Trap argument? The Thucydides Trap refers to political scientist Graham Allison's thesis that when a rising power threatens to displace an established power, war is often the result - historically, war has occurred in 12 of 16 such historical cases. Xi's invocation of the phrase at the White House - "The Thucydides Trap can be overcome" - is China's preferred rhetorical framing for the US-China relationship: a warning against inevitable conflict dressed as an offer of cooperation.

Trade Truce

Intelligence Reports & Briefings

Tea, Taiwan, and Tariffs: Who Really Won the Trump-Xi White House Summit?