South China Sea, explained.
A Chinese Coast Guard vessel’s crew film a Philippine supply boat near Second Thomas Shoal in the South China Sea, 5 March 2024 – one of regular confrontations between the two countries’ vessels in the area (Ezra Acayan/Getty Images)
Renewed military dialogue is a step forward, but the channels do not yet reach the coast guards and allies where maritime incidents are most likely.
Subscribe to The Informer for monthly expert analysis, and to Events for advance notice of visiting world leaders and distinguished guests.
You may unsubscribe from Lowy Institute newsletters at any time. For information on our privacy practices and how to unsubscribe, see our Privacy Policy.
US–China strategic stability doesn’t reach the water US–China strategic stability doesn’t reach the water
Nearly six months after Xi Jinping and Donald Trump agreed in Beijing to build a “constructive” relationship of strategic stability (Opens in new window), a formula they reaffirmed at the White House in September (Opens in new window), their militaries are talking again. Yet in the last week of October, Chinese navy and coast guard ships were active from Japan’s southern straits to the waters off Mindanao.
These facts are not a contradiction. The framework never promised to freeze military competition, and judged against its actual commitments it is making modest progress. The problem is narrower and more serious: the channels that exist do not fit together where dangerous encounters can occur. Whatever the intent, the effect is that bilateral calm and regional pressure coexist.
Start with what was promised. According to Beijing’s summit readout, the two militaries agreed to conclude a memorandum (Opens in new window) on crisis communication and prevention “as soon as possible”. The White House fact sheet did not mention it (Opens in new window), and a US official reportedly said (Opens in new window) the subject was not discussed. Command-level contact has nonetheless resumed: the US Indo-Pacific commander met his Chinese counterpart (Opens in new window) responsible for the Taiwan Strait in Canada.
PLA air activity around Taiwan was still below last year’s level in September. The Taiwan Security Monitor counted 255 sorties in September, 43% fewer than in September 2025 (Opens in new window), although activity rebounded from August. Maritime pressure presents a different picture: on 31 October, Beijing announced coast guard patrols east of Taiwan (Opens in new window), which Taipei challenged. The lower sortie count, therefore, offers an incomplete measure of restraint.
The calendar now matters.
The strongest defence of the framework is that crisis management allows rivals to maintain deterrence while preventing accidents from escalating. US Army Pacific still treats 2027 as a readiness benchmark (Opens in new window), although news agency Reuters reported on 1 October, citing unnamed sources, that US officials considered a military operation before 2028 increasingly unlikely (Opens in new window). Pressure below the threshold of war nevertheless continues. On 23 October, Taiwan’s coast guard warned that coordinated Chinese militia and coast guard operations (Opens in new window) were increasing pressure on Taiwan, Japan and the Philippines.
The harder problem is scope. On 27 October, three Chinese warships and a coast guard ship passed together through Japan’s Osumi Strait, the first time Japan’s defence ministry has disclosed (Opens in new window) the two services travelling together through Japanese waters. Further south, the Liaoning carrier group sailed around the Philippines and did not answer Philippine radio challenges (Opens in new window), although Manila said there was no cause for alarm (Opens in new window). By 28 October, Taipei counted 53 Chinese vessels, 34 of them warships (Opens in new window), between southern Japan and the Philippines.
The Liaoning ran a similar loop (Opens in new window) in October 2024. Ship movements alone do not prove coercion. But the US intelligence community did expect Chinese pressure on Japan to intensify (Opens in new window) through 2026, while US–China channels do not directly connect the vessels involved.
Regional channels do exist. Japan and China opened a defence hotline in 2023 (Opens in new window), and Manila and Beijing agreed in 2024 (Opens in new window) to improve maritime communications, including a future coast guard link. In August, The Interpreter reported that the Japan–China hotline had barely been used (Opens in new window) since its first call in May 2023. Manila said in March that amendments to its coast guard agreement were under discussion (Opens in new window). The gap is not an absence of channels but their coverage, reliability and coordination. A US–China memorandum alone cannot establish procedures for Chinese coast guard captains and Philippine or Japanese commanders when their ships meet.
Two further gaps compound the problem. The first is meaning. Xi has defined stability as including competition within proper limits (Opens in new window), and within days of the summit Beijing’s foreign ministry invoked the leaders’ consensus on Taiwan (Opens in new window). A US military official described Washington’s aim as enabling “both strategic stability and deterrence by denial (Opens in new window)”. Washington views allied exercises as credible deterrence; Beijing can portray them as breaching that consensus.
The second is implementation. Nuclear risk offers another test. US and Chinese officials met at least three times this year (Opens in new window) on the margins of the five nuclear powers’ dialogue, with Washington seeking more advance notice of missile tests. The reported meetings establish engagement, but not a shared launch-notification procedure.
The calendar now matters. Trump is expected in Shenzhen (Opens in new window) for APEC on 18–19 November, and Xi in Miami for the G20 on 14–15 December (Opens in new window).
Two indicators deserve attention. The first is operational: whether both governments agree on the memorandum’s purpose and scope, sign it, and use its procedures when an interception goes wrong. The second is whether the PLA’s November–December peak drill season (Opens in new window) passes without a large exercise around Taiwan. Neither would prove the framework caused restraint, but together they would suggest stability is becoming practice rather than language.
The region should also look beyond the bilateral line. Incidents involving coast guards and allies can also escalate. Extending rules for unplanned encounters to coast guards, and linking allied, regional and US–China channels, would do more for stability than another summit communiqué. Without that, the formula will have stabilised the conversation, not the competition. High-level stabilisation will remain incomplete where maritime incidents are most likely to occur.
About the author
Sophie Wushuang Yi
Dr Sophie Wushuang Yi is a Postdoctoral Teaching Fellow at Schwarzman College, Tsinghua University. She is also a Non-Resident Research Fellow at the Grandview Institution in Beijing and the Wang Gungwu Visiting Fellow at the ISEAS – Yusof Ishak Institute in Singapore.
The most-pressing world events explained by Lowy Institute experts and global contributors, in your inbox, every Thursday.
You may unsubscribe from The Interpreter at any time. For information on our privacy practices and how to unsubscribe, see our Privacy Policy.
