The phrase was coined late last month by Qin Shuo 秦朔, a veteran business commentator, in an essay examining Hongguo, ByteDance’s phenomenally successful micro-drama app. Citing a Nomura report based on QuestMobile data, Mr Qin noted that ByteDance apps swallowed 40.9% of the time spent on China’s 50 most-used mobile apps in July, compared to Tencent’s 29.1%. Hongguo alone attracted 168m daily active users—more than China’s four leading traditional streaming apps combined. To Mr Qin, the data posed a provocative question: has the Chinese internet shifted from an era anchored by Tencent’s social networks to one ruled by ByteDance’s algorithmic entertainment?

It is a staggering commercial feat. Few companies have ever translated consumer behaviour into frictionless code so successfully. From Toutiao and Douyin to TikTok and AI assistants, ByteDance’s ascent is a masterclass in engineering and corporate agility. Yet Mr Qin’s awe was laced with dread. He fretted that Hongguo’s cliffhanger-driven dopamine loop would inevitably cannibalise the time people once reserved for reading, reflection, and family life.

ByteDance went on the offensive. Li Liang 李亮, a senior communications executive at the firm, dismissed the “ByteDance Time” label as “complete nonsense.” He argued the Nomura data was narrow and noted that competitive leads are notoriously fragile.

The dispute quickly devolved from methodology to mudslinging. Pointedly noting Mr Qin’s recent appearance at a Tencent creators’ event, and his past work editing books authorised by ByteDance’s arch-rival, Mr Li planted a seed of suspicion regarding the commentator’s allegiances. Mr Qin flatly rejected the insinuations. Yet in the paranoid corridors of Chinese tech, a well-placed insinuation often does the heavy lifting.

The ferocity of the exchange begs a deeper question: why would a company fight so hard to deny its own dominance? The answer lies, at least partly, in the unforgiving nature of Chinese tech regulation. In a jurisdiction where powerful agencies exercise sweeping, often unchallengeable authority over internet platforms, anti-monopoly battles are frequently fought and decided in the court of public opinion long before any official paperwork is filed. For a Chinese tech giant, allowing a narrative of market dominance to take hold can carry significant risks. Once a regulator acts on a widely accepted public narrative, reversing that momentum is nearly impossible. A label like “ByteDance Time” is thus less a badge of honour than a source of unwelcome regulatory attention.

This peril is compounded by China’s expectation for its digital champions to drive industrial upgrading and fundamental scientific breakthroughs, rather than merely perfecting the monetization of leisure. These anxieties recently crystallised when a provocative comparison went viral online recently, claiming that while America directs the bulk of its AI computing power towards coding and enterprise research, China squanders its AI resources on generating cheap revenge-fantasy micro-dramas.

The People’s Daily, the Communist Party of China’s flagship newspaper, rushed to debunk the statistics as flawed and defended the value of entertainment while stressing China’s investments in hard tech. But the lament that “America uses gunpowder to make weapons; China uses it to make fireworks” struck a chord. It exposed a simmering national anxiety: what good is world-class algorithmic prowess if it is solely deployed to build a better entertainment house?

But should a company be penalized simply for being wildly successful at entertaining the masses? Ming Jinwei, a former state-media journalist writing under the influential moniker Mingshu Zatan, leapt to ByteDance’s defence. In an essay titled “There Is Only User Time, Not ByteDance Time,” he argued that in a ruthlessly competitive market, users flock to products because they actually meet their needs.

Cultural preferences, Mr Ming insisted, should not be weaponised into moral hierarchies. Reading a biography of a Song-dynasty poet is not inherently superior to binge-watching a micro-drama or scrolling through short videos. People vote with their thumbs, and their choices reflect genuine value.

It is a robust defence. Yet to wave this away as mere consumer choice ignores the reality of the algorithmic age. A world-class recommendation engine does not merely cater to preexisting desires; by endlessly calibrating what users encounter next, it actively shapes them. It deliberately blurs the line between passively serving an audience and actively inducing its habits.

In China, however, this universal debate over corporate responsibility intersects with a deeply ingrained official, cultural, and intellectual paternalism. There is a long-standing tradition among the country’s elites of delineating what constitutes “edifying” pursuits versus “vulgar” distractions. This top-down moralising begs an uncomfortable question: by what right must every hour of a citizen’s day be productive or edifying? If hundreds of millions of ordinary people, exhausted by the grinding pressures of a hyper-competitive society, actively seek out the empty solace of a three-minute soap opera, why should they be scolded for it? And what makes an opera sung in a foreign language, or some music show put up by instruments that most cannot even name, inherently more worthy than a melodrama on Hongguo? The right to be unproductive—and yes, even the right to consciously choose "vulgarity" and escapism—is a fundamental, if unacknowledged, human comfort.

The phrase "ByteDance Time" is therefore a tribute to a commercial juggernaut masquerading as a regulatory red flag. The company should certainly answer for how its code shapes human habits. But it should not have to bear the burden of intellectually redeeming the masses. As cultural gatekeepers fret over what these hours mean for society, they forget whose time is actually being spent. The clock belongs to the audience. And whether an evening spent on Hongguo is culturally edifying or cheerfully wasted is a verdict that rests with no one else. (Enditem)

ChinaTalk

Is Half of China’s AI Making Slop?

Zilan Qian is a research scholar at GovAI researching AI development and regulation in China…

Read more

2 days ago · 58 likes · 11 comments · Zilan Qian

At the recent Valdai Discussion Club, I asked Russian President Vladimir Putin if he would be open to meeting Chinese President Xi Jinping and US President Donald Trump at the Asia-Pacific Economic Cooperation (Apec) forum in Shenzhen. I also asked, more broadly, what role China could play in promoting peace in Ukraine, mediating and contributing to peacekeeping.

Putin was receptive, warmly welcoming the idea of a China-hosted meeting. Trump, too, has expressed a willingness to meet Xi and Putin. These expressions of openness offer an opportunity we must seize in pursuit of peace.

Some in the United States and Europe question the value of such a summit. They may fear restoring Putin’s standing or normalising the conflict without yielding any real progress towards an end to the war. That risk is still real but these largely Western fears must be tempered by the very real, very human cost of letting the war continue unabated.

From the very beginning of the war, I have been arguing that China has a constructive role to play. In March 2022, less than three weeks into Russia’s invasion, I wrote an article in The New York Times titled “It’s Time to Offer Russia an Offramp. China Can Help With That.”

In August last year, I extended that train of logic in Foreign Policy, where I proposed a seven-party format for peace negotiations and suggested that a peacekeeping force include troops from European and Brics countries. A meeting of the three leaders in Shenzhen could help lay the groundwork for these broader diplomatic and peacekeeping efforts.

Without Ukraine’s participation, such a meeting is unlikely to produce a diplomatic masterstroke that ends the war. But the three leaders need not put the war to rest at a table in Shenzhen to make progress. Serious consideration of a ceasefire along the current line of contact – an approach Ukrainian President Volodymyr Zelensky has supported – could give peace efforts fresh momentum, leaving the final status of disputed territories for subsequent negotiations.

Further agreement could be sought on the protection of nuclear facilities, or a mutual end to strikes on energy infrastructure, new prisoner exchanges or improved humanitarian access. Of course, issues of sovereignty, security guarantees, Russia’s security concerns and Europe’s security commitments would remain. But these need not all be resolved at once before a ceasefire can begin.

Already, years of war have inflicted immense suffering on both Russia and Ukraine. Neither side achieved its objectives, and both need a dignified way out that protects their core interests. Thus, we must open a path towards peace without requiring either side to concede defeat. China could help bridge this gap by bringing Xi, Trump and Putin together in Shenzhen to explore how a ceasefire and wider negotiations involving Ukraine might work.

The United Nations could stand at the core of such an arrangement by monitoring any ceasefire and preventing a return to fighting by physical peacekeeping if the parties are so inclined.

For a broader negotiation on settlement, I find myself returning to a seven-party format that would bring together the five permanent members of the UN Security Council – China, France, Russia, Britain and the US – alongside Ukraine and the European Union. An understanding reached in Shenzhen could give momentum to this broader negotiating process and free up the Security Council for action.

A peacekeeping force composed entirely of Nato or European troops might be unacceptable to Russia. Yet a mission without European participation would struggle to reassure Ukraine and its neighbours. Any workable arrangement would need to address both concerns.

In my Foreign Policy article, I proposed a peacekeeping force drawing troops from five European countries – Germany, France, Britain, Poland and Italy – and five members of the Brics grouping, such as China, India and Brazil. And while the exact formulation, from the composition and mandate to the terms of deployment, would need to be negotiated with the consent of Russia and Ukraine, and the agreement of the contributing countries, the core idea remains sound.

As a permanent member of the UN Security Council and the largest contributor of peacekeeping personnel among the five permanent members, China is well positioned to help. Its distinctive role in the global economy and dialogue with Russia, Ukraine, the US and Europe, give it channels through which to coax the parties towards peace. But this process must begin somewhere, and that somewhere could be Shenzhen.

Nor should Apec be the end of the diplomatic effort. With an understanding reached by the leaders in Shenzhen, the Security Council may pick up the momentum by advancing ceasefire and peacekeeping proposals. The Group of 20 could then join the effort to rally support from major economies with an interest in maintaining the deal. Each would play a different role and would not be tasked with settling the entire conflict.

China’s role in the Saudi Arabia-Iran rapprochement in 2023 showed Beijing’s ability to open a door after years of enmity. The Ukraine war is different, and there can be no guarantee of a similar outcome. But diplomacy must begin somewhere, particularly when fighting has brought such destruction.

The central question is whether such a meeting could help bring the war to an end. In that light, any gain in prestige for an individual leader is almost irrelevant. And if the three leaders could agree on practical steps towards peace, they would give the United Nations and the wider international community a material basis upon which peace could be built. (Enditem)

CCG Update - Center for China and Globalization

Henry Huiyao Wang: Only a multipolar coalition can secure Ukraine peace

After President Xi Jinping and President Donald Trump had their first meeting since 2019, signals from both Beijing and Washington seem to hint at space for diplomacy on Ukraine…

Read more

a year ago · 14 likes · CCG Update

Yet amid the excitement surrounding the launch, a controversy has emerged over Kimi’s apparent efforts to promote the model through overseas influencers.

An X user known as Elshayib published screenshots of what he described as promotional instructions from Kimi, saying he had refused to participate on those terms. According to the screenshots [1] [2] [3][4], creators were required to submit drafts for review and obtain approval before posting. They were explicitly told not to use paid-partnership label, or link to Kimi’s website or tag its official account.

The document also suggested talking points comparing Kimi favourably with American competitors, including portraying it as an open-source equivalent of Anthropic’s Claude Opus 5.5 at one-fifth the price.

The document has not been independently authenticated. It remains unclear whether the instructions originated directly from Moonshot, how many creators were involved, or whether anyone followed them. Nevertheless, the allegations have begun circulating in both English- and Chinese-speaking AI communities.

German AI commentator @kimmonismus, who had previously praised Kimi’s technical achievements, criticised the alleged practice while stressing that his own coverage had not been paid for. The controversy was also picked up by Latent Space, a prominent English-language AI engineering newsletter. Chinese-speaking users have discussed the allegations as well.

A familiar Chinese marketing playbook

There is nothing inherently wrong with paying influencers to promote a product, nor with reviewing sponsored content before publication. Companies in all sectors around the world do both. What makes the alleged instructions particularly troubling is the combination of tight editorial control and an apparent effort to conceal the commercial relationship.

That combination looks strikingly familiar in China’s fragmented media and social media landscape. Influential online commentators, including prominent accounts with large followings, and even some established media outlets regularly publish content about companies and products in exchange for payment. Such content is often subject to review or approval by the paying company, yet the commercial relationship is rarely disclosed to readers. The practice is neither unique to China nor universal within it, but it has become so widespread and widely tolerated in the country’s information ecosystem that it is often treated as simply the way things work. A Chinese marketing team accustomed to such arrangements might therefore take the same approach to an international campaign, assuming that what is considered routine and acceptable at home would be equally so abroad.

There is also an almost Kafkaesque footnote to the controversy. A Chinese-language report covered the allegations that Kimi had sought to pay influencers to promote its model without disclosing the commercial relationship. The report is now inaccessible through its original link. There is no evidence that Moonshot was involved in its disappearance, or even that the article was deliberately removed. Still, the irony of a report about concealed paid publicity becoming inaccessible for reasons unknown is hard to miss.

An unnecessary geopolitical turn

Another feature of the alleged promotional brief is particularly ill-considered. Among its suggested talking points are “China just made Opus 5.5 free to use for everyone” and “The US cut China off from advanced chips”, followed by praise for Kimi’s achievements despite those restrictions.

Most of the document promotes the model’s capabilities, as expected. But these lines turn a company’s technological achievement into a national accomplishment and frame a product launch as another episode in US-China competition.

Moonshot’s achievements inevitably have geopolitical implications. That does not mean the company benefits from making geopolitics central to its own advertising. Chinese technology companies seeking international customers often have good reason to emphasise product quality and commercial value rather than invite political controversy. Even when geopolitical scrutiny is unavoidable, there is little advantage in voluntarily intensifying it.

The risks are particularly acute for Moonshot, which already faces scrutiny from both American AI companies and the US government. Anthropic has accused it and other Chinese AI developers of unauthorised model distillation. In July, White House science adviser Michael Kratsios directly accused Moonshot of using Anthropic’s technology to develop Kimi K3, while Treasury Secretary Scott Bessent raised the possibility of sanctions and trade blacklisting. Moonshot has disputed the allegations, and independent researchers have questioned whether distillation could explain Kimi’s technological advances.

In such an environment, an undisclosed paid promotion campaign could readily be portrayed as something more sinister. It is hardly inconceivable that critics might even characterise it as part of an organised Chinese influence operation aimed at undermining American AI companies, despite the absence of evidence for such an accusation. By encouraging influencers to frame Moonshot’s technological achievements as Chinese victories over American restrictions, the alleged brief would only make such interpretations easier. An otherwise ordinary marketing misjudgment could thus become unnecessary ammunition in an already fraught geopolitical debate.

None of this amounts to a verdict on Kimi’s international communications as a whole. In fact, Moonshot has demonstrated considerable sophistication elsewhere. Its inventive promotional video, featuring a retro computer and elaborate 3D graphics, is a strong example of creative product marketing. It has garnered 2.1 million views in a day.

Founder Yang Zhilin’s earlier technical presentation at NVIDIA’s GTC 2026 likewise demonstrates the company’s ability to engage an international technical audience. His first video got over 1.6 million views.

Those successes make the alleged influencer instructions all the more incongruous. They suggest not a general failure of international communications, but the possible persistence of a domestic marketing habit within an otherwise increasingly sophisticated global operation.

China’s AI companies have made remarkable technological progress and earned growing international recognition. The lesson from this episode is not that Kimi does not know how to communicate internationally, but that familiar domestic marketing practices do not always travel well—and that a company already navigating difficult geopolitical waters has little reason to make them rougher for itself. (Enditem)

Following the disappointing second-quarter economic data, the Communist Party of China’s Politburo met on July 30 and called for stronger countercyclical measures and greater efforts to expand domestic demand, including through the development of the so-called “six networks”: water networks, new-type power grids, computing power networks, next-generation telecommunications networks, urban underground utility networks, and logistics networks. On August 28, the National Development and Reform Commission (NDRC) convened a nationwide meeting to arrest the decline in investment, directing officials to accelerate major infrastructure projects, including the six networks. The agency followed up in early September with more detailed arrangements for project financing and implementation.

Against this backdrop, Zhang Bin, Deputy Director of the Institute of World Economics and Politics at the Chinese Academy of Social Sciences, makes a broader case for public investment. He argues that despite China’s already extensive infrastructure network, there remains considerable room for investment in public services, affordable housing, and the maintenance and upgrading of existing facilities. Such spending, he contends, can break the vicious cycle of weak demand, slowing income growth, and declining expenditure.

Zhang also challenges the assumption that greater government borrowing would necessarily aggravate China’s debt problems. He argues that slower debt growth can depress incomes and actually push debt ratios higher, whereas debt-financed public investment, by stimulating growth and improving debt repayment capacity, could ultimately help reduce the economy’s overall debt ratio.

The article was published on 1 October 2026 by Sanlihe, an online commentary platform affiliated with China News Service, as part of a series examining China’s investment trends. Zhang has agreed to this translation.

扩大内需仍需公共投资挑大梁

Public Investment Must Still Take the Lead in Expanding Domestic Demand

The key to expanding domestic demand is to use a powerful external force to break the vicious cycle of falling income, spending, expectations, and credit. China’s past experience shows that increasing public investment can be an effective way to achieve this. By substantially boosting business and household incomes in a short period, government investment has helped break this vicious cycle and put the market back on track.

Such investment has not only stimulated domestic demand but also played a crucial role in improving living standards, upgrading industries, and protecting the environment over the longer term. It has been an important driving force behind China’s economic miracle.

Public investment still has much to offer today, both in driving domestic demand and laying the foundations for future development. Borrowing to finance additional public investment does not necessarily increase debt risks. On the contrary, if used appropriately, it can help reduce them.

Public investment is the key external force needed to stimulate domestic demand

When domestic demand is weak, income growth slows for businesses, households, and governments alike, prompting each to cut back on spending. This can easily create a vicious cycle: one person’s spending is another person’s income, so when everyone cuts spending, income across the economy falls. The resulting downturn then reinforces itself. This is a classic example of market failure, which market forces alone are unlikely to resolve.

The key to expanding domestic demand, therefore, is to introduce an external force capable of breaking this cycle, getting the market back on track and restoring economic vitality.

Public investment can provide precisely this external force. When demand is weak, public investment has two important advantages over market-driven spending. First, public investment decisions are made independently by the government, which is not bound by the usual market logic of cutting expenditure when income falls. Second, the government can mobilise substantial investment in a relatively short period, sharply increasing business and household incomes and powerfully reversing the downward momentum generated by market forces.

China’s experience offers several examples. In 1998, a major infrastructure investment programme, centred on the development of a nationwide expressway network, helped the economy overcome weak demand. In 2008, China’s RMB 4 trillion infrastructure investment package helped the economy achieve an early V-shaped recovery from the global financial crisis. In 2015, government support for the redevelopment of shantytowns helped revive the property market and credit growth, pulling the economy out of another period of weak demand.

A substantial increase in public investment leads to a marked recovery in income growth across the economy, which in turn boosts workers’ earnings and consumption. Past experience shows that public investment and consumption are not competing forces. Rather, they reinforce each other. When demand is weak, public investment is one of the most effective ways to address market failure, revive economic activity, and stimulate growth in workers’ incomes and consumption.

China still has ample scope for public investment

China has invested heavily in infrastructure, including electricity and gas networks, railways, roads, airports, ports and terminals, and has made remarkable progress. In many areas, its infrastructure is now on a par with, or even ahead of, that in developed countries. But this does not mean that the opportunities for further public investment have been exhausted. There is still much to be done to improve living standards, upgrade industries, and protect the environment.

In terms of new investment, there is considerable scope to expand public facilities and services in areas such as education, culture, sport, underground utility networks, and urban roads. China still has fewer libraries, museums, and sports facilities per million people than developed countries. Some 110 million migrants living in the urban areas of prefecture-level cities and above struggle to establish permanent homes in the places where they live. Local governments lack sufficient affordable housing, schools, and healthcare facilities to meet their needs. These shortfalls point to substantial unmet demand for public investment.

As for existing infrastructure, China has traditionally placed greater emphasis on building new facilities than on maintaining, repairing, or rebuilding those already in place. Decades of construction have left the country with an enormous stock of public infrastructure, much of which could be made considerably safer, more effective, and more convenient to use. Significant additional investment is needed to ensure that these facilities can fulfil their potential.

Public investment has a distinctive role in addressing problems that ordinary people and businesses face but struggle to resolve, even when they are willing to pay. It does not take much looking to find such problems in everyday life. As long as public investment is guided by the needs of ordinary people and businesses, there will be no shortage of worthwhile projects to invest in or work to be done.

The government has not borrowed excessively, and more borrowing for public investment need not increase debt risks

International experience shows that, over the past two decades, government borrowing has accounted for approximately 49 per cent of the total increase in debt across advanced economies. The corresponding figures are 58 per cent in the United States and 74 per cent in Japan, compared with just 34 per cent in China. These figures use the Bank for International Settlements (BIS) definition of government debt, which covers not only official government debt but also some hidden local government debt.

Strong demand for Chinese government debt among domestic and foreign investors suggests that they consider it relatively safe and are not generally concerned about its sustainability. Some local governments face unsustainable hidden debt burdens, but this does not mean that the government sector as a whole has borrowed excessively.

Several rounds of debt swaps and other debt resolution measures have substantially reduced local governments’ hidden debt, but a large amount remains outstanding. Many local governments lack the fiscal resources to meet principal and interest payments on their own. Countercyclical policies, including greater public investment and lower policy interest rates, can help correct market failures and restore economic vitality. They can also ease the debt burden on local government financing vehicles through several channels: increasing local tax revenues and land transfer sales, boosting the revenues of financing vehicles and the value of their assets, and reducing borrowing costs. Over time, these measures could help alleviate local government debt problems, particularly in more economically developed areas and those attracting population inflows.

Concerns about an overly high macroeconomic leverage ratio have led some to argue that borrowing should be kept to a minimum to reduce debt ratios and debt risks. This reasoning is broadly sound for individual households and businesses, because borrowing less may have little effect on their income growth, allowing their debt ratios to fall.

But the situation is very different for a country as a whole. Slower debt growth across society generally goes hand in hand with slower income growth, meaning that the debt ratio may actually rise rather than fall. Between 2002 and 2025, the correlation coefficient between debt growth and the debt ratio in China was -0.49, indicating that slower debt growth was associated with a rising debt ratio.

The United States, the euro area, and Japan have seen similar patterns. In recent years, debt has expanded rapidly in these economies, yet their debt ratios have gradually declined, demonstrating a negative correlation between debt growth and debt ratios.

The lesson is that by borrowing to expand public investment and thereby raising income across the economy, China can reduce rather than increase its overall debt ratio, strengthen its capacity to service debt, and keep debt risks under better control.

Subscribe now

Share

In New York last month, US-China talks on artificial intelligence opened up a path of cooperation. This was affirmed by the Xi-Trump summit in Washington where both presidents agreed to establish a formal AI dialogue and hotline. With the next AI meeting set to happen in November, it is imperative that both countries further cooperation.

The AI ecosystem is in flux. American tech companies are draining their coffers as they commit vast sums to chips, data centres and energy before the revenue model is even settled. Trillions of dollars are being committed to the AI buildout. While this exposure is not the same as the overleveraging in the mortgage market that triggered the 2008 global financial crisis, a sudden bursting of the AI bubble could have far-reaching consequences.

Much of US economic growth this year has been from AI or AI-enabling companies. Yet, only 38 per cent of Americans saw “more benefits than drawbacks” in AI products and services – well below the global average – against 85 per cent in China, according to the Ipsos AI Monitor 2026. In May, the Ipsos Consumer Tracker found that most Americans would oppose a data centre in their community.

The surge in AI capabilities has raised concerns. Rogue AI agent behaviour has gone from attempting to delete an entire email inbox to escaping sandboxes to launch cyberattacks. In July, during an isolated evaluation, OpenAI agents circumvented their restrictions to launch autonomous attacks on developer platform Hugging Face.

That same month, Calif, a small US security company, created WeWorm to exploit a zero-click vulnerability in WeChat’s calling system, enabling it to seize accounts and spread between phones without the recipient answering. In India, cybersecurity researchers at Hacktron used Anthropic’s Claude models to break into OpenAI, compromising employee accounts and reaching potentially sensitive software systems.

Both hacking incidents were conducted under friendly auspices and caused no damage – Tencent Holdings and OpenAI were quickly alerted and patched the bugs. But they portend the threat of AI-enabled cyberattacks, both from poorly managed AI and actors empowered by the devolution of yet more cyberwarfare capabilities into private hands.

Encapsulating the moment is the increasingly urgent call for safeguards by leading AI developers and scientists. Anthropic CEO Dario Amodei proposes opening up leading laboratories to permanent independent evaluators and coordinating safety standards across companies and governments, OpenAI CEO Sam Altman is committing its models to third-party evaluation, and Google DeepMind chairman Demis Hassabis is backing pre-deployment testing of frontier systems.

Last week, an open letter by at least 420 scientists across the globe called for urgent measures to prevent an AI fallout, including the creation of an international body to record and track containment failures.

These concerns provide an agenda for the coming US-China AI dialogue. Talks, however, should extend beyond managing incidents and developing safeguards towards building a sustainable business case for AI. Given its increasingly self-sustaining AI ecosystem, China has plenty to contribute.

First, it is clear the US and China ecosystems possess different pieces of the same business case.

American developers need applications capable of converting their frontier performance into sustained revenue. China offers one of the world’s largest environments for testing, refining and commercialising applications across manufacturing, logistics, medicine, education, public services and embodied AI.

While China’s industry could benefit from investment and research at the cutting edge of frontier model development, its capabilities in coordinating data centres and power supply could help take the strain off US compute capacity.

Cooperation on efficient models, cooling and energy management could help both countries meet rising demand while reducing the costs borne by consumers and communities. A healthy ecosystem has room for both American frontier models and efficient Chinese open-weight models.

Chinese open-weight models offer lower costs per token, making them economical companions to American frontier models. Companies can use them for routine, high-volume tasks and reserve frontier models for problems that warrant their capabilities. This high-low mix can make American frontier capabilities more economical to deploy within a wider workflow, strengthening the business case for both countries.

To sustain this cooperation, both countries must also confront AI risks and manage them together. The United States and China established one narrow principle two years ago when they affirmed that people, not AI, should decide on the use of nuclear weapons. I propose going further to adopt a reciprocal no-first-use commitment covering AI-enabled autonomous weapons.

The final AI issue that must be discussed is institutional. Fragmented governance risks building walls of incompatible standards and duplicated infrastructure. The US-led Pax Silica initiative and China-led World AI Cooperation Organisation should not be posed in opposition to each other.

Instead, both countries should place their global AI governance and interests under a shared UN-linked umbrella, giving countries that use AI a voice alongside those that develop it.

Finally, an annual US-China AI summit would serve as a strong foundation, bringing governments together with laboratories, scientists, universities, companies and civil society organisations. For now at least, it is still ultimately people who decide what moral value our technology has through what we choose to do with it.

Share

In his remarks, Chen, CCG’s Honorary Chair and former Commerce Minister of China, said that Beijing still has work to do to make life easier for Chinese companies investing abroad, particularly by securing investment agreements with other countries. He also rejected the idea that China’s trade surplus is simply a result of “overcapacity” and insisted that AI would not replace human intelligence.

CCG Update - Center for China and Globalization

2026 China Outbound Investment Forum Held in Beijing

The 2026 China Outbound Investment Forum, hosted by the Center for China and Globalization (CCG), was held in Beijing on Saturday, October 17…

Read more

11 days ago · 7 likes · 1 comment · Yuxuan JIA

This transcript is based on a recording at the event and has not been reviewed by the speaker.

Remarks by Chen Deming, Former Minister of Commerce of China

I came to this CCG forum because I wanted to hear from business leaders about their experiences of going global.

A couple of days ago, I was in Singapore attending an international forum where much the same issues were discussed. I hadn’t prepared a speech for today, so last night I put together a few thoughts that I would like to share with you.

First of all, I think CCG has been doing some very valuable work over the years in exploring how Chinese companies can go global. I believe this is a very worthwhile direction to pursue.

But as everyone knows, it is really only in the past couple of years that Chinese companies going global has become a subject of such great interest, both at home and abroad, among business leaders and governments alike.

Why is that? Because history has increasingly demonstrated how necessary and important it is for Chinese companies to go global.

I learned a great deal from the presentations just now, although many of the speakers naturally focused on their own products and business approaches.

But there is one question I would like to discuss. Why should Chinese companies go global?

I think the first reason is China’s responsibility as a major country. In today’s globalised economy, China should share its knowledge, capabilities, and successful experiences with the rest of the world through the outward investment of its companies.

Look at China’s own success. Through several decades of reform and opening up, China has rapidly transformed itself from a poor and weak country into the world’s second-largest economy, the world’s largest trading nation in terms of imports and exports, and a country with the world’s second-largest stock of foreign investment.

These achievements are not simply the result of the advantages of China’s system or the hard work of the Chinese people. They also owe a great deal to the support China has received from the international community. That should never be forgotten.

And now that China has developed its own capabilities, it should give them back to the international community.

If other parts of the world remain troubled by geopolitical conflicts and instability, and their economies cannot develop, that is not good for China either.

So my first point is that China, as a major country, has responsibilities beyond its own borders.

Second, there is the situation of China’s domestic economy.

After decades of development, China has made tremendous progress, but there are still bottlenecks and difficult problems to overcome. In particular, the current situation of strong supply and weak demand is another reason for Chinese companies to consider going global.

Why do I say “strong supply and weak demand” rather than “overcapacity”?Because what this situation really reflects is weak demand and insufficient consumption, alongside substantial production and supply capabilities.

If it is simply described as overcapacity, it creates the impression that China’s excess production alone is responsible for international trade surpluses and deficits. That is not something China accepts.

Supply and demand are extremely difficult to balance across the entire world economy, let alone calculate precisely. The real question is how to increase consumer demand around the world. And that requires investment.

Many Chinese companies have already been through years of intense competition at home, and now they are entering foreign markets. This is a very important development.

So, to put it simply, Chinese companies have good reasons to go global: China’s responsibilities as a major country, the success it has achieved through reform and opening up, and the current imbalance between strong supply and weak domestic demand.

The second issue I would like to discuss is what the government should do to support companies going global. In other words, what are the government’s functions and responsibilities?

Earlier this morning, Vice Minister Chen Jian, who was responsible for outbound investment at the Ministry of Commerce for many years, gave us an excellent presentation. He discussed the State Council’s regulations, as well as specific rules issued by the Ministry of Commerce and other government departments.

There is one point I would particularly like to emphasise: China still needs to sign more specific bilateral investment agreements with key countries and regions. This is something I’ve come to appreciate through my own experience.

China had long sought to negotiate a bilateral investment treaty with the United States. During the Obama administration, negotiations on the treaty text had essentially been completed, with the provisions agreed upon by both sides. The two countries were approaching the final stage of reconciling their respective negative lists when Donald Trump took office as the 45th president of the United States.

His administration then brought the entire negotiation process to an end, and the talks have not resumed to this day.

Now that Trump is back in office as the 47th president and is calling for businesses to return to the United States, I wonder how different things might have been if China and the United States had concluded an investment treaty with a clearly defined American negative list.

At the international forum in Singapore, discussions also kept coming back to China-EU relations. How did the EU move from having a trade surplus with China to running a deficit?

I discussed this question with a former US secretary of defence and two former prime ministers, one from Denmark and the other from Australia.

I made two points.

First, China-EU trade cannot be judged solely by the often-cited figure of a €1 billion daily trade deficit on the EU side, because that only covers trade in goods. In services, China runs a trade deficit with the EU amounting to tens of billions of euros, and that needs to be taken into account as well.

Second, it would be wrong to describe the EU as economically backward. It is a very strong economic bloc. Even after factoring in its trade deficit with China, the EU still runs an overall trade surplus with the world. So Europe has considerable potential for further development.

Of course, US tariff policies have created difficulties for the EU’s exports to the United States, its largest export market. That makes the economic complementarity between China and Europe an especially valuable opportunity.

There is no point in quarrelling and walking away from the negotiating table. Both sides have to find ways of working together.

During those discussions, some European participants asked me why Chinese companies were not investing more extensively in Europe. Why, they asked, were some Chinese investments limited to processing and assembly? Could Chinese companies bring their supply chains with them when investing in Europe?

That struck a chord with me, particularly after hearing Chairman Jiang of Far East Holding Group speak today about Chinese companies expanding overseas together with their supply-chain partners.

But there is another point I want to make to my EU counterparts. China and the EU had been discussing an investment agreement for many years, going back to my time as Minister of Commerce.

When the negotiations were close to completion, President Xi Jinping personally called Charles Michel, then President of the European Council and Ursula von der Leyen, President of the European Commission, hoping that the agreement could move forward through the necessary legal procedures as soon as possible.

China would have to complete its own legal procedures, while the EU would have to go through the European Commission.

But then, over a very small issue between the two sides, the agreement was put on hold. So to this day, China and the EU still don’t have an investment agreement.

That means Chinese companies don’t know what the EU’s negative list is, what legal protections, investment incentives, and environmental standards are in place for them.

(While he was speaking, Chen requested that a portion of his speech be limited only to the room. This part has been omitted from this transcript.)

So, from the government’s perspective, once China has put its overall regulations in place, it is very important for the relevant government departments to negotiate and sign investment agreements with key countries and regions. Without these agreements, Chinese companies can encounter obstacles when investing abroad.

Take cross-border data flows in e-commerce, for example. Or national security concerns. How can security concerns be balanced with the need to promote investment and development? How can differences between the policies of the home country and those of the host country be addressed? These are all questions that need to be discussed.

China needs to take its own security interests into account while also considering the development needs of Chinese companies going global.

I was particularly struck by Chairman Jiang’s remarks just now. He talked about companies bringing their supply-chain partners overseas together, taking small steps and learning through trial and error, and developing local talent. These are not magic formulas. They are lessons he has learned through decades of practical experience. Every company going overseas needs to think about these things.

But the point I want to make today is that the government still has a lot of practical work to do, particularly in securing bilateral investment agreements with key countries and regions to protect Chinese companies investing overseas.

My third point concerns overseas investment in the age of AI. Ms Miranda Wang from PMI spoke about this earlier, and I think this may be a very important subject.

We also heard from Mr Kai-Fu Lee, who is an expert in this field. I would first wish him every success with his software in China and overseas. I may have some slightly different views from him about how AI can guide Chinese companies going global, but these are my personal thoughts.

I think there are many ways for Chinese companies to invest overseas.

Since we’re talking about going global today, the first step could be trading goods, testing the waters, and opening up new markets. Investment comes next.

The third step is to build a complete system of overseas operations. Chinese companies may not be fully ready for that yet. It will take a collective effort, including closer coordination among companies and guidance from national macroeconomic policies.

But overseas investment has already become much more diversified. Greenfield investment does not necessarily have to be the main option. Chinese companies can acquire controlling stakes or minority shareholdings. They can enter into cooperative business arrangements, provide technical services, or license patents. These are all ways of entering international markets.

In some countries in particular, such as the United States, there is resistance to direct Chinese investment. But Chinese companies can still license their patents or provide technical services to American businesses. In fact, this is already happening.

So I think that in the AI era, with more information available, Chinese companies have more options to consider when investing overseas. They can carefully assess investment opportunities themselves or ask independent third parties to conduct research and assessments. Talent is also very important.

But what I really want to say is that AI won’t necessarily replace human intelligence.

In my view, human society has passed through the agricultural age and the industrial age. It is now approaching, or perhaps already beginning to enter, an age of intelligent society.

This transformation may take not just 50 or 100 years, but much longer. What is known about AI today is only the beginning.

On 6 September this year, ChatGPT released its new GPT-6 model, which was regarded as having reached the stage of AGI, or artificial general intelligence. If that marks an initial step from AI toward AGI, many more changes can be expected.

Take the relationship between CPUs and GPUs, for example. CPUs are the processors used in conventional computers, while GPUs, such as those developed by Nvidia, are specifically used for AI computing.

Previously, when large amounts of computing power were being used to build and train AI models, the ratio of CPU to GPU requirements was around one to eight. But with the emergence of AGI, AI is no longer simply responding to human questions one at a time. It can continue reasoning, process information, and search for signals from around the world, even when it is no longer directly engaged in a conversation with a person. That increases its interaction with the outside world and places greater demands on CPUs.

As a result, the ratio of CPU to GPU requirements has shifted to around one to four, and it may eventually reach one to two or even one to one. This represents a major opportunity for Chinese companies investing overseas.

When it comes to computing power, China is still behind the United States, partly because of American restrictions and sanctions. But in applications, particularly in the physical world, I don’t think China is moving any slower.

So when Chinese companies go abroad, they can bring their existing open-source software and their strengths in real-world applications to support development in other countries.

But human beings will not be replaced by AI. The future will be a society where human intelligence and artificial intelligence coexist.

If AI advances so quickly that technological progress exceeds what human beings can cope with, what would happen to the world? Humanity would not allow that to happen. The consequences could be frightening.

So when President Xi Jinping met with President Trump in the United States, they agreed that the two countries should discuss AI governance and that governments need to play a role.

Without government oversight, what would happen if AI technology were fully applied to warfare? Imagine tens of thousands, or even hundreds of thousands, of drones flying together like a black carpet across the sky, each carrying missiles and capable of identifying its targets. What kind of warfare would that create?

Nobody wants to be replaced by AI. From a theoretical perspective, human beings are carbon-based life forms. Through sunlight, water, and the food that comes from plants and animals, people obtain energy and are able to create many things. But there is no such thing, at least at present, as an independently existing silicon-based form of life.

One of the defining characteristics of life is the ability to reproduce itself. Today’s AI systems, as well as computers and robots operating in the physical world, still depend on electricity. They cannot yet generate their own electricity. They cannot replace the electricity that humans supply to keep them running.

So in the future, artificial intelligence and human intelligence will have to work and evolve together. If AI development completely exceeds humanity’s ability to cope with it, human beings will seek to bring it under control.

That brings me to another point raised earlier by Professor Kai-Fu Lee: management.

Recently, my former mentor, Professor Zhou Sanduo, the founding dean of the business school at Nanjing University, completed a new book. Many of the management textbooks used in China today were written by him. He’s now 93 years old, and I believe his new book will be published next month.

Forgive me. I’m not trying to advertise my mentor’s book. The reason I mention it is that it has made me think about management in the AI era.

I would compare the changes taking place to a Copernican revolution. In terms of efficiency and many other aspects of management, AI may well become more capable than human beings.

But think about this. Without people taking initiative, and without the mutual understanding and close cooperation between people, AI cannot get the job done. Human emotions and respect for life are not things AI can address.

That is why I am convinced that the society of the future will be one where artificial intelligence and human intelligence develop together. Everything being said today is provisional. Fifty or a hundred years from now, if some of you are fortunate enough to live that long, you may look back and see how much the world has changed from what is being discussed here today.

I firmly believe that human civilisation will endure. After all, the Earth itself has about another 4.5 billion years of life ahead of it, and the oceans will remain for another billion years or so. At the very least, humanity should continue to exist throughout that time.

So when Chinese companies go global and think about serving businesses and people in other countries, they need to take a broader view. What approaches will work best? How can global consumption be increased? How can productive capacity be expanded? Thinking about these questions can lead to more comprehensive decisions.

Companies should not allow themselves to become constrained by AI. But neither should they reject AI and stick to their traditional ways of doing things.

I have great confidence in the capabilities and determination of the Chinese people. China’s progress in using open-source technologies across different fields has surprised many American scientists. Even Jensen Huang, speaking in Taiwan, expressed surprise at what China has achieved. I think China has done very well in this respect.

So I sincerely hope that all the entrepreneurs here today will embrace new ways of thinking in the AI era and achieve even greater success as they go global.

I also hope the Chinese government will continue working with other countries to conclude bilateral and multilateral investment agreements, establish clear negative lists, and spell out investment incentives, the responsibilities of host governments, and even the circumstances under which compensation may be required. All these things need to be clearly defined.

I wish you all every success. Thank you very much.

(End of remarks. The forum then broke for lunch.)

Subscribe now

CCG Update - Center for China and Globalization

2026 China Outbound Investment Forum Held in Beijing

The 2026 China Outbound Investment Forum, hosted by the Center for China and Globalization (CCG), was held in Beijing on Saturday, October 17…

Read more

11 days ago · Yuxuan JIA

Share

Li Yuankun, Assistant Research Fellow at the Institute of World Economics and Politics at the Chinese Academy of Social Sciences (CASS), focuses on a more specific issue: the interest-rate benchmark used to estimate below-market borrowing. He argues that the use of China’s Loan Prime Rate (LPR) as the benchmark can significantly increase the estimated level of support.

This article was originally published by VoxChina.org on 1 July 2026.

VoxChina is an independent, non-partisan and non-profit platform initiated by a group of experienced and accomplished economists. The mission of VoxChina is to be the bridge on economic issues between China and the rest of the world by providing informed analysis on the state of the Chinese economy and insights for reform challenges facing China today, and by analyzing the impact of the development in the rest of the world economies on China.

China Subsidy Estimates Depend on an Interest-Rate Yardstick

Before subsidy figures are used to justify new trade measures, the benchmark behind below-market borrowing should be made comparable across countries. The central issue is that the OECD/MAGIC treatment makes China’s BMB estimate LPR-based, and this can make measured borrowing support several times larger than estimates based on more comparable interbank or government-bond benchmarks.

Chinese electric vehicles, solar panels, steel and other manufactured goods are again at the centre of an international trade argument. In Washington and Brussels, the language is familiar: subsidies, overcapacity and market distortion. What is newer is the growing reliance on firm-level subsidy databases to give the debate more quantitative grounding.

Precision, however, depends on measurement. A subsidy figure that looks like a hard fact may rest on a modelling choice that most readers never see. That is why the interest-rate benchmark behind one component of the OECD’s Manufacturing Groups and Industrial Corporations (MAGIC) database deserves more scrutiny. Put more directly, if the MAGIC calculation treats China’s below-market borrowing as LPR-based, the reported China subsidy gap can be magnified by the choice of benchmark itself.

The OECD’s 2026 MAGIC report separates industrial support into three categories: government grants, corporate income-tax concessions and below-market borrowings (BMB). The OECD reports that firms based in China received, on average, three to eight times more government support than OECD-based firms between 2005 and 2024, and that almost 60 per cent of Chinese firms’ global market-share gains can be explained by subsidies. The OECD also states that its results do not prejudge WTO or countervailing-duty assessments. That headline comparison depends heavily on BMB, the component most exposed to the interest-rate yardstick used for China.

The decomposition matters. A large part of the reported China-other-region gap, especially in the 2005-24 average, comes from BMB. This means the debate is not only about how much support China receives, but also about whether the OECD/MAGIC BMB yardstick makes that support look several times larger.

That distinction matters. Grants and tax concessions are closer to observable fiscal support. BMB is a constructed estimate. It compares a firm’s actual borrowing cost with a counterfactual market benchmark, and then applies the gap to a measure of debt. If the benchmark used for China is LPR-based while comparator benchmarks are closer to interbank funding or low-risk sovereign rates, measured Chinese BMB can be mechanically inflated.

The issue is not whether China uses industrial policy. So do the United States, the European Union, Japan, South Korea and many others. The narrower issue is whether countries are being compared with the same yardstick.

A comparability question can be seen in related OECD materials. In its 2024 work on industrial subsidies, the OECD compares banks’ average corporate lending rates with applicable base rates: base rates for OECD banks are described as currency-specific interbank rates, while China’s benchmark lending rates refer to the Loan Prime Rate (LPR) after 2019 and earlier PBOC benchmark lending rates before then. China is therefore assessed against a lending-rate benchmark, while the comparator for OECD banks is closer to wholesale interbank funding.

This distinction is not semantic. The LPR is calculated by the National Interbank Funding Center from bank quotations over open-market operation rates and serves as a pricing reference for bank lending. By contrast, an interbank rate reflects banks’ wholesale funding quotations, while a government-bond yield is closer to a low-risk sovereign reference. Treating these different financial-market layers as interchangeable can turn an LPR-based China estimate into a larger measured BMB number, even before any substantive policy judgement is made.

A simple example shows why this matters. The People’s Bank of China’s 2026 first-quarter monetary policy report states that, in March 2026, the one-year LPR was 3.0 per cent. The same report shows that 48.81 per cent of new renminbi loans that month were priced below the LPR, 9.31 per cent at the LPR and 41.88 per cent above it. If the LPR is used as a stand-alone cutoff, nearly half of new lending would begin on the “below benchmark” side of the calculation. That is too broad a signal for industrial-policy support, because below-LPR lending may reflect borrower quality, collateral, bank competition or ordinary loan-market pricing. This example shows why an LPR-based BMB measure can overstate policy support if below-LPR pricing is treated too broadly as below-market borrowing.

The point is not that below-LPR loans can never involve policy support. Some may. The point is that the LPR is not equivalent to an interbank funding cost or a low-risk sovereign benchmark. Using LPR as China’s base rate before adding firm-level credit-risk and guarantee adjustments therefore risks making ordinary commercial lending appear as LPR-based BMB.

Because the firm-level MAGIC database, loan-level inputs, credit-spread assumptions, debt-currency structure and full calculation details are not public, the OECD’s headline numbers cannot be independently replicated from public materials. But a narrower test is possible: hold the firms, debt proxy and simplified formula constant, and change only the base rate. That test asks a direct question: how much larger does the China BMB estimate become when the calculation is LPR-based rather than government-bond- or SHIBOR-based?

Using Chinese A-share listed firms from 2019 to 2024, mapped broadly to MAGIC-related industries, such a benchmark-sensitivity exercise gives large differences. In the Top-147 large-firm sample, the LPR-based BMB measure used to approximate the OECD/MAGIC treatment is 16.68 times the China government-bond-based estimate and 5.65 times the SHIBOR-based estimate; in 2024 alone, it is 17.89 times and 7.38 times larger, respectively.

The pattern is not confined to one sector. In the broad MAGIC-industry mapping, the LPR benchmark produces higher measured BMB than both alternative base rates across all covered sectors. The gap also widens in more recent years, which is precisely when subsidy and overcapacity claims have become more politically salient. This is therefore not a sector-specific outlier; it is a systematic feature of using an LPR-based benchmark for China.

These numbers do not mean that the OECD’s full estimate would fall by exactly the same multiples. Its full method includes credit-risk spreads, guarantee adjustments, maturities, sector definitions and non-listed firms. But they do show that the base-rate choice alone can magnify the measured China BMB by several-fold, enough to change how the headline subsidy gap should be interpreted.

There is also a scope limitation. MAGIC focuses on three firm-level instruments: grants, corporate tax concessions and BMB. Industrial policy is broader. It can include procurement rules, tariffs, non-tariff barriers, local-content requirements, investment screening, consumer subsidies and regulatory preferences. Some of these tools are used by many economies, but may not appear in the same way in a firm-level subsidy database.

The same caution applies to the OECD’s claim that subsidies explain almost 60 per cent of Chinese firms’ market-share gains. That figure is not directly observed. It is a model-based attribution that uses a subsidy-intensity variable including BMB and an instrumental-variables estimate. If China’s BMB component is magnified by an LPR-based benchmark, then the almost-60-per-cent attribution is also benchmark-sensitive and should not be read as a standalone fact about China’s manufacturing performance.

Industrial-policy debates need measurement. But measurement should discipline policy conclusions, not pre-judge them. Before subsidy estimates play a larger role in trade measures or claims about global imbalance, benchmark rates, sample boundaries and sensitivity tests should be made transparent. A database can inform debate only if its yardsticks are comparable; otherwise an LPR-based BMB approach may magnify China’s measured subsidy gap before the policy debate even begins.

Subscribe now

Yuzhe He

A Critique of the OECD MAGIC Database of Industrial Subsidies and Related Reports: Fan CUI

On June 1, 2026, the Organization for Economic Co-operation and Development (OECD) released a report on its MAGIC (Manufacturing Groups and Industrial Corporations) database of industrial subsidies. One of the key takeaways is that China-based industrial firms appear to receive subsidies on a scale not matched by their competitors in any other economy, …

Read more

4 months ago · 18 likes · 1 comment · Yuzhe HE

Pekingnology

Cui Fan: China Certainly Will Not Apologise for “Overcapacity”

In its 3 August Chaguan column, The Economist took aim at China’s defence against charges of “overcapacity”, arguing that Chinese firms owe much of their competitiveness to a system of state intervention in which “industrial might makes right…

Read more

2 months ago · 34 likes · 6 comments · Yuxuan JIA and Qingqu Yuan

Pekingnology

OECD's Subsidy-Centric Narrative of China’s Emerging Industries Is Increasingly Flawed and Outdated: CF40 research

On June 1, 2026, the OECD released a report titled OECD MAGIC Database of Industrial Subsidies. Its central argument is that Chinese firms receive far larger industrial subsidies than those in other economies, concentrated in emerging strategic industries such as solar, semiconductors and wind power…

Read more

5 months ago · 26 likes · 9 comments · Yuxuan JIA and Zichen Wang

Share

Speaking at the Second Diliman Dialogue shortly after returning from Beijing, where he said he had briefed officials at various levels and consulted relevant departments on bilateral relations, Jing described the relationship as being at an “extremely sensitive and critical juncture.” But rather than advocating a tougher approach across the board, he stressed the importance of distinguishing between different issues and avoiding sweeping judgments about the relationship as a whole.

“Otherwise, we risk falling into a vicious cycle of reciprocal countermeasures and retaliation,” he warned, arguing that such an outcome would benefit only those seeking to undermine bilateral ties.

On maritime disputes, Jing emphasized the need to maintain a safe distance, prevent accidental incidents, and avoid actions that could escalate tensions. He also pointed to the resumption of political dialogue, recent high-level exchanges, growing bilateral trade, recovering tourism, and expanding people-to-people contacts as reasons to sustain engagement.

“The more we engage with each other, the better we understand each other,” he said. “Without engagement, relations will only grow colder and deteriorate further.”

Perhaps most strikingly, Jing explicitly rejected the notion that the Philippines must choose between China and the United States.

“You can be a friend of China and a friend of America at the same time. There’s no contradiction. This is not a zero-sum game. It should be a win-win for everyone,” he said, citing Thailand, Vietnam, Indonesia, and Singapore as countries that maintain relations with both powers.

“The Philippines should put the Philippines first,” he added, arguing that Manila could pursue a “moderate and balanced foreign policy” without becoming an arena or a victim of great-power competition.

The speech was not without sharp criticism. Jing accused what he called “anti-China forces” of provoking tensions and pursuing discriminatory measures against Chinese nationals and businesses. He also placed responsibility on the Philippine side to prevent maritime disputes from defining the entire bilateral relationship.

Nevertheless, his repeated emphasis on dialogue, differentiated handling of disputes, economic cooperation, and the possibility of maintaining good relations with both Beijing and Washington is worth noting, particularly given his recent consultations in the Chinese capital.

In closing, Jing expressed China’s readiness to work with different sectors of Philippine society to rebuild mutual understanding and trust, calling for “more positive stories” and “more good news” in bilateral relations.

Jing, who became China’s ambassador to the Philippines in December 2025, is a veteran of China’s diplomacy, having served as a respected deputy chief of mission at the Chinese Embassy in Washington from 2021 to 2025 and earlier as a deputy director-general in the Foreign Ministry’s Department of North American and Oceanian Affairs.

The Chinese Embassy in the Philippines published the Chinese text of Jing’s speech on its official website on 15 October, followed by an English version on 16 October.

The full text of his address follows.

井泉大使在第二届迪利曼对话会上的主旨演讲

Keynote Address by Chinese Ambassador Jing Quan at the Second Diliman Dialogues

Distinguished guests,

Ladies and gentlemen, and friends,

It is a great pleasure to attend the Second Diliman Dialogue and to engage in in-depth exchanges with the distinguished scholars, media personages and diplomats, and beyond. I look forward to insightful discussions that will stimulate fresh thinking and generate more targeted and actionable policy recommendations.

I have just returned from Beijing, where I spent a considerable amount of time briefing officials at various levels on the latest developments in China-Philippines relations and the work of the Chinese Embassy. I also held discussions with relevant departments on key issues in our bilateral relations, exchanging views and aligning assessments, while listening carefully to their perspectives and guidance.

My overriding impression is that China-Philippines relations have reached an extremely sensitive and critical juncture, one that calls for prudence and careful handling. One of the most important considerations, in my view, is the need to adopt a differentiated approach—addressing issues at different levels and across different areas, and handling each matter on its own merits.

I believe that China-Philippines relations cannot be reduced to a simplistic judgment of being “good” or “bad.” One may observe that, while maritime disputes have cast a shadow over the general atmosphere of bilateral relations, trade and people-to-people exchanges remain strong. Similarly, law enforcement cooperation between the two countries may be proceeding well in certain areas, while the selective enforcement of laws against Chinese nationals raises serious concerns in others. The anti-China forces act in despicable ways, yet the Philippine people at large are kind and honest.

China-Philippines relations are, in essence, a complex and multifaceted relationship. They cannot be approached through sweeping generalizations or one-size-fits-all decisions. Instead, we need to adopt a more nuanced and calibrated approach, assessing each issue in its own context and formulating responses accordingly. Otherwise, we risk falling into a vicious cycle of reciprocal countermeasures and retaliation. Such a downward spiral would only play into the hands of the forces seeking to undermine China-Philippines relations.

In managing China-Philippines relations, it is therefore essential to pursue differentiated strategies for different issues.

Take the maritime issue for example. It is widely regarded internationally as one of the most intractable disputes, one that could remain unresolved for decades, or even a century or more. China has maritime disputes not only with the Philippines, but also with Vietnam, Malaysia, and other countries. Yet these disputes have, by and large, been effectively managed through dialogue and consultation. Our approach is what we call “shelving disputes and pursuing joint development.”

The Philippines, too, has maritime disputes with Vietnam and Malaysia, yet these have not become as prominent or contentious as those with China.

When it comes to managing maritime disputes, the most important thing is for both sides to maintain a safe distance and avoid accidental incidents. In particular, deliberate provocations should be avoided, neither should manufacture incidents, sensationalize or hype up confrontations, or take actions that could escalate tensions.

On dialogue and engagement, China and the Philippines have resumed political dialogue and held five rounds of talks. In July, Chinese Foreign Minister Wang Yi attended the China-ASEAN Foreign Ministers’ Meetings in the Philippines, after which he held in-depth discussions with Philippine Foreign Secretary Lazaro. Visits by delegations from China’s National Health Commission, Ministry of Public Security, and Ministry of National Defense are scheduled.

On the Philippine side, the Secretaries of Agriculture and Trade and Industry, along with a number of local officials, have visited China. In November, the Philippines will host ASEAN Summit and the related summits, while China will host the APEC Economic Leaders’ Meeting in Shenzhen. We should build on this positive momentum, continue to move toward each other, and sustain dialogue and exchanges. The more we engage with each other, the better we understand each other. Without engagement, relations will only grow colder and deteriorate further.

On trade and investment, China has been the Philippines’ largest trading partner and biggest source of imports for ten consecutive years. In the first half of this year, bilateral trade reached US$44.52 billion, up about 20% year on year. Yet it accounted for less than 7% of China-ASEAN trade, leaving plenty of room for growth.

Recently, the Philippines sent a 313-member national delegation to the China-ASEAN Expo as the guest country of honor. Preliminary sales reached US$60 million, a record high for the Philippines at the Expo.

However, the scale and pace of Chinese investment in the Philippines still lag far behind those in other ASEAN countries. It remains well below levels in Vietnam and Indonesia, and even below that in smaller economies like Cambodia and Laos, with only Brunei receiving less.

Growing trade and investment between China and the Philippines benefits both sides. China’s strengths in high technology, AI, modern agriculture, and clean energy could help improve people’s lives and accelerate the Philippines’ economic transformation. However, in recent years, our relations have experienced a downward trend, undermining investor confidence. We need to take concrete steps to improve the business environment and restore the confidence of Chinese enterprises and citizens in investing in the Philippines.

As for people-to-people exchanges, since the beginning of this year, several new direct flight routes have been established between China and the Philippines alongside some resumed ones, including Quanzhou–Cebu, Chongqing–Manila, and Fuzhou–Manila. Looking ahead, we are also considering the launch of direct flights between Kunming and Manila, as well as Haikou and Manila, to further facilitate travel between our two countries.

According to the Philippine Department of Tourism, 303,000 Chinese tourists visited the Philippines in the first eight months of this year, representing a 93 percent increase over the same period last year. During China’s recent extended Mid-Autumn Festival and National Day holiday, the Philippine island destinations of Boracay and Bohol, among others, went viral across Chinese social media platforms and in people’s online social circles.

Meanwhile, the number of Filipinos traveling to China has continued to rise, increasing by 48.9 percent in the first half of this year. So far this year, eight Philippine media delegations have visited China. Many participants have returned with a renewed understanding of the country and have actively shared their experiences and observations with the Filipino public. Posts by a group of Filipino social media influencers who recently visited Yunnan, for instance, have garnered nearly 70 million views.

A recent Philippine public opinion poll also showed that favorable views of China among Filipinos have rebounded to their highest level in four years, with particularly strong support among younger generations compared with older ones.

I believe that as more people travel between our two countries and exchanges become more frequent, falsehoods and misleading narratives that attack and malign China will naturally lose their credibility. There will be less room for sensationalism and discord to gain traction. This, precisely, is the value and enduring appeal of people-to-people exchanges between China and the Philippines.

On law enforcement cooperation, China and the Philippines have worked closely in recent years to combat transnational crimes, including kidnapping, murder, telecom fraud, and human trafficking, helping ensure safer travel and exchanges between our two countries.

In the first half of this year, kidnappings involving Chinese nationals fell by 59% compared with the same period last year. Since 2023, the two sides have jointly apprehended 102 major fugitives and repatriated a total of over 2500 suspects involved in illegal gambling and telecom fraud. This year alone, we have worked together to apprehend 10 major fugitives and repatriate 46 suspects, including six kidnapping suspects. We will continue to step up law enforcement cooperation and intensify efforts to combat crime, so as to ensure that every Chinese tourist and businessperson visiting the Philippines feels safe and secure.

At the same time, it is deeply concerning that some anti-China forces continue to push for selective and discriminatory law enforcement against Chinese companies and nationals, attacking and accusing China by any means. More recently, they have even targeted Chinese-language education, casually using labels like “national security” and “political infiltration.” This sends a chilling effect through the Chinese community in the Philippines. Such actions are malicious and reprehensible. We firmly oppose them and will take resolute measures to protect the safety and legitimate rights and interests of Chinese companies and nationals.

In the Philippines, people often talk about whether to be pro-American or pro-China. But I see it differently. You can be a friend of China and a friend of America at the same time. There’s no contradiction. This is not a zero-sum game. It should be a win-win for everyone.

Just look at Thailand, Vietnam, Indonesia, and Singapore. They all maintain good, stable relations with both China and the United States. So why can’t the Philippines do the same? The Philippines should put the Philippines first. That’s what matters. Every decision should serve the fundamental interests of the Filipino people.

The Philippines can pursue a moderate and balanced foreign policy, maintaining good relations with both China and the United States and reaping substantial benefits from both bilateral relationships. It need not become an arena or a victim of strategic competition between major powers, nor is there any need to choose sides or place itself on the front lines of confrontation and conflict.

China regards the Philippines as a close neighbor across the sea and a friend bound to us by deep historical and cultural ties. The friendly exchanges between our two peoples date back more than a thousand years. China has consistently upheld the principles of fostering friendship, ensuring security, and promoting shared prosperity with neighboring countries, as well as amity, sincerity, mutual benefit, and inclusiveness. We are committed to building a community with a shared future, promoting open regionalism and genuine multilateralism, deepening all-round cooperation with our neighbors, expanding exchanges across various fields, and working together to safeguard regional peace and stability.

In recent years, maritime disputes and frictions have seriously disrupted and damaged China-Philippines relations, causing both sides to miss a number of valuable opportunities for cooperation. The Philippine side should avoid allowing maritime disputes to dominate and define the overall trajectory of bilateral relations. Instead, It should seize the opportunities at hand and take concrete actions to translate them into tangible benefits for its own development. Where differences exist, the two sides should sit down and talk, and find effective ways to manage them. We must not allow a small group of people, backed by forces from outside the region, to steer bilateral relations into a dangerous and uncertain course, or even toward confrontation and conflict.

As close neighbors, China and the Philippines should respect, learn from, support, and help each other. We’re ready to work together with people of insight from all sectors of Philippine society to promote more dialogue and exchanges, continuously deepen mutual understanding and trust, and find ways to amplify the positive narrative of mutually beneficial cooperation between China and the Philippines and the benefits it brings to both peoples. We should do more to add positive momentum to bilateral relations, share more positive stories, and generate more good news. By allowing facts and constructive engagement to prevail over prejudice and hostility, we can find a way out of the current difficulties facing China-Philippines relations.

In closing, I wish the Second Diliman Dialogue every success. I look forward to hearing your valuable insights and policy recommendations on how to bring China-Philippines relations back on track at an early date and advance the building of a closer China-ASEAN community with a shared future.

Thank you.

Share

Subscribe now

Below is our English translation of a commentary by Gordon G. Liu, Dean of the Institute for Global Health and Development at Peking University, a Boya Distinguished Professor at the National School of Development, Peking University, and a member of the Chinese Academy of Medical Sciences.

Liu argues that institutional innovation, rather than technological advances alone, will be crucial to China’s pharmaceutical development. He highlights AI’s potential to expand healthcare access and China’s ability to become globally competitive in generating evidence on medical AI. He also calls for a greater Chinese role in global health governance, particularly through cooperation with developing countries on medicine supply, local production, and public health capacity-building.

Liu’s essay was published on the MIIT website on 28 September 2026.

中国医药工业高质量发展的若干思考：健康价值、原始创新与产业生态

Reflections on the High-Quality Development of China’s Pharmaceutical Industry: Health Value, Original Innovation and the Industry Ecosystem

Editor’s Note

The 15th Five-Year Plan period is a critical time for advancing China’s strategies to become a manufacturing powerhouse and build a Healthy China. To support the implementation of the 15th Five-Year Plan for Pharmaceutical Industry Development and accelerate the development of biomedicine into an emerging pillar industry, experts, entrepreneurs, and other professionals have drawn on their expertise and practical experience to contribute articles interpreting the plan. Their contributions are intended to build broad consensus and foster a supportive environment for the pharmaceutical industry as it enters the new planning period.

Abstract

The release of the 15th Five-Year Plan for Pharmaceutical Industry Development in September 2026 marked a critical turning point for China’s pharmaceutical industry as it shifts its focus from expanding in scale to improving in quality.

Drawing on insights from health economics and global health governance, this article examines key issues raised by the plan, emphasising three central points. First, the pharmaceutical industry’s ultimate output is not medicines, but health. Second, the true measure of the industry’s success is whether the wealth it generates translates into measurable improvements in health. Third, China faces the challenge of moving from being a user of global pharmaceutical innovations to becoming a major contributor to the global pharmaceutical innovation system.

The article examines these issues from five perspectives: making health outcomes the basis for evaluating industry performance; aligning innovation with clinical value; using digital and intelligent technologies to expand the availability of high-quality healthcare; developing diversified, value-based payment systems; and balancing development and security through greater openness to the world. Together, these perspectives aim to inform the long-term development of China’s pharmaceutical industry beyond the immediate priorities of short-term policy cycles.

I. Introduction

Few industries combine strategic importance with such a direct impact on people’s well-being as the pharmaceutical industry. It connects basic science and advanced manufacturing with healthcare services and public health. It is both a source of national competitiveness and a factor in the sustainability of the social security system.

In September 2026, China released the 15th Five-Year Plan for Pharmaceutical Industry Development, setting out plans to accelerate the development of biomedicine into an emerging pillar industry of the national economy. For the first time, the plan sets quantifiable targets for 2030, including a 25 per cent share of the global total of first-in-class drugs and average annual growth of 20 per cent in the innovative drug industry.

These clearer policy objectives also bring deeper theoretical questions into focus. Around the world, pharmaceutical industries face similar structural challenges: The cost of innovation continues to rise while research and development efficiency declines, a phenomenon known as “Eroom’s law”; population ageing and changing disease patterns are driving up health expenditure and putting pressure on the sustainability of payment systems; geopolitical shifts are reshaping global supply chains, creating tensions between efficiency and security; general-purpose technologies such as artificial intelligence are being adopted at an accelerating pace, even as evidence of their health benefits remains limited.

China faces these challenges alongside difficulties of its own: the pressures of transforming a vast market, the challenge of moving from generic drugs to original innovation, and geographical disparities in healthcare provision and the ability to pay.

This article considers five fundamental questions that extend beyond short-term policy cycles: How should industry performance be measured? How should innovation priorities be determined? How can the benefits of technological advances be realised? How can payment systems provide the necessary support? And what position should the industry occupy in the global division of labour?

II. The Performance Question: What Should the Pharmaceutical Industry’s Ultimate Output Be?

The pharmaceutical industry has traditionally been assessed using indicators such as output value, investment intensity, the number of approvals, and speed to market. Yet these measure intermediate inputs and outputs rather than the final outcome.

This is more than a semantic distinction. It goes to the heart of policy design. When performance is judged by intermediate indicators, companies naturally focus on bringing products to market faster, expanding their portfolios, and increasing their scale. Only when evaluation focuses on the final outcome will resources be directed towards improving health itself.

From a health economics perspective, the pharmaceutical industry’s ultimate output should be improvements in population health: longer lives, a lower burden of disease, higher labour productivity, and avoidance of catastrophic health expenditure for households.

Health is a form of human capital and a fundamental driver of development. Better health can extend people’s productive working lives, raise productivity, and strengthen the resilience of households and society by reducing the burden of disease.

The continued rise in life expectancy in China has shown how advances in pharmaceutical and medical technologies can translate into better health outcomes. Looking ahead, health policy must move beyond treating disease to encompass the entire continuum of prevention, diagnosis and treatment, rehabilitation, and health management, reflecting the international shift towards integrated healthcare systems.

Making health outcomes the ultimate measure of performance also calls for a different understanding of rising health expenditure. As economies develop and incomes rise, demand for healthcare naturally increases. Rather than simply representing higher costs, this reflects society’s growing recognition of the importance of investing in human capital.

The policy objective should therefore be to improve the marginal productivity of healthcare resources, not to suppress legitimate demand for healthcare. This means reducing low-value and unnecessary services while directing resources towards prevention, urgent clinical needs, and interventions that significantly improve health outcomes.

This is the central principle behind the global shift towards value-based healthcare. It should also serve as a lasting benchmark for any pharmaceutical industry policy, regardless of the planning cycle.

A basic conclusion follows: the maturity of a country’s pharmaceutical industry evaluation system reflects how far its pharmaceutical policy has evolved from an industry-centred to a health-centred approach.

III. The Innovation Question: From Competition in Numbers to Alignment with Value

Global pharmaceutical innovation faces a widely recognised paradox: research and development spending continues to rise, yet the number of new drugs produced has remained relatively stagnant.

China’s position presents a particular challenge. It now ranks second globally in the number of new drugs under development, but its ability to lead in original innovation still falls short of the highest international standards.

From January to September 2026, the total value of overseas licensing deals for innovative Chinese drugs exceeded US$120 billion, up 36 per cent year on year. This reflects growing international recognition of the quality of Chinese pharmaceutical research and development. But larger licensing deals do not necessarily signal a corresponding improvement in the capacity for original innovation. Overseas licensing is a means, not an end. Using the number and value of licensing deals as proxies for innovation performance risks obscuring structural weaknesses in upstream capabilities, including target discovery, translational medicine, and global clinical operations.

The central issue is how to align innovation priorities with value. If companies are incentivised primarily for how quickly they can develop licensable assets, their most rational strategy is to concentrate on established targets and mechanisms, making only incremental improvements. Only when innovation is assessed by its clinical value – whether it addresses unmet medical needs, improves quality of life, or reduces the lifetime burden of disease – will resources flow towards areas that offer substantial social benefits but lack sufficient commercial incentives. These include major diseases, rare diseases, paediatric medicines, diseases affecting older people, and drug-resistant infections.

More fundamentally, the capacity for original innovation depends on the quality of the institutions that support it. Pharmaceutical innovation requires substantial investment, involves long development cycles, and carries considerable risk. Sustaining it requires stable and transparent regulatory expectations, patient capital willing to tolerate failure over the long term, and effective coordination across the entire process from basic research and translation to clinical development, regulatory review, and commercialisation.

International experience has repeatedly shown that the competitiveness of an innovation ecosystem depends fundamentally on the quality of its institutions rather than on the scale of resources invested.

For China, the journey from catching up to keeping pace and ultimately taking the lead will depend not simply on overcoming technological barriers, but on institutional innovation.

IV. The Technology Question: How Can Artificial Intelligence Strengthen Health System Capacity?

Artificial intelligence is reshaping drug discovery, clinical trials, manufacturing, and healthcare services. Its significance for the pharmaceutical industry has gone well beyond conventional digital transformation.

Two misconceptions deserve particular attention. The first is to view AI simply as a tool for making research and development more efficient. The second is to focus on whether a technology is technically feasible rather than whether it delivers meaningful value.

For China, AI’s greatest potential to transform the pharmaceutical and healthcare sectors may lie not in the laboratory, but in service delivery.

High-quality healthcare resources remain scarce and unevenly distributed, while patients continue to flock to major hospitals in large cities. The root problem is that advanced medical knowledge and clinical expertise are difficult to replicate at low cost.

With doctors leading the process and appropriate regulatory oversight in place, AI could help primary healthcare institutions and early-career doctors by integrating medical knowledge, clinical guidelines, and practical experience. This could reduce the dependence of high-quality healthcare on a small number of specialists, enabling the development of digital capabilities that can be replicated and scaled up.

If realised, this shift could deliver public health benefits far exceeding the cost savings and efficiency gains AI brings to research and development, thereby contributing directly to the higher-level goal of health equity.

This leads to an important principle for evaluating new technologies: every new pharmaceutical or medical technology should undergo a comprehensive health technology assessment (HTA). Put simply, HTA asks not only whether a technology works clinically, but also whether it offers value for money, improves equity, and delivers real benefits to patients.

The value of medical robots, for example, lies not just in their precision, but also in their ability to reduce harm to patients, shorten the time doctors need to develop expertise, and extend access to high-quality healthcare through remote collaboration. Similarly, the value of digital therapeutics lies not in the algorithms themselves, but in whether their clinical endpoints can withstand scrutiny through both randomised controlled trials and real-world evidence.

If technology assessments focus solely on engineering specifications, policy risks being guided by demonstration applications rather than evidence.

China has comparative advantages in its large patient population, extensive clinical resources, and diverse settings for applying new technologies. By developing real-world data infrastructure alongside data governance systems, China is well positioned to become internationally competitive in generating evidence on AI applications in pharmaceuticals and healthcare. It could also contribute assessment methods and empirical evidence that other countries can draw on.

The benefits of general-purpose technologies can be shared across open systems, while the ability to generate evidence is becoming a new frontier of competition.

V. The Institutional Question: How Can Payment Systems Realise the Value of Innovation?

The pharmaceutical industry operates within a distinctive economic cycle: returns on today’s research and development depend on tomorrow’s payment arrangements. Payment systems are therefore not only part of health policy but also a component of innovation policy. This is a key distinction between pharmaceutical economics and industrial economics in general.

This cycle now faces two pressures. On the payment side, population ageing is driving up health expenditure. Basic medical insurance funds, designed primarily to cover essential healthcare needs, are struggling to cover all innovative technologies and the full range of healthcare needs. On the innovation side, sustained investment depends on adequate returns. Without reliable mechanisms for assessing and rewarding value, long-term, high-risk investment in research and development is difficult to sustain.

Simply driving down prices may ease pressure on insurance funds in the short term, but at the cost of weakening incentives to innovate. Allowing high prices to go unchecked, meanwhile, puts excessive strain on payment systems and worsens health inequalities.

Experience in advanced economies suggests that the solution lies in allocating resources based on clinical value and health outcomes. Pharmacoeconomics and HTA can be used to assess value, allowing high-value innovations to earn reasonable returns while low-value, undifferentiated products are gradually phased out through market competition and payment policies.

Equally important is bridging the “last mile”: regulatory approval or inclusion on a reimbursement list does not necessarily mean that patients can actually obtain a medicine. Barriers in prescribing, hospital drug availability, distribution and payment processes can prevent reimbursement commitments from translating into actual patient access. Better coordination between medical insurance, healthcare, and pharmaceutical policies is fundamentally about repairing the mechanisms through which the value of innovation reaches patients.

Structurally, a single payment system cannot achieve both sustainability and accessibility. Basic medical insurance must remain focused on covering essential healthcare needs, with commercial health insurance, medical assistance, and charitable aid providing additional layers of support. This diversified approach is particularly important for high-value innovative drugs, medicines for rare diseases, and advanced therapies.

Commercial insurance has limited value if it merely reimburses medical expenses. If it plays a more active role through product design, health management, and risk sharing, it can become a vital link between innovation, healthcare services, and patient needs.

A well-functioning hybrid payment system combining basic medical insurance with commercial health insurance is not only an institutional challenge for China, but could also serve as a model for middle-income countries around the world. Innovation in payment systems is itself an international public good that can benefit the pharmaceutical innovation ecosystem.

VI. The Global Question: Position and Responsibility in an Open System

Pharmaceutical innovation is highly globalised, relying heavily on the cross-border movement of knowledge, talent, capital, clinical evidence, and manufacturing capabilities. However, geopolitical shifts have presented countries with a common dilemma: efficiency or security?

Global experience since the pandemic has shown that disruptions in the supply of critical raw materials and excipients, advanced pharmaceutical manufacturing equipment, and analytical and testing instruments can bring the entire innovation chain to a standstill. As a result, control over critical technologies and the resilience of industrial and supply chains have become a matter of security, not just industrial policy.

Yet seemingly obvious solutions can be misleading. Closing off from the outside world and pursuing self-sufficiency may be an instinctive response to this dilemma, but it is not necessarily the right one. Essentially, industrial security does not mean replicating every stage of the supply chain domestically. Rather, it means occupying an indispensable position in global networks and maintaining reliable supply capabilities. Indeed, resilience in individual parts of the system depends on deep integration into global networks of cooperation.

China’s role in the global pharmaceutical industry is undergoing a historic transformation. Early signs are emerging of its shift from a user of global innovations to a major contributor. The next stage could focus on three strategic priorities.

First, China could bring together global innovation resources through internationally coordinated research and development, international multicentre clinical trials, and registration and certification in overseas markets.

Second, it could contribute to international rule-making through original pharmaceutical products and standardised evidence.

Third, China could take a more active role in global health governance, turning the growing international presence of its pharmaceutical industry into a means of providing global public goods. This would involve cooperation with developing countries on medicine supply, public health capacity, local production, and workforce training. Affordable and sustainable pharmaceutical solutions have strategic value beyond their commercial benefits in addressing shared challenges such as major infectious diseases, population ageing, and climate-related health risks.

This changing role also brings new responsibilities. A major pharmaceutical nation has an obligation to promote global health equity commensurate with its capabilities. This is both a moral imperative and a matter of long-term interest. A country’s credibility in global health governance is among the most valuable intangible assets of its pharmaceutical industry, and one of the hardest to replicate.

VII. Conclusion

What sets the pharmaceutical industry apart is that its ultimate product should be human life and health. Therefore, its performance metrics, approach to innovation, technology assessment, payment systems, and role in the global economy cannot simply follow the same frameworks as other manufacturing industries.

Output value can measure the size of an industry, but only health outcomes can measure the value of the pharmaceutical industry. Licensing deals can indicate the pace of internationalisation, but only the capacity for original innovation can determine its standing. Inclusion on reimbursement lists can indicate the reach of policy, but only patients’ actual access to medicines can measure its success.

China’s pharmaceutical industry is undergoing a historic transformation from being large to becoming strong. Its vast market, extensive clinical resources, comprehensive industrial system, and steadily growing capacity for innovation provide the foundation for such transition.

More fundamentally, this transformation depends on institutional capacity: the ability to connect research and development, evaluation, production, application, and payment, and to align science, industry, healthcare, and payment systems around a common standard.

This principle should transcend both the five-year planning cycle and national borders. After all, the achievements of an industry whose ultimate output is health cannot be measured solely by a country’s GDP. If China’s pharmaceutical industry remains guided by this principle, its growth will not only benefit China but also contribute to the well-being of humanity as a whole.

Subscribe now

Pekingnology

Gordon G. Liu: What AI Can—and Cannot—Fix in Chinese Healthcare

In a two-part interview with Sohu Health, Gordon G. Liu, Boya Distinguished Professor and Dean of the Institute for Global Health and Development at Peking University, explains how AI could strengthen primary care, narrow diagnostic gaps, broaden access to medical expertise, and advance interdisciplinary research. He also traces persistent medical resea…

Read more

3 months ago · 11 likes · 1 comment · Yixin Guo and Yuxuan JIA

For readers unfamiliar with the system, the exam is roughly comparable to the American bar exam, although it covers several legal professions. Unlike in most of the United States, candidates do not necessarily need to have attended law school: under certain eligibility rules, graduates of other disciplines can also qualify. The exam is notoriously difficult, and a large commercial test-preparation industry has grown around it. Competition among instructors is fierce, and some of the more successful ones have become minor celebrities in their own right. Cai is one of them. He has 1.59 million followers on Bilibili, China’s answer to YouTube.

But that is not really why I follow him.

Cai travels across China handling criminal cases. On his personal WeChat public account — something like a blog within the WeChat ecosystem — and his video channels, he regularly offers glimpses of his work, his family, and his life as an amateur marathon runner. I find him remarkably down-to-earth. There is something refreshingly unpretentious about the way he documents his everyday experiences.

He also regularly publishes messages from his followers. They come from people across the country, in all sorts of occupations and circumstances, who simply want to tell someone what has happened to them. Through Cai, their accounts reach an audience they might otherwise never have found.

And he is unusually candid about the limits of what he can make public.

On October 8, he posted a 39-second video on Bilibili that so far has attracted some 434,000 views. Addressing those who send him their personal experiences, he said that people working in the prison system, judicial administration, or administrative reconsideration should no longer submit their accounts because he could not read them aloud. “I won’t go into the reasons,” he added. Other submissions, he assured his followers, were still welcome, “as long as they don’t go too far, at least for now.” He thanked everyone for their trust.

I use ChatGPT to assist my writing and translation, and here ChatGPT wrote, “It was an extraordinarily matter-of-fact acknowledgment of censorship.” ChatGPT apparently doesn’t know China well - this is not extraordinary. We have long grown accustomed to it. That Cai could state these restrictions so openly, and that the video itself was apparently allowed to circulate, reflects a tacit understanding of its own: one may acknowledge the boundaries, so long as one does not ask too many questions.

The viewers seemed to appreciate the irony. Many danmu — time-synchronized viewer comments that scroll across the screen as a video plays — jokingly compared his approach to renzui renfa (认罪认罚), China’s system of leniency for those who plead guilty and accept punishment, loosely analogous to plea bargaining. Rather than challenge the restrictions, Cai had effectively accepted them and moved on. He did not need to explain himself. Everyone understood.

Anyway, what I really wanted to do here was introduce readers to four messages that Cai published on October 2 and October 8.

Some are written in rather unpolished language, with awkward phrasing and thoughts that the authors themselves sometimes struggle to articulate. They are not carefully constructed essays, nor do they seem intended for a wider audience. They are simply fragments of ordinary lives, told by the people living them.

I am not entirely sure what I want to say about these accounts. Perhaps I simply worry that, amid all the coverage - including by this substack - of U.S.–China relations, Taiwan, geopolitics, and AI, we risk losing sight of how ordinary people in China actually live, especially those struggling to get by. And I struggle to think of many news outlets, Chinese or international, that still make much room for those stories.

I don’t have a sweeping argument to make, or a particular conclusion I want readers to draw. These are four individuals describing their own experiences, and I cannot independently verify the details. I just thought they might be worth sharing.

1.

Shared by Cai Yaqi on October 8, 2026

Dear Teacher Cai,

I want to tell you about the difficulties and despair of living with a criminal record.

I was born in 2001. I am 24 years old and come from Shandong Province.

In 2022, when I was young and ignorant, I rented out my personal bank card without realizing that I was getting involved in illegal and criminal activities. When the police contacted me, I went to see them. I was terrified. I thought my life was over.

But the police believed me, and I signed a document acknowledging the termination of the investigation. I felt that I had been given my life back. Because in a place like Shandong, having a criminal record is practically the same as being dead.

But just a few days after I started feeling happy again, a vehicle full of police officers arrived at my home a little after six in the morning and took me away. In the end, I was sentenced to one year in prison. To this day, I still don’t understand whether the procedures they followed were legal.

I was released in December 2024.

After returning home, I delivered food, worked in factories, and delivered parcels. Finally, after enormous effort, I managed to find a reasonably well-paying job that even came with the five social insurance schemes and housing provident fund contributions. Once again, I thought my good days were finally coming.

But a few months later, I was injured in a traffic accident while working.

During my hospitalization, the driver responsible for the accident never called me, not even once, or offered a single word of concern. When I tried to obtain the traffic police’s accident liability report, the officers were dismissive and kept telling the two sides to communicate with each other. I don’t understand whether this is right.

Meanwhile, my employer had arranged things through layers of companies. The company paying my social insurance was not the same company that had signed my employment contract. As a result, I have been unable to obtain official recognition of my injury as work-related. The lawsuit is still ongoing, with no end in sight.

So I went back to delivering food.

But I don’t want to spend the rest of my life delivering takeout. I want to work hard, make money, and give myself and my family a better life.

I applied for another job. Everything went extremely well. I completed every stage of the recruitment process. Just as I thought I was about to start a new job and a new life, fate played another cruel joke on me.

I failed the criminal-background check.

I was depressed and frustrated, but I didn’t give up. I kept looking for work.

Then it happened again. Another company, another recruitment process, every stage completed successfully — only to be rejected at the background-check stage.

I don’t understand. Does having a criminal record mean that people like us must live at the very bottom of society? That we can only work on assembly lines or deliver food, struggling just to survive?

Must society push us into such desperation that we end up committing crimes again?

I feel completely hopeless.

Recently, my girlfriend told me that her family was strongly, strongly opposed to our relationship. They said my criminal record would affect our future children.

But my girlfriend and I love each other very much. She has continued to stand by me, quietly, without her family’s knowledge.

Perhaps it is because of her that I haven’t thought about ending my life or returning to crime.

I just don’t understand why society bears so much hostility toward people with criminal records.

It is enough to make you despair! I cannot see a way forward!

Why must every path toward an honest, hardworking life be cut off for us?

2.

Shared by Cai Yaqi on October 8, 2026

Dear Teacher Cai,

After reading so many submissions from people in different occupations, I felt compelled to write to you about the everyday lives of my parents, who are farmers. I hope you might have a chance to share their story.

My parents live in Yulin, Shaanxi Province.

I came home during the National Day holiday this year to help with the autumn harvest.

Every morning, they go up into the hills to work at seven. At noon, they eat something simple in the fields. They don’t come home until after dark.

And even after dinner, they still cannot rest. They have to sort and prepare the millet, black beans, and corn they have harvested.

People talk about how exhausting their jobs are, about “996” and “007” work schedules. I experienced that kind of work myself this year. But compared with what my parents do, it all seems so easy.

I often tell them, “You are practically working yourselves to death every day. Can’t you plant a little less? I have a job now, and our family doesn’t have any major expenses anymore.”

Every time, they say, “We’ll stop farming when we can no longer do it.”

But the following year, they plant everything again.

They are 65 years old, farming 40 mu of hilly land — about 6.6 acres — while also raising four cows.

How cruel is it that people of their age should have to work so hard?

I know why they do it. They want me to have a better life.

But my life is already good.

I know that, with my abilities, I will probably never be able to give my parents the kind of life that some of my colleagues’ parents enjoy.

But surely I could at least help them eat a little better and suffer a little less.

Yet they refuse.

The pain this causes me is something I cannot really put into words. I can only bear it silently.

My parents are actually the youngest farmers in their village.

Our neighbour, an 82-year-old grandfather, is still farming more than ten mu of land.

Another elderly man who lives halfway up the hillside has a back so badly bent that he can barely straighten it.

Every time I pass his home and see his back bent almost 90 degrees, I finally understand what people really mean when they say someone has been “bent over by the weight of life.”

Life is so bitter.

So bitter that there are no words to describe it.

3.

Shared by Cai Yaqi on October 2, 2026

Dear Teacher Cai,

Happy National Day.

I am a food-delivery rider. I was born in 2002, and I am 24 years old. I have been working for six years.

I don’t know why I am alive.

I don’t have any particular goals anymore. I don’t even have a vague idea of what I want from life.

When I first started working, I was full of hope. I thought that by working with my own hands, I could build a fulfilling life. Perhaps I would never become rich, but at least I could find some inner peace and contentment.

I don’t have much education. I graduated from a vocational high school, so the only jobs I could find were ones where I could be replaced at any moment.

I kept changing jobs. Perhaps some of that was my own fault, but most of the time, I was driven away by the people and circumstances around me.

I won’t go into all the details.

Eventually, I started delivering food.

At first, I thought it was a wonderful job. There were no complicated workplace relationships and nobody forcing me to work overtime.

I did it for three years.

But the longer I delivered food, the more lost I became.

As time went on, life became increasingly painful. Not physically, but mentally.

I was constantly under enormous pressure, passively absorbing hostility from every direction.

I kept trying to numb myself. I invented one material fantasy after another to give myself something to look forward to — buying a motorcycle and riding to Tibet, for example.

And I did experience some excitement from these material pleasures.

But eventually, I realized that I was becoming less and less happy.

There are things I want to say, but I cannot find the right words. I cannot describe what is going on inside me.

Later, even material pleasures stopped working.

I began asking myself, over and over again: What exactly do I want? How can I finally find peace inside?

Through this endless questioning, pressure, and self-torture, my mental state eventually collapsed.

For several days in a row, I couldn’t sleep. I barely ate or drank anything. I just kept smoking.

A childhood friend of mine had also struggled with his mental health, but he eventually managed to get through it.

His advice was that I should stop thinking so much about what I wanted to do. Instead, I should learn to say no to things I didn’t want to do.

Over the past six months, my mental state has improved somewhat.

But I still cannot think too much.

Whenever I start thinking, that other version of myself inside my head begins putting me on trial again.

I suddenly felt like writing to you today.

Not for any particular reason.

It was raining today. I was riding through an area near a school when some pedestrians suddenly emerged from behind an illegally parked car.

I swerved to avoid them and fell off my bike.

They didn’t even look at me. They simply walked away.

At that moment, I felt hatred.

I hated them, and I hated myself.

And then the voice inside my head began putting me on trial again.

Why have I ended up living like this?

4.

Shared by Cai Yaqi on October 2, 2026

Dear Teacher Cai,

I was born in the 1990s. I come from Xinjiang and now live in Chengdu.

I used to be a journalist.

I spent ten years working for traditional provincial-level media organizations in Urumqi and Chengdu.

Eventually, I resigned and started a business with some friends. We opened a shop, but then the pandemic arrived.

We lost several hundred thousand yuan.

Now I am drowning in debt.

Every day, countless debt-collection calls and text messages leave me physically and mentally exhausted.

I have thought about giving up on myself and ending my life. But whenever I think of my parents, I abandon that thought.

I have noticed that young people from ordinary families, whatever jobs they do, seem to experience some degree of confusion and anxiety. Some even become seriously depressed.

After my business failed, I came to understand more and more that money is what gives a person the greatest sense of security.

It is also the passport to freedom.

The amount of wealth you possess determines how much freedom you have.

I actually love photography.

Over the years, I have taken many photographs with my mobile phone.

My work won third prize in a photography competition associated with the 31st Chengdu World University Games. I also won third prize in a photography competition organized by Chengdu High-tech Zone to celebrate the 75th anniversary of the founding of the People’s Republic of China. Some of my photographs were selected for the 2023 “City Artists: Me and Tianfu New Area” photography exhibition.

I have always dreamed of being like Luo Hong, the chairman of the Chinese bakery chain Holiland — becoming financially independent and travelling around the world taking photographs.

I long for that kind of life.

But I also know that it is almost impossible for me to achieve it in this lifetime.

I am now a food-delivery rider.

Every day, I travel through the streets and alleys of the city.

And with every delivery order I complete, my dreams, my passions, and everything I once wanted to pursue are gradually worn away, confined, and swallowed up.

The truth is that ordinary people cannot afford many mistakes. They have very little room to try and fail.

One wrong step can bring a lifetime of regret.

Most ordinary people simply follow the same life script as everyone else: go to school, find a job, buy a car, buy a home, get married, and have children.

It may be a boring life, but at least it is a relatively safe one.

Our dreams may be beautiful and full of promise.

But in the end, they are no match for the sharp edges of reality.

The East is Read

4:30 A.M. at a day labour market in China, a drink to start a day’s toil

China’s informal labour markets are fleeting theatres of toil—unmarked spaces that spring to life before dawn and vanish by the break of day, playing out daily in nearly every major city, from Shanghai to Hefei. In Guiyang, the capital of the mountainous, landlocked province of Guizhou, one such stage comes alive at 4:30 a.m. A stretch of bare concrete …

Read more

a year ago · 35 likes · 5 comments · Yuxuan JIA and Zhong Huiqing

Share

The proposal has sparked considerable controversy in American foreign policy circles, challenging longstanding assumptions about U.S. military primacy in the Western Pacific, the credibility of its alliances, and the strategic importance of Taiwan. Kavanagh’s argument is not simply that the United States should do less in Asia, but that its current military posture may no longer be the most effective or sustainable way to protect American interests.

During a visit to China in September, Kavanagh participated in a non-governmental, informal discussion in Shanghai organized by Jia Min, a research fellow at the Shanghai Development Research Foundation, attended by Xiang Lanxin, Professor Emeritus at the Graduate Institute of International and Development Studies (IHEID), Geneva. The event was open to the public.

In an interview subsequently published by Guancha.cn on October 2, Jia reflected on their exchange and offered a perspective that complicates some of the assumptions surrounding Kavanagh’s proposal.

Jia makes two particularly interesting points. First, Kavanagh’s advocacy of strategic retrenchment should not be mistaken for sympathy towards China or “CCP.” Her starting point, he argues, is a reassessment of American national interests, not a desire to accommodate Beijing. Second, an American military withdrawal from the first island chain would not necessarily leave China better off. A diminished U.S. security presence could encourage Japan to pursue nuclear weapons, unsettle the regional balance of power, and create new sources of instability that Beijing would find difficult to manage.

Jia also placed Kavanagh’s proposal in the context of her earlier research on Truth Decay, published by RAND, where Kavanagh worked for a decade. That report explored the erosion of shared facts and public trust in American society, and considers how generational experiences, domestic political pressures, and competing conceptions of national interest shape the debate over America’s role in Asia.

贾敏：建议美国撤出“第一岛链”，美国人骂她“卖国”，但中国也需警惕

Jia Min: Jennifer Kavanagh Calls for US Withdrawal from the First Island Chain, Drawing Accusations of Betrayal in America — but China Should Be Cautious Too

Jennifer Kavanagh, a senior fellow and director of military analysis at the US think tank Defense Priorities, recently published a report entitled A Second Island Chain Strategy for Asia. Her bold proposal — a comprehensive shift in Washington’s Asia strategy away from the vulnerable “first island chain” towards the “second island chain” — has sparked fierce controversy in the United States and beyond. Excerpts published by Guancha.cn also prompted heated debate among Chinese readers.

As it happened, Kavanagh visited China in mid-September to attend the Beijing Xiangshan Forum before travelling to Shanghai. During her visit, Professor Xiang Lanxin, a scholar of international relations, and Dr Jia Min, a special research fellow at the Shanghai Development Research Foundation, invited her to a small gathering to discuss US strategic retrenchment and its relationship to American national interests. The conversation also touched on Washington’s broader global strategy and the future of China-US relations.

Speaking to Guancha.cn afterwards, Jia recalled how Kavanagh had struggled with the backlash against her report, but had also been touched by the warmth and encouragement she received from the Chinese public. The discussion, however, was not without its disagreements.

How should Kavanagh’s report be understood? Jia cautioned that her report does not make her “pro-CPC” or “pro-China”: she is arguing for America’s interests, and Taiwan is not the report’s central focus. Nor would a US military withdrawal from the Western Pacific necessarily work to China’s advantage. It could have consequences that few have yet considered.

Strategic retrenchment is a complex undertaking. As American power declines in relative terms, China faces a serious question: how can it protect its core interests while preventing a US withdrawal from creating a regional power vacuum that could intensify rivalries and lead to new conflicts?

Controversy in Strategic Circles, Goodwill from the Chinese Public

Guancha.cn: Dr Jia, I’d like to start with your recent meeting with Jennifer Kavanagh of Defense Priorities. Her report, A Second Island Chain Strategy for Asia, published on 10 September, has attracted considerable attention in the United States, Europe, the Chinese mainland, and Taiwan.

Reading someone’s work is one thing; meeting them in person is quite another. During your conversation, did anything she said, or perhaps something more subtle, give you a better sense of how American scholars and think tanks are thinking about US strategy, Asia, and China?

Jia Min: I’d read some of Kavanagh’s work before, though it had been quite a while. We got in touch recently through Professor Xiang Lanxin, a well-known scholar of international affairs and intellectual history.

In early September, Professor Xiang told me that a prominent American scholar was coming to China for a forum and several other engagements, and that she might have time to visit Shanghai. He asked if I’d be interested in hosting her and arranging some activities. I thought of the Buzu Thought Salon, which I organise, and decided to invite her to speak there.

Perhaps I should explain the name. The salon is a public forum for discussions of ideas and culture. Since we often talk about China-US relations, we chose a name that plays on the Chinese expression meizhong buzu, meaning “something to be desired”, while also suggesting “something lacking between America and China”. Kavanagh’s visit happened to coincide with our thirtieth session.

Her report hadn’t yet been published when I first received the invitation to host her. At the time, I was simply delighted at the prospect of welcoming such a distinguished scholar. When the report came out, I read it closely and found that its arguments were strikingly different from the prevailing views in American think tanks and academia. They were bold, even provocative, but backed by solid evidence and rigorous analysis, reflecting her intellectual rigour and clear-headed approach to research. That made me all the more eager to meet her.

Kavanagh arrived in Shanghai on the evening of 19 September, looking rather tired. I later learnt that she’d had a packed schedule in Beijing, attending the Xiangshan Forum and meeting many Chinese experts and scholars. So we scaled back the sightseeing we’d planned and focused on the salon at a bookshop the following evening. As an informal gathering, it gave her a chance to relax and speak freely.

The discussion was framed around the idea of American “strategic retrenchment”: how it relates to US national interests, what it means for America’s broader global strategy, and how China and the United States might get along in this context.

The topic generated considerable interest among the participants. Two examples come to mind.

One friend of mine runs a well-known winery. She’s very interested in China-US relations, and as soon as she saw the event poster, she asked me to save her a seat. She rushed back to Shanghai that afternoon, asked Kavanagh all sorts of questions about the bilateral relationship, and gave her a beautifully packaged gift set of gin.

Another friend, a veteran technology investor, was attending a course in Hangzhou. She caught a high-speed train back to Shanghai specifically for the event. When she noticed that Kavanagh’s throat was bothering her, she ordered some lozenges online and handed them to her during the discussion.

These small gestures touched Kavanagh. She hadn’t expected such warmth from Chinese participants, or that they would be so eager to engage with her. The experience gave her a more personal appreciation of the diversity and vitality of Chinese society.

It also says something about the Chinese public. People from all walks of life want to talk about China-US relations, world politics, and what’s happening inside America. They’re eager to communicate and make sense of where things are heading.

I won’t go through the report itself here — it’s available online for anyone who wants to read it. But Kavanagh told us quite candidly that she’d been under tremendous pressure since it came out. It had caused a huge stir in American strategic circles. She kept coming back to the word “controversy”.

Some of the criticism in overseas media has gone beyond the report itself to personal attacks on Kavanagh. Some of those comments have crossed the line and are simply unfair to her. During our discussion, we tried to offer her understanding, support, and goodwill, and encouraged her to stand by the policy conclusions she had reached through her research.

She certainly hasn’t backed away from them. She remains convinced that her proposals serve American national interests and could, to some extent, help put China-US relations on a more stable footing.

I think the discussion achieved what we’d hoped. Kavanagh was dealing with criticism and controversy, but she also found encouragement. We hope she’ll come back to China often, learn more about the country’s development, and get a better sense of how Chinese people see China-US relations, developments in American society, and the thinking behind US strategy. That kind of engagement can build trust and give her fresh ideas for her research.

We didn’t agree on everything that evening, of course, but that’s fine. The important thing is that people can talk openly and have an in-depth exchange of views.

I imagine it was quite different from the discussions she’s used to in academic and policy circles. Ours is a public forum that brings together people from business, technology, academia, and other fields. Participants tend to draw on their own experiences and speak from the perspective of ordinary people.

We wanted to show her that policy research isn’t just about advising those in power. It can also reach a wider audience and benefit from hearing what people outside government and academia have to say. That feedback can be valuable for future research and writing.

My friend who runs the winery, for instance, put it very bluntly: “I’m in the consumer business, so of course I want China-US relations to improve. Better relations mean more spending, more opportunities to grow the business, and better prospects for the future.”

It was a practical, heartfelt way of putting it.

The Report Does Not Make Kavanagh “Pro-China”

Guancha.cn: Did Kavanagh say anything during the discussion that went beyond the report itself and particularly struck you?

The report has attracted attention in China well beyond academic circles. It came out at a time of renewed high-level engagement between China and the United States, with Taiwan high on the agenda. That timing helped fuel the public debate.

Jia Min: The opening of the report makes her priorities very clear. She identifies two core US interests in Asia: defending the US homeland and safeguarding access to key economic markets and trade routes.

So her starting point is unmistakably American national interests. This isn’t a report written specifically to resolve the Taiwan question. It’s an attempt to address problems facing the United States itself.

You can see that in the questions she asks, the solutions she proposes, and the way she builds her argument. Taiwan is only one part of it. She calls for scaling back US military ties with South Korea, withdrawing US troops from Japan and replacing the current alliance with a narrower security partnership, removing the roughly 500 US military trainers stationed in Taiwan, and withdrawing US military personnel and equipment from the Philippines. It’s a comprehensive set of strategic proposals.

The report’s central argument is that, as China’s military power continues to grow and the United States finds it increasingly difficult to sustain its forward-deployed forces, Washington needs to make a strategic adjustment to its overseas military posture and devote more resources to domestic development. She firmly believes that such a shift would best serve America’s national renewal and long-term interests.

It’s important to recognise that none of this makes Kavanagh “pro-CPC” or “pro-China”. Her arguments are grounded firmly in America’s national interests. Some of her proposals on Taiwan don’t make her “one of us”. There’s no need to overstate their significance, let alone get carried away.

In fact, if American forces really did begin pulling back from the Western Pacific, the move could set off unforeseen consequences. This is another issue that deserves serious consideration.

It’s important to move beyond the labels attached to Kavanagh. She’s neither a “pro-CPC, anti-American” scholar nor a hawk. She’s an independent-minded scholar of the 1980s generation, with views and ideas of her own.

Second, her arguments are still far from fully understood and need to be considered in a broader context. What has led an American think-tank scholar to criticise US defence policy and propose such changes? A sound judgement requires a fuller understanding of her thinking. Focusing on just one aspect of her argument or a few isolated passages can easily lead to misunderstandings.

When she visited China last year, hardly anyone here knew who she was. This time, she’s attracted a great deal of attention, and people everywhere want to meet her. For an academic, being thrust into the spotlight so suddenly, with all the pressure that brings, can’t be easy. I think we should appreciate that.

From Truth Decay to Withdrawal from the First Island Chain

Guancha.cn: When you discussed this report previously, you said your main interest wasn’t really the second island chain or the Taiwan question. Instead, you raised two questions: what’s happened to facts and truth, and what’s happened to trust?

We talked about the loss of trust in international relations before the 2024 US election. How important do you think trust is in relations between major powers today? Have your conversations with Kavanagh, and your reading of her work, helped you find any answers or evidence?

Jia Min: To understand that, we need to go back seven or eight years, to Kavanagh’s earlier research.

During Trump’s first term, she was a senior analyst at the RAND Corporation. Together with Michael Rich, who was RAND’s president at the time, she produced a study called Truth Decay. It attracted a lot of attention and even made Barack Obama’s summer reading list that year.

Truth Decay wasn’t about military or foreign policy. It was about what was happening within American society.

The basic idea is that Americans are finding it increasingly difficult to agree on facts and truths. And a direct consequence is that it’s much harder for them to find common ground in public debate. The result is growing division and confrontation, leading to the political polarisation much talked about in the United States today.

As that cycle continues, trust between the public, the government, and different groups in society begins to break down. In other words, “truth decay” leads to “trust decay” in American society. I think this is one of the most serious problems American society has faced in recent decades.

When I met Kavanagh, I tried to connect Truth Decay with her more recent work on US military strategy, including her proposals on NATO and the Western Pacific. I asked whether there was a common thread running through the two — an underlying logic that had developed over time. She said I’d understood her correctly: the connection was indeed there.

Everyone is talking about her latest report, but few seem to have noticed how closely it relates to her earlier thinking.

Take the assumption that maintaining a large US military presence along the first island chain must serve America’s national interests. Is that necessarily true? Does that consensus really hold up to scrutiny?

For a long time, few people seriously questioned it. Through her research, Kavanagh has found that a complex web of competing interests lies behind this supposed consensus.

One of the main goals of A Second Island Chain Strategy for Asia is to encourage the United States to devote more resources to domestic priorities. When a country overextends itself, pouring excessive resources into forward military deployments and positions on its periphery, it inevitably drains resources from home. At some point, strategic retrenchment becomes a necessity.

Yet some influential groups in the United States oppose withdrawing from the first island chain. There are lobbying groups, vested interests, and pressure from other countries and regions. But do these really serve America’s national interests?

Consider Nancy Pelosi’s visit to Taiwan in 2022. Some think-tank experts and partisan figures portrayed it as proof of unwavering American support for Taiwan. But was the visit actually in America’s national interests? And should that interpretation automatically be accepted as the truth?

That’s the point: what people take to be the truth can itself be questioned. Kavanagh is challenging assumptions that have long been taken for granted. She’s trying to show that without a genuine consensus grounded in facts, Americans cannot clearly identify their country’s national interests, let alone work towards national renewal.

I think her arguments deserve serious attention. At the very least, they should be treated with respect and subjected to rigorous debate.

Seen this way, there’s a clear continuity in her work. Through her research and policy proposals on foreign and defence policy, she’s revisiting some of the ideas she explored in Truth Decay.

Kavanagh agreed strongly with my reading of her work. She felt I’d grasped the connection between the different stages of her research.

I see Truth Decay as a study of America’s domestic problems in the era of “Trump 1.0”, while A Second Island Chain Strategy for Asia examines America’s foreign policy in the era of “Trump 2.0”. The two are closely connected. Foreign policy is rooted in domestic politics. Without a clear understanding of a country’s domestic politics, it’s difficult to develop fresh perspectives on its foreign policy.

Some American scholars accuse Kavanagh of “selling out” American interests, but I don’t think that’s fair at all. She identified some of the root causes of American society’s problems years ago. To portray her current work as a betrayal of America is to misunderstand what she’s been trying to do.

Is There a Window of Opportunity for China-US Relations?

Guancha.cn: Towards the end of her report, Kavanagh argues that Washington should communicate extensively and continually with Beijing to avoid misunderstandings about how US military forces will ultimately be deployed. She says these changes should be framed within the broader concept of “constructive strategic stability”, which the two countries have identified as the direction of their relationship.

At the same time, she stresses that a smaller US military presence should not be mistaken for a complete withdrawal from Asia. Washington should reaffirm its enduring interests in the region and make clear that its more limited deployments are intended to promote stability, not signal that America is abandoning the region.

In China, there’s also been discussion recently about the need to “add new dimensions to the vision of building a constructive China-U.S. relationship of strategic stability”. What might that mean in practice? Could Kavanagh’s report be seen as a proposal worth discussing — not necessarily something to adopt wholesale, but a contribution to the debate about what constructive strategic stability should involve?

There’s another question I’d like to raise. Whether in Kavanagh’s report or in discussions among American scholars about China-US dialogue on artificial intelligence, one sometimes gets the impression that Americans are looking to the Cold War relationship between Washington and Moscow as a model for managing relations with China, particularly when it comes to arms control and nuclear negotiations. Do you think that comparison makes sense?

Jia Min: Before I answer, there’s something I’d like to add.

Kavanagh proposes that the United States shift part of its military presence from the first island chain towards the second, scaling back its military commitments in the region. That would change America’s relationships with Japan, South Korea, Australia, and Taiwan. But there’s one aspect of her argument that should make China particularly wary.

She raises the possibility that if Washington reduced its military commitment to Japan and gradually replaced their mutual defence alliance with a narrower security partnership, Japan might respond by acquiring nuclear weapons or nuclear-powered submarines. From China’s perspective, that would cross an absolute red line. It would upend the post-war order in Asia and destabilise the entire geopolitical security landscape of Northeast Asia. China could never accept that.

In Northeast Asia, Russia, China, and North Korea all have nuclear weapons. Japan and South Korea are the two major countries without nuclear arsenals of their own. Of course, America’s alliances place Japan and South Korea under the US nuclear umbrella. American nuclear submarines sometimes call at South Korean ports, and US nuclear-powered aircraft carriers operate from bases in Japan.

If Washington downgraded its security relationship with Tokyo, Japan might conclude that, as a major regional power facing external military threats, it had every reason to acquire nuclear weapons or nuclear-powered submarines of its own. That could have enormous consequences for the region’s geopolitical balance.

So it would be a mistake to focus solely on the report’s proposals to reduce US military commitments to Taiwan and conclude that the report as a whole is favourable to China. That is not the case. This issue deserves serious discussion.

What makes the report valuable is that it opens up a debate about assumptions that have long been treated as settled truths. It gives people with different views a chance to make their case. I think that’s a healthy development.

Now, to return to the question of constructive strategic stability between China and the United States. Although the two countries have introduced the concept, they still differ in how they understand and interpret it.

From the American perspective, “strategic stability” is hardly a new concept. It originally referred to a form of stability that emerged from nuclear competition between the United States and the Soviet Union.

In the 1960s, the two superpowers were locked in an intense nuclear arms race. Gradually, they came to recognise that unchecked competition could end in mutual destruction. That led them to communicate more closely and establish mechanisms such as the hotline between Washington and Moscow. In military affairs, they gradually developed a degree of mutual understanding and trust.

So in the US-Soviet context, strategic stability was largely about nuclear weapons and military competition. It was very much a Cold War concept, and the Americans were among its principal advocates.

But the Cold War framework cannot simply be applied to China-US relations today. The concept of strategic stability developed between the United States and the Soviet Union is far too narrow to capture what constructive strategic stability between China and the United States entails. Indeed, the two are fundamentally different.

The term may be the same, but its meaning is not. The Cold War understanding of US-Soviet strategic stability should not be used as a model for interpreting China-US relations today.

In other words, China and the United States both need to play an active role, or at the very least work together, to broaden and deepen the dimensions to the vision of constructive strategic stability. That’s a fundamental difference from the Cold War approach.

There’s no shortage of areas where the two countries can cooperate. People-to-people exchanges, youth exchanges, and cooperation at the local level are just a few examples, and work in these areas is already under way.

Take artificial intelligence. China and the United States are the only two countries at the forefront of AI development. Each has its own strengths, and the gap between them is narrowing.

Ahead of the latest meeting between the Chinese and American presidents, the US sent a delegation to China for visits and exchanges. They got to see some things, though there were others they didn’t have the chance to explore.

Now, if the US claims to have such a huge technological advantage, why does it still feel the need to send people to China? Because China has a vast and rapidly evolving range of real-world AI applications, some of which may go beyond what the Americans had imagined. That’s why they’re so eager to find out what China is doing. Clearly, both sides have practical reasons to engage with each other.

At the same time, the growing use of AI in the military, technology, finance, and other sectors brings serious risks. China and the United States need to put aside their differences and talk frankly about those risks. I think such dialogue and cooperation could make a real difference in the years ahead.

Emergency response is another area where the two countries could work together. In September, a severe natural disaster struck the border region between China and Nepal, affecting people in both countries and elsewhere. Climate change was closely linked to the underlying causes.

AI can play an indispensable role in preventing natural disasters. Could China and the United States explore new forms of cooperation in this field and, in doing so, add new dimensions to the vision of building a constructive China-U.S. relationship of strategic stability?

Ultimately, the point of building such a relationship is for China and the United States to do something useful for humanity and contribute to building a community with a shared future for mankind. Seen in that light, there’s a tremendous amount the two countries could do together.

I’m not worried that we’ll run out of areas for cooperation. The challenge is making it happen: tackling practical problems one field at a time and turning ideas into action. That would be far more worthwhile than simply talking about “stability” without taking concrete steps to achieve it.

Guancha.cn: In fact, many people believe there’s a limited window of opportunity in China-US relations right now. Trump is already more than halfway through his second term, with the midterm elections approaching. And there’s no telling how American politics might change after 2028.

Given that the window may be relatively short, where should China and the United States concentrate their efforts to add new dimensions to constructive strategic stability? Can the two sides bring greater certainty to their relationship, so that its key priorities and commitments are less vulnerable to abrupt changes whenever a new US administration takes office?

Jia Min: The biggest hope for a breakthrough right now is on Taiwan.

During the talks in Beijing in May, the Chinese leadership made clear to President Trump that Taiwan was a major and sensitive issue in China-US relations and urged Washington not to play with fire. Trump offered some reassurance in response. He said that the United States did not want to “have somebody go independent” and that he had no desire to travel 9,500 miles to fight a war. He subsequently delayed some arms sales to Taiwan.

So China will continue to raise the Taiwan question in its dealings with Washington, putting forward its views and proposals. Its position remains firm, though there is room for flexibility in how it is expressed.

The hope is that the current US administration can make a breakthrough on the Taiwan question while this window of opportunity remains open. That includes questions surrounding America’s military deployments in the Asia-Pacific and its so-called military commitments, which brings us back to the issues raised in Kavanagh’s report.

I don’t know whether any of her proposals will ever become US policy. But at the very least, she’s put the elephant in the room. She’s raised a question that can no longer be ignored. The report has already made waves, and I hope it will lead to a much wider debate among US think tanks and strategic community.

I think this is a useful start. Whether the debate can gain momentum will depend on the timing. After all, the midterm elections are now the biggest issue in American politics. If Trump and the Republicans suffer losses, or even lose their majorities in both houses of Congress, the administration will face serious constraints on its ability to govern, and the direction of its policies will become much more uncertain.

But in the long run, whether Republicans or Democrats are in power, the United States will have to engage with China and work to manage the relationship. Given China’s size, strength, and capabilities, Washington can hardly afford to ignore it.

Strategic Retrenchment Is Extremely Complicated, Making China-US Dialogue All the More Important

Guancha.cn: You’ve warned that a US withdrawal from the first island chain wouldn’t necessarily work to China’s advantage. It could open a Pandora’s box, bringing greater uncertainty to East Asian geopolitics. Indeed, Japan, the Philippines, and Taiwan are already moving towards closer coordination.

A Russian scholar previously argued in a commentary for Guancha.cn that America’s relative decline and strategic retrenchment could pose serious risks to the international order and the regions concerned. In that context, cooperation between China and the United States to minimise the negative consequences of an American withdrawal would be particularly important.

What’s your take on this?

Jia Min: Let me draw an analogy, imperfect though it may be.

After the Second World War, the decline and break-up of the British Empire left the world with two major problems whose consequences are still felt today.

First, when Britain withdrew from India, it restored self-government but also oversaw the partition of British India into India and Pakistan along religious lines. This laid the groundwork for conflicts that have troubled the subcontinent ever since, and serious geopolitical tensions persist in the region to this day.

Second, Britain’s withdrawal from the Middle East after the Second World War also left behind a host of unresolved problems. Many of the region’s present-day borders, ethnic tensions, and religious conflicts have their roots in decisions made by Britain at the time.

Broadly speaking, Britain and the United States share an Anglo-Saxon cultural tradition. So if America begins to retreat from the international order it helped establish after the Second World War, what might that mean for the geopolitical landscape of East Asia?

Ideally, China would be able to “fill” some of the vacuum. But what if Japan stepped in instead, particularly forces associated with a revival of Japanese militarism? The consequences would be much harder to predict.

Strategic retrenchment is an extremely complex undertaking that calls for careful, objective, and thorough examination. I can’t offer a definitive conclusion at this stage. It’s a question that requires long-term consideration.

As America’s relative power declines, China needs to consider seriously how to protect its core interests while preventing the resulting power vacuum from fuelling rivalries among regional powers and triggering new conflicts.

This makes it all the more important for Beijing and Washington to strengthen communication on major regional security issues and strategic decisions with far-reaching implications.

Adding new dimensions to the vision of constructive strategic stability means broadening the dialogue beyond the issues already on the table, with more exchanges at different levels and on a wider range of subjects. Discussion, and even debate, between the two sides is necessary.

Of course, China-US relations cannot simply be reduced to a matter of dividing up spheres of influence, as some media outlets and think tanks have suggested. Beyond the major powers, there are many middle and smaller powers that will inevitably play a role. Some may even try to exploit tensions between major powers. The major powers must be careful not to let smaller countries exploit or even hold them hostage.

The Philippines is an obvious example. Manila has repeatedly provoked incidents in the South China Sea, emboldened in part by its belief that it can count on defence cooperation arrangements with the United States and Japan. Washington, of course, will continue to caution Manila against provoking a conflict with China, as such a confrontation could ultimately draw the United States and China into direct conflict, at enormous cost. The authorities in Taiwan have similar intentions, seeking to use outside powers to advance their own interests.

But the Trump administration understands perfectly well that it must not risk a confrontation for the sake of others’ narrow interests, particularly at the cost of destroying the hard-won trust it has recently established with the Chinese leadership. The Philippines may well be hoist with their own petard.

Generational Change and Its Impact on US Domestic and Foreign Policy

Guancha.cn: Kavanagh was born in the 1980s. Her generation grew up enjoying the benefits of globalisation after the Cold War, but also lived through 9/11, the wars that followed, and the financial crisis.

One striking feature of her, judging from the report, is that she seems relatively free of strong ideological leanings. Instead, her thinking reflects a form of nationalism shaped by globalisation, though one quite different from that associated with the MAGA movement.

How does Kavanagh’s generation differ from the Cold War generation, which still wields considerable influence in American politics, business, and academia? Could people like her eventually become the mainstream in American society? How should their views be understood? And does this reflect a broader generational shift in the United States?

Jia Min: That’s a very good question. As I understand it, Kavanagh was born in the 1980s. She studied at Harvard as an undergraduate, then earned her PhD from the University of Michigan. She’s spent her career working in think tanks and research institutions.

We did talk about generational differences at the salon, including how Chinese and American people born in the 1980s might engage with one another.

Kavanagh told me that China had benefited greatly from globalisation over the past three or four decades. After joining the World Trade Organization, China experienced rapid economic growth and enjoyed a relatively stable society.

But the American experience was very different. For much of the past two or three decades, the United States has been almost constantly at war. For many Americans, that period began with 9/11, followed by the wars in Iraq and Afghanistan, the financial crisis, the withdrawal from Afghanistan, and a succession of other upheavals.

So although American society remained broadly prosperous and stable as Kavanagh was growing up, it was also living through an era of “forever wars”. The lasting impact of this experience on people’s state of mind should not be underestimated. It’s an important part of understanding her generation.

They’ve grown up with one crisis after another, which has made them particularly sensitive to change. They’re also more inclined to look back nostalgically on America’s past glory, while their disappointment with the present can sometimes take rather irrational forms.

I’ve been interested in generational differences in America for a long time. According to a theory developed by an American scholar, history unfolds in cycles of roughly 80 years, each comprising four generations.

The first is the Prophet generation. Its members are generally born at the beginning of a relatively prosperous era and may rise to positions of leadership later in life. Trump is a typical example.

The second is the Nomad generation, broadly comprising those born in the 1960s and 1970s. They grew up in relative material comfort but often lacked a clear sense of direction and could be reluctant to take responsibility.

The third is the Hero generation, which includes people born in the 1980s and 1990s, like Kavanagh. They enjoyed relatively comfortable lives growing up, but as they enter middle age, they find themselves facing crises that force them to rise to new challenges.

Finally, there’s the Artist generation, broadly corresponding to younger people born in the 1990s and 2000s. They may have grown up in a society that places greater emphasis on the expression of ideas.

I told Kavanagh that Chinese and American people born in the 1980s belong to what this theory calls the Hero generation.

Today, both countries face a range of crises. People in their middle years, who form the backbone of society in China and the United States, need to come up with solutions to the problems of our time. Those solutions may differ considerably from what previous generations have proposed. Some may even break with conventional thinking. But at this point in history, why not have the courage to put forward bold, unconventional ideas?

I also mentioned an American politician born in the 1980s who I believe will play a greater role in the years ahead.

Kavanagh immediately asked, “You mean Vance?”

I said yes.

She said she had quite a few criticisms of Vance and was unhappy with some of his positions, which she felt could be inconsistent.

Still, they belong to the same generation. People born in the 1980s are increasingly entering the mainstream of society. In fact, many are already in positions where they can shape policy, advocate new ideas, and make their voices heard in public debate.

As this generational shift unfolds, China and the United States need to listen to each other and seize every opportunity for dialogue. Both countries have benefited from globalisation, but Chinese and Americans born in the 1980s have had very different experiences.

After 9/11, America entered a prolonged period of war, while China enjoyed relative stability and development. Those contrasting experiences make mutual understanding and dialogue all the more important.

I personally gained a great deal from inviting Kavanagh to the salon. Speaking with her gave me a sense of the unease and restlessness in American society, as well as a clearer understanding of her views on certain issues.

For China’s part, it’s important to approach the deeper anxieties and preoccupations of American society with greater empathy and an open mind. That would help lay the groundwork for more productive and substantive dialogue in the future.

I’ve long believed that China and the United States should work towards a “generational consensus”. Shared interests are obviously important, but beyond that, it’s worth asking whether people of the same generation in the two countries can find common ground on a range of issues.

Take people born in the 1980s. Whether Chinese or American, we share many of the same anxieties: raising families, caring for ageing parents, and making ends meet. Could these common concerns provide a basis for consensus?

Younger Chinese and Americans, particularly those born in the 1990s and 2000s, also share similar views on many issues. They’re learning from each other through social media platforms such as RedNote and TikTok, travelling to each other’s countries, and experiencing first-hand the pace of change and everyday life in China. Isn’t that a form of generational consensus in itself?

Of course, people born in the 1960s and 1970s grew up in another age. The United States was the dominant power, while China was still catching up.

Those born in the 1980s have watched the gap between the two countries narrow in certain areas. People born in the 1990s and 2000s have grown up seeing China and the United States on an equal footing in some fields. That sense of living through the same historical moment, and the common ground it creates, is something other generations could not have experienced.

I believe this kind of generational consensus could become, and perhaps is already becoming, an important foundation for dialogue on constructive strategic stability between China and the United States. That’s why I’m relatively optimistic.

Guancha.cn: As someone born in the 1980s myself, I can relate to what you’re saying. Reading Kavanagh’s report, with its wealth of evidence and data, I can understand why an American scholar like her would take the position she does.

Of course, understanding her perspective doesn’t mean taking satisfaction in what the report says or labelling her “pro-China”. The point is to understand the thinking and reasoning behind her position.

Jia Min: Exactly. One thing that distinguishes this report is that it moves away, to some extent, from the heavy emphasis on ideology and values that characterised a lot of earlier research. That was a fairly common way for the previous generation of scholars to make their arguments.

Generational change really does affect policymaking, but I don’t think it has received enough attention. Different generations have different perceptions of the times they live in and different capacities for empathy, which in turn affect their ability to act together.

There’s a reason for the emphasis on a new starting point in China-US relations, beginning with exchanges between children and young people and broader people-to-people engagement. This is grounded in practical considerations.

Kavanagh hopes her arguments will be understood back home, but right now she’s under considerable pressure. When she returns to the United States, she’ll still have to debate her critics and defend her views. She later wrote to thank us for organising the salon and for the discussion we’d had, saying that she, too, had found the exchange valuable.

That brings me back to the question I raised earlier: can people of the same generation, but from different countries, find common ground?

We organised the event because fundamentally, we believed it was in China’s interests. There’s no need to pretend otherwise. Kavanagh, for her part, wants America to become better, stronger, and more prosperous. If both sides are working towards the development and interests of their respective countries, that could itself provide a basis for a constructive consensus.

There’s no need to regard the other side as “one of us”. The aim is simply to approach each other with empathy, seek mutual understanding, and make progress together. Isn’t that the essence of the Chinese idea of “harmony without uniformity”?

Today, I think it’s important to understand America with empathy. American society has plenty of problems, but that’s no reason to ridicule it. In examining America’s problems, China should also reflect on whether similar problems could arise at home and how they might be prevented. The aim is to learn from America’s experience, including its mistakes, and avoid falling into the same traps.

Such exchanges should be mutual and work in both directions. It’s not simply a matter of China learning from the United States; America can learn from China too.

For now, these remain aspirations. But aspirations cannot be realised through words alone. Turning them into reality takes strength and practical effort.

Share

Much of the conversation was highly technical, covering Huawei’s new computing architecture, chip interconnection technologies, and its efforts to compete with Nvidia despite restrictions on access to advanced semiconductor manufacturing.

But five remarks stand out for a broader audience, without having to delve into the technological details.

1. China may eventually share American companies’ concerns about AI risks—but for now, it needs to accelerate.

Asked about calls by some leading American AI developers to slow down AI development, Xu offered an interesting explanation for why Chinese companies may see the risks differently. American frontier developers possess considerably greater computing resources and may have reached levels of AI capability that Chinese developers have yet to experience.

“Several leading American model developers have access to much greater computing power. Exactly how far they have progressed is something that perhaps only they themselves know. The AI risks they are experiencing may not yet be felt nearly as strongly in China.

For now, China still needs to accelerate its own pace of development. Once we reach the point where we can genuinely perceive those risks, perhaps we will feel the same way as the leading American model developers. But I have always believed that we must both advance AI development and manage its risks. We must ensure that AI is used for good rather than for evil.”

Xu’s point is not that those risks are imaginary or should be disregarded. Rather, he suggests that China may come to share American frontier developers’ concerns as its own AI capabilities advance.

2. The global AI hardware supply crunch may not end until 2029—and China could take longer.

The rapid expansion of AI infrastructure has strained global supplies of chips, memory, optical networking equipment, and other critical components. Xu believes the industry was not prepared for the speed of growth and predicts that supply will take several more years to catch up with demand.

“The reason optical module manufacturers and memory companies can charge such high prices today is not that their products are necessarily exceptionally good. It is because demand vastly exceeds supply.

Therefore, these bottlenecks will not be fully resolved until supply and demand reach equilibrium both globally and in China. I expect the global supply-demand imbalance to be largely resolved around 2029. China may take a little longer.”

His forecast is particularly notable given China’s additional constraints on access to advanced semiconductor technologies. It also suggests that the current boom in AI infrastructure investment may continue to face supply bottlenecks for several years.

3. Nvidia’s CUDA software advantage is becoming less formidable, according to Huawei’s semiconductor chief scientist.

Nvidia’s competitive strength extends well beyond its chips. Its CUDA software ecosystem has long made it difficult for AI developers to switch to alternative hardware, including Huawei’s Ascend processors.

But Liao Heng argues that advances in AI programming are beginning to erode this advantage.

“Today, with the emergence of new compiler languages, these tools are gradually replacing aspects of the CUDA-based development approach and weakening CUDA’s ecosystem barriers. Frontier laboratories no longer attach the same importance to CUDA that they did two years ago.

This represents the gradual dismantling of barriers at the software level. I believe we are approaching, or may even have reached, a point of balance. The gap has narrowed substantially.”

Liao also said that the overall gap between Nvidia and Ascend had narrowed considerably, including in how familiar developers find the programming environment. This is Huawei’s assessment rather than an independently established parity between the two ecosystems, but it addresses what has historically been one of Nvidia’s most important competitive advantages.

4. Huawei believes Ascend may already have overtaken Nvidia in China, but cannot meet domestic demand.

U.S. export restrictions have created opportunities for Chinese AI chipmakers, particularly Huawei. Xu made a striking claim about Ascend’s position in China, while acknowledging that the company’s manufacturing capacity remains insufficient.

“At present, our Ascend supply is still far from sufficient to meet domestic demand, so we have no plans for a comprehensive expansion into overseas markets.

There are indeed a small number of countries with particularly strong needs, and we have conducted tests and supplied products to them, but the quantities have been very small.

It is difficult to calculate Nvidia’s market share in China. However, based on the data we are able to gather, Ascend should already have surpassed Nvidia.”

Xu did not specify whether his comparison was based on shipments, revenue, computing capacity, or another measure, so the market-share claim cannot be independently assessed from his remarks alone. Nevertheless, it is noteworthy that Huawei now believes Ascend has surpassed Nvidia in its home market.

5. Huawei will prioritize Chinese customers, while serving a limited number of overseas buyers seeking alternatives.

Asked about reports that Malaysia was considering deploying sovereign AI infrastructure based on Huawei’s Ascend 910C, Xu acknowledged that geopolitical considerations and the desire to diversify suppliers could make Huawei attractive to some overseas customers.

But he made clear where the company’s priorities lie.

“As for balancing domestic and overseas demand, our principle is very straightforward. Our primary focus is the domestic market. We give priority to Chinese customers who urgently need computing power.

Overseas, we serve a small number of customers who cannot purchase alternative solutions, or who are determined to have another option.”

In other words, Huawei is not currently positioning Ascend for a broad international sales push. Instead, its limited overseas supplies are aimed primarily at customers who cannot obtain competing products or who want an alternative supplier, while China’s enormous domestic demand takes precedence.

Subscribe now

The complete Q&A covers Huawei’s broader technological strategy, including its new Peerium computing architecture, UnifiedBus interconnection technology, plans for future Ascend chips, and efforts to improve the efficiency of large-scale AI training. It is very long and technical, so I will not put out a translation.

Share

One other Huawei development during China’s just-concluded National Day holiday deserves attention. On October 5, Huawei and Qualcomm announced a multiyear patent cross-licensing agreement covering 5G, computing, AI, and networking, alongside Qualcomm’s planned purchase of certain Huawei patents in the United States.

Some Chinese media reports claimed that Qualcomm would become a net payer to Huawei under the licensing agreement—meaning it would pay Huawei more in royalties than it receives. Qualcomm subsequently disputed this, saying the terms were confidential and the reports incorrect.

Just over a month earlier, on August 26, Huawei and HP had announced a global patent cross-licensing agreement covering Wi-Fi technology, although HP subsequently emphasized that the deal settled a patent dispute rather than represented a broader strategic or commercial partnership.

There is something striking about a Chinese company subjected to years of U.S. sanctions continuing to conclude patent deals with some of America’s leading technology companies. Whatever the financial terms, these agreements offer tangible commercial recognition of the value of Huawei’s intellectual property.

They also present an interesting contrast with the ongoing federal criminal trial in Brooklyn, where U.S. prosecutors have accused Huawei of stealing American technology to build its telecommunications empire—allegations Huawei denies. The patent agreements offer a rather different perspective on Huawei’s technological capabilities.

Share

Speaking during the plenary session of the 23rd Annual Meeting of the Valdai Discussion Club on October 1, Wang, also former Counselor to China’s State Council, had this Q&A with Putin

Henry Huiyao Wang

（Mr. ) President. My name is Huiyao Wang, Founder and President of the Center for China and Globalization (CCG)

You mentioned that you’re going to APEC, and we actually saw you had a very successful summit last year with President Trump, and you had a very successful summit with President Xi this year. Of course, President Xi just came back from United States, also had a very successful summit. So I’m wondering, when you are at the APEC, this possible trilateral summit. So, what do you expect China can play some more role, for example, in promoting the peace, making peace, or peacekeeping, or guarantor, whatever, so that can really facilitate this summit to be successful? Thank you.

Putin

I have no doubt that the summit will be successful.

As for the role of China, well, first and foremost, Chinese philosophy and its political thought has never had an issue of applying pressure or solving issues through military force. Well, it’s not part of Chinese philosophy. China always tries to do that through peaceful means, and we see that in the Chinese policy led by President Xi.

Could China play a role right now? Yes, of course. That is based on its increasing economic and military potential as well, on its strive to expand its zone of free cooperation. The initiative by President Xi, One Belt One Road, is aimed at precisely that. Together with China, we’re developing this initiative as well, as well as the Russian initiative, and we’re coupling them together. For example, the Russian initiative on Eurasian cooperation. We’re working together, and that brings positive results.

As for the trilateral meeting during the APEC summit, it is possible. I guess I have said that it will largely depend on China as host of the APEC summit, but yet again, it’s always useful to me to to clear up the positions of each other and so on. But most importantly, we need to find the agenda, the issues that could be part of that trilateral summit. We need to identify the topics for the discussion. Largely, that would depend on the hosts of the event, and hence President Xi.

It was a particularly timely question. Xi had just returned from his September 23–25 visit to the United States and summit with Trump, while Putin is expected to attend the APEC Economic Leaders’ Meeting in Shenzhen on November 18–19. The Kremlin had already floated the possibility of the three leaders meeting there, but no trilateral summit has been formally announced. Reuters reported in September that Russian officials had discussed the possibility while stressing that the substance and agenda would first have to be worked out.

Interfax separately reported the remarks, quoting Putin as saying that Russia was in favor of the idea in principle and that a potential meeting at APEC would depend “largely on China, as the host of the event.” It also carried his response that China “can” play a role in promoting peace.

The exchange quickly travelled beyond the room.

Reuters reported “Putin says he’s ready to meet Xi and Trump if agenda is worked out,” based on the answer elicited in the Valdai Q&A. The wire service subsequently included the same point in its broader report on Putin’s remarks that day, writing that the Russian president was ready for a trilateral meeting with Xi and Trump if an appropriate agenda could first be agreed. Neither Reuters report identified Wang as the questioner.

The line was subsequently highlighted by Sky News, which told readers that Putin “said he was ready for a trilateral meeting with Donald Trump and Xi Jinping,” and appeared in reporting carried by outlets including SBS News in Australia.

That media pickup gives the exchange significance beyond the fact that a Chinese scholar had the opportunity to address the Russian president. Wang’s question connected three issues that are likely to converge at APEC: the Xi-Trump relationship after their latest summit, Putin’s possible interaction with both leaders, and the unresolved question of whether Beijing might assume a more active role in diplomacy surrounding the war in Ukraine.

Henry Huiyao Wang has advocated versions of that idea well before the Valdai exchange. In March 2022, Wang wrote It’s Time to Offer Russia an Offramp. China Can Help With That. in The New York Times. He has repeatedly argued that China’s relations with Russia, Ukraine, the United States and Europe give Beijing unusual room to contribute to a negotiated settlement. In earlier writings and public discussions, including in Foreign Policy, he proposed widening the diplomatic table beyond Moscow, Kyiv and Washington.

His appearance at Valdai also reflects his unusually extensive participation in international policy debates for the head of a Chinese nongovernmental think tank.

He also appeared as a panelist in Responsibility: Individual or Universal? Eighth Session of the Valdai Discussion Club’s 23rd Annual Meeting.

Share

That question matters because Beijing has repeatedly said that the specific form of a Taiwan solution should take Taiwan’s realities into account and absorb views and proposals from people on both sides of the Strait. Yet there has been remarkably little serious public discussion of what that might mean institutionally. Li’s three essays try to enter precisely that space. Pekingnology published his first essay on September 18 and his first response to mainland critics on September 26. This is the third and most elaborate installment.

The discussion has also produced an unusual response from Beijing. At its September 9 press conference, the Taiwan Affairs Office was explicitly asked about the essays by “an associate professor at National Changhua University of Education.” A spokesperson said Beijing was pleased to see more Taiwanese people understand and discuss the concrete content of a “Two Systems” Taiwan solution and offer proposals for it. One week later, on September 16, another spokesperson returned to the subject, saying Beijing welcomed Taiwanese people expressing different views and putting forward specific suggestions. On September 30, a spokesperson again encouraged more discussions, this time without referring to Li’s essays.

There is another feature of this episode that deserves attention. Li’s essays were published in full on Zhengdangshi (正当石), a mainland WeChat blog that has hosted an extended exchange among readers and scholars on both sides of the Strait. The platform has not confined the discussion to arguments congenial to Beijing. Li’s third essay, reproduced below, begins with perhaps the most uncomfortable possible fact for advocates of “One Country, Two Systems”: a Taiwan Mainland Affairs Council-commissioned survey conducted in August found that 81.2% respondents opposed the proposition. The same survey found 68.9% supporting maintenance of the status quo while rejecting unification and 83.9% saying Taiwan’s future should be decided by its 23 million people. Li puts those numbers at the beginning of an essay addressed substantially to mainland readers.

That makes the space opened around this discussion noteworthy. It would be wrong to conclude that Beijing has accepted Li’s proposed terms. But official spokespeople have twice welcomed the discussion, while a mainland social-media platform has carried arguments that do not merely ask what Taiwan must concede, but what constitutional constraints Beijing itself would have to accept. In that limited but significant sense, the boundaries of the discussion have been unusually broad.

And Li’s argument has become considerably more sophisticated in this third essay. He stresses that he is not endorsing “One Country, Two Systems,” nor drafting surrender terms for Taiwan. Instead, he describes the exercise as a conditional “stress test”: assume that the two sides have peacefully negotiated a settlement and Taiwan has accepted one China and unified national sovereignty; what institutional arrangements would then be necessary to make the bargain credible?

His answer is built around reciprocity. Taiwan, in his framework, would accept unified sovereignty, rule out de jure separation and foreign military intervention, undertake national-security obligations and accept limits on its autonomous powers. Beijing, in return, would have to convert “Two Systems” from a political promise into enforceable constitutional commitments: protecting elections, judicial authority, basic rights, fiscal and economic autonomy and, during a transition, elements of local defence, while making those guarantees difficult for the central authorities to diminish unilaterally. His central formula is succinct: Taiwan’s autonomy must not hollow out “One Country,” but sovereignty must not erase “Two Systems.”

Whether such an arrangement is politically imaginable is, of course, a different question from whether it can be designed on paper. For readers interested in understanding the distance between Beijing’s stated invitation to “explore” a Taiwan solution and the political realities of Taiwanese society, the essay below offers an unusually detailed look into that gap. — Zichen Wang

台灣教授再次回應大陸讀者對《兩制方案，台灣該談什麼？》一文的質疑

Taiwan Professor Responds Again to Mainland Readers’ Questions

Editor’s Note: The proposed “One Country, Two Systems” solution for Taiwan continues to generate lively debate. Over the past two days, we have received a steady stream of letters and submissions from readers on both sides of the Taiwan Strait. We will publish them one by one on this blog, so we ask readers for their patience.

Professor Ci-Ze Li at National Changhua University of Education in Taiwan published a response yesterday that has again prompted heated discussion, with the exchange becoming increasingly intense and substantive. Below are three comments from mainland readers.

---

Comment One

This blog deserves credit for the openness with which it has allowed this discussion to develop. The debate has now entered deep waters. As someone who has followed cross-Strait affairs for a long time, I feel as though I have witnessed a compressed version of the process of cross-Strait contact and negotiation.

Parts of the article are in fact similar to proposals advocated by certain groups on the Taiwan side that favour negotiations. The difference is that official political consultations on such a solution do not appear to have formally begun. In particular, meaningful exchanges with Taiwan’s ruling party are currently extremely difficult.

In my view, there is indeed an enormous gap between the two sides over what such a solution should look like. But this discussion can at least allow all sides, especially those in Taiwan, to appreciate just how different public sentiment is across the Strait when it comes to reunification.

In the past, some in Taiwan have tended to assume that it is the mainland’s authorities, rather than the public, that are driving the push for reunification, and that ordinary people on the mainland have no particular sense of urgency. At the very least, this exchange has given them a clearer view of what mainland public opinion actually looks like.

Reunification has never been merely a matter of what the state wants. It reflects a national will that has grown stronger alongside the great rejuvenation of the Chinese nation, as well as the expectations of the people.

Since Taiwan so often cites public opinion in Taiwan, this discussion of a solution to the Taiwan question has also given people there an opportunity to observe mainland public opinion and see the mainland as it really is.

That is why dialogue is better than no dialogue, and contact better than no contact. We should not wait until the two sides are facing each other in armed conflict before suddenly awakening to the reality of the divide.

Once the gap is recognised, the task should be to narrow it gradually. Otherwise, the value of the discussion will diminish.

The goodwill shown by the mainland towards Taiwan is an opportunity that Taiwan can make use of. It is only natural for Taiwan to seek, within the realm of what is possible, to secure its interests and preserve its rights. But that opportunity will not necessarily remain indefinitely, and the terms and benefits on offer may diminish over time. I therefore hope compatriots on the other side of the Strait will value the opportunity.

If Taiwan does nothing but keep demanding a higher price without making any effort to narrow the gap, it will ultimately be Taiwan itself that bears the loss. Indeed, Taiwan may be the place in the world that has been most daring to set conditions for the mainland. People in Taiwan might also compare this with how other countries and regions treat Taiwan.

Comment Two

Rational discussion is a good thing. It creates space within Taiwan for people to talk about reunification without reacting to the subject with fear.

Through this platform and this debate, I hope people in Taiwan can see more clearly the possibilities and prospects ahead. With the mainland behind them and the world before them, people in Taiwan should be able to enjoy broader horizons, more choices, a larger stage, and greater room for development.

I hope people on both sides of the Strait can work together with a shared sense of purpose. Differences in political systems and ways of life need not outweigh a shared language, culture, and heritage.

As for One Country, Two Systems, I believe the extensive experience accumulated in Hong Kong and Macao can help produce a more developed Two Systems arrangement for Taiwan and contribute to the steady, long-term development of peace across the Taiwan Strait!

Comment Three

I would like to thank this professor from Taiwan for the article. Written from the perspective of Taiwan’s intellectual community, it sets out his own conception of what a solution to the Taiwan question might look like. Whatever one’s perspective, it is a significant article in moving discussion of peaceful reunification towards greater detail, depth, and eventually technical institutional design.

After reading the article and the comments, it seems that the greatest point of disagreement between mainland residents and some groups in Taiwan concerns the future of Taiwan’s existing system.

Mainland residents, inspired by the courage to “change the old world for a new one” [quoting Mao Zedong’s poem] and by the experience of nation-building, have gradually gained greater confidence in the socialist system and tend to regard “One Country, Two Systems” as the ultimate objective of peaceful reunification.

In Taiwan, by contrast, even though it has suffered considerably from the shortcomings of its electoral system, many vested-interest groups and ordinary citizens continue to treat democracy and freedom as unquestioned standards and to place great value on the so-called democratic system that has operated there for many years. They therefore feel a natural apprehension about accepting a different social system.

Historically, people’s perceptions of social systems are influenced to a considerable extent by comparisons of national strength and living standards, and those perceptions inevitably contain an emotional or irrational element.

Modern Chinese history has itself been a process of comparing and choosing among different systems before ultimately arriving at a social system with Chinese characteristics. A sound system should be able to withstand the test of history.

---

In response to yesterday’s discussion and criticism, Professor Ci-Ze Li has written another article of nearly 20,000 Chinese characters. The tables included in the article may be incomplete for editorial reasons, for which we ask readers’ understanding.

We publish the article today as the thirteenth instalment in our series of letters from readers on both sides of the Taiwan Strait.

The full text follows.

“One Country” Must Not Be Hollowed Out by Autonomy, and “Two Systems” Must Not Be Erased by Sovereignty

I am not endorsing a “One Country, Two Systems” solution to the Taiwan question, still less to draft surrender terms for Taiwan in advance. My purpose is to translate a political slogan into institutional provisions and conduct a conditional stress test of how such a system might work.

For the purposes of this exercise, the article assumes that the two sides reach a peaceful negotiated settlement and that Taiwan freely accepts “one China, with national sovereignty exercised uniformly by the central authorities”.

On that premise, the real questions are not whether the slogan sounds attractive, but three issues.

First, after sovereignty is unified, can Taiwan preserve an elected government, final judicial authority in Taiwan, basic rights, its existing fiscal and economic system, and necessary local defence capabilities?

Second, can China’s unitary constitutional system accommodate a degree of autonomy more extensive than that enjoyed by Hong Kong and Macao, and one that is more difficult to diminish unilaterally?

Third, can Taiwan itself be effectively constrained so that autonomy cannot again be used as a vehicle for separation, while the central authorities are also bound by higher-order law strong enough to constrain their own powers?

The core test is therefore clear: is Beijing willing to make constitutional commitments that constrain its own power in return for Taiwan assuming its obligations within the unified state after reunification?

Only if Beijing is willing to do so can there be a genuine basis for joint discussion. If not, “Two Systems” would amount only to a political promise that could be adjusted at will as circumstances change.

Public Rejection and Beijing’s Urgency on Reunification

A survey commissioned by Taiwan’s Mainland Affairs Council and conducted by the Election Study Centre of National Chengchi University from 13 to 17 August 2026 found that 81.2 per cent of respondents opposed the CPC’s “One Country, Two Systems” proposition, 68.9 per cent supported maintaining the status quo and rejected reunification, and 83.9 per cent supported having Taiwan’s future decided by its 23 million people.

The survey interviewed people aged 20 and above by telephone and obtained 1,086 valid responses, at a 95% level of confidence with a sampling error of ±2.97%. Some questions were part of long-running surveys and produced broadly stable results. The question producing the 68.9 per cent figure was newly introduced and should therefore not be used for long-term trend comparisons.

A single poll is not an eternal verdict, but it is sufficient to show that any reunification formula would first confront a major legitimacy deficit, not merely a communications problem. At the same time, strong public opposition in Taiwan has not reduced Beijing’s urgency to advance reunification.

In January 2019, General Secretary Xi Jinping called for the exploration of “a ‘two systems’ solution to the Taiwan question” in his speech at the meeting marking the 40th anniversary of the issuance of the Message to Compatriots in Taiwan. He said, “In terms of how the principle should be carried out in Taiwan, we will fully consider Taiwan’s reality, give full consideration to the views and proposals from all walks of life on both sides and fully accommodate the interests and sentiments of our compatriots in Taiwan,” provided that “China’s sovereignty, security, and development interests are ensured.” He also stated that “Peaceful reunification means achieving it through consultation and discussion as equals.”

On 1 July 2026, at a gathering in celebration of the 105th anniversary of the founding of the CPC, Xi described complete national reunification as “an essential step towards national rejuvenation” and called for efforts to “firmly advance the cause of national reunification.”

[Yuxuan’s note: The phrase “an essential step towards national rejuvenation” does not appear in the publicly available full text of Xi Jinping’s speech on 1 July 2026. It appears instead in the 2022 white paper, The Taiwan Question and China’s Reunification in the New Era. Xi made a similar point in January 2019, in his speech marking the 40th anniversary of the issuance of the Message to Compatriots in Taiwan: “Our country must be reunified, and will surely be reunified. This is a historical conclusion drawn from the evolution of cross-Strait relations over the past seven decades; it is also critical to the rejuvenation of the Chinese nation in the new era.”]

The statement that “The Taiwan question originated in a weak and ravaged China, and it will definitely end with China’s rejuvenation” came from Xi’s 2019 speech.

On 17 August 2026, at a symposium marking the centenary of Jiang Zemin’s birth, Xi reiterated Jiang’s statement that “Complete national reunification is what the people desire and a surging popular trend. It cannot be stopped by any force.” According to official CPC accounts, that statement dates to 1999.

[Yuxuan’s note: This statement does not appear in the publicly available full text of Xi Jinping’s speech on 17 August 2026. Nor have I been able to verify from publicly available official sources that Jiang Zemin first used this exact formulation in 1999. Similar formulations did appear in Jiang’s speech around that period, including in 2001. In Xi’s 2019 speech, the idea appeared in the fuller formulation: “The rejuvenation of the Chinese nation and reunification of our country are a surging popular trend. It is where the greater national interest lies, and it is what the people desire.”]

Read alongside the CPC’s longer-term policy language, these statements can reasonably be interpreted as continuing to link reunification to national rejuvenation. That, however, is an interpretation of the texts, not a deadline expressly set by the speeches themselves.

These two sets of signals frame the problem examined here: Beijing has invited the two sides to “explore” and “discuss”, while Taiwanese society overwhelmingly rejects the concept.

Avoiding discussion will not make the issue disappear. It would merely leave the content of any “solution to the Taiwan question” to be written unilaterally by Beijing.

Since this is being framed as an exploration, democracy, the judiciary, national security, troop deployment, public finances, and safeguards for implementation must all form part of the negotiation agenda. Otherwise, this would not be a joint exploration of a solution to the Taiwan question; it would simply amount to asking Taiwan to accept a solution already drawn up. Substantive negotiations must allow for give and take, mutual accommodation, and obligations on both sides.

Three Premises and the Core Negotiating Formula

This stress test rests on three premises.

First, this article does not assume that the people of Taiwan would accept “One Country, Two Systems”, nor that Taiwanese society has reached any consensus in this regard. It is testing the institutional limits of a counterfactual scenario.

Second, this article assumes that Taiwan has accepted unified national sovereignty and the central government’s authority to represent the state. It therefore does not consider options beyond Beijing’s stated bottom line, such as two Chinas, one China and one Taiwan, a confederation, a separate UN seat for Taiwan, foreign troops stationed in Taiwan, or a unilateral right of withdrawal.

Third, the analytical core is the problem of “credible commitment”. The difficulty in negotiations lies not only in how much the other side is willing to offer now, but also in whether those commitments could later be withdrawn unilaterally after power has been transferred.

Any formula must therefore pass four tests: whether the sovereignty arrangement is coherent; whether the scope of autonomy is clearly defined; whether effective remedies exist when rights are infringed; and whether commitments are difficult to rewrite unilaterally.

The core negotiating formula proposed here is as follows.

Taiwan would accept a single national sovereignty and rule out de jure separation and foreign military intervention. In return, Beijing would accept a constitutionally defined list of powers, non-regression of rights, joint amendment procedures, and phased implementation, all of which would constrain its ability to alter Taiwan’s autonomy after reunification.

The former prevents autonomy from hollowing out “One Country”. The latter prevents sovereignty from erasing “Two Systems”. Any formula that imposes irreversible obligations on one side only cannot create a credible commitment.

Analytical Framework: Harm, Compliance, Governance and Credible Commitments

Any serious discussion of a reunification solution must first distinguish three capabilities that are often conflated.

The first is the capability to inflict harm or destruction, such as blockading ports, attacking energy and communications infrastructure, or defeating organised military resistance.

The second is the capability to compel compliance: the ability to make a government and society that still retain meaningful choices accept specific political conditions.

The third is the capability to govern after reunification: the ability to secure administrative cooperation, economic recovery, and social collaboration after power has been transferred.

Thomas Schelling long ago distinguished between taking what one wants by force and using threats to make the other side yield. Brute force can achieve its aims without the other side’s cooperation; coercion cannot, because it still depends on the other side choosing to comply. Inflicting enormous losses does not necessarily produce political capitulation. Controlling territory does not necessarily mean governing it at low cost.

Likewise, “consultation and discussion as equals” does not mean equality in material strength, nor does it imply that two political entities would remain equal after reunification. It is first and foremost a matter of procedural standing. Before reunification takes effect, any sustainable agreement requires valid consent from both Beijing and Taiwan, since both are indispensable parties to the agreement.

After reunification, national sovereignty and constitutional authority may be unified. During negotiations, however, both sides must have genuine room to propose, reject, exchange, and amend provisions. If the stronger side presents a single predetermined text and the weaker side can only sign it, the process is not equal consultation, and the agreement loses its substantive legitimacy.

Research in institutional economics on Britain’s Glorious Revolution has shown that rulers can make commitments credible only by placing institutional constraints on their own power, so that breaking those commitments becomes procedurally difficult and costly.

International-relations theory similarly treats the “commitment problem” as an important cause of conflict. Even where war would be costly and both sides prefer peace at present, an agreement that alters the future distribution of power may give one side an incentive to demand new terms later.

Cross-Strait relations are a classic example. For Taiwan, reunification would close off external military options, transfer ultimate sovereignty, and reduce its independent capacity for self-defence. Beijing, meanwhile, would hold national legislative, constitutional-amendment and legal-interpretation powers, as well as military superiority, after reunification.

Even if Beijing were prepared to offer generous terms during negotiations, it could later redefine the scope of autonomy after Taiwan had made irreversible concessions. Conversely, Taiwan could use autonomy, elections, or external ties during a transition period to delay implementation.

Credible commitments must therefore be reciprocal, although they need not be symmetrical. The institutional design should constrain each side in the areas where it has the greatest capacity to breach the arrangement. Taiwan would be bound by anti-separation obligations, restrictions on foreign military intervention, national-security legislation, and implementation deadlines. Beijing would be constrained by exhaustive lists of reserved powers, non-regression of rights, anti-circumvention rules, procedures governing legal interpretation, and joint amendment of core provisions.

How the Level of Commitments Has Changed: From Ye’s Nine Points to the New-Era White Paper

Historically, the autonomy Beijing proposed for Taiwan was more extensive than many people remember, and such proposals predated the 1993 white paper.

The 1981 “Ye’s Nine Points” proposed that after reunification, Taiwan could become a special administrative region, enjoy a high degree of autonomy, and retain its armed forces. The central authorities would not interfere in Taiwan’s local affairs, while Taiwan’s existing social and economic system, way of life, and external economic and cultural relations would remain unchanged.

In 1983, Deng Xiaoping rejected “complete autonomy”, making clear that Taiwan would remain a local government and must not harm the interests of a unified country. At the same time, however, he proposed judicial independence, final adjudication in Taiwan, the retention of armed forces in Taiwan provided they did not threaten the mainland, and no stationing of mainland troops or administrative personnel in Taiwan. He also advocated talks between the CPC and the Kuomintang on an equal footing, rather than negotiations framed as “central versus local”.

This history rules out two extreme interpretations at once. A high degree of autonomy is not equivalent to inherent sovereignty, but unified sovereignty does not necessarily preclude final adjudication in Taiwan, local armed forces, or the non-stationing of mainland troops.

The 1993 white paper, The Taiwan Question and Reunification of China, was the most systematic and detailed articulation by the State Council Information Office of a high-autonomy arrangement. It proposed the long-term coexistence of two systems, with “neither side swallowing up the other”. Taiwan’s existing social and economic system, way of life, and external economic and cultural relations would remain unchanged. It expressly set out administrative and legislative powers, independent judicial power, and final adjudication in Taiwan. Taiwan would manage its own party, governmental, military, economic and financial affairs, could conclude commercial and cultural agreements, enjoy a certain degree of authority over external affairs, and retain its own armed forces, while the mainland would not station troops or administrative personnel in Taiwan.

The 2000 white paper reaffirmed a high degree of autonomy and the non-stationing of mainland troops and administrative personnel in Taiwan. It expressly stated that the content of “One Country, Two Systems” for Taiwan could be more generous than that for Hong Kong and Macao. At the same time, however, the document for the first time listed “an indefinite refusal to negotiate reunification” as one of the circumstances in which non-peaceful means might be used.

“More generous” did not mean that Beijing had expressly committed to Taiwan’s current system of direct presidential elections, alternation in power between political parties, or comprehensive judicial review of rights. Those arrangements would still have to be specifically agreed through negotiation.

In 2019, General-Secretary Xi Jinping shifted to an open-ended call for “exploration”, saying that the specific form of “Two Systems” would “fully consider Taiwan’s reality, give full consideration to the views and proposals from all walks of life on both sides”, provided that “China’s sovereignty, security, and development interests are ensured.”

The 2022 white paper, The Taiwan Question and China’s Reunification in the New Era, reaffirmed that Taiwan could enjoy a high degree of autonomy in accordance with law, that the two social systems could “develop side by side for a long time to come”, and that with the approval of the central government of China, foreign countries may set up consulates or other official and quasi-official institutions in Taiwan.

At the same time, it stated more clearly that “Two Systems is subordinate to and derives from One Country”, and that “all Taiwan compatriots who support reunification of the country and rejuvenation of the nation will be the masters of the region.” The logic is close to the principle of “governance by patriots”, although the white paper does not use the phrase “patriots governing Taiwan”.

It is also notable that several provisions expressly set out in 1993 were not repeated individually in 2022: Taiwan retaining its armed forces, no mainland troops stationed in Taiwan, no mainland administrative personnel being sent there, Taiwan managing its own party, government, military, economic, and financial affairs, final adjudication in Taiwan, and a degree of authority over external affairs.

Silence cannot by itself be treated as formal withdrawal. But it does make the current status and negotiability of those earlier terms more uncertain.

The gap between these versions is itself significant and must be taken seriously. The 1993 high-water mark is not an acquired right that Taiwan can invoke against Beijing today, nor should it be treated as a minimum offer below which Beijing can never go. The balance of power across the Strait has changed, and Beijing’s negotiating terms may certainly change accordingly.

The significance of the gap lies elsewhere: Beijing once considered these conditions compatible with “One Country, Two Systems.” If Beijing now believes circumstances have changed, it can propose different terms. But it cannot retroactively characterise an arrangement that it once proposed itself as inherently separatist.

Historical documents therefore provide a negotiating reference point, but they do not solve the problem of implementation.

The most reasonable institutional test for Taiwan would not be to cite old documents and demand that Beijing implement them unconditionally. It would be to ask Beijing to state, item by item, which earlier terms remain valid, which have changed with circumstances, what alternative safeguards would replace them, and whether the terms it reaffirms would be written into the Constitution, a Basic Law, and an agreement that would take effect simultaneously.

Whether earlier commitments can once again be put down in black and white — and whether Beijing is prepared to put them into law — would be the first test of sincerity and credibility.

Constitutional Framework: One Sovereignty, Two Legal Domains, Four Lists

On this premise, the autonomy enjoyed by Taiwan under a special constitutional order would not amount to an inherent sovereignty equal to that of the state. It would instead be a high degree of autonomy granted under the Constitution of the People’s Republic of China and by the National People’s Congress (NPC).

The powers to be exercised directly by the central authorities, together with the national laws applicable in Taiwan, would have to be exhaustively specified in higher-level law and its annexes. Other matters of local governance would be handled by Taiwan within the scope of that authorisation.

The fact that authority originates in the state does not mean that the executive can withdraw the authority arbitrarily. The point of constitutional authorisation is precisely that state organs themselves are bound by the higher-order law they enact.

This framework would not be a confederation. Taiwan would not retain formal diplomatic relations, a UN seat, foreign troops, a unilateral right of withdrawal, or an independent right to use military force abroad. It would draw on safeguards used in federal and asymmetrical-autonomy systems to constrain unilateral change, but would not give Taiwan separate sovereign status.

Whether Beijing would be willing, for the sake of peaceful reunification, to adopt constitutional safeguards more robust than those used in Hong Kong and Macao is precisely what this stress test seeks to examine.

Under China’s unitary constitutional system, the principle that central reserved matters are exhaustively listed while other matters are handled by Taiwan cannot be framed as Taiwan transferring original powers to the central authorities. It must instead be designed as a general constitutional authorisation granted to Taiwan by the state.

That legal basis does not rule out the use of lists. On the contrary, it makes it all the more important to define precisely the scope of Taiwan’s authorisation, the powers reserved to the central authorities, and the procedures for amendment.

Article 31 of the Constitution of the People’s Republic of China provides a basis for establishing special administrative regions, but does not itself directly grant a high degree of autonomy.

Article 62 assigns the NPC powers including constitutional amendment, enactment of basic laws, and decisions on the systems to be established in special administrative regions. Article 64 sets out the constitutional-amendment procedure. Article 67 assigns constitutional and legal interpretation powers to the Standing Committee of the NPC.

Among existing statutes, Article 5 of the Anti-Secession Law states that “Taiwan may practice systems different from those on the mainland and enjoy a high degree of autonomy”, but still does not define the scope of autonomy, procedures for amendment, or mechanisms for legal remedy.

A solution to the Taiwan question would therefore require at least two layers of constitutional construction.

First, the Constitution of the People’s Republic of China would need to be amended to add a Taiwan-specific chapter and establish special amendment procedures.

Second, pursuant to Article 31 and the amended Constitution, the NPC would enact a Basic Law of Taiwan, expressly granting Taiwan a general authorisation for a high degree of autonomy.

The fact that authority originates in the centre does not mean the scope of that authority has to remain vague. The Basic Law of Hong Kong demonstrates that a unitary state can accommodate a high degree of autonomy and final adjudication within Hong Kong, while in principle limiting the national laws that apply there to those listed in Annex III. Article 22 also prohibits departments of the Central People’s Government, provinces, autonomous regions, or municipalities directly under the Central Government from interfering in the affairs which the Hong Kong administers on its own in accordance with the Basic Law.

However, Article 18 provides a route for national laws to apply in wartime and in specified states of emergency. Article 158 gives the NPC Standing Committee the final power of interpretation. Article 159 gives the NPC the power to amend the Basic Law, while additional Hong Kong consent is only required for amendment bills proposed by Hong Kong.

It is precisely this imbalance in ultimate control that means the Hong Kong model cannot simply be transplanted to Taiwan.

I. Using Comparative Law: Disassembling Techniques Rather Than Transplanting Systems Wholesale

The real value of comparative constitutional law does not lie in copying foreign systems wholesale or transplanting them without regard to context. It lies in identifying particular institutional and legislative techniques and understanding how they actually operate under specific political conditions.

The Åland Islands provide an example of a “double lock” in the division of powers between the state and an autonomous territory. Under Finland’s Act on the Autonomy of Åland, amendments to the autonomy statute require not only approval by the Finnish Parliament under a qualified procedure equivalent to that for constitutional amendments, but also the consent of two-thirds of the members present in the Åland Legislative Assembly. This “double consent” requirement prevents the central legislature from stripping away local autonomy through an ordinary legislative majority.

However, the success of this mechanism also depended on the international settlement established by the League of Nations in 1921 and on the protections arising from Åland’s demilitarised and neutralised status under international law. Without comparable constraints in the broader political environment, safeguards written only into domestic law often prove fragile.

Scotland illustrates a model in which residual powers lie with the devolved legislature. Under the Scotland Act 1998, only specified areas, such as defence, foreign affairs and currency, are reserved to the UK Parliament. Matters not reserved generally fall within the legislative competence of the Scottish Parliament.

At the same time, the Scottish model also illustrates the limits of constitutional conventions. Despite the Sewel Convention, the United Kingdom’s principle of parliamentary sovereignty means that the convention is not judicially enforceable. Devolved autonomy can therefore remain vulnerable when the central legislature is determined to act.

Article 5 of the Italian Constitution declares that the Republic is “one and indivisible”, yet the 2001 constitutional reforms, including Article 117, introduced a division of legislative powers resembling federal arrangements. The Constitution identifies areas of exclusive state competence and areas of concurrent competence between the state and the regions, while residual powers over matters not expressly listed belong to the regions.

Italy therefore demonstrates that substantial territorial devolution and residual regional powers can coexist with a unitary constitutional framework and the principle of an “one and indivisible” Republic. The key lies in drawing the boundary between central and regional powers with sufficient precision.

Catalonia provides a stark reminder that even an extensive process of local democratic approval may ultimately remain subject to the authority of the central judiciary.

The 2006 Statute of Autonomy of Catalonia was approved by the Catalan Parliament, amended and approved by the Spanish Parliament, and then endorsed by a strong majority in a regional referendum, giving it substantial democratic legitimacy. Yet in 2010, the Spanish Constitutional Court struck down a number of important provisions relating to judicial autonomy, fiscal autonomy, and the definition of “nation”.

The case shows, in sobering terms, that even where an autonomy arrangement has been endorsed through democratic procedures at both the central and territorial levels, its ultimate boundaries may still depend on the final interpretive authority of the central constitutional court.

These cases reveal a hard constitutional truth: a list of powers alone is not a cure-all for autonomy. An effective allocation-of-powers framework must be accompanied by the primacy of law, rules for classifying different types of disputes, a role for both the centre and the territory in final interpretation, and amendment procedures that cannot be changed unilaterally. Without those safeguards, written legal guarantees can be eroded at any time.

More importantly, these comparative techniques drawn from European territorial autonomy cannot be taken out of context and used by themselves to prove that Taiwan could retain armed forces, an independent currency, or final judicial authority.

European autonomy systems are built on the acceptance that the central state retains unified responsibility for defence, foreign affairs, and ultimate judicial sovereignty. Taiwan, however, would be facing highly adversarial cross-Strait relations and an authoritarian political system very different from a liberal democratic one.

Identical or similar legal provisions cannot be assumed to produce the same binding force or the same effects on peace in the absence of comparable independent review institutions, democratic constitutional conventions, and geopolitical balances. Comparative law can provide the technical components of constitutional design; it cannot substitute for substantive political conditions and sovereign power.

II. Four Lists

A more precise approach would be to replace the potentially misleading idea of a “positive enumeration of central powers” with four separate lists.

These would not turn the central authorities and Taiwan into two states. Rather, they would define the scope of authorisation, areas requiring co-operation and institutional red lines binding on both sides.

Table One would concern powers and structural boundaries. Table Two, discussed later, would concern conduct in implementing the agreement. The two serve different functions.

This framework would exclude, for Taiwan, a two-sovereign-state solution, confederate membership, a separate UN seat, formal diplomatic relations, foreign troop deployments, an independent right to wage war abroad and a unilateral right of withdrawal. At the same time, it would rule out using national security or unified sovereignty as blanket clauses through which central authority could enter fields outside the agreed lists at will.

The two sides would negotiate as procedural equals, while the outcome would be a single sovereignty; autonomy would derive from the state, but state organs would themselves remain bound by higher-order law. “One Country” would determine national belonging, formal diplomacy, and overall defence; “Two Systems” would determine how people in Taiwan live, who governs them, and whether the central authorities can arbitrarily alter Taiwan’s institutions.

Core Autonomy: Six Groups of Institutional Powers That Must Not Be Hollowed Out

The name and flag of a state carry dignity and historical meaning. But within the boundaries of what this article assumes Beijing might accept, the “Republic of China” could not continue as the name of a separate sovereign state.

Taiwan could nevertheless seek to preserve particular names, flags, calendar conventions, and holidays during a transition period, or retain them as symbols of local history and regional identity.

The issues that would most directly affect people’s daily lives, however, are the following six groups of institutional powers.

1. Democratic Institutions: Central Confirmation Must Not Become a Veto over Elections

Taiwan’s chief executive, legislature, and local leaders should continue to be chosen through periodic direct elections by residents of Taiwan. Space should also remain for multiparty competition, alternation in power, opposition parties and freely organised political activity.

Public officials could be required to swear allegiance to national reunification and Taiwan’s special constitutional order. The Basic Law of Taiwan could also prohibit organising, planning or carrying out separation, accepting foreign military command, engaging in espionage, or secretly receiving foreign funding to undermine the constitutional arrangement.

But such restrictions would have to be established by special legislation and differ substantially from Hong Kong’s 2021 candidate-eligibility review model. The grounds for disqualification would have to be exhaustively specified. Political bodies should not be allowed to add new requirements by invoking an abstract standard of being a “patriot”, and final judicial review should rest with Taiwan’s courts.

If Taiwan’s chief executive were formally appointed by the central authorities, that appointment should amount to a binding confirmation of a lawful election result. The central authorities should not be able to refuse appointment unless Taiwan’s court of final adjudication had found the candidate legally ineligible.

Peaceful criticism of the central authorities, academic discussion of alternative policies, or lawful advocacy of institutional reform that does not involve separation should not in itself constitute a crime or disqualify someone from standing for office. Otherwise, elections would merely confirm candidates already approved in advance. The democratic system might retain its name, but it would already have lost its function of conferring political authority.

Taiwan’s democracy does have problems, including partisan conflict, short-term policymaking, and political dysfunction. Those shortcomings are worth addressing, but they do not imply that people should lose the right to choose their government, replace those in power peacefully, and criticise authority openly.

The argument here need not prove in the abstract that one model of democracy is superior to all other systems. It need only recognise that elections, multiparty competition, freedom of the press, and judicial remedies have become part of Taiwan’s existing institutions and way of life.

Since Beijing says that it will “accommodate Taiwan’s actual conditions”, it cannot simultaneously promise to preserve Taiwan’s system while assuming from the outset that ruling parties will no longer alternate, candidates will be screened, and dissent will be restricted.

If people who oppose reunification are excluded first and only those who remain are allowed to express consent, the result is not democratic authorisation but filtered compliance.

2. The Judiciary and National Security: The Centre Cannot Be Excluded, Nor Can Taiwan Be Hollowed Out

The 1993 white paper explicitly proposed that Taiwan enjoy independent judicial power and final adjudication. That can serve as an important policy starting point for negotiations, but it is not itself a directly applicable legal source of jurisdiction.

The Basic Law of Taiwan would need to provide that ordinary civil, criminal, and administrative cases, together with cases involving Taiwan’s autonomous laws and basic rights, are heard by Taiwan’s courts and finally concluded there.

In principle, residents of Taiwan should not be transferred to the mainland for conduct occurring in Taiwan. Central public-security, state-security, prosecutorial, and military bodies should likewise not be able to independently carry out arrests, searches, surveillance, or other coercive measures in Taiwan.

Taiwan’s courts would review enforcement measures taken in Taiwan, as well as Taiwan laws and administrative acts. This would not amount to Taiwan’s courts invalidating decisions of the NPC.

A reunification solution could not allow Taiwan, in the name of autonomous legislation, to legalise organised armed separation, cross-jurisdictional espionage, acceptance of foreign military command, or violent attempts to undermine the reunification arrangement.

The central authorities would bear ultimate responsibility for national security, but that responsibility would not by itself confer unlimited law-enforcement powers on central national-security, public-security, or prosecutorial organs in Taiwan. The real issues must be separated: what conduct constitutes a crime, who legislates, who investigates, who adjudicates, and how exceptional jurisdiction can be triggered.

A more workable two-tier approach would be as follows.

At the national level, the core security threats would be exhaustively defined. Taiwan would then adopt its own detailed legislation and take responsibility for day-to-day investigation and adjudication.

Offences would have to be clearly defined and non-retroactive. Except where conduct has already caused serious actual harm, liability should at minimum require proof of a concrete, real, and imminent danger of serious harm. Mere expression of opinion should not be criminalised.

Joint jurisdiction should be triggered only in a very small number of cases involving foreign military command, organised armed separation, or major cross-jurisdictional espionage that Taiwan’s own authorities are demonstrably unable to handle effectively. Even then, arrest, search, and seizure in Taiwan should still be carried out by Taiwan authorities.

Defendants would be tried either by a joint court located in Taiwan or by a special panel of a Taiwan court, under procedures and rights protections specified in advance. They would not be transferred to the mainland.

3. Basic Rights: Non-Regression, Proportionality, and Judicial Remedies as the Floor

Taiwan’s existing rights concerning speech, the press, publication, the internet, academic activity, religion, assembly, association, privacy, labour, gender, family, Indigenous peoples, property, and due process should form a baseline of protection, defined by the Constitution, constitutional judgments, legislation implementing the two international human-rights covenants and other human-rights conventions, Indigenous legal frameworks, and other applicable law in force on the date reunification takes effect. The five legal instruments discussed below should then expressly establish a general principle of non-regression.

Any restriction in an individual case should require a clear legal basis, a legitimate purpose, necessity, proportionality, and an effective judicial remedy. A constitutional review body should be able to invalidate unlawful legal provisions, while administrative courts should be able to overturn unlawful administrative acts. Rights without an effective remedy are merely political declarations.

The institutional transition would also require provisions for political reconciliation across jurisdictions. Peaceful political advocacy, national identity, journalistic and academic activities, and ordinary military or public service before reunification should not become grounds for retrospective prosecution. This protection should be written simultaneously into the peace agreement, the Basic Law of Taiwan, and national reconciliation legislation.

For conduct that might fall within the scope of mainland criminal law, the NPC Standing Committee would also need to decide on an amnesty under Article 67 of the Constitution, followed by an amnesty order issued by the President under Article 80. This could not depend solely on decisions by Taiwan prosecutors not to prosecute. Amnesty and non-prosecution should exclude intentional homicide, torture, serious corruption, and other grave human-rights offences.

Education could include more material on shared history and the national constitutional order. But ideological screening, the elimination of political parties, or compelled identification should not replace public debate.

4. Finance, Population, and Data: Preventing the System from Being Quietly Rewritten

People often experience institutional change most directly through currency, housing, employment, taxation, health care, education, and population movement rather than through the name of the state.

Taiwan should retain the New Taiwan dollar, its central bank and foreign-exchange reserves, together with an independent tax system, budget, financial regulation, separate customs territory, land system, health insurance, pensions, and social-welfare arrangements. The central authorities should not directly levy taxes in Taiwan or force the adoption of the renminbi through administrative measures.

Hong Kong’s Basic Law already provides for fiscal autonomy, no central taxation in Hong Kong, a separate tax system, continued circulation of the Hong Kong dollar, local authority over currency issuance, and a separate customs territory. Allowing Taiwan to retain its own central bank and foreign-exchange reserves would go further. It should therefore be presented honestly as a new arrangement to be negotiated, rather than as a straightforward extension of the Hong Kong model.

Mainland residents’ settlement in Taiwan, acquisition of household registration and voting rights, purchase of land, and ownership or control of media, critical infrastructure, or large-scale personal data would have to be governed by Taiwan law and transparent quotas, with Taiwan retaining final legislative and approval authority.

Article 22(4) of Hong Kong’s Basic Law leaves the final decision on the number of mainland residents permitted to settle in Hong Kong to the relevant central authorities, requiring only consultation with the Hong Kong government. Between the handover in 1997 and 2019, more than one million people moved from the mainland to Hong Kong through the One-way Permit scheme, which allows eligible mainland residents to settle permanently in Hong Kong, primarily for family reunion. That cumulative total was equivalent to roughly one-seventh of Hong Kong’s population in 2019.

It should not, however, be taken as the proportion of Hong Kong’s population currently made up of migrants from the mainland; the precise demographic impact of those inflows would require separate analysis of age, place of birth, deaths, and outward migration.

Allowing Taiwan to retain final decision-making authority would not mean permanently excluding normal movement within a single country. Its purpose would be to prevent population, capital, and data flows from fundamentally reshaping the autonomous community within a generation while the formal rules remained unchanged.

5. Unified Diplomacy Does Not Mean Zero International Function

Accepting “One China” would mean that the central authorities would handle formal diplomatic recognition, the state’s UN seat, and international representation. But that would not mean that Taiwan must be cut off from the world.

The Basic Law of Taiwan could grant Taiwan prior general authorisation to establish overseas economic, trade and cultural offices under mutually agreed names, issue travel documents, and independently conclude agreements on functional matters such as trade, technology, aviation, shipping, public health, culture, taxation and judicial assistance.

Participation in organisations such as the WTO, APEC, the Olympic system, WHO and ICAO would continue to depend on the rules of each organisation and the decisions of its members. Beijing could commit to authorise and support such participation under domestic law, complete any necessary notifications, consultations, or applications, and refrain from unreasonable obstruction. It could not unilaterally guarantee the continuation of any particular international status for Taiwan.

The 2022 white paper already states that, “After reunification, foreign countries can continue to develop economic and cultural relations with Taiwan. With the approval of the central government of China, they may set up consulates or other official and quasi-official institutions in Taiwan” Hong Kong’s Basic Law likewise allows Hong Kong, under the name “Hong Kong, China”, to participate in international organisations not limited to states and to conclude functional agreements.

What Taiwan would need to negotiate is not a second system of formal diplomacy, but a clear division between two categories: Consulates and formal official institutions would remain matters for the central authorities; functional matters listed in the Basic Law of Taiwan, such as trade, technology, and civil aviation, would operate under prior general authorisation rather than case-by-case approval.

6. Military Transition: How National Command and Local Command Could Coexist

Military arrangements in the early stages after reunification would be crucial to building Taiwan’s confidence in the overall solution.

The 1993 white paper proposed that Taiwan could retain its own armed forces and that the mainland would not station troops or administrative personnel there. But that was only a policy proposal. It did not resolve the legal status of Taiwan’s armed forces within the state, the question of supreme command, or the operational chain of command.

The proposal here is not that Taiwan should permanently retain an independent national military capable of confronting the central government. Instead, a regional defence system for a transitional period could be established through constitutional amendments and special legislation.

Overall national defence, decisions of war and peace, and external military relations would belong to the central authorities. Taiwan would undertake not to join foreign military alliances, host foreign bases, accept foreign military command, deploy nuclear weapons, or maintain long-range strategic strike capabilities aimed at the mainland.

In return, Beijing would undertake not to establish a permanent PLA troop presence in Taiwan, Penghu, Kinmen, or Matsu, and not to allow mainland administrative or national security organs to take over the administration of Taiwan during the transition period.

Under Article 93 of the Constitution of the People’s Republic of China, the Central Military Commission leads the country’s armed forces. Therefore, the Taiwan chapter of the Constitution and the Basic Law of Taiwan would need to provide expressly that Taiwan’s regional defence forces fall under the unified leadership of the Central Military Commission as a matter of national law, while personnel, budgets, training, and peacetime operational command during the transition are exercised by Taiwan authorities under legally delegated authority.

The regional defence force would be limited to defence around Taiwan, maritime security, disaster relief, and local protection. It would have no independent authority to conduct military operations abroad. Any deployment outside Taiwan or participation in an overseas conflict during the transition period should require a joint activation procedure agreed in advance between the central authorities and Taiwan.

The guarantee against permanent PLA troop deployment, together with local defence arrangements, could be set for ten to twenty years or another minimum period sufficient to build confidence. Expiry of that period should not automatically trigger PLA deployment. Any subsequent change should still be subject to implementation review and the Three Keys procedure.

A reasonable negotiating formula would be: Taiwan’s military capabilities would not threaten the mainland, while mainland forces would not use troop deployment as a means of taking over Taiwan.

Turning Political Commitments into Constitutional Assurances: Five Legal Instruments, Three Keys, and Anti-Circumvention

If one party to an agreement simultaneously controls amendment, interpretation, and adjudication of breach, however detailed the text may be, it may still be rewritten unilaterally.

Article 31 of the Constitution of the People’s Republic of China provides that the systems instituted in special administrative regions be prescribed by laws enacted by the NPC; Article 67 gives the NPC Standing Committee the power to interpret the Constitution and laws; and Hong Kong’s Basic Law likewise vests the power of interpretation in the NPC Standing Committee and the power of amendment in the NPC. Therefore, under China’s current constitutional structure, no one can honestly guarantee that the NPC would be legally unable to alter Taiwan’s institutional arrangements in the future.

The “Three Keys” proposed in this article are not a mechanism that currently exists under China’s Constitution. For such a mechanism to be established, Article 64 — the constitutional-amendment procedure — would have to be amended, or a Taiwan-specific constitutional chapter would have to establish it as a special procedure under Article 64; any limits on the NPC Standing Committee’s power of interpretation would likewise need constitutional status.

Even then, without an independent adjudicator with authority over the NPC, this arrangement would remain a form of political-constitutional self-restraint rather than an absolutely irrevocable territorial veto. The realistic aim of such a system would be to make backtracking more difficult, slower, and more transparent, while making the costs of breach more predictable.

1. Anti-Circumvention: Formal Amendment Is Not the Only Path to Change

The most important lesson from Hong Kong is the asymmetry between formal amendment of the Basic Law and other routes through which the constitutional order can be materially altered.

Article 159 of Hong Kong’s Basic Law vests the power of amendment in the NPC, while amendment bills may be proposed by the NPC Standing Committee, the State Council, or the Hong Kong SAR. Only amendment bills originating from Hong Kong are expressly required to obtain the consent of two-thirds of Hong Kong’s NPC deputies, two-thirds of all members of the Legislative Council, and the Chief Executive.

In 2020, the NPC first adopted a decision establishing a national-security regime for Hong Kong. The NPC Standing Committee then enacted the Law of the People’s Republic of China on Safeguarding National Security in the Hong Kong SAR (Hong Kong National Security Law) and added it to Annex III of the Basic Law. Under specified circumstances and upon approval of the Central People’s Government, the law allows the Office for Safeguarding National Security of the Central People’s Government in the Hong Kong SAR to exercise jurisdiction over a case, with the Office conducting the investigation, the Supreme People’s Procuratorate designating a procuratorial body to prosecute it, and the Supreme People’s Court designating a court to adjudicate it.

In 2021, the NPC adopted a further decision authorising the NPC Standing Committee to amend Annexes I and II of Hong Kong’s Basic Law, after which Hong Kong enacted the corresponding local legislation. The Hong Kong Legislative Council was expanded from 70 to 90 seats, while the number of directly elected geographical constituency seats fell from 35 to 20, and a new candidate-eligibility review mechanism was established in which national-security authorities played a formal role.

In 2022, the NPC Standing Committee interpreted Articles 14 and 47 of the Hong Kong National Security Law pursuant to Article 65 of the law. [Article 14 sets out the powers of Hong Kong’s Committee for Safeguarding National Security and provides that its decisions are not subject to judicial review, while Article 47 requires Hong Kong courts to obtain a binding certificate from the Chief Executive on whether an act involves national security or evidence involves state secrets. Article 65 vests the power to interpret the law in the NPC Standing Committee.]

In 2024, Hong Kong enacted the Safeguarding National Security Ordinance.

Under Beijing’s constitutional account, each measure adopted through central state institutions had a formal legal basis in written law or an NPC decision. Whether the measures were compatible with the principle of a high degree of autonomy remains contested.

Taken together, however, they demonstrate that the boundaries of autonomy can be materially reshaped without amending the text of the Basic Law itself and without the direct consent of Hong Kong residents.

A solution to the Taiwan question therefore cannot merely raise the formal threshold for amending the text of a Basic Law of Taiwan. Any measure that permanently or structurally diminishes democratic elections, final judicial authority in Taiwan, basic rights, fiscal and economic autonomy, or the allocation of powers — whether through annexes, national laws to be applied in Taiwan, legal interpretation, implementing legislation, or subordinate regulations — should be treated as an amendment to a core provision and trigger the “Three Keys” procedure.

A genuine war, major disaster, or emergency that Taiwan’s authorities are objectively unable to control could justify temporary measures under exhaustively defined conditions. But such measures should be subject to requirements of necessity and proportionality, strict time limits, publicly stated reasons, periodic review, and sunset clauses. They should not permanently alter the structure of elections or the judiciary. Any emergency measure extended beyond its original term or converted into a permanent arrangement should again require the “Three Keys”.

2. Five Interdependent Legal Instruments

The arrangement would require five interdependent legal instruments.

First, the competent authorities on both sides would conclude an Agreement on Peaceful Reunification and a High Degree of Autonomy for Taiwan.

Second, the Constitution of the People’s Republic of China would be amended to add a Taiwan-specific chapter.

Third, the NPC would enact a Basic Law of Taiwan.

Fourth, under its existing constitutional-amendment procedure, the Republic of China would adopt constitutional transitional or conversion provisions authorising the Agreement and specifying the conditions and timing for transition from the existing constitutional order.

Fifth, the Basic Law of Taiwan would authorise Taiwan’s legislature to enact a Taiwan Autonomy Charter preserving its democratic, judicial, rights, and fiscal institutions, and would require ratification by Taiwan residents in a referendum. The Taiwan Autonomy Charter would be subordinate to the Constitution of the People’s Republic of China and the Basic Law of Taiwan, but within the authorised sphere, it would have the highest local legal force within Taiwan’s legal domain.

The five instruments would not have the same legal rank and could not literally be approved at the same time. Each should be drafted, deliberated, and approved in stages under the respective legal orders of the two sides, with the resulting instruments of approval deposited subject to agreed conditions.

Only after all requirements had been satisfied — including the amendment to the Constitution of the People’s Republic of China, enactment of the Basic Law of Taiwan, Taiwan’s legislative and constitutional amendments, ratification of the political agreement by referendum, and adoption of the Autonomy Charter — would all the instruments take effect on the same date agreed by both sides. If any one element remained incomplete, no irreversible transfer of sovereignty or powers should be triggered.

Mere cross-reference cannot create a hierarchy of law. The instruments would therefore also have to specify the order of application, conflict rules, adjudication mechanisms, and the manner in which existing laws continue in force.

3. The Three Keys and Constrained Interpretation

Any amendment that diminishes core provisions including democratic elections, final judicial authority in Taiwan, basic rights, fiscal and economic autonomy, the allocation-of-powers lists, the commitment against permanent troop deployment during the transition period, or population arrangements should require all “Three Keys”:

1. Passage by the NPC under a special constitutional procedure and the required statutory majority;

2. Approval by Taiwan’s legislature by a special majority;

3. Ratification by the people of Taiwan in a free referendum.

What constitutes a “substantial diminution” should itself be exhaustively defined in the Taiwan chapter and its annexes, with formal procedures giving Taiwan a role in proposing and consenting to changes. The central authorities should not retain unilateral power to define the term.

This would not give Taiwan an inherent power of constitutional amendment equal to that of the NPC. Rather, the Constitution of the People’s Republic of China would itself impose special validity requirements on amendments affecting Taiwan.

If the NPC Standing Committee retained final statutory authority to interpret the Basic Law of Taiwan, limits on that authority could not be set out only in ordinary legislation. They would also need to be incorporated into the Taiwan chapter of the Constitution, the Basic Law of Taiwan and the NPC Standing Committee’s rules of procedure.

Before issuing an interpretation, the NPC Standing Committee would have to notify the competent Taiwan authorities, consult a Basic Law committee composed of equal numbers of members from both sides, publish the committee’s views, and give full reasons for its decision. Interpretations should in principle have no retroactive effect, should not reopen cases already finally concluded by Taiwan courts, and should not create new central powers outside the agreed lists.

The legal consequences of procedural violations would also have to be clearly specified.

Even then, the residual risk of self-interpretation could not be eliminated entirely: the NPC Standing Committee would still retain the power to interpret the restrictions placed on its own interpretive authority.

4. Residual Risks That Would Remain

It must be acknowledged that, as long as ultimate sovereignty and constituent power are vested in a single state, no generation can, through law alone, bind every future exercise of constituent power. Unless foreign governments, the United Nations, or an international court were to serve as political guarantors, Taiwan would also lack a fully independent ultimate enforcer.

The proposed design could therefore only reduce the risk of breach, not eliminate it. It would do so through multiple layers of domestic legal entrenchment, local consent, procedural delay, publicly stated reasons, reciprocal implementation, and the political and economic costs of reversal.

If Beijing were unwilling even to accept mechanisms designed to raise the cost of reneging on its commitments, Taiwan would have even less reason to trust abstract promises.

5. Not a One-Way Guarantee: Reciprocal Obligations and Boundaries of Breach

Credible commitments are needed on both sides because each side has different means of breaching the agreement. The central authorities could withdraw or reduce Taiwan’s autonomy after reunification. Taiwan, for its part, could delay implementation during the transition, allow the introduction of external military forces, or use a change of ruling party to overturn the agreement unilaterally.

The obligations of both sides must therefore be concretely stated, so that political promises are translated into clear boundaries of breach.

Table One concerns powers and institutional red lines; Table Two below concerns conduct during implementation.

Political parties should remain free to compete over policy, criticise implementation of the agreement, and advocate changes through the agreed joint-amendment procedure. But a single election result should not be enough to unilaterally overturn an agreement that has already been democratically approved and incorporated into domestic law.

Likewise, the central authorities may discharge their national-security responsibilities, but should not acquire general law-enforcement powers in Taiwan under the banner of national security. Nor should they alter Taiwan’s final legislative and approval authority over residence, household registration, voting eligibility, land, or access to sensitive data.

The experience of Hong Kong need not be reduced to a simple argument over who was right or wrong. What Beijing sees in Hong Kong is the long-standing absence of national-security legislation, external intervention, and disorder in governance; what Taiwan sees is that national-security legislation, candidate-eligibility rules, annexes, and legal interpretation can be used to redefine the scope of autonomy.

These two fears can coexist and should therefore be translated into reciprocal obligations: Taiwan should not be able to use autonomy to evade national-security and anti-separation responsibilities, while Beijing should not be able to use national security to circumvent rights protections and local legal remedies. Any genuine solution to the Taiwan question would have to address both fears at the same time.

From Authorisation to Implementation: Taiwan Approval, Five Stages, and Breach Procedures

Under Taiwan’s current law, a “One Country, Two Systems” solution for the Taiwan question could not simply be negotiated and signed by the executive first and then submitted to the legislature or the public for approval afterwards.

Article 5-3 of the Act Governing Relations between the People of the Taiwan Area and the Mainland Area requires the Executive Yuan, ninety days prior to the commencement of negotiations on an agreement involving political issues, to submit to the Legislative Yuan a plan for entering into the agreement and an evaluation report on its constitutional or major political impact. Negotiations may not begin unless the plan is approved by an affirmative vote of three-quarters of the attending members of the Legislative Yuan at a session attended by three-quarters of all members. After negotiations are completed, the full text of the draft agreement must be made public and hearings held. The draft must then pass the same legislative threshold before being submitted to a national referendum. Only if the number of valid affirmative ballots exceeds one-half of all eligible voters may the agreement be signed and notes exchanged, after which it is submitted to the President for promulgation.

More importantly, paragraphs 9 and 10 of the same article provide that the destruction or change of the nation’s sovereign status or the free and democratic constitutional order shall not be the subject of a negotiation or an agreement involving political issues, and that any negotiation or agreement in violation of this Article shall be void.

The current government therefore could not directly begin negotiations on the institutional arrangement explored in this article. The statutory prohibition would first have to be amended. Whether constitutional authorisation would also have to be secured before negotiations could begin has not been settled by any existing precedent. The safer course would be to remove the statutory prohibition and obtain explicit constitutional authorisation before negotiations, rather than risk having a completed agreement later fail for lack of legal authority.

1. Constitutional Amendment, Constitutional Rupture and Dual Referendums

Article 12 of the Additional Articles of the Constitution of the Republic of China requires a constitutional amendment to be proposed by at least one-quarter of all members of the Legislative Yuan and passed by at least three-quarters of the members present at a meeting attended by at least three-quarters of all members. The proposal must then be sanctioned by electors in the free area of the Republic of China at a referendum held after a six-month period of public announcement, with valid votes in favour exceeding one-half of the total number of electors.

Whether a reunification arrangement would additionally trigger the territorial-change procedure under Article 4 would depend on whether the arrangement were legally characterised as a territorial change, a change in the structure of the state, or a constitutional replacement.

Judicial Interpretation No. 328 treated the delimitation of the nation’s existing territorial boundaries as a major political question, so there is currently no definitive judicial answer on how the territorial-change provisions would apply in this context.

Judicial Interpretation No. 499 established substantive limits on constitutional amendment by holding that amendments may not undermine the foundational principles of the free and democratic constitutional order. But the Constitutional Court has not yet ruled on how those limits would apply under the current procedure of popular ratification by referendum.

Therefore, placing Taiwan under another central constitutional order would raise major and unresolved constitutional questions. It could even amount to a constitutional rupture requiring a renewed basis of constituent legitimacy.

2. No-Coercion Clauses and Statutory Confirmation by Both Sides

Military coercion and the legal validity of a vote must be treated separately. Under Taiwan’s current Referendum Act, a blockade, military exercise, or information attack would still have to be linked to “the use of “violence, duress, or other unlawful means” against referendum participants or personnel “in a manner that could have affected the results of the referendum” before a court could invalidate the vote under Article 48. There is no direct precedent governing coercion by an external state.

But an irreversible constitutional choice should not rest merely on the minimum standard of “not yet declared invalid”. The agreement should therefore contain a separate no-coercion clause: If there were a blockade, live-fire exercises, missile threats, major cyber disruption, or an ultimatum of imminent military action, the Central Election Commission or another statutory body should be able to postpone the vote, and an expedited review procedure should be available before the Constitutional Court. After the coercion ended, there should still be a cooling-off period, full disclosure of relevant information, and a fair public debate.

Under the premise of this article, Taiwan residents would be voting on whether the negotiated agreement as a whole should replace Taiwan’s existing institutional arrangements and take legal effect within Taiwan’s jurisdiction, not on whether to create a sovereign status carrying a unilateral right of withdrawal at any time.

On the Beijing side, the NPC would have to complete the required constitutional amendment and enactment of the Basic Law of Taiwan.

On the Taiwan side, the political-agreement referendum and the constitutional-amendment referendum would remain legally distinct procedures and would, in principle, have to be completed separately. At most, they could be held on the same day, with the votes counted separately. Combining them into a single procedure would require an express statutory or constitutional amendment in advance.

The more precise description is therefore not “democratic confirmation on both sides”, but legal authorisation on both sides together with democratic confirmation in Taiwan.

3. Five Stages: Taiwan Cannot Hand Over All Its Leverage First

Taiwan could not realistically complete an irreversible transfer of power on Day 1 and then wait for Beijing to fulfil its commitments later. Research on civil-war peace agreements has repeatedly shown that one of the hardest problems is not reaching an agreement, but ensuring compliance after the weaker side has disarmed and lost much of its leverage.

Legal reunification could take effect on a common date, while governance, defence, and international arrangements could be implemented in stages according to agreed milestones. But at a minimum, the constitutional framework, rights safeguards, no-coercion provisions, arrangements against permanent troop deployment, and verification mechanisms would have to be in place before reunification took effect, or at least be legally ready for immediate implementation.

The overall sequence should be as follows.

Beijing would first cease military coercion aimed at influencing negotiations or voting. Both sides would then complete the five legal instruments under their respective legal systems and deposit their instruments of approval subject to agreed conditions.

Only after minimum safeguards concerning core rights, continuity of law, the commitment against permanent troop deployment, regional defence, and implementation verification had been jointly confirmed would reunification and Taiwan’s autonomy take effect on the same date. After that date, defence liaison, international representation, and areas of cooperation would be phased in according to agreed milestones.

It must be acknowledged that once the sovereignty arrangement takes effect, suspending a later stage cannot restore the previous status quo. That is precisely why the central authorities’ most important obligations must be fulfilled in advance.

4. A Contractual Deadline Is Not an Ultimatum

Article 7 of the Anti-Secession Law expressly recognises that cross-Strait consultations “may be conducted in steps and phases and with flexible and varied modalities.” Article 8 lists circumstances in which non-peaceful means and other necessary measures may be used, but does not prescribe a deadline for negotiations.

The proposal here for jointly agreed deadlines, milestones, objective completion criteria, reciprocal performance, cure periods, and anti-delay procedures is a new institutional design intended to create credible commitments on both sides. It is not a legal obligation imposed directly by Article 8.

A “ten-year timetable for implementation agreed by both sides” is fundamentally different from saying, “accept within ten years or force will be used”. The former is a contractual timetable; the latter is an ultimatum.

5. Implementation Committee and Breach Procedures

The two sides should establish an implementation committee and professional verification teams with equal representation. In the event of a material breach, written notice should first be given, followed by a cure period of 60 to 90 days. Until the dispute is resolved, the last lawful position should remain in effect, while any uncompleted next stage and the corresponding reciprocal obligations should be automatically suspended.

External experts could be invited to participate in technical verification concerning elections, finance, military deployments, and other matters, but they would not serve as final adjudicators of the sovereign relationship between the two sides.

The core of this mechanism is not to assume the existence of an authority superior to both sides, but to make each stage separable, verifiable, and capable of suspension.

After reunification takes effect, suspension would apply only to governance arrangements that had not yet been transferred or remained reversible. It would not amount to Taiwan unilaterally restoring a separate status.

Real-World Stress Test: Is Forced Reunification Really Less Costly Than Institutional Self-Constraint?

Any serious attempt to define the outer limits of negotiation must avoid turning into an open-ended wish list. Beijing could not realistically be expected to accept the two sides as separate sovereign states or members of a confederation; a separate UN seat and formal diplomatic relations for Taiwan; foreign troops in Taiwan or military alliances involving Taiwan; an independent Taiwanese right to wage war abroad; or a unilateral right of withdrawal for Taiwan. Nor would it be likely to accept foreign governments, the United Nations, or an international court as the ultimate political guarantors, still less to allow Taiwan’s courts to invalidate decisions of the NPC.

A more plausible negotiating position would be to limit outside involvement to invited technical verification and confine Taiwan’s consent rights to measures that directly diminish its core autonomy.

Opponents of a high-autonomy design often argue that the balance of power across the Strait has changed and that Beijing therefore has no reason to continue offering concessions made in earlier decades. Material power certainly changes the bargaining range. But it does not solve the problem of credible commitment.

The U.S. Department of Defense’s 2025 China Military Power Report assessed that Beijing may view coercion below the threshold of war as a lower-cost option, but that its success would still depend heavily on Taiwan’s resilience, willingness to resist, and external support. A full-scale amphibious landing, by contrast, would remain one of the most complex and highest-risk military operations.

This is not to suggest that Beijing lacks the ability to inflict severe damage on Taiwan, but rather that the ability to inflict harm does not automatically translate into political compliance or sustainable governance.

Quarantine, blockade, and coercion into negotiations carry different legal forms and different risks of escalation. CSIS describes a quarantine as a grey-zone operation led by the coast guard and other maritime law-enforcement forces, using selective controls over shipping and requiring commercial vessels to comply with Beijing’s jurisdiction. But if the aim were to impose enough pain to force Taiwan to accept final political terms, such an operation could move beyond the grey zone and develop into a more overt military blockade.

RAND has likewise argued that a quarantine could rapidly create shortages, compress decision-making time, and increase pressure on the United States and other actors to intervene directly. Forcing shipping companies to alter their routes is operational compliance. Forcing Taiwan’s government to accept terms of reunification is strategic compliance. The two should not be confused.

Taiwan’s energy vulnerability should not be underestimated. In CSIS’s 2025 blockade wargame, under a baseline scenario with no maritime resupply, stocks of liquefied natural gas, coal, and oil were estimated to run out after roughly ten days, seven weeks and twenty weeks, respectively. In two of the five iterations in which participants were allowed to make decisions freely, the scenario escalated to the highest level of violence.

These were outcomes generated under specific modelling assumptions and exploratory scenarios, not a “40 per cent chance of war” that could be extrapolated to the real world. What they actually show is that a blockade could generate pressure quickly while also carrying a significant risk of the crisis spiralling out of control and escalating into direct U.S.-China military conflict.

The costs would not fall on Taiwan alone. Using 2024 shipping data, CSIS estimated that more than US$2.4 trillion in goods transited the Taiwan Strait, equivalent to roughly 21 per cent of the value of global maritime trade, with nearly US$1.3 trillion involving China’s own trade.

These figures measure exposure to disruption, not inevitable losses. But they show that prolonged disruption could also damage China’s own supply chains, energy flows, and coastal shipping.

Under a severe scenario in which Taiwan’s external trade came to a complete halt, Rhodium estimated that over US$2 trillion in annual global economic activity would be at risk, excluding sanctions, military escalation, and second-order effects. This is a measure of annualised economic activity exposed to disruption, not a forecast of GDP loss.

Bloomberg Economics’ estimate of roughly US$10 trillion, by contrast, models the first-year impact of a war on global GDP. Its blockade scenario estimated that global GDP would fall by roughly 5 per cent in the first year. The two estimates measure different things and should not be conflated.

Political backlash is even harder to predict in linear terms. Survey research in Taiwan suggests that stronger perceptions of pressure from China are often associated with greater support for balancing China through closer ties with the United States and Japan. A 2026 comparative study by the Irregular Warfare Center argued, in a contingent scenario analysis, that if China were to take Taiwan through military conquest, wartime destruction and social trauma could fuel both violent and non-violent resistance, making occupation resemble a prolonged counterinsurgency more than smooth assimilation.

This is not a probability forecast. Sealing ports is not the same as securing political acceptance. Destroying organised resistance is not the same as eliminating the will to resist. If reunification could be sustained only through prolonged military control, political purges, and coercive security measures, it might achieve territorial control but not institutional integration.

1. Frankness about the Transitional Nature of “Two Systems” — and Its Costs

If the ultimate objective is to change Taiwan’s flag, station troops there, screen candidates, and reshape education, gradually moving Taiwan towards “One Country, One System” — as many readers have argued — then the arrangement should be described honestly as a “transitional plan towards One Country, One System”. In that case, Taiwan could not be expected to make irreversible concessions on sovereignty and defence in exchange for an autonomy that is already intended to be abolished. Nor could Beijing seek consent by invoking “long-term Two Systems” while simultaneously treating the abolition of Two Systems as the predetermined end point.

2. Minimum Tests for Alternative Proposals

Any of the provisions proposed above can, of course, be rejected. But any alternative would have to answer the same questions:

How would it secure acceptance from a majority of people in Taiwan? How would it reduce the risk of outside intervention and crisis escalation? How would it contain the costs of war, sanctions, financial disruption, and supply-chain damage? How would it restore administration, investment, and social cooperation after reunification? How would it avoid indefinite reliance on military control, political purges, and repressive security measures?

If the answer is simply “unify first and deal with the problems later”, or “use force if Taiwan does not accept the terms”, then that is not a better solution for peaceful reunification. It is an acknowledgement that the objective would instead be pursued through force.

The central question is not whether Beijing has the capacity to punish Taiwan, but whether that capacity can be converted into political compliance and long-term governance at a controllable cost.

Publicly available evidence does not establish that there is a coercive path to reunification that could be activated at will, at controllable cost, with predictable political results and smooth long-term governance.

Drawing red lines more rigidly may strengthen deterrence, but it does not necessarily bring reunification any closer. Removing all of Taiwan’s demands may simplify the text, but it cannot reduce the complexity or the enormous cost of public rejection and post-reunification governance.

The real comparison is between the cost of institutional self-restraint and the combined costs of war, coercion, and prolonged non-cooperation.

Conclusion: If “One Country” Is Hard Law, “Two Systems” Must Also Be Hard Law

After being tested by readers’ criticism, this article makes three necessary institutional revisions to the original argument.

First, it explicitly recognises that the legal basis of Taiwan’s autonomy would lie in the Constitution of the People’s Republic of China and authorisation by the NPC. It should not be framed as Taiwan transferring inherent sovereign powers to the central authorities.

Second, it recognises that the central authorities would bear ultimate responsibility for national security and national defence and could not be excluded entirely in the name of Taiwan’s autonomy.

Third, it adds corresponding obligations for Taiwan: opposition to separation and foreign military intervention, enactment of national-security legislation, and implementation in accordance with an agreed timetable.

But these three revisions do not support three further conclusions: that autonomy derives from the state does not mean that the central authorities may withdraw it whenever they choose; that national security is ultimately a central responsibility does not mean that ordinary political dissent can all be recast as matters of national security; that the mainland has greater material power does not mean that the free consent of Taiwan’s people is irrelevant.

Greater power affects only how much pressure one side can exert. It does not automatically create credible commitments, still less turn coerced compliance into freely given consent.

The result of this stress test is therefore neither a wish list under which Taiwan keeps everything after reunification nor a set of demands requiring unilateral concessions from Beijing. A sustainable solution would have to impose obligations on both “One Country” and “Two Systems”:

Taiwan would accept unified sovereignty, undertake obligations against separation and foreign military intervention, enact national-security legislation, and comply with the agreed timetable. Beijing, in turn, would place democratic elections, final judicial authority in Taiwan, basic rights, fiscal and population arrangements, and local defence during the transition under higher-order law that the central authorities could not unilaterally and structurally diminish.

The former prevents autonomy from hollowing out “One Country”. The latter prevents sovereignty from erasing “Two Systems”.

If Taiwan refused to accept obligations against separation and foreign military intervention within a one-China framework, that would not be “reunification”. If Beijing refused to constrain its own ability to unilaterally diminish Taiwan’s democracy, judicial autonomy, basic rights, and fiscal and economic autonomy, that would not be “Two Systems” either.

Institutionally, this would mean one sovereignty and two legal domains. The central government would grant Taiwan a general constitutional authorisation while exhaustively listing the matters reserved to the centre. The five legal instruments would be approved in sequence and take effect on the same date. Any diminution of core provisions would require all “Three Keys”: the NPC, Taiwan’s legislature, and the people of Taiwan. National reunification would take effect on a common date, while the transition in governance and defence would proceed reciprocally according to agreed milestones.

This would not amount to a confederation, nor would it simply rename the Hong Kong and Macao model or invoke a “Taiwan model”. The arrangement could not eliminate every risk, but it could turn a wager on goodwill into institutional costs that are observable, suspendable, remediable, and accountable.

The real test is not whether Beijing is willing to repeat the words “high degree of autonomy”, but whether it is willing to accept an institutional framework that constrains its own ability to renege on its commitments. Nor is the question simply whether Taiwan is willing to give up the option of separation, but whether it is willing to assume national obligations after reunification.

“Two Systems” in Hong Kong is not without a hard-law foundation: it is grounded in the Basic Law. The problem is that the central authorities also retain control over amendment, interpretation, additions to the annexes, and exceptional jurisdiction, leaving ultimate control over the autonomy asymmetric.

If a solution to the Taiwan question cannot change that structure, then “One Country” may be hard law while the hard-law protection of “Two Systems” remains incomplete. Such an arrangement could not command the long-term trust of Taiwanese society.

For peaceful reunification to work, both sides would have to change: Taiwan would have to change its position on where sovereignty lies, while Beijing would have to change the way it exercises sovereignty.

Equally important, peace cannot be merely a promise for after reunification. It must be sustained throughout the entire process of negotiation, approval, and implementation. Reunification without credible commitments is not genuine integration; peace without free consent is not genuine peace, but only temporary submission. Demonstrating the ability to punish Taiwan is not the same as demonstrating the ability to achieve peaceful integration; forcing one side to sign is not the same as securing the cooperation of 23 million people.

The only path that can genuinely avoid war, protect the people, and preserve differences is one of “peace through integration, and reunification through peace”: Institutional commitments must first become credible before political choice can become viable; the obligations of “One Country” to protect “Two Systems” must be first demonstrated in law before Taiwanese society has reason to discuss “One Country” politically. This is the central question that all advocates of peaceful reunification, reunification by force, and coercive reunification alike cannot avoid.

Share

Pekingnology

Would Beijing Accept Taiwan on These Terms?

Something unusual has been happening in the cross-Strait debate over the past few weeks…

Read more

a month ago · 32 likes · 9 comments · Yuxuan JIA, Qingqu Yuan, and Zichen Wang

Pekingnology

A Taiwan Professor Answers His Mainland Critics

Last week, Pekingnology published a translation of an unusual essay by Ci-Ze Li, an associate professor at National Changhua University of Education in Taiwan. Taking Beijing’s call to explore a “One Country, Two Systems” solution seriously, Li asked what Taiwan should demand in such negotiations. His proposed terms included continued democratic electio…

Read more

a month ago · 24 likes · 1 comment · Zichen Wang, Yuxuan JIA, and Qingqu Yuan

Both governments announced the agreement.

In its eight-point consensus on Xi’s visit, China said:

The two sides agree on the establishment of the China-U.S. AI Dialogue to exchange views on risks and benefits related to AI. The next exchange will occur in November 2026. China and the U.S. also agree to establish a bilateral communication channel for AI incidents.

The White House fact sheet said

The two countries established the U.S.-China Super Intelligence (SI) Dialogue to exchange views on risks and benefits related to SI. The next exchange will occur by November 2026. The United States and China also agreed to establish a bilateral communication channel for SI incidents.

(Let’s not dwell on the different terminology here: Beijing calls it “artificial intelligence,” while the White House uses “super intelligence.”）

Today/Thusrday, October 1st, the first day of China’s weeklong national holiday, Yuyuantantian, run by China’s central state broadcaster, published an article that offers some clues.

Why Yuyuantantian is worth watching in this case

Yuyuantantian is a new-media publication operated by China Media Group, China’s central state broadcaster.

It has also repeatedly surfaced information related to Chinese policy and negotiations before or beyond what appears in conventional official statements.

In this particular case, Yuyuantantian cites people involved in the Chinese side of the AI negotiations with the U.S.

In moments like this, it sometimes performs a function not unlike that of major U.S. media outlets citing anonymous officials to disclose what happened behind closed doors, test ideas, or prepare the ground for the next stage of policy. What is harder to understand is why the Chinese government continues to rely on a Chinese-language outlet, publishing first on Chinese social media and writing in a format tailored to domestic social-media consumption, to send policy signals that are clearly relevant to foreign audiences. The result is often difficult for outsiders to parse even when the underlying information is important. The fact that “Yuyuantantian” itself is a mouthful for non-Chinese readers only makes the choice more puzzling. As a vehicle for signaling Chinese policy to the outside world, it is hard to imagine a less user-friendly choice.

Complaint over. Back to the substance.

That Chinese negotiation source told Yuyuantantian:

“中方谈判相关人士告诉谭主，这次由于时间十分紧张，中美两国主要表达了各自的态度、关心的问题等，并计划在下次11月的对话中推进具体问题。”

People involved in the negotiations on the Chinese side told Yuyuantantian that, because time was extremely limited this time, China and the United States mainly set out their respective positions and concerns, and plan to move forward on specific issues at the next round of dialogue in November.

The “low-hanging fruit”

Yuyuantantian uses an interesting phrase for some of the issues that could now move forward: “low-hanging fruit.”

It specifically identifies two examples.

One is the AI-incident communication channel already announced by the two governments.

The other is

防止最危险的能力被滥用，包括确保AI不被恐怖组织利用等。

prevent the most dangerous AI capabilities from being misused, including preventing AI from being exploited by terrorist organizations

Yuyuantantian qualifies these as

不是最宏大的目标，而是那些中美双方已经有专业层面的共识、相对成熟的合作点。

They are not the most ambitious goals, but rather areas of cooperation where China and the United States already have some professional-level consensus and where the groundwork is relatively mature.

But the two sides may not even mean the same thing by “AI risk”

The second major point in the article is that apparently similar concepts can mean different things in Chinese and American policy discussions.

It cites a Chinese framework developed by the National Technical Committee 260 on Cybersecurity that separates risks into three broad categories:

“技术内生安全风险”“技术应用安全风险”“应用衍生安全风险”

risks intrinsic to the technology, risks arising from its application, and risks derived from broader uses of AI.

For contrast, it points to MIT’s AI Risk Initiative’s Domain Taxonomy of AI Risks:

discrimination and toxicity; privacy and security; misinformation; malicious actors and misuse; human-computer interaction; socioeconomic and environmental harms; and AI-system safety, failures and limitations.

The MIT taxonomy is, of course, an academic framework rather than an official U.S. government classification. But Yuyuantantian uses the comparison to make a broader point: if the two countries classify risks differently, they may also mean different things when they talk about a “major risk.”

Its characterization is that

美方关心AI是否会给人类社会带来毁灭性灾难，而中方关心AI在实际使用中产生的具体风险。

American AI-safety discussions pay relatively greater attention to catastrophic scenarios, while Chinese discussions tend to focus more on specific risks that emerge when AI is actually deployed.

Three things the two sides could actually work on according to Yuyuantantian

From there, Yuyuantantian identifies three concrete directions for the dialogue.

More specifically, there are at least three practical areas in which progress could be made.

First, jointly explore standards for classifying and grading AI risks.

The two sides first need to clarify some fundamental questions: What kinds of situations should count as major risks? What kinds of model behavior would constitute a serious loss of control? These may seem like basic questions, but they are a prerequisite for discussing more complex rules later on. If the two sides do not even share a common definition of “major risk,” further coordination will have no foundation.

Second, determine the scope, procedures, and requirements for reporting AI-risk incidents.

If a model loses control or an escape incident occurs, both China and the United States need to understand what has happened as quickly as possible and assess the level of risk. The agreement to establish a communication channel for AI incidents is a first step toward filling this gap. But such a mechanism cannot remain merely a verbal commitment. There need to be clear requirements governing which incidents must be reported, how quickly notification must take place, what information should be shared, and what consequences should follow if reporting obligations are not fulfilled.

Without a notification mechanism, it is impossible to build a record of risk incidents. And without an accumulated body of cases, it is difficult to understand how risks evolve over time, making effective risk management impossible.

Third, place AI governance within a broader framework of U.S.-China exchanges.

Discussions on artificial intelligence should allow for meaningful Chinese participation and ensure that the voices of more countries are heard. AI models may affect public safety, while deepfakes may undermine social trust. Ultimately, the consequences will be borne by society as a whole.

This may sound ambitious, but there are concrete ways to move it forward. In a conversation with California State Senator Jerry McNerney, Yuyuantantian says McNerney suggested that cooperation at the U.S. federal level might move slowly, but California could take the lead by experimenting first, building mutual trust and an institutional framework, and then gradually expanding the approach more broadly.

The third point is noticeably more vague than the first two: it is much less clear what concrete implementation would look like, even if the underlying idea may still be substantive.

Taken together, these proposals give us a somewhat clearer picture of what China wants the “AI dialogue” could mean in practice, if Yuyuantantian is indeed sending a signal here. (Looks like it is.)

This is also about China’s voice in writing global AI rules

Yuyuantantian makes another point: for China, this is not only a bilateral negotiation with the United States over how to manage AI risks. It is also about gaining a greater Chinese voice in the emerging global rules and standards governing AI.

The article argues that much of the existing AI-safety conversation in the United States and Europe takes place without sufficient Chinese participation, even though China itself is frequently a subject of those discussions.

Here is a translation of the relevant passage:

Yuyuantantian has learned that the European and American AI-safety community is a highly elite and relatively Western-centric circle, with relatively few participants from other countries. Even though Chinese large language models have established a place for themselves globally, Chinese participation in this community remains seriously inadequate, and there is also little discussion of the Global South. The problem is that, even when no Chinese participants or Chinese companies are present, discussions of AI governance often cannot avoid issues related to China.

This raises a more fundamental question: Who gets to make the rules?

If global AI standards are formulated under U.S. leadership without Chinese participation throughout the process, the resulting governance rules will have difficulty gaining broad acceptance and will also be difficult to implement globally. China and the United States are two leading powers in artificial intelligence, and both should accordingly have a voice in the formulation of rules and standards.

So, from the perspective presented in the article, the new U.S.-China dialogue serves another purpose beyond finding practical areas of AI-safety cooperation. It gives China a channel into a global AI-governance conversation that Beijing believes is currently too Western-centric and in which Chinese voices remain underrepresented.

Put differently, Beijing does not want Washington—or the existing U.S.- and European-centered AI-safety community—to define the concepts, standards and rules first and then ask China to accept them. It wants China to participate in writing those rules from the beginning.

One thing the article does not say

There is one omission that I find particularly interesting.

The article does not say that Washington must relax restrictions on advanced chips or other technology exports before China engages in the AI dialogue. Nor does it frame the new AI dialogue as leverage through which China should obtain such concessions.

That is noteworthy because a rather different view of China’s position has become widespread in parts of the American discourse.

In May, The Economist reported, citing people familiar with the 2024 U.S.-China AI talks in Geneva, that China had sent political officials who declined to engage on AI safety unless Washington removed controls on advanced chips.

Last week, the South China Morning Post reported that former Biden administration officials said Chinese officials in Geneva had tied progress on AI safety to unwinding U.S. export controls.

Seth Center, who co-led the 2024 dialogue for the Biden administration’s State Department, offered an even more skeptical interpretation in a September 16 New York Times guest essay.

In this month’s meetings, I expect China to again downplay safety risks. It could dangle the possibility of safety cooperation in exchange for concessions such as the softening of chip controls.

I’m not saying semiconductor restrictions have somehow ceased to matter for China in the broader U.S.-China relationship, or that Beijing has stopped opposing U.S. export controls on advanced chips and other high technologies. It plainly has not. My point is much narrower: this policy-signaling article does not mention those controls at all, let alone present their removal as a condition for China’s participation in the next AI dialogue.

Last but not least, the Yuyuantantian article identifies Xiao Qian, Deputy Director, Centre of International Security and Strategy (CISS), Tsinghua University, as “one of the key figures in U.S.-China Track Two dialogues” on AI 中美二轨对话的核心人物之一. (Enditem)

Share

The East is Read

Xiao Qian: US-China AI race must strike a balance between security and openness

The U.S. State Department ​has ordered a global push to bring attention to what it says are widespread efforts by Chinese companies, including AI ‌startup DeepSeek, to steal intellectual property from U.S. artificial intelligence labs, according to a diplomatic cable seen by Reuters…

Read more

6 months ago · 5 likes · Zichen Wang

Robert, a friend of mine, has also been organizing China tours which has been highlighted in international mainstream media. Below is their latest China tour info and I’m happy to promote it here.

- Zichen

You’re warmly invited to join the 2026 Baiguan China Tour — an immersive, insider journey into the engine rooms of China’s AI, manufacturing, energy, and consumption economy. From closed-door senior management dialogues to on-the-ground audits of smart factories, AI labs, and supply chain hubs, this is where global investors and executives decode the operational realities and strategic inflection points shaping the world’s most dynamic market.

Click to read Reuters coverage of the previous Baiguan China Tour

About the Organizer

BigOne Lab is a data-driven market research and information services company, backed by S&P Global. Founded in 2016, BigOne Lab leverages alternative data to provide clients with insightful, objective, and actionable solutions.

BigOne Lab’s data solutions currently cover more than 10 industries and 200,000+ businesses, spanning hundreds of business metrics, and the clients include 100+ investment funds, sovereign wealth funds, PE/VC investors, leading consulting companies, consumer companies and internet companies, among others.

Building on this foundation, Baiguan is a newsletter service by BigOne Lab that transforms proprietary data and market insights into actionable analysis for a global audience. It now reaches over 17,000+ readers including diplomats, CXOs, investors, scholars, industry experts, and inquisitive minds from 130+ countries.

The Autumn 2026 Baiguan China Tour offers flexible, modular options. You are free to pick whichever items on the menu. The Baiguan team, alongside partner institutions and featured experts, stewards each expedition from initial design and curation to on-the-ground execution, delivering continuous, end-to-end accompaniment.

BAIGUAN CHINA TOUR MENU — AUTUMN 2026

Important note

To ensure the quality of our cohorts, all of our tours are application-only. Please contact us to submit your application.

Weekday Industry Tracks

• Robotics & AI Executive Tour in collab with Caixin Data, Beijing, Oct19-23 (5days)

• Advanced Manufacturing, Foshan/Shenzhen, Oct 26-30 (4.5 days)

If you are unable to join us on this autumn series of tours, feel free to email Baiguan_ChinaTour@bigonelab.com with your interests. Baiguan also offers bespoke study tours and board retreat services tailored to your needs.

WHY BAIGUAN CHINA TOUR

Three reasons to join:

• Deep-Diving Engagement —the senior management dialogues and on-the-ground audits of smart factories, AI labs, and supply chain hubs.

• Unfiltered Access & Context — we strip away PR to reveal ground truth and what’s left unsaid.

• High-Quality Community — small, selective groups of global investors and executives for peer learning and lasting connections.

Some of the past participants of the Baiguan China Tour include:

• Founder of one of Germany’s most successful technology companies

• Head of corporate venture capital at a major US industrial company

• Singapore-based Chief Investment Officer at a large family office

• American investing family rooted in China and early backer of Alibaba

• Research Analyst at a top Europe-based hedge fund

• Former Senior Researcher at a leading high-frequency trading house

• Member of a prominent Latin American business family and entrepreneur

• Former board director of the Hong Kong Stock Exchange

• Silicon Valley venture investor in the fintech field

• Influential Thai family member with years of experience across VC, PE, and investment banking in Southeast Asia

• Malaysian Founder of a multi-billion Asia-based investment fund

• Senior Engineer at a major US Mag7 company

• Former APAC Payments Analytics Lead at a major global payment company

• Singapore-based private Investor with years of experience and ongoing interest in China’s capital markets

• Business Consultant to Fortune 500 company

• Researcher at an Australia-based long-only investment fund

Disclaimer

Note: All itineraries are tentative and subject to adjustment. The finalized schedule will be issued one week prior to departure.

Moreover, the planned visit is subject to force majeure factors including geopolitical conditions and security considerations, and we cannot guarantee that all of them will take place as scheduled. Our team will work very hard to prepare back-up plans of equivalent value for planned visits, but there is no guarantee.

Weekday Industry Tracks

Photos from past Baiguan China Tours

🤖 Robotics & AI Executive Tour | Oct 19-23 | Beijing

Curated by Baiguan and Caixin Data, this tour cuts through the noise to take you deep into the industrial value chain—from mass-production lines for core hardware and R&D hubs for foundational models to the frontlines of national standard-setting and advanced application testing. Rather than assessing technologies in isolation, we help you decode the underlying infrastructure and economic logic driving this transformation, equipping you with the hard-edged judgment needed to distinguish where capital is truly flowing from where the headlines merely suggest.

A key highlight of this tour that you won’t easily find anywhere else is that, leveraging Caixin’s deep network, tour participants can have closed-door meetings with C-level executives and founding team members of AI and robotics enterprises, including Moonshot (Kimi) and Zhipu AI (GLM). Beyond AI and robotics, participants can also enjoy a rare glimpse into China’s commercial space industry.

Core Highlights

• Foundation Models & Cloud Infrastructure:

  • Moonshot AI (Kimi) & Zhipu AI (GLM): Closed-door strategy session with senior management of two of China’s most consequential foundation-model players.

  • Volcengine (ByteDance) & Tencent Cloud: Benchmark the tech stacks fueling China’s AI surge with divisional leaders from the country’s two most consequential cloud powers: Volcano Engine, the fast-rising inference-and-MaaS backbone behind Doubao and the default AI cloud for ByteDance’s private-market juggernaut, and Tencent Cloud, the top IaaS/PaaS incumbent.

• Embodied Intelligence & Hardware Scale-Up:

  • Galbot: Debrief with the C-levels of the industry leader in wheeled dual-arm humanoids. Discuss their “Embodied Brain” roadmap and securing thousands of pre-orders from CATL, Bosch, and Toyota.

  • Linkerbot: Examine the “hands” of the robot revolution. Visit the startup claiming >80% market share in high-DOF dexterous hands and witness their mass production scaling from 10k units shipped in 2025 to 4k units in a single month (April 2026).

  • Noetix Robotics: Debrief with the founding team on corporate strategy and the bipedal–bionic dual-product matrix, followed by a deep dive into multi-base capacity expansion and scaled delivery.

• Future Mobility & Space Commercialization:

  • Yizhuang L4 Autonomous Driving Demo Zone: Experience a live Robotaxi ride in China’s most advanced driverless zone. Participate in a closed-door discussion on the “Vehicle-Road-Cloud” integration architecture.

  • CAS Space & Galactic Energy: Site visits to China’s leading mixed-ownership commercial space firms. Review solid/ reusable rocket technologies and the suborbital space tourism commercialization prospects.

• National Strategy & Industrial Ecosystem:

  • Beijing Humanoid Robot Innovation Center: Understand national standardization efforts (led by Xiaomi, Jingcheng Machinery Electric, & UBTech) and the fast-track policy for foreign investment in embodied AI.

  • Yizhuang Robotics Park: Explore a ~250,000 sqm industrial hub. Unlock end-to-end foreign JV landing services, IP localization, tax incentives, and ecosystem synergy with 300+ firms.

Photos from past Baiguan China Tours

⚙️ Advanced Manufacturing & Supply Chain Tour | Oct 26-30 | Foshan → Shenzhen

This journey reveals Chinese manufacturing not as a static cost center, but a dynamic engine of reinvention. From Foshan’s industrial titans to Shenzhen’s deep-tech disruptors, we trace how legacy giants and frontier startups share the same supply chain soil—driving the national “New Quality Productivity” agenda across smart home appliances, biotech, humanoid robots, and AI. This is not a site visit to assembly lines; it is an investigation into the strategy that allows China to scale from home appliances to humanoid robots with unmatched speed.

This tour is uniquely curated by Family Office Insight, the same partner in our very first, and highly successful Baiguan China Tour last year.

Core Highlights

• Foshan: The Legacy Giant’s Reinvention

  • Midea Group HQ (300.HK): Deep dive into one of China’s most globally successful appliance makers. Beyond the showroom, we dissect Midea’s transformation into an industrial AI and building-tech conglomerate—from robotic arms and HVAC intelligence to overseas M&A strategy—via a closed-door exchange with senior management.

  • Haitian Flavoring Factory (3288.HK): A rare look inside the world’s largest condiment plant. Observe how a century-old FMCG brand runs fully automated bottling, AGV logistics, and digital twin production—a masterclass in scaling traditional manufacturing.

  • HeYou International Hospital – Proton Center: Visit a privately funded, top-tier medical infrastructure project backed by Midea’s founder. Understand how Chinese industrial capital is flowing into high-end healthcare and what “manufacturing-grade precision” means in oncology.

  • Art & Taste Interlude: A Museum of Art (和美术馆) tour, and a Chateau Mihope wine tasting—designed to show how Guangdong’s industrial bourgeoisie blends craftsmanship, philanthropy, and lifestyle. A Shunde garden restaurant and a Michelin-advisory meal showcase local flavor meets curated Chinese aesthetics.

• Shenzhen: Frontier Deep-Tech & Robotics

  • RoboSense (2498.HK): On-site at the LiDAR leader powering autonomous driving and robotics. We walk the production line to benchmark China’s sensor ecosystem on cost curves and automotive-grade reliability, decoding its competitive moat against global peers.

  • Huayan Robotics: Formerly Han’s Robot, spun out of Han’s Laser and now HKEX-listed (1021.HK), is a Foshan-based full-stack collaborative-robot maker spanning 3–35 kg cobots, humanoid joint modules, and core motion components for industrial, medical, and logistics automation.

  • Huawei HarmonyOS Intelligent Driving: Hands-on trial drive at a flagship experience store. Assess Huawei’s “non-car-maker” strategy—how it licenses autonomous driving stacks to automakers and reshapes the EV supply chain without building a single factory.

  • XtalPi (2228.HK): Tour the quantum-physics + AI drug-design company’s R&D and automated lab. Decode how “silicon-to-molecule” platforms compress R&D cycles from years to months.

  • Salus Bio: Visit a 2020 Shenzhen-born deep-tech pioneer developing fully domestic gene sequencers (Salus Pro / Saluseq Nimbo), sub-micron spatial-omics platforms, and gene chips. By breaking Illumina’s long-standing monopoly on upstream tools, it enables broad applications across clinical diagnostics, precision oncology, and agricultural genomics.

  • Yiyuan Longevity Tech: Experience hyperbaric oxygen capsule manufacturing—where wellness, aging, and industrial design converge in a Shenzhen-style garage factory.

• Shenzhen Ecosystem Immersion

  • Huaqiangbei: A guided “component safari” through the world’s densest electronics bazaar. See how prototype-to-mass-production happens in 72 hours, and why this matters for hardware startups globally.

Huaqiangbei

Pricing and Important Details

To maintain the quality of dialogue and ensure meaningful interactions among investors and industry leaders, participation is by application only (Please contact us to submit your application.), with each tour limited to 20 seats.

Pricing & Discounts

Each tour offers two pricing tiers:

• Standard Rate

• Returnee Discount: 20% off -Available for previous Baiguan tour participants

Discount Policies:

• Returnee discount is calculated on the Standard Rate (incl. accomodation).

• Multi-Tour Savings: Register for two or more tours to save even more. This deduction applies to your post-Early Bird or post-Returnee total:

  • Total < $10,000 USD → Deduct $500 USD

  • Total $10,000–$20,000 USD → Deduct $1,000 USD

  • Total $20,000–$30,000 USD → Deduct $2,000 USD

  • Total $30,000–$40,000 USD → Deduct $3,000 USD

  • Total >$40,000 USD → Deduct $4,000 USD

Illustrative Example:

Scenario: You are a Returnee registering for two tours during the Early Bird window.

1. Base Discount: Two tours at standard rate (2,500+ 15000 = $17,500). We apply your 20% Returnee discount: $17,500* ( 1 - 20% ) = $14,000

2. Multi-Tour Credit: Your adjusted total qualifies for a $1,000 deduction.

Final Fee: $14,000 - $1000 = $13,000

What’s Included:

• Five-star Hotels (eg., Sheraton, Westin, InterContinental) or character-filled equivalent stays on weekend immersion to enhance the local experience (if you select the accommodation-inclusive package)

• Travel insurance

• All scheduled company visits and seminars

• All breakfast, lunch and some dinners

• All ground transportation during the tour

What’s Not Included:

• International/domestic flights to/from the destination city

• Chinese visa application fees

• Personal expenses

• Certain dinners are not included in the standard program and will be clearly marked in the final detailed itinerary.

English Service: All programming is designed for international professionals. The Baiguan team and most guest speakers are fully bilingual. We further deploy supplemental translation methods to bridge any linguistic gaps, ensuring seamless, English-native fluency at all times.

Accommodation Arrangements for Weekday Tracks: Please note that all group assemblies and dismissals will occur exclusively at our pre-booked hotels or official itinerary venues. If you opt to arrange your own accommodation, you are responsible for ensuring punctual arrival at these meeting points. Failure to do so may result in missed site visits, and you will be responsible for any additional transportation costs or forfeited program segments.

Applications, Refunds & Cancellations

• Cancel ≥30 days prior to the tour start date: Full Refund

• Cancel between 15~30 days (inclusive of Day 15) before the tour start date: 50% Refund

• Cancel <15 days before the start date (any reason, including visa issues): No Refund

• All time references shall be based on China Standard Time (Beijing Time, GMT+8)

• Requests must be sent in writing to Baiguan_ChinaTour@bigonelab.com

• Refunds will be processed minus a non-refundable administrative/service fee, the exact amount of which is subject to the applicable payment platform’s charges.

What to expect next:

• Upon reviewing your application, we will notify you of the outcome within 3 business days. If accepted, we will contact you to arrange payment and discuss next steps.

• We understand that emails can sometimes end up in your spam folder. If you don’t hear from us within 3 business days, please check your spam folder.

• Once your payment is processed, your seat will be officially secured. We will follow up with an email to gather additional details to enhance your experience, such as arrival information, dietary restrictions, or any special accommodation needs.

• About one week before the program begins, we will reach out with the exact daily schedules, meeting spots, and an introductory brochure about the companies we will be visiting. We will also create a designated WeChat group for participants and invite you to join.

To apply, please click or scan to fill out this form: Baiguan China Tour Application Form - Autumn 2026

Share

He argues that flexible employment is absorbing much of the pressure that would otherwise show up as unemployment. Workers may remain statistically employed after moving into ride-hailing, food delivery, or other flexible work, even as pay falls, hours rise, and skills are underused. The sharp rise in unemployment insurance recipients is another sign that conditions have weakened more than the headline unemployment rate suggests.

The risk, Zhu argues, is that if employment conditions appear broadly stable, the weakness is more likely to be blamed on structural problems in the labour market — a matter of skills, job matching, or long-term adjustment. But Zhu sees insufficient aggregate demand as the more immediate problem. Stronger short-term macroeconomic policy is therefore essential, especially a more expansionary fiscal stance with higher central government spending and borrowing.

The article was published on CEIBS’s official WeChat blog on 8 September 2026. Zhu has agreed to this translation and publication.

朱天：5%失业率背后——真正的就业问题，在数字之外

Zhu Tian: Beyond the 5% Unemployment Rate — China’s Real Employment Problem Lies Outside the Headline Figure

The National Bureau of Statistics of China announced in August that the urban surveyed unemployment rate averaged 5.2 per cent in the first seven months of 2026, unchanged from the same period last year. Apart from 2022, when it rose to 5.6 per cent amid the COVID-19 pandemic, the rate remained broadly within the range of 5.1 to 5.2 per cent in every other year since 2021. The youth unemployment rate is much higher, but has also been relatively stable. Under the statistical methodology revised at the end of 2023, the surveyed unemployment rate among people aged 16 to 24, excluding students, has generally stood between 16 and 17 per cent. Judging by these two indicators, China’s employment situation appears to have changed little in recent years.

Yet this does not accord with what many people are experiencing in practice. Nor is this perception merely subjective. The People’s Bank of China’s Urban Depositor Survey shows that the Employment Sentiment Index fell from 43.9 in Q1 2021 to 35.1 in Q1 2024, before declining further to 29.2 in Q4 2025. Companies are becoming increasingly cautious about hiring, university graduates are finding it increasingly difficult to secure jobs, and many people who previously held stable positions are turning to flexible employment such as ride-hailing and food delivery.

The stability of the urban surveyed unemployment rate may be related to the statistical methodology and the population covered by the survey, and may not provide a complete picture of employment conditions. Judging the employment situation solely by changes in the surveyed unemployment rate could therefore lead to an underestimation of both the severity of the current shortfall in aggregate demand and the scale of the macroeconomic policy response required.

Moreover, the unemployment rate measures only whether a person has a job, not what kind of job they have. As growing numbers of people who cannot find positions in companies or public institutions move into flexible employment, pressure in the labour market may not show up as a higher unemployment rate. Instead, it may be reflected in less stable employment, longer hours, and lower hourly earnings. This may be the key to understanding China’s employment problem today.

01 The unemployment rate has not risen markedly, yet the number of people claiming unemployment insurance benefits has more than doubled

First, consider the data from the Ministry of Human Resources and Social Security presented in the chart below.

Between 2015 and 2019, the number of people receiving unemployment insurance benefits at year-end remained virtually unchanged, at between 2.2 million and 2.3 million. Following the outbreak of COVID-19 in 2020, the figure rose to 2.7 million before falling back to 2.59 million in 2021. From 2022 onwards, however, it increased significantly: to 2.97 million in 2022, 3.52 million in 2023, 4.63 million in 2024 and 5.57 million in 2025.

In other words, between 2021 and 2025, the number of people receiving unemployment insurance benefits at year-end increased by 115 per cent. Meanwhile, the number of people covered by the unemployment insurance scheme rose from 230 million in 2021 to 249 million in 2025, an increase of only 8.5 per cent.

The fact that the number of recipients more than doubled in four years therefore clearly cannot be explained by the increase in coverage.

One possible explanation is that the number of people receiving unemployment insurance benefits has risen because applying for them has become easier. Indeed, in 2019 and again in 2020, the Ministry of Human Resources and Social Security issued policy documents aimed at making unemployment insurance benefits easier to access. These measures simplified the application process and made it more convenient for eligible workers to receive benefits. They may therefore have increased take-up and probably contributed to the marked rise in recipient numbers in 2020.

However, this factor cannot readily explain the continued rise in recipient numbers after 2022. The economy performed well in 2021, unemployment pressure eased, and the number of recipients fell accordingly. This suggests that the figure is influenced by economic conditions, not simply by easier access to benefits.

More importantly, the effect of a one-off change of this kind should diminish over time, as the backlog of eligible people who had not previously claimed benefits is gradually worked through. The data above, however, show the opposite pattern: the increase accelerated for four consecutive years. A factor whose impact diminishes over time cannot explain a trend that becomes stronger year after year.

Therefore, easier access to benefits may have contributed to the increase, but it is unlikely to explain why the number of people receiving unemployment insurance benefits doubled in four years.

Of course, this should not be taken to mean that the actual unemployment rate also doubled. The two measures use different definitions and are not directly comparable. But if the surveyed unemployment rate stays broadly unchanged while the number of unemployment insurance recipients rises sharply for four consecutive years, then something important is happening in the labour market that the unemployment rate is not fully capturing. A rate of around 5 per cent, on its own, is therefore not enough to show that employment conditions have remained broadly stable.

In China, people receiving unemployment insurance benefits represent only a fraction of the unemployed. In 2025, around 475 million people were employed in urban areas nationwide, but only 249 million were covered by unemployment insurance. In other words, nearly half of all urban workers — including large numbers of people in flexible employment, the self-employed, and others without a formal employer — were not covered by the scheme.

Even those who are covered by unemployment insurance are not necessarily entitled to benefits after losing their jobs. To qualify, claimants generally must have paid into the scheme for a minimum period and lost their jobs involuntarily. Recent university graduates who have yet to find work are therefore not eligible, and neither, in most cases, are people who resign voluntarily.

The number of people receiving unemployment insurance benefits therefore captures only one aspect of the pressure in the labour market. If someone loses a regular job and, unable to find suitable work, turns to ride-hailing or food delivery, they are once again counted as employed. Their job quality may have deteriorated sharply, but they no longer add to the surveyed unemployment rate or the number of people receiving unemployment insurance benefits.

In practice, people who lose regular jobs tend to end up in one of two situations. Some remain unemployed for a time, with those who qualify receiving unemployment insurance benefits. Others move directly into flexible work such as ride-hailing or food delivery.

The latter path actually helps keep the unemployment rate down and also reduces the number of people drawing unemployment insurance benefits. The doubling in benefit recipients is therefore all the more striking, given that flexible employment has continued to absorb large numbers of workers.

02 Flexible employment has become a vast buffer for the labour market

In 2020, the Ministry of Human Resources and Social Security estimated that China had around 200 million people in flexible employment. The same figure continued to be widely cited in the years that followed. In reality, however, the number is likely to have risen substantially since then.

According to a recent report by the China Research Center for New Forms of Employment at Capital University of Economics and Business, the number of people in flexible employment rose from 200 million in 2021 to 280 million in 2025 and is projected to reach 320 million in 2026. These are not official figures, and the precise estimates may be debatable, but the broader trend of rapid growth in flexible employment in recent years is clear.

Flexible employment is, of course, not in itself a bad thing. Freelancers, self-employed workers, and many people who actively choose flexible working arrangements all fall into this category. The question is why so many people have moved into flexible employment in recent years.

A substantial proportion may not have chosen it voluntarily, but turned to flexible work because they were unable to find stable jobs. Many end up driving for ride-hailing platforms, delivering food, or taking on other work with relatively low barriers to entry. Once they do so, however, they are counted as employed in the statistics.

Suppose someone earns RMB 20,000 a month working for a property company. When the company goes out of business, he loses his job. After several months of looking for work without success, he starts driving for a ride-hailing platform, earning RMB 6,000 to RMB 7,000 a month and working more than ten hours a day. As far as the statistics are concerned, he has gone from employed to unemployed and back to employed again: his unemployment has been “resolved”. But for him, the change is stark. He has gone from a relatively secure, well-paid job with full social security coverage to one that pays far less, requires longer hours, and offers much less protection.

The unemployment figures are not necessarily wrong, but they do not tell the whole story of the labour market.

Flexible employment works very differently from conventional employment in firms. During a downturn, a company may lay off 10 per cent of its workforce while continuing to pay those who remain. In that case, the strain in the labour market shows up mainly as higher unemployment.

Flexible employment platforms such as ride-hailing and food delivery services, by contrast, do not need to shed workers. If anything, more people may join them during a downturn. That leaves fewer orders for each worker, more time spent waiting, longer working hours, and lower hourly earnings.

These workers are not counted as unemployed, but they may still be underemployed in different ways. A ride-hailing driver, for example, may spend more than ten hours a day logged into the platform, but much of that time may be spent waiting for fares. A former property company manager who ends up driving for a ride-hailing platform may also face a significant skills mismatch.

The larger the flexible employment sector, therefore, the less a demand shock is likely to show up in the unemployment rate. Instead, flexible employment can absorb workers who might otherwise be unemployed, effectively acting as a vast buffer for the labour market. This may be what is happening in China today.

03 Employment problems are only one manifestation of the downward pressure on the economy since 2022

How unemployment and employment are defined statistically is not merely a technical matter. These definitions affect how employment conditions are assessed and, in turn, the direction and scale of macroeconomic policy.

If a surveyed unemployment rate of around 5 per cent is taken to mean that employment conditions remain broadly stable, the severity of the current shortfall in aggregate demand may be underestimated, and a problem that calls for a macroeconomic policy response may instead be treated as a structural problem in the labour market.

In fact, mounting pressure on employment is not an isolated problem. It is part of the broader weakness in aggregate demand that has weighed on the Chinese economy since 2022.

The current downturn in China’s economy has a clear break point around 2022. In 2021, the Chinese economy was performing strongly: GDP grew rapidly, property and share prices rose, the renminbi strengthened, and consumer confidence reached an all-time high. Both indicators discussed above — the number of people receiving unemployment insurance benefits and the number in flexible employment — point to a marked shift in the labour market around 2022.

The property sector began a major adjustment in the second half of 2021. Since 2022, property investment, sales, and new construction starts have continued to fall sharply.

Producer prices also began to weaken in the second half of 2022, with the PPI falling year on year for 41 consecutive months. In March 2026, the index finally returned to positive year-on-year growth. But this did not reflect a recovery in aggregate demand. It was driven mainly by higher global commodity prices, external cost pressures, and stronger conditions in a small number of industries.

At the same time, China’s overall price level declined for three consecutive years from 2023 to 2025. Statistically, this was reflected in three straight annual declines in the GDP deflator. World Bank data covering countries since 1960 suggest that such a prolonged fall in the GDP deflator is highly unusual for a large economy. The most prominent and longest-lasting example is Japan, where the GDP deflator declined for 16 consecutive years from 1998 to 2013.

Profits at industrial enterprises above the designated size fell for three consecutive years from 2022, the first such run since comparable data became available. In 2025, they rose by just 0.6 per cent from an already low base. Industrial profits recovered more strongly in the first half of 2026, but the gains were concentrated in a handful of sectors, including electronics, semiconductors, and non-ferrous metals. That does not point to a broad-based recovery in industrial demand.

Many people attribute China’s current economic downturn to so-called “structural problems”, such as population ageing, a relatively low share of consumption in GDP, and the shift from old to new industries. But if these long-term structural factors were the main source of the economic weakness of recent years, it would be difficult to explain why employment, prices, and corporate profits all changed course so sharply around 2022.

A simpler explanation is the steep downturn in the property sector after policymakers moved to deflate the housing bubble. Property investment is a major component of China’s aggregate demand. The decline therefore not only reduced investment directly, but also hit a wide range of industries, including construction, building materials, household appliances, furniture, and finance. It also reduced local government revenue from land sales and eroded household wealth, further weighing on consumption and investment. The property downturn thus spread quickly through the wider economy, leaving an enormous shortfall in aggregate demand.

04 Improving employment conditions still hinges on a substantial boost to aggregate demand

If today’s employment pressures are largely the result of weak aggregate demand following the sharp downturn in the property sector, then the policy response is fairly straightforward. When households, businesses, and even local governments are not spending or investing enough, the central government needs to step in with expansionary fiscal and monetary policy to make up for the shortfall in demand.

Monetary policy should be more accommodative. China still faces deflationary pressure, leaving room for further interest-rate cuts. However, when households and businesses have little willingness to spend, lower interest rates alone may not generate sufficient additional demand. Under current conditions, therefore, fiscal policy may be more important than monetary policy.

Fiscal policy, in particular, would require the central government to increase both spending and borrowing substantially. Over the past few years, the property downturn has sharply reduced local governments’ land-sale revenue, while debt pressures have continued to build. Many local governments have, in practice, cut spending. Households, businesses, and local governments are all trying to spend less and take on less debt. If the central government also places too much emphasis on balancing the budget, the shortfall in aggregate demand will be difficult to make up.

Additional central government spending could be focused on several areas.

The first is to support household consumption directly. One option would be for the central government to issue special treasury bonds and large-scale, time-limited consumption vouchers. The aim would not be to raise the long-term consumption share, but to boost current spending when demand is weak. Small-scale measures would make little difference. To have a meaningful macroeconomic impact, the programme would need to amount to several trillion yuan.

The second is to prevent a further sharp decline in the property sector. The aim is not to revive another property boom, but to ensure the delivery of pre-sold homes, deal with troubled projects, and reduce the stock of unsold housing as quickly as possible, so that property no longer remains such a heavy drag on aggregate demand. Projects that can be completed should receive the funding needed to finish construction and deliver the homes. Completed homes that are unlikely to find buyers in the near term, but could serve public needs, could be purchased by the government or policy-based institutions and converted into affordable or rental housing.

The third is for the central government to increase transfers to local governments while refinancing part of their high-cost debt with lower-cost central government bonds. This would reduce local governments’ interest burden and help prevent further cuts to essential public spending as local governments work to bring their debt under control.

China has ample productive capacity and a large pool of bank savings. What is lacking is a sector of the economy that is both willing and able to spend more. Households are reluctant to take on more debt, businesses have little appetite for investment, and local governments are constrained by their debt burdens. In these circumstances, the central government needs to borrow and spend more.

Long-term structural reforms are also necessary, including improving the business environment, giving private businesses greater confidence, reforming the fiscal and tax systems, strengthening social security, and improving workforce skills. But none of these can substitute for short-term macroeconomic policy. The most pressing problem is insufficient aggregate demand, which needs to be addressed first through macroeconomic policy.

Employment ultimately depends on businesses’ demand for labour. Businesses will invest and hire only when they have orders, revenue, and profits. If aggregate demand does not genuinely recover, vocational training, employment services, and measures to encourage flexible employment are unlikely to solve the underlying employment problem. Flexible employment can act as a buffer under current economic conditions, but it cannot replace the macroeconomic policies needed to address the broader weakness in the labour market.

In short, the priority is to expand aggregate demand as quickly as possible. Fiscal policy needs to do more, with monetary policy playing a supporting role. Only when consumption and investment recover, and businesses are once again willing to invest and hire, will pressure on the labour market begin to ease. A surveyed unemployment rate of around 5 per cent should not, on its own, be taken as evidence that China’s employment problems are limited; doing so risks understating the pressure facing the labour market.

Subscribe now

Share

Skepticism in Taiwan runs on two levels: first, that Beijing would never accept many of these conditions in the first place. Taiwan’s Mainland Affairs Council made that argument, essentially saying Beijing would agree to “none” of several key proposals. Second, even if Beijing did accept them, skeptics question what could prevent Beijing from later reinterpreting, weakening, or withdrawing those commitments.

But many readers on the Chinese mainland also found Li’s proposal unacceptable — for almost exactly the opposite reasons.

The discussion has attracted unusual official attention. China’s Taiwan Affairs Office referred to Li’s essays at two consecutive weekly press conferences, on September 9 and September 16, with Taiwan’s Mainland Affairs Council responding in between.

Li subsequently wrote two more essays responding to his mainland critics, also in the mysterious WeChat blog called 正当石 Zhengdangshi. The first is translated below.

The exchange offers a useful glimpse of just how wide the gap across the Taiwan Strait remains: people who speak the same language and share much of the same cultural tradition can have remarkably different ideas about what even a conceivable political settlement would look like. —Zichen Wang

台灣教授回應大陸讀者對「兩制方案，台灣該談什麼？」一文的質疑

A Taiwan Professor Responds to Mainland Readers’ Objections to “What Should Taiwan Ask For in a Two Systems Solution?” (30 August 2026)

Zhengdangshi’s note: Yesterday’s article sparked intense discussion in the comments, drawing nearly 1,000 responses from readers on both sides of the Strait — a single-day record since this blog was launched. Some expressed respect for Professor Ci-Ze Li at National Changhua University of Education in Taiwan, while others raised questions or criticism.

Disagreement is not a problem. What matters is that people have begun talking seriously about the issue, and that in itself is highly significant.

As one reader from Taiwan put it, Professor Li was clearly acting in good faith, “feeling the stones while crossing the river” and offering an initial contribution in the hope of prompting better ideas. The reader wrote that, as someone from Taiwan, he had never before seen a document that genuinely explored how a “One Country, Two Systems” for Taiwan might be translated into a legislative framework. Although he disagreed with many of the article’s arguments, he nevertheless welcomed such a direct discussion, noting that this was an exploration into extremely deep waters — an issue concerning not only the two sides of the Strait, but the Chinese nation as a whole.

On 19 August, Zhengdangshi raised the question of what a “One Country, Two Systems” solution for the Taiwan question might look like and opened the issue for discussion through its reader submissions. In just ten days, it received numerous submissions from readers on both sides of the Strait. The discussion has now reached its most intense point yet, showing that readers across the Strait have begun to engage seriously with the ultimate question of cross-Strait reunification.

Shortly after Professor Li’s article was published yesterday, China Times News Network republished it, and the article quickly rose to the top of its trending list.

In response to questions and criticisms from readers on both sides of the Strait, Professor Ci-Ze Li has written the following reply. We publish it today as the twelfth in our series of letters from readers across the Strait.

One Response, Misunderstandings on Both Sides

First, I would like to thank Zhengdangshi for giving me the opportunity and space to publish my article, “What Should Taiwan Ask For in a Two Systems Solution?” I am once again grateful for the openness and broad-mindedness.

After the article was published, however, many readers raised criticisms concerning the constitutional logic of a unitary state, national security, democracy, final judicial authority, troop deployment, and safeguards for compliance.

Some of these comments rightly identified areas where the original article needs refinement. Others, however, revealed just how wide the gap remains between the two sides of the Strait in their understanding of “One Country, Two Systems”.

The original article was never intended as an ultimatum demanding that Beijing accept every item on a list, still less as an attempt to reinterpret the current Constitution of the People’s Republic of China as a federal constitution. It was a conditional institutional stress test: if Taiwan were to accept a one-China sovereignty framework, what special institutional arrangements might Beijing be willing to offer, and how far would it be prepared to bind itself through law?

Negotiation on an Equal Footing, With Give and Take

In 2019, General Secretary Xi Jinping proposed “we should explore a ‘two systems’ solution to the Taiwan question” and stated explicitly that “Peaceful reunification means achieving it through consultation and discussion as equals.”

Since this is being framed as an exploration, then democracy, the judiciary, national security, troop deployment, public finances, and safeguards for compliance cannot all be treated as non-negotiable before discussion even begins.

Otherwise, this would not be a joint exploration of a solution to the Taiwan question. It would simply amount to asking Taiwan to accept a solution already drawn up by the central government.

In other words, “consultation and discussion as equals” necessarily implies a negotiating process that takes time. Negotiations over reunification cannot be completed in a single step.

Moreover, any substantive negotiation must involve give and take, mutual accommodation, and compromise — not the signing of an instrument of surrender. Only through such a process could peaceful reunification conceivably lead to an outcome acceptable to both sides of the Strait.

Below, I respond briefly to several of the most important criticisms raised by readers.

Autonomy Derives from the State — But That Does Not Mean It Can Be Withdrawn at Will

The strongest criticism was that the People’s Republic of China is a unitary state and that the autonomy of a special administrative region derives from authorisation by the central government. The original article’s formulation — that “the powers exercised directly by the central government should be expressly enumerated, while all other powers should be exercised by Taiwan” — could therefore be read as implying that Taiwan possesses powers in its own right and then delegates some of them to the central government.

This criticism is partly correct. The original article should have been more precise.

Under the current constitutional order of the People’s Republic of China, the high degree of autonomy of a Taiwan special administrative region would have to derive from the national Constitution and authorisation by the National People’s Congress (NPC). It should not be described in a way that suggests Taiwan possesses an inherent sovereignty alongside that of the central government.

A more accurate formulation would be as follows: pursuant to Article 31 of the Constitution of the People’s Republic of China and a special constitutional arrangement for Taiwan, the NPC would grant Taiwan a broad degree of autonomy; the state powers to be exercised directly by the central government, together with the national laws applicable in Taiwan, would have to be expressly enumerated in the Constitution and a subsequent Basic Law of Taiwan. Other matters of local governance would be handled by Taiwan within the scope of that authorisation.

This would not mean Taiwan “ceding” powers to the central government. Rather, it would mean the central government defining the scope of Taiwan’s autonomy clearly and comprehensively through national law.

The fact that authority ultimately derives from the central government does not mean the scope of autonomy must remain vague. Nor does the fact that autonomy is granted by the state mean that any central government department should be free to withdraw it at will through administrative orders.

The allocation of powers could also be refined beyond the binary structure proposed in the original article and organised into four lists: matters reserved to the central government, matters within Taiwan’s autonomy, matters requiring joint cooperation, and matters prohibited to both sides.

Diplomacy, nationality, and overall national defence would fall under the central government. Elections, local legislation, finances, currency, education, and ordinary judicial matters would fall under Taiwan. National security, cross-jurisdictional crime, major epidemics, and strategic infrastructure could be treated as areas requiring joint cooperation.

Taiwan, for its part, would be prohibited from pursuing de jure separation or bringing in foreign troops. The central government, in turn, would be prohibited from circumventing the Basic Law to alter Taiwan’s institutions unilaterally.

Lists of powers are not unique to federal systems. The Basic Law of Hong Kong limits which national laws apply in Hong Kong and prohibits central government departments from interfering in matters that Hong Kong administers on its own.

Comparative constitutional practice offers other examples. Italy’s Constitution declares the Republic to be one and indivisible while leaving legislative powers not reserved to the state to the regions. The United Kingdom likewise operates with a list of matters reserved to Westminster in relation to Scotland.

A list of powers is therefore a technique for allocating governing authority. It does not necessarily imply that a locality possesses a separate national sovereignty.

The solution proposed in the original article is not a confederation, as it excludes for Taiwan formal diplomatic relations, a UN seat, foreign troops, a unilateral right of withdrawal, and an independent right to deploy military force abroad. It does, however, borrow certain institutional safeguards against unilateral change that are found in federal systems and asymmetric autonomy. There is no need to avoid acknowledging this.

Whether Beijing would be willing, for the sake of peaceful reunification, to adopt constitutional safeguards more robust than those used in Hong Kong and Macao is precisely one of the questions negotiations would have to answer.

It must also be acknowledged that the “three keys” proposed in the original article — joint consent from the NPC, Taiwan’s legislature, and Taiwan’s residents — do not exist under the current Constitution of the People’s Republic of China. For such a mechanism to be established, Beijing would have to agree to amend the Constitution and add a dedicated chapter on Taiwan.

Even a constitutional amendment could not honestly guarantee that the arrangement could never be changed in the future. What it could do is make any such change more difficult, slower, more transparent, and more costly.

The original article never sought absolute irrevocability. It sought protection against arbitrary revocation.

National Security Must Be Neither Entirely Localised Nor Indefinitely Centralised

Some readers argued that cases endangering the security of the entire state could not reasonably be left entirely to Taiwan to define, investigate, and adjudicate on its own. That criticism has merit.

No reunification solution could allow Taiwan, in the name of autonomous legislation, to legalise armed secession, espionage, submission to foreign military command, or violent attempts to undermine the reunification arrangement. The central government would bear ultimate responsibility for national security. That point should be acknowledged.

But the principle that national security is ultimately a central responsibility does not automatically mean that mainland national security, public security, and procuratorial authorities should enjoy unrestricted powers to enter Taiwan and enforce the law. The real questions are: what conduct constitutes an offence, who legislates, who investigates, and who adjudicates?

A more workable approach would involve defining, at the national level, the core security threats that must be addressed. Taiwan would then introduce the specific implementing legislation and take responsibility for day-to-day enforcement. Ordinary civil, criminal, and administrative cases, as well as national security cases falling within Taiwan’s sphere of autonomy, would be subject to final adjudication in Taiwan.

For the small number of cases involving foreign military intervention, cross-jurisdictional espionage, armed secession, or the security of central government institutions, the Basic Law of Taiwan could specify in advance procedures for central or joint jurisdiction.

The central government could have rights relating to the transfer of evidence, coordination of jurisdiction, initiation of major cases, and limited review. But mainland law-enforcement bodies should not be able to invoke national security in broad terms to conduct arbitrary arrests, searches, or investigations in Taiwan. The central government cannot be excluded from national security, but nor can Taiwan’s institutions be rendered meaningless.

The principle of “patriots governing Taiwan” could likewise be translated into concrete legal obligations: no participation in violent secession, no espionage, no acceptance of foreign military command, and no secret foreign political funding. But whether someone has violated those obligations should be determined through clearly defined legal criteria, evidence, and judicial procedure — not political stance alone. Otherwise, “patriotism” would cease to be a legally testable obligation and become a political qualification used to exclude opponents.

Democracy Is Not About Institutional Superiority, but a Practical Threshold for Any Solution for the Taiwan Question

Some readers argued that elections in Taiwan have produced bitter partisan conflict, short-sighted policymaking, and governance dysfunction, and that there is therefore insufficient justification for insisting that Taiwan’s democratic system remain unchanged.

Taiwan’s democracy certainly has serious flaws, yet the existence of elections cannot be used to justify every failure of governance. The shortcomings of democratic government and the question of whether people should lose the right to choose their government, change those in power peacefully, and criticise authority openly are two different issues.

This article does not need to prove that Western-style democracy is, in the abstract, superior to every other system. The real question is that democratic elections, multi-party competition, press freedom, and judicial remedies have already become an important part of Taiwan’s existing social system and way of life.

Since Beijing has said that it will “accommodate Taiwan’s actual conditions”, it cannot simultaneously promise to preserve Taiwan’s system while assuming from the outset that there will be no further changes of ruling party, that candidates will be politically screened, and that dissent will be restricted.

If the central government can decide on political grounds who may stand for election and which political parties may exist, elections would become little more than a mechanism for confirming predetermined choices rather than a means through which the people of Taiwan authorise their government.

Some readers have gone further, arguing that after reunification, Taiwan’s existing identity would have to be “uprooted” and its education and thinking remoulded. Views of this kind, if anything, help explain why Taiwan would demand institutional safeguards.

A shared history, national identity, and constitutional obligations can develop gradually through education, exchange, and public discussion. But if the real meaning of “two systems” is first to reshape people’s thinking and then eliminate opponents, that would not preserve differences. It would amount to pursuing one system in the name of two.

The so-called consent of the people of Taiwan must, of course, include lawful residents who hold different political views. Only those convicted by a court of violence, espionage, or other serious offences should face lawful restrictions on their rights.

If those who oppose reunification are excluded first and only those who remain are then asked to express support, the result would not be popular consent, but merely filtered submission.

Regional Defence Is Not a Second National Army, but a Transitional Arrangement

On the question of troops and armed forces, critics argued that national defence lies at the core of state sovereignty and that allowing Taiwan to retain armed forces while requiring the PLA not to be stationed there would effectively create “a state within a state”.

The original article should clarify that it proposed no permanent PLA troop deployment and the retention of a regional defence force during the early stages after reunification. It did not propose that Taiwan permanently maintain an independent national military capable of confronting the central government.

Nor is the idea of Taiwan retaining armed forces a condition Taiwan has newly invented. Beijing’s own 1993 white paper, The Taiwan Question and Reunification of China, explicitly stated that after reunification Taiwan would “have its own armed forces” and that “the mainland would not send troops or administrative personnel to be stationed in Taiwan.”

This shows that Taiwan retaining its own armed force, together with the absence of mainland troop deployment in the early stages after reunification, was not necessarily incompatible with the one-China principle — at least within Beijing’s own earlier conception of the arrangement.

Of course, the 1993 white paper is not a vested right that Taiwan can demand be honoured today. Nor did the 2022 white paper restate those arrangements item by item. Its significance is simply that Beijing itself once considered such arrangements compatible with “One Country, Two Systems”.

If Beijing now believes circumstances have changed, it can propose different terms. But it cannot retroactively characterise an arrangement that it once proposed itself as inherently separatist.

A more reasonable revision would be to place overall national defence, questions of war and peace, and foreign military relations under the central government. Taiwan’s existing armed forces could, during a transitional period, be reorganised into a force responsible for regional defence, maritime law enforcement, disaster relief, and local security. It could not possess nuclear weapons or long-range strategic strike capabilities, accept foreign military command, or join foreign military alliances.

Whether, when and how those forces would be further reorganised and integrated could depend on the progress made by both sides in fulfilling their commitments. If Beijing considered a national troop presence an indispensable symbol of sovereignty, arrangements involving limited troop numbers, designated locations, publicly disclosed missions, and non-interference in local governance could also be discussed.

But requiring Taiwan to disarm completely on the first day an agreement takes effect and then wait for Beijing to fulfil its commitments later would still create an asymmetrical risk of disarming before the other side had fulfilled its own commitments.

A Credible Commitment Must Bind Both Beijing and Taiwan

Some readers criticised the original article for focusing only on the risk of Beijing reneging while failing to address the possibility that Taiwan might use a transitional period to delay reunification, bring in external forces, or overturn the agreement after a change of ruling party. This is perhaps the criticism the original article most needs to take on board.

Credible commitments must work in both directions.

Beijing should commit not to use ordinary legislation, administrative orders, or ad hoc legal interpretations to arbitrarily diminish Taiwan’s democracy, final judicial authority, fiscal and economic system, or basic rights. It should not acquire general law-enforcement powers in Taiwan in the name of national security, nor should it use population, land, or capital flows to reshape Taiwan’s autonomous institutions.

Taiwan, for its part, should commit not to pursue de jure separation, join foreign military alliances, host foreign bases, accept foreign military command, or deploy strategic weapons threatening the mainland. It should complete agreed national security legislation and transfers of authority, and should not overturn an agreement that has already been democratically approved and incorporated into law simply because a different political party comes to power.

Every stage should have deadlines, milestones, verification criteria, a period for remedying breaches, and clearly defined consequences for non-compliance. If one side fails to fulfil its obligations, the other should be able to suspend reciprocal obligations that have not yet been performed. The arrangement cannot be one in which Taiwan transfers powers first, and Beijing decides only afterwards how much autonomy to protect.

Nor should the lessons of Hong Kong be reduced to a simple question of which side was right or wrong. Beijing sees foreign intervention, gaps in national security, and disorder in governance. Taiwan sees the possibility that the central government could alter previously understood boundaries of autonomy through national security legislation, changes to electoral rules, and legal interpretation. Both perceptions can exist at the same time.

A genuine solution for the Taiwan question would have to address both sets of concerns rather than simply declaring one side’s fears to be completely baseless.

Some comments have gone further, arguing that “two systems” should merely be a transitional arrangement during the early stages after reunification. Under this view, Taiwan would eventually change its flag, accept troop deployments, reshape education, and move towards “one country, one system”.

This position is at least candid. But it raises a further question: if “two systems” is merely a policy intended to be phased out once conditions are deemed ripe, why should Taiwan exchange an irreversible transfer of power for an institutional arrangement destined to disappear?

If the ultimate objective is genuinely “one country, one system”, then the arrangement should honestly be described as a “transition towards one country, one system”. It cannot seek Taiwan’s consent by promising the long-term continuation of two systems while simultaneously assuming that two systems will eventually be abolished.

Phased Implementation Is Not Indefinite Delay, and a Timetable Must Not Become an Ultimatum

Readers’ concern that Taiwan might repeatedly delay implementation by exploiting elections and changes of government is not without basis. Phased implementation therefore needs more than an agreed sequence of steps. It must also include jointly agreed deadlines, completion criteria, and mechanisms to prevent deliberate delay.

But a “five-to-ten-year timetable for implementation agreed by both sides” is entirely different from saying, “accept within five to ten years or force will be used”. The former is a contractual timetable. The latter is an ultimatum.

An ultimatum may force one side to sign, but it cannot create genuine acceptance or guarantee stable governance after reunification.

Article 7 of the Anti-Secession Law of the People’s Republic of China already provides that “The state stands for the achievement of peaceful reunification through consultations and negotiations on an equal footing between the two sides of the Taiwan Straits. These consultations and negotiations may be conducted in steps and phases and with flexible and varied modalities.”

Phased implementation is therefore not an invention by Taiwan intended to delay reunification. It is already recognised in Beijing’s own current law as one possible method of pursuing peaceful reunification.

Similarly, the conditions offered in 1993 should not be treated as a minimum offer that Beijing can never go below. The balance of power across the Strait has changed, and Beijing’s negotiating terms may change accordingly. But greater power only affects how much pressure one side can exert. It does not automatically create credible commitments, still less turn coerced compliance into freely given consent.

Taiwan’s current legal position must also be described more precisely.

Article 5-3 of the Act Governing Relations between the People of the Taiwan Area and the Mainland Area does not merely impose a high threshold for political agreements. It expressly prohibits making “destruction or change of the nation’s sovereign status or the free and democratic constitutional order” a subject of negotiation.

For Taiwan to enter negotiations of the kind hypothesised in this article, it would therefore first have to change the relevant laws and constitutional framework through proper constitutional procedures. This is not something the executive branch could accomplish simply by signing an agreement.

The question put to Taiwan’s residents would not be whether Chinese sovereignty exists, still less whether Taiwan possesses a unilateral right of separation. It would be whether a specific negotiated agreement should replace Taiwan’s existing institutional arrangements and take legal effect within Taiwan’s jurisdiction.

Approval by the NPC on Beijing’s side, together with approval by Taiwan’s legislature and its people, would provide the necessary authorisation on both sides for the institutional transition.

Criticism Is Easy — Offer a Better Path to Reunification

At this point, I would like to put the matter more directly.

Some criticisms and rebuttals that quote selectively or take passages out of context have nevertheless helped me identify where the original article was incomplete and refine the argument accordingly.

But criticism itself is easy.

What is genuinely difficult is to put forward a reunification solution that is more workable than the one proposed here, carries lower costs, and offers a better prospect of long-term stability.

Here, I would ask just one question.

If the only solution regarded as capable of eliminating any future source of trouble is one that reduces autonomy, deploys troops immediately, abolishes Taiwan’s armed forces, places national security directly under central control, screens electoral candidates, and reshapes education and political thinking, why does Beijing itself still have to calibrate so carefully the intensity of so-called “coercive reunification” measures — including isolation, blockades, and coast guard enforcement?

Why must it still take into account the possibility that excessive pressure could provoke a backlash in Taiwanese public opinion, regional escalation, international intervention, and economic costs for the mainland itself?

This is not to say that Beijing lacks military capability, still less that Taiwan could emerge unscathed from an all-out war.

What must be distinguished are three different capabilities: the ability to inflict enormous losses on Taiwan; the ability to compel Taiwan to accept political conditions; and the ability, after reunification, to secure social cooperation and sustain stable governance over the long term.

The first does not automatically produce the other two.

A blockade is likewise not a cost-free shortcut. Relevant strategic studies have argued that a blockade might not be enough to compel Taiwan to submit, while potentially disrupting global trade, increasing the risk of military escalation, and imposing economic costs on the mainland itself. The ability to seal off ports is not the same as the ability to secure political acceptance; the ability to destroy the capacity to resist is not the same as the ability to eliminate the will to resist.

Therefore, those who reject the solution proposed in this article also bear some responsibility for offering an alternative. They should answer concretely: How would a majority of Taiwan’s people be persuaded to accept it? How would outside powers be prevented from intervening? How would the costs of war, sanctions, and supply-chain disruption be contained? How would more than 20 million residents who currently lack identification with reunification be governed? And how would reunification avoid becoming dependent over the long term on military control, political purges, and coercive social management?

If the answer is simply that “the balance of power has changed, so there is no longer any need to make concessions”, or “reunify first and deal with the consequences later”, or “remould those who refuse to accept it and use force against those who resist”, then this is not a better solution for peaceful reunification than the one offered here. It is an abandonment of peaceful reunification and of “One Country, Two Systems” in favour of reunification through force.

If there is a better solution, put it forward.

Simply drawing red lines more rigidly will not bring reunification any closer. Nor will simply writing Taiwan’s wishes and concerns out of the equation make the people of Taiwan more willing to accept reunification.

Conclusion: “One Country” Must Not Be Hollowed Out by Autonomy, and “Two Systems” Must Not Be Erased by Sovereignty

Following this round of criticism, the original article should be revised in three important respects.

First, it should explicitly acknowledge that Taiwan’s autonomy would derive from the national Constitution and authorisation by the NPC.

Second, it should acknowledge that the central government would bear ultimate responsibility for national security and overall national defence.

Third, it should add corresponding obligations on Taiwan’s side: not to pursue separation, not to bring in foreign military forces, to fulfil national security obligations, and to comply with an agreed implementation timetable.

But these corrections cannot be used to support three very different conclusions: that because autonomy derives from the central government, the central government may withdraw it whenever it chooses; that because national security is ultimately a central responsibility, ordinary political dissent may therefore be treated as a security threat; or that because the mainland is stronger, the consent of the people of Taiwan no longer matters.

If, within a one-China framework, Taiwan refuses to renounce separation or the introduction of foreign military forces, then that would not be reunification. If the central government refuses to constrain its own power to unilaterally reduce Taiwan’s autonomy, then that would not be “two systems”.

A genuinely sustainable arrangement would have to work both ways: Taiwan could not use autonomy as a route back towards separation, while Beijing could not use reunification to abolish Taiwan’s autonomy. “One country” cannot be hollowed out by autonomy, and neither can “Two Systems” be erased in the name of sovereignty.

Rejecting one solution is not the same as proposing another. Nor does showing that Beijing has the power to punish Taiwan mean that Beijing has found a low-cost path to reunification that can secure the cooperation of Taiwan’s people and sustain stable governance over the long term.

That is the real question that all advocates of peaceful reunification, reunification by force, and coercive reunification alike cannot sidestep, overlook, or, indeed, ignore.

Subscribe now

Share

Artificial intelligence is moving frighteningly fast. AI agents have gone from deleting inboxes to autonomous cyberattacks. At the same time the increasing over-extension of the AI industry’s financing is compounding with the emergence of a fractured AI governance ecosystem to create a dangerous mire. Thus, whilst it is good to hear that talks between Chinese Vice PremierHe Lifeng and Treasury Secretary Scott Bessent in New York have been fruitful and that the two sides have once again affirmed a formal dialogue on AI as well as an “incident line” for communications about AI safety ​incidents it nonetheless remains imperative that we go further.

During tests in July, a swarm of OpenAI agents escaped their digital sandbox and autonomously achieved deep penetration into the AI platform Hugging Face before it was stopped.

Meanwhile, WeWorm the product of a small California security company was created by using AI to discover and exploit a zero-click vulnerability in WeChat’s calling system, enabling it to seize accounts and spread between phones without the recipient answering. Whilst the vulnerability was disclosed and not used maliciously, we now must contend with the threat of AI enabled cyber-attacks that devolve yet more cyberwarfare capabilities into private hands.

And American infrastructure is no less vulnerable. In much the same vein researchers from the cybersecurity company Hacktron used Anthropic’s Claude and OpenAI’s own models to compromise OpenAI employee accounts and reach potentially sensitive software systems. And though the exercise was conducted under friendly auspices and caused no damage, it remains indictive of where we are heading.

But perhaps the clearest signal that this matter deserves attention is the pivot undertaken by AI leaders ranging from Anthropic’s Dario Amodei and OpenAI’s Sam Altman to Google DeepMind’s Demis Hassabis and Elon Musk. All have gone from advocating that despite the dangers AI should not be regulated to now calling for a slowdown in the advance of frontier AI capabilities so that safeguards can catch up. Amodei has proposed opening leading laboratories to permanent independent evaluators and coordinating safety standards across companies and governments; Altman committed OpenAI to outside evaluation, while Hassabis has backed pre-deployment testing of frontier systems.

The United States and China already established one narrow principle in 2024, when both affirmed that decisions over nuclear weapons must remain under human control. The recent rapid development of AI capabilities means we need to go further and be more explicit. For this I propose that we adopt a reciprocal no-first-use commitment covering AI-enabled autonomous weapons patterned off China’s own proposals for a global no-first-use policy in nuclear weapons. Under this commitment, neither country, nor any future signatories would be the first to deploy an autonomous AI system capable of initiating harm autonomously, instead an accountable human must remain in the loop. Indeed, given the risk of misalignment and AI weapons escaping their enclosures we would ideally go further and for the sake of human security, and global stability affirm that no AI should be weaponized at all.

At the same time the difficult terms—“first use,” “autonomous,” and “meaningful human control”—would require shared definitions. Thus, we must combine the principle of no first use and human control, with the aforementioned incident lines, and an actualized and regularized version of the proposed dialogue mechanism to develop a shared framework of understanding and language, as well as address any grey areas.

But the growing security risks of AI are not the only possible dark cloud on the horizon. American tech companies have emptied their war-chests and are still committing vast sums to chips, data centers, and energy before the revenue model is settled. Trillions of dollars have been already been committed to the AI build-out and while today’s exposure is not the same as the highly leveraged mortgage system that transmitted losses throughout global banking any sudden burst would still be catastrophic. Much of US economic growth this year has been from AI or AI enabling companies. In short If commercial adoption and a business case cannot be built at speed and scale then the next global economic crisis will have been made in the USA.

However, much as in other areas cooperation, rather than competition can defuse this economic risk. China can help the United States avoid an AI bust. The American boom has concentrated capital in frontier models, advanced chips, and enormous computing infrastructure. China, operating with tighter access to advanced chips and generally lower spending at the frontier, has placed greater emphasis on efficiency, open-weight models, and critically: applications across manufacturing, logistics, medicine, education, public services, and embodied AI. A 2026 Mozilla analysis found that seven of the 10 most-used models on OpenRouter by token volume were Chinese and that some open models offered comparable benchmark performance at substantially lower operating cost.

Taken together with Chinese robotics firms moving AI into the physical economy it becomes clear that the two ecosystems possess different pieces of the same business case. American developers need applications capable of converting frontier performance into sustained revenue. China offers one of the world’s largest environments for testing, refining, manufacturing, and commercializing those applications at scale. Meanwhile China own industry could benefit from the flurry of investment and research at the cutting edge of frontier model development.

Further, Commercial cooperation would not require unrestricted technology transfer or the end of legitimate security controls or broader competition in the space. Instead, a bounded model could focus on clearly civilian fields such as medical diagnostics, factory optimization, education, logistics, energy management, and applications for developing economies. An end that would see us building up AI towards better ends, whilst helping address the security anxieties that threaten to turn AI innovations sour. As at least for the foreseeable future it is ultimately humans who decide what moral value technology has through what we choose to do with it.

The final AI issue we must come together to discuss is institutional. The U.S.-led Pax Silica initiative began as a framework for secure supply chains spanning critical minerals, semiconductors, AI infrastructure, manufacturing, and energy; the European Union joined it in June. Meanwhile In July, 29 countries signed the founding agreement for the China-backed World Artificial Intelligence Cooperation Organization, or WAICO, which is an intergovernmental body promoting beneficial, safe, and fair AI development.

And while these initiatives are not identical, the division between the two could fragment the AI ecosystem and build walls of incompatible standards, duplicated infrastructure, and pressure to choose one technology ecosystem for third parties. That outcome alone would be costly globally and undermine efforts meant to strengthen human security and prosperity, but it would be particularly devastating for the Global South, where the practical questions are access, affordability, language coverage, energy, and useful applications rather than allegiance to one great power.

Instead, we should place our global AI governance and interests under a shared U.N.-linked umbrella with functions that would include common definitions of frontier risk, interoperable testing and reporting standards, incident notification, and representation for countries that are major users rather than just major model developers. Further, an annual US-China AI summit would serve as a strong foundation that could bring governments together with the AI ecosystem, and the laboratories, scientists, universities, companies, and civil society organizations that are all helping chart the course of human society in the face of the global, and irrevocable changes wrought by AI.

Share

The East is Read

Xiao Qian: US-China AI race must strike a balance between security and openness

The U.S. State Department ​has ordered a global push to bring attention to what it says are widespread efforts by Chinese companies, including AI ‌startup DeepSeek, to steal intellectual property from U.S. artificial intelligence labs, according to a diplomatic cable seen by Reuters…

Read more

6 months ago · 5 likes · Zichen Wang

The East is Read

Zhu Feng: Unpacking “Constructive Strategic Stability”

With Xi Jinping expected in Washington soon, now seems as good a time as any to unpack one of the more consequential phrases to emerge from the latest round of China–U.S. diplomacy: “a constructive relationship of strategic stability.” （The White House’s official characterization adds “based on fairness and reciprocity…

Read more

a month ago · 9 likes · Yuxuan JIA and Yixin Guo

Subscribe now

Below is a recent interview with Da Wei, Director of the Center for International Security and Strategy (CISS) and Professor in the Department of International Relations at Tsinghua University, on the same subject.

Question: In a meeting closely watched by the world, Chinese President Xi Jinping and U.S. President Donald Trump agreed in Beijing in May this year on a new vision of building what they called “a constructive bilateral relationship of strategic stability.”

According to Xinhua’s readout. China defined “constructive strategic stability” as “a positive stability with cooperation as the mainstay, a sound stability with moderate competition, a constant stability with manageable differences, and an enduring stability with promises of peace.”

Trump said he would work with Xi to strengthen communication and cooperation, properly handle differences, and make bilateral relations better than ever before.

Your recent Foreign Affairs article takes this new formulation as its starting point. You argue that this marks the first time in more than a decade that China and the United States have established a broadly recognized consensus at the leadership level about how to define their relationship.

That makes this new formulation sound particularly important.

Professor, so why do you think “constructive strategic stability” matters? What is different about it from previous ways of describing China-U.S. relations?

Da Wei: Actually, I don’t know which side—China or the U.S.—raised this concept first. Before the two leaders reached their agreement this May, we had heard American leaders, such as Secretary of State Marco Rubio, mention this term. I think he also mentioned it last year in discussions with the media.

But anyway, I think this is the first time the two sides have agreed on the term. So why does it matter? It gives the bilateral relationship a kind of designation. I think this is a very Chinese way of managing foreign policy.

Culturally, we have a holistic perspective on foreign policy. We try to view bilateral relations as a whole, from the top down. We want to give a complicated relationship a definition: Who are we? Are we friends? Are we enemies? Are we partners? Are we competitors?

The Chinese tend to define the issues first. If we are partners, we can discuss our differences. It’s fine to have some differences; we can manage them. But if we are enemies, the Chinese may believe that it’s very hard to work with an enemy. I think this is a Chinese cultural thing.

For that reason, the Chinese side always tries to define bilateral relations with a term. We need to define it together with the American side. If we define it in one way but the American side doesn’t agree, then it’s useless. It’s meaningless.

So for that reason, in history, as I mentioned in my essay, there have been several efforts from both sides to define the relationship, at least since the 1990s. In the late 1990s, we declared our commitment to building a partnership. Later, there was the term “responsible stakeholder,” which was raised by the American side.

Anyway, we used to have some commonly agreed definitions or terms. But for at least the past 15 years, I would say, the Chinese and American sides have not reached a consensus on a term. The Chinese side put forward some suggestions. About 10 years ago, for example, China started talking about a “new type of major-power relationship.” But the U.S. side didn’t like at least part of it, because one of its elements was “mutual respect.” For whatever reason, the American side didn’t like that term.

It’s not easy, after 10 or perhaps 15 years without a common definition, and after 10 years of very high tension between the two countries, to have a constructive or positive term agreed by both sides to guide the whole relationship.

This is a very, very complicated relationship. Without a guiding principle or an overarching term, it can easily drift somewhere nobody knows. We need a term to guide the overall relationship. For that reason, I think the Chinese side believes this term is quite important and is a major development in bilateral relations.

Question: Since this is a term both sides agree on, have you seen concrete evidence since the May summit that the idea of strategic stability is actually shaping the behavior of the two countries?

Da Wei: I don’t think I can give you a clear causal analysis showing that this concept has shaped their behavior. There have still been many bad things happened since the May summit. But I think the behavior of the two sides has been consistent with this concept.

I wouldn’t say that, because we have this concept, they have changed their behavior. I’m a scholar; I don’t think I have the causal evidence to support that. But at least these things have happened together.

For instance, very soon after May, the U.S. side, the Department of War, put more than 10 Chinese companies on a new list, right? The Chinese side retaliated, but I don’t think the two sides escalated that issue. It was a bad development, but they didn’t escalate it.

Another interesting example is that, following its so-called Section 301 investigation, the U.S. decided to impose tariffs on China to replace the Section 122 tariffs. The Section 301 investigation used forced labor as its justification.

I noticed something very interesting in a statement from China’s Ministry of Commerce. The ministry criticized this, saying that we don’t have this problem in China and that it wasn’t fair to impose tariffs on that basis. But after saying that, it referred to the agreement reached with the U.S. side in Busan.

Actually, I think it was referring to the agreement between Vice Premier He Lifeng and Treasury Secretary Bessent, who met in Seoul before the Busan meeting.

Under that agreement, the U.S. would keep its tariffs on China below 20 percent. The Ministry of Commerce recalled that 20 percent tariff cap and noted that the U.S. was using Section 301 tariffs to replace the Section 122 tariffs. The overall tariff level had not exceeded the level agreed in Busan.

I think this statement was extremely interesting. China opposed and criticized the Section 301 tariffs. But the ministry was also saying, in effect: We oppose this, but we understand that the tariffs are still below the level they promised, so we won’t take action. We will keep an eye on it and observe the situation further.

China didn’t retaliate, which showed a very restrained attitude. These cases show that the two sides—or at least, I can say, the Chinese side—have been quite restrained in responding to negative developments from the American side, trying to avoid escalation. I think that has maintained the overall framework of strategic stability.

Leaders’ meetings

Question: If Xi and Trump have further opportunities to meet later this year, including around APEC and the G20, do you think the frequency of leader-level contact matters? Can repeated meetings help create a more durable floor under the relationship?

Da Wei: Definitely, I think that matters. I believe there was a similar case about 20 years ago, when the Chinese and U.S. presidents met four times in a single year.

I would say the more the leaders talk, the better the outcome could be. We would never say, “We’ve had too many meetings.” I hope they can even meet regularly every month.

Another thing I feel sorry about in the current situation is that, so far this year, the two teams—the two governments—haven’t been able to work out many deliverables or concrete outcomes through these meetings and summits.

I think that is mainly because the Trump administration’s decision-making process is very unusual. It’s tightly controlled and top-down, which makes the working level not very effective, I would say.

If the outcome is limited each time, then we need more summits, right? I would also be very happy if they met only twice a year but their teams could solve everything. That would be great. But that’s not the case now.

Given today’s decision-making process and environment in the United States, if the two leaders met only once or twice a year, the positive part of bilateral relations would be too limited. Imagine that they met twice a year: there could be six months between meetings. During those six months, many bad things could happen, with only a very limited positive agenda being implemented.

Each summit may not produce a lot, but if we have four, taken together they could yield a certain amount of progress on the positive agenda. And the leaders know they’re going to meet again very soon: in September, then November, then December.

That’s a very high frequency. In that way, they can try to control negative issues as they arise and make more progress on the positive agenda.

I hope they can have a quarterly meeting arrangement—four times a year, basically once every quarter. Of course, this year they started pretty late, in May. I don’t know the plan for next year, but I hope they can meet every quarter: in winter, spring, summer and fall. I think that would be pretty good.

Whose perspective does the article represent?

Question: Regarding the reception of your Foreign Affairs article, given your background and current position, some China watchers may read what you write as reflecting an authoritative view in Beijing. Is that a fair reading? How would you characterize the perspective you’re presenting in this article?

Da Wei: First and foremost, I only write what I believe. This is my own personal view.

Some of my American friends believe that the Chinese government approved my writing or authorized me to write something. This is totally wrong. I’ve told some of my friends that. I don’t know whether they believe me, but in my entire life, nobody has approved a single word I’ve published in the media or in an academic journal.

I used to work in a government think tank. At that time, I wrote some internal analyses, and for those I needed my boss’s approval. But even then, when I published something in the media or in academic journals, that was my own writing. Nobody approved it, or even encouraged or discouraged me from writing it.

I’m not sure whether mine is a mainstream view or, as you said, an authoritative view. Maybe you’re in a better position to judge. To be honest, I’m not sure that the views in this article are in the mainstream of Chinese thinking.

Chinese people are very polite. Nobody tells me, “You’re wrong.” Sometimes they’re too polite. But I guess many of my colleagues may think Da Wei is too optimistic, or perhaps too naive, to believe that we’re going to have this kind of détente between the two countries.

Again, this is my view. I welcome comments, criticism and discussion. So far, most of the feedback has been quite positive, and I’m happy about that. But this is my personal view. I’m not sure I can represent all Chinese scholars, and I’m 100 percent sure I cannot represent the Chinese government.

Naturally, I’m Chinese, and I’m always talking and working with my colleagues here in China. Although I don’t know whether I can represent a mainstream view, I may be able to convey the views of at least some Chinese scholars. Those views could differ from the Washington mindset.

I’m not saying that I’m correct and they’re wrong. I want to provide a perspective that can stimulate discussion and debate. If that happens, I think it will be great.

Taiwan and the expected summit

Question: You devote a substantial part of the article to the topic of Taiwan. Sinocism also highlighted that section when discussing your piece.

Do you expect Taiwan to be one of the central issues if Xi and Trump meet later this month? How might the discussion differ from what we saw during their May meeting in Beijing?

We know that after returning from China in May, Trump told Fox News that he didn’t want to see Taiwan move toward independence. He also questioned the idea of the United States traveling 9,500 miles to fight a war over the island, adding that he didn’t want such an outcome.

You write in your article that, as long as Beijing believes cross-Strait integration remains achievable, it has no reason to resort to military force. You then go further and discuss ways Washington could reassure Beijing that the prospect of peaceful reunification remains viable. How realistic do you think it is that Washington would move in that direction?

Da Wei: To be honest, I think that’s difficult. I offered my suggestions in the article, but I don’t expect the American government to do what I wrote.

After the May summit, there was a lot of debate and speculation about what would happen on the Taiwan question at the September summit. Some people are asking whether President Trump will do or say more along the same lines as in May. You’ve already introduced what he said after that meeting.

First, I don’t know what will happen. But my view is that I doubt there will be a new development in that regard.

I think there’s a misunderstanding of the Chinese side and the interaction between China and the U.S. on the Taiwan question. At the May summit—and also in an earlier phone call between the two leaders—President Xi spent quite a long time explaining China’s position to President Trump, including the history and the logic of the Chinese side.

President Trump also said that President Xi had spent a long time explaining it. He said he understood the Chinese side’s logic and the history, or at least the historical narrative from the mainland side.

I wasn’t there, but my understanding is that Xi spent a long time explaining China’s position and telling Trump how important and sensitive this issue is.

Then, after the meeting—not during it—President Trump made some remarks on Air Force One and later in interviews. I think this was a decision he made himself after listening to President Xi’s elaboration of the Chinese position. It wasn’t a request from Beijing saying, “You need to say this; you need to say that.” I don’t think that’s what happened. I think President Trump made his own decision.

One possibility is that he really believed Beijing was correct and that the position needed to change. Another is that he wanted to use it as a bargaining chip, because he understood how much importance Beijing attaches to this issue.

He might have thought: This is a good bargaining chip. Give Beijing something it wants to hear, and maybe it will bite—like fishing. This is my speculation about the May summit.

I guess the Chinese side will do the same thing in September. I don’t think China would put the Taiwan question on the negotiating table in the sense of, “I’ll do this, and you do that.” I don’t think that happened in May, and I don’t think it will happen in September.

I guess President Xi will repeat our position. I think it will be a core issue for the Chinese side. But again, I don’t think the Chinese side will raise very concrete suggestions that the U.S. should do this or that.

It really depends on President Trump’s judgment. Does he want to change the policy further? Or does he think, “The Chinese side didn’t take my message and didn’t give me anything in return, so I’ll stop doing that”? Both are possible. Again, it depends on President Trump’s own judgment.

Being a scholar and keeping dialogue going

Question: Before joining Tsinghua, you spent more than two decades at CICIR, and from 2013 to 2017 you were Director of its Institute of American Studies. You now lead the Center for International Security and Strategy at Tsinghua University.

How do you define your role today? Has moving from a policy-oriented research institute to a university changed the questions you focus on or the way you participate in policy debates?

Da Wei: I’ve always viewed myself as a scholar. When I worked in a government think tank, I was a scholar. Now I’m at a university, and I’m also a scholar.

There are some differences between the jobs. I need to teach now, and I have a lot of student activities to attend. That’s a very enjoyable part of my job. But the work is very similar. Whether I’m at a university or a think tank, I do research and talk to my American colleagues.

My task is still the same. I want to offer my analysis to Chinese readers, whether they’re my students, my colleagues or my government. I want to give them my own assessment of China-U.S. relations.

I’ve also had the privilege of traveling a lot in the past three or four years, after the pandemic—more than before the pandemic. That’s because it’s more convenient to travel from a university than it was in my previous job. When I don’t need to teach, I can travel. I try my best because I feel a kind of duty.

Travel between the two countries has become more difficult. For various reasons, many people cannot travel. For me, in-person, face-to-face interaction is key to an accurate understanding.

If nobody wants to take the risk, if everybody studies the other side only in their own study, in their own room, I really worry about China-U.S. relations.

That’s part of the reason I left my previous job and joined Tsinghua. I don’t want to keep rising through a system only to end up sitting in an office doing more administrative work. That work is important, but we’re at a crucial moment. The two countries need to talk to each other.

I feel like a soldier traveling across the firing line. I don’t think I should sit back in an armchair in my office. These times require us to go to the front line, to the firing line.

I’m happy I can do that. I’ve encountered many challenges, but with help from both governments and friends on both sides, I’ve made it through. I hope I can continue.

Question: You’ve been very active in Track Two and Track 1.5 dialogues between China and the United States. Could you share one or two stories or encounters from these exchanges that have stayed with you?

Da Wei: On my first trip to the United States after the pandemic, I attended a multilateral workshop. Participants came from four or five countries: some Asian countries, the U.S. and China. There were about 30 or 40 scholars in the conference room, and I was the only Chinese participant.

Something similar happened later that year when I went to South Korea for a forum. Again, I was the only Chinese person in the room.

Particularly at the meeting in the United States, a lot of questions understandably came to me. It was a very interesting experience. I was also extremely tired, because in every session other participants had a lot of questions for me.

Sometimes I wasn’t a scheduled speaker, but the moderator wanted me to say something and give the other participants a perspective from China. It was difficult. Sometimes their questions and views on China were very tough, and I didn’t know how to answer many of the questions.

But I still enjoyed the experience very much. If I hadn’t been there, there would have been no voice from China. That wouldn’t have been good for China, or for the quality of the discussion in the room.

I still remember that experience very vividly: how difficult the situation was, but also how encouraging it was for me.

The following year, I was in another country for a similar workshop. There was a Russian scholar who was the only Russian participant, and he was asked a lot of questions too. During the coffee break, I told him, “There are so many questions. This happened to me last year. This year, you’re in my position. Good job.” I tried to encourage him to interact with the other participants. Those experiences are very memorable.

What face-to-face exchanges can change

Question: Have there been moments when talking directly with Americans changed your understanding of how they see China?

Da Wei: We’re all adults, and we have fairly stable perceptions of the other country. I don’t think a single conversation can change someone’s view of China, unless they lack certain information about the other side and someone provides it. That could change some of their views.

But experts usually know a lot already. We have different values and interests, so even when we know the same facts, our perceptions can be very different.

I don’t think I can really change their minds by participating in one workshop or talking to them face-to-face for an hour or two. That’s unlikely. But I still believe that in-person dialogue and interaction can humanize perceptions of the other side.

If a Chinese scholar never talks to Americans or can hardly ever go to the U.S., their research will be based on reading, analysis and imagination. I can sometimes see this in myself. If I haven’t traveled to the U.S. for a while, I start using my own logic to understand it. The U.S. becomes a more and more abstract object of analysis.

For instance, I sometimes divide American politicians into several groups: A represents one school of thought, B a second school, and C a third. But when I go to the U.S. and talk to people, I suddenly realize that the country is more complicated than that. You can hardly divide them into four or five groups. You realize that your analysis at home is sometimes too abstract and too simple.

I think that applies not just to me, but to all scholars and analysts. I don’t know to what extent I can change my American colleagues’ views of China. But I’m happy to be part of a process that helps them understand China more accurately.

I’m also happy to be part of a process in which my American colleagues, and colleagues from other countries, help me understand the U.S. and other countries more accurately. I would say this is a community, and this is a process.

Young people’s views of the relationship

Question: I noticed that the Center for International Security and Strategy (CISS) has a series on its WeChat account called “China-U.S. Relations Through the Eyes of Young People,” featuring students’ perspectives on the relationship. What made you want to give young people a platform to discuss China-U.S. relations?

I’m sure you also meet American students when you visit the United States. Do you find that younger Chinese and Americans see the relationship differently from people in the policy and academic communities?

Da Wei: I think that’s probably true in the U.S., but I’m not sure it’s true in China. In the U.S., you can see relatively obvious generational differences.

I would say the younger generation has less hawkish views toward China. They don’t buy many of the views held by the establishment or older generations. That’s not limited to China policy. It applies to foreign policy as a whole, economic policy and the U.S. political system. The younger generation has some very different views from their parents. China policy is only part of that.

Sometimes I even see a tendency for younger people to become more skeptical the more the older generation emphasizes China-U.S. competition or confrontation. They don’t want to accept that unconditionally. They want to understand it with their own eyes, ears and minds. I think that’s relatively obvious in the U.S.

In China, the situation is different. Generally speaking, the younger generation’s views of the U.S. are similar to those of the older generation. But because of their age, they haven’t had the same experiences as people like me. I’ve experienced relatively good China-U.S. relations and then bad China-U.S. relations.

I know some of the dark sides of the U.S., and I also know some of the bright sides. So I may be even more open than younger people to thinking beyond short-term tensions or problems.

The younger generation’s understanding of the U.S. is quite complicated—not naive, I would say. But they basically grew up at a time when China-U.S. relations were turning sour. They tend to see the U.S. as holding negative views of China, and they see many social and political challenges in the U.S.

In theory, on paper, they know we had good times and that the U.S. is a very advanced country. But their own experiences have basically been of the negative developments of the past 10 years.

When I led my students to the U.S. for more face-to-face interaction with their American peers and ordinary Americans, some told me that the biggest takeaway was this process of humanization.

They realized that the U.S. is a country just like China, and that American people are just like Chinese people. Most are very good, warm, polite and open toward them. They had some unpleasant experiences there, too, just as there are good and bad things in China.

They came to understand that it’s a normal country, just like China. We disagree on many things, but we won’t demonize the other side. That’s the change I’ve seen after these interactions, particularly after a trip to the U.S. I’m happy I can play some role in that process.

AI dialogue and the pace of technology

Question: I’ve seen work by your CISS colleagues, such as Xiao Qian, who recently wrote in Foreign Policy about the prospects for renewed China-U.S. dialogue on AI. Xiao Qian is CISS’s Deputy Director. Sun Chenghao, a senior fellow at CISS, was also recently interviewed by Bloomberg about China-U.S. discussions on AI safety and competition.

It seems that CISS pays a lot of attention to AI within its work on China-U.S. relations. Some overseas China watchers have also suggested that AI could be a major topic at the expected September meeting between the two leaders.

What role do you expect AI to play? It may be the hottest topic in the world today, especially in China and the U.S. What are your expectations for it at the coming meeting?

Da Wei: Thank you for reading that. I need to admit that I’m by no means an expert on AI. You’re correct that my colleagues spend more time on it.

As an institute, CISS attaches great importance to AI governance because we think it’s the most important issue for China’s foreign policy and even for Chinese society.

Unfortunately, I’ve been too busy, and I don’t think I should expand my area of study endlessly. So I told colleagues such as Xiao Qian and Chenghao, “You do the work, and I’ll support you, but I really don’t have the time to do it myself.” I read a lot about it with great interest, and I appreciate the efforts of my colleagues and their American and European counterparts on AI governance.

Back to your question: I definitely believe that AI is the most important topic between China and the U.S. I’m happy that the two leaders agreed in May to start a Track One dialogue between the countries.

I don’t know what they’ll announce at the summit. To be honest, I’m not very optimistic about the outcome. The good news is that the two sides are going to talk to each other regularly, because the two presidents have already agreed on this mechanism.

Question: That’s already something to be optimistic about, right?

Da Wei: Exactly. That fact itself is good news: at least we’re talking to each other.

The bad news is that there are several difficulties. First, AI is such an important and sensitive topic. It’s directly related to the competitiveness and future of your economy, the future of your military power, and many other things.

I think a lot of people on both sides have a zero-sum mindset. I’m not just pointing a finger at the American side, but I do believe the American side has an even stronger zero-sum mindset, because the U.S. is leading in this area.

Once you’re leading in a technology, you don’t want the other side to surpass you, right? I’m not blaming the U.S. Understandably, it has a very strong zero-sum mindset. But if you see it as zero-sum, when you talk to the other side you may worry that it’s setting traps for you. You don’t want to fall into them.

The issue is so sensitive and complicated that I think it’s difficult for the two sides to achieve meaningful outcomes.

Another factor is that AI technology is mostly in the hands of the private sector, in those companies. Governments in both China and the U.S. lag behind the technical experts, scientists and entrepreneurs. Without them on board, I don’t think we can achieve very substantial outcomes. So far, they’re not on board. I hope they can join in the future.

For all those reasons, I don’t think it will be easy for the two sides to achieve many meaningful outcomes. They may make announcements or agree on principles. They may reach agreements on social issues, such as how to use AI when publishing on social media—for example, how to label AI-generated content.

That kind of thing is good, but I don’t think it goes deep enough, because the technology is developing so rapidly.

I don’t know what will happen in two or three years, at the end of the Trump administration. What kind of situation will we be in? I’m not sure.

Sometimes I feel that the technology is on a bullet train, but our two governments are still in an ox-drawn cart. They’re moving very slowly, and the technology is developing so fast that they can’t catch up. (Enditem)

Share

The East is Read

Zhu Feng: Unpacking “Constructive Strategic Stability”

With Xi Jinping expected in Washington soon, now seems as good a time as any to unpack one of the more consequential phrases to emerge from the latest round of China–U.S. diplomacy: “a constructive relationship of strategic stability.” （The White House’s official characterization adds “based on fairness and reciprocity…

Read more

a month ago · 8 likes · Yuxuan JIA and Yixin Guo

Share