China, AI, and Assessing Policy Trends:
An interview with Dr. Ryan Manuel

Dr. Ryan Manuel is the founder and managing director of Bilby, a technology company that uses machine learning and big data to read and analyze Chinese policies and regulations. He was previously the Chief Asia Strategist at Silverhorn Investment Advisors, and an academic at Oxford University, the Australian National University, and Hong Kong University. Earlier in his career, Ryan was a Senior China Analyst with the Australian government and a management consultant with the Boston Consulting Group. He holds a doctorate from Oxford University, where he was a Rhodes Scholar.

Q1. Please introduce yourself and tell us about your work and interests.

I run Bilby, a startup that converts what governments tell their own bureaucrats into trading and reporting signals for financial institutions. I also teach Chinese politics.

Q2. For those who not familiar with China’s government, can you give a brief explanation of how the CCP governs the country and what are the mechanisms through which it exerts control?

There are two main mechanisms: the first is recognizable to observers of Western politics, which uses governance techniques such as policies, regulations, laws, etc. The other is through Chinese Communist Party actions. These are focused on individuals and ideological functions. In other words, you can get people to do what you want through giving them a state and government framework, or through training them all as individuals to think a certain way. In the former, you change policies; the latter, minds.

Q3. How are orders communicated from the top down to the provincial and local levels?

Slowly. The centre issues signals that are passed down level-to-level with some tweaks as they go down. (China has five formal levels of government, and an informal level called the village; most of the action happens at the county, which is the fourth level down). Then the various actors affected respond with their specifications of what they are doing to make these changes flesh. Similar to call-response in music, each time a signal is issued you see different reactions to it. There is a system of approval built in, too. Should the centre like a reaction from the lower level, they can re-issue it, telling everyone that this is more the sort of thing it is after.

Q4. Experimentation seems to be a big part of how the CCP tries out new policies. Can you talk a bit about this and cite a few recent examples?

Experimentation is a part of this call-response logic; when Beijing wishes to try something out, or when there is a desire from a lower-level leader to catch the eye of those above, then a pilot is created. There are a couple hundred of these running at any given time, covering everything from eye surgery to the digital renminbi. They act as good early signals of what the nation will be doing in future years, as well as showing how more vague central orders are being interpreted by those in charge of making them happen. Some recent examples include the demonstration pilot for improving the circular economy in urban districts (园区循环化改造示范试点) and the “Program for Strengthening National Transport” (交通强国建设试点).

Q5. Does the CCP have different messaging domestically than internationally and what do you think is the impact of this when it relates to foreign policy or the ability of foreigners to understand the country?

Hugely different. It is often less a matter of messaging than of incentives. Chinese officials, even those in charge of international messaging, care first about the local impact and then the national impact of whatever they need to say. International considerations are the remit almost solely of Xi Jinping. Foreign policy is written for domestic Chinese ears as much as for foreign consumption. A good example was the lauded Belt and Road policy, in which overseas investments were not considered suitable by Chinese local leaders, and hence despite being Xi’s international priority policy, investment to Belt and Road countries actually fell year on year for three consecutive years when Chinese outbound investment was overall increasing. The reason is that the economic incentives of a project mattered more than the foreign policy implications, hence local leaders didn’t like BRI projects. Instead of BRI leading to more investment, it often had the reverse effect.

This “think local, talk global” paradigm hinders foreign observers, who are more used to thinking of China having some sort of nefarious intent or grand plan behind their actions; rather, it is more likely to be a clumsy, large country fumbling around in the dark looking for the light switch and breaking something in the process. This doesn’t change the fact that things have been broken, and this is ultimately a bad outcome.

Q6. There is a belief that decision-making in China is opaque and that it is impossible to pierce the veil on issues of any importance. Do you believe this and if not, what are the signals or clues you look for when assessing a new policy or trend?

There are 96 million Party members in China. If decision making is opaque, they are unable to know what to think or do on issues of any importance. So, it is clearly possible to pierce that veil; the question is how. The signals and clues I look for are what is initially said (as in the language used); who said it, and whether or not it is an area of policy in which they have decision making mandates. The forum used to issue the policy or statement is also important, as is the stage of the policy lifecycle the statement exists at and if it is an early or a late signal. I also look at how many other parts of the system are discussing it or quoting it; and finally, what is the level of specificity of the policy.

Q7. How much Chinese policy is reactive and made quickly in response to new challenges or opportunities, and how much is baked in from decisions made months or even years in advance? Is policy flexibility built into the system or is governance more inflexible and rigid?

This is very hard to assess without first knowing which political level we are discussing. At the central level, change is very slow and baked in over years. At the local level, you have a tremendous amount of discretion in what you choose to prioritise — there are nearly 3000 counties, whose leaders are judged on roughly 2000 measurements annually; how on earth can they know what to do?! — and that means that there is considerable flexibility if you have cash in the budget.

Q8. How well does the Chinese leadership understand the outside world, and the West in particular? Within the senior ranks of the leadership, is there much intuitive understanding of the outside world or how China’s actions might be perceived by specific countries or segments of the global population?

I haven’t met all that many senior Chinese leaders personally, and even should I have, I would need hundreds of interactions to form that sort of judgment. So, I confine my comments to what they write about or say in official media. They appear to have an excellent understanding of the formal nature of Western systems, of the rank and hierarchy, and of what things look like on paper. Their intuitive grasp of the informal is limited, in the classic mirroring problem of “if I were X of course I would do Y”, where Y is dependent on history, language, culture etc.

When dealing with overseas reactions to events, as before, I’m not sure they really care as much how well matters are received overseas. There are 1.4 billion critics that they have to deal with every day, not to mention the Party members they are meant to be instructing. The domestic ramifications will nearly always trump everything else. This is different for Xi, who as the head of the system is the person to be held accountable for failures on the international stage. He has a more interesting set of calculations to perform as to which envoy might send what message, and how much of the credit or blame he might receive.

Q9. When analyzing China, what role do you think technology (data analysis, machine learning, etc.) can play and how do you think these tools can help augment intuition and experience?

I’m biased, as I left full-time academia solely to develop the tools you are asking me about. So, take the following with a hefty shakerful of salt. That said: China is a nearly perfect use case for supervised machine learning and big data analysis. All of the formal rules about messaging, communications, methods, rank, etc. that I discussed above need to be followed in order for the system to work. And, absent computers, the scale of the country is far too large for any analyst or team of analysts to make assertions that hold outside very small N studies. Added to these tools there needs to be a deep feel or judgment, and sophisticated communication skills, that only humans can provide. Where I’m most excited is how we can bring intuition and experienced judgment into artificial intelligence models in order to try and scale them. Rather than looking at one or two documents or having a couple of conversations, and making empirical generalisations from them, we can now initiate the same types of hypotheses and test them across thousands of counties, decades of data, and millions of documents.

Q10. Domestically in China, the use of technology in daily life has become ubiquitous (payments, gaming, social media, streaming, etc.). The government has sought to reign in the power of the big tech companies and control both usage and behavior. Why do you think they do this?

Because they also use the big tech companies to govern, and indeed, it would be very hard to run China without them. Xi’s signature program, his greatest achievement according to his own speeches, is an extraordinary household transfer policy designed to end abject poverty. The cash was mostly funded and nearly solely delivered by Tencent and Alibaba using mobile phones. And for all the ink spilled discussing the woes of China’s tech giants, their founders remain wealthy and at liberty. China is a nicer and easier place to live in because of their products. That means that the government would like to use them to govern (and yes, that will often include to surveil, but that is not the only reason). This push-pull dynamic of wanting to gain the benefits of the technology but also wanting to control it is not unique to China.

Q11. We often get seduced by the belief that new technology will change or fundamentally alter human nature or social behavior. What are your thoughts on this? How much of today’s China is still very much a reflection of its exceedingly long history and culture, complete with all the dynastic dramas and Confucian/Daoist/Buddhist legacies?

I don’t think the tech giant products discussed above have altered Chinese society; they have just brought everything together into one app and saved time. The so-called China Dream (note: a concept first discussed in Chinese culture by Chiang Kaishek, and with far more grounding in US culture) does not seem affected by technology as much as by fairly quotidian aspirations for a better life, with some deep historical and cultural affectations. These themselves are rarely changed by leaders — for all the talk of their power, Mao’s and Xi’s constant complaints of people not doing what they want are shared by nearly all Chinese emperors — and instead are the accretion of family and local values. These more parochial concerns are such a broad church that calling them Chinese seems misleading. In a nation of 1.4 billion patched together only a hundred years ago, there are always going to be so many historical and cultural touch points that one can pick and choose whatever value one wants to follow. Technology, however, helps this dissemination. So today you see smart Western-educated bankers using their iPhones to pick the most auspicious day for moving house or changing life direction or to consult a fortune teller.

Q12. How much importance do you place on the new slate of leaders as an indication of significant future policy shifts? What do you think will be different in the next 5 years from the 5 preceding?

New leaders could give us a good idea of macro priorities. That is because in China you are promoted in the ruling Communist Party to a higher rank and then after that allocated to a government position commensurate with your rank. But the problem is that it is very hard to use biographical or other information to divine policy shifts. So, for example, the new head of the government appears to be Li Qiang, taking over from Li Keqiang (no relation). He was promoted in from Shanghai without any real central government experience, but also worked in the most dynamic and economically progressive local government areas. He’s also very close with Alibaba, by all accounts — as well as being close with Xi Jinping, his old boss. Any of these simple biographical details could indicate a new policy direction for China. The question is which direction. Given how haphazard and soothsayer-like this all seems, I prefer to trust more in reading what China’s leaders have already said they will do as the most accurate indication of what they will do. Based on that, the next 5 years will see a continuation of Xi’s preoccupation with conservative Chinese values; a redoubling of efforts to ensure that China is unable to be blocked from accessing foreign technology and markets; and a push to advance the manufacturing industries using better tech.

Q13. The Covid-19 pandemic has effectively isolated China from the rest of the world for nearly three years. Has this isolation changed China and its population in any fundamental ways?

I don’t know. I’m not sure anyone does yet. There is definitely a renewed emphasis on community health care, which is for the best as China’s system was for too long too strong on the high-end of care and the basic primary medicine tasks. China’s hospitals are often pretty good, but the clinics that make up the skein of the health care system are weak. Patients traditionally flood the hospitals for care, overloading them, and cutting out the step-up, step-down nature of a more robust system. China has long planned to improve community-level health care and make it more attractive for physicians to work in, but COVID has increased the spending yet further. COVID has also increased the Party’s presence in people’s lives as it has been highly active in distributing goods, taking tests, and enforcing lockdowns.

Q14. How do you see China’s relations with foreign countries evolving over the next several years and will there be a tiered system of relationships spanning friendly to antagonistic? Are you optimistic that China and the West will find a way to peacefully co-exist?

I don’t think there is a way not to co-exist. Whether it is peaceful or not is mostly presumed; within Chinese official media and thought there are no drums of war beating (i.e., there are not pieces talking about preparing for conflict) but there is constant pressure on the military to protect China’s sovereignty, and what is considered sovereign keeps expanding. There is also a persistent and nasty nationalist streak of coverage of the US and Japan in particular. So, I don’t see the antagonism between the West and China dying down any time soon. But at the same point, the world is completely intertwined with China’s economy. So, what’s most likely is a spectrum of relations spanning friendly to antagonistic and where one sits being dependent on how much China’s leaders perceive it to be in their interests to get along with that nation. You will therefore see a lot of picking and choosing: countries being allied with the US on some issues and with China on others. That will be issues based, though. I don’t see any sort of strategic shift of relationships or change in overall foreign policy from China, if only because China’s foreign policy is reactive and not really driven by any determinable strategy. Rather than a vision of “this is what we want the world to be”, it is instead a series of bilateral relationships.

What is said in the course of prosecuting these relationships is often contradictory. So, with the United States at the moment, for example, the line is that China is the country upholding the multilateral regime. However, that is not coherent with China’s stance on Ukraine in the United Nations. Part of what feeds into this, if I can return to the earlier question on messaging, is that we are not used to China’s international behaviour yet. Take the problematic “no limits friendship” line from the Putin-Xi bilateral prior to the Ukraine invasion. This looks terrible in isolation. But it is exactly the sort of vacuous statement that China makes in nearly all bilateral meetings. Talks with South Korea at the same time had nearly identical, somewhat over-the-top language — and that was with a newly elected South Korean president who got into office promising to improve solely the relationship with the US and who took two weeks even to call China. Perhaps that is why I am quietly optimistic that things can improve, because there doesn’t seem too much further they can fall, and much could be done through better communication.

If you or your company has a due diligence requirement or needs assistance on a fraud issue, corporate investigation, risk advisory, or other related matters, please contact us at info@kalavinkaadvisors.com or +852 2196 2727