By Stella Hong Zhang (@StellaHongZhang), edited by Callan Quinn.

While there is keen interest in China’s industrial policies, one rarely gets a chance to hear how Chinese thinkers discuss the topic. However, a recent WeChat post by Peking University’s Institute of New Structural Economics (INSE) of discussions between Deputy Dean Wang Yong and at least 20 other senior university professors, think tank researchers, and executives in leading private firms has offered a glimpse into their thoughts.

INSE was founded by Justin Yifu Lin to promote New Structural Economics (NSE), a theory of economic development he developed during his tenure as the Senior Vice President and Chief Economist at the World Bank (2008-2012).

Justin Yifu Lin lectures students at a summer school at INSE.

Following his return to China in 2012, Lin and his team have ardently promoted NSE and industrial policy. With NSE extolled as the Chinese contribution to development economics and a “third way” between the old-school structuralist approach to economic development and neoclassical economics, Lin has toured developing countries across different continents offering his NSE advice on national development strategies.

Lin was part of a high profile debate with well-known Chinese economist Zhang Weiying five years ago regarding the merit of industrial policy, with Lin advocating for an active role for the state while Zhang dismissed its effectiveness.

The debate was spectacular in the sense that it attracted wide attention both within China and in international media, which has been extremely rare for theoretical debates in China after the 1980s. However, this intellectual schism, with its focus on whether the state should intervene in the economy, has fallen by the wayside— as far as China’s policymakers are concerned, that the state should be proactively managing and guiding the economy is a given.

But while the Chinese government has been enacting industrial policies regardless of such debates, the country is searching for a theory that can justify these practices, if not guide them.

There is a political need for such theorization as China seeks to position itself as a development leader on the world stage through its ambitious Belt and Road Initiative.

The INSE plays an important role in building up an institutional base and shaping the discourse about industrial policy and development in China. But as will be seen in the discussion, there are still fundamental disagreements among Chinese thinkers about the basic conceptualization of industrial policy and how to do it appropriately.

*****

I enjoy putting ChinaTalk together. I hope you find it interesting and am delighted that it goes out free to seven thousand readers.

But, assembling all this material takes a lot of work, and I pay my contributors and editor. Right now, less than 1% of its readers support ChinaTalk financially.

If you appreciate the content, please consider signing up for a paying subscription. You will be supporting the mission of bringing forward analysis driven by Chinese-language sources and elevating the next generation of analysts, all the while earning some priceless karma.

Subscribe now

Hefei model?

The WeChat discussion started with the sharing of a recent CCTV interview with the Party Secretary of Hefei, the capital city of Anhui Province, located in the downstream of the Yangtze River.

Traditionally a city with low visibility on China’s economic map, Hefei recently made headlines by joining 22 other Chinese cities in the elite “trillion-yuan GDP” club, making it one of the few member cities from China’s inland provinces.

The CCTV interview, a carefully scripted PR stunt, portrays Hefei as a city with a smart strategy of economic catching-up: the audience are informed that Hefei has been generously supporting basic scientific research at the local universities (Hefei hosts the prestigious University of Science and Technology of China, a rare case of an elite university located outside of China’s “top-tier” cities), which helps build up the city’s human capital.

Hefei has also been aggressively soliciting firms to relocate to the city using financial and policy incentives as part of its effort to upgrade the local industrial structure. Several leading home appliance firms have set up factories there, making Hefei China’s largest base for home appliance manufacturing.

But most interestingly, and controversially, is Hefei’s equity investment in pioneering tech firms through its municipal government-owned “urban construction” company.

These types of local state-owned enterprises (SOEs) have played a major role in financing infrastructure and urbanization in China, and have been responsible for China’s “shadow banking” problems.

Some, Hefei’s company among them, have transformed into industrial investment funds. The city’s first notable investment was in 2008, when it invested in BOE Technology, which then set up several state-of-the-art thin-film-transistor liquid-crystal display (TFT-LTD) production lines in Hefei. This was believed to have facilitated the clustering of home appliance manufacturing in Hefei.

BOE showcases products in Fuzhou, Fujian province.

The latest instance of this was a $1 billion investment in the electric-vehicle (EV) company NIO in 2020, saving it from a cash crunch. As part of the deal, NIO agreed to relocate its entire operation from Shanghai to Hefei. This investment appeared to be immediately profitable as the NYSE-listed NIO’s stock price increased ten times over in the 12 months following the deal.

Hefei municipal government promised to re-invest the returns from the investment in developing the smart electric vehicle industry in the city. A new industry park for this purpose, jointly developed by NIO and Hefei, was inaugurated in April 2021.

As some Chinese observers put it, Hefei municipal government has virtually turned itself into an investment bank or a venture capitalist.

What Hefei has been doing goes beyond what we conventionally understand about China’s “state capitalism”, which primarily describes the way the state maintains ownership and control of enterprises inherited from the command economy, while restructuring them into modern corporations to participate in the market economy.

In Hefei’s case, the local state invests in firms in emerging industries to receive their production, R&D and other capacities, with the strategic purpose of attracting other firms up and downstream the value chain to agglomerate and thereby upgrade the industrial structure of the city.

Such investments also take place in the context of a globalized capital market, giving the state investor access to financial returns unavailable in the case of a traditional state cash injection.

A conversation among experts

The following are some excerpts from the discussion (digressions from the topic of industrial policy were removed and the text was slightly edited for brevity). As can be seen, there is actually considerable disagreement among the group participants about the definition and scope of industrial policies, and what specific problems they are supposed to solve.

Tang Jie (formerly Vice Mayor of Shenzhen, currently Economics Professor at the Shenzhen campus of Harbin Institute of Technology):

Three questions are worth careful consideration: First, should industrial policies be widely applied? Should such policies intervene in market mechanisms in an extensive and regular manner? If so, how does the market work? If not, how is the will of the government manifested?
Second, are industrial policies suitable for frontier innovation industries? How does the government know the comparative advantage of frontier industries and how does it coordinate? Or are industrial policies only for the “catching-up” industries?
Finally, is there a spatial limit in the implementation of industrial policies? Assuming that A and B belong to two administrative units with relatively independent fiscal and administrative powers. How to implement industrial policies separately without splitting the market and [resorting to] beggar-thy-neighbors?

Wang Yong (Deputy Dean of INSE):

The Hefei case… shows that there is still a lot of room for improvement in our country's venture capital market. Neither NIO nor BOE belongs to an industry involving national security or economic security. Therefore, unless they meet the comparative advantage of [Hefei’s] endowment structure, they should not be the target of industrial policy.
The government should also be prepared to exit after an initial period of guidance and return it to market competition. What needs to be studied in particular is what among the things the Hefei government did in the NIO investment could not have been done by the venture capitalists in the market, and why not. Was it because the market was over-developed, or under-developed?
It seems there are many industrial guiding funds set up by various local governments, but not many of them seem successful.

Zhu Yanmei (Executive Director and Executive Vice President of BGI Group):

We should also encourage diversity in the industrial policies, as well as the products and factors they promote, in different regions in order to generate flows in the commodities, talents, capital, etc., which can then stimulate the vitality of the market and the entrepreneurs.
The degree to which the government should intervene should be determined by whether it can spur entrepreneurship.

Tang Jie:

I have always had doubts about the boundary between industrial policy and innovation/development policy, as well as whether there is a limit for the industrial policy. It seems that it’s better to further break down government behaviors rather than lumping everything under “industrial policy”?
… During the Second World War, the U.S. enacted policies that helped with the aerospace and semiconductor industries in the 1960s and 1970s, which played an important role in ensuring the frontier status of the US.
However, the US did not call them industrial policies, but rather efforts to build a system for innovation that can be sustained in the long term… How the US formulated and implemented major frontier science and technology policies warrants our careful study.

Shen Ling (Associate Professor at East China University of Science and Technology):

I would define industrial policy as a biased allocation of public resources. In fact, it is what the government has to do given budget constraints—the government does not have a choice not to do it just because the result isn’t ideal.
After all, there will be sequencing in the government’s allocation of public resources. If the government prioritizes sectors with comparative advantage, then such industrial policies can help grow the public resources, which is the gist of NSE.
Many critics of industrial policy tend to cherry-pick failed cases to reject industrial policies. But then what should be the criteria for public resources allocation?
Once embarking on a path of industrial upgrading, you can't stop in the middle. Many amaze at what a great decision Hefei’s bet on BOE was, but in fact at that time Hefei also invested in plasma technology [which was not successful]. However, there is little documentation of how much money they invested into that and the price they paid.

Tang Jie:

The government should be considered the agent, not the principal. When there are investment losses, the government should be questioned.
Even if in ten investment cases one was successful and made up for all the losses, taxpayers still have reasons to question if the investor was captured. This is why the government cannot be a VC, and our audit regime is also based on this notion.

Wang Yong:

It is necessary to distinguish the different roles of the government and VC based on the actual conditions of different industries at different development stages. In many cases, our government needs to play a greater role regarding frontier innovative industries because our capital market, especially the venture capital market, is under-developed.
In other cases, the factor markets such as labor and land, as well as public services such as education, health care and old age care are still subject to policy regulations.
Without [local] governments’ facilitation and special policy treatments to overcome these constraints, companies can hardly build up their operation and recruit the necessary talents simply by the power of the market.

Shen Ling:

I think these are two different scenarios. If government regulation on the factor market was unreasonable to begin with, then such regulation should be reduced or improved in order to promote industrial development. Classic economics would consider it a case of market reform-induced growth, rather than the effect of industrial policies.
On the other hand, if innovative tech companies are hampered due to under-development of the capital market, then we need the government to help lower transaction costs and spur the development of the market and the industry. I guess this is what New Structural Economics calls “proactive government”?
This is why I think the “biased allocation of public resources” advocated by NSE should work on cultivating the market and reducing transaction costs…
We shouldn’t think about industrial policies as the government directly investing in companies. The “industrial funds” should be just one of the ways, but not the only way, of conducting industrial policy. Direct equity investment can easily lead to corruption and efficiency loss.

Ma Xiaoye (founding director of the Academy for World Watch and a former representative of China to the World Trade Organization):

How to define industrial policy is a critical question. What [Tang Jie] said about the boundary between industrial policy and development policy is spot on.
The discussion about a “proactive government” is essentially a debate about big or small government. It would be missing the point to put this at the center of the discussion on industrial policy. This is why, despite so much having been written about industrial policy, the discussion remains shallow.
Of course, for those advocating an important role of industrial policy, they would want to frame the discussion in this way since it gives them power under the current situation. But if we want to push the discussion deeper, we need to get rid of these noises.
… Given the consensus that the government has a necessary role to play in making economic development policies, we can frame the discussion in terms of the core function of the government: the trade-off between safeguarding market competition in order to enhance efficiency of the economy on the one hand, and the implementation of targeted industrial policies on the other.
… The two functions of the government are not free of friction (here we should avoid confusing industrial policy with general development policies that do not target specific industries).
If a specific industrial policy in a certain sector causes changes in the competitive environment of related industries, can the government make sure that the entire market adjusts in an orderly manner following the initial changes in the first industry? Can it be guaranteed that the international market can also adjust without problems? Is the “proactive government” supposed to preemptively intervene in all related industries or simply let them go their ways?
Imagine that the government follows this path, even if we are not returning to a fully planned economy, it won’t be easy to reverse the course.
We have already had such experiments with government corporatization during the first Five-year Plan period and the socialist transformation of industry and commerce.
We should not resort to such a way of intuitive thinking when encountering new problems.
I have always questioned those arguments that treat the US and other countries’ economic development policies as industrial policies for the sake of “enriching” the discussion.
A basic scientific research policy is not an industrial policy; the military contractor system related to the defense industry is not an industrial policy; the impact of the Defense Advanced Research Projects Agency’s (DARPA) military-use research projects on the development of cutting-edge technologies is not an industrial policy; a regional economic development policy is not an industrial policy.
We have been doing all these things too, and how much we can accomplish depends not on the level of knowledge but on the availability of financial resources and competition among the bureaucracies… I think a key difference between development policy and industrial policy is that the former does not lead to a significant bias in the market conditions for specific industries, rather, it works on improving the overall competitiveness of multiple sectors.
We should not include these types of policies in our discussion of industrial policies. The development policy is exactly what the National Development and Reform Commission (NDRC) has been doing, as reflected in the set-up and evolution of its internal divisions.
It is in the advantage of countries like China with a big government to engage with development policies. These kinds of policies will not necessarily result in negative consequences undermining the functioning of the market, even though there might be waste in how the money is spent.
Just this month, the European Union has formally initiated talks with other major countries to formulate a multilateral response to the government-led "market disruption" of trading partners by using state-owned enterprises to implement industrial policies…
If we want to take down the multilateral disciplines that target our industrial policies, in the spirit of countering “the rules made by a minority of countries” in the words of [China’s top diplomat] Yang Jiechi, there is still a lot of difficult work to do…
We will need to provide the theoretical foundation for our actions and convince people that our industrial policies are both necessary for our development policies and will not affect the basis of market competition. Only then can we confidently say that the criticisms of China’s policies are motivated only by the desire to contain China’s development.
The notion and methods of industrial policy advocated by NSE have been increasingly used in the domestic setting in recent years. In my opinion, if you make industrial policies based on comparative advantages and latent comparative advantage while using only supportive tools (such as subsidies and investment by state-owned enterprises) without the restrictive policy options (even though there might still be administrative approval that can be regarded a restrictive measure), then it becomes a regional development policy. This is what the National Development and Reform Commission and the local Development and Reform Commissions have been doing...
Of course, if the NSE’s Growth Identification and Facilitation Framework (GIFF) can be used for retrospective research on regional development policies, then it can still make a significant methodological contribution.
But one should caution against reinventing the wheel in the field of regional development policies. Sure, it may be necessary to test the water in the domestic setting by trying the new method to analyze the industrial structure. But without the borders and customs, regional development problems are quite different from the scenarios that involve different countries.
Findings from research on regional development cannot be extrapolated for industrial policy research that is premised on international comparative advantage.
The idea of “proactive government” calls for a government that is not just the guardian of market order, but also guides economic development based on comparative advantage using various methods.
This is a good idea, but what people in academia and industries really care about is whether such practices of the government, including the direct intervention in the market through SOEs, contradict with the government’s role in terms of safeguarding a level playing field to ensure the efficiency of the industries and the general economy? How does the government safeguard fair competition and high efficiency while practicing industrial policies? Where is the limit of industrial policies?
After all, growing national power depends on the efficiency of the economy, and a competitive environment cannot be replaced by carefully formulated industrial policies.
Some may advocate the government to continuously adjust and formulate new industrial policies to correct the distortions in the market from the previous industrial policy, then we would need a very powerful planning committee, equipped with [Alibaba’s] cloud computing and [JD’s] vision of a planned economy.
A feasible big government theory for economic development is nonexistent. China’s economic development in the past four decades does not support such a theory. The domestic and international economic conditions are constantly changing.
Under the current circumstances, should we prioritize spurring market competition or making a greater number of more sophisticated industrial policies, especially when there is a trade-off between the two? How should the priority be adjusted when the external economic conditions change? What should be the priority under the “dual circulation”?

Wang Yong:

Your point is well taken that the scope of industrial policy needs to be clearly defined. … I think NSE has always had a clear definition of industrial policy …
NSE’s classification of industries is meant for a more comprehensive analysis of the industrial policies of a developing economy.
In my opinion, Mr. Zhang Weiying's definition of industrial policy is too narrow, as he equates it with government intervention on private sector production…. It is perhaps not a bad thing for different scholars to have different definitions, because no consensus has emerged yet for this important question.
But when engaging in debates, one should be clear about where the disagreement lies and ground the discussion in areas where we do agree.

Ma Xiaoye:

The example of potatoes you gave is a typical economic development support policy that is commonly seen in many countries. The US has policy support for everything, from soil, seeds and planting to the forecast of international markets and policy. These are mandated by the Agricultural Act 1952 [translator’s note: there wasn’t such legislation in 1952. The year may be a mistake].
When New Zealand’s kiwifruit was threatened by the cheaper species from Latin America, their government led scientific research on the technology to change its color. Such policies were probably not controversial.
It is better to distinguish “non-neutral” industrial policies that target specific industries. I agree with you that Zhang Weiying’s argument based on entrepreneurship was off the point. But didn’t people on this side of the debate also digress?

Wang Yong

I disagree that industrial policy should be limited to the industrial sector only. Even standard textbooks of international economics treat the US agricultural subsidies as a typical example of industrial policy, such as Paul Krugman's textbook on international economics.

Ma Xiaoye:

But the fact is that the evidence NSE has used to support its conclusion is China’s manufacturing takeoff. I have no objection to the inclusion of services and agriculture in the research, but they need to be studied separately.
The international consensus divides policies related to industrial development into support policies and development policies. There were two plurilateral agreements prior to the WTO: the International Bovine Meat Agreement and the Agreement on Trade in Civil Aircraft. These were two internationally recognized protocols for development.
The Better Cotton Initiative that we have heard so much about recently also mainly covers development protocols for the cotton industry. It was right for Krugman to include agricultural policy in his textbook, as the US has both development support and subsidies, and import restrictions for agriculture, which has a major impact on international agricultural trade.
As a result, agricultural trade has long been excluded from the free trade principles. But we can’t be trying to emulate the US policy for agriculture from the 1950s in our policies for manufacturing, can we?

Wang Yong:

You are right that many of the policy documents issued by our government ministries are often criticized by the US and other developed countries as evidence of breaking international rules.
One of the most important reasons is that the planned economy mindset remains when drafting policies, which they may not even realize. The Made in China 2025 strategy set clear targets for the localization rate of the strategic industries, which were taken by the US as a clear import substitution strategy that violated the spirit of the WTO. In fact, our original intention might be just to upgrade the industry and produce higher value-added products.
Setting the quantitative targets was for the purpose of evaluation, but they became used by others as evidence of wrongdoing.
I would also like to respond to your point about basic scientific research, defense industry, and regional economic development policies falling out of the realm of industrial policy…
We know that most of the industries in developed countries are at the global technological frontier, which the NSE would classify as the “leading” industries. The upgrading of these industries mainly depends on technological progress, which primarily comes from their own R&D.
Therefore, the best government support for these industries would be for basic scientific research… As most companies are unwilling to invest on a large scale on the “R” (basic research) and prefer to invest in the “D” (product development), which can be more profitable in a shorter time frame, the “R” is predominantly carried out by higher education and research institutions.
But the government still intentionally and “discriminatorily” supports areas of research deemed more important through mechanisms such as the National Science Foundation. These areas of research are more or less related to their industrial upgrading. For example, the “Endless Frontier Act'' and the other legislation to support advanced manufacturing in the US are de facto “industrial policies”, even though they call them “innovation policies.”
One day when China reaches a similarly high income level, the most appropriate industrial policies for China will also be similar to those policies used by the developed countries like the US.
However, since our per capita income is less than one quarter of the US, China still has many “catching-up” industries where the technology is far from the global frontier. The appropriate industrial policy for these industries is not to reinvent the wheel, but to better introduce, learn from and absorb the existing technologies.
In terms of the defense industry, my team and I have been studying the industrial policies of various countries in the past few years and found that many industrial policies in the US are led by the Department of Defense (DoD). One might think that the DoD only deals with weapon production, but in fact the DoD supports a wide range of industries, including artificial intelligence, telecommunications, genetic technologies and so on, many for both military and civilian uses.
The US is particularly good at applying military technologies in civilian industries. For example, Boeing produces both civilian airliners and fighter jets. Many technologies are interlinked, so we cannot simply exclude the defense industry from industrial policy analysis. The five major industrial classifications of NSE include “strategic industries”, which can be further subdivided into two major categories: national defense security and economic security.
For this type of industry, even if it does not conform to the comparative advantage in the economic sense, it should still be promoted for the overall national interests… Since China is a great power in the geopolitical sense, “strategic” industries cannot be ignored.
The US encircled Huawei globally based on the idea that Huawei threatens the US national security. We cannot abandon our industrial policy research on the 5G technology which Huawei is in simply because the US has included Huawei in the field of national security.
Regarding whether there is any overlap between regional economic development policies and industrial policies, I think it also needs to be discussed case-by-case.
If the relevant policy is a place-based industrial policy (such as industrial parks, financial pilot zones, special economic zones, etc.), which involves supporting specific industries, then it falls under the scope of industrial policy analysis. Just like some of the trade policies are also industrial policies.

How relevant is NSE as an economics theory for China’s industrial policy?

What I take away from the above discussion is that NSE has yet to convince many other scholars and practitioners of its basic definition of industrial policy, let alone how effective its policy prescriptions are.

While NSE advocates agree that industrial policy as a theoretical concept should refer to only those policies targeting specific industries, in practice, they tend to include all kinds of government intervention in their industrial policy analysis.

In other words, there is a disconnect between the theoretical foundation of the NSE and its policy prescriptions.

To summarize the gist of NSE, it argues that the government should identify the “latent comparative advantage” (LCA) of an economy based on its factor endowment structure, provide infrastructure and other policy support to correct whatever market failure that has caused the sectors with LCA to not receive the level of investment they deserve.

Fundamentally, NSE doesn’t challenge the idea that the market knows how to best allocate resources, even though there might be hiccups, which it argues is where the government should step in and deal with them.

The problem is, however, NSE does not have a good way of determining what specific industries have LCA. Its method of identifying LCA is based on referring to other economies with similar endowment structures but higher income levels: the industries that have been exporting in these economies would be those with comparative advantage and thus should also be developed in the economy in question.

This method is based on a highly reductionist understanding of the economy inherited from neoclassical economics. Practically speaking, however, the analysis of the endowment structure will only make sense if there is a specific industry in mind so that relevant endowment factors (the kind of human capital and resource) can be identified, and have a stylized “foreign economy” to compare with (thus comparative advantage).

The problem is circular: without knowing the industry, one cannot have an empirical strategy for identifying the relevant endowment factors; without properly assessing the endowment factors, one cannot identify the industry that best fits the endowment structure.

Indeed, its indeterminacy in identifying the sectors or paths of development has been a main critique of the NSE.

But a more serious problem for NSE to be really useful for the Chinese industrial policymakers is that the policymakers have a very different concern now than what NSE conceptualizes. NSE assumes a late-industrializing economy that needs to catch up, based on which assumption it identifies comparative advantage by looking at economies with higher income levels.

But as reflected in Made in China 2025, the kind of policy China is pursuing is about cultivating the so-called “emerging strategic industries” and moving to, or even defining, the global technological frontier. It is about “leading”, not “catching up”. For the government, the task is to improve the endowment structure to facilitate the growth of the frontier industry, rather than to decide whether to pursue a frontier industry based on its existing endowment structure.

Interestingly, there have been attacks from some Chinese economists who argue that NSE is antithetical to China’s strategy of pushing for frontier industries, because NSE advocates “comparative advantage conforming” development strategies and therefore would advise against developing industries of high technological intensity, as they are not in the comparative advantage of developing countries like China.

Such questions about whether the “industrial policy” framework is appropriate for frontier industries were also raised by Tang Jie, a former vice mayor of Shenzhen and now a professor at the Shenzhen campus of Harbin Institute of Technology, in the group discussion.

As he is likely the only person in the group who has policy making experience, Tang made several interesting points. He stresses the need to delineate the boundary for “industrial policies”. He suggests that for the development of frontier industries, it may be more appropriate to talk about “innovation policies” that support basic research and promote sustained innovation in the long term, rather than “industrial policies” that target specific industries.

He cites US policies in the mid 20th century that resulted in the development of the aerospace and semiconductor industries as an example, and that those should be considered innovation policies, not industrial policies.

Similarly, Ma Xiaoye, founding director of the Shanghai-based Academy for World Watch and a former representative of China to the World Trade Organization (WTO), shares this view.

Ma notes that there are trade-offs between the government’s role in maintaining competition to enhance economic efficiency and the government’s role to support specific industries for development purposes.

To him, it is necessary to distinguish “development policies” that address system-level issues and improve competitiveness across sectors from “industrial policies” that, if inappropriately implemented, could easily undermine the functioning of the market. He also argues that the real advantage of China’s “big government” is in this kind of system-level “development policies” rather than sector-specific “industrial policies”.

Ma also contends that “industrial policies” based on “comparative advantage” need to be applied in the context of international trade, and thus the tools of industrial policies can include supportive measures such as subsidies as well as restrictive measures such as trade barriers.

These are exactly the policy areas where an elaborate system of rules has been established, embodied by the WTO, which China is part of. Therefore, he argues that if China is to make industrial policies using these tools, it is necessary to justify them according to international rules.

He also notes that the European Union and other major countries are pushing for new rules in multilateral institutions in response to “market disturbance” caused by industrial policies conducted through SOEs (no doubt with China in mind), and suggests that China’s industrial policy theories need to explain how such policies do not disrupt the basis of international market competition.

Re-imported theory?

Ma shares an interesting observation. In recent years, NSE has been increasingly applied in China’s domestic settings to guide the development of regions, even though NSE is based on a theoretical foundation of comparative advantage, originally a theory of international trade.

This is also reflected in Chinese-language academic publications on NSE. Searching the CNKI database, the closest thing Chinese academia has to Google Scholar or JSTOR, I found 204 journal articles applying the NSE theory published between 2010 and 2020, only 4 of which analyzed development strategies in foreign developing countries. The rest are all about domestic development in China, including regional development strategies for Chinese provinces, cities and even townships.

It seems the theory, which was originally meant to target the international audience as part of the “Chinese solution” to development, has been reimported for domestic consumption.

Applying NSE in their research, the researchers seem to be identifying the so-called “comparative advantages” for various regions in a rather liberal way. For example, in the regional development strategy that Lin designed for Jilin province based on his NSE theory in 2017, he argues that Jilin needs to both catch up in the light industries (such as textiles), where Jilin has been traditionally weaker, and keep up with sectors that Jilin was already leading (including everything from economic crop plantation to railway, shipping, and aerospace industries); on top of that, Jilin still needs to develop emerging industries and invest in information technology and electronics.

It is hard to see how an analysis of Jilin’s endowment structure can be replicable. Even if these recommendations all turn out to be good, it is probably due to other insights beyond what NSE has to offer.

All these unanswered questions in the theory have not stopped NSE from being promoted in China. According to Wang Yong, the Chinese Ministry of Education is supporting the compilation of NSE textbooks for undergraduate and PhD students.

Courses on NSE can be expected to be offered in economics departments across Chinese universities. Led by INSE, there is also a national “NSE research alliance” whose members include 24 institutions located in universities across China (as well as one in Poland), with 12 more having shown interest in joining.

Within the INSE, there are teams specializing in domestic development and international development advisory, which have been actively working with local governments in China and foreign governments. Such institutional efforts to canonize a theory seems unprecedented and how that will shape the next generation of China’s economic policy practitioners remains to be seen.

I’d love to have more economists from China on the ChinaTalk podcast. If anyone has suggestions or could make introductions, please do reach out!

ChinaTalk

Chinese Industrial Policy: How A City Turned…