Whether expressed as “strong supply, weak demand”, overcapacity or some form of a “K-shape”, a consensus has emerged that insufficient demand is the central weakness of the Chinese economy. The persistent debate centres on how best to fill that demand gap.
Economists of a Keynesian bent, such as Yao Yang, Yu Yongding and Zhang Ming, all concede to varying degrees that the pre-2021 economic model had problems. However, they all argue that the most urgent task is to repair local government balance sheets in order to restore economic activity in traditional construction and manufacturing. They suggest that reform to the economic model, whether in favour of emerging industries or household consumption, would be too slow and inadequate to meet the current challenges of stagnant demand, investment and employment.
Such thinking was prominent in a recent, widely shared speech by the Tsinghua economist Li Daokui. Li argues that while the new industries have failed to expand sufficiently to replace the old ones, the deflationary “black hole” of local government finances has exerted a cooling effect on the entire economy. For Li, the idea of a “K-shaped” split between the high-growth emerging sectors and low-growth traditional sectors is a distraction from the need to restore heat to the economy as a whole. In response, he proposes heavy, centrally backed stimulus on an aggregate level.
Not everyone agrees. The following piece by retired Tsinghua professor Sun Liping is framed as a riposte to Li Daokui’s diagnosis. Sun is a liberal-minded sociologist and public intellectual known for his work analysing Chinese society through the lens of inequality and social fracture. In his view, the K-shaped split is not only genuine, but reflects the breakdown of the old economic transmission process, whereby investment in the economy’s “upper body” (formerly manufacturing and real estate) could once “pull up” the “lower body” of consumption and employment. Aggregate stimulus or additional liquidity would therefore fail to revive the broader economy.
Sun sees the breakdown of this old mechanism as driven by technological change, which is siphoning capital towards priority industries in the upper arm of the K. His conclusion is one shared by pro-rebalancing economists such as Li Xunlei and Liu Shijin: “the crucial task is to change the model of development and give priority to people’s livelihoods [民生], employment and building a sound social-security system”.
The following is a translation of Sun Liping’s recent article, followed by a translation of Li Daokui’s speech from last month for context.
—James Farquharson
Key Points
The diagnosis of China’s economy as “overall cold” was correct during the pandemic and its immediate aftermath, but since 2025 the economy has moved towards an increasingly pronounced “fire-and-ice” split between rapidly growing and struggling sectors.
The economy’s “upper body” and “lower body” increasingly resemble independently operating systems, with circulation between advanced industries and ordinary livelihoods weakening.
As such, the main problem is not simply that emerging sectors are too small to “pull up” the rest of the economy, but that the past mechanism whereby this would occur is failing.
Instead of boosting the rest of the economy as construction and manufacturing once did, emerging industries are now siphoning capital away from other parts of the economy.
The effect of this split is that traditional industries and even earlier high-growth sectors are left with diminished access to investment, harming economy-wide employment and livelihoods.
AI is further weakening livelihood security by eroding the link between economic growth and employment, thereby encouraging greater precautionary saving.
Additional liquidity would likely remain circulating within the economy’s “upper body”, producing a limited effect on reviving employment, consumption and investment activity across the wider economy.
Rather than relying primarily on aggregate-level stimulus, policy should prioritise livelihood spending, employment and social security as a way to rebalance towards a new development model.
The Scholar
Name: Sun Liping (孙立平)
Year of birth: 1955 (age: 71)
Positions: Retired Professor, Department of Sociology, School of Social Sciences, Tsinghua University (active faculty: 2000–2020)
Other: Specially Invited Research Fellow, China Society of Economic Reform (CSER); former Director, Centre for Research on Oral Materials of Social Life, Peking University; former Director, Development Research Centre, Department of Sociology, Peking University
Research focus: Social modernisation; social transformation and structural change; market transition; oral social history; state–peasant relations in contemporary China
Education: BA in Journalism, Department of Chinese Language and Literature, Peking University (1982); Sociology Special Programme, Nankai University (1981)
ON THIS POINT, I DO NOT ENTIRELY SHARE PROFESSOR LI DAOKUI’S VIEW
By Sun Liping (孙立平）
Published on his Public Account on 13 August 2026
Human-edited machine translation
(Illustration by ChatGPT)
I. An Important Shift Has Taken Place
At a recent forum, Professor Li Daokui offered up an important economic assessment: the greatest problem facing China’s macroeconomy is not K-shaped divergence, but that the economy as a whole is running cold [不是K型分化，而是整体偏冷]—and that this “running cold” has been ongoing for three years.
As an old friend, I understand his point: he wants to stress that we must confront the problems before us and maintain an alertness to crisis. But as a specific diagnosis, I disagree with his view.
Is China’s economy currently running cold across the board, or is it undergoing K-shaped divergence—or, as I have repeatedly put it recently, is it split between fire and ice? These are two very different diagnoses. I think we may need to examine the question dynamically.
Viewed from an evolving perspective, I think that the diagnosis of an economy running cold across the board was appropriate during the pandemic and the period immediately afterwards, broadly up until 2024. But since the beginning of 2025, the situation has changed markedly: the economy has gone from running cold as a whole to two worlds of fire and ice [冰火两重天].
The heat is concentrated in two main areas. One is high tech and advanced manufacturing, and the other is exports. Consider the following data:
In the first half of 2026, value added in high-tech manufacturing grew by 13.3% year on year, 7.9 percentage points faster than among all large-scale industrial enterprises [Note: 规上工业, or enterprises with annual revenue exceeding RMB 20 million]. Within this, aerospace vehicle and equipment manufacturing grew by 16.3%, while the electronics and communications equipment industry grew by 17%. AI-related industries such as integrated-circuit manufacturing and intelligent in-vehicle equipment manufacturing maintained growth of over 30%, while industrial robot output rose by 28%.
During the first seven months of 2026, exports grew by 14% year on year, with growth reaching 27.0% in June alone. Within this, cumulative integrated-circuit exports surged by 91.6% year on year, partly due to price effects. Vehicle exports rose by 49.1%, lithium-battery exports by 35.8%, wind-turbine exports by 34.8%, ship exports by 32.7% and industrial robot exports by 13.2%.
Professor Li Daokui is, of course, aware of this development. He simply believes that these sectors still account for a relatively small share of the economy. As he puts it, the new industries are indeed expanding, but they remain too small to fill the hole left by the extinguishing of the traditional growth engines.
But I believe that what matters most is the trend, and the structural characteristics that the trend reveals.
II. Different Diagnoses Reveal Different Problems
Two different diagnoses may stem in the main from different ways of viewing the problem. Professor Li Daokui broadly approaches it from an aggregate perspective. From this perspective, he emphasises two problems. The first is employment: he points out that the broad unemployment rate has reached 10.2%. The second is fixed-asset investment: he says that its negative growth is almost unprecedented in recent decades.
Viewed in aggregate terms, I agree with these conclusions. But I believe we must look more closely beneath the surface.
Take fixed-asset investment. In 2024, overall investment in China’s high-tech industries grew by 8.8%. Within this, investment in high-tech manufacturing rose by 8.2% and investment in high-tech services by 10.2%. Investment by centrally administered state-owned enterprises in strategic emerging industries grew by 17.6% year on year. In 2025, investment accelerated across all these fields, while nationwide investment in equipment renewal grew by as much as 11.8%. In 2026, investment in electronic-circuit manufacturing rose by 55.6%, lithium-ion battery manufacturing by 24.4%, aerospace vehicle and equipment manufacturing by 24.7% and information services by 15.5%.
Therefore, while recognising that fixed-asset investment is running cold overall, we must pay even greater attention to the ice-and-fire split across different sectors.
Where, then, does the problem lie? Professor Li Daokui’s reasoning is that the hot parts of the economy remain too small to pull up the economy as a whole. One therefore needs to act at the aggregate level [从总量上做文章] and heat up the entire economy. My view, however, is that there exists a structural problem behind this phenomenon. In other words, the question is not whether sectors are able to pull up the rest of the economy, but whether the “pulling-up” mechanism even still exists at all.
Several years ago, I proposed the concept of a new dual structure—the problem of what I referred to as the economy’s “upper body” and “lower body”. Judging from current trends, these two parts look more and more like two independently operating systems: you run your way and I run mine [你运行你的，我运行我的]. In the past, growth could create employment and increase society’s purchasing power, thereby promoting circulation and prosperity throughout the economy. But this relationship has grown progressively weaker. Instead, resources across society are continually being siphoned into high-end industries. This is particularly clear in the stock market.
One point needs particular emphasis: we may now be confronting a very new situation and a very new problem, rather than the traditional economic logic of the past. A few days ago, I wrote an article entitled The US Economy Is Also Split Between Fire and Ice. In the first quarter of 2026, AI-related investment contributed 0.75 percentage points to annualised quarter-on-quarter US real GDP growth, while in the same period private consumption contributed only 0.37 percentage points, its lowest level since the second quarter of 2022.
Why should economies that differ so much in so many respects display this common characteristic? Because a common factor, the current technological wave centred on artificial intelligence, underlies it. It is simply that this factor manifests itself differently in different contexts.
III. Different Diagnoses, Different Approaches
Different diagnoses naturally suggest different approaches to solving the problem.
On the basis of his aggregate analysis, Professor Li Daokui argues that over the past 20 years, local infrastructure investment and public fiscal expenditure consistently amounted to around 41% of GDP and served as an important driver of regional growth. That proportion has now fallen to 35%, with an even sharper contraction in capital expenditure. Burdened by heavy debts, local governments have gone from engines generating heat to entities that instead absorb it.
As such, Professor Li Daokui’s proposed solutions are to more than double central government bond issuance, enabling local governments to emerge from contraction and once again become a force driving transformation; to increase welfare provision and remedy its deficiencies; and to expand the supply of high-quality public services.
Many people attribute weak consumption to inadequate household incomes, but Professor Li rejects this logic. Household consumption demand has now moved upmarket. Demand for ordinary everyday goods is approaching saturation, while people are more willing to spend on healthcare, high-quality education and elderly-care services. The issue is that there is insufficient high-quality supply in these areas. People want to spend but have nowhere suitable to do so. This, he argues, is the real reason consumption has failed to pick up.
I agree with some of Professor Li Daokui’s specific policy proposals, such as remedying shortcomings in welfare provision. But I think his basic approach remains one centred on liquidity. In my view, the problems we now face are much more complex than this. The key lies in the aforementioned new challenges created by the technological wave centred on artificial intelligence.
More specifically, the new challenges we face include at least the following three:
First, new technologies are siphoning up capital. Leaving aside state investment, China’s venture-capital market recorded 2,865 funding deals in the first quarter of 2026, totalling RMB 256 billion—year-on-year increases of 52% and 48% respectively. The overwhelming majority occurred in new-technology sectors. Not only are traditional industries being neglected, but even what were once called the “new three” [Note: EVs, solar and batteries]—which are, in reality, now the “old new three” [老新三样]—are being left out in the cold.
Second, it is becoming increasingly difficult for wealth to spread through society by means of employment. In the past, when the economy improved, employment would improve and people’s incomes would rise. The current problem is that when the economy performs poorly, jobs are naturally harder to find and incomes decline, but even if the economy improves, none of this will necessarily change fundamentally. It is not that the new parts of the economy are unable to pull the whole economy along, but that the very logic through which “pulling up” could once occur just does not exist [新的部分不是带不动整体，而是根本没有带动的逻辑].
Third, the uncertainty we now face is unprecedented. Some time ago, I introduced the concept of “livelihoods” [生计模式]. The greatest impact of this technological wave is its assault on certainty, particularly the certainty of people’s livelihoods. Under these circumstances, precautionary saving becomes an even more pressing necessity. The real reason behind weak consumption is that nobody dares to spend money.
I therefore do not particularly favour an approach centred on liquidity. Under present conditions, even additional liquidity would struggle to perform its former function and would merely continue circulating within the economy’s “upper body” [在上半身循环]. I believe the crucial task is to change the model of development and give priority to people’s livelihoods [民生], employment and building a sound social-security system. On this foundation, the economy can be made to circulate in a genuine sense.
What follows is a translation of Li Daokui’s original speech, to which Sun is responding.
Key Points
The main challenge facing China’s economy is not a K-shaped split, but a condition of running cold overall.
Emerging industries, usually described as the “upper arm” of the K-shape, are too small to replace the traditional growth engines of property and infrastructure.
This shows up in a broad unemployment rate of 10.2% (including the long-term unemployed or “discouraged labour force”) as well as a historically unprecedented contraction in fixed-asset investment.
During the two decades of China’s high-speed growth, local-government infrastructure investment and property were instrumental, with the former, alongside routine expenditure, averaging an estimated 41% of GDP.
The impact of the property downturn on consumption has been less catastrophic than expected, with households absorbing large losses without a corresponding collapse in consumption.
The most serious problem is therefore the contraction of local government finances, which are absorbing credit to finance debt repayments and acting as a blockage in the circulation of China’s economy.
Local-government debt has exceeded 100% of GDP, as borrowing to refinance debt and cover interest keeps liabilities rising despite repeated debt-resolution measures, turning local governments into economic “heat absorbers”.
The solution should be broad liquidity stimulus, increasing central bond issuance to refinance local-government debt, stabilise property and fund local expenditure.
The Scholar
Name: David Daokui Li (李稻葵)
Year of birth: 1963 (age 62)
Positions: Director, Academic Center for Chinese Economic Practice and Thinking (ACCEPT), Tsinghua University; Mansfield Freeman Chair Professor of Economics, School of Economics and Management, Tsinghua University; Co-President, Society for the Analysis of Government and Economics (SAGE)
Other: Former Standing Committee Member of the 13th National Committee of the CPPCC; former member of the Monetary Policy Committee of the People’s Bank of China; Founding Dean, Schwarzman College, Tsinghua University
Research focus: Government and economics; the Chinese economy; economic development; international economics; comparative economics
Education: BSc in Management Information Systems, Tsinghua University (1985); PhD in Economics, Harvard University (1992)
Experience abroad: Assistant Professor of Economics, University of Michigan, Ann Arbor (1992–1999); National Fellow, Hoover Institution, Stanford University (1997–1998)
THE BIGGEST PROBLEM FACING CHINA’S ECONOMY TODAY IS NOT K-SHAPED DIVERGENCE
By Li Daokui (李稻葵）
Published on Aisixiang on 12 July
From a speech given at a China Macroeconomic Forum seminar on 11 July
Human-edited machine translation
I. The Problems: Weak Investment and Unemployment
My first central argument is this: the biggest problem facing China’s macroeconomy today is not K-shaped divergence, but that the economy as a whole is running cold [不是K型分化，而是整体偏冷] and has been doing so for three years. Why do I emphasise this? Because once we begin talking about a K-shaped divergence, it is as though we are placing our hopes in the upper arm of the K and expecting it to pull the entire Chinese economy upwards [拉动整个中国经济]. This actually diverts us from the real problem. I agree with all the views expressed just now by the previous speakers. However, that upper arm just cannot pull up the base of the economy as a whole. Regarding this issue, we must maintain a strong sense of alertness to crisis.
The first figure is the broad unemployment rate. Using the National Bureau of Statistics’ underlying data, we have reincorporated the people who have been unable to find work over the past two years and are therefore no longer statistically counted as part of the labour force. We term them “discouraged labour force” [受挫劳动人口]. They are not “lying flat”; they still want to find work. If we treat them as unemployed and add them back into both the numerator and denominator, the resulting broad unemployment rate is currently 10.2%. [Note: China’s official surveyed unemployment rate counts only those classified as part of the labour force, a measure from which the long-term unemployed are excluded, leading to an official rate averaging 5.2% from January to May 2026.] There are approximately 24 million members of the long-term discouraged labour force, which is very detrimental to social stability.
The second figure concerns fixed-asset investment. From January to May this year, cumulative fixed-asset investment fell by 4.1%. Private investment and manufacturing investment both registered negative growth, while cumulative fixed-asset investment recorded positive growth of only 0.4% [Note: possibly a transcription error for a fall in manufacturing investment by 0.4% during the same period]. In the statistical history of the People’s Republic, negative growth in fixed-asset investment has occurred only in 1961 and 1967, and the severity and intensity of this current cumulative contraction are unprecedented. We must pay very close attention to these two developments. Unless they are resolved, China will encounter difficulties in accomplishing its various economic objectives and tasks.
Of course, having said that, I need to emphasise that China’s economy has bright prospects and untapped potential. We have complete confidence in this, but the problems must be confronted directly.
If everyone agrees with the judgement above—that the economy’s overall frigidity has already created relatively serious problems for macroeconomic performance—we must next analyse what exactly has caused the economy to remain cold for three consecutive years, or operate below its potential GDP growth rate.
II. The Solution: Firing Up the Old Growth Engines
I shall first offer an overall assessment and then discuss the specific mechanisms.
My overall judgement is that new drivers of growth have not yet been created, while the old sources of growth of the past several decades have already withdrawn [新的增长动力尚未打造出来，而过去几十年的老增长点都已退出]. The flourishing [红红火火] of China’s economy over the past 20 years was not driven principally by household consumption—which accounted for approximately 35–38% of GDP, excluding services consumption here—but by two major engines. The first was large-scale infrastructure investment, sustained for nearly two decades and arguably representing the largest infrastructure programme in the history of the world economy. The second was the overall rise of the property market, which undoubtedly provided an enormous boost to the economy.
Local governments were the pillar supporting infrastructure investment. Our repeated calculations show that local-government infrastructure investment and routine expenditure have together averaged out at 41% of annual GDP. This far exceeds household consumption and has constituted the single largest driving force behind China’s economy. Infrastructure expenditure accounted for approximately 75% or even more of total local-government expenditure. A relatively small amount went towards routine operations, while the great majority was spent on construction. From construction, building materials, migrant labour and land development, to the transfer of income to households via land development, this exerted an enormous upward pull on the Chinese economy.
Today, however, property and infrastructure have both been “extinguished”. Let us begin with property. Lu Ting was entirely correct in what he said earlier about the effects of falling prices. I agree completely with his specific analysis. But I would like to add one more point: the impact of the property downturn on China’s economy has actually been far smaller than we expected three or five years ago, because households have absorbed a large part of the shock. Chinese households hold approximately RMB400 trillion in net housing wealth. Its value has fallen considerably, yet household consumption has not collapsed. That is no easy feat. Ordinary people have shouldered the pressure [老百姓承担了压力]: house prices have fallen, yet everyone has remained relatively calm. This deserves recognition. If this had happened in the United States or another Western country, after house prices had fallen by this much, people would long ago have handed their keys to the banks, stopped repaying their mortgages and “lain flat”. Overall, China has not experienced this problem.
As such, consumption has fallen only to a certain extent. The effect of the property downturn on local fiscal revenue is also far smaller than the impact of local-government debt, to which I shall turn shortly. The property downturn has had a substantial effect on fixed-asset investment—on housing construction—but even this decline is smaller than the effect of the contraction of local governments as a whole.
Therefore, the property market needs to be stabilised as soon as possible. However, the single largest reason for why China’s economy is currently running cold remains local governments.
As I said earlier, local governments accounted for expenditure equivalent to an average of 41% of GDP over the past 20 years. The problem today is that local governments have become a blockage [瘀堵点] in the operation of the economy. How has this “blockage” arisen? Let us begin with households. Increasing amounts of household income are being deposited in banks. The macroeconomic data clearly show that household deposits have increased in recent years. Some of this money has been used to repay mortgages early, but the total has continued to rise. As for companies, investment is weak and corporate leverage has remained broadly stable. Companies have both deposits and loans, and overall have not extracted a net positive in investment resources from the financial system.
III. The Challenge: Blocked Financial Transmission
Nevertheless, aggregate financing to the real economy is still rising. So where has the money gone?
Large volumes of funds are flowing to local governments for debt repayment. We have repeatedly calculated the figures, extrapolating from microeconomic cases to the macroeconomic level. Our conclusion is that total local-government debt has exceeded 100% of GDP. Historically, this borrowing has taken the form of short-term debt, with maturities of five years or even less than seven years. It must be repaid and carries high interest rates. Large volumes of bank and bond-market financing flow to local governments, which are issuing new debt to repay the old debt. Despite the debt-resolution measures of recent years and the special bond quotas provided by the central government, the level of local-government debt has not fallen, but actually risen—because interest rates are too high. Local governments already depend on fiscal transfers from the central government. If they do not repay their debts, they have difficulty maintaining normal operations, never mind paying the interest. As they issue more new debt to repay old debt, the volume of new borrowing greatly exceeds the debt being retired and the total volume of local debt continues to expand.
This has produced the following situation: households are unwilling to borrow to consume; companies are unwilling to convert financial resources into investment; and much of the money borrowed by local governments merely “circulates idly” [空转] as they issue new debt to repay old debt. This is the “blockage” currently visible at the financial level.
In the real economy, local governments’ total expenditure—routine administrative expenses together with capital expenditure—has fallen from 41% to 35% of GDP, reflecting a sharp decline in capital expenditure. Moreover, local governments are using every conceivable means to extract money from companies. They are clawing back tax concessions previously promised to businesses and even collecting taxes in advance. Consequently, local governments have become a heat-absorbing [吸热] part of the Chinese economy. Just as a chip must dissipate heat when it operates, local governments have instead become “heat absorbers”, black holes for [economic] energy. This is the fundamental reason why the economy is currently running cold.
In essence, the transition between old and new growth drivers has not been completed—China has failed to move beyond property and infrastructure onto a new growth path. The specific mechanism is that, after completing their infrastructure projects, local governments are anxious to repay their debts. Interest rates remain high and the more debt they repay, the larger their debts become. They absorb enormous quantities of economic and financial energy without converting them into actual work performed. They also impose numerous burdens on companies through payment arrears, advance tax collection and related practices.
IV. The Mechanism: What Kind of Stimulus?
What, then, should be done? If everyone agrees with the analysis above, I would like to make the following points:
First, we must treat the problem with great seriousness and maintain a strong sense of alertness to crisis. The economy as a whole is running cold. We must not focus exclusively on the small section that is rising, because it cannot pull the entire Chinese economy upwards.
Second, the crux of the problem is the “blockage” created by local-government debt. The answer is therefore readily apparent: the central government, and indeed the state sector as a whole, must increase its leverage. Central-government debt currently amounts to less than 30% of GDP, leaving considerable room for expansion, while local governments possess large volumes of commercial assets that can be monetised. The central government should issue an appropriately larger volume of debt. Approximately RMB12 trillion is planned for this year, but that is far from sufficient. The amount should be doubled or increased still further. After the bonds have been issued, local governments could propose projects for which the central government would allocate funding. Specific uses could include:
Purchasing unsold properties locally and converting them into inexpensive housing or housing with an appropriate rent-to-price ratio, and then packaging them into REITs for commercial operation;
Investing in projects centred on “people”, such as integrating migrant workers into urban communities, with the central government using additional bond issuance to finance transfer payments;
Allowing eligible local governments to apply to the central government and sign undertakings promising to regulate their future borrowing and refrain from issuing debt recklessly, in exchange for replacing their existing high-interest debt.
These measures would significantly increase local governments’ capacity to assist the economic transition and create new sources of profitable growth. At the very least, they would transform local governments from a source of negative energy into a source of positive energy [从负能量转为正能量]. The money could be used to stabilise the property sector, invest in people, strengthen livelihood-related expenditure or provide local consumer subsidies, including by holding concerts and similar events. There are many ways through this. At present, however, local governments do not possess sufficient capacity to drive the economy.
If we can form a consensus around this approach, I believe the Chinese economy can emerge relatively quickly from its present cold and blocked state. I emphasise once again that the Chinese economy has a positive future, bright prospects and untapped potential. Once these problems have been resolved, the economy can rapidly return to an upward development trajectory and realise its growth potential.
That concludes my analysis. I welcome your criticisms and corrections. Thank you.
READ MORE
Strong Supply, Weak Demand, Little Urgency? | Economic Digest: July 2026
“K shaped” is July’s defining buzzword, with a flurry of articles exploring China’s simultaneous boom and bust in different parts of the economy through this alphabetic metaphor.
Zheng Yongnian on Local Government Overreach and the Private Economy
Recent criticism in official publications of economic mismanagement by some local governments may have opened space for sharper lines of enquiry among Chinese scholars. Last month, official statements on optimising the “unified national market” contained disapproval of “hidden and explicit barriers” in the internal market, a nod to local government protectionism, a point several high-level scholars then developed. A Qiushi article early this month by economist Zhang Jun directly criticised local governments for ploughing resources into favoured new industries and industrial park build-outs, without proper consideration of the economic logic in the local context.