Why China’s Football Problem Is an Economic Problem | Li Xunlei

“If these proposals remain forever confined to principles or words on the page and reform is still further delayed, then cause and effect dictates that a crisis will follow.”

Li Xunlei, chief economist at the securities firm Zhongtai International, is among the most engaging proponents of economic rebalancing in China.

In the following article, he cites China’s World Cup qualification woes as among the consequences of a growth model that prioritises fixed-asset investment, but fails to generate sufficient household wealth or employment opportunities. In his telling, the financial insecurity which results from this model intensifies educational competition and pushes children towards pathways with clear career returns. Youth football participation loses out.

There are, in addition, more serious concerns than the poor performance of China’s first eleven. With returns on investment falling and growth in local government debt outpacing GDP growth, Li presses the case for an urgent shift towards growth anchored in household consumption. He argues that failure to carry out such a shift represents an obvious, but neglected, risk factor in China’s economy—what one might also call a “grey rhino”.

Recently, calls for stronger central government-backed stimulus have been widespread among Chinese economists. Li is highly sceptical of these proposals, arguing that fiscal stimulus might address some proximate causes of the slowdown (by boosting headline investment figures and short-term nominal GDP growth) but not the fundamental causes. As he has previously put it, countercyclical stimulus would “smooth the surface” of cyclical trends, but only postpone a worse economic reckoning.

The central government likely recognises this. While it did authorise the release of special-purpose bonds for infrastructure investment earlier in the year, it has discouraged expectations of a stimulus package on anything approaching the scale advocated by many Chinese economists. On the other hand, signs of a genuine pivot towards “investing in people” are also weak. For his part, Li agrees with the thrust of some recent essays in Qiushi on “investing in people”, but is not convinced that they represent actual government policy. Based on recent signalling, it appears that the central government is banking on gradually transitioning the economy to emerging growth areas, particularly artificial intelligence and “high-quality services”, to address weak demand from the supply side.

Li attributes the overriding focus on investment in high-tech to a bias that prioritises immediate challenges (notably, competing technologically with the United States) over the looming, long-term challenge of rebalancing the economy towards household consumption—a task that policymakers may be more tempted to kick down the road. He makes a strong argument, however, that it is success or failure in establishing a more balanced growth model, and not AI diffusion, that will determine China’s future economic prospects.

—James Farquharson

Key Points

1. Economic analysis must distinguish proximate triggers from fundamental causes. Large trends are usually the result of deep, structural forces instead of conspicuous short-term events, though people often fixate on the latter.

2. China’s weakness in football illustrates this distinction. It is common to blame it on corruption in the football association, but this is only the most visible phenomenon—the fundamental issue being China’s small pool of youth participants.

3. Extending the chain of reasoning further, China’s weak youth football participation is socio-economically rooted in a set of institutions that prioritises investment over boosting consumption, employment and welfare.

4. This is because China’s economy is failing to create sufficient employment, and the resulting lack of opportunities pushes families towards a utilitarian focus on activities where rewards are certain, leaving little incentive for participation in sports like football.

5. Inadequate job creation stems from a growth model heavily reliant on investment, with the resulting overcapacity seriously impacting employment, while advanced manufacturing and AI do not create sufficient jobs.

This post was shared last month with paid subscribers.
To receive early access to selected content and full access to Sinification’s archive, consider becoming a paid subscriber. As a non-profit, Sinification depends on your support.

6. This forms a vicious cycle, with weak employment suppressing household consumption, worsening deflation, squeezing profits, discouraging investment and thereby further exacerbating unemployment.

7. Rebalancing would require boosting household incomes through redistribution and increasing job creation through the expansion of the services industry. Consumption vouchers and trade-in schemes, on the other hand, only treat the symptoms.

8. Currently, a fixation on GDP targets and local industrial support encourages duplicated investment and overcapacity. Overbuilt infrastructure will leave local governments with mounting long-term debt burdens, storing up problems for the future.

9. It may be tempting for China to compensate for its private investment shortfall in AI relative to the US through state funding, but it would be more prudent to spend these resources on a genuine pivot to support for households.

10. With public debt growing far faster than GDP, consumption-oriented reform to establish a less debt-reliant growth model should be the fundamental priority. Time is running out, and further delay will lead to a genuine crisis.

The Author

Name: Li Xunlei (李迅雷)
Year of birth: 1963 (age: 62)
Positions: Chief Economist, Zhongtai Financial International Limited; Deputy Chair, China Chief Economist Forum; Member of Central Committee and Deputy Director, Economic Committee, Jiusan Society; Member of Index Expert Committee, Shanghai Stock Exchange
Formerly: Chief Economist and Head of Research, Zhongtai Securities; Deputy General Manager and Chief Economist, Haitong Securities; Chief Economist, Guotai Junan Securities; Member of Standing Committee and Financial and Economic Affairs Committee, Shanghai Municipal People’s Congress
Research focus: Macroeconomics; finance and capital markets; asset allocation
Education: Undergraduate degree in Statistics, Shanghai University of Finance and Economics (1985); Master’s degree in Economics, Shanghai University of Finance and Economics (1991)

FUNDAMENTAL CAUSES AND PROXIMATE CAUSES: REFLECTIONS ON CHINA’S FAILURE TO QUALIFY FOR THE WORLD CUP
By Li Xunlei (李迅雷)
Published on his public account on 21 June
Thank you to Li Xunlei for his permission to share this article
Translation by James Farquharson

N.B. The first section of Li Xunlei’s article has been slightly abridged in order to focus on his economic arguments. Any omitted paragraphs are marked with an ellipsis.

I. Introduction: Networked Causality

The mainstream concept of causality in the West is that of linear determinism. A single cause produces a single effect, or what Hume called “constant conjunction”. Chinese conceptions of causality, by contrast, are networked and cyclical. Buddhism speaks of the “conjunction of causes and conditions” [因缘和合], whereby a single outcome can have several causes, or several causes can produce several outcomes. Daoism teaches that “good fortune perches on disaster, while disaster crouches beneath good fortune” [祸福倚伏], or put differently, that opposite states may be interchanged. Confucianism speaks of good and evil meeting their respective just deserts [善恶报应].

I am often asked questions like whether an interest-rate increase by the US Federal Reserve will cause the stock-market bubble to burst, whether the appearance of a “golden cross” heralds a major bull market, or whether an uptick in transactions means that the property market has bottomed out and will recover. [Note: A “golden cross” is when a short-term moving average of a market index crosses over the long-term moving average.] I have noticed that the person asking such a question tends to interpret some minor, incidental event as a cause from which they can infer some enormous outcome. As I see it, the causal relationships are genuine, but a single outcome generally has multiple causes. We therefore need to distinguish between “fundamental causes” [大因果] and “proximate causes” [小因果].

The football World Cup is currently under way and, regrettably, China has once again failed to qualify. There has been an all-consuming discussion over why, for so long, Chinese football has been unable to break out of Asia and establish itself on the world stage. Many pin the blame on the Chinese Football Association. But I can’t help but think, that association has been around for more than seventy years, and China still cannot even be regarded as a top footballing nation within Asia. Can all this really be down to faults with the association?

[…]

II. The Economic Causes of China’s Footballing Struggles

Looking at the geographical distribution of the 48 countries participating in the 2026 World Cup, East Asia is represented by only two countries: Japan and South Korea. They are neighbouring countries with exceptionally close historical ties, while Asia’s other participating countries are concentrated in West and Central Asia. Not a single country or territory in South or Southeast Asia qualified. Asia accounts for around 60% of the world’s population, yet the eight Asian participants have a combined population of only 690 million. [Note: The combined population of Japan, Iran, South Korea, Iraq, Uzbekistan, Saudi Arabia, Jordan and Qatar is actually lower, at 402.9 million according to UN World Population Prospects 2024.] In other words, Asian countries and territories that make up over half of the world’s population are absent from the World Cup. In particular, Asian countries with populations exceeding 100 million—including India, China, Indonesia, Pakistan, Vietnam, Bangladesh and the Philippines—have failed to qualify almost every time.

By contrast, the 40% of the world’s population that lives outside Asia is represented by 40 participating countries, of which Europe has the largest number, at sixteen. In terms of the predominant ethnic groups in the participating countries, Japan and South Korea are the only representatives of what has traditionally been called the “yellow race”; all the others are predominantly white, black or brown. A Chinese proverb holds that “every land and climate nurtures its own people” [一方水土养一方人]. As a highly confrontational competitive sport, might football, like other competitive sports, be one in which different ethnic groups possess different strengths and weaknesses?

[…]

[However], how do we explain why Chinese football is so much weaker than that of Japan and South Korea, given that their geography, diet and even cultural inheritance are broadly similar to that of north-eastern and eastern China? On the sole basis of the number of Olympic gold medals won, China ranks first in Asia, Japan second and South Korea third, with an enormous gap separating these three from all other Asian countries and territories. This suggests that China, Japan and South Korea are the most directly comparable cases.

In reality, the gap in footballing standards between China and its two neighbours comes down to the sport’s participation rate. Statistics show that youth-football participation in China is far below that of Japan and South Korea. China’s pool of registered youth players is extremely small and has remained at around 100,000 to 110,000 for years, standing at 109,200 in 2025. Japan has a vast pool of approximately 400,000 to 1.2 million registered youth players, while South Korea has around 500,000 to 800,000.

[Note: Proportionally speaking, Li’s argument holds water. Nonetheless, South Korea’s entire registered football population is around 110,000–120,000. The 500,000–800,000 quoted likely reflects broad recreational participation. Japan’s registered youth pool is about 600,000–700,000.]

Regarding football participation within the general population, China has relatively few amateur players and a shortage of grassroots football pitches. School football is exceptionally widespread in Japan, where competitions such as the national high-school football championship attract interest comparable to that enjoyed by professional leagues, and there exists a deeply rooted footballing culture. In South Korea, around 67% of the population is reportedly interested in football and the sport enjoys high national standing. The fundamental causes of China’s low participation rate are the small number of grassroots participants and serious gaps in the talent pipeline. By contrast, as a result of the integration of sport and education, extensive school-football programmes and a deep footballing culture, Japan and South Korea have reached a situation whereby mass participation begets professionalisation, creating an enormous pool from which to select talented players.

Why, then, is the rate of school football participation so low in China? The essential reason is still the impact of “involution”. [Note: Neijuan 内卷, or intense competition with diminishing marginal returns from each extra increment of invested time and resources.] From kindergarten onwards, parents start off this involutionary competition by planning for their children’s eventual employment. Nowadays, a growing proportion of university graduates is proceeding to postgraduate study, reaching 20.4% in 2024 (including the 2% who studied abroad). In Japan, the corresponding proportion of undergraduates continuing on to a master’s degree after graduation is relatively low, at around 11.8% in recent years. The fundamental explanation for China’s low football-participation rate is the employment pressure that young people expect to face. Some may object that badminton, table tennis, swimming and piano playing are all highly popular among Chinese children, but this relates to another issue—that of utilitarianism. As these are all activities in which China [already] excels and may help a student gain bonus points or special consideration in the senior-high-school or university entrance examinations, [they are more likely to attract participants].

In comparison, the employment rate among Japanese university graduates has reportedly reached 98% over the past two years. Even if the true figure is not quite so high, it is unlikely to be much lower, leaving undergraduates with little incentive to pursue postgraduate study. China’s university-student population, on the other hand, will continue to grow. According to a report by the Standing Committee of the National People’s Congress, the number of students enrolled in Chinese higher education will peak in 2032. This means that university graduates are likely to be facing intense employment pressure for at least the next six years.

In recent years, the Ministry of Education has introduced numerous measures to reduce the burdens on primary- and secondary-school pupils, but involution among parents has not diminished. The central problem remains the pressure surrounding their children’s future employment. Participation is low not only in football but is also declining in other competitive sports. To take an example with which I am familiar, participation in Go among Chinese children is likewise falling, while South Korea remains well ahead.

In summary, the institutional defects and localised corruption frequently discussed in relation to Chinese football are all matters of “proximate causation”. At the level of “fundamental causation”, East Asian countries do indeed lack an innate physical advantage in football. Otherwise, it would be difficult to explain how there are three qualifying nations in this year’s World Cup with populations of only around 150,000 to 500,000 or so. [Note: Of teams drawn from populations of roughly this size, only Cabo Verde and Curaçao qualified in 2026; Iceland qualified in 2018]. Nonetheless, South Korea has qualified for every World Cup since 1986 and Japan for every tournament since 1998, both becoming major footballing nations. For its part, China is also shaped by Confucian culture, its people are no less hardworking and competitive than those of Japan and South Korea, and—to boot—it has the advantage of a sporting system supported by the mobilising power of the state [举国体育]. Why, then, has it become a weak footballing nation? The “fundamental cause” here is the decline in youth-football participation, driven principally by growing employment pressure on young people and intensifying involutionary competition in education.

Otherwise, it would be hard to explain why the footballing gap between China, on the one hand, and Japan and South Korea on the other has widened since the 1980s and 1990s. This is despite enormous improvements in Chinese people’s nutrition over the past thirty to forty years. According to data for 2024–25 from the UN Food and Agriculture Organisation, the average daily calorie intake in the three countries ranks as follows: China at 3,100 kilocalories per person; South Korea at 2,750; and Japan at 2,705.

III. The Institutional Drivers of High Unemployment

If employment pressure were to ease in the future, would involutionary competition in education also subside? In causal terms, it should. Naturally, this will also require certain “conditions” [缘]: favourable timing, conducive circumstances and people pulling together [天时地利人和].

Logically speaking, China’s working-age population has been declining since 2011. Moreover, as normally understood, annual GDP growth generates additional employment, and China has been reporting over 12 million new jobs each year, or 120 million over the decade. Why, then, have the employment difficulties facing young people proved so intractable? And how has the number of people in flexible employment reportedly reached 320 million, or around 40% of the employed population?

Getting an answer to that question requires us to ask what kind of economic-development model is most suited to generating employment. First, we need to be clear about one point: a growth model driven by artificial intelligence simply cannot drive employment. AI depends on computing power, which consumes electricity, whereas human cognition consumes calories. When computing power replaces brainpower—i.e. when electricity consumption replaces calorie consumption—job displacement will increase.

In China, total employment in manufacturing began to decline in 2013, indicating that the country had started moving from labour-intensive towards capital-intensive manufacturing. Over the past decade or more, China’s share of global manufacturing value added has risen by around one percentage point each year, and now stands at nearly one-third. China’s economic growth over this period has depended increasingly on capital inputs rather than improvements in total factor productivity. As private investment growth shifts into negative territory, these capital inputs have become increasingly reliant on government borrowing.

Investment activity can certainly create jobs, but the creation of more than 12 million new jobs each year does not mean a net increase of 12 million—particularly when the number of labour-intensive manufacturing firms is declining and swathes of traditional businesses are going bankrupt. The sector capable of accommodating the most employment is naturally the service sector. Services account for more than 80% of total employment in the United States and more than 70% in both Japan and Germany.

Countries in which services account for a high proportion of employment share two characteristics. First, their household disposable income represents a relatively high share of GDP. Secondly, their Engel coefficients are relatively low. [Note: The Engel coefficient, developed by German statistician Ernst Engel in 1857, measures the share of household expenditure devoted to food, which generally falls as incomes rise.] Put simply, ordinary people must have money and be able to spend more of their discretionary income on things other than food, such as leisure, tourism, sport, wellness, culture and entertainment.

Traditional Chinese values encourage thrift and regard eating, drinking and entertainment as frivolous and wasteful. As such, China’s household saving rate is extremely high, with household bank deposits alone reaching RMB170 trillion. If one simply invests without consuming, overcapacity is inevitable. Overcapacity then depresses prices, leaving firms reluctant to invest and banks struggling to lend, thereby impeding economic circulation and affecting employment. The blocked labour market, in turn, then further exacerbates insufficient consumption.

On 18 June, the Qiushi website ran a commentary entitled “Making Greater Efforts to Stimulate Consumption” [《以更大力度提振消费》]. It examined the 0.6% year-on-year decline in total retail sales of consumer goods in May and the fall in cumulative growth for January to May to 1.4%. It offered four principal recommendations: first, accelerate the formulation and implementation of a plan to increase the incomes of urban and rural residents; secondly, expand the economy’s capacity to generate employment through more proactive employment policies; thirdly, raise household disposable income through an improved redistribution system; and fourthly, alleviate households’ reticence to consume by establishing a more robust social-security system.

Each of these four recommendations goes to the heart of the matter [切中要害], identifying the “fundamental causes” behind the decline in consumption growth over recent years. They are worlds away from measures such as consumption vouchers, consumer loans and trade-in schemes, all of which belong to the realm of “proximate causation”. If the four proposals can be implemented, consumption will certainly receive an enormous boost. Naturally, though, planning accounts for 10% of the work and implementation for 90% [一分部署九分落实]. Implementation is what matters in the end.

Once a period of rapid growth ends, every economy will develop structural problems to a greater or lesser extent. Those structural problems can manifest themselves at a variety of levels, [revealing the underlying trend] just as “one falling leaf can herald the arrival of autumn” [一叶知秋]. For example, performance across China’s A-share market has become increasingly uneven in recent years. Although the price-to-earnings ratio of the CSI 300 is currently very low, return on equity is continuing to decline [Note: The CSI 300 is an index comprising 300 of the largest and most actively traded companies listed in Shanghai and Shenzhen]. This clearly does not support the prospect of a market-wide bull run.

Why has return on equity declined? From the perspective of “proximate causation”, there is no shortage of technical reasons to cite, such as insufficient product innovation, lax financial management, weak markets or companies’ failure to develop a second growth curve in time. From the perspective of “fundamental causation”, however, the causes are mostly institutional. To meet GDP targets and carry out their goals of transitioning from old to new growth drivers [新旧动能转换], local governments establish industrial guidance funds [产业引导基金], using them to attract investment [招商引资] into new projects. The resulting duplication in construction ultimately leads to nationwide overcapacity. Between 2011 and 2025, the cumulative increase in the producer price index was almost zero. From 2023 to 2025, China’s export price index in US-dollar terms fell by 19%. It’s possible to see, then, how these weaker prices could weigh on the profit growth of listed companies.

[Note: Industrial guidance funds are government-backed investment vehicles, often structured as public–private funds, formally introduced as a policy measure by the central government in 2005.]

It is therefore best not to speak too broadly of expanding domestic demand, as domestic demand includes both consumption and investment. The present structural contradiction is one of overcapacity and sluggish consumption, so the priority should be stimulating consumption. In practice, however, investment produces immediate and visible results, whereas in the case of stimulating consumption, policymakers lack an effective lever [抓手]. The basis for expanding domestic demand thus chiefly takes the form expanding investment, thereby further enlarging production, storage and transport capacity, and exacerbating surpluses in all three.

IV. The Risks of Infrastructure Overbuild

Everyone is likely aware of overcapacity in manufacturing, while the issue of overcapacity in transport is more readily overlooked. The rapid growth in government debt in recent years is actually connected to the excessively rapid expansion of transport capacity. This is because infrastructure investment is usually led by the government sector, and local governments in particular.

China accounts for around 17.5% of the world’s population, a figure that could fall to approximately 15% within the next twelve years. Yet China currently has more than 70% of the world’s high-speed railway mileage, more than 40% of its motorway mileage and more than 40% of its total metro mileage. Its global share of transport capacity far exceeds its share of the global population.

Take metro systems, for example. Their extensive development has already caused bus passenger numbers to fall sharply, even though metros are the most expensive form of rail transport to build. Under Ministry of Transport requirements, an existing metro network must have an average daily passenger intensity of at least 7,000 journeys per kilometre before a city may apply to build a new line. Yet data from February this year show that, among the 54 Chinese cities with urban rail systems, only 10 met this requirement: Beijing, Shanghai, Guangzhou, Shenzhen, Chengdu, Chongqing, Xi’an, Changsha, Harbin and Lanzhou. But these account for just 18.5% of the total [Note: In terms of the number of cities, rather than mileage of the networks].

In future, around 80% of cities will experience net population outflows as people concentrate in large cities, while China’s total population is estimated to fall to around 600 million by 2100. A surplus of costly metro infrastructure is therefore inevitable. Annual metro operating and maintenance costs amount to RMB10–15 million per kilometre [Note: According to current levels documented in the 2025 China Association of Metros annual report], imposing a substantial debt burden on local governments.

Investment should therefore be undertaken with a long-term vision in mind [着眼于长远]. Although it might boost GDP in the short term, it can turn into a heavy debt burden over the longer term.

V. Choosing between Consumption and AI Investment

Since last year, even fixed-asset investment has recorded negative growth, with private investment contracting by more than 7% between January and May this year. There are two reasons for this. First, returns on investment are falling and investment may result in losses. Secondly, local governments invest by taking on debt, raising their borrowing costs and increasing the pressure of interest repayments. Official data show that China’s real cumulative GDP growth between 2021 and 2025 was 35.3%, while the outstanding debt of the central and local governments grew by a cumulative 106.3% over the same period—approximately three times the cumulative GDP growth rate.

China’s economy therefore needs high-quality development, and urgently. Put differently, it needs relatively rapid economic growth, achieved through slower debt growth. However, one present phenomenon requires special attention. While the great majority of AI investment in the United States is financed by private capital, in China it is funded mainly by public finances. If China’s AI investment were also financed by private capital, as in the United States, private-investment growth would be boosted into positive territory, and public funds could be redirected towards stimulating consumption and supporting social security.

The problem is that, in the age of AI, technological progress may proceed at an exponential rate, and if capital expenditure fails to keep pace today, the gap between China and the United States in AI could widen further. How, then, should China choose between stimulating consumption and investing in AI? At this historical juncture, we ought to pay greater attention to “fundamental causation” and avoid becoming preoccupied with “proximate causation”. After all, the latter affects the present moment, whereas the former determines the future.

People habitually assign excessive weight to events occurring in the present, mistaking “proximate causes” for the grand narrative itself. Alternatively, they may realise that they should take a long-term view, and yet still remain constantly occupied with immediate problems and trust that “the boat will straighten itself out once it reaches the bridge” [船到桥头自会直]. However, contrary to one’s hopes, problems keep on accumulating, while the time left to act rapidly runs out.

In my view, the four recommendations proposed in the Qiushi commentary are in fact four reform-oriented measures and as such belong to the realm of “fundamental causation”. But if these proposals remain forever confined to principles or words on the page and reform is still further delayed, then cause and effect dictates that a crisis will follow [危机则符合因果律]. This is why I am always fond of saying: “Trust in logic, not miracles.”

---

READ MORE

---

---