China Pursues Mass Automation While Pledging to Protect Worker Employment

In May 2026, the Hangzhou Intermediate People's Court ruled that a technology company had illegally terminated a quality-assurance worker named Zhou after it replaced his job functions with artificial intelligence [1]. The company had first demoted Zhou and slashed his salary by 40%. When he refused to accept the new terms, he was fired. The court ordered compensation, establishing a precedent: in China, you cannot fire a worker simply because a machine can do his job.

That same month, the Chinese government's National Venture Capital Guidance Fund — a $138 billion, 20-year war chest — continued deploying capital into the robotics and AI startups building the very machines that made Zhou's job redundant [2].

This is the central tension in China's industrial policy: the world's most aggressive automation campaign, paired with an expanding legal architecture designed to prevent the mass layoffs that automation typically produces. Whether this constitutes a contradiction or a coherent strategy depends on a demographic clock that Beijing is watching more closely than any production metric.

The Scale of the Machine

China's robot density — the number of industrial robots per 10,000 manufacturing employees — has risen from 49 in 2015, when Beijing launched its "Made in China 2025" initiative, to 470 in 2023 [3]. That nearly tenfold increase vaulted China past both Germany (429) and Japan (419), leaving only South Korea (1,012) and Singapore (770) ahead [3].

Source: International Federation of Robotics

Data as of Nov 20, 2024CSV

In 2024, Chinese factories installed 295,000 industrial robots, a record that accounted for 54% of all robots installed worldwide [4]. The country's operational stock now exceeds 2 million units — roughly 4.5 times Japan's total and more than the combined stock of the United States, Germany, Japan, and South Korea [4].

Source: International Federation of Robotics

Data as of Nov 20, 2024CSV

Domestic manufacturers have captured 57% of the Chinese market, up from approximately 28% a decade ago [5]. "For the first time, Chinese manufacturers have sold more than foreign suppliers in their home country," IFR President Takayuki Ito noted in the 2025 World Robotics Report [4]. China also holds two-thirds of the world's effective robotics patents — roughly 190,000 by August 2024 [6].

Source: IFR World Robotics Report

Data as of Sep 28, 2025CSV

The expansion is no longer confined to automotive assembly lines. Textile factories saw a 29% year-over-year increase in robot installations in 2024. Food and beverage processing surged 86%. Electronics manufacturing added 83,000 units, growing at 16% annually since 2019 [7]. In Guangdong province alone, robot output rose from 44,700 units in 2019 to 246,800 — a fivefold increase [8].

The Workforce That Isn't There

Critics who frame China's automation push as a jobs-destroying policy frequently understate a structural reality: China's working-age population is in irreversible decline. The cohort aged 16–59 peaked in 2013 and has since fallen from 896 million to 876 million [9]. The fertility rate stands at 1.09 births per woman, roughly half the 2.1 replacement level [10]. In 2022, more Chinese people died than were born for the first time since 1961 [11].

Long-term projections are stark. Apollo Global Management estimates that China's working-age population will shrink from 900 million to 250 million over coming decades [12]. The World Economic Forum projects an annual shortfall of 11.8 million workers over the next decade [11]. The IMF expects growth to slow to an average of 3% after 2027, driven partly by demographic headwinds amounting to roughly 1% annual GDP drag [9].

From Beijing's perspective, automation is not displacing workers who would otherwise have jobs. It is filling positions that a shrinking labor pool cannot. This framing — robots as demographic compensation rather than labor substitution — is central to the government's public messaging and shapes its policy design [10].

The Legal Architecture

China's labor law requires that any "mass layoff" — defined as cutting 20 or more employees or 10% of the total workforce — be preceded by 30 days' notice, formal consultation with the enterprise labor union, a government filing, and specific economic justification [13]. Severance is set at a minimum of one month's average salary per year of service [14].

The May 2026 Hangzhou ruling extended these protections into the AI era, building on a December 2025 decision where a separate court found that a mapping company's implementation of AI did not meet the legal threshold for terminating an employee [1]. The emerging judicial principle: companies cannot unilaterally reduce salaries or terminate workers due to technological advancement alone. They must negotiate, offer retraining, or provide reassignment first [1].

Enforcement, however, varies. State-owned enterprises operate under direct party oversight and face explicit "social stability" metrics that discourage layoffs [13]. Private firms, particularly in export-oriented manufacturing hubs like Dongguan and Shenzhen, face less direct scrutiny. The practical result is a two-track system: SOEs retain workers even when automation renders positions redundant, while private manufacturers rely on natural attrition — simply not replacing workers who leave voluntarily — to shrink headcounts without triggering legal thresholds [7].

This approach has produced measurable results. China's manufacturing workforce fell from approximately 115 million in 2013 to below 85 million in 2025, a loss of over 30 million jobs, even as exports hit record highs [7]. Yet the official unemployment rate has remained stable at roughly 4.6% [15]. The mechanism: service-sector absorption. China's service sector now employs 50% of the national workforce and accounts for 61.7% of GDP [16].

Source: World Bank Open Data

Data as of Dec 31, 2025CSV

Where the Money Goes

The scale of state capital directed at automation dwarfs comparable Western programs. In March 2025, the National Development and Reform Commission launched the National Venture Capital Guidance Fund with 1 trillion yuan ($138 billion) earmarked over 20 years for robotics, AI, and advanced manufacturing [2]. This sits atop an existing ecosystem of city- and province-level subsidies.

Beijing established a 10-billion-yuan ($1.4 billion) municipal robotics fund. Guangzhou matched it. Shenzhen committed $630 million to an AI and robotics package covering up to 60% of individual business costs [17]. At least 21 cities and 5 provinces offer direct subsidies covering up to 30% of robot purchase costs for small manufacturers [6].

Source: Multiple Sources

Data as of Mar 1, 2025CSV

For comparison, the U.S. CHIPS and Science Act allocated $52.7 billion for semiconductor manufacturing and $13 billion for R&D and workforce training [18]. The EU Chips Act committed approximately €43 billion, though actual European spending is expected to reach €20–30 billion by 2030 [19]. South Korea's semiconductor commitment — $400 billion through 2030 — is the largest single-country program, but it is narrowly focused on chips rather than the broader robotics and AI manufacturing stack that China's fund targets [19].

China's cost advantages compound the spending gap. Chinese humanoid robots are priced at $36,000–$62,500, compared to over $120,000 for Boston Dynamics' Atlas [17]. Unitree's G1 model sells for approximately $11,650 [17]. Material costs for robotic arms in China are less than half those in the United States, according to SemiAnalysis [6]. In 2025, Chinese manufacturers produced 12,800 humanoid robots — 90% of the global total [20].

Lessons from the Coal Cuts

China has managed large-scale industrial displacement before. Between 2016 and 2020, Beijing's campaign to reduce overcapacity in coal and steel targeted 1.8 million workers for redeployment [21]. The government established a 100-billion-yuan ($14 billion) Industrial Special Fund and pledged an additional 10 billion yuan for worker retraining [21].

The outcomes were mixed. Steel capacity cuts exceeded targets: 65 million tons closed in 2016 alone against a 45-million-ton goal, with 280,000 steel workers eventually redeployed [22]. Mining employment fell from 6.4 million in 2013 to 4.1 million in 2018 [22].

But a study by the China Association for Promoting Democracy found that many displaced workers struggled to find new employment because they lacked transferable skills, and vocational training programs were often restricted to recent graduates rather than the older workers who needed them most [22]. The gap between policy design and ground-level execution — a recurring theme in Chinese industrial transitions — is now repeating in the automation era.

The Skills Gap at the Factory Gate

China's 10 strategic manufacturing industries need an estimated 62 million skilled workers by the end of 2025. Nearly half that demand — approximately 30 million positions — is expected to go unfilled [11]. The AI sector alone faces a shortage of 5 million qualified workers [20]. Youth unemployment hovers around 17%, yet highly skilled workers constitute only about 8% of China's labor force [11].

The mismatch is partly structural. China produces over 10 million university graduates annually and leads globally in STEM degree output, but university curricula emphasize theory over the practical skills needed to operate automated production lines [11]. The Chinese Academy of Social Sciences has described manufacturing's "notoriously high turnover rate" as creating a "hire-and-replace rather than train-and-retain" culture [11].

State Council regulations require companies to allocate at least 2.5% of payroll to vocational training. Most firms spend below 0.5% — less than half what U.S. and U.K. companies invest per worker [11]. Guangdong province has pledged 5 billion yuan over three years to reskill 3 million industrial workers, but trainers on the ground report limited engagement. "When it's mandatory, they show up, but they don't engage," said Lao Youqian, a vocational trainer in Foshan [8].

Multinationals and the Party Committee

Foxconn's 2016 decision to replace 60,000 workers with robots at its Kunshan factory remains the most cited case of automation-driven displacement in China [23]. By 2018, the company had deployed over 80,000 industrial robot units across its facilities, automating roughly 30% of operations [23]. BYD, meanwhile, hired 280,000 workers in a single year to reach a workforce of 570,000 in 2022 — while simultaneously scaling robotic battery production and vehicle assembly [7].

The pattern among multinationals operating in China is to automate aggressively while avoiding headline layoff numbers. Nearly 80% of companies in Foshan's factory belt expected to adopt automation by year's end, with production lines shrinking by approximately 15% when automation is introduced [8]. Digital upgrades delivered a 16% productivity increase and 17% cost reduction across surveyed firms [8].

Local party committees have generally accelerated rather than slowed automation plans, viewing productivity gains as aligned with national objectives. The friction emerges not at the enterprise level but at the municipal level, where cities dependent on manufacturing employment face population losses. Towns in Dongguan and Guangzhou have experienced 20–30% population reductions compared to a decade earlier, as workers who are not replaced migrate elsewhere [7].

The Western Policy Question

If China manages to automate its manufacturing sector while maintaining nominal employment stability — absorbing displaced workers into services, platform economy jobs, and state-directed hiring programs — it presents a challenge to the Western assumption that automation inevitably destroys net employment.

The Chinese model, however, rests on conditions that are difficult to replicate: a shrinking working-age population that reduces the denominator of the employment equation, a one-party state capable of directing capital at scale without legislative gridlock, a legal system that can simultaneously fund automation and prohibit the layoffs it causes, and a service sector expanding fast enough to absorb tens of millions of displaced manufacturing workers.

Source: ILOSTAT — International Labour Organization

Data as of Dec 31, 2024CSV

The service sector has absorbed at least 20 million former manufacturing workers into platform economy roles — ride-hailing, food delivery, livestream commerce [7]. But wage growth in these sectors is stagnant, and the work offers less stability than factory employment [16]. The question is whether this represents genuine re-employment or a reclassification of underemployment.

For Western policymakers, the Chinese experience suggests that the timeline of automation-driven job loss matters as much as its magnitude. A country losing 30 million manufacturing jobs over 12 years while its working-age population simultaneously contracts by 20 million produces a different political outcome than one where the same job losses occur against a stable or growing labor force. The Chinese model does not disprove the claim that automation destroys jobs. It demonstrates that demographic decline can mask the destruction — at least for a time.

What Comes Next

China's 15th Five-Year Plan (2026–2030) lists embodied AI — robots that can perceive and interact with physical environments — as a priority technology [17]. Approximately 150 companies are now manufacturing humanoid robots, and price competition is intensifying [17]. MERICS, the European think tank, has warned that "European firms could easily find themselves facing a repeat of the race for EVs, outpaced by China's blend of industrial capacity and state support" [17].

The bet Beijing is making is that it can run two clocks simultaneously: an automation clock that accelerates factory productivity and a demographic clock that naturally reduces the workforce needing those factory jobs. If the clocks stay synchronized — if attrition outpaces displacement — the strategy works without the social instability that accompanied the coal and steel cuts. If automation moves faster than demographics, the 300 million migrant workers with minimal social safety nets become a political problem that no court ruling can contain [20].

Chen Zhen, a 38-year-old worker in Foshan's factory belt, put it plainly: "Automation is evolving fast. If we don't keep up, we'll be eliminated" [8]. The question for Beijing is whether "keeping up" means retraining workers or simply waiting for them to retire.