Subsidy-Dependent Sectors: Where Policy Enables Demand

China's new-energy vehicle (NEV) sector is frequently cited as the clearest global example of state subsidy building an internationally competitive industry from a standing start, and the underlying policy commitment was real and sustained: national and local purchase subsidies date back to pilot programs beginning around 2009, scaling through the 2010s, before the national program's scheduled phase-out completed at the end of 2022. The precise scale, however, is more modest than some retrospective commentary suggests — direct central government NEV subsidies between 2010 and the end of 2022 totaled approximately ¥152.1 billion (roughly $21 billion), supporting around 3.17 million vehicles; including the value of tax breaks alongside direct subsidies over a slightly longer 2009–2022 window, the combined figure rises to more than ¥200 billion (approximately $28 billion). Either figure is a real, sustained, multi-decade policy commitment — but it is meaningfully smaller than the scale that would be implied by treating NEV subsidy spend as comparable in size to, say, China's semiconductor Big Fund program, and any comparative claim about subsidy intensity across sectors should reflect that distinction accurately.

This sustained demand-side support ran in parallel with supply-side industrial policy — preferential financing, land grants, and local government co-investment supporting battery and vehicle manufacturing capacity buildout — that directly underwrote the early scale-up of companies including BYD and CATL well before either achieved the global market position they hold today. The relatively modest direct-subsidy total, set against the scale of the industry China's NEV sector has become, is itself informative: it suggests execution, manufacturing scale advantages, and the supply-side industrial policy layer did at least as much work as the headline purchase subsidy in producing the sector's current global competitiveness.

Direct purchase subsidies were not simply eliminated at the end of 2022 — they were substantially replaced by an NEV purchase tax exemption (waiving the standard 10% vehicle purchase tax for qualifying NEVs), a policy that has itself been extended multiple times beyond its original planned expiration, reflecting continued state commitment to demand-side support even after the more visible direct-subsidy program ended. A less visible but equally important condition attached to much of this sector's state support has been localization content requirements — battery and component sourcing thresholds that had to be met domestically to qualify for various support programs — which shaped supply chain structure well beyond what pure market economics would have dictated on its own.

The semiconductor sector's state-support intensity, discussed in more depth in relation to the "Big Fund" program elsewhere in this series, follows the same demand-creation-through-subsidy logic but at a dramatically larger scale relative to sector size — Big Fund III alone (¥344 billion, ~$47.5 billion) committed more capital in a single 2024 fundraising than the entire NEV sector received in direct subsidies over 13 years, reflecting both the semiconductor industry's far higher capital intensity and its more acute status as a national-security priority given US export controls. Solar and broader renewables followed a subsidy trajectory (feed-in-tariff support driving manufacturing capacity expansion well ahead of pure demand economics) similar in shape to the NEV story but roughly a decade earlier in its timeline, and is now a more mature, globally dominant Chinese manufacturing sector with comparatively less ongoing direct subsidy dependency.

Solar manufacturing's subsidy history is worth its own brief treatment because it is the sector that most directly foreshadowed the NEV playbook roughly a decade earlier. Domestic feed-in-tariff support (guaranteeing a premium price for solar-generated electricity fed into the grid) drove a first wave of Chinese solar manufacturing capacity expansion in the late 2000s and early 2010s, well ahead of what domestic or even global demand alone would have justified at the time — a pattern that produced a painful overcapacity and price-crash cycle for solar manufacturers around 2011–2013 (a cautionary precedent worth remembering when evaluating any subsidy-driven capacity buildout, including today's semiconductor and battery expansion) before the sector matured into the genuinely globally dominant, largely subsidy-independent manufacturing base it represents today. The lesson for a foreign investor in a currently subsidy-supported sector: state-driven capacity buildout ahead of demand has, in China's own solar precedent, produced both an eventual global manufacturing champion and a painful intermediate overcapacity shakeout — and it is not obvious in advance which sector currently receiving similar support will follow the same eventual trajectory versus stalling at the overcapacity stage.

Within the semiconductor sector, localization requirements have a specific and increasingly consequential focus on manufacturing equipment rather than just chip design or fabrication capacity — given US and allied export controls on advanced lithography and other fabrication equipment, Chinese state support (including guidance-fund capital) has directed meaningful resources specifically toward domestic semiconductor equipment manufacturers, an area where China has historically lagged furthest behind global leaders like ASML, Applied Materials, and Tokyo Electron. This equipment-specific push is a distinct and, in some respects, more strategically urgent priority than chip design or fabrication capacity alone, because equipment self-sufficiency is what would ultimately insulate Chinese chipmakers from further tightening of foreign export controls — a foreign investor evaluating a Chinese semiconductor equipment company specifically, as opposed to a chip designer or fabricator, should expect an even more concentrated and urgent state-support posture than the sector average this fact otherwise describes.