China's New Innovation Integration Opinion: What It Actually Changes for Metals and Materials

China's two central offices published a joint 意见 (guiding opinion) on 21 October on integrating technological and industrial innovation. It is a framework document — no numeric targets, no budget line, no compliance deadlines — and it does not mention steel once. Read only for its sector list, it looks like boilerplate. Read for its governance mechanics, it contains several changes that will shape how state-funded materials and metallurgical R&D is actually selected, paid for and judged over the next five-year cycle.

What the document is, and what it is not

The text was released by Xinhua on 21 October, with the approval of the CPC Central Committee and the State Council (经党中央、国务院同意). It runs to eight sections and seventeen numbered tasks. It is a 意见, not a regulation or a statute: binding detail — ratios, catalogue revisions, fund rules — will come later from ministries such as MIIT, NDRC, MOST and MOF, and none of those implementing documents are referenced in the text.

Three properties matter for anyone trying to trade or invest off the back of it:

• It is cross-sector, not industrial-structure policy. There is no capacity, output, energy-consumption or decarbonisation language. Nothing on iron ore, coking coal, EAF share or capacity replacement.
• It is a supply-side innovation document with a demand-side hook. The distinctive mechanism is "scenario" (场景) opening — local governments are told to open administrative services, smart-city and rural-construction use cases to new products.
• Its centre of gravity is the enterprise. The stated core is an "enterprise-led, multi-actor industry-academia-research integrated innovation system" (企业主导、各主体协同的产学研融通创新体系).

The four mechanisms that matter most

1. Money follows application, and enterprises hold the pen. National S&T programmes are to be given an explicit industrial-application orientation (产业应用导向). Enterprise needs are to be collected on a routine basis; more experts from companies and venture-capital firms are to sit on guideline-drafting, project-approval and acceptance panels. Where an industrial application goal is clear, projects should "in principle" be enterprise-led, with the enterprise also driving commercialisation. Acceptance is to be judged by a two-part test: "application assesses the technology, the market assesses the product" (应用考核技术、市场考核产品). "Open-call" (揭榜挂帅), "owner" (业主制), milestone management and staged disbursement are confirmed as the organising toolkit.

2. SOE R&D gets partial insulation from profit metrics. Two provisions stand out for state-owned metals producers. SOEs are encouraged to set up R&D reserve funds that are separately accounted, exempt from value-preservation-and-appreciation assessment (免于增值保值考核), and carry error tolerance. Separately, the ratio at which joint R&D spend with universities and institutes is added back as deemed profit (视同利润加回) is to rise. For a sector where the largest flat-product, special-steel and alloy producers are SOEs and where R&D intensity is judged against a profit target, this is the most operationally consequential language in the document.

3. A tax lever for smaller innovators. The opinion states that the R&D expense super-deduction ratio (研发费用加计扣除比例) for technology-based SMEs will be raised. This is a stated intention, not an enacted rate — the current rate and any future change must come through MOF and STA. Treat the direction as firm and the number as unknown.

4. Standards are pulled into the project design. Standards layout (标准布局) is to be embedded in major national S&T projects, with standards research tasks given assessment targets, and frontier-field national and international standards and pre-standard research strengthened. For materials, where qualification and certification — not invention — is usually the bottleneck, this is a more useful clause than it first appears.

Where metals and materials actually sit

Steel is absent by name, but three channels carry it.

Two further channels are demand-side rather than materials-side. The continuation of "major manufacturing technical transformation and large-scale equipment renewal" supports machinery demand — and equipment replacement inside steel plants themselves. And the full rollout of the "AI+" initiative covers AI process control, yield and quality prediction and predictive maintenance in steelmaking, which is where much of the near-term productivity gain in Chinese steel actually sits.

The more interesting signal is the ranking. Advanced materials and key basic materials appear in the breakthrough tier alongside integrated circuits and machine tools, not in the emerging-industry tier. New materials also appear in the strategic-emerging-industry list (with new energy and aerospace), where clusters and commercialisation support apply. The future-industry list — quantum, biomanufacturing, hydrogen and fusion, brain-computer interfaces, embodied intelligence, 6G — deliberately excludes conventional materials. That is a hierarchy worth noting: conventional metallurgy is framed as a traditional industry to be upgraded by advanced technology, not as a frontier to be subsidised.

The part most readers will skip, and shouldn't

Task (九) tells large enterprises to open R&D platforms, application scenarios and supply chains to SMEs and to be the "first user" (先试首用) of SME innovations; Task (三) adds "use first, pay later" (先使用后付费) licensing of university and institute results to SMEs. For specialty steel, alloys and advanced ceramics, this is a direct attack on the qualification bottleneck: the difficulty is rarely the metallurgy, it is getting a new supplier onto an approved-vendor list at a large OEM. Whether "first user" guidance becomes procurement practice at central SOEs is the metric to watch.

Task (十三) also names proof-of-concept centres and pilot/testing (中试) verification platforms, alongside national manufacturing innovation centres. Pilot-scale verification capacity is precisely what the Chinese advanced-materials chain has lacked relative to its laboratory output — a point made repeatedly in domestic industry discussion, though this document itself sets no capacity target.

Task (十五)'s trade-secret and patent-open-licensing provisions, combined with the standards push, are the quiet enablers: a materials innovation that cannot be specified in a standard cannot be procured at scale.

Analytical judgement

The opinion's importance is not in any single clause but in the direction of travel across three axes: who decides (enterprises and investors gain panel seats and project leadership), how it is judged (market acceptance, not paper output), and how it is financed (R&D reserves shielded from SOE profit metrics; state venture capital told to invest early, small, long-term and in hard tech).

For metals professionals outside China, two implications are worth carrying:

• The competitive threat is upstream of steel, not in steel. The named targets — machine tools, high-end instruments, advanced materials, basic components — are the segments where Chinese producers have historically been weakest and where import substitution has been slowest. Policy attention concentrated there, with enterprise-led project design and standards leverage, is a more credible route to capability than another round of capacity subsidies. This is analysis, not a forecast the document makes.
• Expect implementing documents, not the opinion, to set the numbers. The super-deduction ratio, the deemed-profit add-back ratio, the SOE R&D reserve rules and the innovation-points (创新积分制) evaluation methodology will each arrive through ministry and tax documents. Those are where the investable detail lives.

The honest caveat: this reading rests on a single official reprint of the Xinhua release. No implementing rules, no budget allocations and no provincial responses were available at the time of writing, and the document's own Section 八 leaves implementation to departments and localities "in light of actual conditions" — which in practice means the gap between the text and the outcome will be wide and uneven.

Sources

• 中共中央办公厅 国务院办公厅关于推动科技创新和产业创新深度融合的意见, Xinhua release republished by MIIT, 21 October (source of all quotations above).