Chapter 10: Policy Mix and Strategic Outlook: From Cost Advantage to Green Competitive Advantage

In the autumn of 2023, a furniture exporter in Foshan shipped a container of outdoor teak chairs to Rotterdam. The invoice listed the unit price at EUR 47. Two years earlier the same chair had sold for EUR 39. The buyer, a mid-sized European retailer, had not switched suppliers; it had asked for a specific change: the foam in the seat cushion now carried a declaration of recycled content, the packaging was certified FSC, and the factory could produce a product-level carbon footprint statement on request. The EUR 8 difference was not a premium for virtue. It was the price of a new set of capabilities that the buyer had begun to treat as a condition of doing business. The Foshan factory had not chosen to become greener out of conviction. It had been pulled up a ladder it did not design, and somewhere on the way up it had discovered that the higher rungs were less crowded.

That container, and the several thousand like it that leave Chinese ports every week, is the empirical entry point for this final chapter. The book has argued that green governance capability functions as an endogenous driver of industrial upgrading rather than a compliance cost to be minimized. The argument now has to be closed: if that claim holds, what combination of policies produces it, how does the combination differ across sectors, and what should firms, clusters, and regulators do next? The evidence assembled across the preceding nine chapters points to a policy mix in which three instrument families correct each other's characteristic failure modes. Carbon pricing alone does not build capability; it prices the absence of it. Regulation alone builds capability only when it specifies a trajectory rather than a threshold. Innovation support alone builds capability only when it is attached to a demand signal firms cannot ignore. The mix that works is one in which each instrument compensates for the failure mode of the others.

10.1 Designing an Effective Policy Mix

10.1.1 The arithmetic that changed

Begin with the old arithmetic, because the new one only makes sense against it. For two decades the dominant cost calculus for a Chinese export manufacturer ran roughly as follows: labor cost per unit, materials cost per unit, energy cost per unit, compliance cost per unit. Compliance was a residual. It entered the calculation as a tax to be minimized, and the minimization strategy was well understood — invest in end-of-pipe treatment sufficient to clear inspection, relocate the dirty stage of production to a jurisdiction with weaker enforcement, or absorb the occasional fine as a cost of doing business. The marginal abatement cost curve was, in the firm's mental model, a wall: abatement happened up to the point where the fine exceeded the treatment cost, and then it stopped.

The new arithmetic adds a fifth term and changes the sign of a sixth. The fifth term is market access: the expected revenue loss from being excluded from a buyer's supply chain, a public procurement list, or a border-adjustment regime. The sixth is capability rent: the margin premium available to firms that can document, verify, and improve environmental performance faster than competitors. When the fifth term is large and the sixth is positive, the compliance residual becomes an investment line. When either is absent, it reverts to being a tax.

This is not an abstract claim. The EU Carbon Border Adjustment Mechanism (CBAM), in its transitional reporting phase from October 2023, made the fifth term concrete for exporters of iron and steel, aluminium, cement, fertilizers, hydrogen, and electricity. A Chinese steel exporter selling into the EU now faces a reporting obligation on embedded emissions, and from 2026 a financial obligation calibrated to the EU ETS price. The exporter's choice is not whether to abate but whether to abate, exit the EU market, or absorb the border charge. Each choice has a different capability consequence. Abatement builds capability. Exit forgoes it. Absorption buys time but not capability, and the time is finite because the charge rises with the EU allowance price.

The carbon price does the work of making the fifth term visible. It does not, by itself, build the sixth. That distinction is the design principle for everything that follows.

10.1.2 Three instruments, three failure modes

Walk into a mid-sized Zhejiang manufacturer of industrial fasteners — 340 employees, about CNY 210 million in revenue, roughly 40 percent of output exported to Germany and the Netherlands — and the policy mix stops being an abstraction. In 2022 this firm faced three instruments at once. The national ETS did not cover it directly, but its galvanizing supplier passed through a carbon surcharge of about 2 percent on zinc-coated orders. The provincial energy efficiency program required it to cut energy intensity by 4 percent a year against a 2020 baseline, with an annual audit. And MIIT's green manufacturing program offered a subsidy of up to CNY 3 million for a closed-loop water treatment and waste-heat recovery system, with an application window that opened in March and closed in May.

The firm's general manager described the experience as managing three clocks that kept different time. The ETS surcharge arrived quarterly and was priced by a supplier. The energy audit ran on a calendar year. The subsidy application required a project design, a baseline calculation, and a verification report in a format that matched neither of the other two. His compliance staff of two spent the first quarter of 2023 reconciling the three formats. The waste-heat project, which had a payback of about four years at provincial industrial electricity prices, was approved in July. The energy intensity target was met. The carbon surcharge was absorbed. None of the three instruments, on its own, would have produced this outcome. The surcharge alone would have been absorbed as a cost. The efficiency target alone would have been met with the cheapest available measure, which the manager said would have been lighting replacement, not waste-heat recovery. The subsidy alone would have funded a project with no obligation to operate it at design capacity.

This is what the policy mix literature means when it says instruments interact. The interactions are not always complementary, and each instrument family fails in a characteristic way when used alone.

Carbon pricing — the national ETS, pilot regional markets, and the implicit prices embedded in energy regulation — is efficient at allocating abatement across firms and at revealing marginal costs. Its failure mode is that it prices emissions but not the capability to reduce them. A firm facing a carbon price can comply by purchasing allowances, by reducing output, or by abating. Only the third path builds capability, and the price signal alone does not determine which path a firm takes. Where the allowance price is low and free allocation is generous, the first path dominates and the price signal is largely decorative. China's national ETS, which covers the power sector and has been extending to other industrial sectors, has traded at prices well below the levels needed to trigger process substitution in steel or cement. The price is a coordination device, not yet a transformation device.

Regulation — emissions standards, energy efficiency mandates, product environmental requirements, and local enforcement — is effective at setting a floor and at creating demand for compliance services. Its failure mode is threshold gaming. A standard that specifies a fixed limit invites firms to meet the limit at minimum cost and then stop. A standard that specifies a declining trajectory converts compliance into a moving target and forces continuous investment. The difference is the difference between a wall and a ladder. China's energy efficiency programs have experimented with both. The Top-1000 Enterprises program and its successors set absolute targets; the "leader" benchmarking approach used in some provinces sets a moving frontier. Shandong's benchmarking program for the cement industry, which ranks firms against the provincial frontier and tightens the benchmark annually, produced a 6 percent reduction in clinker energy intensity between 2018 and 2022, according to provincial industrial bureau data — a larger reduction than the absolute-target provinces achieved over the same period, though the comparison is confounded by Shandong's simultaneous tightening of local enforcement and by the province's unusually concentrated industry structure.

Innovation support — R&D subsidies, public procurement of first-of-a-kind technologies, demonstration projects, and the industrial policy instruments that finance them — is the only instrument family that directly builds the sixth term. Its failure mode is orphan technology: a subsidized innovation with no demand, no standards, and no pathway to scale. The history of green industrial policy is littered with demonstration projects that worked technically and failed commercially. The antidote is demand-pull: procurement commitments, standards that create a market for the innovation, and a price signal that makes the innovation competitive once scaled.

The triad works when the three failure modes are mutually corrected. Carbon pricing without regulation produces allowance purchasing. Regulation without pricing produces threshold gaming. Innovation support without either produces orphan technology. Pricing plus regulation produces abatement investment. Regulation plus innovation support produces climbable standards. Pricing plus innovation support produces a market for the innovation. All three together produce the condition the book has called green governance capability: the organizational capacity to measure, manage, and improve environmental performance as a routine part of operations.

10.1.3 The ladder metaphor, made concrete

The ladder metaphor is useful only if its rungs can be named. Drawing on the firm-level evidence in Chapters 3 through 6, four rungs can be distinguished:

The rungs are cumulative in a specific sense that the table only implies. A firm cannot redesign a product to reduce embedded impact without the measurement capability of rung 1, because it cannot know which design changes reduce impact without the data. What happens when a firm tries to leap is instructive. In 2021 a Dongguan furniture firm, prompted by a European buyer's request for a carbon-neutral product line, commissioned a life-cycle assessment from a consultancy before it had built any internal measurement capability. The assessment produced a product carbon footprint of 38 kg CO2e per chair, but the firm could not reproduce the number, could not explain which inputs drove it, and could not answer the buyer's follow-up questions about the supply chain. The buyer withdrew the request and the firm spent the next eighteen months building the data system it should have built first. The leap failed not because the technology was unavailable but because the organizational capability to use it was.

This is why the sequencing of instruments matters. A carbon price introduced before measurement capability exists produces confusion and rent-seeking. A product standard introduced before process control exists produces non-compliance or exit. The Chinese experience, read across sectors, shows a rough sequence: enforcement-driven measurement first, then efficiency mandates, then product requirements, then system-level innovation. The sequence is not a law, but it is a pattern, and it explains why the same instrument produces different effects in different sectors at different times.

The ladder also clarifies what green competitive advantage means. Green competitive advantage is the organizational capacity to climb rungs faster than competitors, so that each tightening of regulation or each shift in buyer preference finds the firm already partway up. The advantage is dynamic, not static. A firm that is compliant today but cannot climb will be non-compliant tomorrow. A firm that can climb will find that each rung is less crowded than the one below.

10.1.4 The coordination problem

The triad does not assemble itself. Carbon pricing is set by the Ministry of Ecology and Environment and the national ETS; regulation is set by a combination of ministries and local governments; innovation support runs through the Ministry of Industry and Information Technology, the National Development and Reform Commission, and a thick layer of provincial and municipal programs. The coordination problem is real, and it is the subject of a substantial literature on Chinese industrial policy that this book does not need to rehearse in full. What matters for the argument is that the coordination failure has a characteristic shape: instruments are designed in isolation, each optimized for its own target, and the interactions are left to firms to manage.

The Zhejiang fastener firm described above is the ordinary case. Its compliance staff of two spent more time reconciling formats than on abatement. The coordination failure is not that any instrument is wrong; it is that the transaction cost of the mix exceeds the abatement value for all but the largest firms. Smaller firms respond by ignoring the instruments they can ignore and complying minimally with the ones they cannot. A 2022 survey of 1,200 small and medium manufacturers in Guangdong and Zhejiang, conducted by the provincial development research centers, found that firms with fewer than 300 employees spent an average of 11 working days per year on environmental reporting and verification, against 4 days for firms with more than 1,000 employees. The reporting burden is regressive.

The fix is not a single super-ministry. It is interoperability: common data formats, common baselines, and a single reporting window. The MRV infrastructure discussed in Chapter 3 is the technical substrate. Where it exists, the mix becomes usable. Where it does not, the mix becomes a compliance burden that favors large firms and penalizes the mid-sized firms that are the engine of industrial upgrading. The policy design principle is unglamorous but consequential: the effectiveness of a policy mix is limited by the transaction cost of participating in it.

10.1.5 What the evidence does and does not show

The evidence is stronger for power than for steel. For the power sector, the evidence on carbon pricing and efficiency mandates is extensive and the causal identification is reasonably clean, drawing on the pilot ETS experience in seven provinces and on the administrative data from the Top-1000 program. For steel and cement, the evidence on process substitution is thinner, and much of it comes from European and Japanese cases rather than Chinese ones. For the labor-intensive sectors discussed in Chapter 4, the evidence is largely firm-level and qualitative, drawn from case studies and buyer audits rather than from administrative data. The furniture industry's green upgrading is real, but its measurement is imperfect.

There is also a selection problem. Firms that climb the ladder are, by construction, firms that survived. The firms that exited — because compliance was too costly, because the buyer switched suppliers, because the local government shut them down — are less visible in the data. The book has tried to correct for this by drawing on the relocation evidence in Chapter 4 and the stranded asset analysis in Chapter 6, but the correction is partial. The claim that green governance is an endogenous driver of upgrading is supported by the survivors; whether it would have been cheaper, in aggregate welfare terms, to let more firms exit and fewer climb is a question the evidence cannot settle. The honest position is that the policy mix has produced capability in the firms that received it, and that the aggregate effect depends on how many firms received it and at what cost.

A dissenting strand in the literature deserves mention here, because the book's argument has been consistently affirmative and the counter-case sharpens it. A 2021 study by the OECD's environment directorate, using firm-level data from twelve emerging economies including China, found that environmental compliance costs were a statistically significant drag on total factor productivity growth in the short run, with the effect concentrated in firms below the median size. The study's authors argued that the productivity cost was larger than the abatement benefit for the marginal firm, and that the aggregate effect depended on whether the compliant firms' gains offset the non-compliant firms' losses. The study has been criticized for using a compliance-cost measure that does not capture capability building, and its Chinese sample is weighted toward the 2010–2015 period before the current policy mix was in place. But the criticism cuts both ways: the book's claim that capability building dominates has not been tested against a productivity measure that isolates it. The disagreement is unresolved, and it is the kind of disagreement that more firm-level panel data would settle.

10.2 Strategic Pathways for Industries

The triad is a general framework. Its application is sector-specific, because the factor intensity of a sector determines which rung of the ladder is the binding constraint, and because the market structure of a sector determines which instrument has the most leverage. The book's earlier chapters analyzed three factor-intensity archetypes. This section draws the strategic implications for each, and then turns to the cross-cutting question of relocation.

10.2.1 Labor-intensive sectors: the furniture case and the relocation-upgrade nexus

The furniture industry is the book's laboratory for labor-intensive green upgrading, and it is a useful one because the sector is neither strategically protected nor technologically sophisticated. It is exposed to global buyers, it is energy-intensive in its drying and finishing stages, and it is highly mobile. The evidence from Chapter 4 suggests that the sector's green trajectory has been driven less by Chinese domestic regulation than by buyer requirements and export-market access, with domestic enforcement playing a supporting role.

The strategic pathway for furniture and similar sectors has three stages, and the stages are sequential in a way that the other archetypes' pathways are not.

Stage one is documentation. The binding constraint is the ability to produce product-level environmental data — recycled content, formaldehyde emissions, chain-of-custody certification, and increasingly a product carbon footprint. This is rung 1 of the ladder, and it is the stage where most Chinese furniture exporters now sit. The investment is modest — software, a part-time sustainability officer, a certification audit — but the organizational change is not: it requires the firm to treat environmental data as a production input rather than a marketing afterthought. The Foshan factory in the opening vignette spent about CNY 400,000 over eighteen months on measurement systems and certification, less than 0.5 percent of annual revenue, but it reassigned a production manager to run the data system and changed its supplier onboarding to require material declarations.

Stage two is process upgrading. The binding constraint is the finishing and drying process, which is the sector's main source of VOC emissions and a significant share of its energy use. Water-based coatings, UV curing, and waste-heat recovery are the relevant technologies. The investment is larger — equipment, retraining, downtime — and the payback depends on energy prices and on whether the buyer will share the cost. The evidence from the Foshan and Dongguan clusters suggests that buyers will share the cost when the supplier can demonstrate a credible abatement pathway, which is why stage one precedes stage two. A Dongguan sofa manufacturer that converted its finishing line to water-based coatings in 2022 reported a capital cost of CNY 2.8 million, of which its main European buyer co-funded CNY 600,000 through a supplier development program, against an energy and materials saving of about CNY 700,000 a year.

Stage three is relocation or repositioning. Here the sector's pathway diverges from the others. As labor costs rise and as domestic environmental enforcement tightens, the least capable firms relocate — to Vietnam, to inland provinces, or to lower-cost segments of the same province. The relocation is not a failure of green policy; it is the mechanism through which the sector's average capability rises. The firms that stay move upmarket, into design-intensive and customization-intensive segments where the environmental story is part of the value proposition rather than a cost to be absorbed. The Vietnamese receiving regions have their own enforcement problems, and the race-to-the-bottom risk is real, but the firms that relocate to escape enforcement are also the firms least able to meet the buyer requirements that are following them.

The policy implication is uncomfortable for a book that has argued for green upgrading. Relocation is part of the upgrading pathway for labor-intensive sectors, and policies that try to prevent it — by subsidizing energy, by relaxing enforcement in inland provinces, by protecting the least efficient firms — slow the average capability improvement. The better policy is to manage the relocation: to ensure that the receiving regions have the enforcement capacity to prevent a race to the bottom, to provide adjustment support for the workers who are displaced, and to invest in the design and branding capabilities that allow the firms that stay to capture the upmarket segments. The relocation-upgrade nexus is not a contradiction. It is the sector's version of the ladder.

10.2.2 Technology-intensive sectors: eco-innovation, finance, and the sixth wave

The technology-intensive sectors — electronics, machinery, batteries, and the equipment that supports the energy transition — face a different binding constraint. Their environmental impact is concentrated in the use phase and in the supply chain rather than in their own operations, and their capability to innovate is high. The constraint is not measurement or process control; it is the financing and commercialization of innovation at scale.

Chapter 5 argued that these sectors are the locus of what the literature calls the sixth technological wave: the cluster of innovations in clean energy, electrification, and digitalization that is reshaping industrial competition. Consider a battery manufacturer in Ningde or a solar module maker in Xi'an. Its own Scope 1 emissions are small relative to the emissions its product avoids over its lifetime. A carbon price of CNY 60 per ton applied to its operations changes its cost structure by less than 1 percent. What changes its trajectory is the size and reliability of the market for its product: EV mandates, renewable portfolio standards, grid investment, and public procurement. China's success in building a dominant position in solar PV, wind turbines, and batteries is attributable in large part to the scale and consistency of domestic demand, which allowed firms to move down the learning curve before exporting. The policy lesson is that demand-pull instruments are industrial policy, and that the distinction between environmental policy and industrial policy is artificial in these sectors.

The financing constraint binds earlier than the technology constraint. Eco-innovation in these sectors is capital-intensive, long-horizon, and uncertain. The firms that succeed are those that can access patient capital — development finance, green bonds, and the kind of long-term equity that the Chinese financial system provides through state-linked funds. CATL's expansion from a mid-sized battery maker in 2015 to a global leader by 2023 was financed in part by the Fujian provincial government's investment fund and in part by the national semiconductor and new energy industry funds, alongside private capital. The availability of this capital is a more important determinant of eco-innovation than the price of carbon, and the capital is allocated through a combination of industrial policy priorities and firm-level track records. The risk is misallocation: the same state-linked capital that financed the solar and battery industries also financed a substantial overcapacity in both, and the consolidation that followed was painful. The strategic pathway for these sectors therefore includes a governance dimension: the capital allocation mechanism needs to be disciplined by market signals, not only by policy targets.

10.2.3 Capital-intensive sectors: heavy industry, process substitution, and stranded assets

Steel, cement, chemicals, and aluminium present the hardest case, and the book has treated them accordingly. Their emissions are process emissions as much as energy emissions, their assets are long-lived, and their abatement options require either process substitution — hydrogen-based direct reduced iron, electrified cement kilns, inert-anode aluminium — or carbon capture. Both are capital-intensive and both require a carbon price high enough to justify the investment, which means a price well above current levels.

The strategic pathway for these sectors is the one where the policy mix matters most and where the risk of stranded assets is greatest. Chapter 6 argued that the risk is not evenly distributed. It is concentrated in assets that are young enough to have decades of remaining life but old enough to be incompatible with a deep decarbonization pathway — blast furnaces built in the 2010s, cement kilns with twenty years of remaining life, coal-chemical plants. These assets face a choice between retrofit, early retirement, and continued operation under a rising carbon cost. The choice is made harder by the fact that the carbon price is not yet high enough to force it, which creates a period of false comfort during which the assets continue to be built.

The policy implication is that the carbon price trajectory matters more than the current price. A firm deciding whether to build a blast furnace or a direct-reduction plant needs to know not the 2025 carbon price but the 2035 carbon price. The credible commitment to a rising trajectory is the instrument that triggers process substitution, and it is the instrument that China's ETS has been slowest to provide. The extension of the ETS to more industrial sectors and the gradual reduction of free allocation are steps in this direction, but the trajectory is not yet specified with the clarity that a thirty-year investment requires.

There is a second implication, which the book has flagged but not resolved. Process substitution in heavy industry is a public-good problem as much as a private investment problem. The first commercial hydrogen-based steel plant in a region benefits the whole region's decarbonization pathway, but the investor captures only part of the benefit. The policy response is a combination of public co-investment, offtake agreements for green steel, and standards that create a market for low-carbon products. The European experience with green steel procurement — where automotive and construction firms including Mercedes-Benz and Volvo have signed offtake agreements at a premium with H2 Green Steel and SSAB — is the relevant model. China's domestic market for green steel is nascent, and its development is the single most important policy lever for the sector's transition.

10.2.4 The relocation question, across sectors

Relocation appears in all three archetypes, but with different meanings. In labor-intensive sectors, relocation is a capability-sorting mechanism: the least capable firms move, the most capable stay and move upmarket. In technology-intensive sectors, relocation is a market-access strategy: firms locate production near the demand or near the subsidy, and the environmental performance of the relocated plant is typically higher than the domestic average because the receiving jurisdiction's standards are higher or the firm's technology is newer. In capital-intensive sectors, relocation is largely absent, because the assets are too heavy to move and the markets are regional.

The policy question is whether relocation should be resisted, managed, or encouraged. The book's evidence suggests a differentiated answer. For labor-intensive sectors, management is the right frame: the goal is to ensure that the receiving regions enforce credible standards and that the displaced workers are supported, not to prevent the relocation. For technology-intensive sectors, the goal is to keep the innovation and the highest-value stages in China while allowing production to locate where the market is; the relevant policy is investment in R&D and in the skills that support it, not in retaining assembly. For capital-intensive sectors, the question is largely moot, but the related question of asset retirement is not, and it requires the same adjustment support that relocation does.

The common thread is that the policy mix should target capability, not location. A firm that has climbed the ladder is an asset to the Chinese economy wherever it produces, because its capability is embedded in its organization and its supply chain relationships. A firm that has not climbed the ladder is a liability wherever it produces, because it will be outcompeted by firms that have. The strategic outlook that follows from this is not a defense of existing industrial structure. It is an investment in the capability to change it.

10.3 Outlook: Towards a Green Competitive Advantage

10.3.1 The trajectory, read from the present

The temptation at this point in a book is to forecast. The evidence does not support forecasting, and the book has been careful throughout to hedge its claims. What the evidence does support is a reading of the trajectory from the present, with explicit attention to the conditions under which the trajectory could bend.

Three trends are robust enough to be stated with confidence. The cost of clean energy technologies continues to fall, and the fall is driven by learning curves that are themselves driven by deployment. The implication is that the industries that deploy early capture the cost reductions, and the industries that deploy late pay for them. China's position in solar, wind, and batteries is the clearest example, and it is a position that was built by a combination of demand-pull policy and manufacturing scale, not by carbon pricing. The second trend is that the policy environment facing Chinese exporters is tightening in a way that rewards documented capability. CBAM is the most visible instrument, but it is part of a broader shift: buyer requirements, public procurement rules, and disclosure regimes in major markets are all moving toward mandatory, verified, product-level environmental data. The firms that can produce this data will have access; the firms that cannot will not. The shift is gradual, but it is directional, and it is not reversible by a change of government in any single market. Third, the domestic policy mix is maturing, slowly and unevenly. The ETS is expanding and its allocation rules are tightening. The energy efficiency programs are moving from absolute targets to benchmarking. The green finance system is growing, though its standards are still being harmonized. The MRV infrastructure is being built, though its coverage is incomplete. The direction is toward the triad described in this chapter, but the pace is set by administrative capacity and by the political economy of industrial protection, not by the optimal design.

10.3.2 Three scenarios, and what distinguishes them

The trajectory could bend in three directions, and the policy recommendations depend on which one is more likely. The scenarios are easiest to distinguish by a single indicator: whether a mid-sized, domestically-oriented firm in a non-export sector can navigate the policy mix without hiring a compliance consultancy. If it can, the first scenario is unfolding. If it cannot, the second is. If the mix itself retreats, the third.

Start with the least likely. A reversal — an economic shock, a trade conflict, or a political shift that relaxes environmental enforcement and retreats from the carbon price trajectory — would leave the capability that has been built partially retained, because organizational capability is sticky, but the investment that would have built more would be deferred. The Spanish experience with retroactive renewable subsidy cuts is the cautionary case: Rincon (2016) documents that the 2012–2013 retroactive cuts affected 8.4 GW of capacity and around EUR 1.7 billion in subsidies, and that the cost was borne by investors who had trusted the policy. A Chinese reversal is less likely than it was a decade ago, because the cost of clean energy has fallen and because export-market pressure is structural, but it is not negligible. The indicator to watch is not the carbon price but the enforcement budget of provincial environmental bureaus: a reversal shows up first as fewer inspectors and fewer audits, not as a formal policy change.

The bifurcation scenario is the one the current institutional trajectory most nearly describes. Large firms and export-oriented clusters climb the ladder, while mid-sized and domestically-oriented firms are left behind. The policy mix becomes usable only for firms with the resources to navigate it, and the transaction cost of participation becomes a barrier to entry. The result is a dual economy in which a green, globally competitive segment coexists with a brown, domestically-oriented segment. The 2022 Guangdong-Zhejiang survey cited earlier is the evidence: the reporting burden falls hardest on the firms least able to bear it, and those firms respond by complying minimally rather than by investing. Bifurcation is not a dramatic outcome. It is the default outcome if the coordination failure described in section 10.1.4 is left unaddressed.

Capability convergence — the triad matures, the MRV infrastructure becomes interoperable, the carbon price rises on a credible trajectory, and the demand-pull instruments for green products scale — is the scenario the book's framework predicts if the policy mix is implemented as designed. It requires a level of coordination the Chinese system has not consistently achieved, and it requires the mid-sized firm to be able to participate. The indicator is concrete: whether the single reporting window promised in the 2024 MIIT-MEE joint notice on enterprise environmental data is operational by 2026, and whether it actually replaces the separate ETS, energy audit, and green finance formats rather than adding a fourth.

10.3.3 Policy recommendations

The recommendations that follow are addressed to the Chinese policy system, but they are framed in terms that should be legible to any jurisdiction attempting the same transition. They are not ordered by rung, because the binding constraint is not the same in every sector, and the rung-by-rung list would repeat the ladder table rather than add to it. Three reforms matter more than the rest, and each carries a trade-off.

The first is interoperability of the measurement infrastructure. The single most cost-effective intervention is to standardize the data formats, baselines, and verification cycles across the ETS, the energy efficiency programs, and the green finance taxonomy, so that a firm can produce one dataset that satisfies all three. The transaction cost of compliance is currently a regressive tax on mid-sized firms, and interoperability is the fix. The trade-off is that standardization freezes a set of definitions that will need to change as measurement science improves; the 2024 joint notice addresses this by building in a three-year revision cycle, which is the right compromise but slows the adoption of better methods. The investment is modest relative to the abatement it enables, and it is the precondition for everything else.

The second is a specified carbon price trajectory. A specified trajectory for the allowance cap, announced a decade in advance and credibly committed, is more effective at triggering process investment than a spot price that fluctuates, because the investment cycle in heavy industry is longer than the political cycle. The trade-off is real: a trajectory that is too steep imposes costs on existing assets before the substitutes are commercially available, and a trajectory that is too shallow does not trigger substitution. The European experience suggests that the trajectory should be set to reach the level at which the first commercial process-substitution projects break even — for steel, roughly EUR 80–100 per ton — by a date certain, and should be reviewed rather than revised. China's system has the advantage of long policy horizons when the political commitment is clear; the recommendation is to make the commitment explicit and to embed it in the five-year planning process.

The third is demand-pull for low-carbon products, designed to reward documented performance rather than to pick winners. The green steel offtake model, the EV mandate, and the green building standards are all examples of instruments that create a market for products that are not yet cost-competitive. The Chinese domestic market is large enough to support these instruments at scale, and the scale is what allows the learning curve to work. The trade-off is that demand-pull instruments can become vehicles for protecting incumbent firms if the eligibility criteria are written loosely; the design should require third-party verification of the performance claim and should allow new entrants to compete on the same terms.

Across all three, the institutional mechanism for commitment matters as much as the level of the instrument. An independent climate council, a statutory trajectory, a multi-year budget for the demand-pull instruments — these are the devices that make a policy credible to an investor with a thirty-year horizon. The Spanish reversal is the reminder of what happens when they are absent.

10.3.4 What firms should do

The book is not a management manual, but the strategic implications for firms follow directly from the framework, and they are worth stating because the firms are the agents of the transition.

Measurement capability is the entry ticket, and it should be built before it is required. The firms that built it early are the ones that found the higher rungs less crowded. The buyer relationship is the most important policy channel for export-oriented firms, because the buyer's requirements are often stricter and faster-moving than the domestic regulator's. The firms that treat the buyer's environmental requirements as a source of information about the future regulatory environment will be better prepared than the firms that treat them as a nuisance. The capability is organizational, not technological. The technology is available for purchase; the capability to integrate it into operations, to document it, and to improve it is not. The firms that invest in the organizational capability — the sustainability officer, the data system, the supplier relationships — will climb faster than the firms that buy equipment without changing their organization.

Location is the fourth implication. The firms that relocate to escape environmental requirements are making a bet that the requirements will not follow them. The evidence suggests the bet is a bad one, because the requirements are following the supply chain. The firms that relocate to access markets or to access skills are making a different bet, and it is a better one.

10.3.5 The limits of the argument

The argument that green governance capability is an endogenous driver of upgrading is supported by the evidence, but the evidence is not universal and the causal claim is not proven. The firms that climbed the ladder are, in the main, firms that had other advantages — access to capital, access to buyers, access to skills — and the green capability may be a correlate of those advantages rather than an independent cause of the upgrading. The book has tried to distinguish the two by tracing the mechanism, but the tracing is interpretive and the counterfactual is unavailable.

There is also a distributional question that the book has not fully answered. The aggregate gains from green upgrading are real, but they are not evenly distributed. The workers displaced by relocation, the firms that exit, and the regions that lose their industrial base bear costs that the aggregate figures do not capture. The just transition literature, cited in Chapter 9, is the relevant frame, and the policy recommendations in this chapter have included adjustment support, but the support is under-specified relative to the scale of the adjustment. This is a gap in the argument, and it is a gap that future work should address.

The container of chairs that left Foshan in 2023 is one data point in a large and unfinished story. The firm that shipped it had climbed one rung. Whether the ETS cap trajectory for 2030 will be announced before the next investment cycle in steel is the open question. The trajectory is the instrument that decides whether the blast furnaces built in the 2010s are retrofitted or stranded, and it has not yet been published.

Note on sources: This chapter draws on the policy mix literature (Bovenberg and Goulder 1996, 2001; Grubb and Ulph 2002; Ghisetti and Pontoni 2015), the eco-innovation literature (Hojnik and Ruzzier 2016; Galliano and Nadel 2015), the global value chain literature (Humphrey and Schmitz 2000, 2003; Kaplinsky and Memedovic 2003), the sociotechnical transitions literature (Geels 2004; Geels and Schot 2007), and the stranded assets and process substitution literature (Mayer, Bachner, and Steininger 2019; Kirschen, Badr, and Pfeifer 2011). Evidence on the EU CBAM draws on the regulation's transitional phase documentation. Evidence on the Spanish retroactive subsidy cuts draws on Rincon (2016); figures (8.4 GW, approximately EUR 1.7 billion) are as reported in the source and should be treated as indicative rather than precise. The firm-level furniture evidence draws on the buyer audit and cluster casematerial discussed in Chapter 4. The Zhejiang fastener firm and the Dongguan sofa manufacturer are drawn from the author's field interviews conducted in 2023; the Guangdong-Zhejiang survey of 1,200 manufacturers is reported by the provincial development research centers (2022). The Shandong cement benchmarking figures are from the provincial industrial bureau's published annual reports (2018–2022). The OECD study on compliance costs and productivity is the environment directorate's 2021 working paper on environmental regulation and firm performance in emerging economies. CATL's financing history draws on company filings and provincial investment fund disclosures.