Chapter 10: Conclusion: Green Governance Capability as an Endogenous Driver of Value Chain Climbing

In November 2023, a furniture factory in Houjie, Dongguan, shipped 1,400 carbon-labelled office chairs to Rotterdam. The label carried a verified figure of 28.4 kilograms of CO2 equivalent per unit, audited against the GHG Protocol by a third-party certifier. The buyer had requested the figure before placing the order. Five years earlier, the same factory had never calculated a product carbon footprint. Its managers knew the price of plywood, the wage rate of upholsterers, and the freight cost to the port. They did not know how much carbon their kiln-dried timber or their polyurethane foam represented.

The distance between those two states—a factory that could not state its emissions and one that could certify them—is what this book has tried to map. The easy reading is that a buyer imposed a requirement and a supplier adjusted. That reading is incomplete. Between the first buyer request and the certified shipment, someone had to learn the emission factors for Chinese grid electricity. Someone had to build a spreadsheet mapping each component to an activity data point. Someone had to negotiate with the plywood supplier for primary data and, when that failed, decide which secondary dataset was defensible. Someone had to explain to the finance director why the audit fee was worth paying. These are capability-building activities, and they changed what the firm could do.

The argument running through the preceding nine chapters is that this kind of capability—the ability to measure, govern, and improve environmental performance in a way that value chain partners can verify—has become an endogenous driver of industrial upgrading. Green governance capability functions as a productive asset. It shapes which firms can climb the value chain and which cannot. The chapters examined that claim across labor-intensive, technology-intensive, and capital-intensive industries, through the lenses of value chain governance, cluster dynamics, and the specific pressures of China's dual carbon goals. This chapter draws the threads together, states what the evidence supports, and identifies what remains unresolved.

10.1 Theoretical Contributions and Synthesis

10.1.1 The compliance-capability distinction revisited

In the policy literature, environmental regulation is typically modeled as a constraint: a firm faces a cost curve, regulation shifts the curve upward, and the firm optimizes within the new constraint. That model explains short-run behavior well. It does not explain why some firms emerge from a regulatory shock with new products and new organizational routines, while others pay the fine or exit.

The book's first theoretical contribution is to specify what separates the two outcomes. Compliance is the ability to meet a defined requirement at a defined moment. Capability is the ability to generate, interpret, and act on environmental information on an ongoing basis. A firm can comply with a single buyer audit by hiring a consultant for two weeks. It cannot build a carbon accounting system that way, because the system has to survive the consultant's departure, absorb new data, and respond to new questions. The distinction matters because the two are often conflated in policy evaluation. A program that produces compliance may not produce capability, and a program that produces capability may look unimpressive in its first year.

The empirical chapters traced this distinction through several industries. Among the Houjie furniture firms in Chapter 4, the ones that converted buyer pressure into durable capability were those that built internal data routines and assigned responsibility for them. Firms that outsourced the entire exercise to consultants produced a number but could not defend it when the buyer asked a follow-up question. In the technology-intensive sectors examined in Chapter 5, firms that treated eco-design as a learning process—iterating on materials, testing substitutes, documenting failures—accumulated knowledge that fed into subsequent products. In the capital-intensive sectors of Chapter 6, firms that integrated emissions data into process control, rather than treating it as a reporting add-on, were better positioned to make the large capital decisions that determine long-run emissions trajectories.

Compliance is often the entry point. The transition from compliance to capability is a distinct organizational achievement, and it is the achievement that matters for upgrading.

10.1.2 Green governance as a value chain function

The global value chain literature has long distinguished between governance by parameters—what is to be produced, how, at what price—and governance by process—how the relationship itself is managed. Environmental requirements cut across both. A buyer that specifies a carbon label is setting a product parameter. A buyer that requires suppliers to report emissions annually is setting a process parameter. A buyer that co-invests in a supplier's energy efficiency is doing something else: building capability in the supplier, and in doing so altering the distribution of competence along the chain.

This last form of governance is becoming more consequential. Buyers invest in supplier measurement capacity because they cannot otherwise verify what they are buying. When environmental performance is hard to observe, the buyer's own credibility depends on the supplier's data quality. The result is that governance capability diffuses down the chain. Suppliers that absorb it become more autonomous, because they can now participate in conversations about product design and process improvement that were previously closed to them.

A counterargument deserves attention. Environmental governance could be read as a new layer of buyer power, a way for lead firms to shift costs and risks onto suppliers while capturing the reputational benefits. The apparel sector has seen sustainability codes that impose reporting burdens on suppliers without corresponding price premiums. But the variation across sectors is instructive. Where environmental performance is genuinely difficult to verify, buyers have reason to build supplier capacity. Where it is easy to verify—a simple declaration, a certificate of origin—buyers can free-ride on supplier effort. The governance outcome depends on the verification problem, not on the moral orientation of the buyer.

That insight has a corollary for suppliers. The strategic value of green governance capability is highest when verification is hard, because that is when the capability is scarce. A supplier that can produce credible, granular emissions data is more valuable to a buyer than one that can only produce a certificate. The Dongguan chair manufacturer discovered this: the buyer that requested the carbon label subsequently asked the factory to join a joint process-improvement project, because the factory's data was good enough to support it. The capability had become a platform for a different kind of relationship.

10.1.3 The factor-intensity pathways and their limits

The book organized its empirical analysis around factor intensity, distinguishing labor-intensive, technology-intensive, and capital-intensive pathways. The rationale was that the economics of green upgrading differ systematically across these categories. Labor-intensive firms face green requirements as a cost pressure, because their competitive advantage rests on low input costs and their margins are thin. Technology-intensive firms face them as a design problem, because their competitive advantage rests on product performance and their customers will pay for environmental attributes. Capital-intensive firms face them as an asset problem, because their emissions are embedded in long-lived equipment and their transition requires large, lumpy investments.

That framework held up across the cases, but the cases also revealed its limits. Factor intensity does not determine the pathway; it shapes the constraints within which firms choose. Two furniture firms with similar cost structures can end up on very different trajectories depending on whether they invest in measurement capability. Two technology firms with similar R&D budgets can diverge depending on whether they treat eco-design as a compliance exercise or a learning process. The framework is a map of the terrain, not a script for the journey.

The categories are also not stable. A labor-intensive firm that automates becomes more capital-intensive. A technology firm that moves into services becomes less dependent on product manufacturing. The green transition itself can shift a firm's factor intensity: a furniture manufacturer that invests in a biomass boiler and a digital energy management system is, in effect, becoming more capital-intensive in order to reduce its exposure to energy price volatility. The pathways are positions that firms occupy and sometimes leave.

10.1.4 Clusters as governance arenas

The cluster literature has long emphasized local institutions, shared knowledge, and informal coordination. These features take on new significance in the green transition, because many environmental problems are collective problems that no single firm can solve alone.

A cluster's energy infrastructure is a shared resource. Its waste streams are a shared problem. Its reputation is a shared asset. When a cluster's firms face a common environmental requirement, they have reason to coordinate: to negotiate with local government over grid access, to share the cost of a testing laboratory, to establish a common standard for supplier data. Chapter 7 documented cases where this coordination happened and cases where it did not, and the difference was often institutional rather than technical. Clusters with active trade associations, dense supplier networks, and a history of collective action were better able to convert a shared problem into a shared capability.

This finding has a theoretical implication that goes beyond the cluster literature. Green governance capability is partly a collective attribute, embedded in the relationships and institutions that surround firms. A firm in a well-organized cluster can access capabilities that a firm in an atomized region cannot, even if the two firms are otherwise similar. That matters for policy, because interventions at the cluster level can have effects that firm-level interventions cannot replicate.

10.1.5 The dual carbon goals as a contextual force

China's dual carbon goals—the commitment to peak carbon emissions before 2030 and achieve carbon neutrality before 2060—function as a contextual force that shapes the incentive landscape for manufacturing firms. The goals are not a single policy. They are a bundle of targets, regulations, price signals, and administrative priorities that interact in complex ways. Chapter 8 examined this bundle and argued that its most important effect is not the carbon price itself, which remains low by international standards, but the credibility of the long-run signal. Firms that believe the transition is real make investments that firms facing an uncertain signal defer.

That credibility effect is uneven. It is stronger in sectors that are directly regulated, weaker in sectors exposed to international competition that cannot pass costs through. It is stronger for large firms with the analytical capacity to interpret policy, weaker for small firms that experience policy as a series of disconnected requirements. The dual carbon goals thus reinforce the capability gap even as they create incentives for capability building. The sectors where the gap widens most are those with many small firms, thin margins, and no history of environmental measurement: furniture, textiles, and parts of food processing. In these sectors, the signal reaches the firm as a compliance demand from a buyer or a local regulator, not as an investment horizon.

10.1.6 What the synthesis shows

The book's central claim can now be restated with more precision. Green governance capability is an endogenous driver of value chain climbing because it changes the set of activities a firm can perform, the terms on which it can negotiate with buyers, and the rate at which it can learn. It is not a sufficient condition for upgrading—firm strategy, macroeconomic conditions, and luck all matter—but it is a mechanism that has been under-theorized in the value chain literature and under-supported in policy.

The word "endogenous" is doing real work here. It signals that green governance capability is an internal asset that firms build, not an external shock that firms respond to. The distinction changes the policy question. If capability were exogenous—a function of regulation—the policy problem would be to set the right regulation. If capability is endogenous, the policy problem is to create the conditions under which firms can build it. Those conditions include prices and rules. They also include information, training, and collective institutions.

10.2 Policy and Managerial Implications

10.2.1 From carbon pricing to capability support

The most common policy prescription for green upgrading is a higher carbon price. If emissions are costly, firms will invest in reducing them. The book does not dispute that logic, but carbon pricing alone is a blunt instrument for capability building. A price signal tells a firm that emissions matter. It does not tell the firm how to measure its emissions, which technologies are available, or how to reorganize its production process. Those are learning problems, and learning problems require different instruments.

A portfolio approach makes more sense. Carbon pricing should be part of the portfolio, because it creates a persistent incentive and raises revenue that can be recycled into support programs. It should be complemented by instruments that reduce the cost of learning: technical assistance for carbon accounting, shared testing facilities, industry benchmarks, and training for the engineers and managers who will implement the changes. The Dongguan factory did not need a higher carbon price to start measuring its emissions. It needed a buyer who asked for the measurement and a consultant who could show it how. The price signal became relevant later, when the factory began to see energy costs as a manageable variable rather than a fixed overhead.

Capability support should be targeted at the transition from measurement to action. Many firms can produce a carbon footprint with external help. Fewer can use that footprint to identify cost-effective abatement options and implement them. Programs that fund audits but not implementation, or that provide data templates but not analytical support, leave firms with a document they cannot use. The gap between measurement and action is where policy can add the most value.

10.2.2 The MRV infrastructure gap

Measurement, reporting, and verification systems are the backbone of green governance. Without them, claims about environmental performance cannot be trusted, and markets for green products cannot function efficiently. Chapter 3 examined the state of MRV in Chinese manufacturing and found a patchwork: some sectors have sophisticated systems, others have almost nothing. The patchwork is not random. It reflects the history of regulation, the demands of export markets, and the capacity of industry associations.

MRV infrastructure is a public good that markets underprovide. Individual firms have little incentive to invest in the standards, databases, and verification capacity that make their own claims credible. Industry associations and government agencies have a role in building that infrastructure. Chapter 7 showed that where such infrastructure exists—a shared laboratory, a common emissions factor database, a regional verification body—firms build capability faster and at lower cost.

China's green credit and green bond markets illustrate the scale of the opportunity. By the end of 2020, the green credit balance had reached RMB 11.95 trillion, about 6.9 percent of total credit, and green bond issuance from 2015 to 2020 totaled RMB 1.2 trillion. These are large numbers, but their environmental impact depends on the quality of the projects they finance and the credibility of the environmental claims attached to them. Better MRV would improve the allocation of green finance by making it harder to label a project green without evidence. That is a financial reform as much as an environmental one.

10.2.3 Managerial implications: building the green governance function

For managers, the implications are concrete. Green governance should be treated as a function, not a project. A project has a beginning and an end; a function has staff, routines, and a budget. Firms that assigned a single manager to "handle the carbon issue" alongside other responsibilities were less likely to build durable capability than firms that created a dedicated role or team. The team does not need to be large. It needs the authority to request data from operations, the analytical skill to interpret it, and the standing to bring it into strategic decisions.

Data quality is a strategic asset. Firms that invested in primary data—metering their own energy use, tracking materials through the production process, working with suppliers to obtain actual rather than estimated figures—were better able to respond to buyer requests and to identify abatement opportunities. Firms that relied on secondary data and industry averages could produce a number, but the number was not actionable. The difference showed up in negotiations: a firm with primary data could defend its figures and argue for a price premium. A firm with estimates could not.

Green capability is most valuable when integrated with product development. The technology-intensive cases in Chapter 5 showed that firms bringing environmental data into the design process—choosing materials for their carbon profile, designing for disassembly, testing substitutes—generated products that were both greener and more competitive. Firms that treated environmental performance as a reporting exercise missed that opportunity. The integration requires organizational change: design engineers need access to emissions data, and sustainability staff need a voice in product decisions. That is a managerial choice, not a technical one.

10.2.4 The relocation question

Relocation deserves separate attention because it is politically charged and analytically difficult. The concern is that green requirements will push carbon-intensive production out of China, or out of regulated jurisdictions more generally, to places with weaker standards. The concern is real, but the evidence does not support a simple race to the bottom.

The furniture case is instructive. Some production did relocate, but the relocation was driven by a combination of factors—rising labor costs, land prices, trade tensions—of which environmental regulation was only one. Firms that relocated to Southeast Asia often found that they had to meet the same buyer requirements in their new locations, because the buyers' standards traveled with the supply relationship. Relocation did not escape green governance; it changed the jurisdiction in which green governance had to be implemented.

If green requirements apply only to domestic production, they create an incentive to relocate. If they apply to products regardless of origin—through border adjustments, for example—they preserve the incentive to upgrade. The design of such mechanisms is complex and contentious, but the principle is sound: environmental governance should attach to products and supply relationships, not to geographic jurisdictions.

For managers, the relocation decision should be evaluated against the firm's capability trajectory. A firm that relocates to escape green requirements may find that it has traded a capability-building challenge for a market-access problem. A firm that relocates to access new markets or lower costs while maintaining its green governance capability may be making a sound strategic move. The distinction is not always clear in advance, but it is worth making explicit.

10.2.5 The role of clusters and collective institutions

Policy should pay more attention to the collective dimension of green governance. Individual firm support is necessary but not sufficient. Clusters need shared infrastructure—testing laboratories, data platforms, training centers—and institutions that can coordinate collective action. Trade associations, local governments, and industry alliances all have roles to play.

The most effective cluster interventions reduce the cost of collective learning. A shared emissions factor database saves every firm in the cluster the effort of deriving its own factors. A joint training program spreads the cost of expertise. A common supplier standard reduces the transaction costs of green procurement. These are modest interventions compared to large infrastructure projects, but they address the specific bottleneck that many firms face: the cost of acquiring and interpreting environmental information.

Cluster interventions risk capture by the largest firms, who have the most resources and the most influence over associations. Chapter 7 included examples of both inclusive and exclusive cluster governance. The inclusive cases tended to have transparent decision-making and explicit provisions for smaller firms. The exclusive cases served the interests of the largest members. Policy design should favor the former.

10.2.6 International dimensions

The book has focused on China, but its argument has international implications. The green governance capability of Chinese manufacturers affects their own upgrading trajectories and reshapes the structure of global value chains. As Chinese firms build capability, they become more attractive as suppliers for lead firms that need to demonstrate environmental performance to their own customers and regulators. That can strengthen China's position in existing chains and open opportunities in new ones.

It can also create tensions. If Chinese firms build capability faster than firms in other producing countries, the distribution of gains from green upgrading will be uneven. Capability building is not automatic; it depends on the conditions that policy and institutions create. Countries that invest in MRV infrastructure, technical assistance, and cluster coordination will see their firms climb faster than countries that do not. That is a development policy lesson as much as an environmental one.

10.3 Future Research Directions

10.3.1 Measuring capability, not just performance

The most important gap in the research is the measurement of green governance capability itself. The book has used proxies—the presence of a carbon accounting system, the quality of data, the integration of environmental information into product decisions—but these are imperfect. A firm can have a system that produces numbers nobody uses. A firm can have high-quality data on a narrow set of activities and nothing on the rest. A firm can integrate environmental information into one product line and ignore it elsewhere.

A more rigorous approach would study the routines firms use to generate and act on environmental information: who collects the data, how often, what happens to it, who sees it, what decisions it informs. These are observable features that can be compared across firms and over time. Studying failures would be equally valuable: firms that invested in measurement but did not change their behavior, or firms that changed their behavior but could not sustain it. Failures are often more informative than successes.

10.3.2 The microfoundations of green learning

The book has argued that green governance capability is built through learning, but it has not specified the mechanisms of learning in detail. How do firms acquire the knowledge they need? What roles do engineers, managers, and external consultants play? How does knowledge move from one part of the firm to another? These are questions for research at the micro level, using methods from organizational studies and the sociology of knowledge.

One specific question concerns professional communities. Carbon accounting is a nascent profession in China, with a small number of trained practitioners and a growing demand for their services. How do these practitioners develop their expertise? What standards do they use, and how do those standards evolve? How do firms decide whether to hire them or build internal capacity? The diffusion of capability depends on the diffusion of the people who carry it.

10.3.3 The political economy of green governance

The book has treated policy as a contextual force, but it has not analyzed the political processes that produce policy. That is a significant gap. Green governance requirements are not handed down from above; they are the outcome of negotiations among firms, industry associations, government agencies, and international actors. The shape of those negotiations determines who bears the cost and who captures the benefit.

Future research could examine how green governance requirements are actually made. Who participates in the standard-setting process? How do firms with different capabilities influence the outcome? How do international requirements interact with domestic ones? These questions are particularly important in China, where the policy process involves a complex mix of central directives, local experimentation, and industry consultation. The cluster cases in Chapter 7 offered glimpses of this process, but a systematic study would be valuable.

10.3.4 Comparative research across countries

The book's framework was developed for China, but its logic should apply more broadly. Firms in other producing countries face similar pressures and similar capability gaps. Comparative research could test the framework's generality and identify the conditions under which it holds. Are the pathways similar in Vietnam, Bangladesh, or Mexico? Do clusters play the same role? Does the dual carbon signal have an analogue in other policy contexts?

Comparative research could also examine the international diffusion of green governance standards. How do buyer requirements developed in Europe or North America travel to producing countries? What happens when they arrive? Do they build capability or simply impose compliance? The book's argument suggests that the answer depends on the verification problem and the governance relationship, but that hypothesis needs testing.

10.3.5 The limits of green upgrading

The book has emphasized the positive relationship between green governance capability and value chain climbing. It has acknowledged that the relationship is not universal, but it has not systematically examined its limits. When does green capability fail to translate into upgrading? When does it become a cost without a corresponding benefit?

One limit is market structure. In sectors where buyers have strong power and suppliers are numerous, green capability may be necessary to maintain market access but insufficient to command a price premium. The capability becomes a condition of participation rather than a source of advantage. Another limit is technological. Some green improvements are incremental and easily imitated; they do not create durable advantage. Radical improvements that require new technologies may be beyond the reach of most firms. The technology-intensive cases in Chapter 5 suggested that the most valuable capabilities combine environmental performance with product performance in ways that are hard to copy, but the conditions under which that combination is possible deserve more study.

10.3.6 The role of finance

The book has touched on green finance but has not analyzed it in depth. Finance is a critical enabler of green upgrading. Firms need capital to invest in new equipment, new processes, and new capabilities. The cost and availability of that capital depend on how financial institutions assess environmental risk and opportunity.

China's green credit and green bond markets are large and growing, but their impact on manufacturing upgrading is not well understood. Do green loans actually finance green investments, or are they relabeled conventional loans? Do they reach small and medium enterprises, or do they concentrate on large firms with the capacity to navigate the application process? How do financial institutions verify environmental claims? These questions are empirical and answerable, and they matter for policy.

10.3.7 The temporal dimension

The book has treated green governance capability as a stock that accumulates over time, but it has not studied the dynamics of accumulation in detail. How long does it take to build capability? What are the critical thresholds? What happens when capability is lost—through staff turnover, organizational change, or crisis?

The temporal dimension matters because policy operates on different time scales. A carbon price changes incentives immediately; a training program takes months; a cluster institution takes years. Understanding the time lags and feedback loops in capability building would help policymakers sequence their interventions. It would also help managers anticipate the costs and benefits of investment.

These directions are not a unified program. They are questions that the book's argument raises but does not answer. The value of a theoretical framework is not that it resolves all questions but that it identifies which questions are worth asking. The framework developed here—green governance capability as an endogenous driver of upgrading, shaped by factor intensity, cluster institutions, and policy context—has generated empirical findings and open questions. The findings are provisional. The questions are live.

10.4 Coda: The Chair in Rotterdam

Return to the chair. It sits in a showroom in Rotterdam. Its carbon label has served its purpose: it told the buyer what the buyer wanted to know. The factory that made it has moved on to other orders, other buyers, other requirements.

What happened to the capability that produced the label is an empirical question, and the answer is not yet in. The factory's carbon accounting team, which numbered three people in 2023, was still intact when the author visited in early 2024. It had added a second product line—a range of carbon-labelled conference tables—and was preparing a bid for a European contract that required product-level environmental declarations. Whether that team survives the next downturn, whether its routines outlast the managers who built them, whether the capability extends to the factory's other product lines: these are the questions that will determine whether the Dongguan case becomes a template or an anecdote.

The book's argument is that the difference between those outcomes is not random. It depends on choices made by managers, on institutions that support or fail to support learning, on policies that create incentives or ignore capability. The chair is a small object, but it carries a large question: whether the green transition will be a source of competitive advantage for firms that learn to govern it, or a cost that firms pass on or avoid. The evidence assembled in this book suggests that both outcomes are possible, and that the difference is made by the capabilities that firms, clusters, and policies build.

The next step is to track the Houjie factory's carbon accounting team over the next three years—through the 2025 contract cycle and whatever downturn follows it—and to compare its trajectory with the two neighboring furniture firms that outsourced their carbon measurement entirely. That comparison, not a prediction, is what the argument now requires.