Chapter 1: Introduction: From Compliance to Capability in Green Governance
In November 2023, the owner of a furniture factory in Shunde, a district of Foshan in Guangdong, received a letter from his largest European buyer. The buyer, a German flat-pack retailer with distribution centers in Hamburg and Rotterdam, wrote that from the next fiscal year every cabinet and upholstered sofa shipped to its warehouses would need a product-level carbon footprint declaration, verified by a third party, with a documented reduction pathway to 2030. The factory had supplied this buyer for eleven years. Its competitive logic was simple: quote a lower price than the next Foshan factory, deliver on time, absorb whatever cost pressure came down the chain. The owner read the letter twice, then called his production manager and asked a question he had never asked before: "How much carbon is in one of our cabinets, and where does it come from?"
The production manager did not know. Neither did the finance director, the export sales team, or the owner himself. They knew the cost of every board, hinge, and container; they could quote a per-unit price to the yuan. They had no idea how much of that price was carbon.
That ignorance is the starting point of this book. The question the owner asked is operational, and it is also disruptive, because answering it requires the firm to reconstruct its entire production process as a flow of energy and materials rather than a flow of costs. The letter did not ask him to comply with a regulation. It redefined the terms of trade: it attached a verifiable carbon cost to every SKU shipped, and it made that cost a condition of continued access to the buyer's shelves. A consultant could produce a number in a spreadsheet within a month. The harder question is whether the firm can turn that number into knowledge, and knowledge into a capability: the ability to redesign products, renegotiate supplier relationships, reconfigure energy inputs, and reposition itself in the value chain so that carbon performance becomes a source of advantage rather than a tax on survival. The difference between a firm that complies and a firm that climbs is the subject of everything that follows.
1.1 The Compliance Trap
For most of the postwar period, environmental regulation was the province of the state. Governments set emission standards, imposed penalties, and firms responded by meeting the legal minimum. The relationship was vertical and adversarial: regulator against regulated. That model still exists, but it has been overlaid by a second system in which lead firms in global value chains act as private regulators, setting environmental requirements that exceed or precede public law and enforcing them through supplier selection. The labor standards literature documented this shift two decades ago. Distelhorst, Hainmueller, and Locke (2015) showed how lean production practices in Nike's supply chain interacted with labor compliance in ways neither the brand nor the state could have achieved alone; Distelhorst, Locke, Pal, and Samel (2015) traced how private regulation in global electronics produced compliance outcomes that varied enormously across sites within the same supplier base. The lesson transfers directly to environmental governance: private standards do not produce uniform compliance, and the firms that do best are not those facing the strictest standards but those that develop the internal capacity to interpret, adapt to, and eventually shape the standards they face.
The environmental version has arrived with more force than the labor version, for reasons that are worth spelling out. The scientific case for decarbonization is not seriously contested; the direction of travel is clear even where the pace is not. The policy instruments now being deployed—carbon pricing, border adjustment, product footprint requirements—carry direct and measurable cost implications that cannot be absorbed through public relations. And the data infrastructure for verifying environmental performance has become granular enough to move the unit of compliance from the facility to the product. When the unit of compliance is the SKU, the compliance burden stops being a fixed overhead spread across output. It becomes a variable cost attached to each product, each batch, each shipment—and therefore a new axis of competition. Two Foshan factories can quote the same price for the same cabinet and face very different carbon costs per unit. The one that can document a lower footprint wins the order.
The conventional response to this redefinition is what I call the compliance trap: the assumption that environmental performance is a cost of doing business, to be minimized and passed through, and that the firm's strategic task is to meet the standard at the lowest possible expense. The assumption is not irrational. It is rooted in a real history of environmental regulation as a cost-imposing exercise, and it is reinforced by policy designs that internalize externalities by making pollution expensive.
What makes the trap dangerous now is that it is static. It treats the standard as a fixed target rather than a moving frontier. A firm that complies with today's product footprint requirement and stops has bought itself one or two years; the buyer's 2030 reduction pathway, the EU's phase-out of free allowance allocation, and China's expanding national emissions trading system all point the same direction. Compliance investments made defensively, as deadweight costs, are also isolated investments—environmental performance gets parked in a separate department rather than folded into product design, process engineering, and supply chain management. The firm ends up with a certificate and no capability.
The empirical record shows the trap in operation. In the first two phases of the EU Emissions Trading System (2005–2012), when allowance prices were low and free allocation generous, abatement came mostly from reducing output rather than from process innovation; Martin, Muûls, and Wagner (2016) found that regulated firms cut emissions partly by cutting production, with little evidence of technology upgrading. When the carbon price is low, the cheapest compliance strategy is to pay it. When the price is high, the cheapest strategy is to relocate. Neither builds capability, because both treat carbon as a cost to be avoided rather than a parameter to be optimized.
The green supply chain literature shows the same pattern from the buyer side. Gimenez and Tachizawa (2012), reviewing the field systematically, found that most studies of supplier sustainability assessment focused on the design of assessment tools rather than on whether assessment led to capability building; the implicit model was monitoring and enforcement, in which the buyer sets the standard and the supplier meets it or loses the order. That model can produce compliance. It does not produce the deep process change that decarbonization requires. Handfield, Walton, Seeger, and Melnyk (1997), in an early study of green value chain practices in the US furniture industry, found that firms adopted green practices in response to customer pressure, but the practices were bolt-on rather than built-in and did not diffuse beyond the specific customer relationship that prompted them.
1.2 The Argument
The compliance trap is a choice, not a necessity. The alternative is to treat green governance as a capability—a set of organizational routines, skills, and relationships that let a firm interpret environmental requirements, redesign products and processes in response, and capture value from the resulting differentiation. My argument is that this capability is endogenous to the firm and its cluster: it is generated from within through learning and organizational change, not installed from outside by a regulator or a buyer, and once built it becomes a source of advantage that competitors cannot easily imitate because it lives in routines and relationships rather than in a piece of equipment or a certificate.
The word endogenous carries weight. It implies that the process of building the capability has effects beyond the environmental problem it was designed to solve. A furniture firm that learns to trace the carbon content of a cabinet has, in the process, built a material-flow accounting system that also cuts waste, a supplier database that also improves quality control, and an energy management function that also reduces cost. The capability spills over.
Three claims organize the book. Green governance capability is a dynamic capability in the sense Sid Winter and David Teece gave the term: it involves sensing changes in buyer and regulator requirements, seizing opportunities to redesign products and processes, and reconfiguring the firm's resource base to sustain the new configuration. It is not an extension of quality management or cost management, because it requires data, expertise, and collaboration that those functions do not generate. Second, the pathway to the capability is differentiated by factor intensity. Labor-intensive firms such as furniture and apparel face the sharpest buyer-driven pressure and the highest relocation risk, but they also have the most to gain from process innovation that cuts energy and material intensity. Technology-intensive firms such as electronics and machinery face standards embedded in product design and performance, and their pathway runs through innovation and the learning curve. Capital-intensive firms such as steel, cement, and chemicals carry the largest absolute emissions and the greatest stranded-asset risk, and their pathway runs through scale, process substitution, and the governance of long-lived capital investments. Third, the capability is built in clusters and governed across levels. A firm is embedded in a local industrial cluster that supplies shared infrastructure, specialized suppliers, and a labor market with relevant skills; it is also embedded in a multi-level governance system that runs from the district government to the national ministry to the international regime. Whether the capability can be built depends on how those layers interact.
The Shunde factory owner is an actor in a cluster that has the resources—suppliers, technicians, local government support—to mount a response. Whether he mounts one depends on whether he reads the letter as a threat to be managed or an opportunity to be seized. That perception is not a matter of personality. It is a matter of the capability architecture the firm and its cluster have already built, and the governance conditions that support or undermine it.
1.3 Why Now
The timing is not accidental. Three developments have converged since 2020, and each one raises the cost of staying in the compliance trap.
The first is China's dual carbon goals, announced in September 2020: peak emissions before 2030, neutrality before 2060. The goals matter beyond the domestic environmental agenda because they set in motion a policy cascade—an expanding national emissions trading system, a green finance system whose green credit balance reached RMB 11.95 trillion by the end of 2020 (6.9 percent of total credit), building standards such as GB/T 50378–2014 that evaluate energy, water, and material performance across a building's life cycle. These instruments are unevenly enforced and not fully integrated. They establish a direction of travel.
The second is the EU's Carbon Border Adjustment Mechanism, which entered its transitional phase in October 2023 and covers steel, cement, aluminum, fertilizers, electricity, and hydrogen. CBAM is designed to prevent carbon leakage, but its practical effect is to extend European carbon pricing into the supply chains of every country that exports to Europe. For a Chinese manufacturer, the mechanism appears as a line item on the invoice for covered products, and the line grows as free allowance allocation in the EU ETS phases out through 2034. Furniture is not yet a covered sector. The buyer's letter shows how the logic travels ahead of the regulation: retailers with their own 2030 scope-3 commitments impose footprint requirements on suppliers before any border mechanism reaches them.
The third is the decarbonization of electricity, which underpins everything else. The International Energy Agency estimates that electricity generation accounts for roughly 27 percent of global greenhouse gas emissions, the single largest emitting sector; in the United States and Europe the share is about 25 percent. If electricity can be made carbon-free, it can power almost everything else, which is why the transition slogan has become "not just low, but zero." For Chinese manufacturing the implication is concrete: the carbon content of a cabinet shipped from Foshan depends heavily on the carbon intensity of the grid that made the board, the coating, and the foam. A firm that cannot trace its electricity source cannot document its footprint, and a firm that cannot document its footprint cannot sell to the buyer who wrote the letter.
China is the world's largest manufacturer and its largest emitter. Its export sector is deeply embedded in global value chains, and its domestic market is large enough to support scale economies in green technology. Export exposure and domestic scale together mean that Chinese firms face foreign carbon regulation and a large home market for low-carbon products at the same time. Some will navigate both. Others will relocate or exit.
1.4 The Empirical Puzzle
If green governance capability were simply a function of regulatory pressure, firms facing the same pressure would converge on the same practices. They do not. The variation is the empirical heart of this book.
Consider the furniture cluster in Shunde. Within a fifteen-kilometer radius, firms face the same local environmental enforcement, sell to overlapping buyer bases, and draw on the same labor pool. Their responses diverge sharply. One firm I visited in 2022 had reformulated its coating line to water-based finishes, installed a heat-recovery system on its drying kilns, and hired a full-time environmental engineer who sits in on product design meetings. Its per-unit carbon footprint, by its own accounting, fell 31 percent between 2019 and 2023, and it used the documentation to win a Dutch contract that two competitors lost. Another firm, similar in size and export orientation, had done the minimum: it paid a consultant for a footprint report, filed it with the buyer, and continued to quote on price. A third had begun shifting assembly to a leased facility in Vietnam, keeping design and sales in Foshan. A fourth had exited altogether.
The EU ETS evidence shows the same dispersion. Martin, Muûls, and Wagner (2016) found that higher energy intensity and higher carbon cost raised the probability of investment in low-carbon technology, but the relationship was not deterministic: some high-cost firms did not invest, and some low-cost firms did. The green supply chain literature finds supplier compliance varying widely within the same buyer's supply base, explained partly by the supplier's existing capabilities and partly by the nature of the buyer relationship (Gimenez and Tachizawa 2012). Cluster studies find firms in clusters with strong institutional support and dense supplier networks adopting green practices more readily than firms in thin, weakly supported clusters.
If capability were just regulatory pressure, the Shunde firms would look alike. They do not. Capability is a separate variable, with its own determinants and its own effects. Identifying those determinants and tracing those effects is the work of the chapters that follow.
1.5 What Existing Frameworks Miss
Three bodies of theory speak to this question, and each explains part of the variation while missing part of it.
The resource-based view, extended to environmental performance as the natural-resource-based view, argues that firms developing environmentally oriented capabilities—pollution prevention, product stewardship, sustainable development—can achieve competitive advantage because those capabilities are valuable, rare, and hard to imitate. The argument is compelling in principle and has been criticized for tautology: successful firms are said to have the capabilities that explain their success, and failing firms are said to lack them. The framework does not specify the conditions under which capabilities are built, and it does not explain why firms under similar pressure develop different ones.
The global value chain framework, extended as green GVCs, argues that lead firms set environmental standards and enforce them through supplier selection, and that suppliers in developing countries upgrade in response. This explains how standards travel. It has been criticized for determinism: suppliers appear as passive recipients of buyer requirements, and the framework does not explain why some upgrade while others do not. It also concentrates on the buyer-supplier dyad at the expense of the cluster and institutional context in which suppliers operate.
Innovation studies, extended as eco-innovation, emphasizes technological learning and the learning curve: firms move down the cost curve as they accumulate experience with new technologies, and the resulting cost reductions make green products competitive. This is powerful for technology-intensive industries and weak on system context. It does not explain how the learning curve is shaped by supply chain relationships, cluster dynamics, and governance structures.
The capability-based view developed in this book integrates the three by treating them as answers to different parts of one question. The GVC literature explains the pressure; the resource-based view explains the heterogeneity of response; innovation studies explain the dynamics of learning. What none of them supplies, and what the book adds, is an account of the conditions under which the three interact to produce capability—and an account of why those conditions differ systematically across industries with different factor intensities.
1.6 The Policy Puzzle
If standards alone produced upgrading, two decades of proliferating environmental standards would have produced a matching proliferation of green capabilities. They have not. Standards create pressure. They do not create capability. Capability has to be built through investment, learning, and organizational change, and those processes respond to factors that standards do not directly touch.
The Chinese policy mix illustrates the gap. Green credit has grown fast, but the system has been criticized for counting the quantity of credit rather than monitoring whether financed projects achieve their stated environmental outcomes. The national ETS, initially covering the power sector and now expanding, has a low carbon price and generous free allocation, which weakens the incentive to invest in low-carbon technology. Building standards such as GB/T 50378–2014 are weakly enforced and cover new construction far better than existing stock. None of these criticisms argues against standards. They argue for designing standards and supporting policies so that they reward capability building rather than merely imposing cost—a question I take up in Chapter 9.
1.7 Research Questions
The central question of the book is: under what conditions does green governance capability become an endogenous driver of industrial upgrading and value chain climbing, and how does the process differ across industries with different factor intensities?
Three components need separate treatment. The conditions operate at the firm level (resources, routines, relationships), the cluster level (shared infrastructure, supplier networks, labor markets), and the governance level (policy design, enforcement, international coordination). The process is path-dependent and unfolds over time: the capabilities a firm can build depend on the ones it already has, and the pathways available depend on its position in the value chain. The comparison across industry types is not taxonomic. It asks which mechanisms operate in which context, and under what conditions.
Five subsidiary questions follow. What is the theoretical relationship between green governance capability and value chain climbing? (Chapter 2.) How does the dual carbon context reshape the competitive landscape for Chinese manufacturing? (Chapter 3.) How can the capability be conceptualized and measured? (Chapter 4.) How do pathways differ across labor-intensive, technology-intensive, and capital-intensive industries? (Chapters 5 through 7.) What institutional and cluster conditions enable or constrain capability building, and how do they operate in the furniture case? (Chapters 8 and 9.)
1.8 Method and Evidence
The book combines case studies, econometric analysis, and structured comparison. The combination follows from the question, which needs both depth—to see the mechanisms through which capability is built—and breadth—to establish the conditions under which they operate across industries.
The case studies supply the depth. Between 2018 and 2024 I conducted firm-level and cluster-level fieldwork in the furniture, electronics, machinery, steel, and cement industries, in clusters across Guangdong, Zhejiang, Jiangsu, Shandong, and Hebei. The work involved semi-structured interviews with firm managers, local government officials, industry association staff, and supply chain partners; site visits to production facilities; and collection of firm documents including environmental reports, carbon footprint declarations, and supplier assessment records. Cases were selected for theoretical replication, not statistical representativeness: they vary on factor intensity, export orientation, value chain position, and engagement with green governance, so that the mechanisms can be observed in operation.
The econometric work supplies the breadth. It draws on the Chinese Industrial Enterprises Database, the China Customs Database, and a proprietary firm-level environmental performance dataset collected with collaborators. Panel models with firm fixed effects control for time-invariant unobservables; difference-in-differences designs exploit policy changes as natural experiments; instrumental variable estimation addresses reverse causality, using the timing of policy changes, the geographic distribution of enforcement, and firm exposure to foreign markets with different environmental standards. Selection into green programs is handled with propensity score matching and Heckman selection models. No single estimate is decisive. The goal is triangulation across methods and data sources.
Three identification problems deserve mention because they shape how the results should be read. Reverse causality: firms that upgrade may invest in green capability rather than the reverse. Omitted variables: firms that invest may differ in management quality, organizational culture, or access to finance in ways that also affect upgrading. Selection: firms that join green programs may be systematically different from those that do not. The econometric chapters state which technique addresses which problem and where the technique falls short.
The limitations define the scope. The empirical analysis is Chinese; the framework is meant to be general, but its application here is specific to a country with a particular institutional environment and position in global value chains. The sectoral coverage is three industry types and a handful of sectors within each; services and agriculture are outside the scope. The temporal coverage runs from the early 2000s to 2024, with emphasis on the period since the dual carbon announcement, and the findings may not hold as the policy environment shifts. The case evidence rests on a limited number of interviews, and the econometric evidence on observational data subject to measurement error. The claims are conditional: the book argues that green governance capability can drive upgrading under specified conditions, and that understanding those conditions matters for firms, policymakers, and scholars navigating the transition.
1.9 Terminology
Green governance capability, as I use the term, refers to the organizational capacity to interpret environmental requirements, redesign products and processes in response, and capture value from the resulting differentiation. It is narrower than green governance in the institutional sense (the rules and bodies that govern environmental behavior) and narrower than green capability in the loose sense of any environmental performance. It concerns the firm's ability to govern its own environmental performance in the context of external requirements.
Value chain climbing refers to movement from lower-value to higher-value activities in the chain. It is broader than process upgrading (doing the same thing more efficiently) and different from diversification (entering new product markets). Climbing combines process, product, and functional upgrading in a way that shifts the firm's position in the chain.
Dual carbon is shorthand for China's goals of peaking emissions before 2030 and reaching neutrality before 2060, and for the policy architecture, market signals, and competitive dynamics that have grown up around them.
Factor intensity refers to the relative importance of labor, technology, and capital in production. I use the conventional three-way classification while recognizing that the boundaries are not sharp and that many industries straddle categories.
1.10 Plan of the Book
Chapter 2 builds the theoretical framework from the GVC literature, the resource-based view, and innovation studies, and sets out a capability-based view of green competitiveness. Chapter 3 examines the dual carbon context: the policy architecture, carbon pricing under the national ETS and the EU ETS, CBAM, and the competitive landscape they produce. Chapter 4 defines green governance capability, proposes a measurement framework, and identifies its endogenous drivers at the firm, cluster, and governance levels.
Chapters 5, 6, and 7 take the three industry types in turn. Chapter 5 examines labor-intensive industries—compliance pressure, relocation risk, and the limits of cost-based upgrading—with the furniture and apparel sectors as the main cases. Chapter 6 examines technology-intensive industries, where green competitiveness runs through innovation and the learning curve, drawing on electronics and machinery. Chapter 7 examines capital-intensive industries—steel, cement, chemicals—where the questions are scale, stranded assets, and the governance of low-carbon process substitution.
Chapter 8 applies the framework to the furniture industry under dual carbon, tracing how CBAM and buyer footprint requirements affect export competitiveness and relocation, and why firms in the same cluster diverge. Chapter 9 examines institutions, clusters, and multi-level governance, identifying the conditions under which capability building is most and least likely. Chapter 10 synthesizes the findings and draws out the theoretical and practical implications.
The book's contributions follow from its argument. For the GVC literature, it shows that suppliers can be active agents in governance, that their capacity to act is shaped by cluster and institutional context, and that pathways differ across industry types in ways a single governance model cannot capture. For the resource-based view, it specifies the conditions under which green capabilities are built, responding to the tautology criticism by identifying firm-, cluster-, and governance-level enablers, and it treats the capability as dynamic—sensing, seizing, reconfiguring—rather than as a static resource. For innovation studies, it shows that the learning curve is shaped by supply chain relationships, cluster dynamics, and governance structures, not only by cumulative production. The practical contributions run in parallel: a framework managers can use to assess their capability and its gaps; evidence for policymakers on when standards and supporting instruments actually induce capability building rather than cost; and evidence on how domestic and international regimes interact, which bears on the design of cooperation on carbon pricing and product standards.
The evidence base is mixed by design, and it is worth being explicit about what each part can support. The case studies show mechanisms in operation and let the reader see the choices managers face and trace their consequences; they are theoretically representative, not statistically representative, and should not be read as population averages. The econometric analysis shows patterns across larger samples and tests the robustness of the case findings; it is subject to the usual limits of observational data, and the chapters that report it state the assumptions and caveats. The policy analysis draws on government documents, industry reports, and secondary data to assess what the instruments can and cannot achieve; it is not a substitute for full program evaluation. No single source settles the argument. Together they support the conclusion that green governance capability is endogenous to firms and clusters, and that its development depends on how firm resources, cluster dynamics, and governance structures interact.
The Shunde owner has a decision to make before the next buying season. He can hire a consultant, file the footprint report, and hope the standard holds still. Or he can put an engineer into the design meetings, trace the carbon in his board and his coatings, and quote the Dutch contract his competitor cannot document. The choice is his. The conditions that shape it—the cluster's supplier base, the district's enforcement posture, the buyer's 2030 pathway, the price of carbon on the EU ETS—are not. This book is about those conditions, and about what it takes for a firm to build, rather than buy, the capability to meet them.