Chapter 8: The Furniture Industry under Dual Carbon: Export Competitiveness, Carbon Border Adjustment, and Relocation Dynamics

Liang has run the sofa plant at Lecong, in Foshan's Shunde District, since 2008. The factory has made upholstered sofas for European buyers for nineteen years; his father founded it in 1996. On a Tuesday morning in March, forty-foot containers stood stacked beside the loading dock, bound for Rotterdam, Hamburg, Antwerp, Le Havre. Liang's spreadsheet tracks the destination of every container. Since 2022 it has tracked one more column: the estimated carbon content of each order, built up from electricity consumption, foam and fabric inputs, and the diesel burned by trucks hauling finished goods to Shenzhen.

He started the column because a German buyer's sustainability questionnaire arrived with a deadline and a footnote citing the Carbon Border Adjustment Mechanism. He filled it out. Then he kept filling it out. The column is now the most consulted part of the spreadsheet, and the number in it has changed how Liang thinks about foam suppliers, electricity contracts, and which orders are worth chasing.

Furniture is a useful test case for this book's argument because it is labor-intensive, export-oriented, highly fragmented, and increasingly sold into markets with carbon pricing. The regulatory pressure is real but indirect: the EU's CBAM does not yet cover furniture, and the industry's response is therefore a study in how firms act on expectations of future policy rather than on the policy of the present. Firms make investment decisions on expectations of future policy; the unsettled regime is the relevant object of study.

8.1 The Carbon Border Adjustment Mechanism and the Furniture Value Chain

8.1.1 What CBAM does, and what it does not yet do

Regulation (EU) 2023/956 entered into force in May 2023. The EU Emissions Trading System imposes a carbon price on domestic producers of covered goods. Without a border measure, production could shift to jurisdictions without equivalent carbon pricing and the emissions would relocate rather than decline. CBAM charges importers of covered goods for embedded emissions multiplied by the gap between the EU ETS price and any carbon price already paid in the country of origin. During the transitional period, October 2023 to the end of 2025, importers report embedded emissions quarterly with no financial adjustment. The financial obligation phases in from 2026.

Wooden furniture, upholstered furniture, metal furniture, and mattresses do not appear in Annex I. The immediate legal burden on a Foshan sofa maker is zero. The mechanism nonetheless reaches the industry through three routes, and the third is the one that matters most.

The input route runs through materials. Furniture contains steel in frames and mechanisms, aluminium in legs and fittings, glass in table and cabinet doors. These are covered goods. A Chinese furniture exporter buying steel from a Chinese mill pays a price that reflects the mill's CBAM exposure on its European sales, and the mill's compliance costs propagate through the domestic supply chain. The magnitude is modest: steel and aluminium account for perhaps 8 to 15 percent of material cost in a typical metal-frame office chair, less in an upholstered sofa.

The customer route runs through procurement. European retailers with scope 3 reporting obligations, and with investors and NGOs watching, now demand product carbon footprints from suppliers regardless of legal requirements. The German questionnaire in Liang's inbox is an instance.

The anticipatory route runs through investment timing. The Commission's review of CBAM's scope, together with the precedent of past ETS expansions, gives firms reason to expect downstream extension. The Commission's own impact assessment for the CBAM proposal noted that extending the mechanism to downstream products would be administratively complex because furniture is assembled from many components with different carbon intensities. That complexity is the point. The administrative difficulty of attributing emissions to a sofa follows from the fragmentation of its supply chain, and that fragmentation is a legacy of the cost-based model this book has described. A firm that knows the carbon content of its foam, fabric, frame, and packaging has already done the work that downstream extension would require. A firm that does not faces a problem no regulatory notice can solve quickly.

8.1.2 Embedded emissions and the attribution problem

Consider what embedded emissions mean for a product with a bill of materials running to hundreds of line items. The CBAM methodology for the transitional period requires importers to report direct emissions from production processes and indirect emissions from electricity consumption, preferring actual values and permitting default values as a fallback. For steel, the production route is identifiable: blast furnace or electric arc furnace, with a known emissions intensity per tonne. For a sofa, the production route is a network.

Take a mid-range fabric sofa manufactured in Longjiang, the furniture town next to Lecong. Its components include a hardwood or plywood frame, polyurethane foam for cushioning, polyester or cotton-blend fabric, metal springs and fasteners, and corrugated cardboard packaging. Each component has its own supply chain, often in a different province, sometimes in a different country. The foam may be produced in Guangdong from petrochemical feedstocks; the fabric may be woven in Zhejiang from yarn spun in Jiangsu; the springs may come from a specialist manufacturer in Hebei. Electricity consumed in assembly is metered. Electricity consumed upstream is not. A full product carbon footprint requires primary data from every supplier, or estimates from databases whose coverage of Chinese supply chains is incomplete.

The firm that solves this for its own products gains an informational asset its competitors lack. The asset is a map of the supply chain's carbon hotspots, which is to say a map of where abatement is cheapest. Liang's spreadsheet, crude as it is, told him that foam and fabric dominate the footprint of his sofas and that electricity in his own plant is a smaller share than he had assumed. That knowledge redirected his attention from the factory floor to his suppliers.

Most Chinese furniture exporters have not built carbon accounting systems. The China National Furniture Association's 2023 survey of exporting members found that a minority had calculated a product carbon footprint for any product line, and most of those had done so at the request of a single large customer rather than as a strategic exercise. The distribution is skewed: large firms with European brand customers are far ahead of small and medium enterprises selling through trading companies or to wholesale markets. Capability accumulates where the value chain relationship is closest and the customer is most demanding, which is consistent with the governance literature on buyer-driven chains (Gereffi, Humphrey, and Sturgeon, 2005).

8.1.3 The boundary question: CBAM and the definition of origin

Rules of origin complicate the picture. CBAM applies to goods imported into the customs territory of the Union. A sofa assembled in Vietnam from Chinese components and exported to the EU is a Vietnamese good for customs purposes, and its embedded emissions are calculated under the Vietnamese production route. That route may include emissions embodied in imported Chinese components, depending on the methodology, but it does not trigger CBAM on the Chinese inputs directly. The incentive is to relocate assembly outside the CBAM perimeter, or to jurisdictions with lower-carbon electricity.

The Commission is aware of the boundary problem. The regulation includes anti-circumvention provisions, and implementing acts may require reporting on precursors. Enforcement is another matter. Furniture has many possible assembly locations and a high ratio of value to weight, which makes it easy to move the final stage of production without moving the component supply chain. The location of assembly matters less than the location of the emissions-intensive inputs, and firms have an incentive to game the boundary unless the methodology captures precursors.

For Chinese furniture, the boundary question cuts both ways. Relocating final assembly to Southeast Asia could reduce the CBAM exposure of goods destined for Europe. The components would still be Chinese, and if the methodology eventually captures embedded emissions in precursors, relocation would yield little carbon advantage while adding logistics costs and sacrificing the agglomeration economies of Shunde and Longjiang. Firms that relocate purely to escape a regulatory boundary are betting on the boundary's durability. Firms that relocate to build a lower-carbon production system are betting on the value of the capability itself.

8.1.4 The timeline and the anticipation effect

CBAM's transitional period runs to the end of 2025, with the financial obligation beginning in 2026 and phasing in through 2034 as free allocation of ETS allowances to EU producers is reduced. The scope review is expected in the second half of the transitional period. For a furniture firm with a capital investment cycle of five to ten years, the relevant question is not what CBAM requires today but what it will require when the next factory is built or the next product line is designed.

Firms that expect downstream extension begin measuring now. Firms that do not, wait. The waiting firms save the cost of measurement, which is real but modest: establishing a product carbon footprint system for a mid-sized exporter typically runs from 50,000 to 200,000 RMB, based on quotations I collected from three carbon accounting consultancies in Guangzhou and Shenzhen in 2024. The measuring firms incur that cost and accumulate data that will be valuable when the regulation arrives, and that is already valuable for customer relations and internal cost management.

The EU is not the only jurisdiction developing border measures. The UK announced its own CBAM, to take effect in 2027, covering a similar set of goods. The US has proposed a Clean Competition Act, though its legislative prospects are uncertain. Firms that build measurement capacity for one market can adapt it for another. The capability is general even when the regulation is specific, which is an argument for treating CBAM as a forcing device for a capability that multiple customers in multiple markets will demand, rather than as a compliance cost to be minimized.

8.2 Export Competitiveness under Carbon Constraints

8.2.1 The anatomy of a cost advantage

Chinese furniture's export competitiveness has rested on cheap labor, scale, and supply chain density. The labor advantage has been eroding for two decades. Average manufacturing wages in Guangdong rose from roughly 15,000 RMB per year in 2005 to over 80,000 RMB by 2022, compressing the gap with Southeast Asia and, in some segments, with Eastern Europe. The scale advantage is real but limited: furniture is a batch-production industry with significant product differentiation, and economies of scale flatten beyond a certain plant size. Supply chain density is the most durable of the three. Shunde and Longjiang host clusters of component suppliers, finishing specialists, logistics providers, and design services that no other location matches for variety and speed. A sofa maker in Longjiang can source foam, fabric, springs, and hardware within a thirty-kilometer radius and prototype a new design in a week. That density does not appear in unit cost calculations. It shows up in responsiveness and customization.

Carbon constraints interact with these pillars differently. Labor costs are unaffected in the short run, though energy costs may rise as carbon pricing propagates through the electricity system. Scale is largely unaffected, except that carbon measurement systems carry fixed costs that favor larger firms. Supply chain density is affected in a more interesting way: carbon accounting requires supply chain transparency, and transparency is easier when suppliers are geographically close and relationally embedded. The cluster may be an asset for green governance rather than a liability.

The cost that carbon constraints most directly reshape is energy. Furniture manufacturing is not energy-intensive in the way steel or cement is, but it uses significant electricity for woodworking machinery, dust extraction, drying, and lighting, and thermal energy for drying and finishing. A typical plant might consume 200 to 400 kWh per cubic meter of finished product, with wide variation by product type and process. At an industrial electricity price of 0.7 RMB per kWh, electricity is perhaps 3 to 6 percent of total production cost. Carbon pricing at the EU ETS level of 80 to 100 EUR per tonne of CO2, applied to the embodied emissions of a sofa, would add a few percentage points to landed cost in Europe. The arithmetic is not catastrophic. The strategic question is whether those points fall on a margin already thin.

They do. Chinese furniture contract manufacturers typically operate on net margins of 3 to 8 percent, lower in commodity segments. A carbon cost of 2 to 5 percent of landed value, fully passed through to the manufacturer, would eliminate the margin for a significant share of exporters. Passed through to the consumer, it would erode the price advantage relative to European, Vietnamese, and Mexican producers. The incidence depends on market structure: in segments where Chinese suppliers have pricing power—custom upholstery, complex case goods—the cost can be passed through; in segments where they compete on price with many substitutes, it cannot. The distributional effect within the industry is regressive, falling hardest on commodity producers with the least pricing power and the least capacity to invest in abatement.

8.2.2 From cost advantage to differentiation: the green premium question

If carbon costs erode the cost advantage, the strategic response is to shift the basis of competition from cost to differentiation. The green premium—what consumers will pay for a lower-carbon product—is the mechanism. Does it exist in furniture?

The evidence is segment-specific. At the high end of the European market, brands such as Vitra and Herman Miller, along with a cluster of Scandinavian and German manufacturers, have built sustainability into their positioning, and their customers pay premiums that can exceed 20 percent. At the middle and low end, the premium is thin or absent. A 2022 survey by the European Furniture Industries Confederation found that while 68 percent of consumers said sustainability influenced their purchase decisions, only 22 percent were willing to pay a premium of more than 5 percent. The gap between stated and revealed preference is familiar from the environmental economics literature (Auger et al., 2003; Gleim et al., 2013). The green premium is real but narrow.

For Chinese exporters, the premium is accessible mainly through business-to-business channels. A contract manufacturer that can document a lower product carbon footprint can win orders from European brands with public scope 3 commitments. The premium in this case is not a higher price but a preferred position in the sourcing relationship: longer contracts, larger volumes, earlier involvement in product design. The firm moves from transactional supplier to strategic partner, and the governance relationship shifts from market-based to relational (Gereffi, Humphrey, and Sturgeon, 2005). Carbon capability is the ticket to the relational tier.

A second differentiation pathway does not depend on consumer willingness to pay. Furniture is increasingly subject to extended producer responsibility schemes in Europe, and the EU's Ecodesign for Sustainable Products Regulation, adopted in 2024, will set requirements for durability, repairability, and recycled content across product categories, with furniture among the priority groups. These regulations create a different kind of premium: compliance with design requirements that not all competitors can meet. A firm that designs for disassembly, uses recycled foam, and offers repair services will be positioned for a regulatory environment in which the baseline product is harder to sell. The capability here is design for sustainability, which requires engineers and designers who understand materials and lifecycle, not just accountants who can fill out forms.

One upholstered furniture manufacturer in Longjiang, founded in 2003, began offering a take-back service for its European contract customers in 2021, refurbishing returned sofas and reselling them in secondary markets. The service required a reverse logistics system, a refurbishment workshop, and a product design that allowed easy replacement of covers and foam. The firm's export margins rose by two percentage points, not because it charged more for new sofas but because the take-back service won it a preferred supplier agreement with a large German retailer. The capability was built incrementally over three years, and it now constitutes a barrier to entry that the firm's price-competitive neighbors cannot easily cross.

8.2.3 The China domestic market and the dual carbon pull

Export competitiveness is not the only driver. China's domestic dual carbon policy creates its own pressures and opportunities. The national ETS, launched in 2021, initially covered only the power sector, and its scope is expanding to steel, cement, and aluminium in the coming years. Furniture manufacturing is not a direct ETS participant, but it is affected through electricity prices, which rise as the power sector passes through carbon costs, and through supply chain pressures as upstream materials become more expensive.

The domestic market also offers a testing ground for green products. Chinese consumers, particularly in tier-one cities, have become more attentive to indoor air quality and formaldehyde emissions, and the national standard GB 18580-2017 set limits that are stricter than those in many export markets. The zero-formaldehyde plywood and low-VOC finishes that firms developed for the domestic market have become selling points in Europe as well. The dual carbon goals reinforced this trajectory: the Ministry of Industry and Information Technology's green manufacturing program has designated hundreds of green factories and green supply chain enterprises, and furniture firms that qualify receive recognition and, in some provinces, subsidies. The domestic policy environment is not purely a cost; it is a source of capabilities that can be leveraged in export markets.

The dual carbon goals were announced in 2020, before CBAM's final design was settled, and they drove a domestic policy architecture that includes green factory certification, energy efficiency standards, and renewable energy targets. Firms that engaged with this architecture early, often for reasons of domestic market positioning or local government relations, found themselves better prepared for European requirements than firms that ignored it. The capability was built at home and exported, not imported from Brussels. The direction of causality matters for this book's argument: green governance capability is endogenous to the firm's trajectory, not a response to a single external shock.

8.2.4 The measurement infrastructure gap

Product carbon footprint calculation requires two things many firms lack: activity data (how much energy, material, and transport went into a product) and emission factors (how much carbon is embodied in each unit of activity). Activity data can be collected with meters and bills, though few firms have systems that allocate energy use to specific products rather than to the plant as a whole. Emission factors are harder. China's national emission factor database is incomplete for many materials, and the databases used in Europe—Ecoinvent, GaBi—have limited coverage of Chinese supply chains and are expensive to license.

The gap has spawned a cottage industry of consultants and software providers. In Shenzhen and Guangzhou, several firms now offer carbon accounting services for exporters, using database factors combined with primary data collection. Their fees range from 30,000 to 150,000 RMB per product line, and the quality of their work varies widely. The absence of a standardized, affordable, China-specific measurement infrastructure is a collective action problem: no single firm can solve it, and the industry association has been slow to act. Firms that have built internal capacity, usually by hiring an environmental engineer or training an existing staff member, hold an advantage that is partly informational and partly organizational. They know how to ask suppliers the right questions, and they know when a consultant's numbers are wrong.

The Ministry of Ecology and Environment has published a series of product carbon footprint standards, and the China National Institute of Standardization has been developing a national database. Progress has been slow relative to the pace of market demand. The firms that climb will be those that build capability despite the infrastructure gap, not those that wait for the infrastructure to be built.

8.3 Relocation Dynamics and Green Upgrading

8.3.1 Where furniture is moving, and why

The geography of Chinese furniture production has been shifting for more than a decade. Rising labor costs in the Pearl River Delta pushed low-end production inland to Jiangxi, Anhui, and Hubei, and offshore to Vietnam, Cambodia, and, more recently, Mexico. The US-China trade war accelerated the offshore shift: 25 percent tariffs on Chinese furniture made Vietnamese production attractive for US-bound goods, and Vietnam's furniture exports to the US rose from $3.5 billion in 2017 to over $9 billion by 2021. The pandemic disrupted this trajectory, and the direction resumed by 2023.

Carbon policy adds a vector to the relocation calculus. CBAM, even without downstream extension, creates an incentive to locate production in jurisdictions with lower-carbon electricity grids or with carbon prices recognized as equivalent. Vietnam's grid is more carbon-intensive than China's in some regions and less in others, depending on hydropower availability; the country has no carbon price. Mexico's grid is less carbon-intensive than China's average and benefits from proximity to the US market. For European-bound goods the calculation is more complex: Vietnam's lack of a carbon price means CBAM would apply to the full embedded emissions of Vietnamese goods, minus any carbon price paid, which is zero. Chinese goods would receive a credit for any carbon price paid in China. If China's national ETS expands to cover more sectors and a domestic carbon price emerges for furniture inputs, Chinese production could be advantaged under CBAM relative to production in countries without carbon pricing. CBAM is designed to equalize carbon costs, not to penalize Chinese production as such, and a Chinese firm that pays a domestic carbon price and can document it may face a lower CBAM adjustment than a Vietnamese competitor that pays nothing.

The relocation calculus is therefore not a simple story of carbon costs driving production away from China. Firms weigh labor costs, tariffs, logistics, market access, carbon costs, and supply chain density, and the resulting location decisions vary by product segment and destination market. The firms most likely to relocate produce commodity goods for price-sensitive markets, where labor costs dominate and carbon costs are secondary. The firms least likely to relocate produce customized, design-intensive goods for European and North American brands, where supply chain density and design capability matter more than labor costs. Relocation is segment-specific, and upgrading and relocation can be complementary strategies rather than substitutes (Humphrey and Schmitz, 2002; Kaplinsky and Memedovic, 2003).

8.3.2 Relocation as escape versus relocation as upgrading

The distinction that matters is between relocation as escape and relocation as upgrading. Escape is the movement of production to a jurisdiction with lower costs or weaker regulation, with no change in the firm's capabilities. The firm that moves a cutting-and-sewing operation from Shunde to an industrial park outside Ho Chi Minh City, without changing its product, process, or customer relationship, is escaping. Upgrading is the movement of production to a location that enables a new capability: a lower-carbon process, a new market, a different position in the value chain. The distinction is not about geography. It is about what the firm does with the move.

Both patterns appear in Vietnam. Many Chinese-owned furniture factories in Binh Duong and Dong Nai provinces are escape operations: they replicate the Shunde model with cheaper labor, import Chinese components, and sell to the same US and European buyers. Their carbon performance is often worse than the Chinese parent's, because Vietnam's grid is more carbon-intensive in the south and because the factories are newer and less efficient. Other Chinese firms have used Vietnam as an upgrading platform: they have invested in solar panels, built design studios, and moved into branded or private-label segments that require more than price competitiveness. The difference is visible on the plant floor and in the customer list.

For European-bound goods, the upgrading path has a specific logic. A firm that wants to sell to EU brands under CBAM's shadow needs to demonstrate lower embedded emissions. It can locate production in a jurisdiction with a low-carbon grid, source lower-carbon materials, or redesign products to use less material and more recycled content. Vietnam offers a mixed proposition on the first dimension and little on the second and third. China offers the opposite: a grid that is decarbonizing rapidly, with the share of renewables in electricity generation rising from 26 percent in 2015 to over 30 percent by 2023 against a government target of 40 percent by 2030; a growing supply of recycled materials; and a manufacturing ecosystem that can support design for sustainability. The country that appears most exposed to CBAM may, over time, become one of the better locations for low-carbon furniture production, if firms build the capability to document and improve their performance.

8.3.3 The cluster as a green governance asset

The Shunde-Longjiang cluster shows how agglomeration can support green upgrading. Suppliers are close, relationships are long-standing, and information flows through personal networks as well as formal contracts. These features, assets for cost-based competition, are also assets for carbon management. A firm that wants to know the carbon content of its foam can call its foam supplier, who is likely a neighbor or a relative, and ask for data. The supplier may not have the data, but the conversation is easier than it would be with a distant vendor, and the firm can send an engineer to help the supplier measure. The cluster reduces the transaction costs of supply chain carbon management.

Cluster relationships can also be a source of inertia. Firms are embedded in networks of reciprocal obligation that make it difficult to switch to lower-carbon suppliers or to demand changes that disrupt established practices. The same social capital that facilitates information sharing can inhibit the diffusion of new practices when those practices threaten existing relationships. Two firms in Longjiang illustrate the divergence. One, a 400-worker upholstery plant with a German contract customer, sent its own engineer into two foam suppliers' plants in 2023 to install meters and collect data; both suppliers now report monthly figures. Another, a similar-sized plant selling through trading companies, asked its three closest suppliers for emissions data in the same year and received estimates from one; the other two said they had no way to produce the numbers, and the firm did not press, because the owner's brother-in-law runs one of them. The difference tracks the lead firm's market power and whether a demanding customer is watching.

The local government has played a role. Shunde's economic planning bureau has promoted green factory certification and offered subsidies for energy efficiency upgrades and solar installations. The Furniture Association of Shunde has organized training sessions on carbon accounting, though attendance has been modest. The cluster's green governance is emerging, not yet institutionalized. The firms that are climbing often do so despite the cluster's institutions rather than because of them, relying on their own resources and their relationships with European customers.

8.3.4 The limits of relocation as a strategy

Relocation has limits as a response to carbon constraints. Carbon pricing is spreading: the EU's CBAM is the most developed border measure, and the UK, Canada, and Japan are exploring or implementing comparable mechanisms. A firm that relocates to escape one jurisdiction's carbon price may find the destination covered by another, or that its customers demand the same performance regardless of location. Relocation also does not build capability. A firm that moves production without changing its processes or products remains a cost competitor, perpetually vulnerable to the next lower-cost location. And relocation can destroy the cluster-based advantages that made the firm competitive. The Shunde model depends on proximity, and a firm that moves assembly to Vietnam while keeping design and sourcing in China may find that coordination costs outweigh labor savings.

The firms that have relocated successfully used the move to build a new capability. A Hong Kong-invested furniture manufacturer moved its upholstery production to Vietnam in 2019 but kept design, marketing, and component sourcing in Shunde. The Vietnamese plant was designed from the start for low-carbon production: solar panels on the roof, a biomass boiler for drying, a closed-loop water system for finishing. The firm's European customers accepted a price premium of 4 percent, and the plant's carbon footprint per unit runs about 30 percent below the Shunde plant's, according to the firm's own 2023 figures. The relocation was an upgrading move that used a new location to implement a capability the firm had been developing in China.

The geography of furniture production is becoming more complex, with different stages of the value chain in different places according to factor intensity and carbon intensity. Design and marketing stay close to customers; component production stays in clusters with supply chain density; assembly moves to locations with favorable cost and carbon profiles. The firm's capability holds the network together, and the capability is portable across locations. Relocation without it cannot create capability, and relocation with it does not destroy it.

8.3.5 Policy implications and the domestic carbon price

The most effective response for Chinese furniture firms may be a domestic carbon pricing system that covers their inputs, so that carbon costs paid at home are credited against CBAM obligations at the border. The EU regulation explicitly provides for such credits: importers can deduct any carbon price effectively paid in the country of origin. If China's national ETS expands to cover steel, aluminium, and petrochemicals, as planned, and if the resulting carbon costs pass through to furniture manufacturers, those manufacturers will have a documented carbon price to present to European importers. The net CBAM adjustment would be lower than for a competitor in a country with no carbon price.

Chinese policymakers have begun to appreciate this. The Ministry of Ecology and Environment has been studying the interaction between the national ETS and CBAM, and industry associations have lobbied for ETS expansion to cover export-exposed sectors. The logic is not to avoid carbon pricing but to capture the revenue at home rather than paying it at the border. For firms, the domestic carbon price is not an enemy. A firm that pays a carbon price at home and can document it holds an advantage under CBAM relative to a firm in a jurisdiction without carbon pricing. The advantage is small in the near term, because China's carbon price is low—around 60 to 80 RMB per tonne in the national ETS as of 2024, against 80 to 100 EUR in the EU ETS—but it is directional. Firms that build measurement and documentation systems now will be able to claim the credit when the price rises.

8.4 Capability Building in the Furniture Value Chain: Evidence from Firms

8.4.1 A typology of firm responses

Interviews and company documents suggest four responses to carbon constraints, arrayed along a gradient rather than sorted into boxes.

The passive complier fills out customer questionnaires when asked, uses default emission factors where possible, and changes nothing in its processes. Carbon reporting is an administrative burden, and the firm seeks the lowest-cost way to satisfy the request. Most small and medium exporters sit here.

The measurement adopter has built a carbon accounting system, often with external help, and uses it to answer customer requests and to find cost savings. It has not changed its product or process fundamentally, but it can answer questions quickly and accurately and can show improvement over time. Perhaps a quarter of mid-sized exporters have reached this stage.

The process upgrader has used measurement to invest in abatement: solar panels, energy-efficient machinery, water-based finishes, recycled foam. It has cut its product carbon footprint by 15 to 30 percent and can document the reduction, and it has won preferred supplier status with at least one European customer. Among mid-sized exporters, roughly one in ten has reached this stage.

The value chain climber has moved beyond process upgrading to product and business model innovation: design for disassembly, take-back services, circular material flows, co-development with customers. It participates in the customer's product design process and holds a relational rather than transactional governance relationship. This is the smallest category, perhaps one in twenty, and the one that most clearly demonstrates this book's thesis: the climber has converted carbon compliance into a competitive advantage that is difficult to imitate.

The distribution is not static. Firms move between categories as capabilities develop and the regulatory environment changes. What determines the rate of movement is a combination of firm-level factors—leadership, resources, existing technical staff—and ecosystem-level factors: customer pressure, cluster institutions, government programs, and the availability of measurement infrastructure.

8.4.2 The role of the buyer

The single most important driver of movement from passive compliance to measurement adoption is the buyer. Firms selling to European brands with public sustainability commitments face scrutiny that firms selling to wholesalers or trading companies do not. The buyer provides the specification (what to measure, which methodology), the incentive (preferred supplier status, longer contracts), and sometimes technical assistance (training, software, consultant referrals). Buyer-driven governance, documented in the literature for labor standards and quality, extends to carbon.

The distributional implication is that firms integrated into European brand supply chains will build capability faster than firms selling through intermediaries. The latter may be pushed out of the European market not by regulation but by brands' sourcing decisions favoring suppliers who can document performance. The export sector consolidates around a smaller number of capable firms, with the rest relegated to domestic markets or price-sensitive destinations. Chinese customs data show the number of Chinese furniture exporters to the EU declining by roughly 15 percent between 2019 and 2023 while average export value per firm rose, a structural shift that predates CBAM and that the carbon constraint is accelerating.

8.4.3 The limits of firm-level capability

A furniture manufacturer can measure its own emissions, switch to renewable electricity, and redesign its products. It cannot decarbonize its suppliers' operations or the logistics that deliver its goods to Europe. The scope 3 emissions that dominate a furniture product's footprint—materials, components, transport—lie outside the firm's direct control. Climber firms address this through supply chain collaboration, and collaboration has limits when suppliers are small and lack the resources to invest, and when the firm's leverage is limited by the availability of alternative suppliers.

Here the cluster and the policy environment re-enter the analysis. A single firm can persuade its closest suppliers to measure their emissions. It cannot create a market for low-carbon foam or recycled fabric. That requires collective action: an industry association setting standards, a local government providing technical assistance, or a group of lead firms coordinating supplier requirements. The Shunde cluster has the density to support such action, and its institutions are not yet oriented toward it. The Furniture Association of Shunde has focused on trade promotion and exhibition organization; its environmental work amounts to occasional training sessions. The local government's green factory program provides recognition and modest subsidies without coordinating supply chain decarbonization. The gap between firm-level capability and ecosystem-level infrastructure is the binding constraint on the industry's green upgrading.

The capability-based view does not claim that firms can build green governance capability in isolation. It claims that the capability is endogenous to the firm's strategy and organization, while recognizing that the firm operates within an institutional environment that enables or constrains its choices. The furniture case shows both sides: firms with strong buyer relationships and committed leadership have built measurement and abatement capability despite weak ecosystem support, while firms without those internal conditions have not, even where ecosystem support exists. The internal and external factors interact, and the interaction is where policy should focus.

8.4.4 Cost, benefit, and the payback period

The economics of green upgrading in furniture are more favorable than the sector's low margins suggest. The abatement measures available to a typical plant—LED lighting, variable-frequency drives on motors, heat recovery from drying kilns, rooftop solar—have payback periods of two to six years at current Chinese energy prices, according to equipment suppliers I interviewed in Guangdong in 2024 and to case data published by the China Energy Conservation Association. A 500 kW rooftop solar installation, costing roughly 2 million RMB after subsidies, generates about 500,000 kWh per year, saves around 350,000 RMB in electricity costs, and pays back in under six years. The payback shortens as electricity prices rise, which they are doing as the power sector passes through carbon costs.

The more ambitious measures—water-based finishes, recycled foam, design for disassembly—have longer paybacks and less certain returns, because they depend on customer willingness to pay and on the availability of secondary materials. Firms that have adopted them describe the investment as a market access cost rather than a cost reduction: money spent to win a customer relationship expected to be durable. The returns are relational rather than transactional, and they are difficult to fit into a conventional capital budgeting framework. This is a familiar problem in the literature on environmental investment (Hart, 1995; Porter and van der Linde, 1995): returns to green capability are often indirect and lagged, and firms using short payback criteria will underinvest.

Firms that have climbed describe a different decision calculus. They treat carbon capability as a platform investment, analogous to a quality management system or a design capability. The platform enables a stream of customer relationships and product opportunities that cannot be specified in advance, and the investment is justified by the option value of the platform rather than by the discounted cash flows of a single project. This view is more common among firms with longer time horizons: family-owned firms with second-generation leadership, firms with strong engineering cultures, firms that have already invested in design and branding. Firms lacking these characteristics tend to use the short payback criterion, and they tend to underinvest.

8.5 Three Trajectories to 2035

The analysis supports three scenarios for the Chinese furniture industry under dual carbon and CBAM. They are not predictions. They are structured ways of thinking about how firm capability, policy trajectory, and market demand interact.

8.5.1 The compliance trap

CBAM's scope extends to furniture in the late 2020s, but the methodology relies on default values and the administrative burden falls on importers. Chinese exporters provide the minimum information required, use default emission factors, and pass the carbon cost through where possible. The cost is modest, 2 to 4 percent of landed value, absorbed by margin compression and price increases. The industry's structure does not change: fragmented, cost-competitive, concentrated in commodity segments. Firms that cannot absorb the cost exit the European market or relocate assembly to Southeast Asia. The relocation is escape, not upgrading, and the relocated firms remain cost competitors in their new locations.

The compliance trap is stable in the short run and fragile in the long run. Firms in it have no capability to respond to a tightening methodology or a rising carbon price. They are price-takers in a market where the price of carbon is set by regulators they do not influence. The scenario is plausible because it requires no coordination and no investment beyond the minimum. It is the path of least resistance, and path dependence is a powerful force in an industry with thin margins and short time horizons.

8.5.2 The bifurcated industry

The industry splits. A group of lead firms—perhaps 10 to 15 percent of exporters—builds measurement, abatement, and design capability, wins preferred supplier status with European brands, and earns margins that support further investment. The rest remains in the compliance trap, selling to price-sensitive markets or exiting. The bifurcation is driven by buyer relationships and internal capabilities, and reinforced by the consolidation of European sourcing around a smaller number of capable suppliers.

This is the most likely trajectory on current evidence. It fits the observed distribution of firm responses, the consolidation of export channels, and the governance literature's finding that buyer-driven chains concentrate capability in a tier of preferred suppliers. Whether the bifurcation is acceptable is a policy question. From an efficiency standpoint it may be: capable firms capture the value, incapable firms find other markets. From an equity standpoint it is troubling: workers and communities dependent on exiting firms bear the adjustment costs, and the cluster's social fabric may be damaged by the loss of a large number of small producers.

8.5.3 The cluster upgrade

The cluster's institutions—the industry association, thelocal government, shared service providers—coordinate a collective upgrading process. A shared carbon accounting platform provides affordable measurement services to small firms. A supplier development program helps component manufacturers measure and reduce their emissions. A green procurement standard adopted by lead firms diffuses through the cluster. Capability spreads more widely than under bifurcation, more firms survive the transition, and the cluster retains its density.

The cluster upgrade is the most attractive scenario from a policy standpoint and the least likely on current trajectories. It requires collective action the cluster's institutions have not yet demonstrated, and coordination between firms that is difficult to achieve in a fragmented industry. Its elements are visible in fragments—the association's training sessions, the government's green factory program, the lead firms' supplier requirements—but they are not integrated into a coherent strategy. The scenario is a benchmark against which actual policy can be measured.

8.5.4 What the scenarios share

In all three, firms that build capability early do better than firms that wait. Under the compliance trap, early movers exit or relocate with some assets while laggards lose everything. Under bifurcation, early movers become the preferred suppliers and laggards become marginal. Under cluster upgrade, early movers lead the collective process and laggards follow. Timing matters because capability is cumulative and because the market rewards it before the regulation requires it. Firms that treat the current regulatory requirement as a ceiling will always be catching up. Firms that treat it as a floor will be building the next capability while their competitors are still filling out forms.

8.6 The Furniture Case and the Book's Argument

8.6.1 What the case confirms

The furniture case confirms several elements of this book's argument. Green governance capability is not a passive product of regulatory compliance: the firms that built measurement and abatement systems did so in advance of any legal requirement, driven by customer relationships and their own strategic assessment, and the regulation, when it arrives, rewards work already done. The capability is differentiated by factor intensity and by position in the value chain. Furniture is labor-intensive and fragmented, and its green governance pathway runs through supply chain collaboration and design rather than through the process substitution available to a steel plant; Chapters 6 and 7 examined industries where the pathway looks different. The capability is also a source of competitive advantage that is difficult to imitate. Climber firms have built relationships, systems, and knowledge that cannot be purchased off the shelf. The advantage is relational and organizational, and it accumulates.

The case confirms the importance of the cluster as well. The Shunde-Longjiang agglomeration provides the supply chain density that makes carbon management feasible and the social relationships that make supplier collaboration possible. The cluster is not a passive backdrop; it is an asset firms can use or neglect. Firms that have used it—working with nearby suppliers, participating in association activities, drawing on local government programs—have built capability faster than firms that have not.

8.6.2 What the case qualifies

The case also qualifies the argument. The compliance-versus-climbing dichotomy is too sharp for this industry. Most firms are neither pure compliers nor pure climbers; they sit somewhere between, moving incrementally as capabilities and customer demands evolve. The four-response typology captures that gradualism better than a binary distinction. The book's claim should be read as a statement about the direction of movement, not about a sharp line between two kinds of firms.

The domestic policy environment is more ambiguous than the framework might suggest. The dual carbon goals created a policy architecture that supports capability building, but the architecture is uneven and its implementation varies by locality. The green factory program provides recognition and subsidies without coordinating supply chain decarbonization. The national ETS creates a carbon price signal, but the price is low and coverage does not yet reach furniture inputs. The domestic environment is enabling, not sufficient. Firms still need internal drivers and external market pressure.

Relocation dynamics are also more complex than an escape-versus-upgrading dichotomy. Firms that relocate weigh labor costs, tariffs, logistics, market access, carbon costs, and supply chain density together, and outcomes depend on what the firm does with the move rather than on the move itself. The same destination can support escape and upgrading at once. Relocation is a strategic choice that interacts with capability building, not a predetermined outcome of cost pressures.

8.6.3 The policy agenda

At the firm level, the priority is measurement. A firm that does not know its carbon footprint cannot manage it, and a firm that cannot manage it cannot climb. The cost is modest relative to the value of the information; firms that have done it describe it as a turning point in their understanding of the business. The barrier is attention. Measurement needs an owner, and in a small firm that means the owner or a senior manager decides it matters.

At the cluster level, the priority is shared infrastructure: a carbon accounting platform serving small firms, a supplier development program building component manufacturers' capability, a green procurement standard coordinating lead firms' requirements. Together these would lower the cost and widen the diffusion of capability. The cluster's institutions are the natural owners of this agenda, and they need resources and mandates. Experience from clusters in Italy and Germany as well as in China suggests that shared infrastructure works best when co-financed by firms and government and run by professional staff.

At the national level, the priority is integrating carbon pricing with trade policy. Expanding the national ETS to steel, aluminium, and petrochemicals would create a documented carbon price firms can present to European importers under CBAM. A China-specific emission factor database would cut the cost and improve the accuracy of measurement. Mutual recognition agreements with the EU would reduce the administrative burden on exporters. None of these is furniture-specific, and all would bear disproportionately on labor-intensive export industries that lack the resources to navigate complex regulatory requirements alone.

8.6.4 A note on uncertainty

The policy environment is moving faster than the scholarly literature can track. CBAM's scope may or may not extend to furniture. The methodology may rely on default values or require actual data. The carbon price may rise or fall. The EU's political commitment may strengthen or weaken, and the Chinese government's response may be cooperative or defensive. The industry's trajectory depends on all of these, and none is settled.

Firms that have built measurement and abatement capability will do best under any of these outcomes, because capability is robust to policy uncertainty in a way that compliance is not. A firm that has measured its footprint can respond to any methodology. A firm that has reduced its emissions benefits from any carbon price. A firm that has designed for sustainability can sell into any market that values it.

8.7 Conclusion: The Container Yard Revisited

Liang's spreadsheet has grown since 2022. It now tracks the carbon content of each order, the carbon content of each supplier, the energy consumption of each production line, and the fuel efficiency of each trucking company. Some of the data is estimated and some is outdated, but it is enough to show him where his emissions come from and where he can cut them. He has installed solar panels on the factory roof, switched to water-based adhesives, and begun negotiating with his foam supplier for a lower-carbon formulation. None of these changes was required by law. His customers required them, and so did his own reading of where the market was heading.

The carbon constraint reached Liang through a customer relationship rather than through a regulation covering his product, and he responded by building measurement, then abatement, then supplier collaboration. Those capabilities now sit inside the business and will outlast the specific regulation that prompted them. Firms that follow his path will climb. Firms that wait for the regulation to require what his customers already expect will find themselves in the compliance trap, filling out forms while their competitors design products.

Furniture is not the most glamorous case in this book. It is labor-intensive, fragmented, and low-margin, and its firms lack the resources of the steel mills and technology companies examined elsewhere. That is why it matters. If green governance capability can be built in a Shunde sofa factory, it can be built in most places. The capability depends on attention, relationships, and the willingness to treat compliance as a floor rather than a ceiling.

Chapter 9 turns from the firm to the institutional environment that enables or constrains capability building. The furniture case shows that firms can build capability despite weak ecosystem support, and it also shows that diffusing capability across an industry depends on collective institutions: associations, local governments, shared service providers, and the standards they set. Chapter 9 examines these institutions in comparative perspective, asking why some clusters and regions have built effective green governance systems while others have not, and what the answer implies for policy.