When Do Carbon Markets Reduce Inequality? Article 6 Transfers Under Alternative Futures

When Do Carbon Markets Reduce Inequality? Article 6 Transfers Under Alternative Futures | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article When Do Carbon Markets Reduce Inequality? Article 6 Transfers Under Alternative Futures Mel George, James Edmonds This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8793231/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract We assess whether Article 6 carbon market transfers reduce between-region income inequality. Using GCAM, we model internationally transferred mitigation outcomes (ITMOs) across 32 regions under three socioeconomic baselines and a more ambitious net-zero 2050 pathway. A maximalist implementation of Article 6 reduces inequality under most scenarios, with Gini reductions of 0.5 to 0.9 points by 2050 under SSP1, SSP2, and net-zero pathways. Under SSP4, where income divergence erodes developing regions' comparative advantage in low-cost mitigation, the progressive effect weakens to near-neutral. The financial transfer effect dominates the policy burden effect across scenarios, operating primarily through the between-region channel. Africa's position as net seller or buyer serves as a diagnostic: swinging between favorable and unfavorable development pathways. These findings suggest Article 6 can reduce global inequality, but outcomes depend on development trajectories rather than the market mechanism itself. Ensuring broadly progressive outcomes requires complementary policies that preserve developing regions' comparative advantage, including technology transfer and access to clean energy finance. Climate Analysis and Modeling International Relations Development Economics Article 6 internationally transferred mitigation outcomes carbon markets NDCs integrated assessment modeling inequality Gini coefficient Theil decomposition Figures Figure 1 Figure 2 Figure 3 1. Introduction 1.1. Motivation Article 6 of the Paris Agreement establishes pathways for voluntary international cooperation in mitigation. Under cooperative approaches (Article 6.2), Parties may authorize and transfer internationally transferred mitigation outcomes (ITMOs) toward achievement of their nationally determined contributions (NDCs), subject to accounting requirements intended to preserve environmental integrity (UNFCCC, 2021 ). In principle, an ITMO market can lower the cost of meeting a given set of national targets by reallocating abatement to where it is cheaper. Modeling studies using integrated assessment frameworks have estimated that cooperative implementation could reduce NDC compliance costs by more than half and, if savings were reinvested, could nearly double global mitigation relative to independent implementation (Edmonds et al., 2021 ; Aldy et al., 2021 ). An ITMO market also generates cross-border financial flows: buyers pay sellers for mitigation outcomes. These flows are increasingly discussed as a potential channel of climate finance, with some analysts positioning carbon markets as vehicles for mobilizing investment in developing countries (Keane and Fernandez, 2025 ; OECD, 2023 ). The UNFCCC itself notes that Article 6 can “unlock financial support for developing countries” (UNFCCC, 2024 ). Yet most quantitative assessments of Article 6 foreground efficiency gains, trading volumes, and accounting design rather than distributional incidence across regions. The question of whether ITMO transfers reduce or exacerbate between-region inequality has not been directly examined in the modeling literature. At the same time, integrated assessment modeling has begun to engage more explicitly with inequality and distributional outcomes. Emmerling et al. ( 2024 ) coordinate an ensemble of eight IAMs to assess how climate impacts and mitigation pathways affect global inequality, finding that unmitigated climate change raises the global Gini coefficient by approximately 1.4 points by 2100, while Paris-aligned mitigation with equal-per-capita revenue recycling can reduce inequality below baseline levels. Gazzotti et al. ( 2021 ) demonstrate that even under economically optimal climate policies with full international cooperation, significant between-region inequality persists, with the ratio of top-to-bottom income deciles remaining 117% higher than in a no-climate-change counterfactual. These findings underscore that efficiency and equity are distinct: aggregate cost savings do not automatically translate to reduced inequality. The theoretical case for international cooperation in climate mitigation rests on the potential for mutual gains from trade in abatement. Nordhaus ( 2015 ) formalizes this through the concept of "climate clubs," showing that coalitions with internal carbon pricing and border adjustments can overcome free-riding that undermines purely voluntary agreements. Article 6 represents a partial implementation of this logic: it enables cost-reducing trades but relies on voluntary participation rather than club-based enforcement. This paper asks a focused question: when a stylized ITMO market is embedded within an integrated assessment framework under NDC-consistent policy constraints, do the resulting financial transfers reduce between-region income inequality? We answer using a population-weighted global Gini coefficient computed across model regions, and we decompose the change to identify which regions drive the result and whether the effect operates primarily through domestic mitigation cost reallocation or through international payment flows. 1.2. Contributions We make four contributions. First, we provide transparent accounting of ITMO trading under NDC-consistent constraints across three socioeconomic baselines (SSP1, SSP2, SSP4) and a net-zero 2050 extension. This design isolates how baseline heterogeneity conditions distributional outcomes, rather than treating Article 6 in a single reference world. Baseline inequality varies substantially across SSPs: SSP1 features relatively high income convergence, SSP2 a middle-of-the-road trajectory, and SSP4 a world of deepening inequality with greater income dispersion across regions (Riahi et al., 2017 ; Rao et al., 2019 ). By running Article 6 experiments across these baselines, we can assess whether the market’s distributional effects depend on initial conditions. Second, we track net financial transfers as an explicit accounting object, making cross-border payment flows visible for distributional evaluation. Third, we move beyond reporting an aggregate inequality statistic by decomposing the change in the population-weighted global Gini into components: income effects from domestic mitigation costs (policy burden) and income effects from international ITMO transfers. This decomposition identifies which regions contribute most to the inequality change and clarifies whether the aggregate result is broad-based or driven by a small subset of large players. Fourth, we employ Theil index decomposition to formally partition inequality changes into between-region and within-region components. This approach, which exploits the Theil index’s exact additive decomposability, allows us to verify whether Article 6 transfers operate primarily through cross-border redistribution or also affect within-region inequality through our distributional assumptions. This methodological contribution complements the Gini analysis and provides a robustness check on our central findings. 1.3. Preview of findings Our headline finding is that Article 6 transfers reduce between-region inequality under most scenarios, with the magnitude contingent on development conditions. Under SSP1 and SSP2, where incomes converge across regions over time, Article 6 produces progressive outcomes: lower-income regions have cheaper mitigation options, become net sellers of ITMOs, and receive financial transfers from higher-income buyers. The Gini coefficient falls by 0.5 to 0.9 points by 2050 relative to baselines. Under net-zero 2050 ambition, transfer volumes and carbon prices rise substantially, strengthening the inequality reduction to approximately 0.8 points. Under SSP4, the progressive effect weakens substantially but does not reverse into a strongly regressive outcome. In this divergent development pathway, lower-income regions lack capital for clean technology deployment, and their mitigation costs rise relative to higher-income regions. By mid-century, Africa shifts from net seller to net buyer of ITMOs. However, large sellers like China and India continue to receive substantial transfers, partially offsetting Africa’s reversal. The transfer effect remains slightly progressive (-0.2 Gini points), but combined with a regressive policy burden (+ 0.3 points), the net effect is near-neutral (+ 0.03 points). Africa’s position serves as a diagnostic indicator of which lower-income regions gain or lose: Africa pays $ 554 billion under SSP4 but receives $ 577 billion under SSP1 by 2050, a swing exceeding $ 1.1 trillion. The decomposition reveals that the transfer channel dominates the policy burden by a factor of three to five, operating primarily through the between-region channel under convergent scenarios. We discuss these findings in light of critiques that carbon markets may not deliver equitable outcomes, and we identify the mechanism determining the magnitude of Article 6’s progressive effect: comparative advantage in low-cost mitigation, which depends on development trajectories. 2. Related Literature Two literatures inform this work: one on inequality in integrated assessment modeling, the other on Article 6 economics and implementation. We bridge them by treating ITMO cooperation as an explicit system of cross-border financial transfers and quantifying its distributional incidence. 2.1. Inequality in integrated assessment modeling A growing body of work has moved distributional questions from the periphery of scenario analysis toward a central object of study. Emmerling et al. ( 2024 ) provide a prominent example, coordinating an ensemble of eight large-scale IAMs that incorporate income heterogeneity to evaluate how climate risks and Paris-aligned policies interact with inequality. They find that unmitigated climate change increases the global Gini coefficient by approximately 1.4 points by 2100, and that stringent mitigation (1.5°C pathways) alleviates much of this exacerbation in the long run but can slightly increase inequality in the short term without complementary redistribution policies. They also show that if carbon pricing revenues are recycled on an equal-per-capita basis, global inequality can fall below baseline levels, demonstrating that policy design matters for distributional outcomes. Gazzotti et al. ( 2021 ) use a cost-benefit IAM with 50 regions to examine inequality under economically optimal climate policies. Even with full international cooperation achieving temperature stabilization around 1.8°C, they find that the ratio of top-to-bottom income deciles remains 117% higher than in a counterfactual without climate change. Their results underscore that optimal policy guided by efficiency criteria does not inherently resolve between-region disparities. Climate damages fall disproportionately on lower-income regions, and cost-minimizing abatement allocation does not compensate for this pattern. Related work shows that carbon pricing with equal-per-capita revenue redistribution can substantially reduce inequality (Budolfson et al., 2021 ; Lang et al., 2025 ), establishing that climate policy design affects distributional outcomes. Our paper complements this literature by examining a distinct channel: cross-border financial transfers generated by international cooperation through carbon markets, rather than domestic revenue recycling or explicit transfer schemes. The Article 6 mechanism differs from equal-per-capita redistribution in that payment flows depend on market outcomes, namely which regions have lower marginal abatement costs and thus become sellers. Whether this market-based channel produces distributional effects comparable to designed redistribution remains an open question. 2.2. Modeling Article 6 cooperation A parallel literature has developed around the economics and implementation of Article 6 cooperation. Early analyses emphasize that cooperative approaches can substantially lower the costs of achieving mitigation targets by reallocating abatement to lower-cost opportunities. Edmonds et al. ( 2019 ) assess the economic potential of Article 6 using GCAM, finding that global trading could reduce NDC implementation costs by more than half (approximately $ 250 billion per year in 2030). Aldy et al. ( 2021 ), in a peer-reviewed extension published in Climate Change Economics, show that if cost savings were reinvested in enhanced ambition, cooperative implementation could yield an additional 9 GtCO2/year of mitigation, with more than half of near-term gains coming from land-use measures. These modeling exercises build on two decades of experience with international carbon market mechanisms. Michaelowa ( 2019 ) traces the evolution from the Clean Development Mechanism through Joint Implementation to Article 6, identifying four phases ranging from "exuberance" (2005–2011) to "hibernation" (2012–2014) to the current "reconfiguration" under the Paris Agreement. This history informs expectations about how Article 6 markets may develop and what distributional patterns might emerge. Piris-Cabezas et al. ( 2023 ) use a partial-equilibrium carbon market model to estimate that Article 6 trading could enable nearly double the climate ambition through 2035 at no additional cost. Our previous work summarizes findings across NDC and net-zero pathways, showing that cooperative implementation substantially reduces total compliance costs and that financial flows between regions could exceed $ 1 trillion per year by 2050 under ambitious scenarios (IETA, 2023 ). These studies also explore market design variations, including limits on ITMO usage by buyers and transaction fees. They find that usage limits raise costs without climate benefit by forcing buyers into more expensive domestic abatement while reducing sellers’ opportunities to earn revenue (IETA, 2023 ). Transaction fees similarly shift burdens toward sellers and shrink trading volumes. Environmental integrity concerns have shaped Article 6 rule development. Schneider and La Hoz Theuer ( 2019 ) provide a framework for assessing integrity across multiple dimensions: accurate baselines, demonstrated additionality, robust monitoring, and avoidance of double counting through corresponding adjustments. These safeguards impose transaction costs but are necessary conditions for ITMO transfers to represent real emission reductions rather than accounting artifacts. The operational framework for Article 6.2 is codified in Decision 2/CMA.3, adopted at CMA3 in Glasgow in November 2021, which establishes requirements for ITMO authorization, corresponding adjustments to national inventories, and reporting cycles (UNFCCC, 2021 ). Subsequent guidance at COP27 and COP29 has clarified implementation details. The UNFCCC Secretariat’s Article 6.2 Reference Manual provides comprehensive technical guidance on accounting procedures (UNFCCC, 2025a ). Schneider et al. ( 2018 ) examine how emissions trading system linkages would operate under Article 6.2 accounting rules, addressing challenges for countries with single-year NDC targets. Distributional modeling in climate policy typically examines carbon pricing with domestic revenue recycling or explicit climate finance transfers. Our contribution is to treat ITMO payments as cross-border transfers operating within an NDC framework and to measure the between-region inequality implication using a population-weighted Gini coefficient. We then decompose the Gini change to identify regional contributions. Chepeliev et al. ( 2021 ), who link CGE modeling with microsimulation to assess distributional impacts of carbon pricing coalitions, represent the closest methodological analog; they find progressive global effects and reduced poverty burdens under cooperation, but their focus is pricing coordination rather than Article 6 mechanisms specifically. Existing Article 6 modeling thus tracks costs, volumes, and abatement but not distributional incidence. The governance literature examines market fragmentation and integrity (Ahonen et al., 2022 ; Michaelowa, 2019 ) but does not quantify inequality effects. We find no direct quantification of between-region inequality outcomes from ITMO trading in the peer-reviewed IAM literature, despite extensive modeling of Article 6 cost savings and volumes (Edmonds et al., 2019 , 2021 ; IETA, 2023 ). This is the gap our paper addresses. Our NDC representation uses the post-Belem generation of submitted NDCs as recorded in the UNFCCC registry through 30 September 2025, the cut-off date for the UNFCCC’s 2025 NDC Synthesis Report (UNFCCC, 2025b ). That synthesis covers 64 new or updated NDCs and notes that 89% of Parties indicated intention to participate in Article 6 cooperation, up from 64% in previous submissions. We treat this as a realism improvement for NDC-world modeling while recognizing that our central focus remains the distributional incidence of ITMO transfers conditional on heterogeneous development pathways. 3. Methods 3.1. Model and regional structure We use GCAM-8s, a variant of the Global Change Assessment Model version 8.2 developed for the ScenarioMIP exercise. GCAM is a market-equilibrium model that integrates representations of energy, water, land, and economic systems across 32 geopolitical regions (Calvin et al., 2019 ). The model solves for equilibrium prices and quantities in energy and agricultural markets subject to resource constraints, technology availability, and policy interventions. It operates on five-year time steps from 2015 to 2100 and uses a recursive-dynamic solution approach without perfect foresight. Population and GDP trajectories follow the Shared Socioeconomic Pathways. We implement SSP1, SSP2, and SSP4 to capture variation in baseline income dispersion. SSP1 represents a sustainability-oriented world with relatively low population growth and high income convergence across regions. SSP2 represents a middle-of-the-road trajectory. SSP4 represents a world of deepening inequality, with greater income dispersion and slower convergence (Riahi et al., 2017 ). GCAM served as the marker model for SSP4 in the original SSP quantification (Calvin et al., 2017 ). Baseline Gini projections for the SSPs are documented in Rao et al. ( 2019 ). For the inequality analysis, the relevant accounting objects are regional GDP per capita and population. GCAM tracks both, enabling computation of a population-weighted global Gini coefficient across the 32 regions. This yields a between-region inequality measure; it does not capture within-region inequality. We interpret changes in this metric as shifts in cross-region income dispersion induced by policy, recognizing that the 32-region aggregation smooths over within-region heterogeneity. GCAM’s 32 regions aggregate multiple countries; our “between-region” inequality measure should not be interpreted as country-level inequality. 3.2. Stylized ITMO market We embed a stylized representation of Article 6 cooperation in GCAM by allowing regions to trade emission permits under a global carbon market that clears at a single price. In the cooperative scenario, each region faces its NDC constraint but can meet part of that constraint by purchasing ITMOs from regions that over-comply and sell. The market clears when global supply of ITMOs equals global demand, yielding an equilibrium carbon price that equates marginal abatement costs across participating regions. We define "transfers" as payments equal to the equilibrium carbon price multiplied by net ITMO volume: sellers receive revenue, buyers incur expenditure. These are compliance payments for mitigation outcomes, not aid or grants. We treat them as finance-like because they constitute cross-border flows that relax budget constraints for sellers and tighten them for buyers, affecting regional mean income. This framing does not imply that transfers generate development co-benefits, reach intended beneficiaries within regions, or map directly to welfare gains. Those outcomes depend on domestic revenue use, which GCAM does not model. Net ITMO positions are determined endogenously by the model. Regions with marginal abatement costs below the global clearing price become net sellers; those with costs above become net buyers (IETA, 2023 ). The financial transfer associated with trading is computed as the product of the equilibrium carbon price and net ITMO volume: a net seller receives revenue equal to the price multiplied by ITMOs sold, while a net buyer incurs expenditure equal to the price multiplied by ITMOs purchased. In the independent implementation counterfactual, each region meets its NDC independently through domestic mitigation only. The regional carbon price in this case is the shadow price of the domestic constraint, varying across regions according to NDC stringency and abatement cost structure. Comparing outcomes with and without trading isolates the effect of Article 6 cooperation on regional incomes and inequality. This implementation follows precedents in GCAM-based Article 6 modeling (Edmonds et al., 2019 ; Aldy et al., 2021 ; IETA, 2023 ). We assume a perfectly competitive global market with full fungibility of ITMOs and do not impose bilateral restrictions, usage limits, or transaction fees in the core scenarios. These simplifications maximize the scope for observing distributional effects; sensitivity to market design features is a subject for future work. 3.3. Scenario design We construct scenarios along two dimensions: socioeconomic baseline (SSP1, SSP2, SSP4) and policy ambition (NDC, net-zero 2050). The NDC case implements national targets as recorded in the UNFCCC registry through 30 September 2025, following the scope of the 2025 NDC Synthesis Report (UNFCCC, 2025b ). NDCs are translated into regional emission constraints in GCAM using standard harmonization procedures consistent with IPCC AR6 scenario protocols (IPCC, 2022 ). Absolute targets are implemented directly; intensity and BAU-relative targets are converted to absolute terms using model-consistent GDP and baseline emissions projections. The net-zero 2050 case implements a trajectory in which all regions reach net-zero CO2 emissions by 2050, representing a high-ambition extension. This case generates higher carbon prices and larger transfer volumes, allowing examination of how policy stringency affects distributional outcomes. For each baseline-ambition combination, we run two variants: one with Article 6 trading (cooperative implementation) and one without (independent implementation). This yields eight core scenario pairs. We also test sensitivity to land-use sector scope in the SSP2 baseline, examining whether including land-use change emissions in the carbon market alters the distributional findings. Table 1 summarizes the scenario matrix. Table 1 Scenario matrix and market characteristics. Baseline Policy Ambition LUC Pricing Article 6 Scenario Label SSP1 NDC No No SSP1 NDC independent NDC No Yes SSP1 NDC cooperative SSP2 NDC No No SSP2 NDC independent NDC No Yes SSP2 NDC cooperative NDC Yes No SSP2 NDC LUC independent NDC Yes Yes SSP2 NDC LUC cooperative Net-Zero 2050 No No NZ2050 independent Net-Zero 2050 No Yes NZ2050 cooperative SSP4 NDC No No SSP4 NDC independent NDC No Yes SSP4 NDC cooperative Note: SSP1 represents convergent development with high income convergence across regions. SSP2 represents middle-of-the-road development. SSP4 represents divergent development with persistent inequality. NDC scenarios implement nationally determined contributions as recorded in the UNFCCC registry through 30 September 2025. Net-Zero 2050 implements a trajectory where all regions reach net-zero CO2 emissions by 2050. LUC Pricing indicates whether land-use change emissions are included in carbon pricing. Market size is total annual financial flow from buyers to sellers (billion 2025 USD). Carbon price is the equilibrium clearing price (USD per tonne CO2). 3.4. Inequality metrics We measure inequality using two complementary approaches: population-weighted Gini coefficients and Theil index decomposition. Between-region Gini coefficient. The Gini coefficient ranges from 0 (perfect equality) to 1 (maximum inequality). For a set of regions indexed by r with GDP per capita y_r and population weight w_r, the population-weighted Gini is computed using the standard formula based on mean absolute differences in incomes (Lakner and Milanovic, 2016 ). We focus on 2030 and 2050 as key policy horizons. Regional GDP per capita is adjusted for policy-induced income effects. In the independent implementation scenario, adjusted income equals baseline GDP per capita minus domestic mitigation cost (computed as the change in regional GDP from baseline to policy scenario in GCAM). In the cooperative scenario, adjusted income additionally reflects net ITMO transfers: sellers gain revenue, buyers incur expenditure. All values are expressed in per capita terms using regional population. This metric captures between-region inequality only. It does not account for within-region income distributions, which GCAM does not represent at the household level. We interpret changes in the global Gini as indicative of shifts in cross-region income dispersion, recognizing that interpersonal global inequality would require additional information on within-region distributions. The Gini coefficient is widely used in this literature (Emmerling et al., 2024 ; Lakner and Milanovic, 2016 ) despite known limitations, including non-decomposability into additive between- and within-group components (Bourguignon, 1979 ). We report results using the Gini for comparability with existing work. Theil index decomposition. To formally partition inequality changes into between-region and within-region components, we employ the Theil T index (also known as GE(1)) is exactly additively decomposable (Shorrocks, 1980 ), unlike the Gini coefficient (Bourguignon, 1979 ). The total Theil index T can be written as: T = T_between + T_within The between-region component treats each region as a unit with its mean income. The within-region component is a weighted sum of regional Theil indices, where weights reflect each region’s share of total income. For this decomposition, we construct 320 observations (32 regions times 10 income deciles) using within-region income distributions from SSP-specific projections (Rao et al., 2019 ). This decomposition serves two purposes. First, it provides a robustness check on our Gini-based findings by using an alternative inequality measure. Second, and more importantly, it allows us to verify that Article 6 transfers operate primarily through the between-region channel. Because GCAM does not model how ITMO revenues or payments are distributed within regions, we make an equal-per-capita (EPC) assumption: transfers are distributed uniformly across all individuals within each region. Under this assumption, transfers have a small mechanical effect on within-region inequality (equal absolute transfers are progressive in relative terms), but we expect the dominant effect to operate through changes in regional mean incomes. The Theil decomposition tests this expectation directly. 3.5. Decomposition of income effects To understand what drives the change in inequality between the independent and cooperative scenarios, we track Gini changes across three sequential stages. Stage 1: Baseline inequality - The Gini coefficient under each SSP baseline without climate policy. Stage 2: Policy burden effect - The change from baseline to policy scenario, capturing how mitigation costs are distributed before trading. Stage 3: Transfer effect - The change from policy to cooperative scenario, isolating the impact of ITMO financial flows. If money flows from rich to poor (poor regions sell), this is progressive; the reverse is regressive. Full procedural details appear in SI Section S4. 3.6. Distributional assumptions and limitations GCAM does not endogenously model within-region income distribution changes from climate policy. We therefore make the following assumptions about how policy effects are distributed across deciles within each region. We assume both mitigation costs and ITMO payment flows are distributed equally per capita across all deciles within each region, equivalent to lump-sum domestic revenue recycling. Second, ITMO payment flows are distributed equally per capita within receiving and paying regions. This equal-per-capita (EPC) assumption is equivalent to assuming lump-sum domestic revenue recycling. Under EPC, equal absolute income changes have progressive relative effects (larger percentage change for poorer deciles within each region). Regions receiving transfers see slight within-region inequality reduction; regions paying for credits see slight within-region inequality increase. The EPC assumption represents a specific policy choice about domestic revenue use. Alternative recycling mechanisms (e.g., proportional to income, targeted transfers to poor households) would yield different within-region effects not captured in this analysis. This is a known limitation: Article 6 determines the magnitude and direction of between-region flows, but domestic fiscal policy determines how those flows affect within-region distribution. Our analysis isolates the between-region channel, which Theil decomposition confirms that 92 to 100 percent of the transfer effect operates through the between-region channel under SSP1, SSP2, and net-zero scenarios (Table S5). Under SSP4, this share falls to 79 percent by 2050. Additional limitations warrant acknowledgment. First, GCAM’s 32-region structure aggregates substantial heterogeneity; results for model regions should not be interpreted as country-specific predictions. Second, we model a stylized global market with perfect competition and full participation; actual Article 6 implementation involves bilateral arrangements, varying levels of readiness, and transaction costs that could alter outcomes. Third, our income measure is GDP per capita adjusted for mitigation costs and transfers; this is a proxy for economic resources but does not capture welfare, consumption, or non-monetary dimensions of inequality. Fourth, we do not model climate damages, so our inequality trajectories reflect mitigation policy effects only. 4. Results 4.1. Market characteristics Before examining distributional outcomes, we characterize the Article 6 markets that emerge under each scenario. Market size, measured as total financial flows from buyers to sellers, varies substantially across socioeconomic baselines and policy ambition levels. Table 1 reports market sizes and equilibrium carbon prices for key years. Under SSP2 with NDC-level ambition, the ITMO market grows from $ 380 billion in 2030 to $ 953 billion in 2040 and $ 1,672 billion in 2050. SSP1 produces smaller markets ( $ 261 billion to $ 706 billion over the same period) because income convergence reduces the dispersion in regional mitigation costs that drives trading gains. SSP4 generates the largest markets ( $ 427 billion to $ 1,744 billion) as income divergence widens cost differentials between regions. Under net-zero 2050 ambition, market size reaches approximately $ 1,770 billion by 2050, comparable to SSP4 NDC levels but with different regional compositions. Equilibrium carbon prices range from $ 30–155/tCO2 (SSP1) to $ 45–207/tCO2 (SSP4) over 2030–2050, with net-zero 2050 reaching $ 285/tCO2 by 2050. Higher prices under SSP4 reflect concentrated mitigation capacity; higher prices under net-zero reflect policy stringency. These market magnitudes are substantial in development finance terms. Annual ITMO flows under SSP2 NDC cooperative represent 0.3 to 1.5 percent of global GDP across the projection period, comparable to current official development assistance flows. For recipient regions, transfers can reach 10 to 20 percent of regional GDP in scenarios where those regions are net sellers. 4.2. Gini trajectories across scenarios Figure 1 presents Gini trajectories under baseline and cooperative scenarios. The vertical distance between lines represents Article 6's net effect. In the convergent development scenario (SSP1), Article 6 produces consistent inequality reduction throughout the projection period. The Gini coefficient falls by 0.32 points relative to baseline in 2030, by 0.44 points in 2040, and by 0.90 points in 2050. This progressive pattern reflects the comparative advantage mechanism: in a convergent development world, lower-income regions have cheaper mitigation options, become net sellers of ITMOs, and receive financial transfers from higher-income buyers. The middle-of-the-road scenario shows a similar progressive pattern. The Gini falls by 0.38 points in 2030, by 0.95 points in 2040, and by 0.48 points in 2050. The pattern remains consistently progressive, with transfers flowing from rich to poor regions throughout the period. The net-zero 2050 pathway produces larger inequality reductions than NDC-level ambition despite imposing greater mitigation costs. The Gini falls by approximately 0.06 points in 2030, 0.24 points in 2040, and 0.76 points in 2050. Higher climate ambition strengthens the progressive effect of Article 6. The mechanism is that transfer volumes scale faster than mitigation burdens as ambition increases. Under net-zero targets, the global carbon price rises substantially, increasing the value of each ITMO traded. Because low-income regions retain comparative advantage in low-cost mitigation under SSP2 baseline conditions, they sell more ITMOs at higher prices, receiving larger transfers. Table 2 summarizes Gini levels and changes across scenarios, decomposing the net effect into policy burden and transfer components (discussed in Section 4.4 ). Table 2 Between-region Gini coefficients by scenario. Scenario 2030 2050 SSP1 Baseline 57.03 50.42 SSP1 NDC cooperative 56.71 49.52 SSP1 ΔGini -0.32 -0.90 SSP2 Baseline 57.44 52.51 SSP2 NDC cooperative 57.06 52.03 SSP2 ΔGini -0.38 -0.48 SSP2 NDC LUC cooperative 57.09 51.96 SSP2 LUC ΔGini -0.35 -0.54 SSP4 Baseline 57.59 56.33 SSP4 NDC cooperative 57.40 56.36 SSP4 ΔGini -0.19 + 0.03 NZ2050 Baseline 57.44 52.51 NZ2050 cooperative 57.38 51.75 NZ2050 ΔGini -0.06 -0.76 4.3. The SSP4 pattern SSP4 presents a notably different outcome from the other scenarios. Under this divergent development pathway, Article 6 produces weaker progressive effects that approach neutrality by mid-century. The Gini falls by 0.19 points in 2030 and 0.77 points in 2040, but the net effect is only + 0.03 points by 2050, essentially neutral. This attenuation of the progressive effect reflects erosion of comparative advantage for some lower-income regions. In SSP4, income divergence means that regions like Africa lack capital for clean technology deployment. Their mitigation costs rise relative to higher-income regions, which continue to invest in low-carbon infrastructure. By mid-century, Africa shifts from net seller to net buyer: the continent faces binding NDC constraints but lacks cheap domestic mitigation options, forcing it to purchase ITMOs. However, the overall outcome does not become strongly regressive because other large developing regions remain net sellers. China and India, despite being lower-income than developed regions, retain comparative advantage in low-cost mitigation and continue to receive substantial transfers. These flows partially offset Africa’s reversal, leaving the aggregate between-region effect near-neutral. Africa’s position serves as a diagnostic indicator of which lower-income regions benefit from Article 6. Under SSP1 in 2050, Africa (aggregated across the four African regions in GCAM) receives $ 577 billion in net transfers as a seller of ITMOs. Under SSP2, Africa receives $ 103 billion. Under SSP4, Africa pays $ 554 billion as a net buyer. This swing of over $ 1.1 trillion between SSP1 and SSP4 represents a fundamental reversal in Africa’s role in the carbon market. When Africa sells credits, Article 6 is strongly progressive; when Africa buys credits, the progressive effect is substantially attenuated. The near-neutral aggregate outcome (+ 0.03 Gini points) reflects offsetting regional dynamics: Africa's shift to buyer status is counterbalanced by continued large seller positions in China and India, which retain comparative advantage even under divergent development. The "poor pays rich" reversal is thus partial, not universal. Table S3 reports the decomposition: burden + 0.26, transfer − 0.23, net + 0.03 in SSP4 2050. The SSP4 pattern is not an artifact of modeling assumptions but reflects a structural feature of how comparative advantage operates in carbon markets under divergent development. If development trajectories concentrate mitigation capacity in already-wealthy regions while leaving some lower-income regions behind, the market will produce differentiated outcomes across the Global South. This finding has direct implications for Article 6 design, which we discuss in Section 5 . 4.4. Decomposition: transfers dominate To understand the mechanisms driving inequality changes, we decompose the net effect into two components: the policy burden effect (change in inequality from mitigation costs before any trading) and the transfer effect (change from ITMO financial flows). Table 3 reports this decomposition across scenarios. The transfer effect dominates the policy burden effect by a substantial margin across all scenarios. Under SSP2 NDC cooperative, the policy burden effect is + 0.22 Gini points by 2050 (slightly regressive, as mitigation costs are not perfectly proportional to income), while the transfer effect is -0.70 Gini points (strongly progressive). The transfer effect is 3.2 times larger in magnitude than the burden effect, and its progressive direction more than offsets the regressive burden to produce a net progressive outcome of -0.48 points. Under net-zero 2050, both effects are larger, but the transfer effect grows faster. The policy burden rises to + 0.30 Gini points, reflecting higher mitigation costs under stringent targets. The transfer effect reaches − 1.06 Gini points, yielding a ratio of 3.5 to 1. This pattern explains why higher ambition produces more progressive outcomes: the transfer channel scales with carbon prices and trading volumes, while the burden channel scales with mitigation effort. Because low-income regions sell more under high-ambition scenarios (they have the cheapest remaining mitigation options as global abatement deepens), the progressive transfer effect outpaces the regressive burden effect. Under SSP1, the pattern is even more pronounced. The policy burden is + 0.18 Gini points by 2050 (mitigation costs are distributed somewhat regressively even in a convergent world), while the transfer effect is -1.08 points, a ratio exceeding 5 to 1. The net effect of -0.90 points is the most progressive across scenarios. Under SSP4 in 2050, the transfer effect remains slightly progressive (-0.23 points) even though Africa has become a buyer, because China and India continue as large sellers. The policy burden is + 0.26 points. Combined, the net effect is + 0.03 points, essentially neutral. This contrasts with the other scenarios where the transfer effect decisively offsets the burden. Table 3 Sequential decomposition of inequality effects by scenario. Scenario Year Burden Transfer Net Ratio SSP1 2030 + 0.03 -0.35 -0.32 11.7 2050 + 0.18 -1.08 -0.90 6.0 SSP2 2030 + 0.03 -0.41 -0.38 13.7 2050 + 0.22 -0.70 -0.48 3.2 SSP2 LUC 2030 + 0.03 -0.38 -0.35 12.7 2050 + 0.25 -0.79 -0.54 3.2 SSP4 2030 + 0.06 -0.25 -0.19 4.2 2050 + 0.26 -0.23 + 0.03 0.9 NZ2050 2030 + 0.05 -0.11 -0.06 2.2 2050 + 0.30 -1.06 -0.76 3.5 Note: Burden = Gini(NDC) - Gini(Baseline); Transfer = Gini(cooperative) - Gini(NDC); Net = Gini(cooperative) - Gini(Baseline). All values in Gini points. Ratio is absolute value of transfer effect divided by burden effect. Negative values are progressive; positive values are regressive. 4.5. Regional dynamics Figure 2 displays net financial flows by region group across scenarios, illustrating the trading patterns that underlie the aggregate Gini results. We aggregate GCAM’s 32 regions into six groups for exposition: Africa (four regions), China, India, Southeast Asia and Indonesia, Latin America, and Developed regions (USA, EU-15, EU-12, Japan, Canada, Australia/NZ). Under SSP2 NDC cooperative in 2050, Africa receives $ 103 billion in net transfers, India receives $ 340 billion, and China receives $ 517 billion. Developed regions pay $ 72 billion (USA), with substantial heterogeneity: some developed regions with cheap mitigation options (notably Canada) are net sellers. The overall pattern is progressive: lower-income regions sell, higher-income regions buy, and financial flows move from rich to poor. Under higher policy ambition, the regional pattern shifts further, revealing how stringency interacts with comparative advantage. Under net-zero 2050, Africa’s receipts increase to $ 894 billion as higher carbon prices raise the value of its mitigation exports. However, China reverses from receiving $ 517 billion under NDC cooperative to paying $ 948 billion under net-zero, a swing of $ 1,465 billion. This is the largest single-region shift across scenarios. Under stringent net-zero targets, China’s massive emissions base requires importing credits despite its domestic mitigation efforts. India also flips from receiving $ 340 billion to paying $ 65 billion. Southeast Asia becomes a net payer despite being in the Global South. These shifts reveal heterogeneity obscured by aggregate statistics. Under net-zero, the cleavage is not simply rich versus poor but regions with large remaining emissions versus those with cheap land-use mitigation. Brazil shifts from paying $ 320 billion (NDC) to receiving $ 109 billion (net-zero); Canada similarly reverses. This realignment complicates simple North-South narratives about carbon market distributive impacts. 4.6. Theil decomposition To verify that Article 6 operates primarily through the between-region channel, we employ Theil index decomposition. The Theil T index is exactly additively decomposable into between-group and within-group components, unlike the Gini coefficient. We construct 320 observations (32 regions times 10 income deciles) using within-region income distributions from SSP-specific projections, then compute total, between-region, and within-region Theil indices under each scenario. The decomposition confirms that the transfer effect operates almost entirely through the between-region channel under convergent scenarios. Across SSP1, SSP2, and net-zero scenarios, the between-region component accounts for the dominant share of the total Theil change. The within-region component contributes a smaller offsetting effect because equal-per-capita transfers are slightly progressive within regions (giving the same absolute amount to each person benefits lower-income deciles proportionally more). Under SSP4 in 2050, the Theil decomposition reveals a different pattern. The between-region component shows a slight decrease in inequality (progressive), while the within-region component shows an increase (regressive). When poor regions pay for credits under the EPC assumption, the uniform outflow is regressive within those regions because equal absolute payments represent a larger share of income for poorer deciles. The within-region effect dominates in SSP4 by 2050 because the large buyer economies have high GDP weights in the within-region calculation. This nuance does not change the headline finding that SSP4 produces near-neutral between-region effects, but it reveals that the mechanism operates differently than in progressive scenarios. Under SSP1 and SSP2, transfers reduce between-region inequality and have small progressive within-region effects. Under SSP4, between-region effects are weak, and within-region effects in buyer regions become regressive. The overall implication is that SSP4 represents not just a weaker progressive outcome but a qualitatively different distributional pattern. Full Theil decomposition tables are provided in the Supplementary Information. 4.7. Sensitivity to land-use sector scope We test whether including land-use change (LUC) emissions in the carbon market alters distributional findings. Under SSP2 with LUC pricing, market size expands substantially: from $ 380 billion to $ 497 billion in 2030 (+ 31 percent), from $ 953 billion to $ 1,471 billion in 2040 (+ 55 percent), and from $ 1,672 billion to $ 1,952 billion in 2050 (+ 17 percent). Despite these large changes in market size, the distributional impact is essentially neutral. The Gini coefficient differs by less than 0.1 points between LUC and non-LUC scenarios. This neutrality reflects heterogeneity within income groups: some lower-income regions gain from LUC pricing while others lose, producing no systematic progressive or regressive tilt. Full LUC scenario results are provided in the Supplementary Information. 5. Discussion 5.1. Efficiency versus equity in carbon markets Our findings speak to a longstanding tension in international climate policy between efficiency and equity objectives. Article 6 is primarily framed as an efficiency mechanism: by allowing abatement to occur where it is cheapest, cooperative implementation reduces the global cost of achieving a given set of targets (Edmonds et al., 2019 ; Aldy et al., 2021 ). The question we have examined is whether efficiency gains coincide with equity gains, or whether they come at the expense of distributional objectives. Under convergent pathways and high-ambition scenarios, Article 6 produces both efficiency gains and progressive distributional outcomes, with transfer effects dominating policy burdens by a factor of three to six. Under divergent development (SSP4), the equity outcome weakens while efficiency gains persist. The market still reduces global mitigation costs by reallocating abatement, but the regional distribution of gains shifts. Some poor regions (notably Africa) lose comparative advantage as their mitigation costs rise relative to rich regions, becoming net buyers rather than sellers. However, because other large developing regions (China, India) retain seller status, the overall between-region effect remains near-neutral rather than becoming strongly regressive. This finding underscore that efficiency and equity are distinct but not necessarily in tension: aggregate cost savings can coincide with progressive outcomes under most development conditions. 5.2. Article 6 as climate finance Some analysts have positioned Article 6 as a channel for climate finance, noting that ITMO payments could mobilize substantial capital flows to developing countries (Keane and Fernandez, 2025 ; OECD, 2023 ). Our results provide partial support for this framing. Under progressive scenarios (SSP1, SSP2, net-zero 2050), transfer volumes reach $ 700 billion to $ 1.77 trillion annually by 2050, with flows generally moving from higher-income to lower-income regions. For recipient regions such as Africa, transfers can reach 10 to 20 percent of regional GDP, magnitudes that exceed typical development assistance. However, these are gross financial flows, not measures of net developmental benefit. Actual outcomes depend on authorization decisions, transaction costs, and domestic revenue allocation, none of which the model captures. Additional caveats apply. The UNCTAD Least Developed Countries Report (2024) cautions that carbon markets do not constitute or replace climate finance and do not adhere to the principle of common but differentiated responsibilities. Evidence from the Clean Development Mechanism suggests that LDCs gained limited capabilities from participation, with project activity concentrated in a small number of middle-income countries (UNCTAD, 2024 ). Our modeling cannot assess whether ITMO revenues translate into development co-benefits or domestic capacity building; we track only aggregate financial flows. More fundamentally, the SSP4 results demonstrate that Article 6 transfers are not guaranteed to flow toward all developing regions. If development diverges and some poor regions lose comparative advantage in mitigation, they become buyers rather than sellers, and Article 6 functions as a drain on their resources even while benefiting other developing regions. This conditionality distinguishes Article 6 from needs-based climate finance mechanisms, which allocate resources according to vulnerability or development status rather than market position. 5.3. The comparative advantage mechanism The SSP4 pattern illuminates the mechanism that determines the magnitude of Article 6’s progressive effect: comparative advantage in low-cost mitigation. This mechanism deserves explicit attention because it identifies the structural condition that policymakers would need to preserve or create if they want carbon markets to deliver broadly progressive outcomes. In SSP1 and SSP2, income convergence means that developing regions accumulate capital, deploy clean technologies, and develop institutional capacity for mitigation. Their marginal abatement costs remain low relative to developed regions, which face harder-to-abate residual emissions after exhausting cheap options. Developing regions thus have mitigation to sell, and developed regions have demand to buy. Financial transfers flow from rich to poor. In SSP4, income divergence means that some developing regions lack capital for clean technology deployment. Their energy systems remain carbon-intensive, and their mitigation costs rise as they face the dual burden of development needs and emissions constraints. Meanwhile, developed regions continue investing in low-carbon infrastructure, driving down their marginal costs. For regions like Africa, the comparative advantage flips: developed regions have cheaper mitigation options. These poor regions must buy credits to meet their NDCs because domestic abatement is too expensive. Africa’s position serves as the clearest diagnostic of this mechanism. When Africa has comparative advantage (SSP1, SSP2), it sells credits and receives transfers; Article 6 is strongly progressive. When Africa loses comparative advantage (SSP4), it buys credits and pays transfers; the progressive effect is substantially attenuated. The $ 1.1 trillion swing in Africa’s position between SSP1 and SSP4 in 2050 is a direct consequence of how development pathways affect mitigation cost structures. If policymakers want Article 6 to deliver broadly progressive outcomes, lower-income regions need access to capital, technology, and institutional support for clean energy deployment. If climate policy inadvertently concentrates mitigation capacity in wealthy regions while leaving others behind, the carbon market will produce differentiated outcomes across the Global South. 5.4. Comparison to domestic revenue recycling How does the distributional effect of Article 6 compare to other climate policy mechanisms? The most relevant comparison is to domestic carbon pricing with equal-per-capita revenue recycling, which Budolfson et al. ( 2021 ) and Emmerling et al. ( 2024 ) have shown can substantially reduce inequality. Budolfson et al. ( 2021 ) find that equal-per-capita redistribution of carbon tax revenues within each region can reduce the global Gini coefficient by approximately 2 points under 2C-consistent mitigation pathways. This effect operates primarily through the within-region channel: lump-sum transfers benefit poor households more than rich households in proportional terms, compressing the income distribution within each country. Our findings suggest that Article 6 achieves inequality reductions of 0.5 to 0.9 Gini points under progressive scenarios, roughly one-quarter to one-half the magnitude of domestic revenue recycling. Critically, Article 6 operates through a different channel: the between-region channel, as confirmed by our Theil decomposition showing the dominant share of the effect in the between-region component under convergent scenarios. Article 6 compresses the income distribution across regions by transferring resources from rich-region buyers to poor-region sellers, rather than compressing distributions within regions. Article 6 and domestic revenue recycling are complements, not substitutes. They address different dimensions of global inequality through different mechanisms: carbon pricing with progressive domestic recycling reduces within-region inequality, while Article 6 cooperation reduces between-region inequality. A comprehensive climate policy package could include both, with combined effects exceeding what either achieves alone. The governance challenges differ as well. Article 6 requires negotiated rules, accounting infrastructure, and credible commitment across sovereign parties; revenue recycling requires domestic political will to implement progressive transfers. 5.5. Design implications Our results have implications for Article 6 design and complementary policies. Market features that restrict trading volumes reduce both efficiency and distributional flows, creating tradeoffs between environmental integrity and equity co-benefits. The share-of-proceeds provision under Article 6.4, directing 5 percent to adaptation and mandating 2 percent cancellation, illustrates how equity considerations can be embedded in market design. Our analysis does not model these provisions explicitly, but they illustrate the scope for layering redistributive elements onto efficiency-oriented mechanisms. The governance architecture of Article 6 also has distributional implications. Ahonen et al. ( 2022 ) document the fragmentation between Article 6.2's bilateral approach and Article 6.4's centralized mechanism, noting tensions between bottom-up flexibility and top-down standardization. This fragmentation may affect which regions can effectively participate: countries lacking institutional capacity for bilateral negotiations may benefit from centralized mechanisms with standardized procedures, while countries with sophisticated climate diplomacy may prefer the flexibility of bilateral arrangements. More fundamentally, our findings suggest the most important determinant of Article 6's distributional outcome is not market design but the underlying development trajectory. Under SSP4, the weaker progressive outcome occurs because some poor regions lack comparative advantage. This points toward complementary policies outside the Article 6 framework: technology transfer, capacity building, concessional finance for clean energy deployment, and other measures that preserve or create mitigation capacity in developing regions. Article 6 can distribute the gains from trade, but it cannot create comparative advantage where none exists. Higher policy ambition strengthens progressive outcomes under favorable development conditions. The net-zero 2050 pathway produces larger inequality reductions than NDC-level ambition because transfer volumes scale faster than mitigation burdens. This finding cuts against concerns that stringent climate targets disproportionately burden developing regions. Under SSP2 conditions, the opposite is true: higher ambition generates larger markets, higher carbon prices, and larger transfers to developing-region sellers. Climate ambition and equity can be complements rather than tradeoffs, but only if developing regions retain comparative advantage in low-cost mitigation. 5.6. Limitations Several caveats must be noted with our analysis. Our inequality measure captures between-region dispersion only; within-region effects depend on domestic revenue allocation, which varies across countries and political contexts. GCAM's 32-region structure aggregates substantial heterogeneity, so results for regions like "Africa_Western" should not be interpreted as country-level predictions. We also model a stylized global market with perfect competition and full participation; actual Article 6 implementation involves bilateral arrangements, varying readiness levels, and transaction costs. We do not model climate damages, so our inequality trajectories reflect mitigation policy effects only. Finally, our analysis is positive rather than normative; we quantify effects but do not assess whether the magnitudes satisfy principles of climate justice. 6. Conclusion This paper quantifies how Article 6 carbon markets affect between-region income inequality. Our central finding is that Article 6 reduces inequality under most scenarios, with the magnitude contingent on development conditions. Under SSP1, SSP2, and net-zero 2050, Article 6 produces progressive outcomes. Lower-income regions sell ITMOs and receive transfers of $ 700 billion to $ 1.77 trillion annually by 2050. The population-weighted global Gini falls by 0.5 to 0.9 points. Higher ambition strengthens this effect: net-zero scenarios produce larger inequality reductions than NDC-level ambition because transfer volumes scale faster than mitigation burdens. Under SSP4, the progressive effect weakens to near-neutral (+ 0.03 Gini points by 2050). Income divergence erodes some developing regions' comparative advantage; Africa shifts from seller to buyer, paying $ 554 billion rather than receiving $ 577 billion as under SSP1. However, China and India retain seller status, partially offsetting Africa's reversal. The transfer effect remains slightly progressive but insufficient to offset the policy burden. Our decomposition reveals that financial transfers dominate policy burden effects by a factor of three to six. Theil decomposition confirms that 92 to 100 percent of the transfer effect operates through the between-region channel under convergent scenarios. Article 6 thus offers a distinct channel for reducing global inequality that complements domestic revenue recycling, which operates through the within-region channel. The progressive character of Article 6 depends on developing regions maintaining comparative advantage in low-cost mitigation. Policies that concentrate mitigation capacity in wealthy regions would undermine conditions for progressive outcomes. Complementary mechanisms, particularly technology transfer and adaptation finance, may be necessary to preserve conditions under which carbon markets contribute broadly to global equity. Whether Article 6 ultimately reduces or increases global inequality depends on choices about development pathways, technology access, and the institutional architecture of international climate cooperation. References Ahonen H-M, Kessler J, Michaelowa A, Espelage A, Hoch S (2022) Governance of fragmented compliance and voluntary carbon markets under the Paris Agreement. Politics Gov 10(1):235–246 Aldy JE, Pizer WA, Tavoni M, Reis LA, Akimoto K, Blanford G et al (2021) How much could Article 6 enhance nationally determined contribution ambition toward Paris Agreement goals through economic efficiency? Clim Change Econ 12(2):2150007 Bourguignon F (1979) Decomposable income inequality measures. Econometrica 47(4):901–920 Budolfson M, Dennig F, Fleurbaey M, Siebert A, Socolow RH (2021) Climate action with revenue recycling has benefits for poverty, inequality and well-being. Nat Clim Change 11:1111–1116 Calvin K, Bond-Lamberty B, Clarke L, Edmonds J, Eom J et al (2017) The SSP4: A world of deepening inequality. Glob Environ Change 42:284–296 Calvin K, Patel P, Clarke L, Asrar G, Bond-Lamberty B et al (2019) GCAM v5.1: Representing the linkages between energy, water, land, climate, and economic systems. Geosci Model Dev 12:677–698 Chepeliev M, Rodarte O, van der Mensbrugghe I, D (2021) Distributional impacts of carbon pricing policies under the Paris Agreement: Inter and intra-regional perspectives. Energy Econ 101:105408 Edmonds JA, Forrister D, Clarke L, de Clara S, Munnings C (2019) The Economic Potential of Article 6 of the Paris Agreement and Implementation Challenges. World Bank, Washington, DC Edmonds JA, Yu S, McJeon H, Forrister D, Aldy J et al (2021) How much could Article 6 enhance NDC ambition? Clim Change Econ 12(2):2150007 Emmerling J, Andreoni P, Charalampidis I, Dasgupta S, Dennig F et al (2024) A multi-model assessment of inequality and climate change. Nat Clim Change 14:1254–1260 Gazzotti P, Emmerling J, Marangoni G, Castelletti A, van der Wijst K-I, Tavoni M (2021) Persistent inequality in economically optimal climate policies. Nat Commun 12:3421 IETA (2023) Modelling the Economics of Article 6: A Capstone Report. International Emissions Trading Association, Geneva IPCC (2022) Climate Change 2022: Mitigation of Climate Change. Contribution of Working Group III to the Sixth Assessment Report. Cambridge University Press, Cambridge Keane J, Fernandez M (2025) How carbon markets can deliver climate investment for developing nations. ODI Insights. Overseas Development Institute, London Lakner C, Milanovic B (2016) Global income distribution: From the fall of the Berlin Wall to the Great Recession. World Bank Economic Rev 30(2):203–232 Lang SF, Dennig F, Budolfson M, Errickson F, Tavoni M (2025) Within-country inequality and the shaping of a just global climate policy. Proceedings of the National Academy of Sciences 122: e2505239122 Michaelowa A (2019) Evolution of international carbon markets: lessons for the Paris Agreement. WIREs Clim Change 10(6):e613 Nordhaus W (2015) Climate clubs: Overcoming free-riding in international climate policy. Am Econ Rev 105(4):1339–1370 OECD (2023) The Interplay Between Voluntary and Compliance Carbon Markets. OECD Publishing, Paris Piris-Cabezas P, Lubowski R, Leslie G (2023) Estimating the power of international carbon markets to increase global climate ambition. World Dev 167:106243 Rao ND, Sauer P, Gidden M, Riahi K (2019) Income inequality projections for the Shared Socioeconomic Pathways (SSPs). Futures 105:27–39 Riahi K, van Vuuren DP, Kriegler E, Edmonds J, O’Neill BC et al (2017) The Shared Socioeconomic Pathways and their energy, land use, and greenhouse gas emissions implications: An overview. Glob Environ Change 42:153–168 Schneider L, Cludius J, La Hoz Theuer S (2018) Accounting for the linking of emissions trading systems under Article 6.2 of the Paris Agreement. ICAP/Oeko-Institut, Berlin Schneider L, La Hoz Theuer S (2019) Environmental integrity of international carbon market mechanisms under the Paris Agreement. Clim Policy 19(3):386–400 Shorrocks AF (1980) The class of additively decomposable inequality measures. Econometrica 48(3):613–625 UNCTAD (2024) The Least Developed Countries Report 2024: Leveraging Carbon Markets for Development. Geneva: United Nations Conference on Trade and Development UNFCCC (2021) Decision 2/CMA.3: Guidance on cooperative approaches referred to in Article 6, paragraph 2, of the Paris Agreement. Adopted at CMA3, Glasgow, November 2021. FCCC/PA/CMA /2021/10/Add.1 UNFCCC (2024) Article 6 of the Paris Agreement. https://unfccc.int/process-and-meetings/the-paris-agreement/article-6 UNFCCC (2025a) Article 6.2 Reference Manual for the Accounting, Reporting and Review of Cooperative Approaches. Version 3. UNFCCC Secretariat, Bonn UNFCCC (2025b) Synthesis report on nationally determined contributions under the Paris Agreement. Synthesis of NDCs submitted by 30 September 2025. FCCC/PA/CMA /2025/8 Additional Declarations The authors declare no competing interests. Supplementary Files article6SIv2.2.pdf Supplementary Information Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8793231","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":586140084,"identity":"ca9fec59-4726-426b-beb7-51a805ca1e18","order_by":0,"name":"Mel George","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAw0lEQVRIiWNgGAWjYDADfiCWYGBgJkGLZAPJWgwOEKtF3r352McvFdsSN9/ITrzBUGGd2EBIi+GZY8mzZc7cTtx2I3ezBcOZdCK0zMgxZpZsu21sdiN3mwRj22EitMx/A9Ty77ax8QyQln9EaJGX4DFm/NhwW85AAqSlgQgtBjxpycwMx27LSZx5u9ki4Vi6MWFb2g8fZvxRc5uHvz13440PNdayhG05AIwLHhgvgZBysC1AQxl/EKNyFIyCUTAKRi4AAJnNQJZ2fBFDAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-2688-3388","institution":"Center for Global Sustainability, School of Public Policy, University of Maryland - College Park, MD USA","correspondingAuthor":true,"prefix":"","firstName":"Mel","middleName":"","lastName":"George","suffix":""},{"id":586140085,"identity":"f25b24ba-2d3f-434f-9336-d1ab3a773870","order_by":1,"name":"James Edmonds","email":"","orcid":"","institution":"Center for Global Sustainability, School of Public Policy, University of Maryland - College Park, MD USA","correspondingAuthor":false,"prefix":"","firstName":"James","middleName":"","lastName":"Edmonds","suffix":""}],"badges":[],"createdAt":"2026-02-05 06:55:03","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-8793231/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8793231/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102035838,"identity":"bdf5024b-720f-4a69-b886-730617965974","added_by":"auto","created_at":"2026-02-06 11:58:31","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":116638,"visible":true,"origin":"","legend":"\u003cp\u003ePopulation-weighted between-region Gini coefficient trajectories, 2025 to 2050. Solid lines show baseline scenarios (no climate policy); dashed lines show NDC cooperative scenarios (with Article 6 transfers). Panels display (a) SSP1 convergent development, (b) SSP2 middle of the road, (c) SSP4 divergent development, and (d) net-zero 2050 under SSP2 baseline. Negative Δ values indicate Article 6 reduces inequality; positive values indicate Article 6 increases inequality. Under SSP1, SSP2, and NZ2050, cooperative lines fall below baselines throughout (progressive). Under SSP4, lines nearly converge by 2050 (Δ = +0.03), indicating a near-neutral outcome as the transfer effect weakens but does not reverse.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8793231/v1/250f1494a14d8bdb407742a3.jpg"},{"id":102035844,"identity":"e780f93c-553f-40a9-bcfe-017405e963f4","added_by":"auto","created_at":"2026-02-06 11:58:38","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":102034,"visible":true,"origin":"","legend":"\u003cp\u003eNet financial transfers by region group across scenarios, 2030 to 2050 (billion 2025 USD). Positive values (above zero line) indicate net payments by credit buyers; negative values (below zero line) indicate net receipts by credit sellers. Panels display (a) SSP1 NDC cooperative, (b) SSP2 NDC cooperative, (c) SSP4 NDC cooperative, and (d) NZ2050 cooperative. Under SSP1 and SSP2, Africa is a net seller receiving transfers. Under SSP4, Africa shifts to net buyer by 2050, paying over $554 billion. China and South \u0026amp; SE Asia remain large sellers across all NDC scenarios but shift to buyers under NZ2050 as stringent targets exhaust low-cost domestic mitigation options.\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8793231/v1/ba0ab9a1b2aee33e3d209c66.jpg"},{"id":102035911,"identity":"dc6d1bdd-d388-4998-91d5-c29692414e94","added_by":"auto","created_at":"2026-02-06 11:58:56","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":80755,"visible":true,"origin":"","legend":"\u003cp\u003eAfrica net ITMO transfers under Article 6 by socioeconomic pathway, 2025 to 2050 (billion 2025 USD). Positive values indicate Africa receives transfers (net seller); negative values indicate Africa pays transfers (net buyer). Under SSP1 (convergent development), Africa receives increasing transfers reaching $1,269 billion by 2050. Under SSP2 (middle of the road), Africa remains a modest seller receiving $226 billion by 2050. Under SSP4 (divergent development), Africa shifts from seller to buyer, paying $1,219 billion by 2050. This $2.5 trillion swing in Africa's position between SSP1 and SSP4 illustrates how development trajectories determine whether Article 6 is progressive or regressive for the poorest regions.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8793231/v1/d91c23d90521144a33905533.jpg"},{"id":102035999,"identity":"0f1ec899-d77b-4b90-a575-9722ae041c5d","added_by":"auto","created_at":"2026-02-06 11:59:33","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1231021,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8793231/v1/03d54d11-23d5-4e55-93e1-72c8ed3ca65b.pdf"},{"id":102035876,"identity":"1555ff5c-5280-4e5b-a6c3-988dcf9c1f79","added_by":"auto","created_at":"2026-02-06 11:58:40","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":511561,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Information\u003c/p\u003e","description":"","filename":"article6SIv2.2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8793231/v1/3f6a6f1fb0ca56121a18e79d.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eWhen Do Carbon Markets Reduce Inequality? Article 6 Transfers Under Alternative Futures\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003e1.1. Motivation\u003c/h2\u003e \u003cp\u003eArticle 6 of the Paris Agreement establishes pathways for voluntary international cooperation in mitigation. Under cooperative approaches (Article 6.2), Parties may authorize and transfer internationally transferred mitigation outcomes (ITMOs) toward achievement of their nationally determined contributions (NDCs), subject to accounting requirements intended to preserve environmental integrity (UNFCCC, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In principle, an ITMO market can lower the cost of meeting a given set of national targets by reallocating abatement to where it is cheaper. Modeling studies using integrated assessment frameworks have estimated that cooperative implementation could reduce NDC compliance costs by more than half and, if savings were reinvested, could nearly double global mitigation relative to independent implementation (Edmonds et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Aldy et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAn ITMO market also generates cross-border financial flows: buyers pay sellers for mitigation outcomes. These flows are increasingly discussed as a potential channel of climate finance, with some analysts positioning carbon markets as vehicles for mobilizing investment in developing countries (Keane and Fernandez, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; OECD, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The UNFCCC itself notes that Article 6 can \u0026ldquo;unlock financial support for developing countries\u0026rdquo; (UNFCCC, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Yet most quantitative assessments of Article 6 foreground efficiency gains, trading volumes, and accounting design rather than distributional incidence across regions. The question of whether ITMO transfers reduce or exacerbate between-region inequality has not been directly examined in the modeling literature.\u003c/p\u003e \u003cp\u003eAt the same time, integrated assessment modeling has begun to engage more explicitly with inequality and distributional outcomes. Emmerling et al. (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) coordinate an ensemble of eight IAMs to assess how climate impacts and mitigation pathways affect global inequality, finding that unmitigated climate change raises the global Gini coefficient by approximately 1.4 points by 2100, while Paris-aligned mitigation with equal-per-capita revenue recycling can reduce inequality below baseline levels. Gazzotti et al. (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) demonstrate that even under economically optimal climate policies with full international cooperation, significant between-region inequality persists, with the ratio of top-to-bottom income deciles remaining 117% higher than in a no-climate-change counterfactual. These findings underscore that efficiency and equity are distinct: aggregate cost savings do not automatically translate to reduced inequality. The theoretical case for international cooperation in climate mitigation rests on the potential for mutual gains from trade in abatement. Nordhaus (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) formalizes this through the concept of \"climate clubs,\" showing that coalitions with internal carbon pricing and border adjustments can overcome free-riding that undermines purely voluntary agreements. Article 6 represents a partial implementation of this logic: it enables cost-reducing trades but relies on voluntary participation rather than club-based enforcement.\u003c/p\u003e \u003cp\u003eThis paper asks a focused question: when a stylized ITMO market is embedded within an integrated assessment framework under NDC-consistent policy constraints, do the resulting financial transfers reduce between-region income inequality? We answer using a population-weighted global Gini coefficient computed across model regions, and we decompose the change to identify which regions drive the result and whether the effect operates primarily through domestic mitigation cost reallocation or through international payment flows.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e1.2. Contributions\u003c/h2\u003e \u003cp\u003eWe make four contributions. First, we provide transparent accounting of ITMO trading under NDC-consistent constraints across three socioeconomic baselines (SSP1, SSP2, SSP4) and a net-zero 2050 extension. This design isolates how baseline heterogeneity conditions distributional outcomes, rather than treating Article 6 in a single reference world. Baseline inequality varies substantially across SSPs: SSP1 features relatively high income convergence, SSP2 a middle-of-the-road trajectory, and SSP4 a world of deepening inequality with greater income dispersion across regions (Riahi et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Rao et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). By running Article 6 experiments across these baselines, we can assess whether the market\u0026rsquo;s distributional effects depend on initial conditions.\u003c/p\u003e \u003cp\u003eSecond, we track net financial transfers as an explicit accounting object, making cross-border payment flows visible for distributional evaluation.\u003c/p\u003e \u003cp\u003eThird, we move beyond reporting an aggregate inequality statistic by decomposing the change in the population-weighted global Gini into components: income effects from domestic mitigation costs (policy burden) and income effects from international ITMO transfers. This decomposition identifies which regions contribute most to the inequality change and clarifies whether the aggregate result is broad-based or driven by a small subset of large players.\u003c/p\u003e \u003cp\u003eFourth, we employ Theil index decomposition to formally partition inequality changes into between-region and within-region components. This approach, which exploits the Theil index\u0026rsquo;s exact additive decomposability, allows us to verify whether Article 6 transfers operate primarily through cross-border redistribution or also affect within-region inequality through our distributional assumptions. This methodological contribution complements the Gini analysis and provides a robustness check on our central findings.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e1.3. Preview of findings\u003c/h2\u003e \u003cp\u003eOur headline finding is that Article 6 transfers reduce between-region inequality under most scenarios, with the magnitude contingent on development conditions. Under SSP1 and SSP2, where incomes converge across regions over time, Article 6 produces progressive outcomes: lower-income regions have cheaper mitigation options, become net sellers of ITMOs, and receive financial transfers from higher-income buyers. The Gini coefficient falls by 0.5 to 0.9 points by 2050 relative to baselines. Under net-zero 2050 ambition, transfer volumes and carbon prices rise substantially, strengthening the inequality reduction to approximately 0.8 points.\u003c/p\u003e \u003cp\u003eUnder SSP4, the progressive effect weakens substantially but does not reverse into a strongly regressive outcome. In this divergent development pathway, lower-income regions lack capital for clean technology deployment, and their mitigation costs rise relative to higher-income regions. By mid-century, Africa shifts from net seller to net buyer of ITMOs. However, large sellers like China and India continue to receive substantial transfers, partially offsetting Africa\u0026rsquo;s reversal. The transfer effect remains slightly progressive (-0.2 Gini points), but combined with a regressive policy burden (+\u0026thinsp;0.3 points), the net effect is near-neutral (+\u0026thinsp;0.03 points). Africa\u0026rsquo;s position serves as a diagnostic indicator of which lower-income regions gain or lose: Africa pays \u003cspan\u003e$\u003c/span\u003e554\u0026nbsp;billion under SSP4 but receives \u003cspan\u003e$\u003c/span\u003e577\u0026nbsp;billion under SSP1 by 2050, a swing exceeding \u003cspan\u003e$\u003c/span\u003e1.1 trillion.\u003c/p\u003e \u003cp\u003eThe decomposition reveals that the transfer channel dominates the policy burden by a factor of three to five, operating primarily through the between-region channel under convergent scenarios. We discuss these findings in light of critiques that carbon markets may not deliver equitable outcomes, and we identify the mechanism determining the magnitude of Article 6\u0026rsquo;s progressive effect: comparative advantage in low-cost mitigation, which depends on development trajectories.\u003c/p\u003e \u003c/div\u003e"},{"header":"2. Related Literature","content":"\u003cp\u003eTwo literatures inform this work: one on inequality in integrated assessment modeling, the other on Article 6 economics and implementation. We bridge them by treating ITMO cooperation as an explicit system of cross-border financial transfers and quantifying its distributional incidence.\u003c/p\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Inequality in integrated assessment modeling\u003c/h2\u003e \u003cp\u003eA growing body of work has moved distributional questions from the periphery of scenario analysis toward a central object of study. Emmerling et al. (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) provide a prominent example, coordinating an ensemble of eight large-scale IAMs that incorporate income heterogeneity to evaluate how climate risks and Paris-aligned policies interact with inequality. They find that unmitigated climate change increases the global Gini coefficient by approximately 1.4 points by 2100, and that stringent mitigation (1.5\u0026deg;C pathways) alleviates much of this exacerbation in the long run but can slightly increase inequality in the short term without complementary redistribution policies. They also show that if carbon pricing revenues are recycled on an equal-per-capita basis, global inequality can fall below baseline levels, demonstrating that policy design matters for distributional outcomes.\u003c/p\u003e \u003cp\u003eGazzotti et al. (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) use a cost-benefit IAM with 50 regions to examine inequality under economically optimal climate policies. Even with full international cooperation achieving temperature stabilization around 1.8\u0026deg;C, they find that the ratio of top-to-bottom income deciles remains 117% higher than in a counterfactual without climate change. Their results underscore that optimal policy guided by efficiency criteria does not inherently resolve between-region disparities. Climate damages fall disproportionately on lower-income regions, and cost-minimizing abatement allocation does not compensate for this pattern. Related work shows that carbon pricing with equal-per-capita revenue redistribution can substantially reduce inequality (Budolfson et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Lang et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), establishing that climate policy design affects distributional outcomes.\u003c/p\u003e \u003cp\u003eOur paper complements this literature by examining a distinct channel: cross-border financial transfers generated by international cooperation through carbon markets, rather than domestic revenue recycling or explicit transfer schemes. The Article 6 mechanism differs from equal-per-capita redistribution in that payment flows depend on market outcomes, namely which regions have lower marginal abatement costs and thus become sellers. Whether this market-based channel produces distributional effects comparable to designed redistribution remains an open question.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.2. Modeling Article 6 cooperation\u003c/h2\u003e \u003cp\u003eA parallel literature has developed around the economics and implementation of Article 6 cooperation. Early analyses emphasize that cooperative approaches can substantially lower the costs of achieving mitigation targets by reallocating abatement to lower-cost opportunities. Edmonds et al. (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) assess the economic potential of Article 6 using GCAM, finding that global trading could reduce NDC implementation costs by more than half (approximately \u003cspan\u003e$\u003c/span\u003e250\u0026nbsp;billion per year in 2030). Aldy et al. (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), in a peer-reviewed extension published in Climate Change Economics, show that if cost savings were reinvested in enhanced ambition, cooperative implementation could yield an additional 9 GtCO2/year of mitigation, with more than half of near-term gains coming from land-use measures. These modeling exercises build on two decades of experience with international carbon market mechanisms. Michaelowa (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) traces the evolution from the Clean Development Mechanism through Joint Implementation to Article 6, identifying four phases ranging from \"exuberance\" (2005\u0026ndash;2011) to \"hibernation\" (2012\u0026ndash;2014) to the current \"reconfiguration\" under the Paris Agreement. This history informs expectations about how Article 6 markets may develop and what distributional patterns might emerge. Piris-Cabezas et al. (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) use a partial-equilibrium carbon market model to estimate that Article 6 trading could enable nearly double the climate ambition through 2035 at no additional cost.\u003c/p\u003e \u003cp\u003eOur previous work summarizes findings across NDC and net-zero pathways, showing that cooperative implementation substantially reduces total compliance costs and that financial flows between regions could exceed \u003cspan\u003e$\u003c/span\u003e1 trillion per year by 2050 under ambitious scenarios (IETA, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). These studies also explore market design variations, including limits on ITMO usage by buyers and transaction fees. They find that usage limits raise costs without climate benefit by forcing buyers into more expensive domestic abatement while reducing sellers\u0026rsquo; opportunities to earn revenue (IETA, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Transaction fees similarly shift burdens toward sellers and shrink trading volumes. Environmental integrity concerns have shaped Article 6 rule development. Schneider and La Hoz Theuer (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) provide a framework for assessing integrity across multiple dimensions: accurate baselines, demonstrated additionality, robust monitoring, and avoidance of double counting through corresponding adjustments. These safeguards impose transaction costs but are necessary conditions for ITMO transfers to represent real emission reductions rather than accounting artifacts.\u003c/p\u003e \u003cp\u003eThe operational framework for Article 6.2 is codified in Decision 2/CMA.3, adopted at CMA3 in Glasgow in November 2021, which establishes requirements for ITMO authorization, corresponding adjustments to national inventories, and reporting cycles (UNFCCC, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Subsequent guidance at COP27 and COP29 has clarified implementation details. The UNFCCC Secretariat\u0026rsquo;s Article 6.2 Reference Manual provides comprehensive technical guidance on accounting procedures (UNFCCC, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2025a\u003c/span\u003e). Schneider et al. (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) examine how emissions trading system linkages would operate under Article 6.2 accounting rules, addressing challenges for countries with single-year NDC targets.\u003c/p\u003e \u003cp\u003eDistributional modeling in climate policy typically examines carbon pricing with domestic revenue recycling or explicit climate finance transfers. Our contribution is to treat ITMO payments as cross-border transfers operating within an NDC framework and to measure the between-region inequality implication using a population-weighted Gini coefficient. We then decompose the Gini change to identify regional contributions. Chepeliev et al. (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), who link CGE modeling with microsimulation to assess distributional impacts of carbon pricing coalitions, represent the closest methodological analog; they find progressive global effects and reduced poverty burdens under cooperation, but their focus is pricing coordination rather than Article 6 mechanisms specifically. Existing Article 6 modeling thus tracks costs, volumes, and abatement but not distributional incidence. The governance literature examines market fragmentation and integrity (Ahonen et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Michaelowa, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) but does not quantify inequality effects. We find no direct quantification of between-region inequality outcomes from ITMO trading in the peer-reviewed IAM literature, despite extensive modeling of Article 6 cost savings and volumes (Edmonds et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2019\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; IETA, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). This is the gap our paper addresses.\u003c/p\u003e \u003cp\u003eOur NDC representation uses the post-Belem generation of submitted NDCs as recorded in the UNFCCC registry through 30 September 2025, the cut-off date for the UNFCCC\u0026rsquo;s 2025 NDC Synthesis Report (UNFCCC, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2025b\u003c/span\u003e). That synthesis covers 64 new or updated NDCs and notes that 89% of Parties indicated intention to participate in Article 6 cooperation, up from 64% in previous submissions. We treat this as a realism improvement for NDC-world modeling while recognizing that our central focus remains the distributional incidence of ITMO transfers conditional on heterogeneous development pathways.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Methods","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.1. Model and regional structure\u003c/h2\u003e \u003cp\u003eWe use GCAM-8s, a variant of the Global Change Assessment Model version 8.2 developed for the ScenarioMIP exercise. GCAM is a market-equilibrium model that integrates representations of energy, water, land, and economic systems across 32 geopolitical regions (Calvin et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The model solves for equilibrium prices and quantities in energy and agricultural markets subject to resource constraints, technology availability, and policy interventions. It operates on five-year time steps from 2015 to 2100 and uses a recursive-dynamic solution approach without perfect foresight.\u003c/p\u003e \u003cp\u003ePopulation and GDP trajectories follow the Shared Socioeconomic Pathways. We implement SSP1, SSP2, and SSP4 to capture variation in baseline income dispersion. SSP1 represents a sustainability-oriented world with relatively low population growth and high income convergence across regions. SSP2 represents a middle-of-the-road trajectory. SSP4 represents a world of deepening inequality, with greater income dispersion and slower convergence (Riahi et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). GCAM served as the marker model for SSP4 in the original SSP quantification (Calvin et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Baseline Gini projections for the SSPs are documented in Rao et al. (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFor the inequality analysis, the relevant accounting objects are regional GDP per capita and population. GCAM tracks both, enabling computation of a population-weighted global Gini coefficient across the 32 regions. This yields a between-region inequality measure; it does not capture within-region inequality. We interpret changes in this metric as shifts in cross-region income dispersion induced by policy, recognizing that the 32-region aggregation smooths over within-region heterogeneity. GCAM\u0026rsquo;s 32 regions aggregate multiple countries; our \u0026ldquo;between-region\u0026rdquo; inequality measure should not be interpreted as country-level inequality.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.2. Stylized ITMO market\u003c/h2\u003e \u003cp\u003eWe embed a stylized representation of Article 6 cooperation in GCAM by allowing regions to trade emission permits under a global carbon market that clears at a single price. In the cooperative scenario, each region faces its NDC constraint but can meet part of that constraint by purchasing ITMOs from regions that over-comply and sell. The market clears when global supply of ITMOs equals global demand, yielding an equilibrium carbon price that equates marginal abatement costs across participating regions.\u003c/p\u003e \u003cp\u003eWe define \"transfers\" as payments equal to the equilibrium carbon price multiplied by net ITMO volume: sellers receive revenue, buyers incur expenditure. These are compliance payments for mitigation outcomes, not aid or grants. We treat them as finance-like because they constitute cross-border flows that relax budget constraints for sellers and tighten them for buyers, affecting regional mean income. This framing does not imply that transfers generate development co-benefits, reach intended beneficiaries within regions, or map directly to welfare gains. Those outcomes depend on domestic revenue use, which GCAM does not model.\u003c/p\u003e \u003cp\u003eNet ITMO positions are determined endogenously by the model. Regions with marginal abatement costs below the global clearing price become net sellers; those with costs above become net buyers (IETA, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The financial transfer associated with trading is computed as the product of the equilibrium carbon price and net ITMO volume: a net seller receives revenue equal to the price multiplied by ITMOs sold, while a net buyer incurs expenditure equal to the price multiplied by ITMOs purchased.\u003c/p\u003e \u003cp\u003eIn the independent implementation counterfactual, each region meets its NDC independently through domestic mitigation only. The regional carbon price in this case is the shadow price of the domestic constraint, varying across regions according to NDC stringency and abatement cost structure. Comparing outcomes with and without trading isolates the effect of Article 6 cooperation on regional incomes and inequality.\u003c/p\u003e \u003cp\u003eThis implementation follows precedents in GCAM-based Article 6 modeling (Edmonds et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Aldy et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; IETA, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). We assume a perfectly competitive global market with full fungibility of ITMOs and do not impose bilateral restrictions, usage limits, or transaction fees in the core scenarios. These simplifications maximize the scope for observing distributional effects; sensitivity to market design features is a subject for future work.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.3. Scenario design\u003c/h2\u003e \u003cp\u003eWe construct scenarios along two dimensions: socioeconomic baseline (SSP1, SSP2, SSP4) and policy ambition (NDC, net-zero 2050). The NDC case implements national targets as recorded in the UNFCCC registry through 30 September 2025, following the scope of the 2025 NDC Synthesis Report (UNFCCC, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2025b\u003c/span\u003e). NDCs are translated into regional emission constraints in GCAM using standard harmonization procedures consistent with IPCC AR6 scenario protocols (IPCC, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Absolute targets are implemented directly; intensity and BAU-relative targets are converted to absolute terms using model-consistent GDP and baseline emissions projections.\u003c/p\u003e \u003cp\u003eThe net-zero 2050 case implements a trajectory in which all regions reach net-zero CO2 emissions by 2050, representing a high-ambition extension. This case generates higher carbon prices and larger transfer volumes, allowing examination of how policy stringency affects distributional outcomes.\u003c/p\u003e \u003cp\u003eFor each baseline-ambition combination, we run two variants: one with Article 6 trading (cooperative implementation) and one without (independent implementation). This yields eight core scenario pairs. We also test sensitivity to land-use sector scope in the SSP2 baseline, examining whether including land-use change emissions in the carbon market alters the distributional findings. Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e summarizes the scenario matrix.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eScenario matrix and market characteristics.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBaseline\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePolicy Ambition\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLUC Pricing\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eArticle 6\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eScenario Label\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSSP1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNDC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSSP1 NDC independent\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNDC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSSP1 NDC cooperative\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSSP2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNDC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSSP2 NDC independent\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNDC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSSP2 NDC cooperative\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNDC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSSP2 NDC LUC independent\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNDC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSSP2 NDC LUC cooperative\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNet-Zero 2050\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNZ2050 independent\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNet-Zero 2050\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNZ2050 cooperative\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSSP4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNDC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSSP4 NDC independent\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNDC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSSP4 NDC cooperative\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003cem\u003eNote: SSP1 represents convergent development with high income convergence across regions. SSP2 represents middle-of-the-road development. SSP4 represents divergent development with persistent inequality. NDC scenarios implement nationally determined contributions as recorded in the UNFCCC registry through 30 September 2025. Net-Zero 2050 implements a trajectory where all regions reach net-zero CO2 emissions by 2050. LUC Pricing indicates whether land-use change emissions are included in carbon pricing. Market size is total annual financial flow from buyers to sellers (billion 2025 USD). Carbon price is the equilibrium clearing price (USD per tonne CO2).\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.4. Inequality metrics\u003c/h2\u003e \u003cp\u003eWe measure inequality using two complementary approaches: population-weighted Gini coefficients and Theil index decomposition.\u003c/p\u003e \u003cp\u003e \u003cb\u003eBetween-region Gini coefficient.\u003c/b\u003e The Gini coefficient ranges from 0 (perfect equality) to 1 (maximum inequality). For a set of regions indexed by r with GDP per capita y_r and population weight w_r, the population-weighted Gini is computed using the standard formula based on mean absolute differences in incomes (Lakner and Milanovic, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). We focus on 2030 and 2050 as key policy horizons.\u003c/p\u003e \u003cp\u003eRegional GDP per capita is adjusted for policy-induced income effects. In the independent implementation scenario, adjusted income equals baseline GDP per capita minus domestic mitigation cost (computed as the change in regional GDP from baseline to policy scenario in GCAM). In the cooperative scenario, adjusted income additionally reflects net ITMO transfers: sellers gain revenue, buyers incur expenditure. All values are expressed in per capita terms using regional population.\u003c/p\u003e \u003cp\u003eThis metric captures between-region inequality only. It does not account for within-region income distributions, which GCAM does not represent at the household level. We interpret changes in the global Gini as indicative of shifts in cross-region income dispersion, recognizing that interpersonal global inequality would require additional information on within-region distributions. The Gini coefficient is widely used in this literature (Emmerling et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Lakner and Milanovic, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) despite known limitations, including non-decomposability into additive between- and within-group components (Bourguignon, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e1979\u003c/span\u003e). We report results using the Gini for comparability with existing work.\u003c/p\u003e \u003cp\u003e \u003cb\u003eTheil index decomposition.\u003c/b\u003e To formally partition inequality changes into between-region and within-region components, we employ the Theil T index (also known as GE(1)) is exactly additively decomposable (Shorrocks, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e1980\u003c/span\u003e), unlike the Gini coefficient (Bourguignon, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e1979\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe total Theil index T can be written as:\u003c/p\u003e \u003cp\u003eT\u0026thinsp;=\u0026thinsp;T_between\u0026thinsp;+\u0026thinsp;T_within\u003c/p\u003e \u003cp\u003eThe between-region component treats each region as a unit with its mean income. The within-region component is a weighted sum of regional Theil indices, where weights reflect each region\u0026rsquo;s share of total income. For this decomposition, we construct 320 observations (32 regions times 10 income deciles) using within-region income distributions from SSP-specific projections (Rao et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThis decomposition serves two purposes. First, it provides a robustness check on our Gini-based findings by using an alternative inequality measure. Second, and more importantly, it allows us to verify that Article 6 transfers operate primarily through the between-region channel. Because GCAM does not model how ITMO revenues or payments are distributed within regions, we make an equal-per-capita (EPC) assumption: transfers are distributed uniformly across all individuals within each region. Under this assumption, transfers have a small mechanical effect on within-region inequality (equal absolute transfers are progressive in relative terms), but we expect the dominant effect to operate through changes in regional mean incomes. The Theil decomposition tests this expectation directly.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e3.5. Decomposition of income effects\u003c/h2\u003e \u003cp\u003eTo understand what drives the change in inequality between the independent and cooperative scenarios, we track Gini changes across three sequential stages.\u003c/p\u003e \u003cp\u003eStage 1: Baseline inequality - The Gini coefficient under each SSP baseline without climate policy.\u003c/p\u003e \u003cp\u003eStage 2: Policy burden effect - The change from baseline to policy scenario, capturing how mitigation costs are distributed before trading.\u003c/p\u003e \u003cp\u003eStage 3: Transfer effect - The change from policy to cooperative scenario, isolating the impact of ITMO financial flows. If money flows from rich to poor (poor regions sell), this is progressive; the reverse is regressive. Full procedural details appear in SI Section S4.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e3.6. Distributional assumptions and limitations\u003c/h2\u003e \u003cp\u003eGCAM does not endogenously model within-region income distribution changes from climate policy. We therefore make the following assumptions about how policy effects are distributed across deciles within each region.\u003c/p\u003e \u003cp\u003eWe assume both mitigation costs and ITMO payment flows are distributed equally per capita across all deciles within each region, equivalent to lump-sum domestic revenue recycling. Second, ITMO payment flows are distributed equally per capita within receiving and paying regions. This equal-per-capita (EPC) assumption is equivalent to assuming lump-sum domestic revenue recycling. Under EPC, equal absolute income changes have progressive relative effects (larger percentage change for poorer deciles within each region). Regions receiving transfers see slight within-region inequality reduction; regions paying for credits see slight within-region inequality increase.\u003c/p\u003e \u003cp\u003eThe EPC assumption represents a specific policy choice about domestic revenue use. Alternative recycling mechanisms (e.g., proportional to income, targeted transfers to poor households) would yield different within-region effects not captured in this analysis. This is a known limitation: Article 6 determines the magnitude and direction of between-region flows, but domestic fiscal policy determines how those flows affect within-region distribution. Our analysis isolates the between-region channel, which Theil decomposition confirms that 92 to 100 percent of the transfer effect operates through the between-region channel under SSP1, SSP2, and net-zero scenarios (Table S5). Under SSP4, this share falls to 79 percent by 2050.\u003c/p\u003e \u003cp\u003eAdditional limitations warrant acknowledgment. First, GCAM\u0026rsquo;s 32-region structure aggregates substantial heterogeneity; results for model regions should not be interpreted as country-specific predictions. Second, we model a stylized global market with perfect competition and full participation; actual Article 6 implementation involves bilateral arrangements, varying levels of readiness, and transaction costs that could alter outcomes. Third, our income measure is GDP per capita adjusted for mitigation costs and transfers; this is a proxy for economic resources but does not capture welfare, consumption, or non-monetary dimensions of inequality. Fourth, we do not model climate damages, so our inequality trajectories reflect mitigation policy effects only.\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Results","content":"\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\n \u003ch2\u003e4.1. Market characteristics\u003c/h2\u003e\n \u003cp\u003eBefore examining distributional outcomes, we characterize the Article 6 markets that emerge under each scenario. Market size, measured as total financial flows from buyers to sellers, varies substantially across socioeconomic baselines and policy ambition levels. Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e reports market sizes and equilibrium carbon prices for key years.\u003c/p\u003e\n \u003cp\u003eUnder SSP2 with NDC-level ambition, the ITMO market grows from \u003cspan\u003e$\u003c/span\u003e380 billion in 2030 to \u003cspan\u003e$\u003c/span\u003e953 billion in 2040 and \u003cspan\u003e$\u003c/span\u003e1,672\u0026nbsp;billion in 2050. SSP1 produces smaller markets (\u003cspan\u003e$\u003c/span\u003e261 billion to \u003cspan\u003e$\u003c/span\u003e706\u0026nbsp;billion over the same period) because income convergence reduces the dispersion in regional mitigation costs that drives trading gains. SSP4 generates the largest markets (\u003cspan\u003e$\u003c/span\u003e427 billion to \u003cspan\u003e$\u003c/span\u003e1,744 billion) as income divergence widens cost differentials between regions. Under net-zero 2050 ambition, market size reaches approximately \u003cspan\u003e$\u003c/span\u003e1,770\u0026nbsp;billion by 2050, comparable to SSP4 NDC levels but with different regional compositions.\u003c/p\u003e\n \u003cp\u003eEquilibrium carbon prices range from \u003cspan\u003e$\u003c/span\u003e30\u0026ndash;155/tCO2 (SSP1) to \u003cspan\u003e$\u003c/span\u003e45\u0026ndash;207/tCO2 (SSP4) over 2030\u0026ndash;2050, with net-zero 2050 reaching \u003cspan\u003e$\u003c/span\u003e285/tCO2 by 2050. Higher prices under SSP4 reflect concentrated mitigation capacity; higher prices under net-zero reflect policy stringency. These market magnitudes are substantial in development finance terms. Annual ITMO flows under SSP2 NDC cooperative represent 0.3 to 1.5 percent of global GDP across the projection period, comparable to current official development assistance flows. For recipient regions, transfers can reach 10 to 20 percent of regional GDP in scenarios where those regions are net sellers.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\n \u003ch2\u003e4.2. Gini trajectories across scenarios\u003c/h2\u003e\n \u003cp\u003eFigure 1 presents Gini trajectories under baseline and cooperative scenarios. The vertical distance between lines represents Article 6\u0026apos;s net effect.\u003c/p\u003e\n \u003cp\u003eIn the convergent development scenario (SSP1), Article 6 produces consistent inequality reduction throughout the projection period. The Gini coefficient falls by 0.32 points relative to baseline in 2030, by 0.44 points in 2040, and by 0.90 points in 2050. This progressive pattern reflects the comparative advantage mechanism: in a convergent development world, lower-income regions have cheaper mitigation options, become net sellers of ITMOs, and receive financial transfers from higher-income buyers.\u003c/p\u003e\n \u003cp\u003eThe middle-of-the-road scenario shows a similar progressive pattern. The Gini falls by 0.38 points in 2030, by 0.95 points in 2040, and by 0.48 points in 2050. The pattern remains consistently progressive, with transfers flowing from rich to poor regions throughout the period.\u003c/p\u003e\n \u003cp\u003eThe net-zero 2050 pathway produces larger inequality reductions than NDC-level ambition despite imposing greater mitigation costs. The Gini falls by approximately 0.06 points in 2030, 0.24 points in 2040, and 0.76 points in 2050. Higher climate ambition strengthens the progressive effect of Article 6. The mechanism is that transfer volumes scale faster than mitigation burdens as ambition increases. Under net-zero targets, the global carbon price rises substantially, increasing the value of each ITMO traded. Because low-income regions retain comparative advantage in low-cost mitigation under SSP2 baseline conditions, they sell more ITMOs at higher prices, receiving larger transfers.\u003c/p\u003e\n \u003cp\u003eTable \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e summarizes Gini levels and changes across scenarios, decomposing the net effect into policy burden and transfer components (discussed in Section\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4.4\u003c/span\u003e).\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\u0026nbsp;\u003ctable id=\"Tab2\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eBetween-region Gini coefficients by scenario.\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"3\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eScenario\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e2030\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e2050\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSSP1 Baseline\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e57.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e50.42\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSSP1 NDC cooperative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e56.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e49.52\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSSP1 \u0026Delta;Gini\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.90\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSSP2 Baseline\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e57.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e52.51\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSSP2 NDC cooperative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e57.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e52.03\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSSP2 \u0026Delta;Gini\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.48\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSSP2 NDC LUC cooperative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e57.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e51.96\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSSP2 LUC \u0026Delta;Gini\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSSP4 Baseline\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e57.59\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e56.33\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSSP4 NDC cooperative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e57.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e56.36\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSSP4 \u0026Delta;Gini\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e+\u0026thinsp;0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNZ2050 Baseline\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e57.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e52.51\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNZ2050 cooperative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e57.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e51.75\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNZ2050 \u0026Delta;Gini\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.76\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\n \u003ch2\u003e4.3. The SSP4 pattern\u003c/h2\u003e\n \u003cp\u003eSSP4 presents a notably different outcome from the other scenarios. Under this divergent development pathway, Article 6 produces weaker progressive effects that approach neutrality by mid-century. The Gini falls by 0.19 points in 2030 and 0.77 points in 2040, but the net effect is only\u0026thinsp;+\u0026thinsp;0.03 points by 2050, essentially neutral.\u003c/p\u003e\n \u003cp\u003eThis attenuation of the progressive effect reflects erosion of comparative advantage for some lower-income regions. In SSP4, income divergence means that regions like Africa lack capital for clean technology deployment. Their mitigation costs rise relative to higher-income regions, which continue to invest in low-carbon infrastructure. By mid-century, Africa shifts from net seller to net buyer: the continent faces binding NDC constraints but lacks cheap domestic mitigation options, forcing it to purchase ITMOs.\u003c/p\u003e\n \u003cp\u003eHowever, the overall outcome does not become strongly regressive because other large developing regions remain net sellers. China and India, despite being lower-income than developed regions, retain comparative advantage in low-cost mitigation and continue to receive substantial transfers. These flows partially offset Africa\u0026rsquo;s reversal, leaving the aggregate between-region effect near-neutral.\u003c/p\u003e\n \u003cp\u003eAfrica\u0026rsquo;s position serves as a diagnostic indicator of which lower-income regions benefit from Article 6. Under SSP1 in 2050, Africa (aggregated across the four African regions in GCAM) receives \u003cspan\u003e$\u003c/span\u003e577 billion in net transfers as a seller of ITMOs. Under SSP2, Africa receives \u003cspan\u003e$\u003c/span\u003e103 billion. Under SSP4, Africa pays \u003cspan\u003e$\u003c/span\u003e554 billion as a net buyer. This swing of over \u003cspan\u003e$\u003c/span\u003e1.1 trillion between SSP1 and SSP4 represents a fundamental reversal in Africa\u0026rsquo;s role in the carbon market. When Africa sells credits, Article 6 is strongly progressive; when Africa buys credits, the progressive effect is substantially attenuated. The near-neutral aggregate outcome (+\u0026thinsp;0.03 Gini points) reflects offsetting regional dynamics: Africa\u0026apos;s shift to buyer status is counterbalanced by continued large seller positions in China and India, which retain comparative advantage even under divergent development. The \u0026quot;poor pays rich\u0026quot; reversal is thus partial, not universal. Table S3 reports the decomposition: burden\u0026thinsp;+\u0026thinsp;0.26, transfer\u0026thinsp;\u0026minus;\u0026thinsp;0.23, net\u0026thinsp;+\u0026thinsp;0.03 in SSP4 2050.\u003c/p\u003e\n \u003cp\u003eThe SSP4 pattern is not an artifact of modeling assumptions but reflects a structural feature of how comparative advantage operates in carbon markets under divergent development. If development trajectories concentrate mitigation capacity in already-wealthy regions while leaving some lower-income regions behind, the market will produce differentiated outcomes across the Global South. This finding has direct implications for Article 6 design, which we discuss in Section \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\n \u003ch2\u003e4.4. Decomposition: transfers dominate\u003c/h2\u003e\n \u003cp\u003eTo understand the mechanisms driving inequality changes, we decompose the net effect into two components: the policy burden effect (change in inequality from mitigation costs before any trading) and the transfer effect (change from ITMO financial flows). Table \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e reports this decomposition across scenarios.\u003c/p\u003e\n \u003cp\u003eThe transfer effect dominates the policy burden effect by a substantial margin across all scenarios. Under SSP2 NDC cooperative, the policy burden effect is +\u0026thinsp;0.22 Gini points by 2050 (slightly regressive, as mitigation costs are not perfectly proportional to income), while the transfer effect is -0.70 Gini points (strongly progressive). The transfer effect is 3.2 times larger in magnitude than the burden effect, and its progressive direction more than offsets the regressive burden to produce a net progressive outcome of -0.48 points.\u003c/p\u003e\n \u003cp\u003eUnder net-zero 2050, both effects are larger, but the transfer effect grows faster. The policy burden rises to +\u0026thinsp;0.30 Gini points, reflecting higher mitigation costs under stringent targets. The transfer effect reaches\u0026thinsp;\u0026minus;\u0026thinsp;1.06 Gini points, yielding a ratio of 3.5 to 1. This pattern explains why higher ambition produces more progressive outcomes: the transfer channel scales with carbon prices and trading volumes, while the burden channel scales with mitigation effort. Because low-income regions sell more under high-ambition scenarios (they have the cheapest remaining mitigation options as global abatement deepens), the progressive transfer effect outpaces the regressive burden effect.\u003c/p\u003e\n \u003cp\u003eUnder SSP1, the pattern is even more pronounced. The policy burden is +\u0026thinsp;0.18 Gini points by 2050 (mitigation costs are distributed somewhat regressively even in a convergent world), while the transfer effect is -1.08 points, a ratio exceeding 5 to 1. The net effect of -0.90 points is the most progressive across scenarios.\u003c/p\u003e\n \u003cp\u003eUnder SSP4 in 2050, the transfer effect remains slightly progressive (-0.23 points) even though Africa has become a buyer, because China and India continue as large sellers. The policy burden is +\u0026thinsp;0.26 points. Combined, the net effect is +\u0026thinsp;0.03 points, essentially neutral. This contrasts with the other scenarios where the transfer effect decisively offsets the burden.\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\u0026nbsp;\u003ctable id=\"Tab3\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eSequential decomposition of inequality effects by scenario.\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"6\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eScenario\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eYear\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eBurden\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eTransfer\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eNet\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eRatio\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSSP1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2030\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e+\u0026thinsp;0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e11.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2050\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e+\u0026thinsp;0.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-1.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSSP2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2030\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e+\u0026thinsp;0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e13.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2050\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e+\u0026thinsp;0.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSSP2 LUC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2030\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e+\u0026thinsp;0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e12.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2050\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e+\u0026thinsp;0.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.79\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSSP4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2030\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e+\u0026thinsp;0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2050\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e+\u0026thinsp;0.26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e+\u0026thinsp;0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNZ2050\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2030\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e+\u0026thinsp;0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2050\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e+\u0026thinsp;0.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-1.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e-0.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003ctfoot\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"6\"\u003e\u003cem\u003eNote: Burden\u0026thinsp;=\u0026thinsp;Gini(NDC) - Gini(Baseline); Transfer\u0026thinsp;=\u0026thinsp;Gini(cooperative) - Gini(NDC); Net\u0026thinsp;=\u0026thinsp;Gini(cooperative) - Gini(Baseline). All values in Gini points. Ratio is absolute value of transfer effect divided by burden effect. Negative values are progressive; positive values are regressive.\u003c/em\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tfoot\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\n \u003ch2\u003e4.5. Regional dynamics\u003c/h2\u003e\n \u003cp\u003eFigure \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e displays net financial flows by region group across scenarios, illustrating the trading patterns that underlie the aggregate Gini results. We aggregate GCAM\u0026rsquo;s 32 regions into six groups for exposition: Africa (four regions), China, India, Southeast Asia and Indonesia, Latin America, and Developed regions (USA, EU-15, EU-12, Japan, Canada, Australia/NZ).\u003c/p\u003e\n \u003cp\u003eUnder SSP2 NDC cooperative in 2050, Africa receives \u003cspan\u003e$\u003c/span\u003e103 billion in net transfers, India receives \u003cspan\u003e$\u003c/span\u003e340 billion, and China receives \u003cspan\u003e$\u003c/span\u003e517 billion. Developed regions pay \u003cspan\u003e$\u003c/span\u003e72\u0026nbsp;billion (USA), with substantial heterogeneity: some developed regions with cheap mitigation options (notably Canada) are net sellers. The overall pattern is progressive: lower-income regions sell, higher-income regions buy, and financial flows move from rich to poor.\u003c/p\u003e\n \u003cp\u003eUnder higher policy ambition, the regional pattern shifts further, revealing how stringency interacts with comparative advantage. Under net-zero 2050, Africa\u0026rsquo;s receipts increase to \u003cspan\u003e$\u003c/span\u003e894 billion as higher carbon prices raise the value of its mitigation exports. However, China reverses from receiving \u003cspan\u003e$\u003c/span\u003e517 billion under NDC cooperative to paying \u003cspan\u003e$\u003c/span\u003e948 billion under net-zero, a swing of \u003cspan\u003e$\u003c/span\u003e1,465 billion. This is the largest single-region shift across scenarios. Under stringent net-zero targets, China\u0026rsquo;s massive emissions base requires importing credits despite its domestic mitigation efforts. India also flips from receiving \u003cspan\u003e$\u003c/span\u003e340 billion to paying \u003cspan\u003e$\u003c/span\u003e65\u0026nbsp;billion. Southeast Asia becomes a net payer despite being in the Global South.\u003c/p\u003e\n \u003cp\u003eThese shifts reveal heterogeneity obscured by aggregate statistics. Under net-zero, the cleavage is not simply rich versus poor but regions with large remaining emissions versus those with cheap land-use mitigation. Brazil shifts from paying \u003cspan\u003e$\u003c/span\u003e320 billion (NDC) to receiving \u003cspan\u003e$\u003c/span\u003e109 billion (net-zero); Canada similarly reverses. This realignment complicates simple North-South narratives about carbon market distributive impacts.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\n \u003ch2\u003e4.6. Theil decomposition\u003c/h2\u003e\n \u003cp\u003eTo verify that Article 6 operates primarily through the between-region channel, we employ Theil index decomposition. The Theil T index is exactly additively decomposable into between-group and within-group components, unlike the Gini coefficient. We construct 320 observations (32 regions times 10 income deciles) using within-region income distributions from SSP-specific projections, then compute total, between-region, and within-region Theil indices under each scenario.\u003c/p\u003e\n \u003cp\u003eThe decomposition confirms that the transfer effect operates almost entirely through the between-region channel under convergent scenarios. Across SSP1, SSP2, and net-zero scenarios, the between-region component accounts for the dominant share of the total Theil change. The within-region component contributes a smaller offsetting effect because equal-per-capita transfers are slightly progressive within regions (giving the same absolute amount to each person benefits lower-income deciles proportionally more).\u003c/p\u003e\n \u003cp\u003eUnder SSP4 in 2050, the Theil decomposition reveals a different pattern. The between-region component shows a slight decrease in inequality (progressive), while the within-region component shows an increase (regressive). When poor regions pay for credits under the EPC assumption, the uniform outflow is regressive within those regions because equal absolute payments represent a larger share of income for poorer deciles. The within-region effect dominates in SSP4 by 2050 because the large buyer economies have high GDP weights in the within-region calculation.\u003c/p\u003e\n \u003cp\u003eThis nuance does not change the headline finding that SSP4 produces near-neutral between-region effects, but it reveals that the mechanism operates differently than in progressive scenarios. Under SSP1 and SSP2, transfers reduce between-region inequality and have small progressive within-region effects. Under SSP4, between-region effects are weak, and within-region effects in buyer regions become regressive. The overall implication is that SSP4 represents not just a weaker progressive outcome but a qualitatively different distributional pattern.\u003c/p\u003e\n \u003cp\u003eFull Theil decomposition tables are provided in the Supplementary Information.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\n \u003ch2\u003e4.7. Sensitivity to land-use sector scope\u003c/h2\u003e\n \u003cp\u003eWe test whether including land-use change (LUC) emissions in the carbon market alters distributional findings. Under SSP2 with LUC pricing, market size expands substantially: from \u003cspan\u003e$\u003c/span\u003e380 billion to \u003cspan\u003e$\u003c/span\u003e497 billion in 2030 (+\u0026thinsp;31 percent), from \u003cspan\u003e$\u003c/span\u003e953 billion to \u003cspan\u003e$\u003c/span\u003e1,471 billion in 2040 (+\u0026thinsp;55 percent), and from \u003cspan\u003e$\u003c/span\u003e1,672 billion to \u003cspan\u003e$\u003c/span\u003e1,952\u0026nbsp;billion in 2050 (+\u0026thinsp;17 percent). Despite these large changes in market size, the distributional impact is essentially neutral. The Gini coefficient differs by less than 0.1 points between LUC and non-LUC scenarios. This neutrality reflects heterogeneity within income groups: some lower-income regions gain from LUC pricing while others lose, producing no systematic progressive or regressive tilt. Full LUC scenario results are provided in the Supplementary Information.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"5. Discussion","content":"\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003e5.1. Efficiency versus equity in carbon markets\u003c/h2\u003e \u003cp\u003eOur findings speak to a longstanding tension in international climate policy between efficiency and equity objectives. Article 6 is primarily framed as an efficiency mechanism: by allowing abatement to occur where it is cheapest, cooperative implementation reduces the global cost of achieving a given set of targets (Edmonds et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Aldy et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The question we have examined is whether efficiency gains coincide with equity gains, or whether they come at the expense of distributional objectives.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eUnder convergent pathways and high-ambition scenarios, Article 6 produces both efficiency gains and progressive distributional outcomes, with transfer effects dominating policy burdens by a factor of three to six. Under divergent development (SSP4), the equity outcome weakens while efficiency gains persist. The market still reduces global mitigation costs by reallocating abatement, but the regional distribution of gains shifts. Some poor regions (notably Africa) lose comparative advantage as their mitigation costs rise relative to rich regions, becoming net buyers rather than sellers. However, because other large developing regions (China, India) retain seller status, the overall between-region effect remains near-neutral rather than becoming strongly regressive. This finding underscore that efficiency and equity are distinct but not necessarily in tension: aggregate cost savings can coincide with progressive outcomes under most development conditions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec25\" class=\"Section2\"\u003e \u003ch2\u003e5.2. Article 6 as climate finance\u003c/h2\u003e \u003cp\u003eSome analysts have positioned Article 6 as a channel for climate finance, noting that ITMO payments could mobilize substantial capital flows to developing countries (Keane and Fernandez, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; OECD, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Our results provide partial support for this framing. Under progressive scenarios (SSP1, SSP2, net-zero 2050), transfer volumes reach \u003cspan\u003e$\u003c/span\u003e700\u0026nbsp;billion to \u003cspan\u003e$\u003c/span\u003e1.77 trillion annually by 2050, with flows generally moving from higher-income to lower-income regions. For recipient regions such as Africa, transfers can reach 10 to 20 percent of regional GDP, magnitudes that exceed typical development assistance.\u003c/p\u003e \u003cp\u003eHowever, these are gross financial flows, not measures of net developmental benefit. Actual outcomes depend on authorization decisions, transaction costs, and domestic revenue allocation, none of which the model captures. Additional caveats apply. The UNCTAD Least Developed Countries Report (2024) cautions that carbon markets do not constitute or replace climate finance and do not adhere to the principle of common but differentiated responsibilities. Evidence from the Clean Development Mechanism suggests that LDCs gained limited capabilities from participation, with project activity concentrated in a small number of middle-income countries (UNCTAD, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Our modeling cannot assess whether ITMO revenues translate into development co-benefits or domestic capacity building; we track only aggregate financial flows.\u003c/p\u003e \u003cp\u003eMore fundamentally, the SSP4 results demonstrate that Article 6 transfers are not guaranteed to flow toward all developing regions. If development diverges and some poor regions lose comparative advantage in mitigation, they become buyers rather than sellers, and Article 6 functions as a drain on their resources even while benefiting other developing regions. This conditionality distinguishes Article 6 from needs-based climate finance mechanisms, which allocate resources according to vulnerability or development status rather than market position.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section2\"\u003e \u003ch2\u003e5.3. The comparative advantage mechanism\u003c/h2\u003e \u003cp\u003eThe SSP4 pattern illuminates the mechanism that determines the magnitude of Article 6\u0026rsquo;s progressive effect: comparative advantage in low-cost mitigation. This mechanism deserves explicit attention because it identifies the structural condition that policymakers would need to preserve or create if they want carbon markets to deliver broadly progressive outcomes.\u003c/p\u003e \u003cp\u003eIn SSP1 and SSP2, income convergence means that developing regions accumulate capital, deploy clean technologies, and develop institutional capacity for mitigation. Their marginal abatement costs remain low relative to developed regions, which face harder-to-abate residual emissions after exhausting cheap options. Developing regions thus have mitigation to sell, and developed regions have demand to buy. Financial transfers flow from rich to poor.\u003c/p\u003e \u003cp\u003eIn SSP4, income divergence means that some developing regions lack capital for clean technology deployment. Their energy systems remain carbon-intensive, and their mitigation costs rise as they face the dual burden of development needs and emissions constraints. Meanwhile, developed regions continue investing in low-carbon infrastructure, driving down their marginal costs. For regions like Africa, the comparative advantage flips: developed regions have cheaper mitigation options. These poor regions must buy credits to meet their NDCs because domestic abatement is too expensive.\u003c/p\u003e \u003cp\u003eAfrica\u0026rsquo;s position serves as the clearest diagnostic of this mechanism. When Africa has comparative advantage (SSP1, SSP2), it sells credits and receives transfers; Article 6 is strongly progressive. When Africa loses comparative advantage (SSP4), it buys credits and pays transfers; the progressive effect is substantially attenuated. The \u003cspan\u003e$\u003c/span\u003e1.1 trillion swing in Africa\u0026rsquo;s position between SSP1 and SSP4 in 2050 is a direct consequence of how development pathways affect mitigation cost structures.\u003c/p\u003e \u003cp\u003eIf policymakers want Article 6 to deliver broadly progressive outcomes, lower-income regions need access to capital, technology, and institutional support for clean energy deployment. If climate policy inadvertently concentrates mitigation capacity in wealthy regions while leaving others behind, the carbon market will produce differentiated outcomes across the Global South.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section2\"\u003e \u003ch2\u003e5.4. Comparison to domestic revenue recycling\u003c/h2\u003e \u003cp\u003eHow does the distributional effect of Article 6 compare to other climate policy mechanisms? The most relevant comparison is to domestic carbon pricing with equal-per-capita revenue recycling, which Budolfson et al. (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) and Emmerling et al. (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) have shown can substantially reduce inequality.\u003c/p\u003e \u003cp\u003eBudolfson et al. (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) find that equal-per-capita redistribution of carbon tax revenues within each region can reduce the global Gini coefficient by approximately 2 points under 2C-consistent mitigation pathways. This effect operates primarily through the within-region channel: lump-sum transfers benefit poor households more than rich households in proportional terms, compressing the income distribution within each country.\u003c/p\u003e \u003cp\u003eOur findings suggest that Article 6 achieves inequality reductions of 0.5 to 0.9 Gini points under progressive scenarios, roughly one-quarter to one-half the magnitude of domestic revenue recycling. Critically, Article 6 operates through a different channel: the between-region channel, as confirmed by our Theil decomposition showing the dominant share of the effect in the between-region component under convergent scenarios. Article 6 compresses the income distribution across regions by transferring resources from rich-region buyers to poor-region sellers, rather than compressing distributions within regions.\u003c/p\u003e \u003cp\u003eArticle 6 and domestic revenue recycling are complements, not substitutes. They address different dimensions of global inequality through different mechanisms: carbon pricing with progressive domestic recycling reduces within-region inequality, while Article 6 cooperation reduces between-region inequality. A comprehensive climate policy package could include both, with combined effects exceeding what either achieves alone. The governance challenges differ as well. Article 6 requires negotiated rules, accounting infrastructure, and credible commitment across sovereign parties; revenue recycling requires domestic political will to implement progressive transfers.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003e5.5. Design implications\u003c/h2\u003e \u003cp\u003eOur results have implications for Article 6 design and complementary policies. Market features that restrict trading volumes reduce both efficiency and distributional flows, creating tradeoffs between environmental integrity and equity co-benefits. The share-of-proceeds provision under Article 6.4, directing 5 percent to adaptation and mandating 2 percent cancellation, illustrates how equity considerations can be embedded in market design.\u003c/p\u003e \u003cp\u003eOur analysis does not model these provisions explicitly, but they illustrate the scope for layering redistributive elements onto efficiency-oriented mechanisms. The governance architecture of Article 6 also has distributional implications. Ahonen et al. (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) document the fragmentation between Article 6.2's bilateral approach and Article 6.4's centralized mechanism, noting tensions between bottom-up flexibility and top-down standardization. This fragmentation may affect which regions can effectively participate: countries lacking institutional capacity for bilateral negotiations may benefit from centralized mechanisms with standardized procedures, while countries with sophisticated climate diplomacy may prefer the flexibility of bilateral arrangements.\u003c/p\u003e \u003cp\u003eMore fundamentally, our findings suggest the most important determinant of Article 6's distributional outcome is not market design but the underlying development trajectory. Under SSP4, the weaker progressive outcome occurs because some poor regions lack comparative advantage. This points toward complementary policies outside the Article 6 framework: technology transfer, capacity building, concessional finance for clean energy deployment, and other measures that preserve or create mitigation capacity in developing regions. Article 6 can distribute the gains from trade, but it cannot create comparative advantage where none exists.\u003c/p\u003e \u003cp\u003eHigher policy ambition strengthens progressive outcomes under favorable development conditions. The net-zero 2050 pathway produces larger inequality reductions than NDC-level ambition because transfer volumes scale faster than mitigation burdens. This finding cuts against concerns that stringent climate targets disproportionately burden developing regions. Under SSP2 conditions, the opposite is true: higher ambition generates larger markets, higher carbon prices, and larger transfers to developing-region sellers. Climate ambition and equity can be complements rather than tradeoffs, but only if developing regions retain comparative advantage in low-cost mitigation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec29\" class=\"Section2\"\u003e \u003ch2\u003e5.6. Limitations\u003c/h2\u003e \u003cp\u003eSeveral caveats must be noted with our analysis. Our inequality measure captures between-region dispersion only; within-region effects depend on domestic revenue allocation, which varies across countries and political contexts. GCAM's 32-region structure aggregates substantial heterogeneity, so results for regions like \"Africa_Western\" should not be interpreted as country-level predictions. We also model a stylized global market with perfect competition and full participation; actual Article 6 implementation involves bilateral arrangements, varying readiness levels, and transaction costs. We do not model climate damages, so our inequality trajectories reflect mitigation policy effects only. Finally, our analysis is positive rather than normative; we quantify effects but do not assess whether the magnitudes satisfy principles of climate justice.\u003c/p\u003e \u003c/div\u003e"},{"header":"6. Conclusion","content":"\u003cp\u003eThis paper quantifies how Article 6 carbon markets affect between-region income inequality. Our central finding is that Article 6 reduces inequality under most scenarios, with the magnitude contingent on development conditions. Under SSP1, SSP2, and net-zero 2050, Article 6 produces progressive outcomes. Lower-income regions sell ITMOs and receive transfers of \u003cspan\u003e$\u003c/span\u003e700\u0026nbsp;billion to \u003cspan\u003e$\u003c/span\u003e1.77 trillion annually by 2050. The population-weighted global Gini falls by 0.5 to 0.9 points. Higher ambition strengthens this effect: net-zero scenarios produce larger inequality reductions than NDC-level ambition because transfer volumes scale faster than mitigation burdens. Under SSP4, the progressive effect weakens to near-neutral (+\u0026thinsp;0.03 Gini points by 2050). Income divergence erodes some developing regions' comparative advantage; Africa shifts from seller to buyer, paying \u003cspan\u003e$\u003c/span\u003e554\u0026nbsp;billion rather than receiving \u003cspan\u003e$\u003c/span\u003e577\u0026nbsp;billion as under SSP1. However, China and India retain seller status, partially offsetting Africa's reversal. The transfer effect remains slightly progressive but insufficient to offset the policy burden. Our decomposition reveals that financial transfers dominate policy burden effects by a factor of three to six. Theil decomposition confirms that 92 to 100 percent of the transfer effect operates through the between-region channel under convergent scenarios. Article 6 thus offers a distinct channel for reducing global inequality that complements domestic revenue recycling, which operates through the within-region channel. The progressive character of Article 6 depends on developing regions maintaining comparative advantage in low-cost mitigation. Policies that concentrate mitigation capacity in wealthy regions would undermine conditions for progressive outcomes. Complementary mechanisms, particularly technology transfer and adaptation finance, may be necessary to preserve conditions under which carbon markets contribute broadly to global equity. Whether Article 6 ultimately reduces or increases global inequality depends on choices about development pathways, technology access, and the institutional architecture of international climate cooperation.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAhonen H-M, Kessler J, Michaelowa A, Espelage A, Hoch S (2022) Governance of fragmented compliance and voluntary carbon markets under the Paris Agreement. Politics Gov 10(1):235\u0026ndash;246\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAldy JE, Pizer WA, Tavoni M, Reis LA, Akimoto K, Blanford G et al (2021) How much could Article 6 enhance nationally determined contribution ambition toward Paris Agreement goals through economic efficiency? Clim Change Econ 12(2):2150007\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBourguignon F (1979) Decomposable income inequality measures. Econometrica 47(4):901\u0026ndash;920\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBudolfson M, Dennig F, Fleurbaey M, Siebert A, Socolow RH (2021) Climate action with revenue recycling has benefits for poverty, inequality and well-being. Nat Clim Change 11:1111\u0026ndash;1116\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCalvin K, Bond-Lamberty B, Clarke L, Edmonds J, Eom J et al (2017) The SSP4: A world of deepening inequality. Glob Environ Change 42:284\u0026ndash;296\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCalvin K, Patel P, Clarke L, Asrar G, Bond-Lamberty B et al (2019) GCAM v5.1: Representing the linkages between energy, water, land, climate, and economic systems. Geosci Model Dev 12:677\u0026ndash;698\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChepeliev M, Rodarte O, van der Mensbrugghe I, D (2021) Distributional impacts of carbon pricing policies under the Paris Agreement: Inter and intra-regional perspectives. Energy Econ 101:105408\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEdmonds JA, Forrister D, Clarke L, de Clara S, Munnings C (2019) The Economic Potential of Article 6 of the Paris Agreement and Implementation Challenges. World Bank, Washington, DC\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEdmonds JA, Yu S, McJeon H, Forrister D, Aldy J et al (2021) How much could Article 6 enhance NDC ambition? Clim Change Econ 12(2):2150007\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEmmerling J, Andreoni P, Charalampidis I, Dasgupta S, Dennig F et al (2024) A multi-model assessment of inequality and climate change. Nat Clim Change 14:1254\u0026ndash;1260\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGazzotti P, Emmerling J, Marangoni G, Castelletti A, van der Wijst K-I, Tavoni M (2021) Persistent inequality in economically optimal climate policies. Nat Commun 12:3421\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIETA (2023) Modelling the Economics of Article 6: A Capstone Report. International Emissions Trading Association, Geneva\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIPCC (2022) Climate Change 2022: Mitigation of Climate Change. Contribution of Working Group III to the Sixth Assessment Report. Cambridge University Press, Cambridge\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKeane J, Fernandez M (2025) How carbon markets can deliver climate investment for developing nations. ODI Insights. Overseas Development Institute, London\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLakner C, Milanovic B (2016) Global income distribution: From the fall of the Berlin Wall to the Great Recession. World Bank Economic Rev 30(2):203\u0026ndash;232\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLang SF, Dennig F, Budolfson M, Errickson F, Tavoni M (2025) Within-country inequality and the shaping of a just global climate policy. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e 122: e2505239122\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMichaelowa A (2019) Evolution of international carbon markets: lessons for the Paris Agreement. WIREs Clim Change 10(6):e613\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNordhaus W (2015) Climate clubs: Overcoming free-riding in international climate policy. Am Econ Rev 105(4):1339\u0026ndash;1370\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOECD (2023) The Interplay Between Voluntary and Compliance Carbon Markets. OECD Publishing, Paris\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePiris-Cabezas P, Lubowski R, Leslie G (2023) Estimating the power of international carbon markets to increase global climate ambition. World Dev 167:106243\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRao ND, Sauer P, Gidden M, Riahi K (2019) Income inequality projections for the Shared Socioeconomic Pathways (SSPs). Futures 105:27\u0026ndash;39\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRiahi K, van Vuuren DP, Kriegler E, Edmonds J, O\u0026rsquo;Neill BC et al (2017) The Shared Socioeconomic Pathways and their energy, land use, and greenhouse gas emissions implications: An overview. Glob Environ Change 42:153\u0026ndash;168\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchneider L, Cludius J, La Hoz Theuer S (2018) Accounting for the linking of emissions trading systems under Article 6.2 of the Paris Agreement. ICAP/Oeko-Institut, Berlin\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchneider L, La Hoz Theuer S (2019) Environmental integrity of international carbon market mechanisms under the Paris Agreement. Clim Policy 19(3):386\u0026ndash;400\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShorrocks AF (1980) The class of additively decomposable inequality measures. Econometrica 48(3):613\u0026ndash;625\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUNCTAD (2024) The Least Developed Countries Report 2024: Leveraging Carbon Markets for Development. Geneva: United Nations Conference on Trade and Development\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUNFCCC (2021) Decision 2/CMA.3: Guidance on cooperative approaches referred to in Article 6, paragraph 2, of the Paris Agreement. Adopted at CMA3, Glasgow, November 2021. FCCC/PA/CMA\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e/2021/10/Add.1\u003c/span\u003e\u003cspan address=\"http:///2021/10/Add.1\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUNFCCC (2024) Article 6 of the Paris Agreement. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://unfccc.int/process-and-meetings/the-paris-agreement/article-6\u003c/span\u003e\u003cspan address=\"https://unfccc.int/process-and-meetings/the-paris-agreement/article-6\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUNFCCC (2025a) Article 6.2 Reference Manual for the Accounting, Reporting and Review of Cooperative Approaches. Version 3. UNFCCC Secretariat, Bonn\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUNFCCC (2025b) Synthesis report on nationally determined contributions under the Paris Agreement. Synthesis of NDCs submitted by 30 September 2025. FCCC/PA/CMA\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e/2025/8\u003c/span\u003e\u003cspan address=\"http:///2025/8\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"University of Maryland, College Park","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Article 6, internationally transferred mitigation outcomes, carbon markets, NDCs, integrated assessment modeling, inequality, Gini coefficient, Theil decomposition","lastPublishedDoi":"10.21203/rs.3.rs-8793231/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8793231/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eWe assess whether Article 6 carbon market transfers reduce between-region income inequality. Using GCAM, we model internationally transferred mitigation outcomes (ITMOs) across 32 regions under three socioeconomic baselines and a more ambitious net-zero 2050 pathway.\u003c/p\u003e \u003cp\u003eA maximalist implementation of Article 6 reduces inequality under most scenarios, with Gini reductions of 0.5 to 0.9 points by 2050 under SSP1, SSP2, and net-zero pathways. Under SSP4, where income divergence erodes developing regions' comparative advantage in low-cost mitigation, the progressive effect weakens to near-neutral. The financial transfer effect dominates the policy burden effect across scenarios, operating primarily through the between-region channel. Africa's position as net seller or buyer serves as a diagnostic: swinging between favorable and unfavorable development pathways.\u003c/p\u003e \u003cp\u003eThese findings suggest Article 6 can reduce global inequality, but outcomes depend on development trajectories rather than the market mechanism itself. Ensuring broadly progressive outcomes requires complementary policies that preserve developing regions' comparative advantage, including technology transfer and access to clean energy finance.\u003c/p\u003e","manuscriptTitle":"When Do Carbon Markets Reduce Inequality? Article 6 Transfers Under Alternative Futures","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-06 11:55:51","doi":"10.21203/rs.3.rs-8793231/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"025700d0-5648-459d-b019-e2f3429ee049","owner":[],"postedDate":"February 6th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":62450767,"name":"Climate Analysis and Modeling"},{"id":62450768,"name":"International Relations"},{"id":62450769,"name":"Development Economics"}],"tags":[],"updatedAt":"2026-02-06T11:55:52+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-06 11:55:51","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8793231","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8793231","identity":"rs-8793231","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}