Approaching national climate targets in China considering the challenge of regional inequality

The study addresses how a nation can meet its climate pledges cost-effectively when substantial regional heterogeneity exists in socioeconomic development, technology, energy structure, and resources. Uniform or overly aggressive regional targets can exacerbate inequality and impose large economic losses, threatening national climate objectives. The research proposes collaborative, economically optimal regional mitigation pathways that both satisfy China’s national climate goals (carbon peak by around 2030 and carbon neutrality by 2060) and balance regional equity and economic outcomes. It highlights the need for interregional strategies that account for differences in maturity and capability to reduce emissions, moving beyond uniform or independently set provincial targets.

Two main strands of literature are reviewed. First, effort-sharing and allocation studies distribute national emission reductions or intensity targets across regions based on equity principles, providing static fair quotas but not dynamic, economically optimal regional pathways. Second, top-down and bottom-up energy-economy models derive multi-regional emission pathways considering heterogeneity but typically optimize for technological cost minimization or market equilibrium, not explicitly maximizing national and regional economic benefits while incorporating equity. Prior work underscores that premature or aggressive targets in less-prepared regions can hinder growth and exacerbate inequality, emphasizing the need for collaborative strategies integrating economic efficiency and fairness.

The study integrates national energy-system optimization with provincial pathway optimization under equity considerations. - National pathway (C³IAM/NET): A bottom-up energy technology optimization model developed by CEEP-BIT is used to derive China’s carbon peak and neutrality pathway under multiple socioeconomic demand trajectories and carbon sink assumptions. It minimizes total system cost across 817 technologies spanning primary energy, power, heating, industry (iron and steel, cement, nonferrous, chemicals), buildings, transport, and other sectors. Outputs include national fossil and non-fossil energy consumption by fuel and fossil-fuel CO2 emissions trajectories, serving as constraints for provincial optimization. - Regional maturity index: To reflect heterogeneity and equity, a provincial carbon peak maturity index is constructed using the TOPSIS method, with indicators such as per capita GDP, energy intensity, coal share, and installed renewable capacity. Provinces are grouped into high-, middle-, and low-maturity categories. In the collaborative optimization (COP) scenario, a carbon peak sequence constraint enforces earlier peaks for higher-maturity groups. - Multi-regional Collaborative Optimization of Emission Pathway (Mr. COEP): A nonlinear optimization model determines provincial energy consumption by fuel and CO2 pathways under national constraints, maximizing national GDP while keeping regional GDP within acceptable bounds. Decision variables are provincial GDP and energy consumption by fuel (coal, oil, gas, non-fossil). Constraints include: (i) national annual energy consumption by fuel equals the sum across provinces; (ii) national fossil CO2 emissions equal the sum of provincial emissions (industrial process emissions excluded sub-nationally); (iii) regional GDP growth bounds; (iv) provincial energy mix bounds where applicable; (v) carbon intensity bounds; (vi) monotonic peaking behavior; (vii) carbon peak sequencing via the maturity index. Emissions are computed using national inventory emission factors (coal 2.66 tCO2/tce, oil 1.73, gas 1.56). Economic linkages operate via shared national energy and emissions constraints and the GDP–energy relationship. - Nonlinear GDP–energy relationship: An extended Kuznets-type nonlinear regression relates provincial energy consumption to GDP (including squared and cubic terms) and demographic/structural drivers (population, urbanization rate, secondary industry share). Heterogeneity in energy intensity, coal share, and renewable capacity is captured through the maturity index and constraints. Estimated relationships for each province are embedded in Mr. COEP to endogenize energy–GDP interactions over time. - Scenarios: Four main scenarios are optimized under the national pathway constraints: (1) AP30—All provinces peak before 2030 (uniform target); (2) FCT—Following currently proposed provincial targets (peak timing and energy goals where announced; others unconstrained); (3) EIP30—Only energy-intensive provinces (about 70% of national emissions in 2020) must peak before 2030; (4) COP—Collaborative optimization with maturity-based peak sequencing to maximize national and regional economic benefits. - Data and solution: Provincial energy data from INEMS (CEEP-BIT), GDP and projections (converted to 2020 constant prices), population and urbanization projections, industrial structure trends, and official provincial policy targets. C³IAM/NET provides national energy/emissions constraints. Nonlinear regression is estimated in SPSS; Mr. COEP solved via GAMS (CONOPT). Tibet, Hong Kong, Macao, and Taiwan are excluded due to data limitations (30 provinces analyzed).

National pathway and energy structure: - Under a medium GDP growth and 1 Gt natural carbon sink in 2060, China peaks CO2 before 2029 at ~12.2 Gt (90.2% fossil, remainder industrial process). Fossil energy peaks at ~4.8 Gtce. 2025–2035 forms a plateau (avg emissions growth ~0.2% pre-2029). Post-peak declines average ~1.1%/yr during 2030–2035; accelerate to ~4.1%/yr during 2035–2050; and >16%/yr during 2050–2060. Non-fossil energy share rises from 15.9% (2020) to 80.1% (2060). In 2030, coal/oil/gas shares are 44.2%/17.3%/11.1%. Economic impacts across regional strategies (2023–2060): - AP30 (uniform peak by 2030): cumulative national GDP ≈ 10,963 trillion RMB (2020 prices). - FCT (follow current independent targets): +0.63% vs AP30. - EIP30 (only energy-intensive provinces peak by 2030): +0.91% vs AP30. - COP (collaborative, equity-aware): +1.54% vs AP30 (largest gains). Thus, collaboration can avoid up to 1.54% of cumulative GDP losses en route to neutrality. Provincial economic outcomes: - COP vs AP30: 27/30 provinces gain GDP (0.34–20.87 trillion RMB; average 6.28 trillion). Three provinces incur small losses to meet national optimum and targets: Sichuan (−0.67T), Hubei (−0.29T), Jilin (−0.06T), each ≈0.05–0.12% of their cumulative GDP under COP. - Largest provincial gains under COP vs AP30: Guangdong (+20.9T), Jiangsu (+18.9T), Zhejiang (+11.7T), Shanghai (+11.6T). - COP vs FCT: national cumulative GDP higher by 101.2T RMB; 27 provinces would otherwise face losses under FCT (avg −3.75T), with only Guangxi, Gansu, Xinjiang seeing small gains (≈0.1–0.5% of their provincial cumulative GDP). Collaborative provincial pathways (COP): - Early peakers by 2027: Shanghai, Zhejiang, Tianjin, Fujian, Jiangsu, Guangdong, Qinghai, Hubei, Sichuan, Yunnan. Five-year CO2 intensity reduction rates rise from ~21% (2020–2025) to ~34% (2035–2040); slight dip (~0.4 pp) in 2030–2035 vs 2025–2030 due to plateau and slower GDP growth. - On-time peakers (by 2030): Jilin, Hunan, Guangxi, Anhui, Jiangxi, Shandong, Henan, Chongqing, Hainan, Shaanxi, Guizhou, Gansu. Five-year CO2 intensity reductions increase from ~17.7% (2020–2025) to ~31.8% (2035–2040). - Later peakers (after 2030 but no later than 2034): Hebei, Shanxi, Inner Mongolia, Liaoning, Heilongjiang, Ningxia, Xinjiang. Five-year intensity reductions: ~9.0% (2020–2025), 11.8% (2025–2030), 15.4% (2030–2035), accelerating to ~24.8% (2035–2040). Energy structure guidance: - High-renewable provinces by 2040: non-fossil shares — Qinghai ~78%, Yunnan ~73%, Sichuan ~67%. - Rapid non-fossil promotion: Zhejiang from ~28% (2025) to >55% (2040); Anhui from ~13% (2025) to >28% (2040). - Energy security and supply roles: coal remains significant through 2040; energy-exporting provinces have minimum coal shares by 2030 (e.g., Shanxi ≥80%, Inner Mongolia ≥71%, Shaanxi ≥57%). Policy alignment suggestions: - Earlier peaks for 10 provinces (e.g., Qinghai 2026; Guangxi 2028; Hunan 2028; Henan 2029 — each 1–4 years earlier than current plans). - Later peaks (but ≤2034) to avoid premature economic losses: Hebei, Ningxia, Hainan, Inner Mongolia, Shandong, Heilongjiang, Liaoning. - For provinces without clear targets: Xinjiang peak after 2030 but ≤2034; Sichuan, Yunnan, Fujian, Jiangsu, Hubei peak by 2027; Anhui, Gansu by 2030. - Energy-share target adjustments (examples): Hunan to 27% non-fossil by 2030 (from 25%); Guangxi to 37% (from 35%); Ningxia to 16% (2025) and 22% (2030) non-fossil; further increases recommended in high-renewable provinces (e.g., Qinghai, Sichuan).

The findings demonstrate that accounting for regional heterogeneity and sequencing provincial peak years by maturity substantially improves national and local economic outcomes while still meeting China’s climate goals. The collaborative optimization (COP) outperforms uniform and independently set strategies by minimizing premature abatement in less-prepared provinces and leveraging advanced, low-intensity, or renewable-rich regions to peak earlier. This approach reduces inequality, protects growth in vulnerable regions, and increases overall GDP. A sensitivity exploration without peak-time constraints suggests even higher national economic gains but requires some less mature provinces to peak earlier and some mature, renewable-rich provinces to peak later—configurations that may be infeasible or inequitable in practice—underscoring the value of the maturity-index-based sequencing. Overall, cross-provincial coordination is essential for cost-effective national decarbonization and balanced regional development.

This study integrates a national energy-technology optimization (C³IAM/NET) with a novel multi-regional optimization (Mr. COEP) constrained by a maturity-based peak sequence to propose equitable, cost-effective provincial pathways toward China’s carbon peak and neutrality goals. Key contributions include: (i) identifying an optimal national pathway peaking before 2029 with deep post-2035 declines and a long-run shift to >80% non-fossil energy by 2060; (ii) quantifying that collaborative regional strategies can avoid up to 1.54% of cumulative GDP losses (2023–2060) relative to uniform peaking, with most provinces gaining economically; and (iii) providing province-specific timelines for peaking, intensity reductions, and energy structure adjustments, alongside policy-alignment suggestions that improve equity and feasibility. Future work could extend the framework to other countries with high regional heterogeneity, incorporate more explicit interregional economic linkages and trade, include industrial process emissions at the subnational level, explore uncertainty in carbon sinks and technology costs, and assess co-benefits (e.g., air quality, health) to further inform equitable regional decarbonization.

- Coverage: Analysis includes 30 mainland provinces; Tibet, Hong Kong, Macao, and Taiwan are excluded due to data limitations. - Emissions scope: Provincial optimization excludes industrial process CO2 emissions (only fossil combustion modeled subnationally), while national peaking includes process emissions. - Assumptions: Results depend on assumed socioeconomic growth paths and a 1 Gt natural carbon sink in 2060; alternative assumptions would shift pathways. - Economic linkages: Interprovincial economic interactions are represented indirectly via national energy and emissions constraints and the GDP–energy relationship, not through a full multi-regional input–output or CGE framework. - Model structure and parameters: Extended Kuznets regressions and maturity index weighting influence outcomes; uncertainty in parameter estimates and indicator weights may affect provincial trajectories. - Energy security constraints: Minimum coal shares for some energy-supplying provinces reflect policy considerations that may evolve, affecting optimized energy mixes.