A cost-effective climate mitigation pathway for China with co-benefits for sustainability

China has committed to the UN 2030 Agenda and SDGs, integrating them into development strategies while pledging to peak carbon emissions before 2030 and reach carbon neutrality before 2060. Numerous climate mitigation pathways (CMPs) for China exist, differing in reliance on energy efficiency, energy mix transformation, and negative emissions. However, prior work often assessed only subsets of SDGs or limited indicators. This study asks: (i) What are the impacts of CMPs on achieving all 17 SDGs? (ii) How can a cost-effective climate pathway be designed to maximize sustainability co-benefits alongside emissions reductions? Using a China-tailored integrated system dynamics model, the study evaluates multiple CMPs’ impacts on all SDGs and identifies a cost-effective pathway that balances sustainability outcomes, CO2 mitigation, and costs.

The authors review work linking climate mitigation and SDGs, noting many studies focus on specific SDGs (e.g., poverty/inequality, food, health, water, economic cost, biodiversity) or use IAMs to assess multi-SDG interactions at global and regional scales. Examples include analyses of synergies and trade-offs between mitigation and sustainable energy objectives, the consequences of 1.5 °C-consistent mitigation for SDGs, and sectoral spillovers from power transitions. At EU and China scales, prior studies assessed decarbonization pathways’ SDG outcomes and SDG-related climate policy effects. Despite progress, most studies consider limited SDGs or indicators, leaving gaps in understanding comprehensive impacts across all 17 SDGs and in designing cost-effective CMPs that maximize sustainability co-benefits. Supplementary Tables synthesize broader SDG analyses and detailed mitigation–SDG relationships.

The study develops the iSDG-Climate-China model, an extension of the Millennium Institute’s core iSDG system dynamics model, calibrated to China’s historical data. Enhancements include: (1) additional cross-sectoral links for climate mitigation, such as CCS impacts on CO2 emissions, biomass yields, water withdrawal, and costs, and PM2.5 concentration impacts on premature mortality; (2) an expanded SDG evaluation framework with 17 goals, 51 targets, and 98 indicators; and (3) China-specific calibration via structural and behavioral validation against historical series. The model covers 30 interacting sectors (economy, society, environment), simulating energy consumption, electricity generation, primary energy supply, emissions, water, health, materials, and government finance, among others. Policies are implemented via parameter changes: energy efficiency (emission factors, material use efficiency, government expenditure), energy mix (generation mix, electrification, end-use sector policies), and negative emissions (LULUCF afforestation and CCS deployment). Simulations are conducted in Stella Architect (v3.0.0). Scenario framework: 189 CMPs to 2060 are evaluated—9 original CMPs compiled from authoritative studies/reports (each with coherent policies for energy efficiency, energy mix, and negative emissions) and 180 “bundled CMPs” formed by independently combining policy clusters for energy efficiency (Mature), energy mix (9 levels from the 9 CMPs), LULUCF (BAU, 1.05, 1.3, 1.65 GtCO2 removal), and CCS (BAU, 0.94, 2, 3.22, 4.2 GtCO2 removal). Where original CMPs ended in 2050, trajectories were linearly extended to 2060. Socioeconomic assumptions use tailored SSPs (tSSPs) removing overlaps with mitigation policies and applied across uncertainty analysis. SDG scoring: 98 indicators are normalized 0–100 using established bounds and aggregation via arithmetic means to targets, goals, and an overall SDG score. A composite SDG-emissions-cost index is computed by rescaling overall SDG score, CO2 emissions, and mitigation costs to 0–100 and averaging, rewarding higher SDG performance, lower emissions, and lower mitigation costs. Calibration and validation include goodness-of-fit across 56 SDG indicators and 16 macro-variables with multiple error metrics. Data and code are provided via Zenodo; the core iSDG model is available upon request.

- Implementing the 9 original CMPs (without other SDG interventions) increased China’s overall SDG score to 71.17–72.14 by 2030 and 77.08–79.61 by 2060 (from 64.70 in 2022), averaging +3.91 (2030) and +7.99 (2060) points over the Reference pathway. - Policy contributions to overall SDG gains by 2060 (vs Reference) averaged: energy efficiency +2.47 points, energy mix +3.73, negative emissions +1.79. The energy mix share of gains ranged 41.2%–60.9% across CMPs; negative emissions 0%–30.4%. - Individual SDG improvements by 2060 (vs Reference) averaged: SDG6 +5.43, SDG7 +15.69, SDG12 +32.22, SDG13 +51.13, SDG15 +8.60; SDG1 +10.47, SDG2 +8.70, SDG9 +7.63, SDG11 +20.02. Trade-offs included declines in SDG10 (−4.13), SDG14 (−9.14), and SDG17 (−15.94) points by 2060. - Energy-mix-driven differences were prominent: the Below 2 °C and NET-led CMPs were the best and worst for SDG7 by 2060 (+18.46 and +10.77 vs Reference), reflecting renewable shares of 83.3% and 35.3% in primary energy by 2060. CMPs without negative emissions (Below 2 °C, RE-led) failed to achieve 2060 carbon neutrality (1.25 and 0.064 GtCO2 remaining in 2060), yielding lower SDG12/13 scores than other CMPs. - Government deficits increased to 7.1%–8.5% of GDP under the 9 CMPs (vs 4.6% in Reference), contributing to lower SDG17 performance (−13.85 to −17.78 points). - Bundled CMPs outperformed originals in SDG synergies. The best SDG performer (EEmature, EMRE-led, LULUCF1.65, CCS2) improved overall SDG score by +9.33 points vs Reference by 2060. - Extremes among all CMPs: best CO2 outcome (EEmature, EM1.5 °C, LULUCF1.65, CCS4.2) achieved −4.59 GtCO2 in 2060 and 14.95 Gt cumulative reductions (2022–2060); lowest mitigation cost (EEmature, EMNET-led, LULUCFBAU, CCSBAU) averaged $0.09 trillion/yr but had poor emissions performance (3.22 Gt CO2 in 2060). - Cost-effective pathway: The EEmature, EMRE-led, LULUCF1.65, CCS2 bundled CMP achieved the highest SDG-emissions-cost index of 76.15 by 2060 (19.90 points above the 9-CMP average), with overall SDG score 80.08, CO2 emissions reported at 2.22 Gt in 2060, and average mitigation cost $0.36 trillion/yr (2022–2060). It outperformed most original CMPs across individual SDGs by 2060 except SDG6 (due to higher long-term water withdrawals). - Uncertainty analysis across five tSSPs and energy-efficiency levels confirmed robustness: Updated NDC to Carbon Neutrality remained the best among the 9 originals (overall SDG up to 82.31 by 2060), while the cost-effective CMP surpassed all originals (overall SDG up to 82.84). The cost-effective CMP retained the top SDG-emissions-cost index under tSSP1–tSSP5 (up to 85.76, 85.94, 82.56, 84.70, 85.30).

The integrated assessment shows mitigation policies generate strong net synergies with SDGs, notably enhancing environmental goals (SDGs 6–7, 12–15) by cutting fossil fuel use, emissions, and expanding protection/afforestation; reducing disaster vulnerability and improving resilience (SDG1); advancing sustainable agriculture (SDG2); lowering PM2.5-related mortality (SDG3); and boosting sustainable infrastructure and urban health (SDGs 9, 11). These cross-sector benefits highlight the value of policy coherence and interdepartmental coordination to align climate action with broader development. Trade-offs arise for inequality (SDG10), life below water (SDG14), and partnerships/finance (SDG17), driven by mitigation costs, fiscal pressures, and reduced spending on water/marine protection. Targeted measures—job creation in low-income regions via forest carbon and renewables, enhanced social protection, regional development strategies, expanded marine protected areas, improved wastewater control, and strengthened climate finance (including carbon markets and pricing)—can help mitigate these trade-offs. The cost-effective CMP (EEmature, EMRE-led, LULUCF1.65, CCS2) best balances sustainability, emissions, and cost, but implementation faces challenges: scaling energy-efficiency investments and innovation, coordinating supply and end-use sectors for a renewable-led energy mix, accelerating CCS (including BECCS) deployment while managing water and land resource conflicts with agriculture and renewables. Recent coal capacity expansion underscores the need to align short-term decisions with long-term neutrality and the identified cost-effective pathway.

By extending and calibrating the iSDG model to China (iSDG-Climate-China), the study systematically quantified co-benefits and trade-offs between climate mitigation pathways and all 17 SDGs, and searched a large policy space (9 original and 180 bundled CMPs) to identify a cost-effective pathway. The selected bundled CMP (EEmature, EMRE-led, LULUCF1.65, CCS2) maximizes an integrated SDG-emissions-cost index, substantially improves overall SDG outcomes by 2060, and controls costs while delivering deep emissions reductions. The findings support policy designs that enhance synergies and address trade-offs (notably SDGs 10, 14, and 17) through targeted social, environmental, and financing measures. The modeling framework can be adapted to other countries, and future research could incorporate additional policy instruments (e.g., carbon pricing, international climate finance), broader mitigation cost components, and diversified carbon sinks, and extend analysis to global scales using tools such as CMIP6 and FeliX.

- Scenario coverage: CMP scenarios omit certain policy instruments and uncertainties (e.g., carbon trading schemes, carbon pricing dynamics, international climate finance). - Cost estimation scope: Mitigation cost accounting focuses on key technologies (renewables, negative emissions) and does not fully encompass all low-carbon energy, energy access, energy efficiency, and carbon removal cost elements. - Carbon sinks: Only forest (LULUCF) sinks are modeled; other sinks (soil, ocean) are not included. - Model nature: The iSDG-Climate-China model is a what-if system dynamics tool capturing nonlinear interactions, not a precise predictive model; results should be interpreted as scenario-based projections subject to structural and data limitations.