Carbon Finance Rises Across Western China, But Gaps Between Cities Persist

Carbon finance development across Western China climbed steadily between 2014 and 2024, but a new study shows the gains are far from evenly spread, with a handful of metropolitan hubs pulling ahead while many resource-dependent cities remain locked into carbon-intensive paths. The research, published in the journal Digital Economy and Sustainable Development, evaluated 79 prefecture-level cities over an eleven-year panel and mapped both the pace and the geography of the region’s emerging carbon finance systems.

The study, led by Liujuan Yu, Qiqing Zhou, and Wen Gao, built a comprehensive evaluation index using the Pressure–State–Response (PSR) framework, a structure originally developed by Canadian statisticians in the 1970s and later adopted by the OECD for environmental reporting. In this adaptation, the pressure dimension captures the constraints that high-carbon development and industrial transformation place on carbon finance, the state dimension reflects regional economic, environmental, and financial conditions, and the response dimension measures how governments, markets, and social actors adapt to promote low-carbon transition. Fourteen indicators were weighted with the entropy weight method, an objective technique that derives weights from the data itself rather than from researcher judgment.

Western China presents a distinctive test case. The region, defined here as the twelve provincial-level areas covered by China’s Western Development Strategy, is simultaneously a major energy supplier and a national ecological security barrier. It remains heavily dependent on coal and energy-intensive industries, with many resource-based cities concentrated in the northwest, yet it also holds abundant forest resources that serve as natural carbon sinks. The authors argue that development mechanisms observed in China’s more mature eastern markets may not transfer directly to this context, since most carbon trading pilots were concentrated in economically developed eastern provinces and provincial capitals.

The results show that the average carbon finance development level rose consistently across the study period, with what the authors describe as stable growth. Xi’an, Chongqing, and Chengdu topped the regional rankings, while the fastest average annual growth occurred in Ulanqab, Guyuan, and Beihai. Overall, Southwestern China outperformed Northwestern China: 38 prefecture-level cities scored above the regional average, 24 of them in the southwest and 14 in the northwest. The authors attribute the southwest’s advantage partly to earlier socioeconomic development in Sichuan Province and Chongqing Municipality, with Chongqing among the earliest western municipalities to pilot a carbon emissions trading market, generating spillover effects for neighboring provinces.

The three PSR dimensions each showed fluctuating upward trends, but with important nuances. Because most pressure-dimension indicators are negatively oriented, a rising pressure score signals easing environmental and industrial constraints, and the data suggest those constraints gradually loosened over the decade. Agricultural cities such as Dingxi and Wuwei in Gansu Province, along with tourism-oriented Guilin, recorded comparatively low development pressure, while Yulin, a national coal production base, and the energy-industrial cities of Jinchang and Jiayuguan faced the heaviest pressure. Notably, cities that absorbed large-scale industrial transfers from central and eastern China, such as Ordos and Wuzhou, saw rising industrialization and energy consumption that increased their development pressure, suggesting that industrial relocation imposes environmental costs unless paired with genuine upgrading.

The state dimension revealed a widening divergence. Provincial capitals started higher and improved faster, while peripheral cities lagged, indicating that financial resources in Western China continued concentrating in central cities rather than dispersing. The authors describe a contradictory configuration: economically strong regions face greater development pressure, while cities with rich ecological endowments tend to have weaker economies. In the response dimension, market activity and policy support did not always align. Tongchuan and Longnan showed relatively active carbon finance markets but insufficient policy backing, whereas Hohhot and Wuzhou had strong policy support but weak market activity. The study concludes that policy assistance alone is insufficient, and that market tools, particularly green credit, are the true driving force of carbon finance development.

Spatial analysis added a crucial layer. Global Moran’s I statistics, computed on a spatial adjacency matrix, were positive and significant throughout the period, confirming strong positive spatial autocorrelation: cities with high carbon finance development cluster near other high-performing cities, and neighboring regions exert positive spillover effects through the movement of talent, capital, and policy learning. Local Moran analysis identified clear high-high and low-low clusters, with Chongqing and Xi’an anchoring high-value zones in 2014 while northern resource-based cities such as Wuhai and Baotou formed low-low clusters described as path locks. Kernel density estimation showed the distribution shifting rightward over time, transitioning from a double-peak to a single-peak pattern, which the authors interpret as a weakening of spatial polarization after the national carbon market accelerated following 2019.

To identify what drives these patterns, the researchers applied a Geographical Detector model, a statistical method designed to analyze spatial heterogeneity without assuming linear relationships. The results were striking. Green innovation, measured by granted green patents, had the highest explanatory power and remained significant across all years, indicating that carbon finance in Western China depends less on financial expansion than on green technology capability. Green consumption, proxied by the share of new energy vehicles in total civilian vehicle ownership, ranked second, operating through both demand for carbon credits and investor appetite for carbon assets. The digital economy, captured by the Digital Inclusive Finance Index, grew markedly in importance between 2018 and 2024, suggesting the region’s carbon finance logic has become digitally driven, with digital tools reducing transaction costs and weakening spatial clustering effects.

Two factors told more cautionary tales. The explanatory power of trade openness weakened over time, which the authors link to shifting trade structures, including escalating China–United States trade frictions from 2018 and the pandemic-era contraction of global capital markets, pushing the carbon finance market toward endogenous rather than externally financed development. Environmental regulation, measured by the proportion of environmental-protection language in local government work reports, was not a significant independent explanatory factor in 2024. However, interaction detection revealed that regulation’s influence strengthens substantially when combined with green innovation, digital infrastructure, and green consumption, with pairwise interactions showing mainly nonlinear enhancement and bifactorial enhancement effects and no independent relationships. In 2018, the interaction between green consumption and green innovation alone explained 0.5 of the spatial differentiation of carbon finance development, far exceeding any single factor, a synergy the authors call a demand-innovation coupling.

The authors are careful about what these findings do and do not establish. They note explicitly that the Geographical Detector explains spatial heterogeneity rather than proving causation, so the results should be read as evidence of spatial association rather than strict causal effects. Data availability also constrained the evaluation system: market-oriented indicators such as carbon derivatives, green investment flows, and firm-level trading participation remain largely inaccessible at the prefecture-city level, and the regional focus limits generalizability to eastern coastal provinces or the national market as a whole. Sensitivity analyses, including fourteen robustness tests that eliminated individual indicators and a CRITIC weighting substitution, produced Spearman rank correlations above 0.95, supporting the stability of the rankings. The policy implications the authors draw are correspondingly structural: build cross-regional carbon finance collaboration networks anchored in Chongqing, Xi’an, and Chengdu; strengthen green innovation through patent support and long-term green lending; and deploy blockchain and big data technologies in carbon trading to lower participation barriers and ease the information asymmetries that still fragment the region’s carbon markets.

Subject of Research: Spatiotemporal patterns and driving factors of carbon finance development across 79 prefecture-level cities in Western China from 2014 to 2024

Article Title: A study on the spatiotemporal differences and influencing factors of carbon finance development in Western China

Article References: Yu, L., Zhou, Q., & Gao, W. (2026). A study on the spatiotemporal differences and influencing factors of carbon finance development in Western China. Digital Economy and Sustainable Development, 4(1), Article 11. https://doi.org/10.1007/s44265-026-00086-9

Image Credits: AI Generated

DOI: 10.1007/s44265-026-00086-9

Keywords: carbon finance, Western China, PSR framework, entropy weight method, spatial autocorrelation, Geographical Detector, green innovation, green consumption, digital economy, environmental regulation, carbon market, regional disparity