TL;DR
This paper moves beyond treating ecological efficiency (EE) as a "black box" by modeling it as a dynamic, interactive network of economic, social, and environmental subsystems. By combining two advanced econometric tools—GSE-NDEA and PVAR—the authors reveal that while China's ecological efficiency is recovering in a "U-shaped" trend, the environmental subsystem remains the primary bottleneck hindering sustainable development.
What Problem Does This Paper Address?
Traditional evaluations of ecological efficiency often simplify the process: resources go in, and economic value (plus some pollution) comes out. This "black box" approach ignores two critical factors:
- The Social Dimension: Most studies focus only on the economy and the environment, neglecting how social factors like health and education feed back into the system.
- Internal Interactions: Subsystems don't exist in isolation. A strict environmental law might help the "Environment" subsystem but inadvertently cripple the "Economic" subsystem if not balanced correctly.
The authors tackle these gaps by analyzing 30 Chinese provinces from 2011 to 2020, a pivotal decade that saw China shift from "growth at all costs" to strategies like "Ecological Civilization" and "Carbon Neutrality."
Methods
The researchers employed a "two-punch" methodological approach:
- GSE-NDEA (Global Super Efficiency Network Data Envelopment Analysis):
- Network: Unlike standard DEA, this looks inside the system at three distinct nodes (Economy, Society, Environment) and the "intermediate" flows between them (e.g., how industrial waste flows from the economy to the environment).
- Global: It uses a global production frontier, making efficiency scores from 2011 comparable to those in 2020.
- Super Efficiency: It allows the "best" performers to have scores greater than 1, enabling a full ranking of provinces.
- PVAR (Panel Vector Autoregression): This is used to identify the direction and duration of influence between subsystems. For example, if the social subsystem improves today, how long does it take to see a positive impact on environmental efficiency?
Datasets
The study utilizes a panel dataset covering 30 Chinese provinces (excluding Tibet, Hong Kong, Taiwan, and Macau) from 2011 to 2020. Data was meticulously scraped from several official sources:
- China Energy Statistical Yearbook
- China Statistical Yearbook (and specific volumes for Industry, Science & Technology, Environment, and Civil Affairs)
- Statistical Communiques on National Economic and Social Development.
The authors selected 17 indicators, including "undesirable" outputs like CO2, SO2, and Nitrogen Oxide, which were aggregated using the entropy weight method to maintain model robustness.
Key Insights and Ideas
- The "U-Shaped" Recovery: China’s overall ecological efficiency didn't just go up. It declined until 2015—likely due to the "growing pains" of structural economic transformation—before beginning a steady recovery.
- The Environmental Bottleneck: Of the three subsystems, the Environmental Subsystem (EnS) consistently shows the lowest efficiency. It is the "weakest link" holding back the entire system's progress.
- Regional Divergence:
- East: The leader in efficiency, leveraging high-tech industries and "intellectual governance" (Big Data/AI).
- West: Catching up fast, but highly vulnerable. Interestingly, economic growth here initially had a negative impact on social development, likely due to widening wealth gaps during early-stage development.
- Inner: Struggling with population pressure. Rapid immigration has created a "lagging" negative effect on environmental efficiency.
- Lack of Synergy: A major "aha moment" is that subsystems currently depend mostly on their own internal momentum. The "mutual promotion" mechanism (e.g., the economy driving environmental tech, which in turn drives social health) is still in its infancy and needs policy intervention to mature.
Key Results
- Efficiency Ranking: On average, the Social subsystem performs best, followed by the Economy, with the Environment lagging significantly behind.
- Regional Performance: The East remains the most efficient. By 2018, the West surpassed the Inner region in ecological efficiency, thanks to aggressive green development policies and "leapfrogging" with newer infrastructure.
- Quantitative Impact: By the 6th period of the PVAR forecast, the contribution of the Economy to its own development dropped to 80.7% in the Inner region, suggesting that other subsystems are starting to play a larger (though still secondary) role in driving growth.
Limitations and Future Work
The authors are transparent about three main hurdles:
- Spatial Correlation: The model treats provinces as independent units, ignoring "spillover" effects (e.g., pollution in one province blowing into the next). Future work should use spatial econometrics.
- Indicator Lag: Some economic indicators have delayed effects that weren't fully captured.
- Sample Size: While 30 provinces provide a solid foundation, the authors suggest expanding the sample to 114 key cities and using bootstrap regression to verify results with higher statistical confidence.
Summary updated