REGULATORY QUALITY AS A SHORT-RUN CATALYST IN CHINA'S RENEWABLE ELECTRICITY TRANSITION: EVIDENCE FROM AN ARDL-BAYESIAN-MACHINE LEARNING FRAMEWORK
This study examines whether regulatory quality conditions the translation of economic capacity into renewable-electricity integration in China, 2000-2024. Using ARDL bounds testing on a logit-transformed renewable-electricity share and a five-variable specification selected after diagnosing severe multicollinearity in a broader candidate set, we find a valid error-correction relationship (adjustment coefficient -0.372, p=0.005; bounds F-statistic 15.29, exceeding the 1% critical bound). Digital readiness is the only variable with a significant long-run association with renewable integration (0.771, p=0.048). Economic capacity has no significant direct long-run or short-run association. Regulatory quality's role is concentrated in the short run and is conditional on economic capacity: a regulatory-quality shock is associated with a significant negative short-run effect on renewable integration at low economic capacity (-1.61, 95% CI [-2.18,-1.04]), a null effect at mean capacity (-0.08, 95% CI [-0.58,0.42]), and a significant positive effect at high capacity (1.09, 95% CI [0.18,2.01]). Granger-causality tests indicate a significant bidirectional relationship between regulatory quality and renewable integration, consistent with a feedback dynamic rather than one-directional causation. Bayesian Model Averaging and blocked time-series cross-validated XGBoost-SHAP are used as robustness and exploratory tools respectively; the latter shows limited out-of-sample generalizability at this sample size and is reported descriptively. Rule of law and control of corruption, tested as alternative moderators, show materially weaker evidence of a cointegrating relationship than regulatory quality, supporting discriminant validity of the regulatory-quality construct specifically. We interpret the results as evidence that regulatory quality's association with renewable-electricity integration operates through short-run adjustment dynamics conditional on prior economic development, rather than as an independent long-run driver. Keywords: Renewable energy integration; regulatory quality; institutional moderation; ARDL bounds testing; Bayesian Model Averaging; China JEL Codes: O13, Q42, Q48, C51
Authors
- Perspectives in Social Sciences
Publication Details
- Journal
- Perspectives in Social Sciences
- Published
- 2026-11-14
- DOI
- https://doi.org/10.5281/zenodo.24218576
- Primary Topic
- Energy, Environment, Economic Growth
- Type
- article
- Field-Weighted Citation Impact
- 0.00