This research investigates the dynamic interactive associations among sustainable investment in the energy sector, air pollution, and sustainable development. To this end, it employs a "one-step" system-generalized method of moments (GMM) and "one-step" differential-GMM estimators, covering the period between 1996 and 2017. In this context, it utilizes the simultaneous equations of the dynamic panel data model for panel data of 27 Chinese provinces and municipalities. We have developed a new model of sustainable development, which incorporates sustainable investment in the energy sector and air pollution to offer a robust theoretical foundation for considering the underlying relations. The system-GMM estimator is used for the full data set; however, differential-GMM is utilized for the subsets of data, in order to tackle the small sample bias problem. The empirical outcomes provide several vital insights in that they yield mixed findings for the aggregated sample and subsets of data. For example, a two-way causal relationship occurs for all the panels, except the central part (medium development regions), between sustainable investment in the energy sector and sustainable development. Contrary to this, causality runs from air pollution to sustainable investment in the energy sector in a full data set and the central part (medium dev.). Nevertheless, the opposite is true in the case of the eastern part (most developed regions) of China. Still, the same relationship runs in either direction in the case of the western part (least developed regions). On the other way around, the feedback hypothesis of causality is confirmed, across all the samples, between air pollution and sustainable development. Hence, sustainable development and air pollution are overwhelmingly interdependent, in the country as well as the province and municipality level of the Chinese economy.
Keywords: Air pollution; Disaggregated analysis; Generalized method of moments; Sustainable development; Sustainable investment in energy sector.