Do inward foreign direct investment and economic development improve local environmental quality: aggregation bias puzzle

Abstract

Since preceding several decades, the carbon emissions based standard Environmental Kuznets Curve (EKC) has been tested and supported by a plethora of studies in countries around the globe. The current study estimated the inward foreign direct investment (IFDI)-augmented EKCs for China’s 27 provincial divisions employing the advanced econometric methodologies involving cross-sectional dependence, slope heterogeneity, and second generation-based estimation procedures. The study has further contributed through a modification to “Stochastic Influence by Regression on Population, Affluence, and Technology” (STIRPAT) in terms of including IFDI to the standard model. Accordingly, this work estimated the standard EKC (involving economic development-carbon emissions linkage) as well as IFDI-carbon emissions linkage within the STIRPAT framework, by employing a panel vector error-correction-based estimation procedure. The findings revealed that (1) the conventional EKC estimates for national and regional samples (i.e., aggregate samples) presented linkages differing from the EKC links for the provincial divisions. It suggested that the EKC at the aggregated levels is likely the consequence of aggregation bias problem. (2) The links between IFDI (in power and non-power sector) and carbon emissions provided inverse U shape for the aggregate samples, while the provincial divisions presented heterogeneous results. This is perhaps because of the aggregation bias. Hence, the aggregation bias puzzle is unriddled. (3) Also, heterogeneous patterns are found in terms of turning points, degree of impact, and nature of the association of income and IFDI with carbon emissions. The meaningful policies can be extracted for the large countries encompassing varied economic development levels, such as China, if the EKC is evaluated at the disaggregate scales.

Appendix 1

Carbon emissions calculation

Let’s start with the argument that the fossil-based fuels’ utilization and the cement production are taken as the principal origins of China’s carbon emissions (Wang et al. 2020). So, the carbon emissions of China’s provincial divisions are computed by the underlying equation:

$$ {CE}_{pd}={\sum}_{k=1}^8\left[\left({UF}_{pd,k}.{CAF}_{pd,k}.{PECE}_{pd,k}-{CSE}_{pd,k}\right).{SOC}_{pd,k}\right]\times 44/12+{ CE C}_{pd} $$

(A.1)

where pd is the representation of provincial divisions, and k is the representation of the energy product’s kind employed in computation. CE stands for the carbon emissions; CAF stands for fuels’ calorific element; UF stands for particular fuels’ utilization; CSE stands for fuels’ carbon sequestration rate; PECE stands for the carbon emissions’ potential element; SOC stands for the share of fuels’ oxidized carbon; CEC is the representation of cement production process¹’ carbon emissions. Finally, the “China Economic Information Network” has been used to retrieve the provincial division-based carbon emissions data from cement production, which is aligned with (Ahmad et al. 2019c).