Study on the spatial characteristics of the digital economy on urban carbon emissions

Abstract

This paper explores the spatial impact effects and spatio-temporal heterogeneity of the digital economy on urban carbon emissions (CO₂). Specifically, firstly, the Digital Economy Index (DEI) system of 285 cities in China was constructed and the Global Principal Component Analysis (GPCA) was applied to evaluate the digital economy level of Chinese cities. Based on spatial correlation and spatial heterogeneity, the paper explores the global spatial impact and spatio-temporal heterogeneity of the impact of digital economy on CO₂ using spatial Durbin model (SDM) and geographic time-weighted regression (GTWR), respectively. And the mechanism variables are used to further demonstrate the mechanism effect and nonlinear characteristics of the impact of digital economy on CO₂. The findings indicate that (1) the advancement of the digital economy is beneficial for achieving carbon abatement goals, and the impact of the digital economy on CO₂ mitigation remains stable across several robustness tests. (2) The spatial spillover effect of the digital economy on the impact of carbon reduction is not significant. And there is significant temporal and spatial heterogeneity in the impact of the digital economy on carbon emissions. (3) According to the mechanism analysis, the digital economy reduces carbon emissions by encouraging the development of green technologies and promoting the modernization of industrial structures. And there are non-linear characteristics of this effect. This study concludes that the digital economy can support China’s goal of achieving “carbon peak and carbon neutrality.” However, it is important to consider the differences in urban development over time and space. Leveraging the city’s strengths to develop a digital economy in a unique way that will help achieve China's carbon reduction goals.

Appendices

Appendix 1

Table 7 Digital economy indicator system

Table 8 Global Moran’s I test

Table 9 Spatial panel model test

Appendix 2

The details of Dagum Gini coefficient

The Gini coefficient used in this paper mainly refers to Dagum (1997) to explore the differences of digital economy among geographical regions in China. Furthermore, the regional differences of digital economy are decomposed into three parts: intra-group difference contribution (Gw), inter-group difference net contribution (Gb) and intensity of transvariation (Gt). The specific Dagum Gini coefficient is calculated in Eq. (5).

$$G=\frac{\sum \limits_{j=1}^k\sum \limits_{h=1}^k\sum \limits_{i=1}^{n_j}\sum \limits_{r=1}^{n_h}\left|{y}_{ji}-{y}_{hr}\right|}{2\mu {n}^2}$$

(5)

y represents the DEI of city in geographic region. μ represents the weighted average of DEI of 285 cities, and represents the number of groups, in this case. The formulas for calculating Gw, Gb, and Gt are as Eq. (6).

$${\displaystyle \begin{array}{l}{G}_w=\sum \limits_{j=1}^k{G}_{jj}{p}_j{s}_j\\ {}{G}_{jj}=\frac{1}{2\overline{y_j}{c}_j^2}\sum \limits_{i=1}^{c_j}\sum \limits_{r=1}^{c_j}\left|{y}_{ji}-{y}_{jr}\right|\\ {}{G}_b=\sum \limits_{j=2}^k\sum \limits_{h=1}^{j-1}{G}_{jh}\left({p}_j{s}_h+{p}_h{s}_j\right){D}_{jh}\\ {}{G}_{jh}=\frac{1}{c_j{c}_h\left(\overline{y_j}+\overline{y_h}\right)}\sum \limits_{i=1}^{c_j}\sum \limits_{r=1}^{c_h}\left|{y}_{ji}-{y}_{jr}\right|\\ {}{G}_t=\sum \limits_{j=2}^k\sum \limits_{h=1}^{j-1}{G}_{jh}\left({p}_j{s}_h+{p}_h{s}_j\right)\left(1-{D}_{jh}\right)\end{array}}$$

(6)

$${p}_j=\frac{c_j}{c},{s}_j=\frac{c_j\overline{y_j}}{c\overline{y}},\sum {p}_j=\sum {s}_j=\sum \limits_{j=1}^k\sum \limits_{h=1}^k{p}_j{s}_j=1$$

where G_jj and G_jhrepresents the Gini coefficient within the region and the Gini coefficient between regions, and D_jh represents the relative influence of the DEI of cities in different geographical regions. The specific formula is shown in Eq. (7).

$${\displaystyle \begin{array}{l}{D}_{jh}=\frac{\left({d}_{jh}-{p}_{jh}\right)}{\left({d}_{jh}+{p}_{jh}\right)}\\ {}{d}_{jh}={\int}_0^{\infty }{dF}_j(y){\int}_0^y\left(y-x\right){F}_h(x)\\ {}{p}_{jh}={\int}_0^{\infty }{dF}_h(y){\int}_0^y\left(y-x\right){F}_j(x)\end{array}}$$

(7)

d _jh represents the weighted average of the difference of the DEI between groups, that is, the mathematical expectation of the sum of all y_ji − y_hr > 0 sample values in different geographical regions. p_jh is expressed as the supervariable first moment, that is, is expressed as the weighted average of all y_hr − y_ji > 0 sample values in different geographical partitions.

Table 10 Analysis of Geographical zones digital economy differences

Table 11 Threshold effect test.

Figures 5, 6, and 7 show the LR plots of the threshold value for the digital economy (DIGE), green technology innovation (innovation), and industrial structure (industry) as threshold variables, respectively. The threshold estimation values are in red line with true values.

Fig. 5

LR figure of threshold value for the digital economy

Fig. 6

LR figure of threshold value for the green technology innovation

Fig. 7

LR figure of threshold value for the industrial structure