Digital Economy, Clean Energy Consumption, and High-Quality Economic Development: The Case of China

Abstract

The digital economy has become an important force driving China’s socio-economic development. From the perspective of sustainable energy development and based on China’s provincial panel data from 2011 to 2020, this paper probes into the relationship and transmission mechanism of digital economy, clean energy consumption, and high-quality economic development and utilizes the spatial Durbin model and intermediary effect model to carry out empirical tests on the paths of their influence. The results indicate that (1) the digital economy has a significant promoting effect and a positive spatial spillover effect on high-quality economic development, and its influence has obvious spatial heterogeneity; (2) the intermediary effect model validates that clean energy consumption is a mediating variable of digital economy and high-quality economy development, and its conduction effect also has obvious spatial heterogeneity. The research conclusions provide a new perspective and empirical evidence for understanding the relationship between digital economy and high-quality economic development as well as policy implications for the realization of digital transformation and low-carbon development goals.

1. Introduction

The booming digital economy has become an important engine for promoting the high-quality development of China’s economy. The central and local governments have issued a series of policy documents aimed at promoting the transformation of digital economy and realizing high-quality economic development. The Sixth Plenary Session of the 19th CPC Central Committee pointed out that the development of the digital economy has become an important focus for promoting China’s economic and social development. With the increasing penetration of the digital economy into people’s clothing, food, housing, transportation, and entertainment, the digital economy has gradually promoted a new pattern of domestic and international double-cycle development and has become a new major booster of China’s high-quality economic development.

At the same time, rapid economic development has put enormous pressure on energy and environment. China’s economy, characterized by investment-driven and heavy industry, has entered into a rapid take-off development mode, but it has also become the world’s largest energy consumer and carbon emitter, and its long-term high level of energy consumption has gradually brought China closer to the "red line" of resource and ecological carrying capacity. With the development of global economic integration, sustainable economic development requires the guarantee of sustainable energy development. Therefore, reducing the level of energy consumption and promoting the clean consumption of energy has become an important component for China’s high-quality economic development and sustainable energy development.

Academic research on how the digital economy drives economic growth can be divided into three main levels: macro, meso, and micro. The macro-level research is mainly based on economic growth theory to study its mechanism of action on resource allocation efficiency and total factor productivity and to explore whether information factors such as data can become new factors of production and whether they will change the traditional input–output relationship. Digital economy is a special economic form in which goods and services are traded digitally [1]. Solow [2] proposed the famous “productivity paradox”; that is, the huge investment in information technology did not bring the expected high rate of productivity growth but rather stagnant or declining productivity growth. Dewan and Kraemer [3] and Stratopoulos and Dehning [4] suggested that the “productivity paradox” disappears for developed countries and for firms or companies that are successful users of IT, but it still exists for developing countries and unsuccessful users of IT. Jorgenson and Motohashi [5] found that information technology capital contributes significantly to the U.S. economy. Forman, Goldfarb, and Greenstein [6] and Ivus and Boland [7] believed that the emergence of the Internet and e-commerce has contributed to economic development. Research at the meso level focuses on analyzing how digital technology can promote the modernization of traditional industries and exploring ways to create economic benefits from new forms of industrial integration and agglomeration. Aghion, Jones, and Jones [8] showed that digital technologies can solve the problems of high research costs and inefficient resource allocation, form customer-centric business models and rapidly replicate and spread them, and become a new service industry that undergoes advanced transformation. A digital economy improves transaction efficiency by reducing transaction costs of enterprises [9], cracking down on regional financial differentiation [10], and improving the availability of financial services and reducing information friction as multiple ways to increase regional entrepreneurial activity, which in turn promotes economic growth. The micro-level research is focused on the field of digital consumption, where the use of digital technologies is fundamentally changing traditional consumer demand. Digital technology has changed consumer preferences and decision making, and the development of the digital economy has made it possible for consumers to personalize their needs, which in turn has led to the upgrading of consumption in traditional markets [11].

In addition to this, there are several studies on the digital economy’s contribution to clean energy consumption. The transparency of information and the ease of information transfer brought by a digital economy has helped inefficient sectors to improve the efficiency of the allocation of labor and capital, enabling balanced and coordinated development in various regions and enhancing green total factor productivity. As an important driving force for energy efficiency, the digital economy plays an irreplaceable role in realizing energy-saving goals [12]. It has been shown that the growth of national carbon emissions can be curbed to a considerable extent by the development of the digital economy and that reductions in emissions are positively correlated with each country’s level of economic development [13]. The impact of the digital economy on carbon emissions is a non-linear inverted “U”-type relationship, and in China, for example, its impact on carbon emissions is mainly in the eastern region, while the impact on the central and western regions is weaker [14]. The digital economy can promote green total factor productivity (GTFP) through green technological innovation, industrial structure upgrading, energy saving, and emission reduction and has a positive spatial spillover effect, which can significantly promote the green development of cities. As an emerging factor of production, data can be capitalized to create revenue and act as an increment in the economic system to achieve green total factor productivity. The informatization and digitization extended by the digital economy have improved the efficiency of capital accumulation [15], thereby enhancing green total factor productivity [16]. Digital technology can also break through time and space constraints, realize the rapid matching of the employment process of workers, and enhance the employment rate of the labor force, which in turn leads to a more reasonable and efficient allocation efficiency of labor factors [17].

Current research on the relationship between digital economy and high-quality economic development mainly focuses on intermediary channels such as technological innovation and industrial agglomeration and rarely includes the transmission path of clean energy consumption, and there is a lack of relevant literature on such issues from the perspective of sustainable energy development. Therefore, based on a panel dataset of 30 Chinese provinces (excluding Hong Kong, Macao, Taiwan, and Tibet due to missing data) from 2011 to 2020, this paper analyzes the mechanism of the digital economy’s impact on the high-quality development of the economy and explores the mediating effect of clean energy consumption in it.

This paper aims at analyzing the impact of a digital economy on high-quality economic development from the perspective of sustainable energy development. It might complement existing studies in two dimensions: first, this paper analyzes and explores the spatial effects of digital economy on high-quality economic development by combining the spatial Durbin model with the improved measurement methodology of digital economy and high-quality economic development in China; second, this paper innovatively incorporates clean energy consumption into the study of the transmission path of how a digital economy affects high-quality economic development, which enriches the study of the impact path between the two and provides new cognitive perspectives and policy orientations for promoting high-quality economic development.

2. Theoretical Basis and Research Hypothesis

The digital economy is the main economic form after the agricultural economy and industrial economy, as it is a new economy with data resources as the key element, modern information network as the main axis, integrated application of information and communication technology and digitization of each element as the important driving force, and a more unified economic form. Advances in digital technology are profoundly changing human production and lifestyle, promoting the double digital transformation of the energy supply and demand sides and having a profound impact on energy consumption. As mentioned above, the digital economy has promoted economic growth through various approaches, such as resource allocation efficiency, total factor productivity [1,2,3,4,5,6,15,16,17], modernization of traditional industries, reduction of transaction costs, regional financial differentiation [7,8,9,10], and promotion of digital consumption [11]. There is also a great deal of relevant literature verifying that the development of the digital economy significantly promotes high-quality economic development [18,19,20]. Therefore, we propose hypothesis H1:

Hypothesis 1 (H1):

The digital economy can significantly drive high-quality economic development.

The progress of digital technology is profoundly changing human production and lifestyle, promoting the dual digital transformation of the energy supply side and demand side and having a profound impact on energy consumption. The development of the digital economy has obvious regional heterogeneity and spatial spillover effect on energy consumption [21]. There is an “n”-curve relationship between carbon dioxide emissions in Chinese cities and the digital economy, which means that the expansion of the digital economy initially strengthens carbon dioxide emissions, but, to a certain extent, it helps to achieve the goal of urban decarbonization. The impact of the digital economy on carbon dioxide emissions has spatial spillovers and regional heterogeneity, and through economies of scale and upgrading of the industrial composition, it can help cities to reduce their carbon emissions in favor of the decarbonization of neighboring cities [22]. The development of digital technology has a significant negative correlation with the structure of energy consumption and the intensity of energy consumption [23]. Therefore, we propose hypothesis H2:

Hypothesis 2 (H2):

The development of the digital economy has a significant role in promoting clean energy consumption.

Digital development has increased the efficiency of energy resource use in carbon-intensive industries, improved production structures, and significantly reduced carbon emissions, but at the same time, as the scale of digital investment has increased dramatically in recent years, digital carbon emissions have also grown significantly [24]. Several studies have pointed to the positive role of ICT in improving energy efficiency [25], while some studies, on the other hand, have concluded that increased electricity consumption is due to ICT [26]. Its energy-saving effects are controversial. Information and communication technologies (ICT) require large amounts of electricity and carbon-intensive materials as direct production inputs, and the implicit carbon emissions of the ICT sector cannot be ignored [27]. However, increased digitization and the rise of ICT services have undoubtedly had a positive effect on reducing energy consumption and promoting sustainability [28], but developing countries will bear higher carbon costs than developed countries [29].

China’s current energy production structure is still dominated by coal and supplemented by oil and natural gas, and its economic development is characterized by strong “energy dependence” and “high energy consumption”, with growth in all types of energy consumption directly leading to an increase in GDP, and there is a unidirectional causal relationship between all types of energy consumption and economic growth [30]. This is also consistent with existing research in other countries [31,32,33]. As China’s economy enters the stage of high-quality development, “green” has been incorporated into the new development concept, and green development and low-carbon transformation have become the main theme of the times. The rapid development of a new generation of digital technology has made the digital economy leapfrog the agricultural and industrial economy and become a new economic form, and the digital economy provides an opportunity for regional low-carbon transformation. At the Fifth Plenary Session of the 19th CPC Central Committee, the CPC Central Committee proposed to build a new development pattern with the domestic macro-cycle as the main body and the domestic and international double-cycle promoting each other, which is an opportunity for our country to follow the trend and realize the sustainable development of energy with high quality. China’s demand for energy will undoubtedly continue to grow in the context of high-quality economic development, and the stimulation of the economy needs to be accompanied by a focus on actions to reduce carbon dioxide emissions and the development of renewable energy sources [34]. Empirical studies from China have proven that optimizing energy structure [35,36] and energy resource allocation [37] has a positive effect on promoting the high-quality development of China’s economy. Therefore, we propose hypotheses H3 and H4:

Hypothesis 3 (H3):

Clean energy consumption can significantly contribute to high-quality economic development.

Hypothesis 4 (H4):

Clean energy consumption has a significant mediating effect on the digital economy and high-quality economic development.

3. Empirical Study Design

3.1. Definition of Variables

3.1.1. Explanatory Variables

The explanatory variable of the article is the level of high-quality economic development ($hq{e}_{it}$), where $i$ and $t$ denote the region and year, respectively, and $hq{e}_{it}$ denotes the level of high-quality economic development of the $i$ region in the $t$ year. This article draws on the measurement method of Sun, Gui, and Yang [38] and selects 18 sub-level indicators from five dimensions of innovation development, coordinated development, green development, open development, and shared development and uses the entropy value method to measure the level of high-quality economic development in 30 provinces and municipalities in China. Further, it constructs the evaluation system and development index of China’s level of high-quality economic development, as shown in

Table 1. Evaluation system and development index of China’s high-quality economic development level.

Level 1 Indicators	Level 3 Indicators	Level 3 Indicators	Indicator Definition	Indicator Properties
High-quality economic development level	Innovative development	GDP growth rate	Regional GDP growth rate	+
		R&D investment intensity	R&D expenditure of industrial enterprises above the scale/GDP	+
		Investment efficiency	Investment rate/GDP growth rate	-
		Technology trading activity	Technology transaction turnover/GDP	+
	Coordinated development	Demand structure	Total retail sales of social consumer goods/GDP	+
		Urban–rural structure	Urbanization rate	+
		Industry structure	Tertiary industry output/GDP	+
		Government debt burden	Government debt balance/GDP	-
	Green development	Energy consumption elasticity coefficient	Energy consumption growth rate/GDP growth rate	-
		Unit output of wastewater	Wastewater discharge/GDP	-
		Exhaust gas generated by the unit	Sulfur dioxide emissions/GDP	-
	Open development	External trade dependence	Total import and export/GDP	+
		The proportion of foreign investment	Total foreign investment/GDP	+
		Degree of marketization	Regional marketability index	+
	Shared development	Share of workers’ compensation	Workers’ compensation/GDP	+
		Residents’ income growth elasticity	Per capita disposable income growth rate/GDP growth rate	+
		Urban–rural consumption gap	Per capita consumption expenditure of urban residents/per capita consumption expenditure of rural residents	-
		Share of fiscal expenditure on people’s livelihood	The proportion of local financial expenditure on education, health care, housing security, social security, and employment to local budget expenditure	+

.

The specific calculation steps of China’s high-quality economic development index are as follows:

The first step is to de-dimensionalize the index. All three indexes are indexed, and the indexation methods for positive and inverse indexes are Formulas (1) and (2), respectively:

$$Y_{ijt}=\frac{X_{ijt}-X_{ijt\min }}{X_{ijt\max }-X_{ijt\min }}$$

(1)

$$Y_{ijt}=\frac{X_{ijt\max }-X_{ijt}}{X_{ijt\max }-X_{ijt\min }}$$

(2)

where $i$ is each region, $j$ is each indicator, $t$ is each year, $X_{ijt}$ is the real data of $j$ indicator in $i$ province in $t$ year, and $X_{ijt\min }$ and $X_{ijt\min }$ are the minimum and maximum values of $X_{ijt}$, respectively, while $Y_{ijt}$ is the $X_{ijt}$-indexed data.

The second step is to determine the index weight. The index system built based on the new development concept is equally important in the dimensions of innovation, coordination, green, openness, and sharing. This paper chooses to use the equal weight method to empower each indicator.

The third step is to calculate the composite index. In this paper, the linear weighting method is used to calculate the economic high-quality development index, as shown in Formula (3):

$$hq{e}_{it}={\displaystyle \sum _{j=1}^{j=18}(X_{ijt}\times w_j)}$$

(3)

where $w_j$ is the weight of $j$ indicator.

3.1.2. Core Explanatory Variables

The core explanatory variable of the article is the level of digital economy development ($de{l}_{it}$). Based on the research method of Zhao, Zhang, and Liang [39], this paper selects the corresponding sub-level indicators from five dimensions, including Internet penetration rate, number of Internet-related employees, Internet-related output, number of mobile Internet users, and digital financial inclusion development. The actual contents of the first four dimensions are as follows: the number of Internet broadband access users per 100 people, the proportion of computer service and software industry employees in urban units, the total number of telecommunications services per capita, and the number of mobile phone users per 100 people. The original data of the above indicators can be obtained from the “China City Statistical Yearbook”. For the development of digital finance, the China Digital Financial Inclusion Index was adopted, which was jointly compiled by the Digital Finance Research Center of Peking University and Ant Financial Group. In this paper, the meanization method is first adopted for dimensionless processing of each index data. Then, the principal component analysis [40] is adopted to determine the weight of each index, and the covariance matrix of the basic index is taken as the input, the data of the above five indicators are standardized for dimensionality reduction processing, and finally, the comprehensive development index of digital economy is obtained. The evaluation system and development index of the digital economy development level are as shown in

Table 2. Evaluation system and development index of digital economy development level.

Level 1 Indicators	Level 2 Indicators	Indicator Definition	Indicator Properties
Digital economy development level	Internet penetration rate	Number of Internet users/year-end resident population	+
	Number of Internet-related employees	Information transmission and soft-ware and information technology service industry employed in urban units/employed in urban units	+
	Internet-related outputs	Total telecom services/year-end resident population	+
	Number of mobile Internet users	Cell phone penetration rate	+
	Digital inclusive finance index	China’s digital inclusive finance index	+

.

3.1.3. Control Variables

The article selected the following control variables: (1) urbanization level ($u{l}_{it}$), expressed by the urbanization rate of each provincial and urban area; (2) FDI intensity ($fd{i}_{it}$), reflected by the ratio of FDI to GDP in each provincial and urban area [41]; (3) economic development level ($gd{p}_{it}$), measured by GDP per capita and logarithmized ($Ingd{p}_{it}$); (4) technology level ($p{a}_{it}$/$tc{t}_{it}$), expressed by the number of patent applications and technology contract turnover by provinces and municipalities and logarithmized ($Inp{a}_{it}$/$Intc{t}_{it}$); and (5) industrial structure ($i{s}_{it}$/$i{t}_{it}$), expressed as the share of secondary industry in regional GDP and the share of tertiary industry in regional GDP.

3.1.4. Intermediary Variables

The mediating variable in the article is the clean energy consumption index ($ge{c}_{it}$), and this paper draws on the research method of Liu, Sun, and Zhu [42], which divides all consumed energy into three categories, namely coal, oil and gas, and other energy consumption. The consumption share of each type of energy in year $t$ is used as a component of a spatial vector, which in turn forms a set of three-dimensional vectors: $E_t=e_t^1,e_t^2,e_t^3$.

Second, we calculate the angles ${\theta }_t^1,{\theta }_t^2,{\theta }_t^3$ and the vectors $E_0^1=\left(1,0,0\right), E_0^2=\left(0,1,0\right), E_0^3=\left(0,0,1\right)$ of energy consumption aligned from high carbon to low carbon:

$${\theta }_t^j=\arccos \left(\frac{{\displaystyle \sum _{i=1}^3(e_t{}^i\times e_0{}^i)}}{{\left({\displaystyle \sum _{i=1}^3\left(e_t^i\right)\ast {\displaystyle \sum _{i=1}^3{\left(e_0^i\right)}^2}}\right)}^{1/2}}\right)j=1,2,3$$

In the end, all vector pinch angles in year $t$ are weighted to form the clean energy consumption index $ge{c}_{it}$, which is calculated as follows:

$$ge{\mathrm{c}}_t={\displaystyle \sum _{s=1}^3{\displaystyle \sum _{t=1}^s{\theta }_t^j}}$$

3.2. Data Sources

This paper aims to study the impact of digital economy on high-quality economic development and the mediating effect of low carbonization of energy consumption structure. The data were obtained from the China Statistical Yearbook of previous years, statistical yearbooks of various provinces and municipalities, the National Bureau of Statistics, the official websites of provincial statistical bureaus and the China Energy Statistical Yearbook, and finally, the panel data of 30 provinces and municipalities in China (excluding Hong Kong, Macao, Taiwan, and Tibet) for the period of 2011–2020. The missing data were supplemented by the interpolation method.

Table 3. Descriptive statistics for each variable.

Variables	Indicators	Indicator Symbols	Average Value	Standard Deviation	Minimum Value	Maximum Value
Explained variables	High-quality level of economic development	$hq{e}_{it}$	0.297	0.130	0.128	0.786
Explanatory variables	Digital economy development level	$de{l}_{it}$	0.939	1.282	−1.027	7.416
Intermediate variables	Clean energy consumption index	$ge{c}_{it}$	5.65	0.39	5.027	6.942
Control variables	Urbanization level	$u{l}_{it}$	58.377	12.158	35	90
	FDI intensity	$fd{i}_{it}$	9.998	1.652	6.556	14.888
	Economic development level	$Ingd{p}_{it}$	10.841	0.436	9.706	12.013
	Technology level	$Inp{a}_{it}$	4.653	0.609	2.865	5.986
		$Intc{t}_{it}$	6.054	0.777	3.753	7.801
	Industry structure	$i{s}_{it}$	41.662	8.441	15.8	62
		$i{t}_{it}$	48.569	9.216	32.6	83.9
Tool variables	Fiber-optic cable line length	$cabl{e}_{it}$	0.394	0.15	0.011	0.731

shows the descriptive statistics of each variable.

3.3. Model Design

To explore the impact of digital economy on high-quality economic development, regression models are constructed:

$$hq{e}_{it}={\alpha }_0+{\alpha }_{1}de{l}_{it}+{\alpha }_{2}ctr{l}_{it}+{\mu }_i+{\delta }_t+{\epsilon }_{it}$$

(4)

where $hq{e}_{it}$ denotes the level of high-quality economic development of region $i$ in year $t$; $de{l}_{it}$ denotes the level of digital economic development of region $i$ in year $t$; $ctr{l}_{it}$ denotes the control variable for high-quality economic development of region $i$ in year $t$; ${\mu }_i$ denotes the individual fixed effect of region $i$, which does not change over time; ${\delta }_t$ denotes the control time fixed effect; ${\epsilon }_{it}$ denotes the random disturbance term; ${\alpha }_0$ is the intercept term; ${\alpha }_1$ is the regression coefficient for digital economy; and ${\alpha }_2$ is the regression coefficient for the control variables.

The results of the spatial autocorrelation test indicate that there is a significant spatial autocorrelation in the level of high-quality economic development in China, so we use the spatial Durbin model (SDM), which can estimate the spatial spillover effects of individual as well as panel data and obtain unbiased estimation coefficients, to investigate the impact of mathematical economic development on the spatial pattern of the level of high-quality economic development. In this paper, we refer to the research method proposed by Zhang, Wang, and Liu [43] and expand Equation (4) as shown:

$$hq{e}_{it}={\alpha }_0+\rho Whq{e}_{it}+{\varphi }_{1}Wde{l}_{it}+{\alpha }_{1}de{l}_{it}+{\varphi }_{2}Wctr{l}_{it}+{\alpha }_{2}ctr{l}_{it}+{\mu }_i+{\delta }_t+{\epsilon }_{it}$$

(5)

where $\rho$ is the spatial autoregressive coefficient; $W$ is the spatial weight matrix (in this paper, we use the Queen collocation matrix); ${\varphi }_1$ and ${\varphi }_2$ are the elasticity coefficients of the spatial interaction terms of the core explanatory variables and control variables, respectively. When estimating the model, the variables are logarithmically treated to avoid the influence of different scales on the results.

In addition to studying the direct effect of the digital economy on high-quality economic development, it is also necessary to test whether clean energy consumption is a mediating variable between these two. Equation (6) is the regression equation of digital economy on clean energy consumption, and Equation (7) is the regression model of digital economy and clean energy consumption on high-quality economic development. The regression model is set as follows:

$$ge{c}_{it}={\beta }_0+{\beta }_{1}de{l}_{it}+{\beta }_{2}ctr{l}_{it}+{\mu }_i+{\delta }_t+{\epsilon }_{it}$$

(6)

$$hq{e}_{it}={\gamma }_0+{\gamma }_{1}de{l}_{it}+{\gamma }_{2}ge{c}_{it}+{\gamma }_{3}ctr{l}_{it}+{\mu }_i+{\delta }_t+{\epsilon }_{it}$$

(7)

where ${\beta }_0$, ${\gamma }_0$ are the intercept terms; ${\beta }_1$, ${\gamma }_1$ are the regression coefficients of digital economy; ${\beta }_2$, ${\gamma }_3$ are the regression coefficients of control variables; ${\gamma }_2$ is the regression coefficient of clean energy consumption.

4. Measurement Results and Analysis

4.1. Spatial Pattern Analysis

With the popularization of the Internet and the continuous improvement of network infrastructure, the economic ties and mass communication between regions are becoming closer and closer, and the information flow and economic ties between provinces and cities and are becoming increasingly rapid and close. The high-quality economic development of each region will not only be affected by various local factors but also be affected by the economic linkages of the surrounding areas. The high-quality development level of the regional economy may have a high correlation in space. If the spatial effect between variables is neglected in the research process, the estimation results may be biased, which will eventually affect the accuracy and scientificity of the conclusion. Therefore, it is necessary to analyze the spatial correlation of data before designing the model.

4.1.1. Moran I Index Estimation

The global spatial correlation reflects the overall characteristics of the spatial association of variables and is often measured by the global Moran index ($GlobalMoran’sI$), the expression of which is given in (8).

$$MoranI=\frac{n}{S_0}\frac{{\displaystyle \sum _{i=1}^n{\displaystyle \sum _{j=1}^n{\omega }_{ij}\left(x_i-\overline{x}\right)\left(x_j-\overline{x}\right)}}}{{\displaystyle \sum _{i=1}^n{\displaystyle \sum _{j=1}^n{\omega }_{ij}}}}$$

(8)

In the equation $S_0={\displaystyle \sum _{i=1}^{\mathrm{n}}{\displaystyle \sum _{j=1}^n{\omega }_{ij}}}$, $n$ is the total number of spatial units, and ${\omega }_{ij}$ is the spatial weight between element $i$ and $j$. When the index value is greater than 0, it means that the attribute values of each region have positive spatial correlation; i.e., the larger the attribute value, the more obvious the spatial correlation. Similarly, when the index value is less than 0, it means that the attribute values of each region are spatially negatively correlated; i.e., the smaller the attribute value, the greater the spatial difference. When the index value is equal to 0, it means that the space is random; i.e., there is no spatial correlation.

The global Moran index of digital economy and the level of high-quality economic development were calculated using stata16 software and tested for global autocorrelation, and the results are shown in

Table 4. Digital economy and high-quality economic development level of the whole Moran index.

Year	Digital Economy Development Level		High-Quality Economic Development Level
	Moran Index	Z-Value	Moran Index	Z-Value
2011	0.156 **	1.884	0.276 ***	2.883
2012	0.160 **	1.950	0.282 ***	2.879
2013	0.126 *	1.561	0.288 ***	2.916
2014	0.106 *	1.420	0.333 ***	3.294
2015	0.104 *	1.418	0.282 ***	2.836
2016	0.113 *	1.510	0.202 **	2.024
2017	0.105 *	1.418	0.308 ***	2.957
2018	0.098 *	1.323	0.363 ***	3.385
2019	0.091	1.276	0.320 ***	3.074
2020	0.092 *	1.323	0.302 ***	2.956

Note: ***, **, and * denote 1%, 5%, and 10% significance levels, respectively.

.

It can be seen that the overall Moran indexes of digital economy and high-quality economic development level are positive and basically pass the significance test at the 10% level, indicating that the spatial aggregation of digital economy and high-quality economic development level is obvious, among which the spatial correlation of high-quality economic development level is high, and both pass the significance test at the 1% level. From the time dimension, the global Moran indexes of digital economy and high-quality economic development level show a fluctuating upward and fluctuating downward trend, respectively.

4.1.2. Local Spatiality Analysis

The global Moran index mainly reflects the overall distribution of variables under the overall space, equalizing the regional differences, so it cannot reflect the local spatial correlation characteristics, while the local spatial autocorrelation can analyze the degree of spatial correlation between each spatial object in the region and its neighboring objects. This paper carries out the analysis of local spatial autocorrelation of the level of digital economy and high-quality economic development in 2011, 2016, and 2020, which can be used to further examine the agglomeration status of the local region, and the results are shown in Figure 1 and Figure 2.

As shown in Figure 1, more than 70% of the provinces and cities are distributed in the first quadrant of high–high agglomeration and the third quadrant of low–low agglomeration, which is consistent with the results of the global Moran index, and the level of digital economy development and the level of high-quality economic development both show a positive spatial correlation. Specifically, the first quadrant of digital economy is concentrated in the Yangtze River Delta region, and the two municipalities of Beijing and Tianjin, which have a higher level of Internet infrastructure and more rapid digital economy, are geographically neighboring and show higher spatial aggregation and spatial spillover effects; the third quadrant is mainly in the central and western regions, which have a relatively backward level of Internet infrastructure compared with that of the eastern coastal regions, with a slower speed of information transmission. Thus, the development of the digital economy generates spatially positive correlation. However, the driving and spillover effects generated by the development of digital economy are very limited. The results of high-quality economic development and digital economy are largely convergent. The first quadrant is also concentrated in the Yangtze River Delta region and Beijing and Tianjin, and in these two municipalities, the general level of high-quality economic development is high, the information flow is fast, and the inter-regional high-quality economic development of the driving effect and spillover effect is obvious; the third quadrant is mainly in the central and western regions, and the level of high-quality economic development of these regions is relatively backward compared to the eastern region, and the driving effect and spillover effect is weaker than that in the eastern coastal region. This result can be observed more intuitively in Figure 2.

As shown in Figure 2, the high–high agglomeration of the digital economy is concentrated in the Yangtze River Delta region, and the low–low agglomeration is mainly distributed in the central and western regions; the high–high agglomeration of high economic quality is concentrated in the Yangtze River Delta region, and the low–low agglomeration is mainly distributed in the central and western regions. Comparison can be found in the level of digital economy development and high-quality economic development level of the high-value agglomeration area and the low-value agglomeration area, which, in the study period have an obviously overlapping part, so it is inferred that the level of digital economy and high-quality economic development has a more significant causal relationship. The level of high-quality development can be improved through the level of digital economy in the region.

4.2. Spatial Effect Analysis

4.2.1. Model Identification Test

The following four models exist for general spatial econometric models: no fixed effects, time fixed effects (tFE), spatial fixed effects (sFE), and spatiotemporal fixed effects (stFE). To better assess the relationship between the digital economy and high-quality economic development, it is necessary to determine an optimal model through testing. Referring to the test proposed by Elhorst [44], the LM test and the robust LM test are first used to determine the spatial autocorrelation of the error term and the lagged term without considering the spatial correlation, and then, the Wald and LR tests are used to test the $H_0:\gamma =0$ and $H_0:\gamma +\rho \beta =0$, respectively, and the results are shown in

Table 5. Non-spatial interaction effect results.

Inspection Method	Statistical Quantities	Probability
LM_spatial error	95.901	0.000
Robust LM_spatial error	36.635	0.000
LM_spatial lag	67.849	0.000
Robust LM_spatial lag	8.582	0.003
Wald_spatial error	93.28	0.000
Wald_spatial lag	128.97	0.000
LR_spatial error	82.16	0.000
LR_spatial lag	104.14	0.000
Hausman	72.85	0.000

.

Based on the spatial correlation test given by Elhorst [44] and the law of selection of fixed and random effects, it can be seen from the test results that, firstly, LM_spatial error, LM_spatial lag, robust LM_spatial error, and robust LM_spatial lag all pass the significance at the 1% level test, which indicates that there is a significant spatial correlation between digital economy and high-quality economic development; secondly, both Wald and LR statistics are significant and pass the Hausman test, which determines that the SDM fixed-effects model should be used.

Considering that the SDM model can also be expanded into three forms of time fixed effects, individual fixed effects, and double fixed effects, this paper selects the optimal model from them by comparing the estimation results of the three types of SDM models. The test results show that the log-likelihood values (Log-likelihood) of the time fixed-effects model, individual fixed-effects model, and dual fixed-effects model are 569.0735, 690.8045, and 708.7931, respectively; the goodness-of-fit (R² within) of the within-group models are 0.0944, 0.5054, and 0.0011, respectively. The SDM double fixed-effects model has the highest log-likelihood value, but its within-group model has the lowest goodness-of-fit; the SDM individual fixed-effects model has a higher log-likelihood value and ranks second after the SDM double fixed-effects model, and its within-group model has a much higher goodness-of-fit than the other two fixed-effects models. Therefore, the SDM individual fixed-effects model is chosen as the optimal model.

4.2.2. Baseline Regression Results

Table 6. Regression results of factors influencing the level of high-quality economic development.

	(1)	(2)	(3)
Variables	SAR	SEM	SDM
$de{l}_{it}$	0.0136 **	0.00998	0.0238 *
	(2.62)	(1.58)	(2.36)
$u{l}_{it}$	0.00115	0.00282	0.00450 **
	(0.75)	(1.78)	(2.58)
$fd{i}_{it}$	0.000674	0.00463	0.00809
	(0.11)	(0.79)	(1.43)
$Ingd{p}_{it}$	−0.0844 ***	−0.0877 ***	−0.0816 ***
	(−4.57)	(−3.81)	(−3.48)
$Inp{a}_{it}$	−0.0178 **	−0.0118 *	−0.0172 **
	(−3.10)	(−1.98)	(−3.01)
$Intc{t}_{it}$	0.0104 *	0.00781	0.0116 **
	(2.55)	(1.95)	(2.91)
$i{s}_{it}$	0.00191	0.00001	0.00405 *
	(1.03)	(0.00)	(2.23)
$i{t}_{it}$	0.00302	0.00297	0.00600 **
	(1.63)	(1.68)	(3.14)
$\rho$/$\lambda$	0.494 ***	0.584 ***	0.333 ***
	(9.19)	(11.45)	(5.35)
Variance sigma2 e	0.000710 ***	0.000665 ***	0.000569 ***
	(12.01)	(11.88)	(12.13)
$W\times de{l}_{it}$			0.0434 ***
			(3.41)
$W\times u{l}_{it}$			−0.00685 *
			(−2.17)
$W\times fd{i}_{it}$			−0.0142
			(−1.35)
$W\times Ingd{p}_{it}$			−0.0603
			(−1.78)
$W\times Inp{a}_{it}$			−0.0366 ***
			(−3.93)
$W\times Intc{t}_{it}$			0.0221 **
			(3.09)
$W\times i{s}_{it}$			0.0160 ***
			(4.36)
$W\times i{t}_{it}$			0.0063
			(1.79)
R-squared	0.213	0.472	0.678
N	300	300	300

Note: Values in parentheses are t-values, where *** indicates p < 0.01, ** indicates p < 0.05, and * indicates p < 0.1.

shows the regression results of the factors influencing the level of high-quality economic development under the three spatial econometric models of SAR model, SEM model, and SDM model (individual fixed). The results show that the SAR model $\rho$ value, SEM model $\lambda$ value, and SDM model $\rho$ value are 0.494, 0.584, and 0.333, respectively, and they all pass the 0.01 significance level test; that is, all three models confirm the existence of the spatial spillover effect of high-quality economic development, and there are obvious proximity and cascade effects of high-quality economic development in neighboring provinces and municipalities, and if the level of high-quality economic development in neighboring provinces and municipalities increases by 1%, it will indirectly promote the high-quality economic development in the region through spatial interaction. Each 1% increase in the level of high-quality economic development in neighboring provinces and municipalities will indirectly promote the level of high-quality economic development in the region by 0.333% through spatial interaction.

Since the regression coefficients of the explanatory variables in the spatial Durbin model do not directly reflect their specific impact effects on the level of high-quality economic development, this paper decomposes the spatial global effects into direct effects and indirect effects through differential processing. The direct effect indicates the influence of the development of digital economy in the region on the high-quality economic development; the indirect effect indicates the spatial spillover effect of the development of digital economy in neighboring regions on high-quality economic development of the region, and the specific results are shown in

Table 7. Direct, indirect, and total effects of factors influencing the level of high-quality economic development.

Variables	$\mathit{d}\mathit{e}{\mathit{l}}_{\mathit{i}\mathit{t}}$	$\mathit{u}{\mathit{l}}_{\mathit{i}\mathit{t}}$	$\mathit{f}\mathit{d}{\mathit{i}}_{\mathit{i}\mathit{t}}$	$\mathit{I}\mathit{n}\mathit{g}\mathit{d}{\mathit{p}}_{\mathit{i}\mathit{t}}$	$\mathit{I}\mathit{n}\mathit{p}{\mathit{a}}_{\mathit{i}\mathit{t}}$	$\mathit{I}\mathit{n}\mathit{t}\mathit{c}{\mathit{t}}_{\mathit{i}\mathit{t}}$	$\mathit{i}{\mathit{s}}_{\mathit{i}\mathit{t}}$	$\mathit{i}{\mathit{t}}_{\mathit{i}\mathit{t}}$
Direct effect	0.029 ***	0.004 **	0.007	−0.090 ***	−0.021 ***	0.014 ***	0.006 ***	0.007 ***
	(2.93)	(2.39)	(1.32)	(−3.90)	(−3.83)	(3.51)	(2.77)	(3.16)
Indirect effects	0.074 ***	−0.008 *	−0.017	−0.125 ***	−0.061 ***	0.038 ***	0.024 ***	0.011 **
	(5.23)	(−1.79)	(−1.18)	(−3.02)	(−5.51)	(3.80)	(4.47)	(2.22)
Total effect	0.103 ***	−0.004	−0.010	−0.215 ***	−0.082 ***	0.052 ***	0.030 ***	0.018 ***
	(8.11)	(−0.79)	(−0.59)	(−5.17)	(−6.32)	(4.39)	(4.44)	(2.88)

Note: Values in parentheses are t-values, where *** indicates p < 0.01, ** indicates p < 0.05, and * indicates p < 0.1.

.

The level of digital economy has a significant driving effect and a positive spatial spillover effect on the level of high-quality economic development, and its direct effect on the level of high-quality economic development is 0.029, the indirect effect is 0.074, and total effect is 0.103, and all of them are significant at the level of 0.01. Each 1% increase in the development level of digital economy will directly increase the level of high-quality economic development of the region by 0.029% and indirectly increase the level of high-quality economic development of neighboring regions by 0.074%. At the same time, the digital economy development of neighboring regions ($W\times de{l}_{it}$) also has a significant driving effect on high-quality economic development of this region: if the level of digital economy development of neighboring regions increases by 1%, it will indirectly promote high-quality economic development of this region by 0.043% through the spatial interaction. This research result verifies the research hypothesis H1 of this paper: digital economy can significantly drive high-quality economic development, which is also consistent with the research results of the related literature mentioned above [18,19,20].

The direct effect of urbanization level on the level of high-quality economic development is 0.004 and significant at 0.05 level, and the indirect effect is −0.008 and significant at 0.1 level. Although some studies have shown that the agglomeration effect can be generated through the resource agglomeration advantages of central cities [45,46], which can promote the coordinated development of regional economy, in recent years, the implementation of the strategy of “strong provincial capitals” in China has also brought about results such as the adjustment of administrative divisions, the war for people, and the competition for the pilot of the national platform policy, leading to the polarization of the agglomeration effect of the capital cities and the widening of digital economy gap between the capital cities and other local cities, increasing the level of urbanization in the region while promoting high-quality economic development. Widening the digital economy gap between provincial capitals and other local cities leads to an increase in the level of urbanization in the region while at the same time promoting the high-quality economic development but also indirectly leading to a decline in the level of high-quality economic development of neighboring regions due to the impact of the digital divide between the regions, the economic agglomeration effect, and the city’s siphoning effect.

The direct effect of economic development level on the level of high-quality economic development is −0.090, the indirect effect is −0.125, and the total effect is −0.215, all of which are significant at the 0.01 level. The results of this study are far from the research conjecture, and the possible reasons are (1) the national situation, which is caused by China’s crude economic growth model of “high input, high energy consumption, high pollution and low efficiency”; (2)the enterprises and factories, which are motivated by the profit-seeking behavior of enterprises and factories to expand production while the regional economic development level increases, which inhibits the green and coordinated development of economy and around which have centered many discussions on economic growth and environmental sustainability in other countries [47,48,49]; and (3) on the public side, as the level of economic development increases and the per capita income rises, the public’s cost of access to resources decreases, which tends to produce more resource consumption and waste and inhibits shared economic development, thus leading to a decline in the level of high-quality economic development. Therefore, although the level of economic development increases, factors such as environmental pollution, resource waste, and urban governments’ tendency to maintain growth at the expense of the environment [50] arising from the process of economic development directly and indirectly lead to a decline in the overall level of high-quality economic development in the region and neighboring areas.

Both technology level and industrial structure have obvious promotion and positive spatial spillover effects on the level of high-quality economic development.

4.2.3. Endogenous Problems

The omission of control variables and the possible reverse causality between the digital economy and high-quality economic development will lead to endogeneity problems, so this paper uses the instrumental variables approach to deal with the endogeneity problem in the regression. The development of Internet technology is related to fiber-optic broadband access technology, and a longer fiber-optic cable line means a stronger information transmission capacity, which is more conducive to the rapid development of the Internet economy [51], and since the development of digital economy is related to the length of fiber-optic cable line, and there is no direct link between the length of the fiber-optic cable line and high-quality economic development, this paper will use the length of the fiber-optic cable line ($cabl{e}_{it}$) as the instrumental variable of digital economy for the two-stage instrumental variable regression.

The regression results are shown in

Table 8. Instrumental variable regression results.

Variables	$\mathit{d}\mathit{e}{\mathit{l}}_{\mathit{i}\mathit{t}}$	$\mathit{h}\mathit{q}{\mathit{e}}_{\mathit{i}\mathit{t}}$
	2SLS Phase I	2SLS Phase II
$de{l}_{it}$	-	0.060 **
	-	(2.09)
$cabl{e}_{it}$	0.002 **	-
	(3.07)	-
N	300	300
Control variables	Yes	Yes
Fixed effects	Yes	Yes

Note: Values in parentheses are t-values, where ** indicates p < 0.05.

. After considering the endogeneity issue, in the first stage of the regression, the regression coefficient of the fiber-optic cable line length is 0.002 and significant at the 0.05 level, indicating that the explanatory variables fitted by the instrumental variables are significant; in the second stage of the regression, the regression coefficient of digital economy is 0.060 and significant at the 0.05 level, indicating that the fitted explanatory variables are significant for the explained variables. The regression coefficient of digital economy is 0.060 and significant at the 0.05 level, indicating that the fitted explanatory variables are also significant for the explained variables. Therefore, the findings of this paper are still robust after overcoming the endogeneity problem by means of the instrumental variable method.

4.2.4. Spatial Heterogeneity Analysis

The diversity of regional policies, development stages, available resources, and social environments may lead to a high degree of heterogeneity characterizing the level of digital economy and the level of high-quality economic development, so a discussion of spatial heterogeneity is warranted. In this paper, the 30 provinces and municipalities in China are divided into eastern, central, western, and northeastern regions according to the National Bureau of Statistics’ division of east–west and central and northeastern regions, and the corresponding spatial weight matrix is split for regression, from which the spatial heterogeneity of the impact of digital economy on high-quality economic development is further analyzed, and the results are shown in

Table 9. Spatial heterogeneity regression results.

Variables	Eastern Region	Central Region	Western Region	Northeast Region
$de{l}_{it}$	0.042 ***	−0.002	0.022	0.057
	(3.54)	(−0.08)	(1.09)	(0.86)
$W\times de{l}_{it}$	0.038 **	0.110 ***	0.003	−0.026
	(2.11)	(4.80)	(0.12)	(−0.44)
$\rho$	0.241 **	0.280 **	0.343 ***	0.391 **
	(2.49)	(2.50)	(3.65)	(2.57)
sigma2_e	0.001 ***	0.000 ***	0.000 ***	0.000 ***
	(7.00)	(5.42)	(7.34)	(3.64)
Direct effect	0.048 ***	0.012	0.024	0.059
	(4.04)	(0.51)	(1.16)	(0.92)
Indirect effects	0.054 ***	0.140 ***	0.016	−0.005
	(3.12)	(6.16)	(0.56)	(−0.09)
Total effect	0.102 ***	0.152 ***	0.04	0.054
	(5.15)	(6.33)	(1.22)	(0.74)
Control variables	Yes	Yes	Yes	Yes
Individual effects	Yes	Yes	Yes	Yes
Time Trends	Yes	Yes	Yes	Yes
Observations	100	60	110	30
R-squared	0.135	0.685	0.157	0.420
N	10	6	11	3

Note: Values in parentheses are t-values, where *** indicates p < 0.01, ** indicates p < 0.05.

.

According to the regression results in Table 9, it can be seen that digital economy has a more obvious spatial heterogeneity in its effect on high-quality economic development of different regions. In the eastern region, the digital economy has an obvious promoting effect and a positive spatial spillover effect on the level of high-quality economic development, with a direct effect on high-quality economic development of 0.048, indirect effect of 0.054, and total effect of 0.102, all of which are significant at the 0.01 level. In the central region, although the impact of digital economy on high-quality economic development is not as significant as that in the eastern region, its indirect effect on high-quality economic development is 0.140, and the total effect is 0.152, both of which are significant at the 0.01 level. Compared with the eastern and central regions, the impact of digital economy on high-quality economic development is not significant in the western and northeastern regions. The development of digital economy depends on the development of information technology, and the eastern and central regions have a better foundation of economic development, industrial structure, and technology level than the western and northeastern regions, among which the level of Internet development and information infrastructure in the eastern region is also far better than other regions, so the degree of impact of digital economy on high-quality economic development gradually decreases from east to west.

4.3. Mediation Effect Analysis

4.3.1. Mediation Effect Test

It has been shown in the literature that digital economy can significantly reduce the intensity of regional carbon emissions [21,22,23], promote regional low-carbon transformation, and at the same time can also promote the low-carbon transformation of neighboring regions, with a significant spatial spillover effect. The previous paper analyzed the transmission mechanism of digital economy on high-quality economic development in order to verify the existence of the transmission mechanism of clean energy consumption; this paper uses the intermediary effect model to further verify it, and the regression results are shown in

Table 10. A test of the mediating effects of digital economy, clean energy consumption, and high-quality economic development.

Variables	(1)	(2)	(3)
	$\mathit{h}\mathit{q}{\mathit{e}}_{\mathit{i}\mathit{t}}$	$\mathit{g}\mathit{e}{\mathit{c}}_{\mathit{i}\mathit{t}}$	$\mathit{h}\mathit{q}{\mathit{e}}_{\mathit{i}\mathit{t}}$
$ge{c}_{it}$			0.011
			(0.46)
$de{l}_{it}$	0.011	0.103 ***	0.010
	(0.97)	(3.43)	(0.85)
$u{l}_{it}$	0.003	0.066 ***	0.002
	(1.37)	(13.86)	(0.73)
$fd{i}_{it}$	−0.001	0.313 ***	−0.004
	(−0.11)	(20.55)	(−0.43)
$Ingd{p}_{it}$	−0.045 **	0.085	−0.046 **
	(−2.07)	(1.52)	(−2.10)
$Inp{a}_{it}$	−0.012 *	0.029 *	−0.012 *
	(−1.77)	(1.67)	(−1.81)
$Intc{t}_{it}$	0.009 *	−0.020 *	0.009 *
	(1.91)	(−1.73)	(1.95)
$i{s}_{it}$	0.001	−0.007	0.001
	(0.64)	(-1.42)	(0.68)
$i{t}_{it}$	0.005 **	-0.004	0.005 **
	(2.36)	(-0.70)	(2.38)
Constant term	0.386	−1.851 **	0.407
	(1.36)	(−2.51)	(1.41)
Time	Fixed	Fixed	Fixed
Province	Fixed	Fixed	Fixed
Observations	300	300	300
R-squared	0.567	0.872	0.567
Number of ids	30	30	30

Note: Values in parentheses are t-values, where *** indicates p < 0.01, ** indicates p < 0.05, and * indicates p < 0.1.

.

According to the results in Table 10, it can be found that digital economy has a significant promoting effect on the intermediary variable, which is clean energy consumption. This result is consistent with the existing literature [52] and also validates the research hypothesis H2: the development of digital economy has a significant role in promoting clean energy consumption. However, the results in Table 10 cannot verify the role of the influence of clean energy consumption on high-quality economic development, and when the stepwise regression method fails, the bootstrap test [31] is performed on the model, and if its confidence interval does not contain 0, there is a mediating effect. The results were obtained using stata16 software: the confidence intervals of both Bs_1 and Bs_2 do not contain 0 (Bs_1 indicates indirect effect and Bs_2 indicates direct effect); i.e., both indirect and direct effects exist. Therefore, it can be concluded that there is a mediating effect of clean energy consumption in the digital-economy-driven high-quality economic development, and this mediating variable acts as a partial mediator here. The test results of bootstrap method verify both the research hypotheses H3 and H4 of this paper: clean energy consumption can significantly contribute to high-quality economic development and has a significant mediating effect on digital economy and high-quality economic development.

Viewed through the lens of globalization, ICT trade has led to a significant imbalance in the distribution of carbon costs and economic benefits across global regions, with developing countries bearing higher carbon costs, with the resulting added value being enjoyed mostly by developed countries. The international inequality reflects the regional comparative advantages in the global ICT production and value chain, where developed economies (e.g., the U.K., U.S., and E.U. countries) dominate the R&D, design, and marketing activities of core ICT components, while emerging economies (e.g., China, India) focus on raw material extraction, intermediate processing, and assembly operations. For the emerging economies, the introduction of low-carbon technologies helps to alleviate the carbon economy brought by the inequality [29].

In terms of total carbon emissions, China, as the world’s top-emitting economy, has seen its carbon emissions rise year after year, from 2122.16 million tons in 1990 to 9570.81 million tons in 2018, with an average annual growth rate of 5.53%. Also a developing country, India’s total carbon emissions are on a rising trend year by year, with an average annual growth rate of 5.39%. Comparatively speaking, except for China and India, two developing countries, the carbon emissions of other economies have not shown a significant upward trend: the United States and the European Union’s total carbon emissions have dropped significantly in recent years, but their per capita carbon emissions are far more than India. But for developed economies such as the U.S., their total carbon emissions do not reflect their true carbon emission status, and developed countries should be subject to more stringent emission reduction standards. The development of digital economy can reduce carbon emissions to a certain extent, which can not only help developed countries such as Europe and the United States to formulate more stringent emission reduction standards but also help developing countries to realize economic development while contributing to emission reduction.

4.3.2. Robustness Tests

To verify the robustness of the mediating effect, the following robustness tests are carried out in this paper: (1) First, we replace the measures of the explanatory variables. This paper recalculates the high-quality economic development index using a new measure [53] and subsequently performs a robustness test by bootstrap method. The test results also show that the confidence intervals of Bs_1 and Bs_2 do not contain 0. In other words, after changing the measurement method of the explanatory variables, there is still a mediating effect of clean energy consumption in the impact of digital economy on economic quality development, and the mediating variable acts as a partial mediator here, indicating that the results of this paper are robust. (2) Next, we perform substitution of mediating variables. In this paper, the consumption of other energy sources (energy sources other than coal and oil) is used and logged, and then, the robustness test is conducted by bootstrap method. The test results also show that the confidence intervals of both Bs_1 and Bs_2 do not contain 0, indicating that the results of this paper are robust. In the fifth session of the 12th National People’s Congress held in 2017, “digital economy” was written into the government work report for the first time and numerous policy documents related to promoting the development of digital economy. Based on this, the interval lengths of 2011–2017 and 2018–2020 are used in this section to verify whether the model results are sensitive to the interval lengths by bootstrap method. The results of the test for the sample with the interval length of 2011–2017 show that the confidence intervals of both Bs_1 and Bs_2 do not contain 0. The results of the test for the sample with the interval length of 2018–2020 show that the confidence interval of Bs_1 does not contain 0, and the confidence interval of Bs_2 contains 0. (This means that the indirect effect is significant, and the direct effect is not significant, and the mediator acts as a full mediator here, most likely due to the small sample data [33].) The results of the test are generally consistent with the previous paper and again prove that the results of this paper are robust.

In summary, it can be verified that the results of this paper are robust, and the basic conclusions of this paper’s transmission mechanism validation have a high degree of confidence.

4.3.3. Tests for Mediating Effects in Different Regions

The development of digital economy can play a very significant role in promoting the reduction of carbon emissions. The digital economy mainly affects carbon emissions through technological innovation, industrial upgrading, optimization of energy structure, improvement of the efficiency of resource allocation, etc. In order to further verify and analyze whether the development of digital economy in different regions contributes to clean energy consumption and thus promotes high-quality economic development, the mediating effects of different regions are also examined, and the results are shown in

Table 11. Tests for mediating effects in different regions.

Variables	(1)	(2)	(3)
	$\mathit{h}\mathit{q}{\mathit{e}}_{\mathit{i}\mathit{t}}$	$\mathit{g}\mathit{e}{\mathit{c}}_{\mathit{i}\mathit{t}}$	$\mathit{h}\mathit{q}{\mathit{e}}_{\mathit{i}\mathit{t}}$
Eastern region
$ge{c}_{it}$			−0.021
			(−0.32)
$de{l}_{it}$	−0.007	0.065 **	−0.006
	(−0.43)	(2.19)	(−0.33)
R-quared	0.732	0.958	0.732
Central region
$ge{c}_{it}$			0.035
			(0.81)
$de{l}_{it}$	−0.026	0.412 ***	−0.040
	(−1.13)	(4.72)	(−1.38)
R-squared	0.901	0.950	0.902
Western region
$ge{c}_{it}$			−0.041
			(−1.24)
$de{l}_{it}$	0.027	0.011	0.028
	(1.10)	(0.13)	(1.12)
R-squared	0.606	0.852	0.613
Northeast region
$ge{c}_{it}$			−0.340
			(−1.11)
$de{l}_{it}$	0.065	−0.227	−0.012
	(0.50)	(−1.71)	(−0.08)
R-squared	0.921	0.995	0.930
Control variables	Yes	Yes	Yes
Individual effects	Yes	Yes	Yes
Time trends	Yes	Yes	Yes

Note: Values in parentheses are t-values, where *** indicates p < 0.01, ** indicates p < 0.05.

.

From the regression results in Table 11, the development of the digital economy has a more significant effect on clean energy consumption in the eastern and central regions than in the western and northeastern regions, and this result is also caused by the fact that the eastern and central regions have a better foundation of economic development and industrial structure than the western and northeastern regions. The coefficients of the impact of clean energy consumption on high-quality economic development in different regions are not significant in Table 11, so we continue to test the mediating effect of clean energy consumption in different regions by bootstrap method. The final test results show that the mediating effect of clean energy consumption in digital economy and high-quality economic development is not significant except for the significant mediating effect in the eastern region. This result may be caused by the limited sample data and the imperfect control variables. However, based on the results of this test, it can still be inferred that the mediating effect of clean energy consumption also has a certain degree of spatial heterogeneity.

5. Conclusions and Policy Recommendations

5.1. Conclusions

Based on the provincial panel data of China from 2011 to 2020, this study deeply investigated the influence process of digital economy on high-quality economic development of China, and based on constructing digital economy development index and high-quality economic development index, it empirically examined the influence process, transmission mechanism, and spatial heterogeneity by using the spatial Durbin model and the intermediary effect model. We analyzed the spatial spillover effect of digital economy on high-quality economic development and the mediating effect of clean energy consumption and explored the spatial heterogeneity of their impact effects. The findings of this paper are as follows:

First, in terms of spatial characteristics, digital economy and high-quality economic development level have obvious global spatial autocorrelation, and the high value agglomerations and low value agglomerations of the digital economy and high-quality economic development have obviously overlapping parts during the study period, and the high–high agglomerations of both are concentrated in the Yangtze River Delta region, while the low–low agglomerations are also basically distributed in the central and western regions. The impact of digital economy on high-quality economic development has obvious spatial heterogeneity. The empirical results verify the hypothesis H1: the digital economy can significantly drive high-quality economic development, and the digital economy has a positive spatial spillover effect on the high-quality development of the neighboring regions. This is an important revelation for local governments to strengthen the application of digital technology and digital infrastructure in all aspects, encourage enterprises to carry out digital transformation, and grasp the initiative of digital economy development.

Secondly, in terms of control variables, the level of urbanization has a significant role in promoting high-quality economic development and negative spatial spillover effects, deciding how to lead the demonstration and radiation of large cities, and reducing the systematic barriers between regions is a test for policymakers; the level of economic development has a significant inhibitory role in high-quality economic development and negative spatial spillover effects, and sacrificing the environment to bring about economic growth is unsustainable. Both the level of technology and industrial structure have significant promotion effects and positive spatial spillover effects on high-quality economic development, and the introduction of high and new technology and the optimization of industrial structure contribute to the region’s high-quality economic development.

Finally, in terms of transmission mechanism, the mediation effect model verifies this paper’s hypothesis H2: the development of the digital economy has a significant role in promoting clean energy consumption; hypothesis H3: clean energy consumption can significantly contribute to high-quality economic development; and hypothesis H4: clean energy consumption has a significant mediating effect on digital economy and high-quality economic development. Moreover, there is a more pronounced space of such transmission effects’ heterogeneity. The development of the digital economy can largely reduce the negative environmental impacts caused by economic development and help towards “carbon neutrality” as a new emission reduction target proposed in recent years. Other countries and not only China should also strive to promote their own carbon-neutral implementation targets, encourage local enterprises to implement carbon-neutral standards, provide emission-reduction strategies, and jointly realize high-quality economic development.

In this paper, we theoretically analyzed and empirically demonstrated the spatial effect and transmission mechanism of digital economy and high economic quality from the perspective of sustainable energy development, but the digital economy is a multidimensional and comprehensive concept of dynamic development, and the connotation of digital economy will be enriched with the continuous updating and iteration of digital technology. Therefore, future research can not only analyze and measure digital economy from different dimensions and study its impact mechanism on high-quality economy development and the promotion of common prosperity, but it can also study the other conduction paths and processes of digital economy affecting high-quality economy development and common prosperity on the basis of giving new features to digital economy, and it can also further expand the research on the impact of the development of digital economy on the carbon emissions of the developing countries and developed countries around the world. The impact of the development of the digital economy on carbon inequality in developing and developed countries around the world can also be further expanded.

5.2. Policy Recommendations

Based on the above findings, this paper proposes the following recommendations to further strengthen the driving role of the digital economy for China’s high-quality economic development:

First, investment in digital infrastructure construction and research and development should be strengthened and the depth and breadth of its application in the real economy expanded. By strengthening the application of digital technology and digital infrastructure in all aspects, oriented by the actual needs of various industries in the whole production process, the government provides financial funds and tax incentives to encourage enterprises to carry out digital transformation, promote network penetration, and improve the coverage and convenience of digital technology application. Enterprises should use digital technology to improve productivity and efficiency, optimize management processes, reduce costs, improve product and service quality, and enhance enterprise competitiveness. At the same time, efforts have been made to reduce the threshold and cost of digital transformation for small- and medium-sized enterprises, strengthen the flow of digital economy factors, and break administrative planning restrictions by relying on platforms such as development zones, industrial parks, and innovation zones, which should be more open to enterprise registration, license issuance, and information sharing. Furthermore, inter-regional cooperation should be strengthened for development and mutually beneficial win–win mechanisms, encouraging leading enterprises in the industry to open up their digital resources.

Second, for jointly realizing the green development goal of energy conservation and emission reduction, the government should strictly restrain local enterprises and factories from engaging in “high-energy-consuming and high-polluting” production through taxation, subsidies, and price adjustments, strictly restraining the landing of high energy consuming projects, collaborating in the development of resource-conserving industries, strengthening the linkage of the development of new types of cross-region industries, enhancing the efficiency of energy utilization to reduce the consumption of energy emissions, and encouraging the development of resource-conserving industries. This also encourages the development of resource-saving industries, improves energy utilization efficiency to reduce energy consumption, promotes the cleanliness of energy supply, and insists on the sustainable development of energy; at the same time, it vigorously promotes the concept of “energy-saving, environmental protection, green and low-carbon” life to the public through the media and various cultural and recreational activities to encourage people to travel in a low-carbon manner, live a green life, and promote clean energy consumption.

Third, the level of digital governance should be enhanced. While the rapid development of the digital economy promotes economic and social prosperity, it also brings serious challenges to economic and social governance due to the imperfections of relevant related laws and regulations. Relevant departments urgently need to incorporate public data services into the public service system and build a unified, national, publicly open data platform and development and utilization ports so as to systematically promote the standardization, standardization, and intensification of the construction of the government platform.

Finally, we should drive sustainable economic development with sustainable energy development. The government should further promote the deep integration of information technology and the real economy and effectively use the promotion of clean energy consumption in the development of digital economy so that more enterprises and factories can feel the economic benefits brought by digital equipment, digital management, and intelligent market forecast, etc. Furthermore, enterprises should spontaneously use energy-saving and emission-reducing production processes and technologies to gradually improve energy-use efficiency, optimize the structure of energy consumption, and use sustainable energy development to drive sustainable economic development effectively. This will promote high-quality economic development and help China transform from a crude economic growth model of “high input, high energy consumption, high pollution and low efficiency” to an intensive economic growth model of “low input, low energy consumption, low pollution and high efficiency”.