How Does Digital Economy Influence Green Mobility for Sustainable Development? Moderating Effect of Policy Instruments

Abstract

The role of green mobility as a low-carbon lifestyle in carbon reduction and sustainable development cannot be ignored. The digital economy effectively promotes green mobility for sustainable energy use in the broader setting of the significant data era and sustainable development. This study utilizes the panel data of 264 cities in China from 2011 to 2021 to construct a two-way fixed-effects regression model to analyze the impact of the digital economy on residents’ green mobility and the indirect impact mechanism of the two policy tools, a low-carbon transportation pilot and carbon emissions trading, from theoretical and empirical aspects. The results show that digital economic development helps promote residents’ green mobility. In addition, the implementation of low-carbon transportation pilots and carbon trading policies has strengthened the role of the digital economy in promoting green mobility. The findings remain after introducing robustness tests such as “smart city” pilots as exogenous shock policies. A heterogeneity study suggests that the effect of the digital economy on green mobility for residents is more significant in economically developed and human capital-rich areas. This study contributes to the literature by providing empirical evidence on the role of the digital economy in promoting sustainable urban transportation and by demonstrating the moderating effects of policy instruments, thereby offering practical insights for policymakers aiming to reduce urban pollution and enhance sustainable development.

1. Introduction

The unsustainable development model of the human economy has led to escalating ecological problems and environmental degradation. Weighing economic development and ecological quality has been one of the critical policy choices faced by countries [1]. While China has made significant economic achievements in the last two decades, its energy demand has been increasing, and the problem of carbon dioxide emissions has become more and more prominent. The continuous increase in CO₂ (carbon dioxide) emissions is the leading cause of the greenhouse effect [2]. According to the World Energy Statistics Yearbook, China emits the most carbon. Yale University and others measured the environmental performance of 180 countries, in which China ranked 120th. Recent studies have explored ways to reduce carbon dioxide emissions in the building sector [3]. For instance, the analysis of embodied carbon in China’s construction industry and optimization of building design parameters for both cost reduction and emission control offer practical insights [4]. Additionally, research on low-carbon insulation design in different ecological zones and assessments of the financial and environmental impacts of exterior building insulation further highlight the importance of low-carbon strategies in the building sector. Sustainable energy use and household travel drive economic growth. Residents’ green energy use awareness is reflected in every consumption behavior [5]. The choice of residents’ travel mode is an essential manifestation of green energy use awareness. According to the UN (United Nations) Emissions Gap Report, household consumption now accounts for two-thirds of total carbon emissions, with transport activities having the most significant impact on carbon emissions [6]. Accelerating the shift in consumption patterns, especially mobility, has become a meaningful way to mitigate climate issues [7]. According to the study, transportation now emits more greenhouse gases in the US (United States) than electricity [8]. The choice of transport mode for the UK (United Kingdom) household activities significantly impacts CO₂ [9]. Road transport contributed the most to China’s 1.14 billion tons of transport sector carbon emissions in 2019. In the face of the continued increase in carbon emissions from global transportation activities, a green transformation in residents’ travel habits must be promoted in China. Since the 14th Five-Year Plan was proposed to promote green development, synergistic promotion of green modes of transportation and conservation of nature has emerged as a crucial support for accelerating the development of green transformation. Green mobility seeks to conserve energy and reduce pollution and can lead to environmental and economic sustainability [6]. In the worldwide environment of low-carbon sustainable development, exploring low-carbon green mobility for residents is vital in alleviating environmental and energy problems.

Sustainability has emerged as a critical global challenge that requires urgent action to balance economic growth, environmental protection, and social equity. Achieving sustainable development involves not only addressing climate change and resource depletion but also transforming how societies operate, particularly in urban areas where the majority of the world’s population resides. In this context, green mobility plays a pivotal role by offering an environmentally friendly alternative to traditional transportation modes that contribute significantly to greenhouse gas emissions. Integrating sustainable practices into urban mobility systems can help cities reduce their carbon footprint, improve air quality, and enhance the overall quality of life for residents. This study explores how the digital economy can facilitate this transformation by leveraging technology and data-driven solutions to promote low-carbon transportation and support the broader goals of sustainable development.

Currently, governments and scholars are conducting research on how to promote green mobility among residents from both governmental and individual perspectives. From the government perspective, researchers have examined the impact of a range of government policy tools on residents’ travel choices and have concluded that residents’ reliance on automobiles has been successfully limited and public transit use has been promoted by government policy tools [10]. Government policy interventions in transport mobility include both transport pricing [10,11,12,13,14] and public transport improvements [15,16,17,18,19] to encourage public transport over private cars. Some scholars have pointed out that Singapore has succeeded in limiting its residents’ reliance on private cars through a series of policy tools, such as road pricing schemes, car purchase restriction schemes, and the improvement of public transport infrastructure [10]. A scholarly study by means of a questionnaire survey found that the city of Madrid, Spain, encouraged residents to shift to public transportation through two restrictive policies, namely low-emission zones and road tolls [20]. Several scholars have found through questionnaires and theoretical studies that tax breaks, direct subsidies, and facilitation policies have promoted the popularity of new energy vehicles and electric vehicles [21,22,23,24]. Shared mobility plays a pivotal role in the transformation of transportation and the promotion of sustainable development. As a key aspect of the digital economy, services like car-sharing contribute significantly to reducing emissions and advancing low-emission transportation solutions [25,26]. In fact, countries around the world have adopted a range of policy tools to support green mobility for their residents. For example, Singapore implemented a permit and road pricing system in 1975; London introduced congestion pricing in 2003; and the United States introduced Transportation Demand Management (TDM) in the 1970s. This shows that a series of policy tools adopted by the government play an essential part in the green mobility of residents. From an individual perspective, some academics have examined the effects of factors such as residents’ environmental awareness, educational attainment, household income, occupational status, positional hierarchy and consumer motivation on residents’ green mobility [27,28,29,30,31].

China has taken a series of measures to support low-carbon development and green mobility. The National Development Commission established eight low-carbon pilot cities in 2010, seeking sustainable urban development. In 2011, the Ministry of Transport launched the first batch of pilot cities for the construction of low-carbon transportation systems in 10 cities, and low-carbon construction gradually penetrated into the field of transportation and travel. The pilot low-carbon transport policy not only includes creating an effective public transport service system but also accelerating the level of urban public transport infrastructure to actively guide the public to choose green modes of transport. The transportation sector is also actively piloting carbon emissions trading. Beijing has included rail transit, buses, rentals, and passenger transportation in its carbon emissions trading system. Shanghai has included public mobile carbon emissions sources such as buses, cabs, and metro in the management of the trading pilot. A series of measures taken by China plays an important role in promoting green mobility.

Along with governmental and personal factors, the development of emerging technologies on the Internet and the digital economy has had a considerable effect on the travel choices, attitudes, and habits of the residents [31,32] that provide new opportunities for realizing low-carbon mobility. According to the “14th Five-Year Plan for Digital Transportation”, digitalization, networking, and intelligence have become a key driving force behind green transportation and low-carbon development. Indeed, China is expanding its investment in new energy vehicles and public transportation, among other things, to strengthen the utility of digital technologies for developing green transportation. For example, the Changsha Traffic Police and Tencent used digital tools to jointly create the Internet Precision Bus. The effect of green mobility is remarkable, and 25% of residents shifted from cars to environmentally friendly transportation.

However, the scholarly community is still divided on how the digital economy affects residents’ green mobility. The first view is that digital technologies can promote green mobility for residents. Digital technologies have led to innovations in building and infrastructure solutions to optimize the built environment and make it smarter [33,34]. An ideal built environment can incentivize residents to use the trolley and public transportation more often [35,36,37]. According to the study, digitization and the improvement of Internet technologies have helped drive the development of Mobility as a Service (MaaS). MaaS is based on the modes of transportation that exist today and uses digital technology to integrate and customize efficient, cost-effective, and low-carbon travel solutions for travelers. MaaS solutions with shared access help shift from private vehicles to public or shared modes [38,39,40,41], contributing to green and low-carbon mobility. Another view is that digitization and broadband Internet access have reduced the use of public and active transportation [32]. The development of the Internet may result in the fragmentation of work activities, and given the flexibility of private vehicles, public transportation is generally not used to implement fragmented activities [42,43]. Therefore, in the worldwide environment of low-carbon sustainable development, it is crucial to study how the digital economy can successfully nudge green mobility.

According to a survey of the pertinent literature in this research, only a few academics have focused on the critical role of digital technology and policy instruments for green mobility. For example, existing research shows that intelligent mobility platforms have great potential for green mobility [41,44,45,46]. Additionally, policy tools favorably affect green mobility [36,37]. In conclusion, although some academics have concentrated on the critical role of digital technology and policy tools for green mobility, they have yet to take an empirical analysis approach to test the relationship between them. Moreover, there is no literature on the moderating role played by policy instruments in the digital economy’s impact on residents’ green mobility. In order to further the research in this sector, this paper examines the moderating function of two policy tools, namely, a low-carbon transportation pilot and carbon emissions trading, from both theoretical and empirical aspects. It also introduces “smart city” pilots as exogenous shock policies and other robustness tests to confirm the validity of the empirical results. This is of great theoretical and practical significance for promoting low-carbon and green transportation for residents.

The research objectives of this paper are (1) to theoretically analyze the transmission mechanism of the digital economy on residents’ green mobility and the moderating role of policy instruments in this transmission mechanism, and (2) panel data of 264 Chinese cities from 2011–2021 are utilized as a macro sample to empirically examine the impact of the digital economy on residents’ green mobility and the moderating role of policy instruments. It provides empirical evidence for the promotion of low-carbon and green mobility modes in China. First, the influence of the digital economy on green mobility is investigated. Second, the moderating effects of policy instruments between the digital economy and green mobility are analyzed. Finally, the paper conducts heterogeneity tests for different economic and human capital levels to further analyze whether the moderating effects of the policy instruments are still significant under the condition of heterogeneity. This paper provides some supportive evidence in discussing the digital economy’s promotion of green mobility for residents. It explains how digital tools and low-carbon policies can be used to promote green mobility and thus promote green and sustainable development in China.

This study is novel in two ways: (1) This paper enhances green mobility research. This paper innovatively links the digital economy, policy tools, and residents’ green mobility to show their essential relationship. This paper highlights the critical role of the digital economy and the moderating effect of policy tools, which provides new perspectives and support for the promotion of green mobility. (2) This paper uses novel research methods. First, this study analyzes the digital economy, policy instruments, and inhabitants’ green mobility using theoretical and empirical methods. The moderating effects of two policy tools, a low-carbon transportation pilot and carbon emissions trading, are examined in depth using Chinese city-level data. The empirical part is analyzed by constructing a two-way fixed effects regression model, which eliminates the influence of factors that vary with individuals and time. In the previous literature, most academics have examined the effect of MaaS on the sharing economy and public transportation only through theoretical analysis. Second, in order to more effectively evaluate the influential role of promoting residents’ green mobility, this paper innovatively adopts the “smart city” pilot as an exogenous shock policy and evaluates this real-world problem by the DID method. Third, considering the time lag in the implementation of policy instruments, this paper lags the explanatory variables by one period to examine the robustness of the results. Finally, this paper examines the essential roles of the economy and human capital in residents’ green mobility and analyzes the heterogeneity from these two aspects. It bridges the gap in this research area.

The remaining sections of the essay are arranged as follows. The second section deals with the theoretical mechanisms. The third section is economic modeling and data sources. The fourth section is the empirical analysis. The fifth section is the discussion. The sixth section is the conclusion and policy implications.

2. Theoretical Mechanisms

2.1. Digital Economy and Green Mobility

The successful use of digital technology as a significant driving force, digitized information as a vital production ingredient, and the Internet as a fundamental carrier are all characteristics of the emerging economic model known as the digital economy. The imbalance between the supply and demand for public transportation and the rise in air pollution caused by China’s fast urbanization can be reduced by its quick development. Digital technologies such as digital twins and intelligent machines are integrated into the process of transportation infrastructure in order to promote low-carbon and green mobility for residents. Specifically, the following is a summary of how the growth of the digital economy has affected local inhabitants’ use of green mobility. First, the explosion of digital technology is the basis for digital applications such as intelligent network support platforms and MaaS platforms. These transportation service platforms combine environmental impacts to provide travelers with green, low-carbon, and shared modes of transportation as a substitute for private car travel options [38,44,45]. In addition, advances in digital technology have led to innovations in building and infrastructure solutions to optimize the built environment [33,34]. An ideal built environment can incentivize residents to use trams and public transportation more often, thus promoting green mobility for residents [35,36,37]. Finally, the growth of the digital economy can make resource allocation more efficient and boost the technical capabilities of new energy cars, which will encourage people to buy these vehicles [47,48,49,50]. General Secretary Xi Jinping emphasized that to hasten the emergence of green and environmentally friendly modes of transportation, it is vital to promote new energy, intelligence, and digitalization and encourage and guide green travel. With the increasing prominence of ecological problems and the government’s focus on green mobility issues, the utility of digital technologies to empower new energy and public transportation infrastructure is being strengthened everywhere. In light of this, we suggest the following hypothesis.

Hypothesis 1. 

The increase in the level of development of the digital economy can effectively promote green mobility among residents.

2.2. Moderating Mechanisms of Policy Instruments

Policy tools are seen as necessary in promoting low-carbon green mobility for residents and contributing to sustainable development. Amid the current explosive growth of the digital economy, a series of government policies is crucial for developing green transportation using advanced digital technologies. Low-emission zone and urban parking restriction policies motivate residents to switch from private cars to green transportation through banning polluting vehicles from specific areas and restricting parking [37]. The combination of automobile purchase restriction policies, which reduce the use of private cars at the root cause, and public transportation infrastructure improvements, which motivate residents to use public transportation, has an even more significant effect on green mobility [36]. In addition, a series of government economic policies can significantly encourage people to buy new energy cars [51,52,53,54]. Examples include direct subsidies for new-energy vehicle purchases, policies to promote the pilot range of new-energy vehicles and exemptions from the purchase tax. Besides the economic strategy of providing subsidies, the license plate control policy restricts consumers from purchasing cars in favor of new energy vehicles, for which licenses are easier to obtain [16,55]. Therefore, a series of government policy tools can limit residents’ dependence on cars and promote public transportation and the usage of new energy cars.

Against the backdrop of a booming digital economy centered on the Internet and big data, a series of policies proposed by the government provides strong support for promoting green mobility among residents. For example, the low-carbon transport pilot policy will expand investment in new energy, data centers and transport infrastructure, accelerating the integration of digital technology and the public transport system. With the promotion of low-carbon transportation policies, measures such as creating a public travel information service system, optimizing urban public transport and improving the carbon emission management system for transportation will effectively promote intelligent low-carbon transportation. When considering the time spent, the MaaS platform based on digital technology provides travelers with efficient, economical, and low-carbon travel solutions through one-stop travel planning services [56]. Travelers are more inclined to use MaaS platforms to plan travel options for themselves, thus promoting low-carbon, sustainable transportation. In addition, the implementation of carbon reduction policies has led to green innovations in digital technology. For example, a digital technology-based “precision bus” provides travelers with convenient, comfortable, and personalized travel services. Changsha traffic police and Tencent use digital tools to jointly create the Internet precision bus. With 25% of residents changing from private cars to public transportation, the green low-carbon travel effect is remarkable. Accordingly, we propose the following hypothesis.

Hypothesis 2. 

Policy instruments play a positive moderating role in the digital economy for green mobility.

3. Economic Modeling and Data Sources

3.1. Variables and Data Sources

3.1.1. Variables

Green mobility (GC) is the explained variable. Green mobility is when residents use modes of travel that have less impact on the environment. The majority of researchers employ questionnaires to look at the effects of various factors on green mobility [20]. Public transportation travel is the most important form of green mobility, and in this study, the number of public transportation operating vehicles per capita was chosen as the explained variable.

Main explanatory variable: the digital economy (DE). Considering the current state of research on the DE, most scholars have created a multi-indicator evaluation system to measure the overall performance of the DE [57]. Therefore, on the basis of the indicator system established by previous studies, this paper selects five sub-indicators to measure the level of the DE in each city. The entropy value method is used to assign weights to each indicator, and

Table 1. DE indicator system.

Variable	Sub-Indicators	Indicator Weights
DE	Telecommunications business (per 100 people)	0.3185
	Percentage of computer workers	0.2788
	Number of Internet users (per 100 people)	0.1827
	Number of cell phone subscribers (per 100 people)	0.1356
	Digital Inclusive Finance Index	0.0844

displays the indicator scheme.

Moderating variables: This paper chooses the low-carbon transportation pilot policy (LCTP) and carbon emissions trading policy (CETP) as the moderating variables. For cities within the scope of the policy, the value is 1 for the year of the policy and thereafter and 0 for all others.

Based on previous research on the DE and green mobility, five control variables were selected for this paper. Economic growth (PGDP): Economic growth is measured by the per capita GDP [58]. Population size (PS): The logarithm of the total population is used to define population size [59]. Human capital (HC) is measured by the share of undergraduate and specialized students in the total population [60]. Industrial structure (IS): The ratio of tertiary sector output to secondary sector output measures industrial structure [58]. Education of residents (EP) uses the logarithm of expenditure on education to measure the education of residents [61]. The variable description is shown in

Table 2. Variable declaration table.

Variable	Abbreviation	Measurement	Units
Digital economy	DE	The results of the entropy value method	/
Green mobility	GC	The number of public transportation operating vehicles per capita	unit/10,000 people
Low-carbon transportation pilot	LCTP	The policy being implemented is represented by a value of 1, and not being implemented is represented by a value of 0.	/
Carbon emissions trading policy	CETP	The policy being implemented is represented by a value of 1, and not being implemented is represented by a value of 0.	/
Population size	PS	The logarithm of the total population	/
Economic growth	PGDP	Per capita GDP	100 million yuan/10,000 people
Human capital	HC	The share of undergraduate and specialized students in the total population	per 10,000 people
Education of residents	EP	The logarithm of expenditure on education	/
Industrial Structure	IS	The ratio of tertiary sector output to secondary sector output	%

.

3.1.2. Data Sources

This analysis uses 2011–2021 panel data from 264 Chinese cities as a macro sample. Missing data for 2020 regarding the number of public bus and electric vehicle operations at year-end were filled in using interpolation, specifically by taking the average of the 2019 and 2021 data, due to data unavailability caused by the COVID-19 pandemic. The information on digital economy indicators in this paper comes from the China Urban Statistical Yearbook and the Center for Digital Inclusion Finance at Peking University. The list of low-carbon transportation pilot cities and carbon emissions trading cities is from the National Development and Reform Commission. Public transportation and other data were obtained from China Urban Statistical Yearbook, China Statistical Yearbook, and provincial yearbooks.

Table 3. Descriptive statistics.

Variable	Observations	Mean	Standard Deviation	Min	Max
DE	2904	0.110	0.066	0.013	0.638
GC	2904	3.639	6.333	0.102	110.5
LCTP	2904	0.125	0.331	0	1
CETP	2904	0.038	0.191	0	1
PS	2904	5.919	0.678	2.970	8.136
PGDP	2904	6.051	5.531	0.618	53.24
HC	2904	197.1	248.3	0.437	1398
EP	2904	13.20	0.794	9.906	16.25
IS	2904	1.037	0.565	0	5.348

shows the results of descriptive statistics for each variable.

3.2. Measurement Model Setting

Before conducting the empirical analysis, this paper determines whether the model has multicollinearity by testing the variables for VIF (variance inflation factor) [60]. The VIF is less than 10, indicating no major multicollinearity between variables. The Hausman test was used to determine whether this paper is suitable for the fixed effects model [62]. The results indicate that this paper is suitable for the fixed effects model. As a result, this study uses a fixed effect model to examine how residents’ green mobility is affected by the DE.

This paper preliminarily establishes a multiple regression model of digital economic development and green mobility. The construction of Model I is as follows.

$${GC}_{i,t}={\beta }_0+{\beta }_1{DE}_{i,t}+{\beta }_2{Z}_{i,t}+{\mu }_i+{\delta }_t+{\epsilon }_{i,t}$$

(1)

${GC}_{i,t}$ stands for the level of green mobility. ${DE}_{i,t}$ is digital economy. ${\mu }_i$ denotes an individual fixed effect. ${\delta }_t$ denotes a time fixed effect.

This paper examines the moderating effect by including an interaction term of the explanatory and moderating variables in the baseline regression to test hypothesis 2. Therefore, the moderating effects of the low-carbon transportation pilot and carbon emissions trading policy are examined by introducing an interaction term. Models II and III are constructed as follows.

$${GC}_{i,t}={\alpha }_0+{\alpha }_1{DE}_{i,t}+{\alpha }_2({DE}_{i,t}\times {LCTP}_{i,t})+{\alpha }_3{Z}_{i,t}+{\mu }_i+{\delta }_t+{\epsilon }_{i,t}$$

(2)

$${GC}_{i,t}={\gamma }_0+{\gamma }_1{DE}_{i,t}+{\gamma }_2({DE}_{i,t}\times {CETP}_{i,t})+{\gamma }_3{Z}_{i,t}+{\mu }_i+{\delta }_t+{\epsilon }_{i,t}$$

(3)

4. Empirical Results

4.1. Baseline Model

The findings of the baseline regression for the digital economy and green mobility are shown in

Table 4. Benchmarking regression results of digital economic development on green mobility.

Variables	(1)	(2)	(3)	(4)
	GC	GC	GC	GC
DE	17.16 ***	15.83 ***
	(9.42)	(8.64)
L.DE			9.014 ***	8.242 ***
			(4.51)	(4.01)
PGDP		0.114 ***		0.075 **
		(3.78)		(2.15)
PS		−6.855 ***		−6.263 ***
		(−9.99)		(−8.26)
IS		−0.414 ***		−0.331 **
		(−2.80)		(−2.03)
HC		0.003 ***		0.003 ***
		(3.14)		(3.02)
EP		−0.257		−0.018
		(−0.93)		(−0.06)
Constant	2.067 ***	45.23 ***	2.678 ***	39.36 ***
	(13.70)	(10.34)	(16.90)	(8.18)
Year FE	YES	YES	YES	YES
City FE	YES	YES	YES	YES
Observations	2904	2904	2640	2640
R-squared	0.063	0.121	0.033	0.078

Note: The value of t-statistics is reported in parentheses; ** p < 0.05, *** p < 0.01.

. Column (1) shows that the impact of the digital economy on green mobility is significantly positive without considering the control variables. After taking into consideration additional factors that can have an effect on green mobility, the results in column (2) demonstrate that the coefficient is still significantly positive. This suggests that the DE’s development has a favorable effect on green mobility. The more advanced the city’s digital economy is, the more technological assistance it can offer for the creation of MaaS service platforms, thus promoting low-carbon green mobility for residents [44]. Digital technology-based “precision transit” provides travelers with convenient, comfortable, and personalized travel services, thus encouraging residents to switch from personal cars to public transportation. In addition, since the change in the mode of travel of residents is a long-term process, the lagged variable model allows for a more objective description of the economic phenomenon [63]. Therefore, this paper lags the relevant explanatory variables by one year to test the role of DE on green mobility, and the results are shown in columns (3) and (4). The digital economy’s impact on green mobility remains significant. This finding supports hypothesis 1. The scatter plot fitting of DE and GC is shown in Figure 1. It can be observed that DE and GC are positively correlated.

The role of economic growth (PGDP) cannot be ignored. This is because economic growth can be accompanied by an increase in public transport infrastructure, the vigorous development of rail transit, and the optimization of route layouts, thus promoting public transport travel for residents. Population size (PS) has a significant negative impact. As the population size continues to grow, the lack of supply of public transportation in the city inhibits its development. Industrial structure (IS) has a negative regression coefficient on green mobility. This indicates that industrial structure upgrading does not promote residents’ green mobility well. Human capital (HC) is significantly positive. This indicates that the higher the level of HC, the more the residents have the right values of green development and thus tend to choose the green mode of transportation.

4.2. Moderating Effects of Policy Instruments

Model II and Model III are estimated to confirm the moderating effects of the low-carbon transportation pilot and the carbon emissions trading policy. As can be seen in columns (1–4) of

Table 5. Moderating effects of policy instruments.

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	GC	GC	GC	GC	GC	GC
DE	12.93 ***	12.04 ***	12.69 ***	11.65 ***
	(6.56)	(6.17)	(6.58)	(6.05)
DE × LCTP	18.68 ***	18.18 ***
	(5.46)	(5.43)
DE × CETP			17.96 ***	17.79 ***
			(6.57)	(6.62)
L.DE					6.573 ***	5.809 ***
					(3.00)	(2.70)
L.DE × L.LCTP					7.917 **
					(2.19)
L.DE × L.CETP						10.99 ***
						(3.72)
Constant	2.225 ***	44.80 ***	2.123 ***	45.98 ***	39.53 ***	40.00 ***
	(14.56)	(10.30)	(14.16)	(10.59)	(8.22)	(8.33)
Year FE	YES	YES	YES	YES	YES	YES
City FE	YES	YES	YES	YES	YES	YES
Control variables	NO	YES	NO	YES	YES	YES
Observations	2904	2904	2904	2904	2640	2640
R-squared	0.074	0.131	0.078	0.136	0.080	0.083

Note: The value of t-statistics is reported in parentheses; ** p < 0.05, *** p < 0.01.

, the coefficients on both policy instruments and the digital economy interaction term are significantly positive. This suggests that the moderating effect of both policy instruments is significant. Considering the lagged nature of policy effects and with the aim to mitigate endogeneity problems, this paper introduces first-order lagged terms for the relevant explanatory variables in the regressions. The coefficients on the first-order lagged terms for both the digital economy and the policy instruments are significant, as shown in columns (5–6). This illustrates that the moderating effect of the two policy instruments selected in this paper still exists after the endogeneity problem is taken into account. This suggests that the positive impact of the growth of the DE on green mobility is positively moderated by policy instruments. The series of policy support proposed by the government will expand investment in building a public travel information service platform, a sound transportation carbon management system, and transportation infrastructure. In the midst of the thriving digital economy centered on the Internet and big data, this policy support can effectively promote residents’ green travel. The conclusion supports hypothesis 2.

4.3. Endogenous Analysis

This study’s fixed effects and lagged variable models partially tackle the endogeneity problem created by omitted variables. However, the level of the DE may also be affected by the green mobility of residents. This implies a bidirectional causal relationship between the two. This paper utilizes the method of instrumental variables to further reduce the estimation error, drawing on the practice of related scholars. The number of landline phones (IV1) and post offices (IV2) in 1984 are used in this study as instrumental variables for the DE [64]. Because the data chosen are in cross-section mode, the data are transformed into panel data by multiplying them by the amount of Internet investment and the Internet users in the prior year [65]. The results show that IV1 and IV2 are highly correlated with the DE. Moreover, both selected IVs passed the weak instrumental variable test and unidentifiable test. In

Table 6. Instrumental variable regression.

Variables	(1)	(3)	(2)	(4)
	DE	GC	DE	GC
DE		32.58 ***		27.78 ***
		(12.29)		(4.84)
IV1	0.021 ***
	(45.93)
IV2			0.324 ***
			(16.48)
Constant	−0.234 ***	59.59 ***	−0.076 *	60.70 ***
	(−6.63)	(11.36)	(−1.68)	(11.38)
Kleibergen–Paap rk LM	1294 ***		272.4 ***
Kleibergen–Paap rk Wald F	2109		271.6
	{16.38}		{16.38}
Year FE	YES	YES	YES	YES
City FE	YES	YES	YES	YES
Control variables	YES	YES	YES	YES
Observations	2904	2904	2904	2904
R-squared	0.723	0.093	0.802	0.107

Note: The value of t-statistics reported in parentheses; * p < 0.1, *** p < 0.01; “{}” show critical values.

, after considering the endogeneity, the DE can still significantly promote residents’ green mobility, and the research findings are reliable.

4.4. Robustness Test

This research conducts the robustness test by adding Province × Year fixed effects, replacing the explained variable and the “smart city” pilot as an exogenous shock policy to ensure the accuracy and reliability of the empirical results.

(1)

Inclusion of Province × Year fixed effects: By setting Province × Year interaction effects, systematic changes in macro factors can be ruled out. The results are shown in columns (1–2) of

Table 7. Robustness check.

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	GC	GC	RGC	RGC	RGC	RGC
DE	15.37 ***	6.154 ***	192.4 ***	127.8 ***
	(6.93)	(2.62)	(8.16)	(5.59)
SC					5.381 ***	4.788 **
					(2.63)	(2.56)
DE × LCTP	16.87 ***		71.70 *
	(4.40)		(1.77)
DE × CETP		42.56 ***		338.4 ***
		(10.63)		(10.61)
SC × LCTP					217.2 ***
					(5.21)
SC × CETP						824.9 ***
						(21.98)
Constant	52.52 ***	44.41 ***	1046 ***	1062 ***	1260 ***	1090 ***
	(9.43)	(8.04)	(19.90)	(20.62)	(19.34)	(18.11)
Control variables	YES	YES	YES	YES	YES	YES
Year FE	YES	YES	YES	YES	YES	YES
City FE	YES	YES	YES	YES	YES	YES
Year × Province	YES	YES	NO	NO	YES	YES
Observations	2904	2904	2904	2904	2904	2904
R-squared	0.195	0.225	0.359	0.385	0.443	0.533

Note: The value of t-statistics is reported in parentheses; * p < 0.1, ** p < 0.05, *** p < 0.01.

, and the previous findings of the article remain robust.

(2)

Replacement of explained variable: The explained variable in the baseline model is replaced with total public passenger transportation (RGC). The regression results are shown in (3–4). The model is robust.

(3)

“Smart city” pilot as exogenous policy shock (SC): Based on the inclusion of Province × Year fixed effects and replacement of the explained variable. This paper adopts the “smart city” pilot as an exogenous policy shock to assess the robustness of the digital economy to promote residents’ green mobility using the DID method. For cities within the scope of the policy, the value is 1 for the year of the policy and thereafter and 0 for all others. Columns (5–6) show regression findings that match baseline regression.

4.5. Heterogeneity Analysis

4.5.1. Economic Development

The population’s per capita disposable income has increased as economic development levels have risen. Residents started to focus on the enjoyment of the natural environment and spiritual literacy once the material quality of life improved. However, the disparity in income between urban and rural Chinese citizens has remained substantial since the country’s opening up and reforming. This means that many low-income people are still more concerned about material life. This section examines whether the digital economy and the role of policy instruments differ significantly across levels of economic development. We categorize each city based on whether GDP per capita is above the median. Regions above the median are characterized as high economic development regions, and regions below the median are characterized as low economic development regions.

Table 8. Heterogeneity test for level of economic development.

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	High economic development regions			Low economic development regions
	GC	GC	GC	GC	GC	GC
DE	22.33 ***	16.75 ***	13.98 ***	−5.620	−5.519	−5.417
	(10.66)	(7.29)	(6.14)	(−1.19)	(−1.16)	(−1.13)
DE × LCTP		15.84 ***			2.632
		(4.40)			(0.15)
DE × CETP			24.56 ***			−0.880
			(7.28)			(−0.15)
Constant	64.02 ***	69.68 ***	70.79 ***	7.831	7.761	7.613
	(10.87)	(11.48)	(11.82)	(1.07)	(1.06)	(1.03)
Year FE	YES	YES	YES	YES	YES	YES
City FE	YES	YES	YES	YES	YES	YES
Control variables	YES	YES	YES	YES	YES	YES
Observations	1452	1452	1452	1452	1452	1452
R-squared	0.232	0.253	0.273	0.026	0.026	0.026

Note: The value of t-statistics is reported in parentheses; *** p < 0.01.

shows that the effect of DE on green mobility is significant in regions with high levels of economic development. The moderating effect of the low-carbon transportation pilot and the carbon emissions trading policy is significant. This is due to the fact that smart transportation infrastructure is better in economically developed regions. The digital economy is better integrated in terms of precision public transport and digital transportation, which is effective in promoting green mobility among residents. On the other hand, in areas with a lower level of economic development, public transportation is underdeveloped, and the role of DE in low-carbon transportation cannot be effectively utilized. Therefore, in regions with higher levels of economic development, the effects of the DE and policy tools are significant.

4.5.2. Human Capital

The degree to which locals are educated is a significant factor in the city’s ability to develop sustainably. Education makes the younger generation aware of the critical role of their own behavior in natural environmental protection and thus choose greener travel modes. Human capital (HC) is measured by the share of undergraduate and specialized students in the total population. This section examines whether the role of the DE and policy tools differs across human capital levels. We categorize each city based on the median human capital level. Regions above the median are high-human-capital regions, and regions below the median are characterized as low-human-capital regions.

The human capital heterogeneity analysis results are in

Table 9. Human capital heterogeneity test.

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	High-human-capital regions			Low-human-capital regions
	GC	GC	GC	GC	GC	GC
DE	20.60 ***	16.97 ***	14.58 ***	−4.513	−4.511	−4.460
	(9.76)	(7.45)	(6.42)	(−1.06)	(−1.06)	(−1.05)
DE × LCTP		14.39 ***			44.41
		(4.09)			(0.31)
DE × CETP			21.47 ***			−1.079
			(6.57)			(−0.20)
Constant	69.59 ***	69.13 ***	69.04 ***	2.472	2.486	2.316
	(11.94)	(11.93)	(12.04)	(0.30)	(0.30)	(0.28)
Year FE	YES	YES	YES	YES	YES	YES
City FE	YES	YES	YES	YES	YES	YES
Control variables	YES	YES	YES	YES	YES	YES
Observations	1452	1452	1452	1452	1452	1452
R-squared	0.219	0.229	0.245	0.039	0.039	0.039

Note: The value of t-statistics reported in parentheses; *** p < 0.01.

. For high-human-capital regions, green travel can be favorably impacted by the digital economy. The moderating effects of the low-carbon transportation pilot and the carbon emissions trading policy are significant. For low-human-capital regions, green mobility is not affected. This indicates that the higher the degree of human capital, the easier it is for residents to master MaaS transportation platforms or other digital transportation modes, and the easier it is for the digital economy to play a role in promoting green mobility.

5. Discussion

The pursuit of green mobility not only saves resources and reduces pollution but also encourages the sustainable development of cities and the building of ecological civilization. Studies have already shown that the growth of digital technology can efficiently encourage green and low-carbon travel modes such as public transportation and sharing. In addition, a series of policies proposed by the government provides strong support for promoting green mobility among residents. Based on the empirical results in Section 4, the research in this paper is discussed as follows.

First, we find that the development of the DE can effectively promote green mobility for residents. Sochor et al. (2018) can prove the point of this article. He pointed out that MaaS platforms based on digital technologies can facilitate the choice of shared mobility and sustainable mobility modes [46]. This is because the development of digital technology is the basis for the advancement of digital applications such as smart grid support platforms and MaaS platforms. These transportation service platforms combine environmental impacts to provide travelers with green, low-carbon, and shared modes of transportation as an alternative to private car travel options [38,44,45]. Therefore, this paper concludes that the growth of the DE can successfully encourage residents’ green mobility.

Second, this paper examines the moderating effects of two policy instruments: the low-carbon transportation pilot policy and the carbon emissions trading policy. By looking at the coefficients of the interaction terms between the policy instruments and the DE, we can infer that both policy instruments are able to positively regulate the promotion effect of the DE on green mobility. This confirms the study of Gonzalez et al. (2022) that low-emission zone policies and urban parking restrictions encourage residents to switch from private cars to green transportation [20]. In addition, Diao (2019) studied that Singapore has succeeded in limiting residents’ reliance on personal automobiles and promoting its adoption of public transportation through a range of policy tools such as road pricing schemes, car purchase restriction schemes, improvements in public transportation infrastructure, and transport planning [10]. Specifically, the low-carbon transportation pilot and the carbon emissions trading policy are studied in this paper. Low-carbon transportation pilot policies include accelerating the construction of green transportation infrastructure, creating an efficiently connected and fast and comfortable public transportation service system and promoting the low-carbon transformation of transportation means promoting green mobility. The carbon emissions trading mechanism incorporates projects such as voluntary certified emission reductions in transportation, promotion of the implementation of new energy vehicles, and transportation of stationary and mobile sources. A series of government policies are crucial to the growth of green transportation using advanced digital technologies as a means. Therefore, this article makes the case that low-carbon transportation pilot policies and carbon emissions trading policies can enhance the DE’s ability to promote green mobility.

Third, the heterogeneity analysis in this paper shows that there are significant differences in the contribution of the DE to the promotion of green mobility among residents in response to changes in the level of the economy and human capital. Specifically, in regions of high economic development and human capital, the DE can effectively promote green mobility among residents. Moreover, policy tools can play a positive moderating role. This is because, in cities with high economic development and education, it is easier for residents to master MaaS transportation platforms or other digital transportation modes and the easier it is for them to take advantage of the digital economy’s contribution to green mobility. The conclusions of this study have significant ramifications for how low-carbon policy and budgetary measures are implemented in various nations.

Furthermore, our results align with recent empirical studies in the field of urban economic development and sustainability. Michalina et al. (2021) emphasize the importance of comprehensive sustainability frameworks [66], which resonates with our finding that the digital economy fosters green mobility by leveraging digital platforms such as MaaS. This is consistent with Sochor et al. (2018) and Amatuni et al. (2020), who found that digital platforms contribute to reducing private car use and encourage green travel modes like shared mobility [46,67]. Additionally, our findings regarding the regional heterogeneity of the digital economy’s impact echo Guo et al. (2023), who highlighted that economic structures and local policies play a critical role in sustainability outcomes [68]. In line with these studies, we propose that future research further explores the broader integration of sustainability metrics to enhance the robustness of such analyses.

When comparing with international experiences, Singapore stands out as a leading example, having successfully implemented policies like road pricing and vehicle quota systems that promote public transportation over private car use [10]. Similarly, London’s congestion pricing, introduced in 2003, has proven effective in reducing traffic congestion and encouraging the use of public transport and cycling. The shift to low-emission zones and the promotion of shared mobility solutions in European cities like Madrid and Stockholm further demonstrate the positive role that policy instruments play in green mobility. These examples illustrate how a combination of regulatory measures and digital technologies can significantly reduce carbon emissions from the transport sector.

6. Conclusions and Policy Implications

6.1. Conclusions

In the context of the era of big data and sustainable development, the Government must formulate appropriate fiscal measures and low-carbon policies to promote the green transformation of residents’ travel modes. The growth of the DE provides an opportunity to promote green travel for residents. Although there is already literature that has preliminarily examined the various factors affecting residents’ public transportation travel, there are still research gaps. This paper empirically examines the DE and the effective role of two policy instruments based on panel data for 264 cities from 2011–2021. The conclusions drawn are as follows:

(1)

The growth of DE can effectively promote residents’ green mobility. After taking into account the fact that the residents’ change of mobility mode is a long-term process, the relevant explanatory variables will be lagged one year to test the role of the DE on green mobility, and the conclusion is still significant

(2)

Low-carbon transportation pilots and carbon trading policies positively regulate the promotion of green mobility by the DE. This indicates that the digital economy is more effective in promoting green mobility for residents in the policy pilot cities.

(3)

This paper further investigates the differences in the roles of the DE and policy tools between regions with different levels of economic and different human capital. The promotion effect of DE on green mobility is more significant in regions with high economic development and regions with high human capital. Furthermore, the moderating effect of policy tools is also more significant.

These findings highlight the pivotal role that the digital economy (DE) plays in driving green mobility, particularly when supported by policy instruments such as low-carbon transportation pilots and carbon trading. The significant effect of these policies in regions with higher economic development and human capital underscores the need for targeted approaches that consider regional disparities. Additionally, the robustness of our findings, even after accounting for time-lagged variables, provides strong empirical evidence that the DE is a key driver in the transition towards sustainable transportation.

6.2. Limitations

This study has several limitations that should be acknowledged. First, the analysis is constrained by the availability of data, which may limit the comprehensiveness of the measures used for both the digital economy and green mobility. While the study employs panel data from 264 cities in China, certain variables may not fully capture the complexities and nuances of these concepts. For instance, the digital economy is measured using a set of indicators that may not cover all aspects of digital transformation, such as emerging technologies or digital infrastructure quality. Similarly, green mobility is primarily assessed through public transportation metrics, which may not encompass other sustainable travel behaviors like cycling or walking.

Second, there could be potential unobserved factors influencing the results. Although the use of a fixed effects regression model helps to control for time-invariant factors, some regional characteristics, cultural differences, or local policies not included in the model may still affect the relationship between the digital economy and green mobility.

6.3. Policy Implications

First, digital technologies need to be integrated into public transport infrastructure, intelligent network support platforms, MaaS transport service platforms, and other areas to enhance the digitization, networking, and intelligence of the public transport system. It is vital to improve public transport travel services, increase investment in the public transport sector, and encourage tram travel to promote green and low-carbon mobility among residents. The government should integrate big data and internet technology to implement spatial planning and set aside public space for transportation systems. A combination of digital technology in transportation system design to encourage walking and public transportation can be used, focusing on the origin and destination of non-motorized trips and transfers midway. Provide residents with comfortable and convenient public transportation to promote green travel.

Secondly, this research makes the following policy recommendations by analyzing the moderating effects of the low-carbon transportation pilot and the carbon emissions trading policy. Government policymakers should guide the public to prioritize green modes of travel such as public transport, walking, and cycling when formulating low-carbon policies. Examples include lowering public transportation fares, implementing congestion pricing programs, restricting the issuance of motor vehicle licenses, and urban parking restrictions. The heterogeneity analysis shows that the government should consider the local level of economic development and human capital when implementing policy instruments. In cities with low economic and human capital, the digital economy and policy tools are not as effective in greening travel. Therefore, before transforming the way residents travel, local economic and educational investments should be increased to improve their disposable income and quality education.