The Impact of Investment Efficiency in the Digital Economy on Urban Waste Reduction: Evidence from China

Abstract

This study aims to explore the relationship between the development of the digital economy and urban waste management practices, with a specific focus on their impact on landfill and incineration disposal methods. The research objective is to enhance our understanding of interdependencies in these domains and offer insights for the formulation of more efficient waste management strategies. Through empirical analysis, the study shows a significant influence of the digital economy on urban waste disposal volumes. Moreover, the provided results show a negative impact of the efficiency of digital infrastructure investment on urban waste disposal volumes. These findings suggest that enhancing the efficiency of investment in digital infrastructure could alleviate the strain on waste disposal strategies, thus helping to reduce environmental pollution. The findings of this study provide valuable insights and suggest directions for future research in pursuit of sustainable waste management practices in the evolving context of the digital economy.

1. Introduction

The exponential growth of the digital economy is bringing about unprecedented transformations in our everyday lives. However, along with the convenience and innovation it brings, this rapid advancement also presents its own set of challenges. One such challenge is the alarming increase in urban domestic waste. The growth of the digital economy has spurred consumption, which in turn has led to a rise in urban domestic waste. It’s worth noting that digital shopping and traditional offline economic activities generate different types of waste (Bai et al. [1]). Digital shopping often results in electronic waste, such as old electronic devices and packaging materials, whereas traditional offline shopping may generate more organic waste and paper trash (Liang and Li [2]). According to the China Internet Development Report 2023 [3] and the China Urban Statistics Yearbook 2022 [4], the scale of China’s digital economy and the volume of urban waste disposal are increasing. Figure 1 illustrates an upward trend in both the digital economy and total waste treatment from 2017 to 2021. However, we observed a decrease in waste processing volume in 2020. This decline can be attributed to the global outbreak of the pandemic, which resulted in a significant reduction in offline economic activities. As the pandemic spread, traditional offline shopping and brick-and-mortar store operations were severely restricted, leading to a shift in consumer behavior. The reduction in these economic activities resulted in a decrease in consumption to some extent, thereby reducing the demand for manufacturing industries such as disposable packaging materials and electronic products. Specifically, the digital economy surged from approximately 29 trillion RMB in 2017 to around 40 trillion RMB in 2021, marking a notable growth of 37.93%. Simultaneously, waste treatment increased from about 200 million tons to approximately 300 million tons, indicating a significant growth of 47.78%. Hence, the surge in urban domestic waste resulting from the expansion of the digital economy imposes a pressing dilemma to be addressed.

Studying the impact of digital economic development on waste management is of great significance for achieving the goals of sustainability and environmental protection. Firstly, the application of digital technology can significantly enhance the efficiency and effectiveness of waste management systems (European Environment Agency [5]). Tools such as smart sensors, real-time data analytics, and optimization algorithms enable cities to collect, transport, and process waste more efficiently, thereby reducing the cost of waste disposal and improving resource utilization efficiency (Mukherjee et al. [6]). Secondly, digitalization can improve waste-sorting and recycling processes by tracking and monitoring the flow of different types of waste, optimizing recycling plans and resource recovery rates, reducing reliance on landfilling and incineration, and contributing to environmental protection by reducing resource waste and pollution (Kurniawan et al. [7]; Maiurova et al. [8]). Furthermore, digital waste management generates a wealth of data that supports data-driven decision-making, helping urban decision-makers better understand waste generation and processing patterns, and formulate more effective policies and strategies to meet the growing demand for waste management (Korherr et al. [9]). The development of the digital economy also fosters innovation in the waste management sector, providing opportunities for businesses and entrepreneurs to develop intelligent, sustainable, and environmentally friendly waste management solutions, creating job opportunities and economic growth (Cheah et al. [10]). Most importantly, the development of the digital economy and improvements in waste management contribute to urban sustainability by reducing waste generation, increasing waste processing efficiency, and reducing environmental impacts (Jiang et al. [11]). This ensures that cities can continue to develop and create better living conditions for future generations. Therefore, the objective of this study is to strive for a new balance in development, one that promotes economic progress while preserving the urban environment and facilitating sustainable development.

Waste management has always been an integral part of urban governance, and in the post-pandemic era, the choice of waste disposal methods has become a focal point of public attention. While there is existing literature pointing to new waste disposal methods, such as the conversion of biomass-derived polymers into functional biochar materials through pyrolysis (Yang et al. [12]), according to the Chinese Urban Statistical Yearbook [4], there are currently two primary waste disposal methods: landfill disposal and incineration.

• Landfill disposal, a traditional waste management method, involves burying waste underground and relying on natural decomposition for disposal (Abdel-Shafy et al. [13]). While this method has relatively low processing costs, it is not without its issues. Landfill disposal requires significant land usage, and the leachate produced from waste decomposition can potentially contaminate groundwater, thus affecting the ecological environment (Parvin et al. [14]).
• Incineration offers higher waste reduction and harmlessness effects. Through high-temperature incineration, harmful organisms such as viruses and bacteria in the waste are eliminated, thereby reducing the risk of pandemic spread (Zu et al. [15]). Additionally, incineration can significantly reduce waste volume and alleviate pressure on land use (Makarichi et al. [16]).

In the context of the pandemic, landfill disposal may increase the risk of virus transmission as landfills can become breeding grounds for the spread of diseases. However, incineration also presents certain challenges, such as the high costs associated with constructing waste incineration sites and the emission of greenhouse gases during combustion, which conflict with the goals of reducing carbon emissions. Furthermore, compared to a landfill for waste disposal, the construction of waste incineration sites necessitates a greater financial investment. Moreover, once operational, a waste incineration plant is not easily dismantled. Consequently, incineration is poised to emerge as the principal method for urban waste management in the foreseeable future. Therefore, this article aims to explore the transformation of waste disposal methods by city managers in the post-pandemic era through empirical research.

In recent years, the waste treatment industry in China has experienced a noteworthy growth trajectory, as illustrated in Figure 2. Between 2002 and 2021, there has been a consistent uptick in the volume of waste treated on a national scale, with a particular surge in total treatment volume (cf., China Urban Statistics Yearbook [4]). The graph reveals a compelling trend: since 2017, the quantity of waste directed to landfills has progressively diminished year over year. Notably, after the emergence of the COVID-19 pandemic, the amount of waste sent to incineration facilities has surpassed that of landfills. Figure 3 further elucidates the dynamics of the waste treatment sector. Between 2002 and 2015, both waste landfills and incineration plants exhibited a steady surge in numbers. However, from 2016 to 2021, while the count of incineration plants continued its upward trajectory, the number of waste landfills maintained relative stability and even commenced a decline in 2020. These trends underscore the evolving nature of waste management strategies in China and the shifting priorities in the sector’s developmental landscape.

In the context of rapid development of digital technology, the contribution of this article lies in revealing the connection between the development of the digital economy and urban waste management. It not only emphasizes the potential of the digital economy to improve the efficiency of urban waste disposal but also discovers mechanisms to reduce waste disposal by improving investment efficiency in digital economy infrastructure. This is an important insight for urban planners and policy makers, especially in an era where addressing urban environmental challenges through technological innovation is sought.

In the structure of this paper, we will organize and present our research in the following manner. In Section 2, we will review related studies to provide readers with an understanding of the previous research and our theoretical foundation regarding the digital economy and urban waste management. Section 3 will provide a detailed introduction to our model setup, including variable selection, assumptions, and modeling methods, to help readers comprehend the methodological basis of our research. Following that, Section 4 will showcase our empirical analysis results, including data analysis outcomes and our primary research findings. Finally, Section 5 will summarize the conclusions and perspectives of this paper.

2. Related Works

2.1. Digital Economy

Assessment and analysis of the digital economy are of paramount importance. Worldwide, an amount of research effort also goes into the digital economy. For example, Herrador and Hernandez [17] delve into the impact of digital information and communications technology on accounting education, while Fidan [18] employs the Gini method to analyze intersectoral digital economic development in Turkey and Lithuania. Coyle and Nguyen [19] bring attention to the complexities that digitalization introduces to conventional economic measurement. Additionally, Otioma et al. [20] provide insights into the development of the digital economy in Kigali through a macro-spatial lens, and Stavytskyy et al. [21] explore how the DESI (Digital Economy and Society Index) influences consumption index growth in Europe.

Particularly, recent studies have highlighted the critical role of the digital economy in driving sustainable economic growth in China. Liu et al. [22] find that digital finance, especially through technological innovation, has a significant and spatially spreading influence on sustainable growth. Jiao and Sun [23] examine how digital economic development positively impacts urban economic growth, noting the heterogeneous effects and the pivotal role of urban employment. Kong and Li [24] reveal that the development of the digital economy enhances green economic efficiency, especially in digitally advanced regions, highlighting the synergy between digital industrialization and environmental sustainability. Liu et al. [25] discover that digital financial inclusion plays a significant role in bolstering economic growth by fostering entrepreneurship and consumer spending. Sun and Tang [26] observe that digital inclusive finance promotes sustainable economic growth by improving financial accessibility and boosting consumer spending among residents. Ma and Lin [27] find that digital infrastructure construction can enable green economic performance in Chinese cities. Dong et al. [28] highlight the digital economy’s effectiveness in reducing energy vulnerability, particularly in higher-income regions.

2.2. Waste Treatment

The management of waste, with a particular emphasis on incineration, stands as a vital component in contemporary waste management strategies across the globe, including in China. The shift from landfill to incineration in waste disposal in China, primarily due to concerns over land usage, is comprehensively discussed by Li et al. [29]. Their research delves into the intricacies of the waste incineration industry, including its upstream and downstream development, and scrutinizes pertinent governmental policies. Kang et al. [30] contribute to this field by assessing the biomass fraction in Korean waste incineration facilities to better estimate greenhouse gas (GHG) emissions. They highlight the variance in biomass fractions across different waste types and stress the necessity of tailored GHG estimations for each incineration facility. Adding to this discourse, Khan et al. [31] provide a thorough review of the advancements in waste-to-energy incineration technologies, particularly in the context of climate change, exploring both the technological evolution and environmental implications.

Additionally, Lu et al. [32] engage in an inventory analysis and social life cycle assessment (SLCA) of GHG emissions emanating from waste-to-energy incineration, underlining the effects of new laws and regulations on GHG management. Yamamoto et al. [33] investigate the potential trade-off between incineration and recycling, using Japan as a case study. Their findings suggest that excess incineration capacity may indeed reduce recycling rates, hinting at a possible conflict between these two waste management strategies. Thomas et al. [34] estimate the socially optimal recycling rate in Japan, finding that the cost-minimizing approach may result in recycling rates lower than currently observed and mandated levels. This research implies that some developed countries, including Japan, may be setting inefficiently high recycling goals. Gradus et al. [35] compare the cost-effectiveness of recycling versus incineration of plastic waste in the Netherlands. Their analysis reveals that the implicit CO2 abatement price for plastic recycling is substantially higher than current carbon prices, suggesting that recycling, in this case, may not be the most economically viable option compared to incineration.

2.3. Urban Development and Urban Waste

Urban development and its intrinsic link with urban waste generation constitute a critical area of focus in modern environmental studies. Liu and Wu [36] embarked on a comprehensive statistical analysis in China, unraveling the myriad factors that influence municipal solid waste generation. Their study sheds light on crucial elements like economic growth, energy consumption, and the scale of urban areas. In a similar vein, Wildeboer and Savini [37] delve into the critical role of state policies in waste valorization within the circular economy paradigm, with a specific lens on construction and demolition waste in Hong Kong and Rotterdam, and its consequential effects on urban development.

Wang and Gong [38] turn their attention to the disparate development and economic strains of urban versus rural wastewater treatment in China, examining this issue through the lens of discharge limit legislation. They underscore the necessity of establishing appropriate discharge limits to equitably distribute the financial burden. Li et al. [39] explore the influence of environmental regulation on the green total factor productivity (GTFP) of Chinese cities, revealing how varying levels of urban economic development impact GTFP. Akbulut-Yuksel and Boulatoff [40] study the effectiveness of a green nudge, specifically the Clear Bag Policy in Canada, in promoting recycling and reducing municipal solid waste. Ihlanfeldt and Taylor [41] analyze the externality effects of small-scale hazardous waste sites on urban commercial property markets, finding significant negative impacts on property values. Kyriakopoulou and Picard [42] investigate the impact of local traffic pollution on the formation of residential and business districts in cities, shedding light on the trade-offs between production externalities, pollution, and commuting costs.

Furthermore, Zhu et al. [43] conducted a comprehensive study on urban-rural coordination in Sichuan Province, China, employing a Principal Component Analysis (PCA)—Grey Entropy measurement model to assess the synchronization of urban and rural development. Peng and Deng [44] innovatively use eco-civilization principles to formulate an indicator system for evaluating urban resource and environmental carrying capacity (URECC) in Guiyang, China. Xu et al. [45] combine nighttime light data with other metrics to assess URECC in Chinese cities, highlighting the impact of economic growth on URECC and its regional disparities.

3. Model Setting

3.1. Sample Description

This study focuses on the evolution of the digital economy and urban waste management, utilizing provincial-level data from across China spanning the years 2011 to 2021 for comprehensive analysis. The digital economy data is sourced from statistical yearbooks such as the Tertiary Industry Statistical Yearbook, the Information Industry Yearbook, the Peking University Digital Inclusive Finance Index, various provincial statistical yearbooks, and the National Bureau of Statistics. Complementarily, additional data is obtained from the National Statistical Yearbook. The resulting sample encompasses 330 observations, encompassing ten variables: total waste treatment, waste landfill quantity, waste incineration quantity, digital economy index, consumption, Gross Domestic Product (GDP), population density, proportion of industrial value-added, urbanization level, and intensity of environmental regulation. A detailed description of these variables can be found in

Table 1. Variable notation and description.

Variable Name	Variable Description
${Total treatment}_{it}$ (million tons)	Dependent variable representing the total amount of waste treated at unit i during time t.
${Burn}_{it}$ (million tons)	Dependent variable representing the amount of waste incinerated at unit i during time t.
${Landfill}_{it}$ (million tons)	Dependent variable representing the amount of waste landfilled at unit i during time t.
${Dig\_econ}_{it}$	Index measuring the level of digital economic activity at unit i during time t.
$T_t$	Time dummy variable indicating the onset of the COVID-19 pandemic at time t.
${GDP}_{it}$	Measure of economic activity, typically the Gross Domestic Product (GDP), at unit i during time t.
${Consumption}_{it}$	Total retail sales at unit i during time t.
${Popu\_density}_{it}$	Population density representing the number of individuals per unit area at unit i during time t.
${Indu}_{it}$	Level of industrialization, measured as the industrial value-added as a percentage of GDP at unit i during time t.
${City}_{it}$	Level of urbanization, representing the proportion of the population living in cities at unit i during time t.
${Regu}_{it}$	Level of environmental regulation, measured as the ratio of investment in industrial pollution control to industrial value-added at unit i during time t.

. This multi-faceted dataset facilitates a comprehensive exploration of the interplay between these variables and their implications.

Table 2. Statistical descriptions.

	Count	Mean	Std	Min	50%	Max
Total_treatment	330	6.440	5.197	0.576	4.918	33.457
Landfill	330	3.410	2.570	0.000	2.997	17.394
Burn	290	3.207	3.766	0.000	1.704	25.541
dig_econ	330	0.119	0.108	0.009	0.084	0.647
Consumption	330	10,541.159	8791.169	413.400	8120.750	44,187.700
GDP	330	26,676.287	21,734.437	1670.440	20,094.000	124,370.000
Popu_density	330	473.307	704.845	7.864	292.897	3925.870
Indu	330	0.321	0.082	0.273	0.378	0.556
City	330	0.596	0.121	0.511	0.656	0.896
Regu	330	0.003	0.004	0.001	0.004	0.031

offers a comprehensive yet nuanced snapshot of China’s economic and societal landscape, underscoring the need for contextualized and regionally tailored policies to address the observed disparities and unlock the country’s full potential. From Table 2 we can observe that:

• An in-depth examination of the ‘total_treatment’ metric reveals a high average value coupled with a notable standard deviation, indicating a certain degree of variability and fluctuation in the overall treatment level. This observed variability can likely be attributed to factors such as disparities in economic development levels and varying degrees of policy implementation effectiveness across China’s diverse regions.
• Moreover, the minimum values of the ‘landfill’ and ‘burn’ indicators being zero highlights the existence of regions or time points where the amount of waste sent to landfills or incinerated is minimal, thus shedding light on regional disparities in waste management practices in China. This apparent imbalance can be tentatively linked to differences in urbanization levels, waste generation rates, and treatment capacities among various regions. Conversely, the relatively high maximum values recorded for these two indicators suggest that certain areas have achieved commendable levels of waste processing, which could be correlated with the level of economic activity and population concentration in those regions.
• Furthermore, the uniformly low values of the ‘dig_econ’ indicator underscore the considerable untapped potential for the growth and development of China’s digital economy. Given the increasingly prominent role that the digital economy plays in the global economic landscape, this represents an opportune and strategic direction for China’s future economic expansion.
• Lastly, the substantial standard deviations observed in the ‘consumption’ and ‘GDP’ indicators unveil noteworthy disparities in consumption levels and economic development across China’s vast regions. These disparities, while significant, can be rationalized by the considerable geographical differences and variations in natural environments, economic conditions, and policy landscapes that exist among regions.

3.2. Study Design

The regression Model (1) employed in this study is a panel data model, which aims to examine the influence of digital economic activity ($dig\_econ$) on urban waste management volume while taking into account temporal trends and individual effects. The defined dependent variables in this model consist of the total urban waste management volume (Total_Treatment), incineration volume (Burn), and landfill volume (Landfill). On the other hand, the explanatory variable utilized is the digital economy index (dig_econ), serving as a proxy for digital economic activity. To account for potential nonlinear associations between digital economic activity and time trends, an interaction term (${dig\_econ}_{it}\times T_t$) is included in the model. Controls_it represents other control variables, including GDP ($GDP$), Total Retail Sales ($Consumption$), Population Density ($Popu\_density$), Level of Industrialization ($Indu$), Level of Urbanization ($City$), and Level of Environmental Regulation ($Regu$). Detailed definitions are shown in Table 1.

$${\mathit{Y}}_{\mathit{i}\mathit{t}}=\mathit{\alpha }+{\mathit{\theta }}_1{\mathit{d}\mathit{i}\mathit{g}\_\mathit{e}\mathit{c}\mathit{o}\mathit{n}}_{\mathit{i}\mathit{t}}+{\mathit{\theta }}_2{\mathit{T}}_{\mathit{t}}+{\mathit{\theta }}_3\left({\mathit{d}\mathit{i}\mathit{g}\_\mathit{e}\mathit{c}\mathit{o}\mathit{n}}_{\mathit{i}\mathit{t}}\times {\mathit{T}}_{\mathit{t}}\right)+{\mathit{\theta }}_4{\mathit{C}\mathit{o}\mathit{n}\mathit{t}\mathit{r}\mathit{o}\mathit{l}\mathit{s}}_{\mathit{i}\mathit{t}}+{\mathit{\epsilon }}_{\mathit{i}\mathit{t}}$$

(1)

The regression Model (2) employed in this study is a panel data model, which aims to examine the influence of the investment efficiency in digital economy infrastructure ($rate$) on urban waste management volume (Total_Treatment), incineration volume (Burn), and landfill volume (Landfill). The formal description is given as follow:

$${\mathit{Y}}_{\mathit{i}\mathit{t}}=\mathit{\alpha }+{\mathit{\theta }}_1{\mathit{r}\mathit{a}\mathit{t}\mathit{e}}_{\mathit{i}\mathit{t}}+{\mathit{\theta }}_2{\mathit{C}\mathit{o}\mathit{n}\mathit{t}\mathit{r}\mathit{o}\mathit{l}\mathit{s}}_{\mathit{i}\mathit{t}}+{\mathit{\epsilon }}_{\mathit{i}\mathit{t}}$$

(2)

Indeed, for the i-th region (i = 1, 2, …, n), the efficiency score is calculated by using a DEA model, through the following linear programming problem:

$$\begin{array}{c}Max {\theta }_i\\ s.t. {\displaystyle \sum _{j=1}^m}{\lambda }_j{x}_{ij}\le {\theta }_i{y}_{is}\\ {\displaystyle \sum _{i=1}^n}{y}_{is}\ge 1\\ {\lambda }_j\ge 0, j=\mathrm{1,2},\dots ,m; {\theta }_i\ge 0, i=\mathrm{1,2},\dots ,n;\end{array}$$

where, $x_{ij}$ is the value of the j-th input variable for the i-th region; $y_{is}$ is the value of the s-th output variable for the i-th region; ${\lambda }_j$ is the weight associated with the j-th input variable; ${\theta }_i$ is the efficiency score for the i-th region. The objective of this model is to maximize the efficiency score ${\theta }_i$. Solving this problem yields efficiency scores for each region, offering insights into the relative efficiency of the digital economy’s output concerning multiple input variables. For example, a score of 1 signifies that a region is fully efficient, while a score less than 1 indicates potential for efficiency improvement.

In our framework, the digital economy index serves as the output variable (i.e., $y_{is}$ and s = 1), while the input variables (i.e., $x_{ij}$) encompass the following factors: the number of mobile phone users per 100 people (households/100 people), the proportion of internet users in the permanent population (%), the density of optical cable lines (km/sq km), the density of mobile phone base stations (number/sq km), the density of internet broadband access ports (number/sq km), per capita fixed asset investment in the information transmission, computer services, and software industry (yuan/person), per capita total telecommunications business volume (yuan/person), and per capita total postal business volume (yuan/person). Estimating the efficiency of digital economy infrastructure investment using the DEA model has several advantages:

• Firstly, the digital economy is a complex system that involves multiple input and output factors. The DEA model is capable of handling situations with multiple inputs and outputs, thereby providing a comprehensive evaluation of the efficiency of digital economy infrastructure investment by considering these multiple factors. This capability makes the DEA model a suitable and comprehensive tool for assessing the digital economy.
• Secondly, as a non-parametric method, the DEA model does not require the pre-setting of production functions or the assumption of specific functional forms. This enhances the flexibility of the DEA model in practical applications and allows it to better adapt to the complexity and dynamics of the digital economy. Consequently, the DEA model accurately estimates investment efficiency.
• Thirdly, the DEA model can identify the efficiency frontier, also known as the best practice frontier, of decision-making units (DMUs) and the gaps between each DMU and the efficiency frontier. This capability enables decision-makers to gain a clear understanding of the efficiency level of digital economy infrastructure investment and identify the causes of inefficiency, as well as the direction for improvement. Moreover, by comparing the performance of different DMUs, the DEA model promotes cross-learning and benchmarking among industries, thereby fostering overall efficiency improvement.

In conclusion, the use of the DEA model to estimate the efficiency of digital economy infrastructure investment offers advantages such as handling multi-input and multi-output problems, not requiring pre-set production functions, and identifying efficiency frontiers and gaps. These advantages make the DEA model a powerful tool for assessing investment efficiency in the digital economy, providing decision-makers with crucial reference points.

4. Empirical Analysis

4.1. Baseline Results

Table 3. Main regression model results.

	Total_Treatment	Burn	Landfill
	(1)	(2)	(3)
dig_econ	0.410 ***	0.314 ***	0.312 ***
	(0.047)	(0.066)	(0.081)
dig_econ × T	−0.130 ***	0.159 ***	−0.560 ***
	(0.039)	(0.054)	(0.103)
Constant	0.065 **	−0.180 ***	0.187 **
	(0.031)	(0.050)	(0.081)
Controls	Yes	Yes	Yes
Province	Yes	Yes	Yes
Observations	330	290	330
R2	0.933	0.870	0.516
Adjusted R2	0.935	0.877	0.502

Note: ** and *** indicate the significance levels of 5% and 1%, respectively.

examines the impact of the digital economy (dig_econ) and the COVID-19 pandemic (T) on waste management practices, specifically total waste treatment (Total_Treatment), incineration (Burn), and landfilling (Landfill). After controlling for other economic factors such as GDP, consumption, and population density, significant effects of the digital economy on all three waste treatment methods are found. The digital economy positively impacts total waste treatment and incineration, suggesting that as the digital economy grows, the volume of waste treated and incinerated also increases. This could be attributed to the promotion of business activities and consumption by the digital economy, resulting in more waste generation.

However, the impact of the epidemic on waste management practices among city administrators is also an important finding. For the overall waste treatment (Total_Treatment), the coefficient of the interaction term is negative, indicating that the presence of the COVID-19 pandemic weakens the influence of the digital economy on the total waste treatment. This could be attributed to the reduction in economic activities caused by the pandemic, which subsequently affects waste generation and treatment. Concerning incineration treatment (Burn), the coefficient of the interaction term is positive, suggesting that the COVID-19 pandemic enhances the positive impact of the digital economy on incineration treatment. This may be due to the increased attention and utilization of incineration as a swift and effective waste treatment method during the pandemic. As for landfill treatment (Landfill), the coefficient of the interaction term is negative and relatively substantial, indicating a significant negative moderation effect of the COVID-19 pandemic on the correlation between the digital economy and landfill treatment. This can be attributed to the heightened restrictions on landfill treatment during the pandemic to mitigate the risk of virus transmission, while the growth of the digital economy propels the advancement of more environmentally friendly and sustainable waste treatment approaches.

4.2. Robustness Analysis

Firstly, we estimate the coefficient ${\theta }_t$ for the interaction term between the digital economy variable and the year dummy variable using the provided formula (see Figure 4). Subsequently, we generate a plot illustrating the changes in ${\theta }_t$ within the 95% confidence interval. The plot reveals that prior to the outbreak of the pandemic in 2019, the values of ${\theta }_t$ largely encompass zero, indicating a lack of statistical significance. Before the pandemic, the digital economy may have exerted similar influences across different provinces, possibly due to the common trends associated with digital economic development, such as the adoption of digital technologies, the growth of online commerce, and the increase in consumer spending. These trends had comparable impacts on various provinces, resulting in a degree of consistency in how the digital economy’s changes affected waste management practices. However, this consistency may have been challenged after the outbreak of the pandemic, as the pandemic likely led to significant changes in economic activities, consumer behavior, and waste generation patterns. Therefore, we can consider the pandemic as one of the key factors contributing to the alteration in the relationship between the digital economy and waste disposal practices. This implies that before the pandemic, the variations in the digital economy index across different provinces adhered to the assumption of parallel trends, thereby validating the application of the continuous double difference model in this context.

4.3. Heterogeneity Analysis

Table 4. Heterogeneity analysis results.

	Total_Treatment		Burn		Landfill
	High	Low	High	Low	High	Low
dig_econ × T	0.269 ***	−0.361 ***	0.173 *	0.212 *	0.272 *	−1.02 ***
	(0.070)	(0.091)	(0.095)	(0.121)	(0.133)	(0.245)
Constant	−0.254 **	0.025	0.195	−0.106 **	0.280	0.025
	(0.157)	(0.031)	(0.213)	(0.049)	(0.296)	(0.083)
Controls	Yes	Yes	Yes	Yes	Yes	Yes
Province	Yes	Yes	Yes	Yes	Yes	Yes
R2	0.950	0.933	0.935	0.850	0.421	0.565
Adjusted R2	0.935	0.931	0.915	0.845	0.245	0.551

Note: *, **, and *** indicate the significance levels of 10%, 5%, and 1%, respectively, with standard errors in parentheses.

showcases the heterogeneity analysis, which delves into the varying impacts of digital economic activity (dig_econ) on urban waste management volume across different regions or subgroups. The implications of these findings are of great significance to policy makers and urban planners. The heterogeneity analysis unveils substantial disparities in the effect of digital economic activity on waste management volume across diverse regions. These differences may stem from variations in infrastructure, population density, and the extent of digitalization in each region. Certain regions exhibit a more pronounced relationship between digital economic activity and waste management volume, indicating their successful utilization of digital technologies to streamline waste management processes. In contrast, other regions may demonstrate a weaker association, potentially due to delayed adoption of digital solutions or other contextual factors. The outcomes of this heterogeneity analysis underscore the importance of an intricate and contextualized approach to waste management policies. Approaches that prove effective in one region may not yield the same outcomes in another. Hence, policy makers and urban planners should thoughtfully consider the unique characteristics and requirements of each region when formulating and implementing waste management strategies.

4.4. Endogeneity Analysis

To address concerns regarding reciprocal causality and selective bias, we use tele (i.e., the fixed phone numbers per 100 people multiplied by national information technology service revenue) as the instrumental variable. The historical quantity of fixed telephones, as an initial regional endowment effect, not only meets the criteria for exclusivity but also correlates with the development of the digital economy. This reflects the significant association with technological infrastructure and revenue generation, fundamental aspects of digital economic activities. Additionally, the year 1984 serves as a baseline for our study, providing a historical reference point to trace the evolution of the digital economy and its infrastructural capabilities. From

Table 5. IV Regression results using tele.

	Total_Treatment
	dig_econ × T	Total_Treatment
	First Stage	Second Stage
dig_econ × T		−0.7016 ***
		(0.230)
Tele	0.0885 **
	(0.031)
Constant	0.0161	−0.1610 ***
	(0.038)	(0.024)
Controls	Yes	Yes
Province	Yes	Yes
R2	0.298	0.917
Adjusted R2	0.283	0.915

Note: ** and *** indicate the significance levels of 5% and 1%, respectively, with standard errors in parentheses.

, we can observe that the results of the first-stage regression exhibit a statistically significant negative correlation between tele and dig_econ × T at a 5% significance level, indicating the absence of a weak instrumental variable problem and satisfying the relevance assumption. In the second-stage regression, the coefficient of dig_econ × T is notably negative at a 1% significance level, providing evidence that our hypothesis remains valid even after utilizing instrumental variables to mitigate endogeneity concerns. This offers further support to the notion that, in the context of a pandemic, the expansion of the digital economy leads to a reduction in waste processing volume.

4.5. Effect of Investment Efficiency in the Digital Economy

Table 6. Analysis of investment efficiency in the digital economy and waste treatment.

	Total_Treatment	Burn	Landfill
	(1)	(2)	(3)
Rate	0.0447 **	−0.0945 **	0.1859 ***
	(0.021)	(0.032)	(0.099)
Constant	−1.0536 ***	−0.3583 ***	−0.0363
	(0.024)	(0.040)	(0.062)
Controls	Yes	Yes	Yes
Year	Yes	Yes	Yes
Province	Yes	Yes	Yes
Observations	330	290	330
R2	0.916	0.840	0.374
Adjusted R2	0.914	0.836	0.360

Note: ** and *** indicate the significance levels of 5% and 1%, respectively, with standard errors in parentheses.

presents a comprehensive analysis of investment efficiency in the digital economy and its potential impact on urban waste treatment. The primary focus of this analysis is to examine the return on investment in digital infrastructure and its implications for the effectiveness and efficiency of waste management practices in urban areas. The efficiency of digital infrastructure investment plays a critical role in determining the overall success of waste management strategies. Higher investment efficiency is typically associated with better integration of digital technologies into waste management systems, leading to optimized resource allocation, improved waste sorting and recycling, and enhanced monitoring and control capabilities. The obtained results indicate that cities with higher investment efficiency in digital infrastructure tend to exhibit more waste treatment practices for landfill waste treatment. Moreover, the analysis highlights that the relationship between digital infrastructure investment efficiency and incineration waste treatment (Burn) is negative. Additionally, for total waste treatment, as investment efficiency reaches very high levels, there may be diminishing returns, suggesting the need for a holistic approach to waste management that combines digital investments with other strategic interventions for sustained improvements. Therefore, policy makers and urban planners should prioritize digital investments that yield high returns in terms of waste reduction, recycling, and resource efficiency.

In

Table 7. IV Regression Results using patent registrations.

	Total_Treatment
	Rate	Total_Treatment
	First Stage	Second Stage
Rate		0.8154 ***
		(0.238)
patent	0. 1818 *
	(0.103)
Constant	0.2689 ***	−0.3115 ***
	(0.099)	(0.049)
Controls	Yes	Yes
Year	Yes	Yes
Province	Yes	Yes
R2	0.254	0.916
Adjusted R2	0.234	0.913

Note: * and *** indicate the significance levels of 10% and 1%, respectively, with standard errors in parentheses.

, we present the endogeneity analysis for investment efficiency in relation to waste treatment, aiming to explore the potential endogenous relationship between digital infrastructure investment efficiency (rate) and patent registrations (patent). In order to address concerns related to reciprocal causality and selective bias, we have employed methodologies to account for endogeneity and utilized the two-stage least squares approach to examine the relationship. The results of the first-stage regression reveal a statistically significant negative correlation between patent and rate at a 10% significance level. This suggests the absence of a weak instrumental variable problem and validates the relevance assumption. In the second-stage regression, the coefficient of rate is notably positive at a 1% significance level. This provides evidence that our hypothesis remains robust even after employing instrumental variables to address endogeneity concerns.

5. Conclusions and Perspective

This study focuses on the relationship between the digital economy and urban waste management through empirical analysis of the relationship between the development of the digital economy, investment efficiency of digital economy infrastructure, and urban waste disposal methods (incineration and landfilling). Firstly, the development of the digital economy exerts a substantial impact on the volume of urban waste disposal, a relationship that remains robust even after accounting for time trends and individual effects. Secondly, addressing endogeneity concerns through instrumental variables reveals a positive impact of the investment efficiency of digital economic infrastructure on landfill waste treatment, but a negative impact on incineration waste treatment. This implies that enhancing investment efficiency in digital infrastructure can contribute to reducing the volume of incineration waste. These findings suggest the potential for optimizing digital economic development and investment efficiency in digital infrastructure to alleviate the pressure on landfilling and incineration, thereby mitigating environmental pollution.

Despite the significant insights garnered from this study, several avenues for future research warrant attention. This study utilizes data from specific regions in the United States, which may limit the generalizability of the research findings. Variations in the level of digital economic development and waste disposal methods across different regions may affect the relevance of our conclusions to other countries or regions. Subsequent research could expand to different countries and regions to obtain a more comprehensive understanding of the impact of the digital economy on urban waste management. Furthermore, this study primarily focuses on the analysis of the digital economy and the efficiency of digital infrastructure investment. It does not consider other factors that may influence urban waste management, such as population growth and changes in consumption patterns. Future research could explore the role of these factors to achieve a more comprehensive understanding of the complexities of urban waste management. Further investigations could delve into the relationship between the digital economy and various waste disposal methods, emphasizing potential pathways and underlying mechanisms. Additionally, exploring the factors influencing investment efficiency in digital infrastructure and their connection to waste disposal and environmental pollution would be valuable. Examining variations in digital economic development, investment efficiency in digital infrastructure, and waste disposal methods among different cities or regions is also worthwhile. Lastly, evaluating the actual effects of digital economy policies on waste disposal and environmental pollution holds practical significance, particularly given the evolving landscape of the digital economy.