Impact of China’s Digital Economy on Integrated Urban–Rural Development

Abstract

This paper measures the comprehensive level of the digital economy and integrated urban–rural development across Chinese provinces from 2013 to 2022. Using a two-way fixed effects model, it explores the nonlinear relationship, regional heterogeneity, and underlying causes of how the digital economy empowers integrated urban–rural development. The study finds that: (1) The digital economy promotes integrated urban–rural development, with a quadratic polynomial regression model and threshold regressive model revealing an inverted U-shaped relationship. The turning point has not yet been reached, and the promoting relationship shows a diminishing marginal trend. (2) In terms of sub-dimensions, the strengths of the promoting effect are ranked as follows: digital infrastructure construction, industrial digitization, and digital industrialization. (3) Regionally, the digital economy’s promoting effect on integrated urban–rural development is strongest in Central China, followed by Eastern China, and weakest in Western China. The innovation and contribution of this paper lie in discovering the nonlinear impact of the digital economy on China’s integrated urban–rural development, analyzing its intrinsic driving forces and regional differences, and providing valuable references for implementing regionally differentiated development policies for the digital economy and integrated urban–rural development in various regions of China.

1. Introduction

The concept of integrated urban–rural development can be traced back to the book “An Inquiry into the Nature and Causes of the Wealth of Nations”, where it was systematically explained that cities and rural areas should be mutually beneficial [1]. Compared to urban–rural coordination, which has been the focus of Chinese urban and rural development policies in the past, integrated urban–rural development emphasizes interactive merging and collaborative sharing. Based on existing literature, integrated urban–rural development can be defined as a status where labor, technology, capital, information, and other factors can flow freely between urban and rural areas [2,3,4], achieving balanced, coordinated, and sustainable development across multiple dimensions. These dimensions include population [5], industry [6], institutions [7], society [8], economy [9], culture [10], and ecology [11], while recognizing the inherent differences between urban and rural areas.

From an international perspective, urban–rural development imbalance is not unique to China; it is also a key issue affecting the comprehensive progress of both developing and developed countries [12,13]. For example, in developing countries such as Nigeria and Kenya in Africa, urban economies are rapidly developing, but rural areas suffer from severe deficiencies in infrastructure and public services [14]. Conversely, in developed countries such as the United Kingdom, the United States, and Japan, rural areas face development bottlenecks, with aging populations and the outflow of young labor leading to over-reliance on agriculture and low-income industries, in stark contrast to urban prosperity [15,16]. To address this global issue, countries have formulated strategies based on their national conditions to promote the coordination and integration of urban and rural development. For instance, from the strategic height of sustainable tourism development, Japan has analyzed the characteristics and strategies of urban and rural destination management organizations (DMOs) to strengthen the coordination and cooperation of local tourism resources [17]. The UK has devised large-scale investment plans in infrastructure, skills, and innovation to strive for balanced urban–rural development [18]. Although these international experiences provide valuable references for narrowing the urban–rural gap, due to China’s unique national conditions and development stage, completely eliminating urban–rural development disparities remain a long-term and complex task.

The trend towards “convergence” in urban–rural integration in China has been emerging in recent years. Rural residents’ disposable income has consistently grown faster than that of urban residents, and rural healthcare infrastructure has significantly improved. However, the restricted flow of resources such as labor, capital, and technology has weakened the intrinsic motivation for rural development due to the remnants of the urban–rural dual system. Significant disparities in urban–rural incomes persist, and issues such as unequal access to basic public services and ecological imbalances remain. In 2022, the “Digital Countryside Development Action Plan (2022–2025)” explicitly emphasized developing the rural digital economy with a focus on coordinated and integrated urban–rural development. The digital economy, characterized by virtualization, permeability, and rapidity [19], can integrate various production factors and address mismatches in resources and spatio-temporal constraints deeply and effectively. This promotes the free flow of labor and other production factors, becoming a core driving force for the reorganization and restructuring of urban–rural elements. Therefore, exploring the mechanisms through which the digital economy impacts integrated urban–rural development has become a crucial topic during this transformative stage.

Substantial achievements have been made in digital economy studies to integrate urban–rural development in the global wave of digitalization, including measuring the indicators of the digital economy and urban–rural integration [20,21,22,23], exploring the theoretical logic and mechanisms through which the digital economy aids integrated urban–rural development [24,25], and empirically testing their impact [26,27]. For instance, Deng et al. (2023) [28], from an income perspective, reveal the positive role of the digital economy on increasing absolute income in urban and rural areas, with a more significant positive effect on urban residents’ income, thereby exacerbating the widening urban–rural income gap. Florido-Benítez (2024) [29], from the unique angle of tourism promotion, emphasizes the importance of Destination Marketing Organizations (DMOs) in utilizing digital technologies to monitor tourism promotion budgets and activities, measuring the efficiency and effectiveness of economic budgets, and opening new channels for urban–rural exchanges. Furthermore, some scholars have also found that the uneven development of digital technology can lead to a digital divide or enhance the siphoning effect of cities on rural areas, thereby widening the urban–rural gap and hindering integration [30,31].

It is worth noting that current research predominantly focuses on linear impact analysis of the digital economy on integrated urban–rural development, which may not comprehensively reveal the intricate relationship between the two. In the limited literature exploring nonlinear effects, scholars recognize the nonlinear trends in the digital economy’s impact on urban–rural integration, yet the delineation of these impact stages remains unclear, and conclusions across different studies show inconsistencies. For instance, Li et al. (2024) found a diminishing marginal effect of the digital economy on integrated urban–rural development across 31 provinces in China from 2011 to 2020 [32], while Yang et al. (2024) identified a significant “increasing marginal effect” in this enhancement [9]. Additionally, the current literature lacks in-depth studies on the decomposition of digital economy indicators. While some studies attempt to decompose digital economy indicators and preliminarily examine their impact on the urban–rural income gap [33,34], research on how the digital economy comprehensively and from multiple perspectives promotes or constrains integrated urban–rural development remains notably scarce.

However, there is still debate on the regional heterogeneity in the impact of the digital economy on integrated urban–rural development. One view, represented by scholars like Huang (2022) [26] and Sun (2023) [35], posits that the digital economy has a stronger promoting effect on urban–rural integration in Eastern China. This is primarily due to the region’s relative advantages in economic development and human capital, which enhance its ability to leverage the digital economy. Conversely, other scholars such as Li (2024) [32], Yu (2023) [36], and Wang (2022) [37] argue that the digital economy’s promoting effect on urban–rural integration is more evident in Central and Western China. Additionally, some studies have found that the digital economy does not significantly impact integrated urban–rural development in Central and Western regions, or may even have a suppressive effect [38].

A comprehensive review of existing studies reveals that most focus on the impact of composite indicators of the digital economy on integrated urban–rural development, without detailing the differentiated impacts caused by different dimensions of digital economic development. Additionally, many studies use a time span of over ten years for their research samples but rely solely on linear models to measure the impact of the digital economy on integrated urban–rural development, overlooking the potential nonlinear effects over long-term development. Finally, the explanations for the regional heterogeneity in the digital economy’s empowerment of integrated urban–rural development are unclear, lacking effective exploration of the underlying reasons. In conclusion, we further clarified the research question in this paper to (1) explore the impact of China’s digital economy on integrated urban–rural development and investigate the nonlinear effects; (2) delve into the contributions of different dimensions of the digital economy in empowering integrated urban–rural development; and (3) investigate the reasons for regional heterogeneity.

Therefore, the potential contributions of this study are as follows: First, this study creatively introduces the squared term of the digital economy and the panel threshold effect model to deeply explore the nonlinear impact of the digital economy on integrated urban–rural development. It reveals the varying intensities of the digital economy’s impact on integrated urban–rural development at different development stages. Second, from the perspective of decomposing the digital economy, the study subdivides the digital economy into three dimensions: digital infrastructure construction, digital industrialization, and industrial digitalization. It explores the contributions of each component of the digital economy in promoting integrated urban–rural development. Third, by combining the decomposition of digital economic indicators with regional analysis, the study compares the differences in the impact of the digital economy on integrated urban–rural development across different regions, deeply analyzing the reasons for regional heterogeneity.

Through this study, we can deeply analyze the differentiated effects of the digital economy on integrated urban–rural development at different stages, providing theoretical support for timely adjustments to digital economy development policies. At the same time, we can deeply explore the impact of various dimensions of the digital economy on urban–rural integration, providing a scientific basis for formulating more targeted policy measures. Furthermore, by combining the analysis of the various dimensions of the digital economy with regional heterogeneity, we can explore the reasons for regional heterogeneity in the digital economy’s empowerment of integrated urban–rural development. These will enable us to propose policy recommendations for promoting the digital economy and integrated urban–rural development in different regions of China in the new era, thereby deepening the practical understanding of improving China’s urban–rural relations.

2. Theoretical Analysis and Research Hypotheses

2.1. Mechanism of the Impact of Digital Economy on Urban–Rural Integration

The empowerment of the digital economy for integrated urban–rural development can be outlined from three perspectives: micro (integrated basics), meso (integrated energy), and macro (integrated results). For clarity, a causal mechanism diagram is constructed in this paper, as shown in Figure 1.

From a micro perspective, the construction and transformation of digital infrastructure form the foundation for enhancing connectivity between urban and rural areas. Core government agencies in China, such as the Ministry of Industry and Information Technology (MIIT) and the Ministry of Science and Technology, have actively led and accelerated the construction of local Internet infrastructure. As of December 2023, the Internet penetration rate in rural China reached 66.5%. The enhancement of digital infrastructure can effectively break various economic, social, and institutional constraints and spatial barriers to urban–rural integration [39,40], accelerating the flow of information elements, optimizing the allocation of factors such as labor, land, and capital, and promoting smooth dual circulation between urban and rural areas [41]. From a meso perspective, the vigorous development of digital industries and the transformation and restructuring of industrial digitization are the core of the digital economy. Based on advantages such as technological universality and penetration, they have deeply penetrated and widely integrated with various sectors of the national economy. This has led to a reduction in urban–rural income and consumption gaps at the macro level, improvement in social services, and enhancement of ecological environments [42,43,44], which are prominently manifested in economic, social, and ecological aspects.

On the economic dimension, local governments actively lead the deep integration of digital technology with agriculture and the rural economy to enhance rural agricultural production efficiency [45], thereby promoting rural economic growth. The Department of Rural Tourism can accelerate the transformation and upgrading of rural industries such as smart agriculture and rural tourism through advanced digital marketing and management strategies, promoting deep integration between agriculture and tourism so as to innovate rural industrial development models, driving significant economic growth and employment opportunities in rural areas, thereby increasing rural residents’ incomes. For example, China’s rural online retail sales reached 249 billion yuan in 2023. At the same time, the network effects of the digital economy are shifting rural residents from subsistence to developmental and discretionary consumption, gradually aligning urban and rural consumption habits [46,47].

On the social dimension, the digital economy reduces information asymmetry barriers. Government departments could utilize digital management to accurately capture the personalized needs of urban and rural residents, leveraging platform effects and long-tail advantages of digital technology to provide precise employment services, social security, and other public service products. In cultural dissemination, local government cultural and tourism departments promote the influx of high-quality educational resources and cultural heritage into rural areas through digital channel marketing and online exhibitions, virtual tours, etc. [48], thereby promoting urban–rural integration in terms of social public services.

On the ecological dimension, digital technology, with its high technical attributes, transforms traditional industries and production modes of existing industries, reducing material and energy consumption and promoting the digitalization, low-carbonization, and greening transformation of urban and rural industrial and agricultural industries [49], thereby reducing pollutant emissions. At the same time, local governments leverage the advantages of digital technology to regularly assess the effects of implementation of low-carbon transformation and green industrial projects, promptly adjusting and improving them to promote the health and sustainable development of urban and rural ecological environments.

Hypothesis 1.

The digital economy significantly promotes integrated urban–rural development, particularly in the dimensions of urban–rural economy, society, and ecology.

2.2. Nonlinear Effect Analysis

There are significant differences between urban and rural areas in terms of resource endowment, historical foundation, and policy environment. The rapid development and increasing penetration of the digital economy may not promote urban and rural development equally. Therefore, the impact of the digital economy on urban–rural integration is not a simple linear relationship.

In the early stages of digital economic development, urban and rural areas gradually build and improve their digital infrastructure construction. Compared to urban areas, rural areas can achieve greater marginal benefits due to their abundant land resources and lower costs, leading to accelerated economic and social development [50]. During this phase, the rapid advancement of the digital economy significantly promotes urban–rural integration. In the mid to later stages of digital economic development, as the benefits of digital infrastructure construction become saturated [51], rural areas face challenges due to weaker information technology applications and digital skills. The initial advantages from resource endowment gradually diminish, making digital transformation and upgrading of traditional industries more difficult [52]. Bottlenecks begin to emerge. In contrast, urban areas, with their higher levels of information technology application, concentrated talent resources, advanced digital technologies, and policy support, quickly integrate into the digital industry era, achieving comprehensive economic and social upgrades. During this process, a “digital divide” forms between urban and rural areas in terms of the speed and quality of digital economic development. The promoting effect of the digital economy on integrated urban–rural development gradually diminishes, potentially slowing down, stagnating, or even hindering the process.

Hypothesis 2.

The digital economy has a nonlinear impact on urban–rural integration development, potentially exhibiting an inverted “U” shape.

3. Research Design

3.1. Methods

3.1.1. Two-Way Fixed Effects Model

We established a two-way fixed effect model of the digital economy and urban–rural integration development based on the above theoretical analysis, as shown in Formula (1):

$$Inte{g}_{i,t}={\alpha }_0+cDi{g}_{i,t}+\delta Z_{i,t}+{\lambda }_i+{\eta }_t+e_{i,t}$$

(1)

Among them, i represents provinces, t stands for the vector of control variables for the year, $Inte{g}_{i,t}$ represents the integrated urban–rural development of region i in year t, $Di{g}_{i,t}$ represents the digital economy of region i in period t, ${\lambda }_i$ and ${\eta }_t$ are individual and time effects, respectively, $Z_{i,t}$ is a series of control variables, and $e_{i,t}$ is a random error term. Additionally, $Di{g}_{1i,t}$, $Di{g}_{2i,t}$, and $Di{g}_{3i,t}$ are selected to represent the decomposition indicators of the digital economy: digital infrastructure construction, digital industrialization, and industrial digitization.

3.1.2. Nonlinear Model

To investigate whether there exists a nonlinear relationship between the digital economy and integrated urban–rural development, we introduce the square term and threshold effect model.

Firstly, by constructing the square term of the digital economy and substituting it into the Two-Way Fixed Effects Model in Equation (1), we generate Formula (2) as follows:

$$Inte{g}_{i,t}={\alpha }_0+c_{1}Di{g}_{i,t}+c_2{Dig}_{i,t}^2+\delta Z_{i,t}+{\lambda }_i+{\eta }_t+e_{i,t}$$

(2)

${Dig}_{i,t}^2$ represents the square term of the digital economy; the other variables are the same as in Formula (1).

Secondly, the panel regression threshold effect model is employed to test whether the impact of the digital economy on integrated urban–rural development is segmented [53]. Thus, a panel threshold model, assuming the existence of a threshold value, is constructed as follows:

$$Inte{g}_{i,t}={\alpha }_0+{\beta }_1^{\prime }Di{g}_{i,t}I(Di{g}_{i,t}\le \gamma )+{\beta }_2^{\prime }Di{g}_{i,t}I(Di{g}_{i,t}>\gamma )+\delta Z_{i,t}+{\lambda }_i+{\eta }_t+e_{i,t}$$

(3)

In Formula (3), the threshold variable is also the digital economy $Di{g}_{i,t}$ and $\gamma$ is the threshold value to be estimated. If there exists a single threshold value, $\gamma$ is set to ${\gamma }_1$. It is divided into two cases based on whether the digital economy is greater than (>) ${\gamma }_1$ or less than (<) ${\gamma }_1$, controlling for other variables. Under different scenarios, the impact of the digital economy on integrated urban–rural development shows significant differences. If there are multiple threshold values for the digital economy, then an extension beyond the single threshold model is necessary. This involves incorporating multiple critical values, ${\gamma }_2$, …, ${\gamma }_n$, resulting in a multi-threshold panel model. Other variables remain the same as in Formula (1).

3.2. Variables

Integrated Urban–Rural Development: Most studies start from the definition of integrated urban–rural development, measuring dimensions such as social, economic, demographic, livelihood, industrial, cultural, institutional, and ecological aspects. Common measurement methods include the entropy weight method [26], factor analysis [54], principal component analysis [55], and urban–rural system composite synergy model [56,57]. Building upon previous research, this study centers around the core definition of integrated urban–rural development, adhering to a “people-oriented” philosophy, focusing on economic, social, and ecological dimensions. It covers various aspects such as urban and rural residents’ income, consumption, education, employment, social security, and living environment. After eliminating dimensional differences between indicators, the entropy weight method is used to construct the urban–rural integration development indicator system, as detailed in Table 1.

Table 1. Indicator system for integrated urban–rural development.

Objective	Dimensions	Indicators	Calculation or Description	Indicator Type	Indicator Properties	Indicator Weights
Integrated Urban–Rural Development	Economic Integration	Income ratio	Per capita disposable income between urban and rural areas	Contrast	Negative	0.076
		Consumption ratio	Rural–urban Engel coefficient difference	Contrast	Negative	0.068
	Social Integration	Education	Teacher:student ratio in urban and rural compulsory education	Contrast	Negative	0.047
		Medical care	Ratio of medical and health institution beds per thousand people in urban and rural areas	Contrast	Negative	0.042
		Employment	Ratio of unemployment insurance population to population aged 15 and above	Comprehensive	Positive	0.448
		Transport	Highway mileage	Comprehensive	Positive	0.218
	Ecological Integration	Energy conservation and emission reduction	Urban/rural domestic waste treatment rate	Contrast	Negative	0.004
		Environmental greening	Green coverage rate of urban built-up areas/green coverage rate of rural areas	Contrast	Negative	0.056
		Industrial pollution	Sulfur dioxide emissions	Comprehensive	Negative	0.042

Data Source: “China Statistical Yearbook (2014–2023)” and “China Urban and Rural Construction Statistical Yearbook (2013–2022)”, as well as Table 2 below.

Table 1. Indicator system for integrated urban–rural development.

Objective	Dimensions	Indicators	Calculation or Description	Indicator Type	Indicator Properties	Indicator Weights
Integrated Urban–Rural Development	Economic Integration	Income ratio	Per capita disposable income between urban and rural areas	Contrast	Negative	0.076
		Consumption ratio	Rural–urban Engel coefficient difference	Contrast	Negative	0.068
	Social Integration	Education	Teacher:student ratio in urban and rural compulsory education	Contrast	Negative	0.047
		Medical care	Ratio of medical and health institution beds per thousand people in urban and rural areas	Contrast	Negative	0.042
		Employment	Ratio of unemployment insurance population to population aged 15 and above	Comprehensive	Positive	0.448
		Transport	Highway mileage	Comprehensive	Positive	0.218
	Ecological Integration	Energy conservation and emission reduction	Urban/rural domestic waste treatment rate	Contrast	Negative	0.004
		Environmental greening	Green coverage rate of urban built-up areas/green coverage rate of rural areas	Contrast	Negative	0.056
		Industrial pollution	Sulfur dioxide emissions	Comprehensive	Negative	0.042

Data Source: “China Statistical Yearbook (2014–2023)” and “China Urban and Rural Construction Statistical Yearbook (2013–2022)”, as well as Table 2 below.

Table 2. Measurement of digital economy indicators.

First-Level Indicators	Secondary Indicators	Unit	Indicator Weights
Digital infrastructure construction	Length of optical cable line	Kilometer	0.075
	Mobile phone penetration rate	Department/100 people	0.034
	Internet access ports	Ten thousand	0.071
Digital industrialization	Enterprises in the information transmission, software, and information technology industries	Piece	0.142
	Employees in information transmission, software, and information technology industries	Ten thousand people	0.164
	Software business revenue	Ten thousand Yuan	0.239
Industrial digitization	Computers used per 100 employees in the enterprise	Desk	0.054
	Websites owned by every 100 enterprises	Piece	0.014
	Proportion of enterprises with e-commerce transaction activities	%	0.034
	Electronic sales	100 million yuan	0.174

Table 2. Measurement of digital economy indicators.

First-Level Indicators	Secondary Indicators	Unit	Indicator Weights
Digital infrastructure construction	Length of optical cable line	Kilometer	0.075
	Mobile phone penetration rate	Department/100 people	0.034
	Internet access ports	Ten thousand	0.071
Digital industrialization	Enterprises in the information transmission, software, and information technology industries	Piece	0.142
	Employees in information transmission, software, and information technology industries	Ten thousand people	0.164
	Software business revenue	Ten thousand Yuan	0.239
Industrial digitization	Computers used per 100 employees in the enterprise	Desk	0.054
	Websites owned by every 100 enterprises	Piece	0.014
	Proportion of enterprises with e-commerce transaction activities	%	0.034
	Electronic sales	100 million yuan	0.174

Digital Economy: The development of the digital economy, characterized by dynamic growth features, is closely linked to the increasingly transformative information technology. To date, evaluation indicators have not been unified. In measuring digital economy indicators, different countries and organizations have constructed diverse evaluation frameworks based on their respective development characteristics and needs. For example, the U.S. Department of Commerce’s Bureau of Economic Analysis (BEA) measures the Digital Economy Index from three perspectives: digital infrastructure, digital transactions (e-commerce), and the creation and acquisition of digital content by digital economy users (digital media) [58]. In academic exploration, the focus is generally on core dimensions related to digital infrastructure construction, digital application proliferation, digital industries, and digital transactions [20,59]. The selection of measurement methods is generally consistent with indicators of urban–rural integration development. Based on this, we chose three dimensions—digital infrastructure construction, digital industrialization, and industrial digitalization—and ten detailed indicators to construct the digital economy indicator system, using the entropy weight method for measurement. See Table 2 for details.

Control Variables: Drawing on existing research findings [43,60], we selected the following control variables: (1) Economic development, measured by the gross regional product. (2) Industrial structure, measured by the ratio of the tertiary industry output to the secondary industry output. Upgrading industrial structures can affect the mutual integration of urban and rural industries, thereby affecting urban–rural integration. (3) Inflation level. Due to differences in consumption concepts and habits between urban and rural residents, their reactions to changes in consumer prices also differ, thereby affecting integrated urban–rural development. Therefore, we included the consumer price index for urban and rural residents as a control variable. (4) Government support, measured by expenditure on urban and rural community construction. Local government expenditures in urban and rural communities may affect the urban–rural integration process. Due to significant differences in government investment intensities across regions, we used logarithmic transformation. (5) Infrastructure construction. The passenger volume of highways in each province serves as an effective metric for assessing infrastructure progress. Enhanced transportation infrastructure construction fosters favorable conditions for the bidirectional flow of urban and rural factors, significantly contributing to the seamless circulation of resources between towns and villages. (6) Degree of opening up, measured by the total volume of import and export trade. This controls for the impact of foreign trade on integrated urban–rural development.

3.3. Data Source

The study covered 30 provinces in China from 2013 to 2022, excluding Tibet, Hong Kong, Macau, and Taiwan. These provinces include Beijing, Tianjin, Hebei, Liaoning, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong, Hainan, Shanxi, Jilin, Heilongjiang, Anhui, Jiangxi, Henan, Hubei, Hunan, Inner Mongolia, Guangxi, Chongqing, Sichuan, Guizhou, Yunnan, Shaanxi, Gansu, Qinghai, Ningxia, and Xinjiang. According to the classification standards of the National Bureau of Statistics, China is divided into three regions: Eastern, Central, and Western. Eastern China includes Beijing, Tianjin, Hebei, Liaoning, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong, and Hainan; Central China includes Shanxi, Jilin, Heilongjiang, Anhui, Jiangxi, Henan, Hubei, and Hunan; Western China includes Inner Mongolia, Guangxi, Chongqing, Sichuan, Guizhou, Yunnan, Shaanxi, Gansu, Qinghai, Ningxia, and Xinjiang. The data used in this study are sourced from the “China Statistical Yearbook”, “China Urban and Rural Construction Statistical Yearbook”, and various provincial statistical yearbooks. For missing data, methods such as linear interpolation and equivalent substitution were employed (For instance, the Engel coefficient in 2013 was derived from the proportion of food expenditure to total cash expenditure; the indicators for urban built-up area greening coverage and rural greening coverage were missing in 2018, so the average value of this indicator in 2017 and 2019 was used for interpolation; the number of students and full-time teachers in urban and rural compulsory education stages in 2013 was missing, thus linear interpolation was used to fill in the missing data; in the instrumental variables, mobile users from 2012 to 2013 were missing in the annual data, thus linear interpolation was also used to fill in the gaps). This was done to ensure continuity of the research indicators. Descriptive statistics for each indicator are shown in

Table 3. Descriptive statistics.

Indicators	Calculations	Numbers	Mean	Standard Deviation	Minimum	Maximum	Skewness	Kurtosis
Integrated urban–Rural development	Index calculated using the entropy weight method	300	0.368	0.093	0.172	0.707	0.934	4.086
Digital economy	Index calculated using the entropy weight method	300	0.163	0.140	0.027	0.839	2.108	7.937
Economic development	Gross regional product (10 trillion yuan)	300	0.294	0.235	0.021	1.291	1.700	6.282
Industrial structure	Value added by the tertiary industry/value added by the secondary industry	300	1.338	0.736	0.314	5.297	3.011	13.708
Inflation level	Consumer Price Index	300	101.97	0.690	100.10	103.90	0.0125	3.044
Government support	Expenditure on urban and rural community construction (100 million yuan)	300	610.92	430.37	61.51	2413.84	1.293	4.776
Infrastructure construction	Highway passenger volume (100 million people/100,000 km)	300	0.419	0.358	0.007	1.572	1.035	3.348
Degree of opening up	Total import and export volume (US $100 billion)	300	1.546	2.410	0.003	12.796	2.456	9.067

.

The skewness of integrated urban–rural development is greater than 0 and the kurtosis is greater than 3, indicating a significantly positively skewed and highly peaked data distribution. In simpler terms, the integrated urban–rural development in most provinces is relatively low, with the majority of data concentrated around the mean, although there are also provinces with significantly higher development. The distribution of the digital economy follows a similar trend to integrated urban–rural development but peaks more sharply and exhibits a severe right-skewed data distribution. This suggests that the digital economy in most provinces is concentrated at lower values and highly centered around the mean, with only a few provinces having very high values. Thus, the imbalance in digital economic development is much higher than that in urban–rural integration.

Figure 2 depicts the annual average changes in integrated urban–rural development and the digital economy across the 30 provinces to visually show the changing trends more intuitively. It can be observed that both the digital economy and the urban–rural integration index in China showed an upward trend from 2013 to 2022.

Due to differences in regional resource endowments, there are evident regional heterogeneity characteristics in both the digital economy and integrated urban–rural development. Overall, the digital economy and integrated urban–rural development in the Eastern, Central, and Western regions showed an increasing trend (Figure 3). Development in Eastern China is significantly ahead, followed by the Central region, while the Western region is slightly behind Central China, demonstrating an “East-Central-West” descending gradient trend. When looking at individual regions, the gap in integrated urban–rural development between the Eastern, Central, and Western regions is gradually narrowing, while the gap in the digital economy is widening. The digital economy in Eastern China far exceeds that in the Central and Western regions.

4. Empirical Test

4.1. Impact of the Digital Economy on Integrated Urban–Rural Development and Its Decomposition

Before examining the impact of the digital economy on integrated urban–rural development, we first used the Levin–Lin –Chu (LLC) unit root test to check the stationarity of each panel series to avoid spurious regression. The unit root test results for each variable gave a p-value of 0.000, which significantly rejects the null hypothesis of non-stationarity (p > 0.05). Additionally, we used the Variance Inflation Factor (VIF) test to check for multicollinearity among the indicators, ensuring the validity of the estimation results.

Following the theoretical analysis above, we conducted robust standard error two-way fixed effects and random effects model regressions to explore the relationship between the digital economy and integrated urban–rural development, as shown in Models (1) and (2) in

Table 4. Impact and role of the digital economy on integrated urban–rural development.

Variables	Integrated Urban–Rural Development
	(1)	(2)	(3)	(4)	(5)
Digital economy	0.129 *** (0.042)	0.115 *** (0.039)	-	-	-
Digital infrastructure construction	-	-	0.482 * (0.258)	-	-
Digital industrialization	-	-	-	0.155 *** (0.048)	-
Industrial digitization	-	-	-	-	0.387 *** (0.111)
Economic development	0.018 *	0.010	0.022	0.023 *	0.021
Industrial structure	0.001	0.015	0.010	0.002	−0.000
Inflation level	−0.002	−0.001	−0.002	−0.002	−0.002
Government support	0.004	0.012 **	0.002	0.004	0.006
Infrastructure construction	0.019 **	0.028 ***	0.025 **	0.017 *	0.009
Degree of opening up	−0.007 ***	0.004	−0.001	−0.006 ***	0.005 ***
Cons	0.516 **	0.312	0.494 **	0.455 **	0.485 **
Years fixed	Yes	Yes	Yes	Yes	Yes
Provinces fixed	Yes	Yes	Yes	Yes	Yes
Years	10	10	10	10	10
Provinces	30	30	30	30	30
R2	0.8997	0.8872	0.8916	0.8962	0.8957

Note: Values in parentheses in the table represent robust standard errors. ***, **, and * indicate that the regression results passed the significance tests at the 1%, 5%, and 10% confidence levels, respectively.

. The xtoverid test results reject the null hypothesis in favor of the random effects model, hence we used the fixed effects model for estimation. The results of Model (1) indicate that the coefficient for the digital economy is 0.129, which is significant at the 1% statistical level. This suggests that higher digital economic development is more conducive to integrated urban–rural development, thereby verifying Hypothesis 1.

We sequentially replaced the digital economy indicator with digital infrastructure construction, digital industrialization, and industrial digitization in the two-way fixed effects model regression to further explore the decompositional impact of the digital economy on urban–rural integration, as shown in Models (3)–(5). The coefficients for digital infrastructure construction, digital industrialization, and industrial digitization are all positive and significant at the 10%, 1%, and 1% confidence levels, respectively. Among these, digital infrastructure construction is the strongest driver for promoting urban–rural integration. Specifically, for every 1% improvement, the degree of integrated urban–rural development increases by 0.482%. Enhanced digital infrastructure construction significantly narrows the digital divide between urban and rural areas, promotes efficient flow of resources and information, and strengthens connectivity between urban and rural areas. Industrial digitization follows next, with an increase of 1% in its index leading to a 0.387% increase in urban–rural integration. Application and transformation through digital technology enhance industrial efficiency and competitiveness, thereby driving coordinated development between urban and rural areas. Digital industrialization is at a lower level of development, with a 1% increase resulting in a 0.155% increase in urban–rural integration. While slightly lower than industrial digitization, this still demonstrates the positive impact of digital industrial development on optimizing the economic structure and promoting rural–urban coordinated development.

4.2. Robustness Checks

To further verify the reliability of the research conclusions, this paper conducted robustness checks using the instrumental variable method, core independent variable replacement method, and exclusion of municipal samples method.

4.2.1. Instrumental Variable Method

To some extent, the two-way fixed effects model can address the issue of omitted variables. However, integrated urban–rural development may also influence the demand for digital products among urban and rural residents, potentially leading to endogeneity problems and biased estimation results. Therefore, drawing on the approach of scholars like Huang et al. (2022) [26] and Yin et al. (2022) [27], we introduced the interaction term between the number of fixed telephone lines per hundred people in 1984 and the previous year’s national internet users (in tens of millions) as instrumental variables for the digital economy. We employed two-stage least squares regression for instrumental variable estimation, as shown in Model (1) in

Table 5. Robustness test.

Variables	(1)		(2)	(3)
	First-Stage	Second-Stage
Instrumental variable	0.006 *** (0.001)	-	-
Digital economy	-	0.223 *** (0.061)	0.014 *** (0.003)	0.101 *** (0.046)
Control variables	Control	Control	Control	Control
Cons	−1.343 ***	0.856 ***	0.552 **	0.533 **
Kleibergen–Paap rk LM statistic	-	15.91 (p-value = 0.0001)	-	-
Cragg–Donald Wald F statistic	-	86.820	-	-
Kleibergen–Paap rk Wald F statistic	-	29.586	-	-
Hansen J statistic	-	0.000	-	-
Years fixed	YES	YES	YES	YES
Provinces fixed	YES	YES	YES	YES
R2	-	0.9834	0.8994	0.9154

Note: Values in parentheses in the table represent robust standard errors. *** and ** indicate that the regression results passed the significance tests at the 1% and 5% confidence levels, respectively.

. The rationale for choosing this variable is that the number of fixed telephone lines reflects the telecommunications infrastructure at that time, and its historical deployment partly influences subsequent digital infrastructure construction and skill application. Fixed telephones, as traditional communication tools, have minimal impact on integrated urban–rural development, thus satisfying the condition of being correlated with the core explanatory variables and possessing exogeneity.

The first-stage regression results from Model (1) in Table 5 indicate a positive and significant coefficient for the instrumental variable at the 1% level, suggesting a positive correlation between the number of telephones per hundred people and the digital economy. The second-stage regression results show that the Cragg–Donald Wald F-statistic is 86.82 and the Kleibergen–Paap rk Wald F-statistic is 29.56. Both exceed the critical value of 16.38 at the 10% level, indicating rejection of the null hypothesis of the weak instrument. The Kleibergen–Paap rk LM statistic is 15.91, with a p-value less than 0.01, suggesting no issues with instrument under-identification. These tests confirm that the instrumental variables selected for this study are reasonable and effective.

4.2.2. Independent Variable Replacement Method

We recalculated the digital economy indicators using principal component analysis (PCA). The analysis results, shown in

Table 6. Principal component analysis effect test.

Components	Eigenvalue	Variance Contribution	Cumulative Variance Contribution
1	5.4088	0.5409	0.5409
2	2.0591	0.2059	0.7468
3	1.0398	0.1040	0.8508
4	0.5997	0.0600	0.9107
5	0.3303	0.0330	0.9438
6	0.1745	0.0175	0.9612
7	0.1379	0.0138	0.9750
8	0.1104	0.0110	0.9860
9	0.0832	0.0083	0.9944
10	0.0563	0.0056	1.0000
KMO	0.836
Bartlett’s Test	0.000

, indicate that the original 10 indicators were compressed into three principal components with a cumulative variance contribution rate of 85.08%, effectively capturing most of the information from the original data. Therefore, we selected these three principal components and used their proportion of eigenvalues as weights to perform a weighted sum, constructing a new digital economy index. Additionally, we conducted a Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy test for the model, which gave a KMO value of 0.832, and a significant Bartlett’s test of sphericity (p-value = 0.000). These results indicate that the construction of the digital economy index is reasonable. Furthermore, we included the weighted digital economy indicator in the two-way fixed effects model to regress integrated urban–rural development. As shown in Model (2) in Table 5, the digital economy continues to have a significant positive impact on urban–rural integration, confirming that the previous conclusions are robust.

4.2.3. Exclusion of Municipal Samples

Compared to other provinces, the municipalities in China (Beijing, Tianjin, Shanghai, and Chongqing) have distinct economic policies and foundations, which may introduce bias in the regression results. Therefore, we excluded samples from these four municipalities and conducted the regression analysis again to verify the robustness of the results. As shown in Model (3) in Table 5, the coefficient for the digital economy remained significantly positive, with only minor changes in its value after adjusting the samples. This indicates that the regression results are robust.

4.3. Nonlinear Regression of the Digital Economy on Integrated Urban–Rural Development

Model (1) in

Table 7. Nonlinear regression of digital economy on integrated urban–rural development.

Variables	(1)	(2)
Digital economy	0.332 *** (0.067)	-
The square term of the digital economy	−0.190 *** (0.064)	-
Digital economy < 0.3392	-	0.346 *** (0.089)
Digital economy ≥ 0.3392	-	0.250 *** (0.069)
Control variables	0.661 ***	0.714 ***
Cons	Control	Control
Years fixed	YES	YES
Provinces fixed	YES	YES
R2	0.9053	0.7594

Note: Values in parentheses in the table represent robust standard errors. *** indicates that the regression results passed the significance tests at the 1% confidence levels, respectively.

presents the regression results of the squared term of the digital economy on integrated urban–rural development. The coefficient for the digital economy is significantly positive, while the coefficient for the squared term is significantly negative. This suggests a preliminary identification of an inverted “U”-shaped relationship between the digital economy and integrated urban–rural development. According to the U-Test [61], the estimated turning point (0.873) is not within the range of the digital economy’s upper and lower limits. This indicates that the current impact of the digital economy on urban–rural integration has not yet shifted from promotion to inhibition. It is still in the left half of the inverted “U”-shaped curve, meaning that the digital economy is currently promoting urban–rural integration development.

We initially employed the Bootstrap method with 300 repeated samples to assess the presence of a panel threshold effect based on the digital economy before conducting the threshold effect analysis. The results indicate a p-value of less than 0.05 for the single threshold test, with a threshold value of 0.339, as detailed in

Table 8. Threshold effect test.

Model	Fstat	p Value	BS Iterations	Threshold
				10%	5%	1%
Single threshold	28.52	0.0167	300	22.058	24.000	30.795
Double threshold	21.78	0.1533	300	26.530	36.935	54.248
Triple threshold	16.93	0.5700	300	37.483	43.606	54.832

. This finding suggests that the promotion effect of the digital economy on urban–rural integration is influenced by a single threshold effect. Specifically, when the digital economy is below the threshold value of 0.339, it exerts a stronger positive driving effect on urban–rural integration, with an elasticity coefficient of 0.346. However, when the digital economy exceeds this threshold value, its promoting effect on urban–rural integration slightly diminishes, with an elasticity coefficient of 0.250. The threshold regression results are elaborated in Model (2) in Table 7, further confirming that the current promotion effect of the digital economy on urban–rural integration exhibits a trend of marginal diminishing returns. Thus, hypothesis 2 is validated.

4.4. Regional Heterogeneity Analysis

Based on the previous analysis, the samples were divided into three parts: Eastern, Central, and Western regions, to further study regional differences in the impact of the digital economy on integrated urban–rural development. The detailed results are shown in

Table 9. Regional heterogeneity of the impact of the digital economy on integrated urban–rural development.

Variables	Eastern Region	Central Region	Western Region
Digital economy	0.109 ** (0.048)	0.348 *** (0.094)	0.211 (0.216)
Control variables	Control	Control	Control
Cons	0.506	1.409 **	0.512
R2	0.9111	0.9529	0.9347

Note: Values in parentheses in the table represent robust standard errors. *** and ** indicate that the regression results passed the significance tests at the 1% and 5% confidence levels, respectively.

. From the regression results (the regional heterogeneity-grouped regression passed the Fisher test, as in

Table 10. Regional breakdown of the impact of the digital economy on urban–rural integration.

Variables	(1)			(2)			(3)
	Eastern Region	Central Region	Western Region	Eastern Region	Central Region	Western Region	Eastern Region	Central Region	Western Region
Dig1	0.235 (0.398)	0.767 * (0.380)	0.894 ** (0.356)	-	-	-	-	-	-
Dig2	-	-	-	0.147 ** (0.063)	0.154 (0.166)	−0.059 (0.192)	-	-	-
Dig3	-	-	-	-	-	-	0.350 ** (0.146)	0.829 *** (0.239)	−0.022 (0.488)
Control variables	Control	Control	Control	Control	Control	Control	Control	Control	Control
Cons	0.401	1.564 **	0.619 **	0.437	1.397 *	0.382	0.594 **	1.478 **	0.394
Years fixed	YES	YES	YES	YES	YES	YES	YES	YES	YES
Provinces fixed	YES	YES	YES	YES	YES	YES	YES	YES	YES
R2	0.8961	0.9522	0.9418	0.9095	0.9480	0.9329	0.9096	0.9517	0.9329

Note: Values in parentheses in the table represent robust standard errors. ***, **, and * indicate that the regression results passed the significance tests at the 1%, 5%, and 10% confidence levels, respectively.

), it can be seen that in the Eastern and Central regions, the digital economy significantly promotes integrated urban–rural development, with Central China demonstrating a stronger promoting effect. Conversely, Western China shows no significant impact. This conclusion aligns with the findings of scholars such as Li et al. (2024) [32] and Wang et al. (2022) [37]. To delve deeper into these findings, further regional decomposition of the effects of the digital economy is warranted.

The regional heterogeneity regression results of the decomposed effects of the digital economy are presented in Table 10. In Eastern China, the promotion of urban–rural integration by the digital economy stems from digital industrialization and industrial digitalization. In Central China, the promotion is attributed to digital infrastructure construction and industrial digitalization. In the Western region, the promotion effect is solely derived from digital infrastructure construction.

Combining the phenomenon of the “East–Central–West” gradient decline in digital infrastructure construction, digital industrialization, and industrial digitalization (The detailed annual trends for these indicators in the Eastern, Central, and Western regions are available upon request from the first author), the regional heterogeneity in the enabling effect of the digital economy on urban–rural integration can be explained as follows:

Western China is still in the early stages of digital economic development, primarily concentrating on the expansion of digital infrastructure. At this phase, the integration of the digital economy with physical industries lags behind, with traditional industries retaining a substantial presence. As a result, only digital infrastructure construction exhibits a certain positive effect on integrated urban–rural development, while digital industrialization and industrial digitalization show no significant impact. Overall, this results in an insignificant effect of the digital economy on urban–rural integration in Western China.

The Central region has relatively favorable urban conditions, development levels, and economic vitality, placing it in the expansion phase of digital economic development. The digital economy’s promotion effect mainly comes from digital infrastructure and industrial digitalization, with the latter having a significantly higher driving effect than in the Eastern region. The digital economy benefits are more fully realized in this region, making the Central region the strongest in promoting urban–rural integration through digital economic development.

As the leading area in information development, the Eastern region has reached a mature stage in digital economic development. It excels in industrial integration and efficiency. The promotion effect of the digital economy on urban–rural integration comes from digital industrialization and industrial digitalization, validating the phenomenon observed. Therefore, the digital economy’s promotion of urban–rural integration is significant and stable in the Eastern region, although the effect is slightly less than in the Central region due to the already high level of integration.

5. Conclusions, Implications, and Future Research

5.1. Conclusions

The core driving force behind the integration of the digital economy and urban–rural development primarily stems from proactive leadership and promotion by government departments. Governments have played a crucial role by fostering infrastructure construction, creating a favorable digital environment, and formulating forward-looking digital policies, thus providing solid support and assurance for the deep integration and extensive application of the digital economy between urban and rural areas. This study analyzed the theoretical mechanisms through which the digital economy impacts China’s urban–rural integration. Based on provincial-level data, it calculated the digital economy and integrated urban–rural development index for 30 provinces in China from 2013 to 2022. Various models, such as panel fixed effects, quadratic models, and threshold effects models, were employed to empirically test the impact of the digital economy on integrated urban–rural development and analyze regional heterogeneity. The conclusions drawn are as follows:

Firstly, the digital economy has significantly promoted the degree of integrated urban–rural development. After robustness checks using instrumental variable methods and substitution variable methods, and excluding samples from direct-controlled municipalities, this conclusion remains valid. This finding aligns with that of most researchers, although there are slight differences in the specific estimated impact. Huang et al. (2022) [26] estimated a coefficient of 0.0578 for the impact of the digital economy on integrated urban–rural development, while Li et al. (2023) [62] estimated a coefficient of 0.170. The reasons can be attributed to two aspects: One is the lack of unified evaluation indicators for the digital economy and integrated urban–rural development. Different studies often select indicators based on their respective research perspectives and emphasis, leading to differences in quantification results for urban–rural integration. Another is the year of study, as the impact of the digital economy on integrated urban–rural development varies between different years. Despite these slight differences, overall, various studies have consistently affirmed the positive role of the digital economy in promoting urban–rural integration.

Secondly, In the internal structure of the digital economy, digital infrastructure construction exerts the greatest pull on integrated urban–rural development, and the promotion effect of industrial digitization is significantly higher than that of digital industrialization. Although the academic community has not yet explored how each dimension of the digital economy specifically affects urban–rural integration, a scholar has analyzed how each dimension of the digital economy affects the urban–rural income gap [51] and consistently concluded that digital infrastructure exerts the strongest driving force.

Third, the nonlinear regression results indicate that the promotive effect of the digital economy on integrated urban–rural development is showing a trend of marginal decline but has not yet reached the ‘inflection point’ where the effect shifts from promotion to inhibition. While some studies have noted the nonlinear trend of the digital economy’s impact on urban–rural integration [36], the inflection point value has not been clearly identified. In a study from the perspective of common prosperity [63], the inflection point value was calculated to be 0.975, which is essentially consistent with this study (0.873), indirectly validating the credibility of our conclusions.

Fourth, in terms of regional heterogeneity, the digital economy has a stronger promotive effect on urban–rural integration in Central China, followed by Eastern China, while Western China shows no significant impact. To deeply analyze the reasons behind this phenomenon, we further decomposed the digital economy indicators. The digital economy in Eastern China has reached a mature stage, and its promotive effect on urban–rural integration comes from digital industrialization and industrial digitization. Central China is in an expansion phase, with its promotive effect coming mainly from digital infrastructure construction and industrial digitization. Western China is still in the stage of digital infrastructure popularization, thus its promotive effect on urban–rural integration comes only from digital infrastructure construction. Although scholars have recognized regional heterogeneity in the digital economy’s impact on integrated urban–rural development, their reasons remain at the theoretical discussion stage, without empirical analysis.

5.2. Implications

The conclusions of this study also provide the following policy insights: (1) Optimize the internal structure of the digital economy. Firstly, emphasis should be placed on strengthening digital infrastructure construction, as it is the primary driver of urban–rural integration and has a significant impact across all regions. Secondly, greater support should be given to industrial digitalization to increase its share within the digital economy. Attention should also be paid to the development of digital industrialization, particularly in central and western regions, to achieve comprehensive digital economic development. (2) Given that the promotion of the digital economy on urban–rural integration shows a trend of diminishing marginal returns, close attention should be paid to changes in this trend. Adjust development policies in a timely manner before reaching the “turning point” from promotion to inhibition effects. Continuously optimize digital economic policies to maximize their promotion of urban–rural integration. (3) Tailor policies to local conditions and classify them accordingly. For Eastern China, focus on enhancing high-end and intelligent development, accelerating the transformation of deep integration between the digital economy and the real economy. For Central China, concentrate on the forefront of digital industrialization strategy, nurture and strengthen emerging digital industries such as artificial intelligence, big data, and blockchain, and enhance the level of communication equipment and key software. For Western China, draw on the development ideas and experiences of Central China, first accelerate the construction of new digital infrastructure, and then leverage the advantages of low digital technology thresholds and strong penetration to promote the emergence of digital trade, smart agriculture, and other emerging business models, thereby promoting the integration of the digital economy with the real economy and advancing industrial digitalization and digital industrialization development.

These policy recommendations have broad applicability and can provide valuable insights for other countries. Due to varying economic, social, and technological development statuses, as well as existing issues in urban–rural integration, the applicability of these suggestions may differ among countries. However, in practical application, each country should adjust and optimize these recommendations according to its specific needs and technological levels. Firstly, local governments should strengthen digital infrastructure construction. This recommendation is relevant to all countries, especially those with weak digital infrastructure in rural areas. Secondly, promoting industrial digitization transformation is particularly applicable to countries reliant on traditional industries that require efficiency upgrades. Thirdly, local governments should monitor the marginal effects of the digital economy in a timely manner, and based on development stages and resource allocation, formulate rational digital investment plans to avoid resource waste.

Overall, these findings offer several practical and theoretical implications. First, they provide robust evidence to support the formulation and optimization of policies related to the digital economy and integrated urban–rural development, aiding government and policymakers. Second, they offer direction for businesses and investors in their investment decisions, helping to steer capital toward more promising integrated urban–rural projects and promoting coordinated economic development between urban and rural areas. Lastly, the exploration of nonlinear effects, the various dimensions of the digital economy, and regional heterogeneity in this study provide a foundation and reference for subsequent related research, advancing academic development in this field.

5.3. Future Research

The current study has some limitations, which should be addressed in future research in the following ways: (1) Emphasize the advantages of DMOs in digital marketing and digital regulation. Incorporate the coordinating functions of DMOs into future studies on urban–rural integration to fully unleash their potential in optimizing resource allocation, enhancing governance efficiency, and accelerating information flow and sharing between urban and rural areas. (2) Strengthen international comparisons. Select developed countries with typical and representative urban–rural development issues as research subjects to explore the common patterns and unique paths of the digital economy and urban–rural integration across different regions and stages. This can provide universally applicable policy insights and development experiences. (3) Expand the sample range. Include data spanning a longer period to gain a deeper understanding of the trends in the digital economy and integrated urban–rural development. (4) Consider the lag and long-term effects of policy implementation. Future studies can further employ the DID model to deepen understanding of the long-term impact of digital economy policies on integrated urban–rural development.