The Impact Mechanism of Digital Rural Construction on Land Use Efficiency: Evidence from 255 Cities in China

Abstract

Based on panel data from 255 prefecture-level and above cities in China from 2011 to 2023, this article innovatively integrates digital rural construction, the digital literacy of farmers, green production of farmers, and land use efficiency into a unified framework for theoretical and empirical research. Our research has found that digital rural construction can indeed have a direct promoting effect on land use efficiency, with an impact coefficient of 0.451, which is significant at the 1% level. The addition of control variables and robustness tests indicates that this result is robust. Building digital villages can also boost land use efficiency in a roundabout way by enhancing farmers’ digital skills and promoting eco-friendly farming practices, with impact coefficients of 0.445 and 0.283, respectively, both significant at the 1% level. The impact of digital rural areas on land use efficiency is heterogeneous. Our studies have indicated that the impact of digital rural development on the shift towards green production among farmers cultivating cash crops is more significant compared to those growing grain crops. Additionally, the influence of digital rural development on improving land use efficiency is more pronounced among farmers who are risk-averse compared to those who are not.

1. Introduction

In this rapid development period of digitization and informatization, digital resources, as a key element for the development and revitalization of rural areas, have become increasingly important globally. Improving their utilization efficiency is not only an important goal of digital rural construction but also an inherent requirement for achieving the high-quality development of the rural economy. To this end, the General Office of the State Council issued the “Overall Plan for Comprehensive Reform Pilot of Digital Rural Construction” in January 2022, which explicitly proposes to “encourage pilot areas to explore the refinement and improvement of digital rural construction standards through detailed evaluation of digital infrastructure construction and application efficiency, and encourage the efficient implementation of digital rural construction through measures such as legal consultation, technical cooperation, and data sharing”. For a long time, the extensive use of chemical fertilizers, pesticides, herbicides, and other chemicals in the utilization of farmland in China has led to various problems such as soil nutrient imbalance, fertility decline, and reduced organic matter content, resulting in increasingly severe soil and water pollution and seriously weakening the supply capacity of high-quality and safe agricultural products in China [1]. Given the limited nature of land resources, only models that effectively utilize land resources can help achieve the mission of sustainable development. The conflict between land resources and human socio-economic development is more evident in urban areas, where more than half of the world’s population resides, producing over 75% of global GDP and generating 71–76% of energy-related carbon dioxide emissions worldwide [2]. Therefore, how to effectively utilize urban land resources in terms of economic, social, and environmental sustainability has become a key issue in addressing the urgent challenge of land scarcity. Enhancing the productivity of urban land utilization necessitates an intensified implementation of digital technologies and a swifter change in the approaches to production, daily life, and management [3,4]. To foster the growth of the digital economy by leveraging digital technology to its fullest, it is essential to unlock consumer potential and elevate the scale of domestic economic circulation [5]. In particular, by expanding the industrial chain and promoting production, circulation, supply, and consumption through leading enterprises, we can form a virtuous cycle of demand-driven production and production-driven supply, continuously creating new demands, meeting new demands, and guiding new demands, forming a good dynamic balance and continuously improving the level of economic development.

The development of digital infrastructure in rural areas, as an inevitable product of the informatization of agricultural and rural economic and social development and the improvement in farmers’ information skills, marks an important process of modernization transformation in agriculture and rural areas [6]. Land use efficiency refers to the comprehensive utilization effect of land resources in agricultural production, ecological protection, residential construction, and other aspects within a certain period of time [7]. In traditional agricultural societies, the improvement of land use efficiency is often limited by factors such as information asymmetry and outdated management methods. However, with the intervention of digital technology, this situation is undergoing fundamental changes. The construction of digital rural areas has achieved the precise monitoring, scientific management, and efficient allocation of land resources by introducing technologies such as the Internet of Things, big data, cloud computing, and artificial intelligence, greatly improving the output and utilization rate of land [8]. Firstly, the application of digital technology has made the monitoring and management of land resources more refined. By installing sensors and cameras in the fields, real-time data such as soil moisture, temperature, and light can be collected to provide accurate decision support for agricultural production [9]. At the same time, the application of unmanned aerial vehicles and satellite remote sensing technology has expanded the monitoring range of land resources, provided more comprehensive data, and provided a scientific basis for land planning and utilization [10]. Secondly, the application of big data and cloud computing technology provides powerful data processing capabilities for the analysis of and decision-making about land resources. Gathering and examining extensive land use information can uncover trends and issues in land use, offering insights to guide the strategic distribution and efficient management of land resources [11]. In addition, by establishing a land resource database, the sharing and exchange of land information can be achieved, improving the transparency and efficiency of land resource management [12]. Once again, the application of artificial intelligence technology has provided the possibility for the intelligent management and automated operation of land resources. Through machine learning and deep learning algorithms, the utilization of land resources can be predicted and evaluated, providing scientific guidance for the rational use of land resources [13]. Meanwhile, the application of intelligent agricultural machinery and automation equipment can reduce manpower input and improve the efficiency and accuracy of land operations.

Finally, developing digital rural regions also encompasses the establishment of rural information systems, such as broadband networks, intelligent terminals, etc. The improvement of these infrastructures provides farmers with more channels for obtaining information and learning platforms, enhances their information literacy and skill level, and provides a talent guarantee for the efficient utilization of land resources [14].

In summary, the impact of digital rural construction on land use efficiency is comprehensive. It not only changes the monitoring and management methods of land resources, improving the utilization efficiency of land resources, but it also promotes the diversified development of the rural economy and improves the living standards of farmers. With the continuous advancement and application of digital technology, the construction of digital villages will undoubtedly provide a stronger impetus for rural revitalization and agricultural modernization. Therefore, in the era of the digital economy, exploring the potential relationship between digital rural construction and land utilization rates from both theoretical and empirical perspectives can provide certain empirical evidence for promoting the improvement of the land utilization rate in China, which has important theoretical significance.

2. Theoretical Analysis and Research Hypotheses

2.1. Analysis of the Impact of Economic Growth on Land Use Efficiency

The impact of economic growth on land use efficiency is multifaceted, and we can analyze it from the following perspectives: Firstly, according to the “China Natural Resources Development Report (2023)”, research has found that there is a strong decoupling between the consumption of arable land, gardens, forests, and grasslands and economic development as a whole. This means that while the economy is growing, the consumption of these land resources has not increased accordingly, which may be due to the combined effects of various factors such as the improvement in agricultural production efficiency, the optimization of industrial structure, government policy protection, ecological compensation measures, improvements in land use efficiency, and the transformation of the economic growth mode. Secondly, the level of urbanization is an important characteristic of urban economic development, and urbanization can significantly promote land use efficiency. In the process of urbanization, factors such as the transfer of labor from rural areas to cities and the increase in government public expenditure will affect the efficiency of urban land use. For example, the degree of labor transfer has a positive impact on urban land use efficiency, while the quality of labor has no effect on land use efficiency. Furthermore, the relationship between economic development and urban land use efficiency can be represented as an inverted “U” curve. In the early stages of economic development, land use efficiency may increase with the development of the economy, but when the economy reaches a certain level of development, land use efficiency may tend to level off or even decline. Finally, there is a significant positive correlation between the level of economic development in different regions and the efficiency of urban land use. The higher the level of economic development in a city, the higher its land use efficiency. At the same time, the input of mobile factors (capital and labor) is conducive to improving the efficiency of urban land use, while the current utilization of non-mobile factors (land) is not conducive to improving the efficiency of urban land use. In addition, the development of the digital economy has significantly promoted the efficiency of green land use in cities, but this impact exhibits regional and batch heterogeneity. Infrastructure construction has a negative regulatory effect on the development of the digital economy and the efficiency of green land use in cities; in particular, the negative impact of information infrastructure construction is the greatest.

2.2. The Direct Impact of Digital Rural Construction on Land Use Efficiency

As an important aspect of the construction of Digital China, the digital countryside is an important measure to drive and enhance the modernization of agriculture and rural areas as a whole. Its impact on land use efficiency is mainly reflected in the following aspects: Firstly, the construction of a digital countryside can achieve the comprehensive monitoring and management of land resources by establishing a sky–ground integrated agricultural and rural observation network and data collection system, which can more reasonably optimize land resource management and enhance the productivity of land use [15]. Secondly, the construction of digital rural areas optimizes the agricultural production process through information technology, such as intelligent irrigation, precision fertilization, etc. The application of these technologies reduces resource waste and improves land output efficiency [16]. Thirdly, the construction of digital rural areas promotes the digital transformation of the agricultural industry chain, directly connecting with the consumer market through e-commerce platforms, shortening the circulation chain, reducing intermediate costs, and increasing the added value of agricultural products, thereby improving the economic efficiency of land use [17]. The fourth aspect is the construction of digital rural areas. By building an integrated digital management platform, the real-time monitoring and efficient management of rural resource allocation, environmental protection, population dynamics, and other aspects can be achieved, improving the efficiency of rural governance and public satisfaction [18], directly affecting land use efficiency. The fifth aspect is to accelerate the informatization process of rural areas in the construction of digital rural areas, improve the infrastructure system, enhance the depth and breadth of network coverage, promote the implementation of various digital transformation projects, improve the overall informatization level of rural areas [19], and provide more efficient information services and decision support for land use. The sixth aspect is the construction of the digital countryside, which helps farmers master necessary digital skills through systematic training and education and cultivates a new type of professional farmer team that adapts to the requirements of the new era. These farmers can use modern information technology to improve agricultural production efficiency [20], thereby enhancing land use efficiency. The integration of digital technology has opened up new ideas and paths for rural governance. Utilizing technologies such as big data analytics, it is possible to deeply explore and accurately understand the real needs and wishes of villagers, provide scientific and accurate data support for government decision-making, ensure that policy formulation is more closely related to people’s livelihoods and in line with reality [21], and improve the rational allocation and utilization efficiency of land resources. The eighth aspect is to coordinate the development of digital villages and smart cities, promote the digitization, networking, and intelligent development of urban and rural production, life, and ecological spaces, accelerate the formation of a digital urban–rural integration development pattern that is co-built, shared, interconnected, distinctive, and mutually beneficial [22], and enhance the efficiency of land resource utilization in urban and rural planning.

Hypothesis 1:

Digital rural construction can effectively improve land use efficiency.

2.3. The Indirect Impact of Digital Rural Construction on Land Use Efficiency Through Farmers’ Digital Literacy

2.3.1. Selection of Indirect Factors Influencing Land Use Efficiency in Digital Rural Construction

This paper believes that the selection of influencing factors indirectly affecting land use efficiency in digital rural construction mainly has three selection bases: a legal and policy basis, theoretical basis, and practical basis. The legal and policy basis refers to the interpretation of the basic concepts of rural revitalization in the documents of the Party and the government, such as the No. 1 central document “Opinions of the CPC Central Committee and the State Council on the Implementation of the Rural Revitalization Strategy” and “Rural Revitalization Strategic Plan (2018–2022)” in 2018, “Suggestions of the CPC Central Committee on the Formulation of the 14th Five-Year Plan for National Economic and Social Development and the Vision Goals for 2035” in 2020, “Opinions of the No. 1 central document in 2021 Comprehensively Promoting Rural Revitalization and Accelerating Agricultural and Rural Modernization”, and People’s Republic of China in 2021 “Rural Revitalization Promotion Law Opinions on Doing a Good Job in the Key Work of Comprehensive Rural Revitalization” in 2022, “Opinions on Doing a Good Job in the Key Work of Comprehensive Rural Revitalization” in 2023, etc. Theoretical basis refers to theoretical achievements based on the reports of the 18th and 19th National Congress, the definition of the connotation of rural revitalization in the academic community, and the impact of digital rural construction on land use efficiency in the academic community. Realistic basis refers to the selection of proxy indicators or approximate indicators based on the availability and practicality of data, mainly used for specific measurements.

This article mainly follows these three selection criteria, and the specific selection process is as follows:

The first step is to select macro factors. Based mainly on the reports of the 18th, 19th, and 20th National Congresses, the “Rural Revitalization Strategy Plan (2018–2022)”, the “Opinions of the Central Committee of the Communist Party of China and the State Council on Implementing the Rural Revitalization Strategy”, the “Rural Revitalization Promotion Law”, the “Outline of the Digital Rural Development Strategy”, and other party and government documents, and following the consensus formed by the academic community on five macro factors, seven primary indicators have been selected, including digital policies, the digital literacy of farmers, digital talent cultivation, agricultural technology penetration, infrastructure construction, the digitalization of production factors, and the green production of farmers.

The second step is to make a selection based on universal factors. The main basis is the academic community’s theoretical achievements related to the scientific connotations of digital villages and land use efficiency. Drawing on and refining Party and government documents such as the “Rural Revitalization Strategy Plan (2018–2022)” and the “Suggestions of the Central Committee of the Communist Party of China on Formulating the 14th Five-Year Plan for National Economic and Social Development and the Long-Range Objectives Through the Year 2035” and the “14th Five-Year Plan for Digital Economic Development”, fully considering the actual situation of digital village construction indicators, and combining the opinions of experts in the research group, five universal factors have been selected: farmers’ digital literacy, farmers’ technology penetration, infrastructure construction, the digitalization of production factors, and farmers’ green production.

The third step is to select the core factors. Mainly based on the selection results of universal factors, field research by the research team, data availability, and practical operability, and considering the comparability of indicators, specific measurement formulas for the two core factors of farmers’ digital literacy and green production were selected through unit conversion processing, greatly improving the practicality of the indirect influencing factors.

2.3.2. The Indirect Impact of Digital Rural Construction on Land Use Efficiency Through Farmers’ Digital Literacy

With the rapid development of technology, digital technology has penetrated into all aspects of agricultural production, becoming a key force in improving agricultural production efficiency, optimizing resource utilization, and enhancing the quality of agricultural products. Having a good digital literacy not only enables farmers to more effectively access and understand modern agricultural information, but it also enables them to adopt new production technologies for agricultural production management and decision-making, thereby indirectly improving land use efficiency. On the one hand, enhancing digital proficiency enables farmers to more effectively comprehend and apply digital technologies like the Internet of Things, big data analytics, and intelligent devices for the oversight and management of their farming operations [23]. For example, by installing soil moisture sensors, farmers can accurately understand soil conditions, arrange irrigation reasonably, avoid water waste, improve crop growth conditions, and increase land productivity. On the other hand, the application of digital technology can also promote farmers’ ability to obtain and analyze market information. Through online platforms, farmers can understand market demand, price changes, and consumer preferences in a timely way and adjust their planting structure and production scale accordingly, making their land use more in line with market demands and improving the economic value of their land. In addition, enhancing farmers’ digital literacy can also promote the dissemination and learning of agricultural knowledge. Online education platforms and agricultural expert systems can provide professional agricultural technology training and consulting services for farmers, helping them master more advanced planting techniques and management methods, improving land use efficiency and output quality [24]. Finally, enhancing the digital literacy of farmers can help attract more young people to stay in rural areas, who typically have a higher acceptance and application ability of new technologies [25]. These young people can become the backbone of digital rural construction, and through their innovation and entrepreneurship, they can develop more new methods and models to improve land use efficiency.

Hypothesis 2:

The development of digital infrastructure in rural regions can significantly boost the efficiency of land use by enhancing the digital literacy of farmers.

2.3.3. The Indirect Impact of Digital Rural Construction on Land Use Efficiency Through Green Production by Farmers

Green production for farmers refers to adopting environmentally friendly, energy-saving, and efficient production methods in the agricultural production process to reduce negative impacts on the environment, improve land use efficiency, and ensure the quality and safety of agricultural products [26]. The application of digital technology helps to plan and utilize land resources more scientifically. Through a digital land management system, farmers can better understand the use of land, undertake reasonable planning and allocation processes, and avoid the waste and abuse of land resources [27]. The construction of the digital countryside promotes the green transformation of agricultural product processing, packaging, cold chains, warehousing, and other links through digital means such as e-commerce platforms, builds green supply chains, promotes green logistics, reduces environmental pollution, and improves land use efficiency [28]. Through online education platforms and agricultural expert systems, we provide professional agricultural technology training and consulting services to farmers, helping them master more advanced planting techniques and management methods and improve land use efficiency and output quality [29]. The construction of the digital countryside encourages the agricultural machinery and equipment industry to develop the industrial Internet, improve the intelligence level of agricultural equipment, and promote the integrated application of informatization and agricultural equipment, agricultural machinery operation services, and agricultural machinery management, so as to improve the production efficiency of land [30]. This will enable to the following measures to be taken: Creating a national platform for monitoring rural ecosystems, employing satellite remote sensing, unmanned aerial vehicles, and high-definition remote video surveillance systems to focus on monitoring vulnerable and sensitive areas within rural ecosystems, thereby enhancing the overall quality of rural beautification efforts [31]. Establishing an integrated surveillance system for the rural living environment. Enhancing the surveillance and safeguarding of the quality of rural drinking water sources, ensuring a continuous and comprehensive oversight of rural contaminants and their sources, guiding the public to actively participate in rural environmental network supervision and jointly maintain a green living environment [32]. We will coordinate the development of digital villages and smart cities, promote the digitization, networking, and intelligent development of urban and rural production, life, and ecological spaces, and accelerate the formation of a digital urban–rural integration development pattern that is co-built, shared, interconnected, distinctive, and mutually beneficial [33].

Hypothesis 3:

Developing digital rural initiatives can significantly boost land use efficiency by facilitating farmers’ adoption of green production practices.

In summary, the path through which digital rural construction affects land use efficiency is shown in Figure 1. There are mainly two paths of direct and indirect impact. According to the three-step screening method of influencing factors, two core factors of farmers’ digital literacy and green production are selected.

2.4. The Causal Mechanism of the Impact of Digital Rural Construction on Land Use Efficiency

Firstly, data-driven decision support. Digital rural areas can collect and analyze a large amount of land use and market demand data through the application of big data and information technology. These data provide a scientific basis for decision-making by the government and farmers. Through data analysis, it is possible to accurately understand the suitability of land to market demands, thereby optimizing the use and allocation of land. For example, decision support tools based on geographic information systems (GISs) can help farmers choose suitable crops to plant, thereby improving land use efficiency.

Secondly, the application of intelligent agricultural technology. Intelligent agriculture is an important component of digital rural construction, which promotes the modernization of agriculture through technologies such as the Internet of Things, artificial intelligence, and drones. The application of intelligent agricultural technology can achieve precision agricultural management, improving crop yields and quality. For example, using sensors to monitor soil moisture and nutrients can achieve precise irrigation and fertilization, thereby improving land use efficiency and resource utilization.

Thirdly, rural e-commerce and market access. The construction of digital rural areas has promoted the development of rural e-commerce, enabling farmers to better access the market. Through e-commerce platforms, farmers can directly sell their agricultural products to consumers, reducing intermediaries and increasing product added value. The convenience of market access encourages farmers to engage in diversified production and improve the comprehensive utilization rate of land.

Fourth, community collaboration and information sharing. The construction of digital rural areas has promoted collaboration and information sharing in rural communities. With the support of digital platforms, farmers can share resources, technology, and information to form an intensive management model. This collaboration not only improves the efficiency of land use but also enhances the sense of cooperation among farmers and the development of a collective economy.

Fifth, government policies and infrastructure construction. The construction of digital rural areas requires government support and policy guarantees. The government’s investment in infrastructure construction, digital technology promotion, and related policy formulation directly affects the improvement in land use efficiency. For example, by improving rural digital infrastructure, the government can provide better Internet access services, promoting the rapid spread of information and the efficient allocation of resources. In addition, government policy guidance can encourage farmers to adopt new technologies and models, promoting the rational use of land.

Through a causal analysis, it can be concluded that there is a significant positive correlation between digital rural construction and land use efficiency. Specifically, for every one standard deviation increase in the completeness of digital infrastructure, land use efficiency increases by about 15%. For every 10% increase in the application of intelligent agricultural technology, land use efficiency improves by 11%. The development level of rural e-commerce has increased by 1 percentage point, and the land use efficiency has increased by about 8%. These results indicate that the construction of digital rural areas has played an important role in promoting the improvement of land use efficiency.

2.5. Boundary Between Urban and Rural Land

The “Regulations on the Classification of Urban and Rural Areas in Statistics” clearly divides China’s regions into urban and rural areas based on administrative divisions, the jurisdiction of residents’ committees and villagers’ committees confirmed by the civil affairs department, and actual construction as the basis for division. Urban area refers to the residential committees and other areas connected to the actual construction of district and municipal government residences in the city’s jurisdiction and cities without districts. Town area refers to the residential committees and other areas connected to the actual construction of the county people’s government and other towns outside the urban area. Independent industrial and mining areas, development zones, scientific research units, universities, and other special areas that are not connected to the actual construction of government residences and have a permanent population of over 3000 people, as well as the headquarters of farms and forest farms, are considered as townships. According to the Law of the People’s Republic of China on the Promotion of Rural Revitalization, rural areas refer to geographical complexes with natural, social, and economic characteristics, as well as multiple functions such as production, life, ecology, and culture, outside of urban built-up areas, including townships and villages.

Definition and characteristics of a city: Cities typically have a high population density. Cities are the centers of economic activity, with a large number of commercial, industrial, and service industries. Cities have a relatively complete infrastructure, such as transportation, education, healthcare, etc. A city is usually an administrative unit with its own local government. Cities usually have rich cultural facilities and social services.

Definition and characteristics of rural areas: The population density in rural areas is relatively low. Rural economic activities are usually dominated by agriculture. The infrastructure in rural areas is relatively limited, which may include fewer transportation, education, and medical facilities. Rural areas usually have more natural landscapes and fewer urbanization features. The community structure in rural areas is usually relatively close, with close connections between residents.

3. Research Methods and Data Sources

3.1. Empirical Model Setting

Firstly, in order to test the direct impact of digital rural construction on land use efficiency, the following benchmark regression model was constructed based on mature research and academic achievements [34]:

$$Lu{e}_i={\alpha }_0+{\alpha }_{1}Dig{r}_j+{\alpha }_{2}Control+\epsilon$$

(1)

In Equation (1), Lue is the dependent variable land use efficiency, $Digr$ is the dependent variable digital rural development level, $Control$ is a set of control variables, ${\alpha }_0$ is a constant term, ${\alpha }_1$ is the estimated coefficient of digital rural construction, ${\alpha }_2$ is the estimated coefficient of the control variable, and $\epsilon$ is the random interference term.

Secondly, in order to test the mediating effect of digital rural construction, this article draws on the research experience of Jiang Ting (2022) [35] in selecting mediating variables and argumentations, mainly identifying the causal relationship between explanatory variables and mediating variables. The selected mediating variables are farmers’ digital literacy and green production. The specific mediating effect model is constructed as follows:

$$Hd{l}_i={\beta }_0+{\beta }_{1}Dig{r}_j+{\beta }_{2}Control+\epsilon$$

(2)

$$Gp{f}_i={\beta }_0+{\beta }_{1}Dig{r}_j+{\beta }_{2}Control+\epsilon$$

(3)

In Equations (2) and (3), $Hdl$ and $Gpf$ are the mediating variables for farmers’ digital literacy and green production, respectively, while $D$igr is the explanatory variable for digital rural construction; $Control$ is a series of control variables; ${\beta }_0$ is a constant term, and ${\beta }_1$ and ${\beta }_2$ are estimated coefficients for the explanatory and control variables; and $\epsilon$ is a random interference term.

3.2. Variable Selection and Measurement

3.2.1. Dependent Variable

This article uses the SBM model of unexpected output super efficiency to measure land use efficiency [36]. Combined with the relevant literature [37,38,39,40], the input indicators for measuring land use efficiency in this paper are selected from three levels: inputs of land, capital, and labor. The expected output indicators are respectively expressed by the area of urban construction land, the amount of urban fixed assets investment, and the number of employees in the secondary and tertiary industries. The expected output indicators are selected from three levels: economic benefits, social benefits, and environmental benefits. They are respectively expressed by the added value of the secondary and tertiary industries, the average wage of urban unit employees, and the area of gardens and green spaces. The unexpected output indicators mainly consider the negative environmental benefits, expressed in terms of industrial sulfur dioxide emissions, industrial wastewater emissions, and industrial smoke emissions.

$${\beta }_{SE}=min{\displaystyle \frac{1+\frac{1}{m}{{\displaystyle \sum }}_{i=1}^m\frac{S_i^{-}}{X_{ik}}}{1-\frac{1}{s_1+s_2}({{\displaystyle \sum }}_{i=1}^{s_1}\frac{y_r^d}{y_{rk}^d}+{{\displaystyle \sum }}_{i=1}^{s_1}\frac{y_r^d}{y_{rk}^d})}}$$

$$s.t\left\{\begin{array}{c}{x}_i^k\ge {\displaystyle \sum }_{j=1,j\ne k}^n{x}_{ij}{\lambda }_j-s_i^{-}\\ y_{rk}^d\le {\displaystyle \sum }_{j=1,j\ne k}^n{y}_{rj}^d{\lambda }_j-y_r^d\\ y_{tk}^{ud}\ge {\displaystyle \sum }_{j=1,j\ne k}^n{y}_{tj}^{ud}{\lambda }_j+y_r^{ud}\\ 1-{\displaystyle \frac{1}{s_1+s_2}}({\displaystyle \sum }_{i=1}^{s_1}{\displaystyle \frac{y_r^d}{y_{rk}^d}}+{\displaystyle \sum }_{i=1}^{s_1}{\displaystyle \frac{y_r^d}{y_{rk}^d}})>0 \\ \lambda ,s^{-},s^{+}\ge 0\\ i=1,2,\dots ,m;j=1,2,\dots ,n\left(n\ne k\right);r=1,2,\dots ,s_{1;}t=1,2,\dots s_2\end{array} \right.$$

(4)

The SBM model provides a more accurate and comprehensive approach for assessing the performance efficiency of decision-making entities, especially when unexpected outputs need to be considered. The application of this model can help enterprises and organizations improve resource utilization efficiency, reduce environmental impact, and achieve sustainable development.

3.2.2. Core Explanatory Variables

The dependent variable of this article is the level of digital rural construction. The current academic basis for measuring the level of rural construction in a certain region mainly comes from two aspects. One is the indicator system independently constructed by scholars, and the core indicators are basically consistent [41,42,43]. The second is the “County Digital Countryside Index” jointly compiled by the New Rural Development Research Institute of Peking University and the Alibaba Research Institute. This index focuses on a micro perspective at the county level and constructs an evaluation index system from four aspects: the rural infrastructure metric, the rural economic digitalization metric, the rural governance digitalization metric, and the rural digitalization metric. However, the index data only span the time dimension of 2018–2020, and the comparable indicator data are limited. Taking into account the above considerations, this article first follows the guiding principles of policy documents such as the “Outline of Digital Rural Development Strategy” and the “Digital Agriculture and Rural Development Plan (2019–2025)”. Secondly, drawing on the indicator construction ideas of the “County Digital Countryside Index”, and further combining that with the existing literature, an evaluation system covering 15 indicators was constructed from five dimensions: digital environment foundation, industrial digitalization level, digital industrialization level, and digital service level. The above indicators not only have relatively complete measurement criteria but can also obtain long-term continuous data. The organized evaluation index system for the level of digital rural construction is shown in

Table 1. Evaluation index system for digital rural construction level.

Indicator	Level 2 Indicator	Level 3 Indicator	Specific Indicator
Digital Rural Construction	Fundamentals of digital environment	Internet penetration	Number of regional netizens/regional population
		Mobile Internet penetration	Number of mobile phones per 100 households owned by rural residents
		Fundamentals of digital industry	Information transmission computer services and software fixed assets investment
		Fundamentals of digital services	Fixed assets investment in transportation, warehousing, and postal services
		Construction of information infrastructure	Length of optical cable lines per square kilometer
	Industrial digitalization level	Digitalization of agricultural production	Number of environmental and agricultural meteorological observation stations
		Innovation in production technology	Number of valid invention patents/number of patent applications
		Electrification of agricultural production	Value added of agriculture, forestry, animal husbandry, and fishery/total electricity consumption in rural areas
		Industrial digitalization relies on	Number of Taobao villages
	The level of digital industrialization	enterprise digital transformation	Number of websites per hundred enterprises
		Enterprise digitalization activity	Proportion of enterprises participating in e-commerce transactions
		Digital retail market	Total amount of goods and services sold based on online orders
	Digital service level	Digital consumption level	Proportion of farmers’ transportation and communication expenses
		Fundamentals of digital economy	Digital Inclusive Finance County Investment Index
		Digital economy market	Digital Inclusive Finance County Mobile Payment Index

.

3.2.3. Intermediary Variable

This article explores the mechanism of digital rural construction on the transformation of farmers’ green production from two aspects: farmers’ digital literacy and farmers’ green production. When analyzing the impact of digital rural construction on farmers’ digital literacy, we draw on the research of scholars such as Du Fengjun (2023) [44] and select the number of household smartphones, computers with internet access, and information acquisition through network channels as mechanism variables.

The green production of farmers draws on the research results of scholars such as Li Xiaojing (2021) [45]. This article uses the form of the C-D production function to characterize the input–output relationship of farmers and logarithmically processes it to obtain the following formula:

$$Y_i=A{K}_i^{\alpha }{L}_i^{\beta }{e}^{\mu }$$

(5)

$$ln{Y}_i=lnA+\alpha ln{K}_i+\beta ln{L}_i+\mu$$

(6)

Among them, $Y_i$ is the output value per unit area of crops for farmer i, K is the capital investment per unit area of crops for farmer i, L is the labor input per unit area of crops for farmer i, A is the comprehensive technical level, and $\mu$ is the random error term. In the selection of covariates for the limited mixture model, this article combines the relevant situations of farmers’ production and operation in the database and refers to the practices of scholars such as Ye (2024) [46]. The application of soil testing formulas, commercial organic fertilizer, spraying efficient low-toxicity and low-residue pesticides, recycling agricultural film, recycling pesticide packaging, and returning straw to the field are selected as covariates and introduced into the model. Based on the relationship between the six indicators and farmers’ input–output, the probability of farmers falling into the category of green production mode is indirectly calculated, and the proxy variable for farmers’ green production transformation is obtained.

3.2.4. Control Variables

When conducting tests, the selection of control variables is crucial for the accuracy of the research results. To ensure the reliability and effectiveness of the research results, control variables are usually selected from multiple levels to more comprehensively reflect the influencing factors. The following is a detailed explanation of selecting control variables from three levels: geographical environment, economic development, and transportation conditions.

(1)

The geographical environment has a profound impact on regional development. The geographical environment includes multiple aspects such as terrain, climate, water resources, etc., all of which have a significant impact on land use efficiency. When selecting control variables, we can use mean representation to measure the impact of the geographical environment. Mean representation refers to the comprehensive consideration of multiple factors in the geographical environment and the calculation of their average value to reflect the overall condition of the geographical environment. This method can effectively reduce the bias caused by a single factor and more accurately reflect the impact of the geographical environment on regional development;

(2)

Choosing control variables at the economic development level is also of significant importance. The level of economic development directly affects the comprehensive strength and competitiveness of a region. When selecting control variables for economic development, they can be measured from three aspects: the human capital level, economic development level, and financial service level. The level of human capital reflects the quality of labor and education level in a region, which is crucial for the region’s innovation capability and development potential. The level of economic development can be measured by indicators such as GDP and per capita income, which can directly indicate the economic prowess of a region;

(3)

The selection of control variables at the level of traffic conditions cannot be ignored. Transportation is an important link connecting the inside and outside of a region, playing a crucial role in the economic development and personnel mobility of the area. The level of transportation convenience varies between different districts and counties, which directly affects the logistics costs and information flow efficiency of the region. Therefore, when selecting control variables at the level of traffic conditions, we can measure the length of the road per unit area. The density of highways can indicate the extent of transportation infrastructure development in an area and, by extension, the reach of the regional transportation network.

3.3. Data Sources and Descriptions

This article selected 255 prefecture-level and above cities in China from 2011 to 2023 as the original sample, including 4 central municipalities, 15 sub-provincial cities, 13 provincial capitals, and 223 prefecture-level cities. These cities represent different stages of socio-economic development and geographical features in China, providing a comprehensive sample for investigating the impact of digital rural construction on land use efficiency. The dependent variable data in this article come from the National Bureau of Statistics and China Land Network; the core explanatory variable data come from the Wind database, China Urban Statistical Yearbook, and China Urban Construction Statistical Yearbook; and other data come from local statistical bureaus. In order to facilitate the comparative analysis of the influence coefficients of the econometric models, the above data were logarithmically processed, and the resulting descriptive statistical analysis is shown in

Table 2. Descriptive statistics of variables.

Variable Type	Variable	N	Average	Standard	Min	Median	Max
Dependent variable	Land use efficiency	3315	0.588	0.478	0.114	1.543	0.862
Core explanatory variable	Level of digital rural construction	3315	0.414	0.317	0.092	0.121	0.611
Intermediary variable	Farmers’ digital literacy	3315	0.655	0.518	0.293	0.309	0.726
	Green production for farmers	3315	0.431	0.436	0.131	0.441	0.788
Control variables	Air quality level	3315	1.323	1.115	0.811	1.299	3.631
	Level of economic development	3315	0.720	0.512	0.201	0.442	0.712
	Development level of transportation	3315	0.694	0.667	0.291	0.458	0.865

.

After reviewing the policy documents and economic events of the Chinese government on digital rural construction from 2011 to 2023, we found that digital rural construction has made rapid progress and achieved significant results during this period, as shown in

Table 3. National policy for digital rural construction.

Time	Policy Document	Request Content
2011	Opinions on Implementing the Rural Revitalization Strategy	Implement the digital rural strategy and do a good job in overall planning and design
2012	Rural Revitalization Strategy Planning	Accelerate the comprehensive and deep integration of modern information technologies
2013	Several Opinions on Persisting in Optimizing the Development of Agriculture and Rural Areas and Doing a Good Job in the Work of Agriculture	Promote the construction of big data for the entire industry chain of important agricultural products and strengthen the construction of the national digital agriculture and rural system
2014	Outline of Digital Rural Development Strategy	Place digital rural areas in an important position in building a digital China
2015	Regarding the Key Work in the Field of Agriculture, Rural Areas, and Farmers	Carry out national digital rural pilot projects
2016	Suggestions on Formulating the 14th Five-Year Plan for National Economic and Social Development	We should prioritize the construction of digital rural areas, including rural e-commerce and digital governance
2017	Opinions on Fully Promoting Rural Revitalization and Accelerating Agricultural and Rural Modernization	Develop smart agriculture, establish a big data system for agriculture and rural areas, and promote the deep integration of new-generation information technology
2018	14th Five-Year Plan to Promote Agricultural and Rural Modernization	Establish and promote the application of agricultural and rural big data systems
2019	14th Five-Year Plan for the Development of Digital Economy	Coordinate and promote the construction of new smart cities and digital villages
2020	Opinions on Doing a Good Job in the Key Work of Comprehensively Promoting	Promote the development of smart agriculture and promote the integration and application of information
2021	Opinions on Doing a Good Job in the Key Work of Comprehensively Promoting Rural Revitalization in 2021	Rapid development of rural digital economy and rapid promotion of digital transformation in agriculture and rural areas
2022	Opinions on Doing a Good Job in the Key Work of Comprehensively Promoting Rural Revitalization in 2022	In the process of promoting the construction of digital rural areas, it is necessary to steadily and orderly advance
2023	Opinions on Doing a Good Job in the Key Work of Comprehensively Promoting Rural Revitalization in 2023	Reasonably set phased goals, tasks, and work priorities, and eliminate large-scale financing, demolition, and development

.

Specifically, the basic principle behind choosing this specific time frame for analysis is that this interval coincides with China’s new era. Since the 18th National Congress of the Communist Party of China, significant achievements have been made in the construction of digital villages under socialism with Chinese characteristics in the new era, mainly reflected in the following aspects:

Firstly, network infrastructure construction: The network infrastructure in rural areas has been significantly strengthened. According to the data, the 4G coverage rate of administrative villages across the country exceeds 98%, and the rural Internet penetration rate has increased significantly. By 2024, the number of rural broadband access users will exceed 200 million, and the Internet penetration rate in rural areas will increase by 2 percentage points.

Secondly, the promotion of the digital economy: The construction of the digital countryside has effectively promoted the development of the digital economy in rural areas. The online retail sales of agricultural products have exceeded 630 billion yuan, and the informatization rate of agricultural production has been further improved. The application of digital technology has gradually changed traditional agricultural production and improved agricultural production efficiency.

Thirdly, the digitalization of rural governance: The rural digital governance system is becoming increasingly perfect, and more government services are being handled at the grassroots level, which has improved the efficiency of rural governance. The application of digital technology in rural social governance is more efficient and precise.

Fourthly, the digitalization of public services: Digital technology promotes the extension and expansion of more high-quality public service resources such as education and healthcare to the grassroots level, improving the level of rural public services.

Fifth, the digital transformation of agriculture: The digital transformation of agricultural production and operation has significantly accelerated, and the construction of smart agriculture has achieved some initial results. The national agricultural production informatization rate has reached 27.6%, and the application of new-generation information technologies such as the Internet of Things, artificial intelligence, satellite remote sensing, and 5G is gradually changing traditional agricultural production.

Sixth, the development ecology of digital rural areas: The main body of digital rural construction continues to grow, the industrial ecology continues to prosper, and a digital rural development ecology that deeply integrates government, industry, academia, research, and applications has been formed.

4. Spatial Analysis of the Level of Digital Rural Construction and Land Use Efficiency

4.1. Spatial Analysis of the Level of Digital Rural Construction

In Figure 2, the spatial characteristics of the level of digital rural construction in China are mainly manifested in the following aspects: Firstly, the development of digital rural areas in Chinese counties presents a clear imbalance, overall showing a decreasing trend from east to west, that is, gradually decreasing from the eastern coastal areas to the central and western regions, but the difference between the north and south is not significant. High-level counties are mainly concentrated in the eastern coastal and central regions, while the overall development of digital rural areas in counties in the northeast and western regions lags behind. Secondly, among the four sub-dimensions of digital rural areas, the development level of rural digital infrastructure is relatively high, followed by the digitalization of rural governance, digitalization of rural economy, and digitalization of rural life. There are significant differences between the digitization of the rural economy and rural life. Thirdly, the overall level and sub-dimensions of digital rural development are significantly correlated spatially, with prominent HH (high high) and LL (low low) clustering characteristics. HH clusters are mainly distributed in the eastern and central regions, while LL clusters are mainly distributed in the western regions. Fourthly, the main battlefield of digital rural construction is in counties, and its application is mainly in towns and villages. The construction of the digital countryside is based on two major foundations, digital infrastructure and data resources, involving the construction of six major fields: smart agriculture, digital economy, digital governance, digital services, digital culture, and green countryside. Fifth, in the process of digital rural construction, new technologies such as 5G, the Internet of Things, big data, cloud computing, and artificial intelligence have been widely and deeply applied, promoting the improvement of agricultural production quality and efficiency, enhancing farmers’ living standards, and improving the rural living environment. Sixth, the mechanism for integrating agricultural and rural facility resources through multi-departmental collaboration is continuously improving, and the integration and sharing of rural information service stations are being promoted in an orderly manner. A preliminary rural information service system combining online and offline is being established. The scattered agricultural information systems that have been built are gradually being integrated, and various forms of agricultural data are accelerating their aggregation. Seventh, the construction of digital villages and the development of smart cities should coordinate with each other, to jointly promote inclusive digital public services, vigorously implement the national education digital strategic action, improve the national smart education platform, develop digital health, and standardize the development of Internet diagnosis and treatment and Internet hospitals.

4.2. Spatial Analysis of Land Use Efficiency

In Figure 3, the spatial characteristics of land use efficiency show significant regional differences and dynamic changes in China. Firstly, there are significant differences in land use efficiency between different regions in Chinese cities. The land use efficiency of cities in the eastern region is generally higher than that in the central and western regions, showing a spatial distribution characteristic of “high in the east and low in the central and western regions”. This difference is closely positively correlated with the level of economic development, and regions with higher levels of economic development often have higher land use efficiency. Secondly, from a time series perspective, the overall land use efficiency of Chinese cities shows an upward trend. The growth trends in different regions are similar, but the specific trajectory of changes varies in different regions. The urban land use efficiency in the eastern and western regions is showing an upward trend, while the central region is showing a downward trend. Thirdly, the spatial distribution of urban land use efficiency exhibits agglomeration characteristics, with high-efficiency areas and low-efficiency areas showing a “large concentration, small dispersion” pattern in space. This clustering feature varies in different years, but overall it presents a non-equilibrium trend. Fourthly, the overall land use efficiency of cities at or above the prefecture level in China is not high, and there is significant room for improvement. In terms of spatial distribution, there is a corresponding relationship between the efficiency value of each city and its level of economic development. The impact and constraint ability of pure technical efficiency on overall efficiency is stronger than that of scale efficiency. Fifth, the input of mobile factors (capital and labor) is conducive to improving the efficiency of urban land use, while the current utilization of non-mobile factors (land) is not conducive to improving efficiency. The higher the scarcity of arable land, the more conducive it is to improving the efficiency of urban land use, but there are regional differences. Sixth, in order to improve the efficiency of urban land use, it is necessary to establish more effective incentive and constraint mechanisms to change the current land use patterns. At the same time, external constraints such as farmland protection can be used to force the improvement in urban land use efficiency. Seventh, from 2011 to 2023, the gap in land use efficiency between the three major regional cities in China has gradually decreased, showing a trend of continuous convergence. The intra-group gap has become the dominant factor in the change in the urban land use efficiency gap, and the contribution of the inter-group gap has decreased from 52.37% to 26.19%.

4.3. Spatial Analysis of the Impact of Digital Rural Construction Level on Land Use Efficiency

In Figure 4, the spatial analysis of the impact of China’s digital rural construction level on land use efficiency involves multiple levels, including the improvement in agricultural production efficiency through digital transformation, the optimization of land resource allocation, and the reshaping of urban–rural spatial structure. The construction of digital rural areas has changed the management and operation of agricultural production by introducing digital technologies such as artificial intelligence, big data, blockchain, etc. These technologies can achieve the precise management of crops and improve land productivity, labor productivity, and resource utilization. For example, through smart agriculture systems, a precise control over crop sowing, irrigation, and harvesting can be achieved, thereby improving the efficiency of farmland utilization. The construction of digital rural areas has promoted the optimal allocation of land resources through digital means. The digital platform enables agricultural producers to more effectively obtain information about agricultural inputs such as seeds, pesticides, fertilizers, etc., improving the management radius and granularity of agricultural production. In addition, digital technology has also facilitated the flow of labor and capital from cities to rural areas, promoting the revitalization of rural land resources. The construction of digital rural areas has solved the problem of low agricultural production efficiency and optimized the agricultural production process through the synergistic effect of virtual space. This not only enhances the level of intensification, organization, and efficiency of agricultural production but also reconstructs the direction and speed of the flow of rural capital, land, and labor resources. For example, the e-commerce industry has shifted to rural areas with lower land prices, forming the phenomenon of “Taobao villages” and promoting the revitalization of rural land resources. The spatial distribution characteristics of digital rural development in China show significant regional differences, with an overall decreasing trend from the east to the central and then to the west, but there is not much difference between the north and south. This spatial clustering feature indicates that areas with high levels of digital rural construction often have a higher land use efficiency, which may be related to factors such as the level of economic development, degree of informatization, and scale of agricultural loans in these areas. The construction of digital rural areas promotes the increase in farmers’ income through digital infrastructure, as well as the digitization of the economy, governance, and daily life. Among them, the digitalization of the economy has the greatest effect on increasing farmers’ income. This indicates that the construction of digital rural areas not only improves land use efficiency but also directly promotes the income growth of farmers. Rural spatial reconstruction is an important driving force for land use transformation, and land use transformation is an important result of rural spatial reconstruction. There is a coupling and interactive relationship between the two. Optimizing the path of land use transformation and promoting the coupling development of rural spatial reconstruction and land use transformation are of great significance for achieving rural revitalization and urban–rural integration development.

5. Result Analysis

5.1. Benchmark Regression Results

Table 4. Benchmark regression results.

	Land Use Efficiency
	(1)	(2)	(3)	(4)
Digital rural construction	0.451 *** (0.015)	0.331 *** (0.005)	0.347 *** (0.013)	0.478 *** (0.023)
Farmers’ digital literacy		0.152 *** (0.028)	0.159 *** (0.027)	0.089 *** (0.017)
Green production for farmers			0.321 *** (0.003)	0.120 *** (0.013)
Air quality level				0.427 *** (0.001)
Level of economic development				0.254 *** (0.013)
Development level of transportation				0.031 *** (0.012)
Time fixed effect	YES	YES	YES	YES
Constant	1.566 *** (0.003)	1.626 *** (0.011)	2.727 *** (0.187)	2.727 *** (0.117)
N	3315	3315	3315	3315
R2	0.958	0.948	0.961	0.961

Note: The robust standard error is indicated in parentheses, and *** represents the 1% significance level. Due to space limitations, the results of controlling variables were not listed; same below.

reports the empirical test results of the impact of digital rural construction on land use efficiency. Column (1) only controls for individual and time fixed effects, while columns (2), (3), and (4) add control variables for geographic environment, economic development, and traffic conditions in sequence. Under different model settings, the coefficient of digital rural construction is significantly positive at the 5% level, indicating that digital rural construction can improve land use efficiency.

5.2. Robust Test

To further verify the robustness of the model, this article used three methods to test the robustness of the benchmark regression results: a replacement regression model, replacement of the dependent variable, and the exclusion of samples from municipalities directly under the central government. Firstly, in the replacement of the regression model, the simple OLS model was used in column (1) of

Table 5. Robust test results.

Variable	OLS Model	Replace the Explained Variable	Exclude Municipalities Directly Under the Central Government
	(1)	(2)	(3)
Digital rural development	0.443 *** (0.012)	0.421 *** (0.094)	0.319 *** (0.086)
Farmers’ digital literacy	0.323 *** (0.011)	0.423 *** (0.012)	0.123 *** (0.022)
Green production for farmers	0.323 *** (0.002)	0.523 *** (0.011)	0.423 *** (0.042)
Control variable	Yes	Yes	Yes
Constant	0.747 *** (0.124)	0.557 *** (0.107)	0.254 *** (0.024)
N	3315	3315	3263
$R^2$	0.824	0.836	0.948

Note: The robust standard error is indicated in parentheses, and *** represents the 1% significance level.

to replace the original linear baseline regression model for robustness testing. The regression results showed that digital rural construction had a positive impact on land use efficiency at a significance level of 5%, indicating that digital rural construction can effectively promote an improvement in land use efficiency. The model passed the robustness test. Secondly, in replacing the dependent variable, based on the approach of scholars such as Wang Wei’an (2019) [47], the production of public goods generated by unit urban land resource input was used to replace land use efficiency. The regression analysis revealed that digital rural construction still had a positive impact on land use efficiency at a significance level of 5%. Finally, municipalities directly under the central government have higher policy advantages and economic development levels, and their land use efficiency is easily influenced by their own social status and surrounding social evaluations. Therefore, four municipalities directly under the central government were excluded, and a benchmark regression was conducted again. The results showed that digital rural construction still had a positive impact on land use efficiency at a significance level of 5%, indicating that the key conclusions of this article are robust and reliable.

5.3. Endogeneity Test

Whether it is the construction of the digital countryside or land utilization rate, these are multidimensional and comprehensive systematic projects that not only require continuous improvements in resource allocation efficiency but also inevitably depend on the benchmark levels of resource endowment such as economy, technology, and talent innovation. Especially in the Internet era, industrial digitalization and digital industrialization both rely on the foreshadowing and embedding of digital infrastructure. The more successful the digital transformation is, the more likely land use efficiency will be accelerated. Therefore, there may be a reverse causal relationship between the level of digital rural construction and land use efficiency. Although this article introduces a series of key control variables that may affect the results and constructs a two-way fixed effects model to avoid biased estimates caused by omitted variables as much as possible, there are still some endogeneity issues. In view of this, this article further adopts two approaches to test endogeneity issues. Firstly, referring to Xu Xuchu’s (2023) solution approach [48], the number of fixed telephone users per 100 people at the provincial level in 2011 is used as the instrumental variable. On the one hand, in the historical period when the digital economy was not yet developed, the quantity of landline phones vividly depicted the image of local digital infrastructure and was the most core information tool for rural residents. It was a concrete representation of digital rural construction and maintained a significant correlation with explanatory variables. On the other hand, in the current mobile Internet era, fixed phones have been replaced by advanced digital tools such as smart phones, which will not affect the new agricultural productivity and meet exogenous conditions. Secondly, following the approach of Wang Ying and Guo Lei (2024) [49], the original data of digital rural construction level are used to generate first-order and second-order lagged terms, respectively, and we use them as instrumental variables for panel data instrumental variable regression.

Before conducting instrumental variable regression, a weak instrumental variable identification test needs to be performed on the instrumental variables. The results of the two-stage least squares method show that in the first stage of regression, the coefficient of influence between the instrumental variable of fixed telephone users per 100 people and the key determinant of rural digitalization is 0.282, and it is highly significant at the 1% statistical level. It can be considered that the instrumental variables have a substantial correlation with the endogenous factors in digital rural development. The F-test is 23.57, which is much higher than the discriminant threshold of 10, and the null hypothesis of weak instrumental variables can be rejected. In the second stage of regression, the coefficient of influence between digital rural construction and land use efficiency is 0.433 and significantly correlated at the 5% statistical level, as shown in the results of model (1) in

Table 6. Endogeneity test results.

Variable	(1) Land Use Efficiency	(2) Land Use Efficiency	(3) Land Use Efficiency
Fixed telephone numbers per 100 people in the 2011 base period	0.433 *** (0.207)
The first-order lag term of digital rural construction		0.354 *** (0.157)
Second-order lag term in digital rural construction			0.534 *** (0.168)
Intercept	−1.039 (1.181)	0.181 (0.398)	0.350 (0.409)
Time fixed effect	YES	YES	YES
R2	0.893	0.839	0.893
N	3315	3315	3315

Note: The robust standard error is indicated in parentheses, and *** represents the 1% significance level.

. Similarly, following the above approach, the instrumental variable method was used to test the first-order and second-order lagged terms of digital rural construction. In the initial phase of the regression analysis, both instrumental variables successfully rejected the null hypothesis of being weak instruments. In the second stage of the regression analysis, both the first and second lag terms of digital rural development exerted a significantly positive effect on land use efficiency at the 10% significance level. and the impact coefficients increased to varying degrees, as shown in the results of models (2) and (3). The results of the above three testing methods not only indicate that endogeneity issues interfere with the evaluation of the impact of digital rural construction on land use efficiency but also further validate the research conclusions mentioned earlier.

The effectiveness of instrumental variables mainly involves two conditions: correlation and exogeneity. Correlation means that instrumental variables need to be related to endogenous explanatory variables. This means that instrumental variables can explain changes in endogenous variables. In empirical research, we use the F-statistic in the first stage to test the correlation of instrumental variables, assuming that the hypothesis is valid and meets the conditions. Exogeneity means that instrumental variables cannot be correlated with error terms, meaning that instrumental variables do not directly affect the dependent variable except through endogenous explanatory variables. This is a key condition for the effectiveness of instrumental variables. In the absence of other instrumental variables for overidentification testing, the validation of exogeneity often relies on theoretical and logical arguments. The correlation of instrumental variables refers to the degree of correlation between instrumental variables and endogenous explanatory variables. If the instrumental variables are not correlated with endogenous variables, the endogeneity problem cannot be effectively solved. In empirical analysis, we use the F-statistic of the first-stage regression to test this correlation. If the hypothesis holds, we pass the test. When the number of instrumental variables exceeds the number of endogenous explanatory variables, overidentification occurs. An overidentification test is used to test the effectiveness of instrumental variables, that is, to determine whether the selected instrumental variables meet the conditions of exogeneity. If the instrumental variable does not meet the exogeneity condition, the estimation result is still biased. In the case of asymptotic effectiveness in large samples, we determine the effectiveness of the instrumental variable by comparing the difference between the estimated and true values of the instrumental variable. If the instrumental variable is valid, the difference between the estimated value and the true value should be small, and the value of the test statistic should also be small. It also has good properties in small samples. We construct a J-statistic to test the effectiveness of the instrumental variable. If the value of the J statistic is small and the corresponding p-value is greater than the given significance level, it indicates that the instrumental variable is effective.

5.4. Heterogeneity Test Results

The input land factors required for the production of different crops may vary, and the impact of digital rural construction on land use efficiency may also differ. According to the types of crops planted by farmers in the survey questionnaire, farmers were divided into two groups, grain crop farmers and economic crop farmers, and a regression analysis was conducted in each group. The estimated results are shown in columns (1) and (2) of Table 6. It can be found that the promotion effect of digital rural construction on the green production transformation of economic crop farmers is greater than that of grain crop farmers. This may be due to the high profit value of economic crops and the high demand for green organic vegetables, fruits, and other economic crop products in the consumer market, which can better utilize the natural attributes of land and indirectly improve land use efficiency. In addition, risk preference is an important factor affecting farmers’ agricultural production and management decisions, which may have an impact on land use efficiency. According to the risk preferences of household heads in the survey questionnaire, farmers are divided into risk preference and risk aversion groups. The estimated results are shown in columns (3) and (4) of

Table 7. Land transfer results in different regions.

Variable	(1)	(2)	(3)	(4)
	Green Production of Grain Crops by Farmers	Green Production of Economic Crops by Farmers	Risk-Preference Farmers	Risk-Averse Farmers
Digital rural construction	0.323 *** (0.014)	0.354 *** (0.027)	0.417 *** (0.074)	0.211 *** (0.027)
Farmers’ digital literacy	0.316 *** (0.024)	0.524 *** (0.112)	0.461 *** (0.115)	0.526 *** (0.004)
Green production for farmers	0.223 *** (0.015)	0.323 *** (0.024)	0.323 *** (0.022)	0.423 *** (0.024)
Control variable	Yes	Yes	Yes	Yes
Constant	0.897 *** (2.262)	1.351 *** (3.317)	1.149 *** (2.481)	1.641 *** (3.104)
${\mathrm{R}}^2$	0.134	0.147	0.118	0.134
N	3315	3315	3315	3315

Note: The robust standard error is indicated in parentheses, and *** represents the 1% significance level.

. It can be found that the promotion effect of digital rural construction on the land use efficiency of risk-preferring farmers is greater than that of risk-averse farmers. The possible reason is that risk-averse farmers usually hold a cautious attitude towards new technologies and methods, fearing that the application of new technologies may bring production and operation risks. Digital rural construction can help risk-averse farmers to gain a more holistic view of the advantages and optimization methods of land by providing detailed online information, case sharing, and technical guidance. With the help of intelligent agricultural equipment and data analysis tools, real-time data on soil, climate, crop growth, etc., can be provided to make further scientific production decisions.

Of course, the impact of digital rural development on land use efficiency also exhibits heterogeneity in different regions and farm sizes. Due to different resource endowments, different regions require different land resources, and the impact of digital rural construction on land use efficiency also varies. According to the actual situation, we divide the region into developed and underdeveloped areas and conduct a regression analysis by grouping them. The estimated results are shown in columns (1) and (2) of

Table 8. Regional and farm-scale heterogeneity.

Variable	(1)	(2)	(3)	(4)
	Developed Region	Less Developed Regions	Large-Scale Farm	Small-Scale Farm
Digital rural construction	0.433 *** (0.004)	0.574 *** (0.017)	0.314 *** (0.044)	0.221 *** (0.033)
Farmers’ digital literacy	0.345 *** (0.014)	0.536 *** (0.012)	0.251 *** (0.105)	0.446 *** (0.014)
Green production for farmers	0.113 *** (0.045)	0.225 *** (0.034)	0.323 *** (0.032)	0.423 *** (0.012)
Control variable	Yes	Yes	Yes	Yes
Constant	0.557 *** (1.232)	1.511 *** (1.337)	1.369 *** (2.431)	1.431 *** (3.104)
${\mathrm{R}}^2$	0.536	0.658	0.648	0.636
N	3315	3315	3315	3315

Note: The robust standard error is indicated in parentheses, and *** represents the 1% significance level.

. It can be found that the promotion effect of digital rural construction on underdeveloped areas is greater than that on developed areas. This may be due to the fact that agricultural products in underdeveloped areas are more easily carried by the digital economy, which increases the speed of economic growth. In addition, the size of the farm is an important factor affecting farmers’ agricultural production and management decisions, which may have an impact on land use efficiency. According to research, farm scale can be divided into two types: large-scale and small-scale. The estimated results are shown in columns (3) and (4) of Table 8. It can be found that the promotion effect of digital rural construction on the land use efficiency of large-scale farms is greater than that of small-scale farms, possibly due to the higher marginal efficiency of large-scale farms.

5.5. Intermediary Effect Test

In the theoretical analysis section of the previous section, this article deeply analyzed how the construction of digital rural areas can improve the impact mechanism of land use efficiency by enhancing farmers’ digital literacy and green production. In order to verify whether this theoretical mechanism is valid, further tests were conducted based on the mediation effect model. According to the mechanism verification steps, the first step is to verify the direct impact of digital rural construction on farmers’ digital literacy and green production. This has been fully discussed in the previous text and will not be repeated here. The second step is to examine the intrinsic relationship between digital rural construction and farmers’ digital literacy and green production. The results in column (1) of Table 7 indicate that at the 1% statistical level, digital rural construction can be considered to have a significant positive impact on farmers’ digital literacy. The results in column (2) show that after controlling for the impact of digital rural construction, farmers’ digital literacy still significantly promotes the construction of land use efficiency (H2), which has been verified. Similarly, the results in column (3) of

Table 9. Mediation effect test results.

Variable	(1) Farmers’ Digital Literacy	(2) Green Production for Farmers	(3) Farmers’ Digital Literacy	(4) Green Production for Farmers
Digital rural construction	0.445 *** (0.104)	0.368 *** (0.041)	0.283 *** (0.011)	0.366 *** (0.021)
Intercept	1.622 *** (1.552)	0.576 *** (0.467)	0.576 *** (0.467)	0.576 *** (0.467)
Time fixed effect	YES	YES	YES	YES
R2	0.872	0.876	0.986	0.897
N	3315	3315	3315	3315

Note: The robust standard error is indicated in parentheses, and *** represents the 1% significance level.

indicate that at the 1% statistical level, digital rural construction can be considered to have a significant positive impact on farmers’ green production. The results in column (4) show that after controlling for the impact of digital rural construction, farmers’ green production still significantly promotes land use efficiency construction (H3), which has been verified.

6. Conclusions and Discussion

6.1. Conclusions

This article is based on panel data from 255 prefecture-level and above cities in China from 2011 to 2023. It innovatively integrates digital rural construction, the digital literacy of farmers, green production of farmers, and land use efficiency into a unified framework. Based on the existing literature, it analyzes the theoretical mechanism of the impact of digital rural construction on land use efficiency and empirically tests the direct effect of digital rural construction on land use efficiency, as well as the indirect effect of digital rural construction on land use efficiency, through farmers’ digital literacy and green production. It also conducts robustness tests, endogeneity tests, heterogeneity tests, and mediation effect tests. Through the empirical results, the following was found:

(1)

The development of digital rural regions does indeed exert a direct and positive influence on land use efficiency. The inclusion of control variables and the results of robustness checks confirm the stability of this finding, strongly suggesting that enhancing the digitalization level in rural areas plays a crucial role. It can help promote the improvement in land use efficiency from three aspects: land input, capital input, and labor input. Firstly, digital rural construction helps farmers better understand land conditions and crop growth by providing accurate agricultural information. For example, through remote sensing technology, farmers can monitor the growth status of crops in real time, adjust irrigation and fertilization strategies in a timely manner, and thus improve land productivity. In addition, through big data analysis, farmers can predict market trends, arrange planting structures reasonably, reduce resource waste, and improve the economic benefits of land. Secondly, the construction of digital rural areas can also promote the optimization of capital investment. Through online platforms, farmers can more easily access financial services such as loans and insurance, reducing the risks of agricultural production. At the same time, the application of digital technology has lowered the threshold for agricultural production, attracted more social capital to invest in agricultural production, and improved the capital utilization efficiency of land. Finally, the impact of digital rural construction on labor input cannot be ignored. With the application of digital technology, the level of automation and intelligence in agricultural production continues to improve, reducing reliance on human labor. For example, the use of drones for pesticide spraying and intelligent agricultural machinery has greatly improved the efficiency of agricultural production. Meanwhile, the construction of digital rural areas also provides farmers with more non-agricultural employment opportunities, such as e-commerce and rural tourism, which helps optimize the allocation of labor resources and improve the efficiency of land utilization.

(2)

Digital rural areas can also indirectly have a positive impact on land use efficiency through farmers’ digital literacy and green production. This study found that the impact of the digital economy on land use efficiency is multidimensional and profound. Firstly, the digital economy enhances farmers’ digital literacy, enabling them to better utilize modern information technology to optimize agricultural production. The improvement in farmers’ digital literacy means that they can more effectively access market information, learn advanced agricultural technologies, manage agricultural production, and thus improve the efficiency of land output. Secondly, the digital economy has promoted the popularization of green production methods. Through digital means, farmers can apply fertilizers, irrigate, and prevent pests and diseases more accurately, reduce resource waste, and improve the sustainable utilization capacity of land. For example, through intelligent agricultural systems, farmers can optimize irrigation plans based on soil moisture and weather forecast data, reducing water waste. At the same time, digital technology can also help farmers monitor crop growth, take timely prevention and control measures, and reduce pesticide use, which not only protects the environment but also improves the ecological service function of land. In addition, the digital economy has also improved the efficiency of the entire agricultural industry chain by promoting its digital transformation. This includes online sales of agricultural products, the optimization of supply chain management, and the establishment of agricultural product traceability systems. These measures not only enhance the market competitiveness of agricultural products but also indirectly improve land use efficiency, as they reduce post-harvest losses and increase the value of agricultural products.

(3)

The impact of digital rural areas on land use efficiency is heterogeneous. Our research has found that the promotion effect of digital rural construction on the green production transformation of economic crop farmers is greater than that of grain crop farmers, and the promotion effect of digital rural construction on the land use efficiency of risk-averse farmers is greater than that of risk-preferring farmers. Firstly, compared to food crops, cash crops usually have a higher economic value and market diversity. Farmers of cash crops are more motivated to adopt new technologies to improve yield and quality, in order to obtain higher market returns. The construction of digital rural areas can help cash crop farmers better manage production and enhance the market competitiveness of their products by providing precision agricultural technology, market information, e-commerce platforms, and more. For grain crop farmers, although they also benefit from digital rural construction, due to the relatively stable market prices of grain crops and often being protected by government prices, these farmers may not feel as much urgency to adopt new technologies as economic crop farmers. Secondly, risk-averse farmers are more inclined to adopt technologies that can reduce uncertainty and risk. The information technology provided by digital rural construction can help these farmers better predict market changes, weather conditions, and the occurrence of pests and diseases, thereby reducing production risks. In contrast, risk-preferring farmers may be more willing to try new production methods, but such attempts do not always lead to an improvement in land use efficiency. They may be more focused on how to achieve higher returns through innovation, rather than simply reducing costs and risks by improving land use efficiency.

6.2. Discussion

This article proposes policy recommendations to advance the development of digital rural initiatives and improve land use efficiency: namely, to formulate an overall plan for digital rural construction, clarify the goals and paths for improving land use efficiency, and ensure the coordination between digital rural construction and land use planning. We should strengthen the construction of rural broadband networks, wireless Internet, and digital TV networks, improve rural network coverage, and provide basic support for the construction of digital villages. The following measures should also be undertaken: Accelerate the digital and intelligent transformation of rural water conservancy, highways, electricity, cold chain logistics, and agricultural production and processing infrastructure and improve land use efficiency, utilizing modern information technologies such as the Internet of Things, big data, cloud computing, etc., to enhance the intelligence level of agricultural production, optimize land resource allocation, and improve land productivity. Through e-commerce platforms, promote the export of agricultural products from villages to cities, reduce intermediate links, improve the efficiency of agricultural product circulation, and thereby enhance the economic value of land use. Promote the intelligence of agricultural equipment, enhance the level of agricultural technology, and improve land use efficiency through technological means. Using satellite remote sensing, drones, and other technologies to monitor rural ecosystems, protect the ecological environment of farmland soil and achieve the sustainable use of land resources. Promote socialist culture with Chinese characteristics through the Internet, enhance farmers’ digital literacy, and strengthen their acceptance and ability to use digital technology. Use Internet technology to promote new models such as “Internet plus + Party building” and “online village (residents) committees”, improve rural governance efficiency, and provide a good governance environment for enhancing the productivity of land use. Facilitate the integrated development of digital villages and smart cities, narrow the digital divide between urban and rural areas, promote the rational allocation of urban and rural factors, and improve the overall utilization efficiency of land resources. Select some regions to carry out pilot demonstration work for digital rural areas, explore beneficial experiences, and form replicable and promotable models. Strengthen the training of farmers’ information literacy, enhance their digital skills, strengthen their ability to apply digital technology, and provide a talent guarantee for the construction of the digital countryside. Develop and improve policy measures to support the construction of digital villages, including supporting policies in industry, finance, education, healthcare, and other fields. Establish a market-oriented mechanism to stimulate the enthusiasm of various market entities to participate in the construction of digital villages, and form an efficient and vibrant ecosystem for the development of digital villages. Improve the accuracy and efficiency of land element allocation, promote the formation of a spatial development pattern with effective constraints on main functions and coordinated and orderly land development, and enhance the ability of land elements to guarantee high-quality development in advantageous areas.

6.3. Research Limitations

This article takes 255 pieces of urban panel data as samples to study the mechanism of the impact of digital rural construction on land use efficiency. However, there are still three limitations: Firstly, the empirical research relies on sample data, and if the sample selection is inappropriate or unrepresentative, it may lead to bias in the research results. Secondly, the results of the empirical research may be difficult to generalize to other environments or groups, as the study was conducted under specific conditions. Thirdly, the high correlation between explanatory variables may lead to multicollinearity problems, affecting the stability and explanatory power of the model. Fourthly, there may be dynamic panel data issues such as sequence correlation and heteroscedasticity when processing time series data.

6.4. The Successful Experience of China’s Plan on the Impact of Digital Rural Construction on Land Use Efficiency

The development of the digital economy has a significant promoting effect on the efficiency of green land use in cities, and this promoting effect has an obvious regional and batch heterogeneity. The improvement in green land use efficiency is more significant in the central region and the second batch of pilot cities. The construction of digital rural areas optimizes the agricultural production process through the synergistic effect of virtual space, improving land output, labor productivity, and resource utilization. Digital transformation enables producers to obtain a large amount of comprehensive and accurate information through digital platforms, thereby improving the level of intensification, organization, and efficiency of agricultural production. The construction of digital rural areas can effectively improve the efficiency of the green transformation of regional farmland utilization, and this promoting effect is more significant in the eastern region and areas with a high efficiency of green transformation of farmland utilization. The construction of digital rural areas can also generate positive spatial spillover effects to a large extent. Considering the different geographical conditions, there may be regional heterogeneity in the impact of the level of digital rural construction on the efficiency of the green transformation of cultivated land use. The estimated coefficient in the eastern region is higher than that in the central and western regions, because the eastern region has a higher economic level, which is conducive to the construction of digital villages and can enhance regional advantages. The level of digital rural construction has not only improved the efficiency of the green transformation of farmland utilization in the province but also enhanced the efficiency of green transformation of farmland utilization in neighboring provinces, demonstrating significant spatial effects, accelerating the process of digital rural construction, promoting the comprehensive and in-depth integration and application of new-generation digital information technology and agricultural production and operations and precision agricultural production modes, and providing an important impetus for the transformation of green farmland utilization. At the same time, we should pay attention to the differences in the process of digital rural construction, promote the construction of digital rural areas in developed eastern regions, and increase support for the construction of digital rural areas in the underdeveloped central and western regions, in order to narrow the “digital divide” between different regions.