Digital Economy, Factor Allocation, and Sustainable Agricultural Development: The Perspective of Labor and Capital Misallocation

Abstract

As an emerging economy, the combination of the digital economy and industrial development can lead to a variety of new industries and new formats and form new momentum. This paper aims to analyze how the digital economy affects sustainable agricultural development, through what path, and what role the factor allocation plays in this process. Based on the analysis of the impact mechanism of the digital economy on sustainable agricultural development, this paper takes 30 provinces in China from 2013 to 2020 as the research object, builds a panel data model and mediation effect model, empirical analyses on the impact of the digital economy on sustainable agricultural development, and verifies the mediation effect of factor allocation in this process. This study found that the digital economy significantly promoted sustainable agricultural development. At the national level, this obvious promotion effect was achieved by reducing the misallocation of labor and capital. At the regional level, sustainable agricultural development in the eastern region of China depended on the optimization effect of the digital economy on the allocation of labor and capital, but the situation was slightly different in the central and western regions of China. The digital economy in the central and western regions of China significantly improved sustainable agricultural development by allocation effect of labor factors. Although the allocation effect of capital factors had a positive effect on sustainable agricultural development, it was not significant. In order to promote sustainable agricultural development, the digital industry development should be accelerated, the allocation effect of the factor market should be brought into play, and the digital economy should be better integrated with sustainable agricultural development so as to realize the modernization of agricultural development.

1. Introduction

With the progress of digital technology, the promotion of rural economic development has gradually become an important factor. The development of industrial digitalization and digital industrialization has further promoted the formation of data as a key production factor. As the digital economy is leading the transformation of the rural agricultural production mode, accelerating the structural optimization and innovation of traditional agriculture, it is playing an important role in improving the total factor productivity of agriculture and realizing the deep integration of real industries. In recent years, emerging technologies, such as big data and artificial intelligence, have become important breakthroughs in upgrading the economic structure and accelerating traditional industries’ transformation. The development of the digital economy has formed a variety of new industries and new formats. The expansion of the scale effect has promoted industrial integration, and the platform effect has fostered the deep integration of industries. Due to the combination of digital technology and industrial integration, the industry has developed rapidly and formed “new industry integration”. As an emerging digital industry, digital technology is rapidly breaking through and being widely applied, swiftly penetrating into different industries. The digital economy has formed a scale, and the scale effect can be brought into play. It has become an important factor in promoting new economic development and is conducive to promoting the formation and development of new industries and new formats [1]. The digital economy has become an important driving force for economic growth and scale expansion, rapidly transformed the old driving force into a new driving force, caused changes in the connotation, space, and field of industrial development, and played a key role in upgrading low-end industries to mid-to-high-end industries, thereby enhancing industrial competitiveness [2]. The improvement and promotion of the digital infrastructure and technology level have increased, promoted the flow of information factors and the allocation efficiency of factor resources, and laid a solid foundation for industrial development [3]. With digital technology as the starting point, industries’ deep integration as the development direction, and the digital service platform as the support, a digital economic model and pattern integrating network, service, and precision will be formed [4]. With the continuous improvement of the network infrastructure, digital technology has accelerated its diffusion to the agricultural field, which has promoted farmers’ entrepreneurial probability, increased their income, optimized the agricultural production system, and improved agricultural quality, efficiency, and competitiveness [5,6]. It can be concluded that the rapid development of the digital economy accelerated the digital transformation, drove the economic society, profoundly changed the production, life, thinking, and governance mode of human beings, promoted the agricultural transformation and rural industries integration, and affected agriculture’s sustainable development [7]. How the current digital economy development affects agriculture’s sustainable development and through what path, as well as what role the factor allocation plays in this process, are still issues worthy of attention.

Agricultural development needs to solve the problems of agricultural production efficiency and agricultural development power, improve the cultivation and operation efficiency of business entities, and ensure efficient allocation of resource elements and agricultural product supply efficiency. The digital economy plays an important role in this process, but there are still problems, such as the lag of digital agricultural transformation, insufficient technological innovation ability, and lack of talent [8]. With the deepening of the digital economy’s development, the agricultural industry is seeking balanced development between industries and regions. However, the flow and convergence of high-quality resources from rural areas to cities leads to the loss of a large number of high-quality resources in rural areas and causes rural hollowing out. At the same time, the problem of farmers’ aging is becoming increasingly serious and leading to the decline of rural areas [9]. With the implementation of the rural revitalization strategy, preferential policies were put forward in terms of talent, land use, and capital, which led to the return of some factors. Through the allocation of talents, technology, and information, the interaction and complementarity between industries and regions will promote the in-depth development of industrial integration. The digital economy plays an important role in this process. However, at the present stage, there is a big gap between provinces in agricultural figures’ development, showing that one province takes the lead and starts rapidly. The digital economy is important in increasing farmers’ income channels, transforming the development mode of agricultural production, and promoting the modernization of rural areas [10]. Lin and Mao (2022), based on provincial panel data, believed that the digital economy supports the improvement of agricultural total factor productivity, but the performance differs between regions, especially in the eastern and central regions [11]. Jiang (2022) thought that the digital economy promoted the agricultural transformation and agricultural industry integration, thus fostering agricultural development, but there were some problems in this process, such as the digital divide, infrastructure, talent support, and so on [7]. Li and Zhou (2022) assert that the digital economy promotes the change in agricultural technology conditions and the effective integration of digital technology, digital technology supports the digital transformation of agricultural mechanization [12,13], and the digital economy promotes high-quality development of agriculture through different paths [14,15].

Based on the above literature, the research objectives of this paper are twofold. First, it is necessary and important to develop the digital economy, but little attention has been paid to how it affects the sustainable development of agriculture. With regard to the digital economy and the sustainable development of agriculture, research has mainly focused on the micro-subject of the combination of the digital economy and actual production [16]. Most of the relevant studies have not completely solved the problems of agriculture and rural areas. The only relevant studies have focused on the integration of the digital economy and a certain link with agricultural development and have not obtained an effective empirical research basis. There is still a lack of research on the production, circulation, and distribution links between the sustainable development of agriculture and the digital economy. This paper theoretically confirms their relevance to each other, brings the digital economy and sustainable agricultural development into the same framework and introduces emerging digital technology into sustainable agricultural development. Second, the digital economy is an important factor in production. How can it promote the sustainable development of agriculture? This needs to be verified empirically. Based on the dynamic perspective, this paper reveals the correlation mechanism between the two, emphasizes the direct effect of the digital economy on the sustainable development of agriculture, and explores how the digital economy affects the allocation of factors and, thus, shows the indirect effect of the sustainable development of agriculture from the perspective of the mismatch of labor and capital factors.

To achieve the expected goal, this paper will expand on and supplement the following aspects. First, the measurement of the digital economy has not yet formed a unified standard, and it is difficult to give a standard definition of the digital economy in theory, which leads to confusion in the definition and an unclear division, increasing the complexity involved in the measurement of the digital economy [17,18,19]. In this paper, a digital economy indicator system is constructed from the multi-dimensional aspects of digital construction, digital application, and digital industry development, which to some extent, mitigates the difficulty in measuring the digital economy. Agricultural development needs to consider multiple links of production, circulation, and distribution. Sustainable agricultural development cannot be measured with a single indicator. Reconstruction of the indicator system of sustainable agricultural development from the perspective of agricultural production, the circulation of agricultural products, and farmers’ income can reflect the process of agricultural development and promote sustainable agricultural development. The quantification of the digital economy and sustainable agricultural development uses the entropy method to synthesize them and calculate a comprehensive index. Second, research on the development of the digital economy in agriculture has been relatively limited, mainly due to the complexity of agricultural development and strong resource endowment [15,20]. The development and application of the digital economy in agriculture are relatively backward. Relevant research has focused on the digital economy and industrial structure, industrial chain, industrial integration, and so on [2,4,7,21]. Few documents have taken the digital economy as an element from the perspective of its impact on the allocation of elements and then on agricultural development. From the perspective of factor allocation, this paper describes the direct and indirect impacts of the digital economy on the sustainable development of agriculture, provides a theoretical basis and practical evidence showing how the digital economy enables the sustainable development of agriculture, and explores the feasible path.

The structure of the article is as follows: (1) theoretical construction and research hypotheses; (2) research methods and data; (3) empirical analysis; and (4) research conclusions and policy implications.

2. Theoretical Analysis and Research Hypotheses

2.1. Digital Economy and Sustainable Agricultural Development

Agricultural development needs to be based on the advantages of local agricultural resources and form related secondary and tertiary industries. With the promotion of agricultural industrialization, agricultural products need to enter the big market, and the exporting of agricultural products has become an inevitable trend. The exporting of agricultural products cannot be separated from digital technology in creating an e-commerce platform, realizing the wide circulation of agricultural products, which will promote the rapid development of logistics, finance, and other industries, and driving the transformation and upgrading the traditional agriculture.

In the era of the digital economy, digital production, digital circulation, and digital consumption have become important trends, promoting the high-quality development of industries, cities, and regions [22,23] and exerting an important impact on sustainable agricultural development [21,24]. The digital economy optimizes agricultural technical conditions, improves agricultural production efficiency, gives consideration to standardized agricultural production, drives the upgrading of primary processing and deep processing of agricultural products, optimizes the circulation mode of traditional products, innovates the digital auction and electronic settlement mode of agricultural products, and promotes agricultural products’ consumption [25]. Digital technologies, represented by big data, the Internet, and cloud computing, are integrated into agricultural development, breaking through the constraints of the small-scale decentralized operation of traditional agriculture with low efficiency, high cost, and high risk and promoting an efficient cross-border flow of industrial factors and industrial integration, scale, and integration development through model optimization and system reform to realize agricultural scale economy [26,27,28].

The essence of the digital economy is digital knowledge and information. With the improvement of digital technology, it will increase in value. Digital technology promotes high-quality agricultural development and forms economies of scale, scope, aggregation, and labor division [8]. An agricultural platform based on big data creates a new mode of information sharing and improves the profitability of farmers [29,30]. The digital economy leads the value reconstruction of the agricultural and industrial chain, optimizes the layout of the industrial chain, and accelerates the cross-integration between agriculture and other industries [31,32]. From the perspective of digital application, the digital economy has become a key factor of production, which can promote the development of advanced productive forces, accelerate the accumulation of social wealth, and reconstruct the pattern of economic development. The use of digital technologies, such as big data, the Internet of Things, and block refining, can significantly improve the level of intelligence, informatization, and automation of agricultural production, promote the transformation and upgrading of agricultural production and operation activities, activate the flow of agricultural and rural factors, effectively enhance the industrialization, socialization, and intensification of agricultural management, and support the vertical and horizontal extension of agricultural production [8].

Hypothesis H1 is thus proposed:

 Hypothesis 1 (H1): 

The digital economy has a positive impact on sustainable agricultural development.

2.2. The Intermediary Effect of Factor Allocation

In many developing countries, one-sided emphasis on the openness and versatility of digital technology may cause people to pay too much attention to the upgrading of technology itself and to neglect the comparative agricultural advantage based on factor endowment conditions [33]. With the innovation of the agricultural development mode and the wide application of agricultural mechanization, the labor force engaged in agriculture is gradually reduced, and the allocation of production factors is increasingly unbalanced. Digital technology supports the efficiency of the circulation and sharing of resource elements, especially the integration of territory-dependent resources and the efficiency of heterogeneous resource allocation, which are conducive to improving the overall operational efficiency of agriculture [34]. The application of the digital economy to sustainable agricultural development, to a certain extent, needs to rely on the existence of labor factors, capital factors, and so on; that is, the digital economy needs to penetrate into the traditional production factors to drive the optimization of traditional factors’ configuration structure, improve the configuration efficiency, and drive the return of the labor force and capital. After the digital economy is put into the real economy, the proportion of factor inputs and the structure of factor allocation will change, and the substitution, penetration, and integration of factors will increase the demand for high-end labor and improve the efficiency of traditional factor allocation. The digital economy can, to some extent, correct the problem of the unbalanced allocation of production factors. With the wide application of the digital economy, factors have been digitized, and this plays a key role in finding an effective solution to information asymmetry and in promoting the optimization of element allocation. The high permeability and dependence of the digital economy make the coordination and matching of the elements more accurate and effective, which is conducive to the optimization of the allocation between elements and promotes the transformation and upgrading of the industrial structure [35].

The effective carrier of agricultural development is inseparable from factor input and technical efficiency optimization. The digital economy includes the improvement of digital technology, the improvement of information technology, and the use of big data, which are all high-level production input factors. Through the measurement of technical efficiency from the perspective of input, it can be concluded that, with other conditions unchanged, due to the improvement of technical efficiency, a producer needs to reduce the input of factors to maintain the original output. Low-efficiency producers need to tap the internal potential to improve technical efficiency and overcome its constraints. The structure optimization and efficiency improvement of the configuration among factors promote sustainable agricultural development. The strong energy intensity of the digital economy aids the transformation of factors from low efficiency to high efficiency, realizes the upgrading of factors, supports the generation of structural upgrading, and then stimulates sustainable agricultural development.

Based on the above two aspects, it can be found that factor allocation may play an intermediary role in the process of the digital economy’s promotion of sustainable agricultural development. The existence of the digital economy reduces the imbalance of traditional factor allocation in a region, and the optimization of factor allocation promotes sustainable agricultural development. Therefore, this paper puts forward hypothesis H2 and hypothesis H3 on the relationship between the digital economy, factor allocation, and sustainable agricultural development:

 Hypothesis 2 (H2): 

The digital economy improves the level of sustainable agricultural development by reducing the misallocation index of regional labor allocation.

 Hypothesis 3 (H3): 

The digital economy reduces the misallocation index of regional capital allocation and then improves the level of sustainable agricultural development.

3. Model Construction and Variable Interpretation

3.1. Model Construction

Sustainable development cannot be separated from the promotion of the digital economy [36,37]. In addition to the impact of the digital economy, rural infrastructure, financial support for agriculture, rural finance, and rural human capital are all important factors affecting sustainable agricultural development. Therefore, the model construction considers not only the digital economy but also other factors as control variables. This paper constructs a benchmark model, as shown in Formula (1), to verify that the digital economy enables sustainable agricultural development [14,15].

$$SA{D}_{it}={\alpha }_0+{\alpha }_1{D}_{it}+{\lambda }_j{\displaystyle \sum _{j=1}^4{X}_{it}}+{\delta }_i+{\eta }_t+{\epsilon }_{i,t}$$

(1)

In Formula (1), $SA{D}_{it}$ is the level of sustainable agricultural development for the province $i$ in year $t$, $D_{it}$ is the digital economy for province $i$ in year $t$, is control variable, ${\delta }_i$ is the time-fixed effect, ${\eta }_t$ is the regional-fixed effect, and ${\epsilon }_{it}$ is the random error term. ${\alpha }_0$ is a constant, ${\alpha }_1$ is an impact coefficient of the digital economy on sustainable agricultural development, and is the impact coefficient of control variables on sustainable agricultural development.

To investigate and better analyze how the digital economy enables sustainable agricultural development, that is to say, to determine the path that the digital economy can take to enable sustainable agricultural development, this paper introduces the intermediary effect model and takes factor allocation as the intermediary variable. The model is set as follows (2)–(4):

$$SA{D}_{it}={\alpha }_0+{\alpha }_1{D}_{it}+{\lambda }_j{\displaystyle \sum _{j=1}^4{X}_{it}}+{\delta }_i+{\eta }_t+{\epsilon }_{i,t}$$

(2)

$$F{A}_{it}={\beta }_0+{\beta }_1{D}_{it}+{\pi }_j{\displaystyle \sum _{j=1}^4{X}_{it}}+{\delta }_i+{\eta }_t+{\epsilon }_{i,t}$$

(3)

$$SA{D}_{it}={\gamma }_0+{\gamma }_1{D}_{it}+{\gamma }_{2}F{A}_{it}+{\rho }_j{\displaystyle \sum _{j=1}^4{X}_{it}}+{\delta }_i+{\eta }_t+{\epsilon }_{i,t}$$

(4)

Here, $F{A}_{it}$ is the intermediary variable, which is the factor allocation level in this paper, and the factor allocation level adopts the labor factor misallocation index and capital factor misallocation index, ${\beta }_0$ is a constant, ${\beta }_1$ is an impact coefficient of the digital economy on factor allocation, and ${\pi }_j$ are impact coefficients of the control variables on factor allocation. ${\gamma }_0$ is a constant, and ${\gamma }_1$,${\gamma }_2$ are the impact coefficients of the digital economy and factor allocation. Other variables are consistent with the benchmark model (1).

For the calculation of the size of the intermediary effect in this paper, according to the intermediary effect model (2)–(4), the methods are $\frac{{\beta }_1\times {\gamma }_2}{{\alpha }_1}$ and $\frac{{\beta }_1\times {\gamma }_2}{{\gamma }_1}$.

3.2. Variable Interpretation and Description

3.2.1. Interpreted Variable and Core Explanatory Variable

The explained variable in this paper is sustainable agricultural development, and the measurement of sustainable agricultural development is mainly carried out with the production, circulation, and distribution links. The production link includes three indicators; specifically, modern production is measured using the total power of agricultural machinery, large-scale production is measured with the rural electricity consumption, and the industrial chain extension is measured using the output value of agricultural, forestry, animal husbandry, and sideline fishery production. The circulation link is measured with the total retail sales of agricultural products, and the distribution link is measured with the per capita disposable income of rural residents.

The core explanatory variable is the digital economy, which is multidimensional [38,39]. The measurement of the digital economy is performed using construction, application, and development links, mainly from the aspects of digital infrastructure construction, digital application, and digital industry development. Digital infrastructure construction consists of eight indicators, which are, respectively, the capacity of mobile phone exchanges, the number of IPV4 addresses, the penetration rate of mobile phones, the capacity of office exchanges, the number of computers used at the end of the period, the rural delivery routes, the number of computers used per 100 people, and the length of the optical cable lines. Ten indicators are used for digital applications, respectively, the number of domain names, the number of web pages, mobile internet users, internet broadband access ports, rural broadband access users, mobile Internet access traffic, the number of informatization enterprises, the number of websites owned by every 100 enterprises, the proportion of enterprises with e-commerce transaction activities, and the number of enterprises with e-commerce transaction activities. The development of the digital industry consists of nine indicators, which are, respectively, software business revenue, total telecom business revenue, information technology service revenue, software product revenue, embedded system software revenue, information security revenue, software business exports, e-commerce purchases, and e-commerce sales forehead.

The measurement method of the digital economy and sustainable agricultural development mainly uses the entropy method, which is divided into four steps (

Table 1. Indicator System of Digital Economy and Sustainable Agricultural Development.

	Index	Definition	Measurement Method
Sustainable agricultural development	Production link	Modern production	Total power of agricultural machinery (10,000 kW)
		Large-scale production	Rural electricity consumption (100 million kWh)
		Industrial chain extension	Output value of agriculture, forestry, animal husbandry, sideline fishery (100 million yuan)
	Circulation link	Circulation and retail of agricultural products	Total retail sales of agricultural products (100 million yuan)
	Distribution link	Farmer’s household income	Per capita disposable income of rural residents (yuan)
Digital economy	Construction link	Digital infrastructure construction	Capacity of office switch (10,000 gates)
			Mobile telephone exchange capacity (10,000 households)
			Rural delivery route (km)
			Length of optical cable line (km)
			Mobile phone penetration rate (unit/100 people)
			Number of IPV4 addresses (10,000pieces)
			Number of computers used at the end of the period (set)
			Number of computers used per 100 people (set)
	Application link	Digital application	Number of domain names (10,000 piece)
			Number of pages (10,000 piece))
			Number of web pages (10,000 pieces)
			Mobile internet users (10,000 households)
			Mobile internet access traffic (10,000 GB)
			Rural broadband access users (10,000 households)
			Number of informatization enterprises (piece)
			Number of websites owned by each hundred enterprises (piece)
			Number of enterprises with e-commerce transactions (piece)
			Proportion of enterprises with e-commerce transactions (%)
	Development link	Digital industry development	Total telecom services (100 million yuan)
			Software business income (10,000 yuan)
			Revenue from software products (10,000 yuan)
			Income from information technology services (10,000 yuan)
			Information security income (10,000 yuan)
			Embedded system software revenue (10,000 yuan)
			Exports of software business (USD 10 thousand)
			E-commerce sales (100 million yuan)
			E-commerce procurement amount (100 million yuan)

). The first step is to build an evaluation matrix. Since the data obtained in this paper are different indicator data of different regions from 2013 to 2020, it is necessary for this paper to build evaluation systems for different regions when constructing evaluation matrices, which are expressed in rows and columns. The second step is to standardize the matrix data. Due to the different quantization methods between indexes, it is necessary to standardize the data. Standardization will solve the problem of the usage of different indicator units. This paper adopts the maximum–minimum method. The third step is to calculate the weight of relevant indicators. Different indicators have different impacts on the digital economy and sustainable agricultural development; therefore, it is necessary to assign different weights. This paper uses the information entropy weighting ($LEW$) method to calculate the weights as it can overcome the defects of subjective judgment. The fourth step is to calculate the comprehensive order parameters of relevant indicators. Through the construction and calculation performed in the first three steps, the digital economy and sustainable agricultural development are integrated into one indicator to judge the level of the digital economy and sustainable agricultural development.

3.2.2. Intermediary Variable

The intermediary variable is factor allocation which is measured using the factor misallocation index, including the labor factor misallocation index and the capital factor misallocation index. The measurement of the mediating variable is based on the relative distortion coefficient of factors ${\widehat{\gamma }}_{Li}$ and ${\widehat{\gamma }}_{Ki}$, calculating the labor and capital misallocation indices, which are, respectively, represented by ${\tau }_L$ and ${\tau }_K$ [40]. However, there are two possible situations of factor allocation: insufficient allocation and excessive allocation. The factor allocation index of excessive allocation is negative. The factor allocation index is made non-negative by taking the absolute value, referring to the research of Ji et al. (2016) and Ma et al. (2020) [41,42]. The specific formulas are the following (5) and (6):

$${\widehat{\gamma }}_{Li}=\frac{1}{1+{\tau }_{Li}}$$

(5)

$${\widehat{\gamma }}_{Ki}=\frac{1}{1+{\tau }_{Ki}}$$

(6)

Here, ${\tau }_{Li}$ and ${\tau }_{Ki}$ are, respectively, the labor and capital misallocation indices in province $i$, and ${\widehat{\gamma }}_{Li}$ and ${\widehat{\gamma }}_{Ki}$ are, respectively, the labor and capital distortion coefficients in province $i$.

The measurement of factor allocation is carried out using the factor misallocation index. If the actual supply and effective demand of factors are not balanced for certain areas, it is considered that there is an imbalance or misallocation between factors, which is actually the difference between the number of input factors and the required number of corresponding output values. The larger the difference, the more unbalanced the factors and the more serious the misallocation. Only two factors, labor and capital, are considered in the issue of factor allocation in this paper, and the distortion coefficients of labor and capital are in accordance with specific calculation formulas, as follows (7) and (8):

$${\widehat{\gamma }}_{Li}=\raisebox{1ex}{$\left(\frac{L_i}{L}\right)$}\!\left/ \!\raisebox{-1ex}{$\left(\frac{s_i{\beta }_{Li}}{{\beta }_L}\right)$}\right.$$

(7)

$${\widehat{\gamma }}_{Ki}=\raisebox{1ex}{$\left(\frac{K_i}{K}\right)$}\!\left/ \!\raisebox{-1ex}{$\left(\frac{s_i{\beta }_{Ki}}{{\beta }_K}\right)$}\right.$$

(8)

Here, $L_i$ and $L$, respectively, represent the agricultural labor force in province $i$ and the whole country, $K_i$ and $K$, respectively, represent the agricultural capital in province $i$ and the whole country, and $s_i$ represents the proportion of the agricultural output of the province in the agricultural output of the whole country. ${\beta }_{Li}$ and ${\beta }_{Ki}$ are, respectively, the contribution rates of $i$ province’s labor and capital to output, and ${\beta }_L$ and ${\beta }_K$ are, respectively, the contribution rates of national labor and capital to output. As the leading force of sustainable agricultural development, the agricultural labor force plays an important role in agricultural production, processing, and sales links. The return of migrant workers and college students is an important factor in rural industries’ development. Therefore, the measurement of the labor force is measured using rural employees. As a necessary condition for sustainable agricultural development, it promotes agricultural industries development. However, agricultural enterprises have relatively high operating risks, large investments, and a long return period, which many business entities are unwilling to bear or find it difficult to bear, and which require the intervention of financial institutions and other intermediaries to develop and strengthen the three rural industry business entities. However, farmers are still widely involved in the production of agricultural products, and decentralized farmers’ operations are more common. Therefore, agricultural capital is measured using the completed investment in fixed assets of rural households.

The meaning of Formulas (7) and (8) is the deviation of the actual input of regional labor and capital factors from the effective allocation, that is, the factor distortion coefficient. If the factor distortion coefficient is greater than 1, it means that the input of the factor is too large and the allocation of the factor is excessive; otherwise, it means that the input is too small and the allocation of the factor is insufficient.

The estimation of ${\beta }_{Li}$ and ${\beta }_{Ki}$ refers to the research of Bai et al. (2018) and Wang and Zhang (2021) [43,44], and it is undertaken by establishing a production function with period and cross-section effects, as shown in Formula (9):

$$\ln \left(Y_{it}/L_{it}\right)=\ln A+{\beta }_{Ki}\ln \left(K_{it}/L_{it}\right)+{\mu }_i+{\upsilon }_t+{\epsilon }_{it}$$

(9)

$Y_{it}$ is the agricultural output of the province $i$, and $A$ is the parameter for which the production function is assumed to be constant returns to scale, that is, ${\beta }_{Li}=1-{\beta }_{Ki}$, ${\beta }_K={\displaystyle \sum _{i=1}^n{s}_i{\beta }_{Ki}}$, ${\beta }_L={\displaystyle \sum _{i=1}^n{s}_i{\beta }_{Li}}$.

3.2.3. Control Variables

The digital economy has an important impact on the sustainable development of agriculture, and the allocation of factors plays an essential role. In addition, the sustainable development of agriculture is affected by other factors, such as the relative backwardness of the agricultural system, the prominent rigid constraints on resources, the lack of technological innovation ability, and the slow development of industrial clusters, all of which will affect the sustainable development of agriculture [45]. Especially in the context of green development, farmland water conservancy, facilitated by construction, environmental protection, and green agriculture and supported by agricultural ecological compensation, has an important impact on the sustainable development of agriculture, especially water conservancy infrastructure construction [46]. To improve the reliability of the research, this article selects a series of control variables to control other factors affecting sustainable agricultural development.

Agricultural resources are an important part of agricultural productivity. With the modernization of agriculture, the quantity and quality of agricultural resources have shown a downward trend. The contradiction between humans and land and the serious resource load has become increasingly prominent, seriously restricting the sustainable development of agriculture. In the face of these contradictions, under the condition of ensuring national food security to achieve sustainable agricultural development, it is imperative to improve the agricultural infrastructure. The agricultural infrastructure has an important impact on the efficiency of grain production, improves the technical level in the process of grain production, expands the production scale, thereby reducing the energy consumption per unit of production resources, improves the level of mechanization and the efficiency of grain production [47,48], and promotes the sustainable development of agriculture. Infrastructure is a necessary condition for the development of agriculture and an important factor affecting its sustainable development. Rural infrastructure is measured by the level of rural roads, and the development of rural roads can improve the transportation level of rural areas and speed up the sales channels of agricultural products.

As the key period of the transformation from traditional agriculture to modern agriculture, the intensity of financial support for agriculture can reflect the support of the state for agriculture to a certain extent, reflect the impact of industrial policies on agricultural development, and play an important role in the sustainable development of agriculture. Effective fiscal support for agriculture policy can better adapt to the development of modern agriculture, change the mode of thinking, and grasp the use direction of special funds [49]. However, in the agricultural investment and financing system, the growth of agriculture-related capital investment is not strong. Although there are many investment funds, there are problems of scattered and repeated use, and there is still a lack of transformation in the direction of sustainable agricultural development. In this paper, the financial support for agriculture is replaced by the intensity of the financial support for agriculture in each province. Financial intermediaries, by optimizing the allocation of capital, improve economic efficiency and promote economic growth, as does rural financial development. The development of the agricultural economy cannot be separated from the support of financial factors. The development of rural financial services can reduce the use cost of rural financial markets, improve the efficiency of capital use, and enhance the liquidity of capital factors. The development of rural financial institutions can improve the efficiency of rural production, optimize the efficiency of capital allocation, and effectively promote agricultural and rural development [50]. It can be seen that the development of rural finance can provide financial support for the sustainable development of agriculture and ensure sufficient agricultural funds to a certain extent. Agriculture-related credit is a strong source of support to reflect the financial elements invested in the development of the rural financial field. The level of rural financial development is replaced by the level of agricultural credit [51]. The rural financial market involves formal financial institutions and private financial institutions. Since the data of private financial institutions are difficult to obtain, the agricultural credit in this paper concerns only formal financial institutions and is measured using the ratio of agricultural, forestry, animal husbandry, and fishery loans in various regions to agriculture-related loans.

Rural human capital has an important impact on farmers’ income, which has obvious regional differences, not only in the level of farmers’ income but also in the income distribution gap [52,53]. Rural human capital has a vital impact on agricultural output value, the circulation of agricultural products, and mechanized production [54,55]. It can be seen that rural human capital, as an important assistant in promoting rural economic development, plays an important role in the sustainable development of agriculture, and this paper uses the per capita education years of rural residents in various regions to measure rural human capital. Generally speaking, a high level of human capital promotes sustainable agricultural development, while a low level of human capital may inhibit sustainable agricultural development. Therefore, the control variables selected in this paper include rural infrastructure, financial support for agriculture, agricultural credit, and rural human capital.

To sum up, the main variables selected in this paper and their measurement methods are shown in

Table 2. Main Variables and Measurement Methods.

Variable	Symbol	Alternative Variable	Specific Measurement Method
Sustainable agricultural development	$SAD$	Comprehensive level	Calculated by the constructed indicator system
Digital economy	$D$	Comprehensive level	Calculated by the constructed indicator system
Labor element misallocation	$LEM$	Labor misallocation index	Calculated according to the formula
Capital element misallocation	$CEM$	Capital misallocation index	Calculated according to the formula
Rural infrastructure	$RI$	Rural road level	Ratio of rural highway mileage to highway mileage
Financial support for agriculture	$AFS$	Intensity of financial support for agriculture	Ratio of fiscal expenditure on agriculture to fiscal expenditure
Rural financial development	$RFD$	Agricultural credit level	Ratio of agriculture, forestry, animal husbandry, and fishery loans to agriculture-related loans
Rural human capital	$RHC$	Per capita years of education	Logarithm of average years of education of rural population aged 6 and above

. The variables in this paper are measured with relative indicators, which can better ensure the reliability and stability of the data.

4. Empirical Analysis

4.1. Sample Selection and Data Source

This paper selects panel data from the eastern, central, and western regions of China from 2013 to 2020. Among them, the eastern region contains 11 provinces (cities), specifically Beijing, Tianjin, Hebei, Liaoning, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong, and Hainan; the central region consists of 8 provinces, namely, Shanxi, Jilin, Heilongjiang, Anhui, Jiangxi, Henan, Hubei, and Hunan; and the western region is made up of 11 provinces (cities and autonomous regions), specifically Inner Mongolia, Guangxi, Chongqing, Sichuan, Guizhou, Yunnan, Shaanxi, Gansu, Qinghai, Ningxia, and Xinjiang. The data regarding the digital economy and sustainable agricultural development are mainly from “the China Statistical Yearbook (2014–2021)”. Some of the missing data are from “the China Rural Statistical Yearbook”, statistical yearbooks of various provinces, and the website of the National Bureau of Statistics. Since most of the data are available until 2020, the article selects the data from 2013 to 2020. At the same time, the sample excludes Tibet because there are too much missing data. The original data for the intermediary variables are from “the China Statistical Yearbook” and “the China Rural Statistical Yearbook”. The control variables are rural infrastructure, financial support for agriculture, and the accounting of rural human capital from “the China Statistical Yearbook, “the China Rural Statistical Yearbook”, “the China Population and Employment Statistical Yearbook”, and so on. The data on agriculture, forestry, animal husbandry, and fishery loans are mainly from “the China Financial Yearbook”, and some data are collated through “the China Rural Financial Services Report”.

4.2. Descriptive Statistics

To clarify the impact mechanism between the digital economy and sustainable agricultural development, this paper describes and analyses the interpreted variables, core explanatory variables, intermediary variables, and control variables statistically, as shown in

Table 3. Statistical Description of Related Variables.

	Maximum	Minimum	Mean	Standard Deviation	Median
Sustainable agricultural development	0.727	0.006	0.276	0.184	0.223
Digital economy	0.830	0.014	0.231	0.175	0.180
Labor misallocation index	0.985	0.915	0.963	0.015	0.966
Capital misallocation index	0.994	0.462	0.954	0.070	0.966
Rural infrastructure	0.857	0.582	0.721	0.070	0.734
Financial support for agriculture	0.204	0.041	0.116	0.034	0.116
Rural financial development	0.400	0.038	0.150	0.079	0.137
Rural human capital	2.276	1.766	2.052	0.077	2.061

.

According to Table 3, the average level of the digital economy is only 0.231, and the standard deviation is 0.175. This shows preliminarily judges that China’s digital economy is still relatively small, and there are significant differences between different regions, which means that digital development is uneven. The development of the digital economy has the great potential [56]. The average value of the sustainable agricultural development level is only 0.276, and the standard deviation is 0.184. The difference between different regions is obvious. Therefore, it is judged that the level of sustainable agricultural development needs to be improved. The digital economy may be an important factor affecting the level of sustainable agricultural development. As can be seen from the labor misallocation index and capital misallocation index, both labor and capital show an over-allocation of factors. However, there are differences between the two. The differences in labor factor misallocation are not obvious, while the differences in capital misallocation are relatively large.

4.3. Analysis of Basic Regression Results

Equation (1) and the data from 2013 to 2020 are used for the benchmark regression, and columns (1) and (2) in

Table 4. Estimated Results of the Impact of Digital Economy on Sustainable Agricultural Development.

	Nationwide		Eastern Region		Central Region		Western Region
Explained variable	$SAD$		$SAD$		$SAD$		$SAD$
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
$D$	0.783 *	0.778 *	0.621 *	0.520 *	1.938 *	0.776 *	1.198 *	0.702 *
Constant	0.095 *	0.668 **	0.149 *	0.471	−0.041 ***	0.858 **	−0.003	−0.588 *
control variable	No	Yes	No	Yes	No	Yes	No	Yes
Individual effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Time effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Adjusted R2	0.553	0.618	0.388	0.756	0.778	0.981	0.813	0.988

Note: *, ** and *** represent significant levels of 1%, 5%, and 10%, respectively.

are based on the national level. Column (1) in Table 4, which does not contain the control variables, shows the impact of the digital economy on sustainable agricultural development. Its coefficient is positive and passes the 1% significance test, indicating that the digital economy can better promote sustainable agricultural development. To test the robustness of this conclusion, column (2) in Table 4 contains the control variables. It is found that the estimated coefficient has decreased but is still positive and passes the 1% significance test, which means that there is a problem of missing important explanatory variables. After adding control variables, the role of the digital economy in promoting sustainable agricultural development is still valid, which indicates that the role of the digital economy in promoting sustainable agricultural development is stable. Digitalization plays an important role in agricultural development [57]. This is consistent with the research from the micro-perspective of agricultural enterprises [16,58].

There may be two main reasons for this result. First, with the progress of technology, the digital economy plays an increasingly important role. There is not only a large-scale economic effect but also an increasing marginal return effect. The use of digital technology improves the efficiency of resource utilization and forms a digital circular economy [37]. Second, the digital economy needs to rely on the existence of traditional factors, such as labor and capital, to a certain extent to promote the rapid aggregation of these production factors, improve the allocation efficiency of labor and capital factors, and transform the idle simple labor subject to technical knowledge into complex utility labor, creating an efficient labor force from the low-skilled labor to achieve an effective allocation of factors [59,60]. Third, the digital economy promotes the continuous optimization of the traditional economic model, forms new industries, new formats, and new models, drives intensive integration, collaborative development, efficient utilization, and other new drivers of factors, and greatly expands the scope of the digital economy.

Due to the differences in the sustainable development of agriculture in different regions, the impact of the digital economy on the sustainable development of agriculture is also different. Therefore, this paper needs to perform regression for different regions, dividing the country into the eastern, central, and western regions. The results are shown in Table 4. According to the results, the digital economy in the eastern, central, and western regions has a positive impact on sustainable agricultural development and passes the 1% significance test, which means that, with the improvement of the digital economy and the deepening of the digital degree, sustainable agricultural development will be further improved. After adding control variables, the impact of the digital economy on sustainable agricultural development in the eastern, central, and western regions is still positive, the impact coefficient decreases and the impact passes the 1% significance test, which verifies the stability of the conclusion. It further shows that the impact of the digital economy on sustainable agricultural development has not changed due to regional differences, indicating that the conclusion is reliable and stable.

Comparing the eastern, central, and western regions, the digital economy has a prominent impact on the sustainable agricultural development of the central region, followed by the western region, and that in the eastern region is relatively weak. The possible reason is that the eastern region mainly focuses on manufacturing and service industries, and the agricultural development foundation is relatively weak. Although the digital economy started earlier in the eastern region, it mainly plays a role in the manufacturing and service industries. The central region is different, with a far larger proportion of agriculture than the eastern region. The agricultural industrial structure and layout have certain advantages, which are conducive to improving the competitiveness of agricultural quality, and the rise of agriculture has become an important support for the rise of the central region [61]. The introduction of the digital economy will accelerate sustainable agricultural development. Affected by natural factors, such as climate, the western region has witnessed a rapid development of characteristic agriculture. The introduction of the digital economy has further accelerated agriculture and sustainable agricultural development.

4.4. The Intermediary Effect of Factor Allocation

Table 5. Mediation effect of labor factor misallocation index.

	Nationwide		Eastern Region		Central Region		Western Region
Explained variable	${\tau }_L$	$SAD$	${\tau }_L$	$SAD$	${\tau }_L$	$SAD$	${\tau }_L$	$SAD$
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
$D$	−0.040 *	0.575 *	−0.002 *	0.403 *	−0.0120 *	1.420 *	−0.002 *	0.908 *
${\tau }_L$		−1.915 *		−0.427 *		−2.117 *		−0.564 *
Constant	0.966 *	1.844 *	0.949 *	0.472	0.983 *	2.030 *	0.975 *	0.496
Control variable	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Individual effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Time effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Adjusted R2	0.547	0.911	0.492	0.905	0.837	0.972	0.672	0.950
Intermediary Effect1		0.117		0.002		0.017		0.001
Intermediary Effect2		0.133		0.002		0.018		0.001

Note: The last two rows in Table 5 are the results of the Intermediary Effect of the mismatch of labor factor allocation calculated by $\frac{{\beta }_1\times {\gamma }_2}{{\alpha }_1}$ and $\frac{{\beta }_1\times {\gamma }_2}{{\gamma }_1}$ two methods, respectively. * represents significant levels of 1% respectively.

and

Table 6. Mediation effect of capital factor misallocation index.

	Nationwide		Eastern Region		Central Region		Western Region
Explained variable	${\tau }_K$	$SAD$	${\tau }_K$	$SAD$	${\tau }_K$	$SAD$	${\tau }_K$	$SAD$
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
$D$	−0.093 *	0.615 *	−0.067 *	0.379 *	−0.016 *	1.461 *	−0.048 *	0.703 *
${\tau }_K$		−0.393 *		−0.365 *		−0.403		−0.415
Constant term	0.944 *	0.365 *	0.906 *	0.398 *	0.977 *	0.342	0.977 *	0.470 ***
Control variable	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Individual effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Time effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Adjusted R2	0.137	0.915	0.109	0.932	0.897	0.965	0.554	0.985
Intermediary Effect1		0.056		0.061		0.004		0.022
Intermediary Effect2		0.059		0.065		0.004		0.028

Note: The last two rows in Table 6 are the results of the Intermediary Effect of the mismatch of capital factor allocation calculated by $\frac{{\beta }_1\times {\gamma }_2}{{\alpha }_1}$ and $\frac{{\beta }_1\times {\gamma }_2}{{\gamma }_1}$ two methods, respectively. * and *** represent significant levels of 1% and 10%, respectively.

list the intermediary effect test results of labor factor misallocation and capital factor misallocation in the digital economy, enabling sustainable agricultural development. Columns (1), (3), (5), and (7) of Table 5 and Table 6, respectively, present the results estimated according to Equation (3) on the national level, in the eastern region, in the central region, and in the western region, that is, the impact of the digital economy on labor and capital allocation. Both the national level and the sub-regional levels show that the estimated coefficients of the digital economy on the labor factor misallocation index and the capital factor misallocation index are negative, and both pass the 1% significance test. These results indicate that the digital economy can reduce the level of labor factor misallocation and capital misallocation, but the impact differs between regions.

Columns (2), (4), (6), and (8) of Table 5 and Table 6 list the estimation results based on Equation (4) at the national level and in the eastern region, central region, and western region; that is, factor allocation is the intermediary effect of the digital economy on sustainable agricultural development. At the national level, the estimated coefficient of the digital economy for sustainable agricultural development is positive and passes the 1% significance test. Compared with Table 5 and Table 6, the estimated coefficient has decreased. Table 5 and Table 6 show that the digital economy significantly reduces the labor and capital misallocation indexes. The estimation coefficient of the digital economy for sustainable agricultural development is positive, and the estimation coefficient of the labor and capital misallocation indexes for sustainable agricultural development is negative. This shows that the digital economy can promote sustainable agricultural development by reducing the level of factor misallocation and verifies Hypothesis H2 and Hypothesis H3. From the regional perspective, the digital economy in the eastern, central, and western regions promotes sustainable agricultural development through labor factor allocation, and this passes the 1% significance test, but the performance of capital factor allocation is different. The digital economy in the eastern region promotes sustainable agricultural development through capital factor allocation and passes the 1% significance test. Although the impact of capital factor allocation on sustainable agricultural development in the central and western regions is positive, it does not pass the significance test.

Through the intermediary effect formulas $\frac{{\beta }_1\times {\gamma }_2}{{\alpha }_1}$ and $\frac{{\beta }_1\times {\gamma }_2}{{\gamma }_1}$, the intermediary effect of factor allocation is obtained. At the national level, the intermediary effect of labor factor allocation is greater than that of capital factor allocation, which means that the digital economy can optimize the allocation of labor and capital factors but that their impacts are different. It also shows that the digital economy is more effective in reducing labor misallocation and improving labor allocation efficiency, but the performance of different regions is slightly different. In the eastern and western regions, the development of the digital economy is more conducive to the allocating of capital elements, reducing the distortion of capital elements, and improving the efficiency of capital element allocation. The digital economy in the central region is more supportive of the allocation of labor elements. From the perspective of the intermediary effect, the intermediary effect of labor factor allocation is greater in the whole country than in the central part and greater in the east than in the west. The intermediary effect of capital factor allocation is shown to be greater in the east than in the whole country and greater in the west than in the center. The continuous flow of the digital economy across the country has a scale effect and improves the allocation of factors. The eastern region is relatively developed, and the development of digital technology facilities is relatively complete. At the same time, the eastern region has greater advantages in science and technology, talents, finance, and so on. The digital economy in the eastern region has a greater impact on the efficiency of capital allocation.

4.5. Robustness Test

To determine whether the above conclusions are robust, the intermediary variable uses factor allocation efficiency. The measure of the factor allocation efficiency follows the DEA method, and the regional regression is conducted according to Equations (2)–(4). The results show that the development of the digital economy can improve the efficiency of factor allocation and that the digital economy can promote sustainable agricultural development through the improvement of factor allocation efficiency. The intermediary effect of the digital economy in the eastern, central, and western regions is tested, respectively. It shows that the digital economy in the eastern, central, and western regions significantly promotes sustainable agricultural development, and the allocation efficiency coefficient is positive, which means that the digital economy optimizes the allocation of labor and capital elements and improves the efficiency of factor allocation. The results of the digital economy optimizing the traditional factor allocation in different regions to promote sustainable agricultural development are still valid. The estimation results verify the above conclusions, indicating that the conclusions in this paper are reliable.

5. Conclusions and Policy Recommendations

Based on the typical fact that the digital economy has driven economic and social development, accelerated digital transformation, profoundly affected the direction and path of agricultural transformation and rural industrial integration, and promoted sustainable agricultural development, this paper used panel data from 30 provinces from 2013 to 2020 to explore the impact of the digital economy on sustainable agricultural development by using a panel data model and an intermediary effect model on the basis of an indicator system of the digital economy and sustainable agricultural development. By introducing the intermediary variable of factor allocation, this paper analyzed the transmission mechanism of the digital economy to sustainable agricultural development and drew the following conclusions.

First, at the national level, the digital economy has played a significant positive role in sustainable agricultural development; that is, the digital economy has significantly promoted the sustainable development of agriculture, which means that, with the deepening of the digital economy, agriculture has also developed. When the control variable was introduced, this conclusion remained valid, which meant that the impact of the digital economy on the sustainable development of agriculture was robust. At the regional level, the digital economy still played a role in promoting the sustainable development of agriculture, but the degree of impact had obvious differences. The central region was the most prominent, followed by the western region, and the eastern region was relatively weak. This indicated that it is necessary to adjust measures to local conditions and formulate relevant policies in different regions.

Second, there were different degrees of labor factor mismatch and capital factor mismatch in all provinces, and the digital economy had a significant positive impact on the optimization of labor and capital factor allocation. This meant that the input of the digital economy increased the input scale of high-end factors and reduced the distortion coefficient of factor allocation, which was conducive to improving the mismatch of capital factors and played an important role in improving the mismatch of labor factors.

Third, at the national level, the digital economy improved sustainable agricultural development by reducing the misallocation of labor and capital allocation. At the regional level, the level of sustainable agricultural development in the eastern region depended on the optimization effect of the digital economy on the allocation of labor and capital. While the level of sustainable agricultural development in the central and western regions depended on the optimization effect of the digital economy on the allocation of labor factors, although the allocation effect of capital factors had a positive impact on sustainable agricultural development, it did not pass the significance test.

With the deepening of digitalization, relevant questions are how to improve the allocation of factors through the digital economy and how to improve the role of the digital economy in promoting the sustainable development of agriculture. Based on the above conclusions, different policy recommendations are put forward.

The first recommendation is to strengthen infrastructure construction and improve the digital technology environment. By strengthening the construction of digital agricultural infrastructure, we can vigorously popularize basic information projects in rural areas, especially remote areas, and comprehensively promote the extension of digital technology to rural areas or agricultural industrial parks. We can pilot new digital technologies in some regions to narrow the “digital gap” between urban and rural areas. Through digital platforms, we can provide digital information services for farmers. According to agricultural resource endowment, we can implement heterogeneous agricultural development policies in different regions to reduce agricultural production costs, optimize the circulation of agricultural products, and increase farmers’ income.

The second recommendation is to strengthen the development of the digital economy and give full play to the allocation effect of digital elements, use digital technology to promote the effective allocation of labor and capital, reduce the degree of mismatch, improve the allocation efficiency, improve the allocation of factor markets, and optimize the allocation structure. We can make full use of the scale economy effect of the digital economy, change the development mode from “input of traditional factors such as labor and capital” to “digital economy + optimization of factor allocation”, and gradually cultivate the digital economy as a new driving force for sustainable agricultural development.

The third recommendation is to pay attention to the effective connection between the digital economy and agricultural development and promote the deep integration of digital technology and agriculture. As an emerging economy, we need to consolidate the foundation further and cultivate the application of digital technology and other emerging elements in rural economic development. In combination with the current development of digital technology, we can strengthen the connection between digital technology and agriculture and make agricultural production, circulation, and distribution progress to achieve high quality. We can gradually promote the deep integration of digital technology and agriculture by way of points, lines, and surfaces, optimize new technologies, new formats, new models, and new products in the agricultural field, improve agricultural productivity and optimize the structure of agricultural products. Compared with the central and western regions, the eastern regions have the advantages of science and technology, talents, and capital. The central and western regions should improve the level of digital technology and strengthen the integration of the digital economy and sustainable agricultural development.