Digital Economy, Factor Allocation, and Resilience of Food Production

Abstract

This paper systematically explores the impact of the digital economy on the resilience of food production and its mechanism of action based on panel data from 30 provincial-level administrative regions in China from 2011 to 2022. This study shows that the digital economy significantly enhances the ability of the food production system to cope with external shocks by improving resource allocation efficiency and mitigating factor mismatch. Specifically, the digital economy directly improves the stability and resilience of food production through the widespread application of digital technology; at the same time, it indirectly contributes to the enhancement of food production resilience by alleviating the mismatch of labor and capital factors. Heterogeneity analyses show that there are regional differences in the impact of the digital economy, in which the main food-producing regions benefit more significantly due to their agricultural resource advantages and policy support, and the improvement of food production resilience in the central region is particularly prominent. This study provides an important theoretical basis and practical reference for exploring the potential of the application of digital economy in agriculture and formulating policies to enhance food production resilience.

1. Introduction

Food production forms the bedrock of economic development, while food security is fundamental not only to social stability and public well-being, but also to sustainable economic and social progress. For a country as populous as China, the importance of food security is even more self-evident. In 2023, China’s total food output reached 695.41 million tons, achieving 20 consecutive years of abundant harvests. This milestone highlights the success of supportive government policies and ongoing innovations in agricultural technology. Food security has consistently been a strategic priority for China, with central government policy documents repeatedly emphasizing the critical role of food production. However, the world is undergoing unprecedented transformations, accelerated by the frequent occurrence of geopolitical conflicts, public health emergencies, and intensified climate change. These factors pose significant challenges to food production [1], threatening the stability of food supplies and drawing increased attention to the resilience of food production systems. Food production resilience refers to the capacity of the system to maintain stability and recover rapidly from external shocks such as climate change, natural disasters, and market volatility. It involves not only stable food output, but also the continuity of supply chains and the safeguarding of farmers’ incomes [2]. Establishing a resilient food production system capable of effectively responding to uncertainty and risk is crucial to ensuring food security under complex conditions. As external challenges grow increasingly intricate, enhancing resilience requires exploring innovative technologies and new models beyond traditional policy approaches [3].

The rise of the digital economy presents new opportunities for enhancing the resilience of food production. The digital economy, characterized by the use of information technology and data resources, relies on technologies such as big data, artificial intelligence, and blockchain to reshape traditional production and consumption models, optimizing resource allocation. Beyond its transformative impact on the manufacturing and service industries, the digital economy is increasingly demonstrating immense potential in agricultural production [4]. At this pivotal moment of technological innovation and industrial restructuring, the Chinese central government has prioritized the integration of the digital economy with traditional agriculture. Policies like the Rural Revitalization Strategy Plan (2018–2022) and the Digital Agriculture and Rural Development Plan (2019–2025) explicitly promote the acceleration of agricultural informatization and digital transformation, the establishment of a modern agricultural industry system, and the enhancement of agriculture’s comprehensive competitiveness to ensure food security. Can the digital economy truly drive improvements in food production resilience? If so, how can we maximize its positive spillover effects to enhance food production resilience? Furthermore, how do these effects vary across different regions? These questions demand urgent scientific analysis. Assessing food production resilience accurately and evaluating the digital economy’s potential as a driver of resilience from a factor allocation perspective hold significant theoretical and practical implications for future policies. Such analysis will contribute to building resilient food systems, ensuring food security, and advancing the construction of a strong agricultural nation. Based on this, this study explores the relationship between the digital economy and the resilience of food production with the help of the panel data of 30 provincial-level administrative regions in China from 2011 to 2022, using a two-way fixed-effects model, so as to reveal in depth the effect and mechanism of the digital economy on the resilience of food production.

2. Literature Review

The current literature on topics related to this study follows three primary lines of inquiry.

2.1. Relevant Studies on the Digital Economy

The first line of inquiry examines the digital economy itself. As a major driving force behind global economic transformation, the digital economy encompasses a wide range of digital technologies, including the internet, mobile networks, e-commerce, and artificial intelligence. These technologies significantly reshape global economic structure and influence development trajectories by enhancing productivity, optimizing resource allocation, and accelerating innovation [5]. With rapid advancements in digital technologies, academic discussions on the digital economy have grown increasingly sophisticated, encompassing multiple dimensions, from foundational theories to practical applications. In foundational research, scholars have focused on defining and constructing theories around the digital economy. At its core, the digital economy is rooted in advancements in information and communication technology (ICT), dating back to the 1950s, and has expanded significantly with the rise of the internet and mobile technologies. It spans not only traditional economic activities related to production and consumption, but also emerging areas such as the platform economy and the attention economy [6]. This theoretical framework has gained broad support in the literature, with scholars emphasizing how the digital economy is transforming market operations and reshaping competitive dynamics, allowing firms to innovate and adapt rapidly in the global market [7]. In terms of impact mechanisms, the rapid growth of the digital economy is largely driven by economies of scale and the spillover effects of the ICT industry. By enhancing the efficiency of traditional industries, the digital economy not only accelerates overall economic growth but also brings about fundamental changes in production and consumption patterns [8]. This transformation is not limited to developed economies; it also has a profound impact on developing countries, providing unique opportunities to leapfrog traditional development pathways [9].

To accurately assess the contribution of the digital economy to national and regional economies, scholars have developed various measurement methods, ranging from foundational theories to specific indicator frameworks. One approach is the “Digital Economy Satellite Account” (DESA), which integrates digital products and services into the national economic accounting system. By constructing Digital Supply and Use Tables (DSUT), this method captures the value-added processes of “intangible assets” and “free digital services” within the digital economy, providing a foundation for more comprehensive statistical evaluation [10]. Another method focuses on applying the Net Present Value (NPV) approach to valuing data assets. By treating data as a non-productive asset, this approach facilitates a more precise evaluation of its long-term economic benefits [11]. Regarding indicator construction, some studies have developed comprehensive evaluation index systems for the digital economy, assessing its performance across multiple dimensions. By introducing a multidimensional evaluation framework that includes “digital users”, “digital platforms”, “digital innovation”, and “digital industries”, researchers have constructed regional digital economy indices to reflect disparities in development across regions [12]. This method quantifies the digital economy development levels of various provinces and cities and highlights the uneven development of the digital economy between regions. Data mining technologies also offer new perspectives for measuring the digital economy. Through the application of big data and deep learning techniques, some studies have proposed data-mining-based methods for measuring the digital economy. By analyzing and processing big data, this approach better reflects trends and changes in digital economic development across regions, particularly gaining traction in studies of China’s digital economy [13]. At the international level, regions such as the European Union have adopted the “Digital Economy and Society Index” (DESI) to monitor digital economy development among member countries. DESI encompasses not only the use of digital technologies, but also the impact of the digital economy on societal development, providing a comprehensive indicator system for assessing the contribution of digital transformation to the overall economy [14].

2.2. Relevant Studies on the Agricultural Resilience and Food Production Resilience

The second area of focus is research on agricultural and food production resilience. Agricultural resilience is a multifaceted concept that captures the ability of agricultural systems to endure and adapt to natural, social, and economic disruptions. Early research on agricultural resilience primarily concentrated on the sustainability of agro-ecosystems, investigating how to preserve the fundamental functions of agricultural production systems amidst environmental changes. However, as global climate change, market volatility, and social dynamics increasingly influence agriculture, scholars have recognized that agricultural resilience extends beyond ecological systems. It now encompasses the adaptive and recovery capacities of the broader socio-ecological framework [15]. In developing agricultural resilience, social and economic dimensions have emerged as pivotal areas of study. Assessing agricultural resilience requires an integrated approach that considers ecological shifts alongside the complexities of agricultural supply chains and the influence of socio-economic structures [16]. Generally, agricultural resilience consists of three essential components: resistance, adaptability, and recovery capacity. Resistance pertains to a system’s capacity to remain stable in response to short-term disturbances, adaptability highlights the system’s evolving responses to long-term environmental changes, and recovery capacity reflects the agricultural system’s ability to safeguard its core functions through structural adjustments amid significant external disruptions [17].

In recent years, scholars have advocated for agricultural resilience frameworks that account for multi-scale complexities—spanning farm, regional, and national levels—and examine the interconnections among ecological, social, and economic factors from a comprehensive perspective. This approach is essential for addressing future uncertainties in agriculture effectively [18]. Food production resilience, originally an offshoot of agricultural resilience, has evolved significantly over time. Early studies primarily focused on the capacity of food production systems to sustain food supply amidst environmental challenges like natural disasters and climate change. With the intensification of global climate change, food production resilience has been redefined as a more comprehensive concept that integrates production management, supply chain stability, and socio-economic considerations [19]. As this field has developed, research on food production resilience has expanded from narrow, single-dimensional analyses to multidimensional perspectives that encompass not only ecosystem health, but also social equity and economic efficiency [20]. The globalization of food trade has further complicated food production systems, particularly in terms of supply chain dynamics. Enhancing the resilience of these systems through policy initiatives, market adaptations, and other strategic measures has become crucial. Increasingly, researchers are highlighting the importance of harmonizing resilience across both regional and global food production systems to address resilience disparities [21].

With the progression of agricultural industrialization, food production systems have become increasingly dependent on centralized supply chain structures. While this intensive production model enhances efficiency, it also amplifies vulnerability to external shocks, including natural disasters and market fluctuations [22]. The resilience of contemporary food production systems hinges not only on ecological and economic stability, but also on factors such as social organization, policy backing, and market dynamics. Enhancing food production resilience involves fostering diversified agricultural models and adaptive supply chain management, which bolsters adaptability across scales—from local to global—to secure food availability [18]. Assessing food production resilience demands a comprehensive approach that integrates ecological, economic, and social dimensions. Recent research has produced various indicator frameworks and tools to evaluate resilience in food production. Ecologically, resilience is assessed through the diversity of land-use practices and the sustainability of ecosystem services. Economically, key indicators include market price stability and resource use efficiency. Social resilience, meanwhile, is shaped by policy support, technological innovation, and the effectiveness of social organization, all of which influence the capacity of food production systems to adapt and respond to external shocks [23,24].

2.3. Relevant Studies on the Impact of the Digital Economy on the Resilience of Food Production

The third area of research explores the impact of the digital economy on food production resilience. With the rapid expansion of the digital economy, various digital technologies are becoming increasingly integrated into agriculture, significantly influencing the resilience of food production systems. Tools such as big data, the Internet of Things (IoT), and artificial intelligence (AI) enhance the adaptive capacity of food production systems, enabling them to respond more effectively to external shocks, including climate change, market fluctuations, and policy shifts. Specifically, these digital tools optimize resource efficiency and streamline production processes, which not only reduce resource waste, but also help mitigate environmental impacts [4]. Data-driven management practices also enable more precise environmental predictions, assisting in minimizing the effects of natural disasters on food production [25]. Economically, the digital economy greatly improves the efficiency and profitability of food production. Digital supply chain management has enhanced transparency and flexibility within supply chains, reducing information asymmetry. By facilitating better access to market data, digital tools empower agricultural producers to align their production decisions with market demand, thereby lowering the risk of economic losses [26]. On a social level, digital technologies have advanced knowledge dissemination and fostered innovation within agricultural communities. Digital platforms not only provide farmers with technical support and policy updates, but also enhance collaboration within agricultural networks, strengthening the social resilience of these systems when facing external challenges [27]. The digital economy enhances food production resilience through several mechanisms. First, advancements in digital technology support precision agriculture and smart farming, optimizing resource utilization within food production systems. This improved efficiency allows farmers to better respond to environmental changes and natural disaster risks [28]. Second, the digital economy bolsters the transparency and flexibility of agricultural supply chains. Digital supply chain management systems enhance the efficiency of each stage—production, transportation, and sales—enabling rapid adaptation to market shifts and unforeseen events, thereby strengthening supply chain resilience. Real-time supply chain monitoring allows farmers and agricultural businesses to adjust promptly based on market demand, mitigating economic losses due to overproduction or shortages [29]. Additionally, digital platforms and technology facilitate faster dissemination and adoption of agricultural innovations. Through these platforms, farmers gain easier access to information on emerging technologies, accelerating their practical application and boosting the innovation and adaptability of food production systems. This mechanism for technology dissemination not only raises production efficiency, but also strengthens the system’s capacity to adapt to environmental shifts [30].

Existing research offers valuable insights for this study; however, certain limitations persist in research perspectives and methodologies. First, a localized theoretical framework for food production resilience, specifically suited to China’s distinctive agricultural and socio-economic context, remains underdeveloped. Further investigation is needed to refine evaluation methods for food production resilience and to explore the mechanisms influencing it. Second, much of the literature has examined the digital economy’s impact on food production resilience, primarily from the perspective of agricultural system factors, such as technological innovation and risk mitigation, often overlooking other critical factors in the production process. This paper seeks to address these gaps by making two key contributions: 1. Using a time-series Markov process model to quantitatively measure regional food production resilience, this study analyzes how digital-era transformations drive resilience through the digital economy, thereby contributing to research on the digital economy’s role in agriculture. 2. By empirically examining the digital economy’s impact on food production resilience and its internal mechanisms, this study offers new empirical evidence to inform discussions on digital economy development and strategies for enhancing industry resilience.

3. Theoretical Analysis and Research Hypotheses

3.1. Direct Effects of the Digital Economy on Food Production Resilience

The emergence and rapid development of the digital economy have had substantial direct effects on agricultural production. From an economic standpoint, technological progress and gains in total factor productivity are key drivers of economic growth. The Solow Growth Model posits technological advancement as a critical factor for long-term economic growth, with technology functioning as an exogenous variable that plays a decisive role in output increases [31]. Digital technologies introduce exogenous technological advancements to agriculture, enabling a shift from labor-intensive to technology-intensive production models. This transition improves agricultural output efficiency and stability, minimizes production waste, and directly strengthens the food production system’s resilience against natural disasters and resource limitation [4,32]. Additionally, technological innovation drives production transformation, enhancing the adaptability of agricultural systems to external changes. The theory of economies of scale further highlights that as production scales expand, unit production costs decrease, thus boosting efficiency and competitiveness. In grain production, the digital economy fosters economies of scale through optimized supply chain management. Technologies like big data, IoT, and blockchain increase transparency and efficiency within agricultural supply chains, reducing waste and uncertainties in distribution [33]. Digital platforms enable farmers to connect directly with markets, minimizing intermediaries and transaction costs. This streamlined approach facilitates market access for small-scale producers and contributes to a more stable food supply [34]. Digital technologies also facilitate automation and standardization in large-scale production, enabling producers to maintain consistent quality across scale and benefit from economies of scale. This approach, which combines large-scale production with supply chain optimization, enhances the food production system’s resilience to external shocks. Furthermore, the digital economy enables more precise supply chain management, reducing logistics and warehousing costs and allowing a broader range of agricultural entities to achieve economies of scale. By expanding market access through e-commerce and other digital channels, the overall efficiency of the grain supply chain improves [35]. Through optimized supply chain processes, the digital economy directly fosters economies of scale in agriculture, strengthening food production resilience [36].

Based on this, Hypothesis 1 is proposed:

Hypothesis 1.

The digital economy can enhance the resilience of food production.

3.2. Indirect Effects of the Digital Economy on Food Production Resilience

Both formal and informal institutional barriers—such as the urban–rural dual structure and imperfections in factor markets—create substantial mismatches in labor and capital allocation within the agricultural sector [37]. Factor allocation theory and total factor productivity theory emphasize that effectively matching production factors like labor and capital is essential for improving production efficiency. Endogenous growth theory and empirical research further indicate that factor mismatches not only reduce production efficiency, but also hinder long-term economic growth and the resilience of production systems [38,39,40]. Current research suggests that the digital economy can reduce barriers to factor mobility, facilitating a more rational allocation of resources between urban and rural areas [41]. As shown in Figure 1,on one hand, digital technologies improve labor supply–demand matching in agriculture, significantly reducing labor mismatch. This technological progress enhances agricultural productivity by optimizing production processes and strengthening information flows, which mitigates the adverse effects of labor mismatches on food production resilience. Additionally, with advancements in artificial intelligence and automation, the demand for repetitive labor in agriculture decreases, while the need for highly skilled labor increases, further reducing labor mismatches and enhancing the flexibility and adaptability of food production systems [42]. By improving labor allocation, the digital economy also promotes sustainable agricultural development, stabilizing food production systems when facing external shocks. Moreover, the digital economy spurs the development of financial technology and improves transparency in capital allocation, facilitating more efficient capital flow and utilization, which, in turn, boosts agricultural production efficiency. Beyond addressing capital mismatches, the digital economy also injects new momentum into sustainable agricultural development. By incorporating digital technologies, capital resources are more effectively directed toward critical stages of agricultural production, thereby enhancing the resilience of food production systems [43].

Accordingly, the following hypotheses are proposed:

Hypothesis 2a.

The digital economy enhances the resilience of food production by mitigating labor factor mismatch.

Hypothesis 2b.

The digital economy enhances the resilience of food production by mitigating capital factor mismatch.

4. Research Design

4.1. Variable Selection

4.1.1. Dependent Variable

Using food production resilience (Fpr). combined with the previous analysis, this paper argues that the resilience of food production can be understood as the ability of the food production system to rely on internal and external dynamics to resist internal disturbances and external shocks in order to reshape the new development path, including the pre-disaster resistance, to maintain the stability of the basic functions, the adaptive power in the disaster to adjust or change the mode of production, and the post-disaster reconstruction power for sustainable development. The food production system is a typical complex dynamic system, whose resilience is demonstrated not only by its short-term response to external shocks, but also by its performance over long-term changes. Time series analysis methods can effectively capture the dynamic characteristics of food production, and the Markov process model can reveal the system’s path dependency and self-recovery capabilities. This method allows for a quantitative analysis of the relationship between production volatility and long-term trends, thereby assessing system resilience [44]. The specific steps are as follows:

First, establish the Markov process model. In the food production system, yield changes often exhibit path dependency. This path dependency stems from the accumulation of technology, the use of inputs, and the long-term development of the ecological environment during agricultural production. These factors collectively result in notable continuity and inertia in food production [45]. The Markov process model can effectively describe such dynamic changes. Changes in per capita food production can be represented by the following equation:

$$Y_{it}={\displaystyle \sum _{j=1}^k{\beta }_M^j}{Y}_{i,t-1}+{\gamma }_M{X}_{it}+{\epsilon }_{Mit}$$

(1)

where $Y_{it}$ represents the per capita food production of region i at time t. $Y_{i,t-1}$ represents the per capita food production in the previous period, reflecting the time dependence of yield; $X_{it}$ represents other variables affecting food production, such as climatic conditions, agricultural policies, and technological inputs; ${\epsilon }_{it}$ is a random disturbance term. This model can not only capture historical changes in food production, but also predict future yields based on previous data.

Second, calculate the conditional expectation and variance. To further assess the resilience of food production systems across regions, the conditional expectation and conditional variance of food production need to be calculated. The conditional expectation represents the expected value of future food production under given external conditions, reflecting the system’s recovery ability in response to external shocks. The conditional variance represents the volatility of food production, while smaller volatility indicates higher system stability. By analyzing the conditional expectation, the future stability of food production can be evaluated, while the conditional variance reflects the potential degree of fluctuation when the system faces external shocks.

Using the assumption of zero mean for the random error term, estimate the conditional expectation prediction value for region i at time t:

$${\mu }_{it}^1=E(Y_{it}|Y_{i,t-1},X_{it})={\displaystyle \sum _{j=1}^k{\beta }_M^j}{Y}_{i,t-1}+{\gamma }_M{X}_{it}$$

(2)

Derive the second central moment using the residuals. Summarizing previous studies, the residuals of the first central moment are used to estimate the variance equation [46]. The form of the variance equation is as follows:

$${\epsilon }_{Mit}^2={\displaystyle \sum _{j=1}^k{\beta }_V^j}{Y}_{i,t-1}+{\gamma }_V{X}_{it}+{\epsilon }_{Vit}$$

(3)

where ${\epsilon }_{Mit}^2$ is the square of the first-order residuals, ${\beta }_V^j$ represents the coefficient in the variance equation, and ${\gamma }_V$ represents the coefficients of other influencing factors.

Similarly, following the assumption of zero mean of the random error term ${\epsilon }_{Vit}$, the conditional variance of food production in area i at time t can be estimated:

$${\mu }_{it}^2={\displaystyle \sum _{j=1}^k{\beta }_V^j}{Y}_{i,t-1}+{\gamma }_V{X}_{it}$$

(4)

This sets the food security threshold. In this study, the internationally recognized food security baseline (400 kg/person/year) is used as the food security threshold. To measure the capacity of each region to maintain food security at different times, the probability of food production exceeding the security threshold can be calculated using the following formula:

$$Fp{r}_{it}={\rho }_{it}=P(Y_{it}\ge Y*)=F({\displaystyle \frac{Y_{it}-{\mu }_{it}^1}{{\mu }_{it}^2}})$$

(5)

Here, Y* represents the food security threshold, while F denotes the distribution function (e.g., normal distribution), which calculates the probability that the yield exceeds this security threshold. This paper applies this approach to quantify the capacity of regional food production systems to maintain a stable output under external shocks. That is, the core explanatory variable of the paper is food production resilience (Fpr).

4.1.2. Core Explanatory Variable

Digital Economy Level (Dig): The digital economy level is the main explanatory variable. Building on prior research [47], this study constructs a digital economy evaluation index system based on three dimensions: digital infrastructure, digital industry development, and digital inclusive finance (see

Table 1. Evaluation index system for digital economy level.

Primary Indicators	Secondary Indicators	Tertiary Indicators	Indicator Attribute
Digital Economy Level	Digital Infrastructure	Number of domain names (10,000)	+
		Number of IPv4 addresses (10,000)	+
		Number of internet broadband access ports (10,000)	+
		Mobile phone penetration rate (per 100 people)	+
		Length of optical cable per unit area (km/km²)	+
	Digital Industry Development	Number of information technology companies	+
		Number of companies with websites (per 100 companies)	+
		Proportion of companies engaged in e-commerce activities (%)	+
		E-commerce sales (billion yuan)	+
		Software business revenue (billion yuan)	+
	Digital Inclusive Finance	Breadth of coverage index	+
		Depth of use index	+
		Digitalization index	+

). Digital infrastructure, represented by metrics such as the number of domain names and IPv4 addresses, serves as a crucial foundation for technology application, diffusion, and digital economy growth. Digital industry development, measured by indicators like the number of IT enterprises and websites per 100 enterprises, drives social production efficiency and supports industrial transformation and upgrading. Digital inclusive finance, assessed using the Peking University Digital Inclusive Finance Index, encompasses efficient payment services and accessible financial resources, essential components of the digital economy. These include the breadth of coverage, depth of use, and digitalization indices. The breadth of coverage index measures the geographic and demographic reach of financial services; the depth of use index assesses the actual use and depth of financial services; and the digitization index reflects the level of digitalization and innovation in financial services. All indicators are positive, and their aggregate measurement is achieved via principal component analysis.

4.1.3. Mechanism Variables

Labor and capital factor mismatch (Lmi and Kmi): The mechanism variables include labor factor mismatch index (Lmi) and capital factor mismatch index (Kmi). With a constant land endowment assumption, this study focuses on the allocation efficiency of labor and capital. Factor mismatch indices serve as key indicators of allocation efficiency. Most existing studies apply the Solow residual method, as described by Bai Junhong and Liu Yuying (2018) [48], to determine the factor mismatch index. Assuming that regional production functions adhere to the constant returns-to-scale Cobb–Douglas (C-D) form, this paper employs a panel model with varying intercepts and slope coefficients to estimate the output elasticities of agricultural capital and labor in each region, subsequently calculating labor and capital factor mismatch indices as follows:

$$Lm{i}_{it}={\displaystyle \raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{${\gamma }_{l_i}$}\right.}-1, Km{i}_{it}={\displaystyle \raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{${\gamma }_{k_i}$}\right.}-1$$

(6)

In Equation (6), ${\gamma }_{l_i}$ and ${\gamma }_{k_i}$ represent the labor and capital price distortion coefficients for region i, respectively, calculated as follows:

$${\gamma }_{l_i}=({\displaystyle \frac{L_i}{L}})/({\displaystyle \frac{s_i{\beta }_{Li}}{{\beta }_L}}), {\gamma }_{k_i}=({\displaystyle \frac{K_i}{K}})/({\displaystyle \frac{s_i{\beta }_{ki}}{{\beta }_k}})$$

(7)

In Equation (7), $s_i$ represents the share of regional agricultural output in the total agricultural output of the entire economy; ${\displaystyle \frac{L_i}{L}}$ and ${\displaystyle \frac{K_i}{K}}$ represent the shares of actual labor and capital usage in the total labor and capital usage of the entire economy for the region, respectively; $s_i{\beta }_{L_i}$ and $s_i{\beta }_{k_i}$ represent the proportion of labor and capital allocated efficiently in region i, respectively; and ${\beta }_{Li}$ and ${\beta }_{Ki}$ denote the output elasticity of labor and capital, respectively, for the region. After calculating the labor price distortion coefficient ${\gamma }_{l_i}$ and capital price distortion coefficient ${\gamma }_{k_i}$ from Equation (7), the mechanism variables, labor factor mismatch index (Lmi) and capital factor mismatch index (Kmi), of this paper can be calculated by combining them with Equation (6).

Considering that the index may contain negative values, this paper takes the absolute value of the index. Agricultural output involved in the calculation is represented by the total output value of agriculture, forestry, animal husbandry, and fishery; the stock of agricultural capital is measured using the perpetual inventory method, represented by multiplying the proportion of fixed-asset investment in agriculture, forestry, animal husbandry, and fishery to the total social fixed capital stock, with a depreciation rate of 9.6%. The stock of agricultural labor is represented by the number of people employed in agriculture, forestry, animal husbandry, and fishery.

4.1.4. Control Variables

Drawing on the existing literature, this paper includes the following control variables in the regression analysis: industrial structure (Is), level of agricultural mechanization (Aml), fiscal support for agriculture Fsa), urbanization rate (Ur), and agricultural management scale (Sab). The industrial structure is represented by the share of value added by the primary industry in the region’s GDP; the agricultural mechanization level is captured through the logarithm of total agricultural machinery power in the region, as agricultural facilities and equipment can influence producers’ operational decisions, thereby impacting food production resilience; and fiscal support for agriculture is measured by the proportion of expenditure on agriculture, forestry, and water affairs relative to the general local government budget. Government support through economic policies and other measures may affect agricultural development, which in turn influences food production resilience; urbanization rate is assessed by the proportion of the urban population, as urbanization can impact food system resilience by altering rural labor structures; and agricultural management scale is represented by the sown area per laborer, a key factor affecting agricultural production efficiency, and thus influencing food production resilience.

4.2. Model Specification

Building on the theoretical analysis above, and accounting for the unique characteristics of the sample data (which passed the Hausman test), this paper employs a fixed effects model to investigate the impact of the digital economy on food production resilience. The model specification is as follows:

$$Fp{r}_{it}={\beta }_0+{\beta }_{1}Di{g}_{it}+{\beta }_{2}Control{s}_{it}+{\mu }_i+{\omega }_t+{\epsilon }_{it}$$

(8)

In Equation (8), $Fp{r}_{it}$ and $Di{g}_{it}$ represent the dependent variable of food production resilience and the explanatory variable of digital economy development level for province i in year t, respectively. $Control{s}_{it}$ represents other control variables. ${\mu }_i$ and ${\omega }_t$ are dummy variables for province and year, respectively, used to control for individual and time effects, while ${\epsilon }_{it}$ represents the random disturbance term.

In addition to exploring the direct effect, this paper examines the indirect effect of the digital economy on food production resilience through improvements in factor mismatch. To address the endogeneity concerns between mechanism variables and the dependent variable—which the mediation effect model cannot fully resolve—and recognizing that an interaction effect model may not comprehensively explain the economic mechanism, this study employs an approach from previous research [49]. Specifically, we directly regress the explanatory variable on the mechanism variable to test for the existence of the mechanism effect. The model specification is as follows:

$$Lm{i}_{it}={\beta }_0+{\beta }_{1}Di{g}_{it}+{\beta }_{2}Control{s}_{it}+{\mu }_i+{\omega }_t+{\epsilon }_{it}$$

(9)

$$Km{i}_{it}={\beta }_0+{\beta }_{1}Di{g}_{it}+{\beta }_{2}Control{s}_{it}+{\mu }_i+{\omega }_t+{\epsilon }_{it}$$

(10)

4.3. Data Sources

This study uses panel data from 30 provinces (including autonomous regions and municipalities) in China for the period of 2011 to 2022. Tibet, Hong Kong, Macau, and Taiwan are excluded due to data availability and comparability constraints. Primary data sources include the China Statistical Yearbook, China Rural Statistical Yearbook, China Rural Management Statistical Annual Report, and individual provincial statistical yearbooks. Data for the “Digital Inclusive Finance Index” is sourced from the Peking University Digital Inclusive Finance Index (2011–2022). Any missing data were supplemented via linear interpolation, and economic indicators were adjusted to 2011 as the base year to ensure comparability. Descriptive statistics for each variable are presented in

Table 2. Descriptive statistics of variables.

Variable	Observations	Mean	Standard Deviation	Minimum Value	Maximum Value
Fpr	360	0.460	0.470	0.000	1.000
Dig	360	0.000	0.700	−1.040	3.370
Is	360	0.100	0.050	0.000	0.260
Aml	360	3.340	0.490	1.970	4.130
Fsa	360	0.110	0.030	0.040	0.200
Ur	360	0.600	0.120	0.350	0.900
Sab	360	0.800	0.410	0.210	2.940
Lmi	360	0.370	0.340	0.000	2.550
Kmi	360	0.690	0.910	0.000	5.700

.

5. Empirical Analysis

5.1. Trends in the Level of the Digital Economy and the Resilience of Food Production

The core explanatory variable in this paper, the level of digital economy, is measured using principal component analysis. Similarly, the resilience of food production, the other key variable, is calculated using Formulas (1)–(5), and its evolution over time is analyzed. The results are presented in Figure 2.

Figure 2 illustrates the national trend in the level of digital economy. Overall, China’s digital economy has shown a steady upward trajectory over the study period, rising from −0.4617 in 2011 to 0.2882 in 2022, marking a notable increase. This suggests that while the digital economy has made significant progress in recent years, its overall development level remains relatively low, indicating it is still in the early stages, with considerable potential for further growth.

Figure 2 also shows the changes in food production resilience at the national level. Overall, China’s food production resilience has remained relatively stable throughout the study period, with occasional fluctuations, but the general trend has been one of stability. This reflects the country’s continued emphasis on the development of the “three rural areas” (agriculture, rural areas, and farmers). In recent years, the steady advancement of the rural revitalization strategy, ongoing agricultural modernization, and the gradual implementation of the strategy for building a strong agricultural nation have provided a solid foundation for enhancing food production resilience.

5.2. Baseline Regression

To ensure the reliability of the estimation results, and to avoid multicollinearity among variables in the regression model, a collinearity diagnostic was first conducted for each explanatory variable. The variance inflation factor (VIF) analysis revealed a maximum value of 2.39, which is well below the threshold of 10, indicating that no multicollinearity concerns are present. Using the baseline regression model, control variables were introduced sequentially to estimate the effect of digital economic levels on food production resilience. The results are presented in

Table 3. Baseline regression results.

	(1)	(2)	(3)
	Fpr	Fpr	Fpr
Dig	0.2142 ***	0.0592 ***	0.0827 ***
	(0.0331)	(0.0190)	(0.0292)
Is			−0.2753
			(0.3557)
Aml			0.3419 ***
			(0.1048)
Fsa			−0.4024
			(0.5424)
Ur			−0.5532
			(0.3902)
Sab			0.1128 **
			(0.0455)
Time Fixed Effects	NO	YES	YES
Province Fixed Effects	NO	YES	YES
_cons	0.4582 ***	0.9762 ***	0.6425
	(0.0232)	(0.0395)	(0.4184)
N	360	360	360
R2	0.105	0.966	0.969

Note: ** and *** indicate significance at the 5% and 1% levels, respectively. Robust standard errors are in parentheses.

.

In Table 3, the impact of the digital economy on food production resilience is reported. Column (1) presents the OLS regression results without province and year fixed effects; Column (2) includes bidirectional fixed effects but omits control variables; and Column (3) incorporates both bidirectional fixed effects and control variables. According to the baseline regression results in Table 3, the coefficient of the core explanatory variable, digital economic level, is consistently positive and significant at the 1% level, regardless of the inclusion of control variables and bidirectional fixed effects. These findings suggest that digital economic development significantly enhances food production resilience, indicating that higher levels of digital economic development are associated with stronger food production resilience, thereby supporting Hypothesis 1. Specifically, for every 1 unit increase in the level of digital economy development, food production resilience increases by 0.0827 units. As a typical “integrative” economy, the digital economy, when closely integrated with grain production, maximizes both factor-driven and industry-driven effects. On one hand, data serve as a crucial production factor, incorporated into the production function and deeply interwoven with traditional production inputs. This integration reduces information asymmetry, restructures the factor system, and enhances labor productivity. On the other hand, technological advancements spurred by the digital economy drive the transformation of agricultural industry structures, extending the grain industry chain both vertically and horizontally. This expansion overcomes traditional development limitations and helps mitigate the impacts of internal and external shocks on food production.

5.3. Robustness Tests

(1) Changing the Estimation Method: Different estimation methods can introduce potential biases in model regression outcomes. Since the dependent variable, grain production resilience, ranges between 0 and 1, it fits the criteria for a constrained dependent variable model. Consequently, a panel Tobit model is suitable for this estimation. The results, displayed in Column (1) of

Table 4. Robustness test results.

	(1)	(2)	(3)
	Fpr Change in Estimation Method	Fpr Adjustment of Calculation Method	Fpr Adjustment of Time Window
Dig	0.0864 **	0.4554 ***	0.0710 **
	(0.0359)	(0.1519)	(0.0325)
Is	−0.3503	−0.0865	−0.1001
	(0.5119)	(0.3433)	(0.3938)
Aml	0.3578 ***	0.3419 ***	0.2494 **
	(0.0883)	(0.1051)	(0.1025)
Fsa	−0.4317	−0.2796	−0.6033
	(0.4910)	(0.5515)	(0.5223)
Ur	−0.3855	−0.4309	−0.6651
	(0.3938)	(0.3893)	(0.4630)
Sab	0.1452 ***	0.0998 **	0.1238 **
	(0.0473)	(0.0452)	(0.0553)
Time Fixed Effects	YES	YES	YES
Province Fixed Effects	YES	YES	YES
_cons	0.5576	0.5245	0.9638 **
	(0.3664)	(0.4211)	(0.4780)
N	360	360	300
R2	/	0.969	0.970

Note: ** and *** indicate significance at the 5% and 1% levels, respectively. Robust standard errors are in parentheses.

, show that after switching the estimation method, the coefficient for digital economic development remains significantly positive. This finding indicates that digital economic development continues to promote grain production resilience, consistent with the baseline regression results.

(2) Adjusting the Measurement Approach for Core Explanatory Variables. In previous analyses, the principal component method was employed to assess the level of digital economic development, which was then incorporated into the baseline regression model. To further affirm the robustness of these regression results, the entropy method was subsequently used to calculate a composite score for digital economic development, grounded in the previously established evaluation index system. This composite score was then integrated into the baseline regression model as a robustness check, with the results presented in Column (2) of Table 4. After modifying the measurement approach for the core explanatory variable, the estimated coefficient of digital economic development on the resilience of grain production remains positive and statistically significant at the 1% level, supporting the reliability of the baseline regression outcomes.

(3) Adjusting the Time Window. The onset of the COVID-19 pandemic had profound effects on the Chinese economy. To minimize potential distortions in model estimates arising from abnormal data fluctuations during this period, data from 2020 onwards were excluded based on a review of the relevant literature. The baseline regression model was then re-estimated, and the results are shown in Column (3) of Table 4. Following this adjustment to the time window, the effect of digital economic development on grain production resilience remains significantly positive. Overall, the baseline regression results in this study are credible.

5.4. Endogeneity Test

Reverse causality, measurement errors, and omitted variables between the dependent and independent variables may lead to endogeneity issues in the model, resulting in biased and inconsistent estimates. This study first examines the impact of digital economic development on food production resilience, noting that enhanced resilience in food production could also, in turn, stimulate regional digital infrastructure improvements and broader adoption of digital technologies in the agricultural sector. Second, while the baseline regression controls for provincial and time-fixed effects—mitigating omitted variables related to regional characteristics constant over time and across provinces—certain province-level factors that could influence food production resilience were not included in the control variables, which may contribute to endogeneity concerns. To address these potential endogeneity issues, two methods are employed.

First, the issue of reverse causality. This study substitutes the current level of digital economic development with its lagged term in the regression model to assess the effect of past digital economic development on current food production resilience. This approach assumes that digital economic growth in the prior period influences current resilience, while the latter is unlikely to impact digital economic development retrospectively. The regression results, presented in Column (1) of

Table 5. Endogeneity test results.

	(1)	(2)
	Fpr Lagged 1-Period Dig	Fpr IV
L.Dig	0.0776 ***	0.4298 ***
	(0.0264)	(0.1609)
Is	0.0022	−0.2754
	(0.2966)	(0.4152)
Aml	0.3704 ***	0.4189 ***
	(0.1178)	(0.1092)
Fsa	−0.1371	0.2060
	(0.3617)	(0.6265)
Ur	−0.3402	0.8510
	(0.2882)	(0.7658)
Sab	0.0582 *	0.1933 ***
	(0.0340)	(0.0657)
Time Fixed Effects	YES	YES
Province Fixed Effects	YES	YES
_cons	0.3930	−1.2564
	(0.3171)	(0.9565)
K-P rk LM		10.169 ***
K-P rk Wald F		19.139 > 16.38
N	330	360
R2	0.981	0.957

Note: * and *** indicate significance at the 10% and 1% levels, respectively. Robust standard errors are in parentheses.

, indicate that even when using the lagged digital economic development level, the estimated coefficient remains significantly positive at the 1% level, which is consistent with the baseline findings.

Second, there is the problem of omitted variables. The instrumental variable method is a primary approach for addressing endogeneity issues; however, its application requires selecting an appropriate instrumental variable that meets both the relevance and exogeneity criteria. Drawing on prior research, this study employs an interaction term between the regional fixed telephone density (i.e., the number of fixed telephones per 100 people in 1984) and the previous year’s national information technology service revenue as the instrumental variable for digital economy development levels. The reasons are as follows: on the one hand, digital technology is a continuation and development of traditional communication technology, and the widespread use of fixed-line telephones, as a traditional communication tool, has strongly promoted the development of the digital economy, thereby meeting the relevance requirement for an instrumental variable. Concurrently, while fixed telephones were once the primary communication tool, their significance has diminished with the rise of mobile and internet technologies, rendering them unrelated to contemporary food production resilience—a condition that satisfies the exogeneity requirement. Moreover, as seen from the empirical results in Column (2) of Table 5, the instrumental variable passed the exogeneity test and the weak identification test. Therefore, both theoretically and empirically, the choice of this instrumental variable is reasonable. Additionally, the estimated coefficient of the instrumental variable is significantly positive at the 1% level, indicating that the digital economy can enhance food production resilience. This conclusion is consistent with the results of the baseline regression, supporting the credibility of the findings.

5.5. Mechanism of Action Tests

The previous analysis has established that digital economic development promotes food production resilience. To further investigate the mechanisms through which the digital economy influences food production resilience, this section examines factor allocation as a key lens, informed by prior theoretical insights. Based on existing research, three primary approaches are commonly employed for mechanism testing: mediation effect models, interaction effect models, and direct regression of the explanatory variable on mechanism variables. However, mediation effect models do not adequately address endogeneity issues between mechanism variables and the dependent variable, and interaction effect models may not fully capture underlying economic mechanisms. Consequently, this study adopts the third approach, with the results detailed in

Table 6. Mechanism test results.

	(1)	(2)
	Lmi	Kmi
Dig	−0.2754 **	−0.8103 **
	(0.1178)	(0.3328)
Is	−0.9997	−3.0626
	(1.6542)	(5.7581)
Aml	−0.1567	0.0583
	(0.3503)	(0.6949)
Fsa	−1.1941	−0.3810
	(1.7267)	(4.9216)
Ur	−3.1394 *	−6.0703
	(1.6636)	(3.6975)
Sab	0.0216	−0.1296
	(0.1780)	(0.4363)
Time Fixed Effects	YES	YES
Province Fixed Effects	YES	YES
_cons	3.6678 **	6.8496 **
	(1.7955)	(3.3668)
N	360	360
R2	0.136	0.107

Note: * and ** indicate significance at the 10% and 5% levels, respectively. Robust standard errors are in parentheses.

. Column (1) of Table 6 reports the estimated coefficient of the digital economy on labor factor mismatch, showing a significantly negative coefficient. This suggests that the digital economy mitigates labor factor mismatch, thereby improving labor allocation efficiency. Similarly, Column (2) presents the coefficient of the digital economy on capital factor mismatch, which is also significantly negative, indicating that the digital economy reduces capital factor mismatch and enhances capital allocation efficiency. These results, in conjunction with prior definitions of mechanism variables and theoretical analysis, suggest that digital economic development promotes food production resilience by alleviating labor and capital factor mismatches, thus confirming Hypotheses 2a and 2b.

5.6. Heterogeneity Analysis

(1) Heterogeneity by grain functional zones: Grain functional zones represent recent adjustments to regional, urban–rural, and human–land relationships within China, embodying strategic shifts in national grain production priorities. To assess the differential impact of digital economic development on grain production resilience across these zones, the sample was divided into major and non-major grain-producing areas for regression analysis, with the results shown in Columns (1) and (2) of

Table 7. Results of heterogeneity analysis.

	(1)	(2)	(3)	(4)	(5)
	Fpr Major Grain-Producing Areas	Fpr Non-Major Grain-Producing Areas	Fpr Eastern Region	Fpr Central Region	Fpr Western Region
Dig	0.1726 ***	0.0932 **	0.0084	0.5312 **	0.1755
	(0.0256)	(0.0418)	(0.0172)	(0.2616)	(0.1486)
Is	0.9868 ***	0.3900	−0.3666	−4.5557 ***	0.0455
	(0.1996)	(0.5677)	(0.3850)	(1.6358)	(1.0057)
Aml	−0.1455 ***	0.7437 ***	0.0085	0.1520	0.4751 *
	(0.0337)	(0.1413)	(0.0662)	(0.1704)	(0.2405)
Fsa	1.1945 ***	−1.0962	0.3417	1.6775	−1.4742
	(0.3306)	(0.7846)	(0.4173)	(1.2769)	(1.1307)
Ur	−0.0809	−1.2744 **	−0.5558 *	9.0499 ***	0.8353
	(0.1769)	(0.5508)	(0.3027)	(1.8262)	(0.8866)
Sab	−0.0696 ***	0.2260 ***	0.0405	0.0815	0.3507 **
	(0.0189)	(0.0851)	(0.0365)	(0.0986)	(0.1493)
Time Fixed Effects	YES	YES	YES	YES	YES
Province Fixed Effects	YES	YES	YES	YES	YES
_cons	0.6114 ***	0.3039	1.4416 ***	−3.5832 ***	−2.1831 **
	(0.1923)	(0.5454)	(0.2248)	(1.0192)	(1.0303)
N	156	204	156	72	132
R2	0.958	0.936	0.993	0.961	0.948

Note: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively. Robust standard errors are in parentheses.

. The coefficients for digital economic development are significantly positive in both major and non-major grain-producing areas; however, the positive effect is more pronounced in major grain-producing areas. This may be attributed to the superior agricultural production conditions in these areas, underpinned by their strategic importance, which enables the digital economy to more effectively bolster agricultural resilience. Furthermore, the state’s recent emphasis on food security, through strategies such as “storing grain in the land and technology” has likely enhanced the benefits accrued by major grain-producing areas.

(2) Heterogeneity by geographic location: Due to differences in foundational conditions and resource endowments across provinces, both digital economic development levels and food production resilience exhibit marked heterogeneity. To further explore whether the impact of the digital economy on grain production resilience varies by geographic region, the sample was segmented into eastern, central, and western regions for regression analysis. As shown in Columns (3), (4), and (5) of Table 7, the results indicate that the digital economy exerts the strongest effect on food production resilience in the central region, relative to the eastern and western regions. This may be due to the central region’s pronounced comparative advantage in food production, making it more likely to harness the benefits brought about by digital economic development.

6. Conclusions

Based on panel data from 30 provincial-level regions in China from 2011 to 2022, this paper investigates the effects and mechanisms through which the digital economy enhances food production resilience. Theoretical analysis suggests that the digital economy influences food production resilience through both direct and indirect channels. The empirical findings are as follows:

First, the level of the digital economy has shown steady growth, but significant regional disparities remain. The data reveal that the digital economy index increased from −0.4617 in 2011 to 0.2882 in 2022, highlighting the success of advancements in information infrastructure and technology adoption. However, there is a marked imbalance in development across regions, with the eastern provinces exhibiting far higher levels of digital economy development compared to the central and western regions. The latter still face challenges, particularly in terms of weak digital infrastructure and delayed technology adoption, which hinder further growth in these areas.

Second, while food production resilience has remained relatively stable, there is considerable potential for improvement. This study shows that, despite the stable trend in food production resilience, driven by agricultural modernization and increasing policy support, its growth has not kept pace with the rapid development of the digital economy. This indicates that the current food production system still has significant room to enhance its adaptability and resilience in the face of complex external shocks.

Third, the digital economy has a substantial positive impact on food production resilience. The regression analysis confirms that a 1-unit increase in digital economy development correlates with a 0.0827-point increase in the food production resilience index. This underscores that the digital economy, through the integration of intelligent technologies, data sharing, and supply chain optimization, enhances the stability and risk resilience of the agricultural system, providing critical support for food security.

Fourth, the optimization of factors of production plays a crucial role. The findings reveal that the digital economy indirectly boosts food production resilience by addressing mismatches in labor and capital allocation. Specifically, the digital economy reduces both labor and capital mismatches, thereby improving resource allocation efficiency. The application of digital technologies not only improves the alignment of supply and demand for production factors, but also strengthens the system’s flexibility and adaptability to external shocks.

Fifth, the benefits of digital economy development are unevenly distributed across regions. This study highlights significant regional differences in the impact of the digital economy on food production resilience. In major food production areas, the effect is notably stronger (0.1726) compared to non-major production areas (0.0932). In the central region, the impact coefficient is as high as 0.5312, driven by advantages in agricultural resources and policy support, which far exceed the effects observed in the eastern and western regions. These disparities suggest that regional differences in resource endowment, policy support, and technology adoption result in spatial heterogeneity in the role of the digital economy in enhancing food production resilience.

7. Recommendations and Discussion

7.1. Recommendations

The above research conclusions reveal that the digital economy helps strengthen food production resilience. Moving forward, efforts should be directed toward accelerating digital economic development strategies, enhancing infrastructure, guiding the digital economy to support rural development, and fostering new drivers for growth in the grain industry. Based on these insights, the following policy recommendations are proposed:

First, strengthen digital infrastructure to reduce regional disparities. To address the inadequate digital infrastructure in central and western regions, the government should increase investment in key areas such as 5G networks, the Internet of Things, and data collection systems in rural areas. This will enable the full integration of digital technology into agriculture. Additionally, partnerships between enterprises and local governments should be encouraged to co-develop information infrastructure, ensuring balanced regional development and providing the technical foundation needed for a resilient food production system.

Second, accelerate the digital transformation of agriculture to improve system efficiency. The government should prioritize the promotion of intelligent agricultural technologies, including precision farming, smart machinery, drone monitoring, and smart irrigation. These technologies optimize resource allocation, reduce waste, and enhance efficiency in production. Furthermore, the use of big data and artificial intelligence should be expanded to improve the real-time monitoring and management of the food supply chain, strengthening the stability and resilience of the agricultural system.

Third, optimize the allocation of labor and capital to boost efficiency. By advancing digital inclusive finance, the government can lower financing barriers for agribusinesses and smallholder farmers, addressing the issue of uneven capital distribution. Simultaneously, policies should encourage the flow of skilled labor into high-value agricultural sectors to increase productivity. Digital platforms should be leveraged to connect agricultural producers with labor and capital markets, ensuring an efficient allocation of resources.

Fourth, adopt regionally tailored policies to maximize digital economy benefits. In grain-producing and central regions, the focus should be on promoting intelligent farming equipment and precision planting techniques to further enhance production resilience. In the eastern region, innovation in digital technologies should be prioritized, while in the western region, infrastructure development must be accelerated to bridge the digital divide. These region-specific strategies will help unlock the full potential of the digital economy in enhancing food production resilience.

Fifth, strengthen digital literacy and provide targeted training for agricultural workers. To address the digital skills gap among rural agricultural workers, the government should implement both online and offline training programs. These programs should focus on helping farmers master digital tools for production and management, thus enhancing their ability to adapt to the digital economy and facilitating the modernization and transformation of the agricultural sector.

7.2. Discussion

This paper provides a comprehensive theoretical and empirical analysis of the impact of the digital economy on food production resilience, shedding light on its mechanisms and offering new insights into this emerging research field. The contributions of this paper can be summarized as follows, highlighting how it advances the understanding of the existing literature:

First, this paper enriches the literature on food production resilience by incorporating the evolving context of the digital economy. Unlike previous studies that primarily focus on the stability of food output or supply chain management, this research systematically investigates the relationship between digital economy development and food production resilience. By employing a Markov process model, it quantifies resilience in terms of dynamic stability and recovery capacity, thereby broadening the theoretical framework for measuring and analyzing food production resilience.

Second, this paper delves into the mechanisms through which the digital economy enhances food production resilience, particularly by alleviating mismatches in labor and capital allocation. While prior research emphasizes the role of technological innovation or policy interventions in agricultural development, this paper uncovers how digital technologies optimize resource allocation, thus improving the capacity of food production systems to withstand external shocks. This represents a novel contribution to understanding the interplay between the digital economy and resource efficiency in the agricultural sector.

Third, the regional heterogeneity analysis reveals that the digital economy has the most pronounced impact in food-producing regions and central China. This finding underscores the strategic importance of the central region within China’s food production system, where favorable policy support and abundant agricultural resources amplify the benefits of digital economy development. By linking the effects of the digital economy with regional resource endowments and infrastructure conditions, this paper offers actionable insights for tailoring regional food production policies.

Despite these contributions, this paper has several limitations that warrant further exploration.

First, the measurement of food production resilience primarily focuses on the dynamic changes in food output, which may not fully capture other critical dimensions such as supply chain resilience, market volatility adaptation, and food reserve systems. Since food production resilience is a multidimensional concept, future research could integrate more comprehensive indicator systems that encompass production, distribution, and consumption dynamics, offering a more holistic perspective.

Second, while this paper emphasizes the alleviation of labor and capital mismatches, it does not delve deeply into other potential mechanisms. For instance, the role of digital technologies in improving land-use efficiency, accelerating technology diffusion, or fostering climate adaptation remains underexplored. In light of increasing climate-related challenges, understanding how the digital economy influences crop structure adjustments and climate-resilient practices would provide valuable insights.

Third, this paper focuses on the context of China’s digital economy without engaging in cross-national comparisons. It remains unclear whether the digital economy exhibits greater efficiency gains in developed countries or emphasizes inclusivity in developing nations. Comparative studies could illuminate the applicability and limitations of the digital economy under varying institutional and developmental contexts, offering broader policy implications.

In summary, this paper provides robust theoretical and empirical evidence on how the digital economy enhances food production resilience, offering critical insights for both academics and policymakers. By addressing the intersection of digital innovation and agricultural resilience, this paper contributes to the ongoing discourse on leveraging the digital economy for sustainable food systems. Future research could build upon this foundation by incorporating cross-country comparisons, constructing multidimensional indicator systems, and further investigating the intricate mechanisms at play. Such efforts would advance the understanding of the digital economy’s role in fostering resilience and inform evidence-based strategies for global food security and economic sustainability.