Rural Digital Economy

The digital economy encompasses a range of economic activities that rely on data resources as primary production factors and modern information networks as key carriers. This sector is driven by the effective use of information and communication technologies, which play a vital role in improving efficiency and optimizing economic structures [1]. Furthermore, the digital economy contributes to rural resource allocation, innovation, entrepreneurship, and carbon emission reduction [2]. Firstly, the rural digital economy facilitates the revival of rural industries by optimally allocating and utilizing rural resources, effectively connecting urban and rural markets, and integrating and developing industries [3]. Secondly, the growth of the digital economy can facilitate rural regeneration, bridge the ‘digital divide’ between urban and rural development, and provide a strong foundation for shaping a digital urban-rural integration development pattern [4]. Drones, Internet of Things sensors, cloud computing, and big data analysis represent some of the most advantageous technologies in agriculture [5]. This technology effectively enhances the rural e-commerce sector of China [6], improves family members’ technological competence, and boosts non-farm employment. Assisting rural families in generating wealth and boosting income [7], it has emerged as an essential propelling force for rural low-carbon development [8]. The implementation of the Internet+agricultural supply chain model [9] has been proven to be effective and has led to the advancement of new industries and business models in rural areas. Additionally, this approach has significantly elevated the innovation level of rural public services. Digital finance simultaneously promotes the upgrading of farmers’ consumption and economic growth [10]. The development of digital inclusive finance establishes a robust rural financial ecosystem, enables the integration of the rural three industries [11], reduces the probability of households experiencing relative poverty, encourages entrepreneurial activities of farmers [12], boosts employment rates [13], and enhances the entrepreneurial environment for farmers.

Agricultural Carbon Emissions

China’s agricultural sector is responsible for 17% of the world’s total carbon emissions. Despite this, it still faces significant challenges. The main factors influencing agricultural greenhouse gas emissions are the total power of agricultural machinery, the amount of agricultural diesel fuel and the amount of fertilizer applied. Additionally, agricultural carbon emissions are significantly impacted by the industrial structure, the level of agricultural economy, the demographic structure, agricultural financial support and the degree of agricultural disaster [14]. Studies have shown that agricultural carbon emissions can be effectively reduced through breakthroughs in agricultural energy efficiency, non-agricultural employment, promotion of agricultural industrial structure upgrading [15], expansion of renewable energy [16], development of digital finance [17], and promotion of agricultural technological progress [14]. This can lead to a reduction in the total amount, intensity, and density of agricultural carbon emissions, as well as a decrease in carbon emissions generated by fertilizer, pesticide, agricultural film, and machinery fuel in the production process [14]. By considering the benefits of carbon emissions in agriculture and addressing the issue of spillover internalization in low-carbon agricultural production, it is possible to increase economic benefits for farmers while reducing agricultural carbon emissions.

Agricultural Green Technology Innovation

Green technology plays a crucial role in addressing China’s latest technological advancements and ecological issues. In agriculture, it serves as a significant impetus for overall eco-friendly transformation and development while improving resource allocation efficiency and reducing carbon emissions [18]. According to the ‘Green List of International Patent Classification’ published by the World Intellectual Property Organization, green agriculture encompasses alternate irrigation technologies, pesticide substitutes, soil improvement, and organic fertilizers made from waste. By implementing environmental regulations and investing in government research and development, along with increasing gross agricultural output and productivity, they can significantly enhance green agricultural innovations [19]. This can improve energy efficiency, reduce environmental pollution caused by agriculture, lower pollutant emissions and effectively promote the growth of green, healthy and sustainable agriculture in the country. Additionally, it can help stimulate soil organic carbon sequestration functions, thereby promoting agricultural production goal of creating a high-carbon sink [20].

Research Gaps in the Existing Literature

The influence of the digital economy and carbon emissions on urban and industrial areas has attracted attention from many scholars, but the impact of the rural digital economy on agricultural carbon emissions and green technological innovation is fragmented. In this study, we explored the impact on agricultural carbon emissions and the transmission mechanism of agricultural green technological innovation. This paper presented a rural-focused index for developing the digital economy, quantified agricultural carbon emissions, defined the boundaries of agricultural green technology innovation, and examined the transmission and consequences of rural digital economy, agricultural green technology innovation, and agricultural carbon emissions through empirical analysis. This would enhance the use of agricultural resources in China’s rural digital economy development and decrease agricultural carbon emissions.

Low Carbon Economy Theory

This thesis analyzes the internal mechanism of the rural digital economy to promote technological innovation and reduce carbon emissions, based on the theory of the low-carbon economy. The theory of low-carbon economics aims to reduce greenhouse gas emissions as its core objective, achieving a balance between economic development and environmental protection. Promoting, operating, and developing the digital economy can unleash innovation and endogenous power in technology, data, and management, based on the principles of low-carbon economics. This can lead to optimized resource and energy efficiency, reduced energy consumption, and decreased energy and resource consumption per unit of output [21]. The outcome is a mutually beneficial scenario for both economic growth and the reduction of carbon emissions, effectively and comprehensively enhancing carbon emission efficiency.

Rural Digital Economy and Agricultural Carbon Emissions

The advancement of the rural digital economy facilitates the digital industrialization of agriculture and the digitization of the agricultural industry by upgrading rural digital infrastructure and constructing a comprehensive industry chain for environmentally friendly production, distribution and consumption. The initial impact of the rural digital economy is the direct boost it provides to low-carbon agricultural development. Firstly, the aim is to encourage the adoption of environmentally friendly and sustainable alternatives in agricultural production methods, with the goal of reducing greenhouse gas emissions and minimizing environmental pollution. Secondly, it accelerates the low-carbon transition of farmers’ lifestyles by using clean energy and minimizing resource waste. Thirdly, it enhances the quality and effectiveness of government oversight, regulates conduct, improves efficiency, and mitigates carbon emissions from agriculture. The digital economy in rural areas utilizes digital technologies such as the Internet, blockchain, big data, and artificial intelligence, to support agriculture. Primarily, it aids in the transformation and modernization of conventional agriculture, highlights the efficacy of scientific advancements, and emphasizes industry advancements. Secondly, it facilitates the efficient allocation of agricultural production resources and enhances agricultural production structure and the framework for agricultural supply. Thirdly, it facilitates the growth of new agricultural sectors, establishes an online and big data platform for distinctive agricultural sectors, and subsequently advances agricultural industrial restructuring. This helps to reduce the prevalence of polluted agriculture, enhance energy utilization efficiency, and mitigate agricultural carbon emissions [22].

Hypothesis H1 suggests that the development of rural digital economy helps to reduce agricultural carbon emissions.

Mechanisms of Green Technology Innovation in Agriculture

Green technological innovation is an essential means for the digital economy to achieve high-quality economic development and upgrade industrial structure [23]. It is crucial to building an environmentally friendly society and achieving the “dual-carbon” goal provides a new perspective [24]. Furthermore, promoting the development of green technology in agriculture is essential to achieve low-carbon and environmentally sustainable agricultural practices, as well as to advance agricultural science and technology towards modernization. The digital transformation of agriculture has a significant impact on promoting agricultural green growth, primarily through green technological innovations [25]. Firstly, innovations can enhance the total factor productivity of agriculture through specialization, scaling, and intensification [26], resulting in reduced costs of agricultural innovation factors. Secondly, they can strengthen research and development of agricultural innovation knowledge, broaden market demand, improve the efficiency of agricultural production, and promote sustainable agricultural development, ultimately improving the quality of agricultural products. Thirdly, the aim is to enhance the diffusion of agricultural innovation knowledge, elevate the scientific and technological culture of farmers, and increase the yield of innovative outcomes. Promoting the acceptance of environmentally friendly products by consumers [24]. The primary route to reduce agricultural carbon emissions in the long run is through the advancement of ‘clean’ agricultural technologies [27-30]. On one hand, the task involves innovating agricultural green technologies to increase energy efficiency, reduce energy consumption, and promote the adoption of renewable energy. The second aspect involves using clean energy, promoting the recycling of agricultural resources, supporting the development of green agriculture, and achieving agricultural emission reduction [31]. Upon synthesizing the literature outlined above, it is evident that the rural digital economy has emerged as a new driving force in modern agriculture. It facilitates low-carbon agricultural development, here achieved through agricultural green technology innovation.

Accordingly, we propose hypothesis H2 that the rural digital economy reduces agricultural carbon emissions through the innovation of agricultural green technology.

Modelling Design

Testing the Model of the Role Mechanism of Agricultural Green Technology Innovation

This study investigates the impact of agricultural green technology innovation on agricultural carbon emissions in the rural digital economy. The findings suggest that the technological advancements play a significant role in reducing agricultural carbon emissions. This paper aims to explore the relationship between the rural digital economy and agricultural carbon emissions. To address the endogeneity issues in the third step of the conventional three-step mediation effect test, a more precise mediation effect model needs to be developed for further empirical examinations. Based on the research conducted by Guo Feng [33], a more prescriptive model should be constructed. This study uses a conventional three-step mediation effect test, focusing only on the second step, to investigate the impact of the rural digital economy and agricultural green technological innovation mechanism. Based on literature and logical reasoning, the study explains how agricultural green technology innovation mediates the relationship between the rural digital economy and agricultural carbon emissions. The basic model is described below:

The given recursive equation represents the mediating variable M_it of agricultural green technological innovation (AGTI), which includes the number of agricultural green patent applications (NOAGPA1) and authorizations (NOAGPA2). The paper focuses on the coefficient is β₁, and if β₁ is significantly positive coefficient, it indicates that the growth of the rural digital economy can effectively enhance the level of innovation in agricultural green technology.

Description of Variable Selection and Data Sources

Core Explanatory Variables

Construction of the rural digital economy indicator system. Based on the findings of Qi Jiang and colleagues [25], a comprehensive rural digital economy indicator system will be developed. This system aims to measure the level of digital economy development in rural areas based on three main indicators: the status of rural digital infrastructure development, the digitalization of agriculture, and the digitalization of the agricultural industry, consisting of eight secondary indicators. Details of the specific indicators can be found in Table 2.

Choice of measurement method: Entropy value method. In order to accurately measure the level of development of the rural digital economy, we need to determine the weights of the specific indicators that make up the digital economy. Based on the research of Guo Feng et al. [33], we have selected the entropy value method of the objective allocation method. This method calculates the weights of the constructed rural digital economy indicators, resulting in an accurate index for measuring its level of development.

The digital economy indicators for rural areas presented in this paper are standardized and have positive and negative values. The calculation methodology is as follows:

Positive indicators:

Negative indicators:

max{x_j} is the maximum value of the j evaluation indicator for all years, min{x_j} is the minimum value of the j evaluation indicator for all years. x_ij is the value of the jth evaluation indicator in year j, dimensionless results. The objective weights of each evaluation indicator were calculated using the procedures of the entropy method.

Calculation of evaluation indicator j’s share in year i:

Calculation of information entropy of evaluation indicators:

Calculating information entropy redundancy:

m is the evaluation year.

Calculation of indicator weights using information entropy redundancy:

Based on the standardized indicators and indicator weights obtained, the level of the rural digital economic development index is calculated via weighted multiple linear functions:

Where ‘RDE_i’ represents the level of the composite index for the development of digital economy in rural areas of province ‘i’, between 0 and 1. A higher value of ‘RDE_i’ indicates a greater degree of development in the rural digital economy, while a smaller value of ‘RDE_i’ corresponds to a lower level of development.

Control Variable

In a truly objective environment, the intensity of agricultural carbon emissions is determined by the synergistic effect of multiple variables. To systematically investigate the impact of the rural digital economy on these emissions, we built on the work of Liu Songqi et al. [39] and added five control variables: rural income level (RIL), rural human capital (RHC), crop disaster severity (EOCD), agricultural industrial structure (AIS) and urbanization level (UL). We obtained data from the China Rural Statistical Yearbook and the China Population and Employment Statistical Yearbook. In the empirical process, to avoid the influence of extreme values and effectively satisfy the classical linear assumption, we adopted a logarithmic treatment for the intensity of agricultural carbon emission intensity, the rural digital economy index, and the control variables. Some missing data were interpolated.

Intermediary Variable

Scholars have used various measures of innovation, such as sales of new products and patents. This paper argues that patent data is a more appropriate indicator of the technological aspects of agricultural innovation, especially when combined with the mediating variable of agricultural green technology innovation. In addition, patents are better suited to integrate technological innovation within the scope of green agriculture. Currently, there is no clear division of agricultural patents within the International Patent Classification (IPC), and no unified standard for IPC classification of agricultural patents has been developed. This paper refers to the study by Liu Lijun et al. [40], which is based on the range of IPC classification numbers of agricultural patents and incorporates the ‘International Patent Classification Green List’ published by the World Intellectual Property Organization.

The level of innovation in agricultural green technology in each province was measured by the number of agricultural green patent applications (NOAGPA1) and the number of agricultural green patents granted (NOAGPA2). The data on agricultural green patents in this paper only include agricultural green inventions and agricultural green utility models. This is because it is impossible to identify the sector to which the patent classification of the design corresponds. For the purposes of this study, it is necessary to start from an agricultural perspective, hence the exclusion of design patents. Table 3 shows the descriptive statistics of the variables.

Regression to Baseline

Empirical data on the effect of the rural digital economy on agricultural carbon emissions using empirical data is presented in Table 4. To ensure comparability with other studies, this report includes regression results for both random effects and double fixed effects models. Column 1 excludes control variables and excludes time and area effects. Column 2 omits control variables but considers fixed effects for time and area. Column 3 includes control variables but still does not adjust for fixed effects related to time and area. Finally, column 4 takes into account control variables and fixed effects for time and area. The regression analyses indicate a significantly adverse impact of the rural digital economy on agricultural carbon emissions, suggesting that the rural digital economy helps to reduce such emissions. In model (4), the coefficient for the digital economy is -0.133, which is significant at the 1% level. This suggests that for every 1% increase in the level of the rural digital economy, the average intensity of agricultural carbon emissions decreases by 0.133%. The results indicate that rural digital economy helps to reduce agricultural carbon emissions. The coefficients for the rural digital economy have a significant negative impact on both random and fixed effects. This supports hypothesis H1, indicating that the growth of rural digital economy contributes to the reduction of carbon emissions in agriculture. Possible improved version: There are potential reasons to support the role of rural digital economy in mitigating carbon emissions from agriculture. Firstly, it could encourage agricultural economic growth and facilitate the green transformation and upgrading of agricultural businesses. Secondly, rural digital infrastructure should be implemented to support rural agriculture and aid in its development, driving the adoption and improvement of agricultural technology and surpassing current limitations. This includes measures to improve the rural Internet penetration rate, as well as the adoption of information technology in agricultural machinery and equipment, enhancement of rural logistics infrastructure, and overall development of the e-commerce environment. With these improvements, the development of the rural economy can be accelerated and carbon emissions in agriculture can be reduced. Thirdly, the rapid expansion of the rural digital economy has gradually changed farmers’ perceptions of production and marketing. This has led to the adoption of online practices as a new trend and has increased farmers’ awareness of ecological conservation, effectively reducing carbon emissions from agricultural activities.

Of the control variables, it was found that rural income and urbanization levels had a significantly negative impact on agricultural carbon emissions. The adoption of low-carbon development techniques in farming and the reduction of agricultural carbon emissions may be influenced by two factors. Firstly, farmers with higher disposable income are more likely to adopt these techniques, which is suggested by the rural income level. Secondly, the level of urbanization reflects the degree of influence on agriculture in the process of industrialized technological progress. As people in cities actively seek environmentally friendly, pollution-free, and high-quality agricultural products, farmers may need to adjust their production techniques, leading to a reduction in agricultural carbon emissions.

Robustness Analyses

Regional Heterogeneity

China’s vast expanse results in significant variations in factor endowments, the resource environment and levels of economic development between regions. As a result, rural digital economy development and agricultural green technology innovation vary throughout the country, leading to regional heterogeneity. To verify the hypothesis that regional variations in the rural digital economy impact agricultural carbon emissions, this study will reference ‘Several Opinions of the Central Committee of the Communist Party of China and the State Council on the Promotion of the Rise of the Central Region’ and ‘The State Council of China Issued the Implementing Opinions on Several Policies and Measures for the Development of the Western Region’ Based on the 16th National Congress of the Communist Party of China’s report, the provinces have been divided into four regions: the East, the Central, the West and the Northeast¹ [41].This categorization enables regression analysis to explore the impact of regional differences in rural digital economy development on agricultural carbon emissions.

The results indicated that the regression analysis of the rural digital economy in the eastern and central regions was not significant (Table 7). However, the coefficients were negative, suggesting that the growth of the rural digital economy led to a decrease in agricultural carbon emissions in these areas. Conversely, the regression coefficients of rural digital economy in the western and northeastern regions were significantly negative at the 5% and 10% levels, respectively. These results suggest that the expansion of the rural digital economy in these regions led to a reduction ¹ The eastern region includes Beijing, Tianjin, Hebei, Shang-

hai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong, and

Hainan; the central region includes Shanxi, Anhui, Jiangxi,

Henan, Hubei, and Hunan; the western region includes Inner

Mongolia, Guangxi, Chongqing, Sichuan, Guizhou, Yunnan,

Shaanxi, Gansu, Qinghai, Ningxia, and Xinjiang; the north-

eastern region includes Liaoning, Jilin, and Heilongjiang. in agricultural carbon emissions. Possible explanations are: The rural digital economy in the eastern and central regions of the country has progressed at a faster rate, resulting in a higher overall level of rural digital economy and improved digital infrastructure development. Furthermore, there is a greater degree of integration and development with agriculture, including a higher utilization rate of emerging information technologies such as the Internet of Things, cloud computing, big data, and mobile internet. A significant number of scientific and technological experts collaborate to optimize the industrial structure, resulting in a reduction of carbon emissions in agriculture. It is important to note, however, that the primary factor influencing agricultural carbon emission reduction is not the rural digital economy. Although the expansion of the digital economy in rural regions of the west is currently in its nascent stage, it is rapidly gaining momentum with vast potential for growth. The digital sector is playing a crucial role in reducing carbon emissions from agriculture. The agricultural sector in the Northeast region of the UK is sizeable and primarily characterized by traditional production methods. The industry structure is relatively homogenous, with a high level of dispersion in agricultural operations and limited uptake of technological innovations, resulting in lower economic returns. Although the rural digital economy is still in its early stages, recent advancements in rural digital infrastructure, digital finance, and the digitization of agriculture have resulted in significant progress. This has led to increased integration and development of the rural digital economy, resulting in a substantial reduction in carbon emissions attributable to agriculture.

Heterogeneity in the Level of Development of the Rural Digital Economy

A regression test was conducted to account for the differences in agricultural economic development, rural digital economy, and technological innovation among the provinces in China. The provinces were divided into two categories based on their 2021 ranking of rural digital economy development: high(provinces ranked 1-15) and low(provinces ranked 16-30)². The regression coefficients of the rural digital economy for high-level provinces are significantly negative at the 10% level, while those for low-level provinces are significantly negative at the 1% level. Please refer to columns (5-6) of Table 7 for the presented results. Meanwhile, the rural digital economy in rural areas of more developed provinces has a higher level of development and agricultural activity, which has a comparatively smaller impact on agricultural carbon emissions.

Examining the Role Mechanism of Agricultural Green Technology Innovation

The regression findings for the rural digital economy mechanism and its effect on agricultural green technology innovation are presented in Table 8. The explanatory variable number used is the number of agricultural green patent applications. The results show a statistically significant positive regression coefficient for the rural digital economy at the 1% level (Table 8 (columns 1-2)). Similarly, it shows that the number of agricultural green patent grants has a significant positive impact on the rural digital economy at the 1% level of significance. This suggests that there is a positive correlation between the growth of the rural digital economy and the level of agricultural green technological innovation (Table 8 (columns 3-4)). These findings are consistent with the results presented in columns (3) and (4) of Table 8. Increasing the lag of the rural digital economy by one period when testing the number of authorized agricultural green patents results in significantly positive coefficients of rural digital economy at a 5% level in column (5) of Table 8 due to a longer time lag effect in the authorization of agricultural green patents. Furthermore, the development of rural digital economy significantly improves the level of green technological innovation in agriculture in the second year. The above demonstrates that the growth of digital economies in rural areas leads to increased innovation in green agricultural technology.

Technological advances are crucial in reducing carbon emissions in agriculture [28, 42]. However, not all forms of technological progress contribute to this reduction. Only ‘clean’ agricultural technology leads to a decrease in carbon emissions [27-30, 43]. Agricultural green technology is one such ‘clean’ technology that can effectively reduce environmental pollution, energy, and raw material consumption when compared to traditional agricultural production techniques [44]. The implementation of innovative green agricultural technologies in agricultural production can reduce or modify the input factors, decrease pollution from surface sources, improve energy efficiency, lower production costs, increase productivity, and reduce agricultural carbon emissions. Secondly, the implementation of green agricultural technology innovations such as new agricultural varieties and the promotion of the green revolution [45], can effectively enhance the transformation and upgrading of the agricultural industry structure. The objective of this approach is to implement environmentally friendly production practices throughout all stages of agricultural product development, including production, processing, packaging, cold chain storage, sales, and beyond. This will enable the creation of a green production model, green supply chain, green logistics, green consumption, and agricultural carbon emission reductions. Technical abbreviations will be explained upon first use. Thirdly, prioritize the integration of agricultural green technology across various sectors such as planting, animal husbandry, fisheries and seed industry, through industrial links. This approach enables the full utilization of the spillover effects of agricultural green technology and fosters interconnectivity and knowledge sharing between neighboring provinces to achieve carbon emission reduction in the agricultural sector. The rural digital economy has the potential to contribute to reducing agricultural carbon emissions through the agricultural ‘technology effect’. Agricultural green technology innovation acts as an intermediary mechanism for carbon emission reduction and is of paramount importance [46]. The development of rural digital economy has a significant positive impact on the reduction of agricultural carbon emissions by enhancing agricultural green technology innovation. This supports hypothesis 2, which suggests that implementing agricultural green technology innovation in rural digital economies can reduce agricultural carbon emissions.

Conclusions

This study analyzes the impact of the rural digital economy on agricultural carbon emissions in China between 2012 and 2021, using panel data collected from 30 provinces. Empirical methods were employed to investigate the relationship between the two variables. The results indicate a significant reduction in emissions, (Table 4). Following robustness tests, such as variable replacements and extended timescales (Tables 5-6), the impact remains consistently significant. Heterogeneity analyses indicate that the impact of the rural digital economy has a more pronounced effect on reducing agricultural carbon emissions in the western and northeastern regions, while it is insignificant in both the eastern and central regions (Table 7). Furthermore, the effect of the rural digital economy on reducing agricultural carbon emissions is more pronounced in provinces with lower levels of rural digital economy development than in those with higher levels of rural digital economy development (Table 7). The mediation process indicates that the development of the rural digital economy improves the level of agricultural green technological innovation and reduces agricultural carbon emissions by the use of agricultural green technology. This is evidenced by the number of agricultural green patent applications and agricultural green patents (Table 8).

Recommendations