1. Introduction
Land is the most essential element for food security and a core resource for economic development especially for developing countries with large populations such as China [
1]. The fundamental predicament in China is the conflict between people and the land. China has to support up to 20% of the world’s population yet only possesses 9% of the world’s arable land. Since the late 1970s, China has gradually established the household contract responsibility system, realizing the “separation of rights” between the collective ownership of contracted land and the contracted management rights of farming households. The household contract responsibility system mobilized the production enthusiasm of farmers, adapted to the needs of economic development at the time, and addressed the issue of Chinese farmers’ access to food and clothing. However, the fragmentation of farmland caused by the equal distribution of land contracts in rural areas has severely impeded the advancement of China’s agricultural modernization.
Land transfer and moderate-scale operation are necessary to develop modern agriculture, which can optimize land resource allocation and improve productivity [
2,
3]. To promote the sustainable utilization of land and the sustainable development of agriculture, China has continuously improved its land transfer policy. In March 2003, the Rural Land Contracting Law of the People’s Republic of China was formally promulgated, and since then, the transfer of land management rights has entered a period of legal protection and policy support. Since 2013, China has reformed its land ownership system. Based on the household contract responsibility system, China has gradually established a “three-rights” pattern of rural land ownership, land contracting rights, and land management rights, whereby the ownership of rural land belongs to the collectives, while farmers have the right to land contracting and are allowed to transfer the right to land management. Land transfer in this paper refers to the transfer of land management rights. According to the Civil Code of the People’s Republic of China, land management rights holders have the right to occupy rural land for a contractually agreed period of time, to carry out agricultural production and management on their own, and to obtain income.
In recent years under the government’s vigorous promotion, the scale of agricultural land transfer in China has been growing rapidly, and the transfer proportion of farmland increased from 18% in 2011 to 37% in 2017, then decreased to 34% in 2020. However, compared with the urgent demand for intensive land use, the marketability of agricultural land elements in China is not high, and the scale and efficiency of agricultural land transfer still need to be improved. There is an urgent need to further explore new driving forces to promote land transfer [
4].
Some of the literature has explored the factors that drive land management rights transfer, such as personal experience [
5], financial literacy [
6], labor mobility [
7], Internet use [
8], agricultural mechanization [
9], and property rights systems [
10]. However, the literature has not yet focused on the relationship between digital finance and farmland transfer.
Digital finance broadly refers to the use of digital technology by traditional financial institutions and Internet companies to achieve financing, payment, investment, and other new financial business models [
11,
12,
13]. Enhancing financial inclusion is the primary goal of digital finance, a new financial sector created by the fusion of digital technology and traditional conventional financial practices [
14]. In contrast to traditional finance, digital finance can effectively address the issues of challenging risk control and the high cost of traditional financial services and serve a wider customer base, especially for long-tail customers [
15]. As a result, there is an international consensus to enhance financial inclusion with the help of digital technology. The G20 High-Level Principles for Digital Financial Inclusion, released by the G20 Global Partnership for Financial Inclusion (GPFI) in 2016, encourages countries to develop national action plans based on their specific national conditions to realize the huge potential that digital technology has for financial services.
Since 2013, China’s digital finance has developed rapidly and the degree of penetration in the social economy is gradually growing. According to the China Internet Network Information Center’s Statistical Report on the Development Status of China’s Internet Network, as of December 2021, there were 903 million Internet payment users, accounting for 87.6% of the total number of Internet users, and 190 million Internet finance users, accounting for 18.8% of the total number of Internet users in China.
Finance is the core of the modern economy, an important force for economic and social development, and a key resource for microeconomic units to achieve optimal goals. High transaction costs are a key barrier to access to financial services for low-income groups [
16,
17]. Digital finance reduces the cost of using financial services, enhances financial accessibility for rural residents [
18], and provides rural households with more livelihood options, thereby reducing the importance of land-based agricultural production and inducing a reconfiguration of household factors.
A few pieces of the literature have discussed how access to credit affects the willingness and behavior of farmland market participants and found that access to credit is more conducive to households renting out farmland, which provides enlightenment for this study. However, this literature focuses on traditional finance and does not explore the internal mechanism [
19,
20].
Against the backdrop of the rapid growth of digital technologies in most parts of the world, how is digital finance affecting household farmland transfers? Are vulnerable farmers benefiting more? What are the mechanisms of action? Answers to these questions will help to understand the relationship between digital finance and farmland transfer in more depth and provide decision support for promoting the sustainable use of farmland from the perspective of digital finance.
Based on the above considerations, this study analyzes and tests the effect of digital financial use on household farmland transfer-out using data from the 2015 China Household Finance Survey. It finds that digital financial use significantly increases the probability of land transfer-out and the proportion of transfer-out for households and that this effect is greater among households with older heads of household and lower per capita household income and financial accessibility, indicating that digital finance has an important role to play in reducing inequality and promoting inclusive growth. Mechanistic analyses reveal that off-farm employment and information channels are mediating mechanisms for digital financial use to promote household land transfer-out.
The main contributions of this paper are as follows: firstly, previous studies on the influencing factors of agricultural land transfer have neglected digital finance. With the increasing popularity of the concept of digital financial inclusion, the impact of digital finance has become more widespread. Based on large-scale survey data in China, this paper investigates the relationship between digital finance and farmland transfer decision-making, which can provide a new perspective for the study of the driving factors of farmland transfer. Second, this study analyses the mechanisms by which digital finance affects farmland transfer and finds that non-farm employment and information effects are important mediating mechanisms, which complements previous studies in the literature and contributes to a more in-depth understanding of the impact of digital finance. Third, this paper analyses the heterogeneous impacts in terms of age of household head, per capita household income, and financial accessibility, and the findings further confirm the poverty reduction effect of digital finance, providing new evidence that digital finance reduces inequality.
The subsequent part of this paper is organized as follows: the second part conducts the theoretical analysis and presents the research hypothesis; the third part is the research design, introducing the data sources, model setting, and descriptive statistics; the fourth part is the empirical results; the fifth part is the mechanism analysis; and the sixth part is the conclusion.
3. Study Design
3.1. Data Sources
The China Household Finance Survey (CHFS), which was conducted nationally in 2015 by the China Household Finance Survey and Research Center of Southwest University of Finance and Economics, provided the data for the empirical analysis.
The sampling program of the China Household Finance Survey (CHFS) adopts a three-stage stratified sampling design. The primary sampling unit (PSU) is the 2585 cities/counties in the country excluding Tibet, Xinjiang, Inner Mongolia, Hong Kong, and Macao. The second-stage sampling will be conducted by sampling neighborhood/village councils directly from the cities/counties; and finally, households will be sampled from the neighborhood/village councils. Each stage of sampling is implemented using the PPS sampling method, which is weighted by the number of people (or households) in that sampling unit. The 2015 survey covers 29 provinces (autonomous regions and municipalities directly under the central government), 351 districts and counties, and 1396 villages (neighborhood) committees, with a sample size of 37,289 households, and the data are representative of the whole country, as well as of provincial and sub-provincial cities. The China Household Finance Survey collects information on household assets and liabilities, income and expenditure, insurance and protection, household demographic characteristics, and employment.
More importantly, the survey questionnaire in 2015 was designed with rich questions related to households’ digital financial use and land transfer, providing data support for this paper’s study (Up to now, the public has access to the survey data from 2011, 2013, 2015, 2017, and 2019; however, the surveys from 2017 and 2019 did not ask about a land transfer, and the surveys from 2011 and 2013 did not ask about digital finance). Since agricultural land transfer involves rural households, this paper’s study only retains the sample of households whose head is an agricultural household registration, and after deleting the missing values of relevant variables, a total of 11,802 valid household samples are obtained.
3.2. Model
Since the dependent variable in this study, land transfer, is a binary dummy variable, using OLS regression would violate the Gaussian Markov assumptions and lead to heteroskedasticity. Therefore, the Probit model is set up in this work to examine the effects of digital finance on household land transfer:
where
denotes whether households transfer out their farmland management rights and take the value of 1 if they transfer out, and 0 otherwise;
denotes whether they use digital finance (use = 1, no use = 0);
represents the control variable;
is the random error term.
Since the land transfer proportion variable is truncated, linear regression of the entire sample using OLS leads to inconsistent estimates. Therefore, the following Tobit model is used in this paper to investigate the effect of digital finance on land transfer proportion:
where
denotes land transfer proportion,
denotes the observed value of land transfer proportion, and the rest of the variables are the same as in model (1).
3.3. Variables
3.3.1. Land Transfer and Land Transfer Proportion
Two explanatory variables, namely land transfer (whether transfer-out = 1, not transfer-out = 0) and land transfer proportion (the area of farmland transfer-out/household farmland area), are set in this paper to properly assess the farmland transfer-out behavior of households.
3.3.2. Digital Finance
Compared with the macro digital finance index at the regional level, the digital finance index constructed from the standpoint of digital finance product use based on micro survey data can reflect the penetration of digital finance in a more detailed and accurate way. Therefore, this paper refers to Song et al. (2020) to measure digital financial use in terms of payment and financial investment [
24]. If households use online banking or mobile banking and invest in Internet wealth management, they are considered to participate in digital finance and are assigned a value of 1, otherwise, they are assigned a value of 0. In addition, this paper also uses digital financial intensity as a substitute for digital finance in the robustness test section.
3.3.3. Mechanism Variables
For the non-farm employment mechanism, the non-farm employment variable is generated based on the number of non-farm employment among household members. Two variables are included in the information channel mechanism: information use and information attention. Two variables are included in the information channel mechanism: information use and information attention. The former is generated based on whether households use smartphones to access information, and the latter is generated based on the degree to which households pay attention to financial and economic information, reflecting households’ active access to information.
3.3.4. Control Variables
The control variables include age, gender, years of education, marital status of the household head, household size, financial literacy, whether or not the household has signed a land contract, whether or not the household has obtained a certificate of land contracting rights, the household off-farm assets and its disposable off-farm income, and control for provincial fixed effects. This paper conducted a two-sided 1% tail-shrinking treatment for continuous variables such as off-farm assets and disposable off-farm income to prevent the impact of extreme values on the regression findings.
Table 1 displays specific variable definitions.
The descriptive statistics for each variable are shown in
Table 2. In the full sample, 18% of rural households transferred out of farmland management rights, and the mean land transfer proportion was 0.142, which was relatively low. Digital finance was used by 9.3% of rural households, and the mean value of digital financial intensity was 0.146, indicating that the penetration of digital finance was low in rural areas.
Figure 2 displays scatter plots of household land transfer (left panel) and Land transfer proportion (right panel) along with digital finance, where digital finance, land transfer, and land transfer proportion are provincial-level means.
Figure 2 shows that both land transfer and land transfer proportion have a stronger positive relationship with digital finance, which at first glance indicates that digital finance use has a positive impact on farmland transfer-out.
5. Conclusions and Implications
Based on the 2015 China Household Finance Survey (CHFS) data, this paper empirically analyzes the effects and mechanisms of digital finance on household farmland transfers. The findings show that digital finance significantly increases the probability and proportion of farmland transfer out for rural households, and this positive effect is more pronounced among rural households with older household heads and lower per capita income and financial accessibility, indicating that digital finance has an important role to play in reducing inequality and promoting inclusive growth. Off-farm employment and information channels are mediating mechanisms for digital finance to facilitate farmland transfer. Further studies of off-farm employment demonstrate that digital finance increases employed employment and entrepreneurship and decreases temporary employment among rural households. Specifically, on the one hand, the financial function of digital finance increases the share of employment and entrepreneurship among rural households. In terms of industry and skill type, digital finance promotes the entry of farmers into tertiary employment, facilitates off-farm employment for low and medium-skilled farmers, and has no impact on the employment of high-skilled farmers. On the other hand, the information function accompanying digital finance broadens households’ access to information, both of which have a favorable effect on farmland transfer-out.
The findings of this study contribute to an in-depth understanding of the intrinsic relationship between the use of digital finance and farmland transfer and have important policy implications for improving the efficiency of farmland management and expanding the marketability of farmland markets. To give full play to the support effect of digital finance on farmland transfer, the following recommendations are made:
First, we should accelerate the construction of the “digital countryside” and improve the foundation for digital financial development. Compared with urban areas, information infrastructure construction in rural areas of China is relatively weak. Government departments should increase financial investment to improve the Internet penetration rate in rural areas and continuously explore the digital application of rural industrial operations.
Second, popularize digital finance education and improve the digital financial literacy of rural residents. A high level of digital financial literacy helps rural households access financial information and use financial services, thus optimizing households’ financial decisions and promoting income growth. The digital financial literacy of China’s rural residents is inherently low, and the aging and hollowing out of the rural population due to urbanization further exacerbates the issue. It is the responsibility of the government as well as various rural financial institutions to increase investment in financial education and adopt various forms to popularize digital financial literacy in the context of rural production and business scenarios. For special groups such as the rural elderly, digital financial APPs such as the rural version and the elderly-friendly version have been developed to simplify the operating interface and reduce the difficulty of operation. At the same time, manual, remote, and door-to-door services for rural elderly groups have been strengthened to improve the convenience of financial services for elderly groups. Training in digital financial knowledge and skills for the elderly has been increased, guiding the elderly to actively integrate into digital life, improving their ability to use digital financial products and services to improve their own lives, and gradually eliminating the digital divide and financial exclusion.
Third, rural financial institutions need to accelerate digital transformation and increase the supply of digital financial services. By introducing and absorbing emerging technologies such as big data and artificial intelligence, they can optimize the risk control model and effectively control credit risks. Combined with the characteristics of rural financial businesses, the digital transformation of outlets is carried out to reduce the cost of financial services. Based on stabilizing traditional advantageous products and accelerating financial innovation, rural financial institutions actively explore digital financial products that fit rural production and business scenarios.
Finally, it is necessary to promote the digitalization of the tertiary industry and actively use digital technology to cultivate new forms and modes of service industries, thereby creating more demand for services and giving full play to the employment function of the tertiary industry. Government departments need to break down barriers to the employment of farmers and create a favorable employment environment, thereby promoting the off-farm employment of farmers.
Due to data limitations, the digital finance indicators in this study include only three categories: mobile banking, online banking, and Internet wealth management. The connotation of digital finance is much more than that, and more digital finance content can be included in the future to measure digital finance indicators more precisely. In addition, the impact of digital finance on other areas, such as farmland management, farmland rent, and employment of farm households, could be further explored in the future.