A Mechanistic Study of the Impact of Digital Payments on Rural Household Development Resilience

Abstract

China has made advancements in the achievement of the poverty relief goal set by the 2030 Agenda for Sustainable Development of the United Nations. However, the rapidly changing socioeconomic environment and the increasingly growing complication of poverty in rural households make it necessary to enhance rural household development resilience to ensure poverty alleviation. With the rapid development of Internet technology, this paper assesses whether digital payments influence rural household development resilience. Based on it, this study takes rural China as its research area, selects samples of farmers’ household registration as its research object, uses the data recorded by China Household Finance Survey (CHFS) in 2015, 2017, and 2019, and constructs the index of “rural household development resilience” based on the family wellbeing model to empirically test the impact of rural households using mobile payments on development resilience. By considering the nonlinear dynamics of event development, this paper addresses the short-term effects of the previous exclusive examination of digital payments, providing a theoretical basis for future research on the relationship between digital payments and development resilience. The results show the following: (1) digital payments can significantly enhance rural household development resilience; (2) relaxing liquidity constraints and promoting farmer-oriented market participation are the primary mechanisms by which digital payments affect rural household development resilience, but digital payments do not have an impact on rural household development resilience by releasing credit constraints; and (3) the heterogeneity regression analysis showed that the promotion effect of digital payments on rural household development resilience is the strongest in the Western areas, at medium level in the Central areas, and the weakest in the Eastern areas. Thus, the influence of digital payments is not significant to rural household development resilience. Based on the conclusions mentioned above, this article proposes countermeasures and suggestions.

1. Introduction

1.1. Research Background and Problem Formulation

Issues relating to agriculture, rural areas, and farmers have always been preoccupations of various countries. The key to solving these problems lies in eradicating poverty and promoting sustainable development in rural areas. As the world’s largest developing country, China has historically eliminated absolute poverty and made advancements in the achievement of the poverty relief goal set by the 2030 Agenda for Sustainable Development of the United Nations under the leadership of the Party and the struggle of the people. However, poverty alleviation does not mean the disappearance of impoverished households. The data have shown that nearly two million people who have been lifted out of poverty in China are still at risk of falling back into poverty, and nearly three million highly vulnerable groups are at risk of returning to a position of impoverishment [1]. The report of the 20th National Congress of the Communist Party of China said that they should “ensure and improve the people’s well-being in the course of pursuing development and encourage everyone to work hard together to meet the people’s aspirations for a better life” and “make public services more balanced and accessible, so as to achieve solid progress in promoting common prosperity”. At this important starting stage of building China into a modern socialist country in all respects, the agenda for China involves how to maintain the bottom line to prevent poverty from returning to ease the poverty vulnerability of marginalized groups [2] and further promote the long-term mechanism for the sustainable development of rural households.

To explore issues related to the sustainable development of rural households, the first step is to determine evaluation criteria. Barrett and Constas [3] constructed the index of “development resilience” based on the theory of poverty trap and fragility, which reflects on how subjects build resilience to resist risks and self-adjust to accommodate the new development and cultivate the ability to achieve sustainable development via alternate paths when encountering various stressors and disturbances. This article continues to use the index of “rural household development resilience”. When it increases, not only can families improve their quality of life, but they are able to respond to emergencies and make poverty impossible. At present, the social and economic environment is in constant variation, and the internal and external environment is increasingly complex. The ability of rural households to effectively prevent and diminish short-term external risks (economic crises and public health safety incidents) and long-term internal risks (disposable income, consumption expenditure, education, and investment) and stimulate, create, and exert endogenous development capabilities is a key factor in enhancing their development resilience. In this context, how to enhance rural household development resilience and promote the sustainability of poverty relief is not only of economic and practical significance to consolidate and expand the achievements of poverty alleviation and establish a long-term mechanism for a smooth transition from poverty alleviation to rural revitalization but also a necessary prerequisite for China to achieve common prosperity in the near future.

However, the weakness of agriculture and the vulnerability of rural households limit their output market participation, internal livelihood capital, and external development capital. The imbalance between urban and rural resources greatly constrains the growth of rural household development resilience. With the arrival of the third technological revolution, the digital era has entered people’s horizons [4]. The digital economy is the main economic form after agricultural and industrial economies, endowing new fields and tracks for economic and social development and shaping new momentum and advantages. As an important engine in the financial technology field, digital payments have gained a large number of users with their convenience, low costs, and efficiency [5]. Not only does this provide convenient payment methods for rural households, but it also creates corresponding business opportunities for rural producers, bridges the urban–rural “digital divide”, and forms a “transfer payments dividend” for economic development [6]. Digital payments also play a significant positive role in rural family decision-making on consumption and entrepreneurship [7], which can effectively enhance the expansion and quality upgrading of the sustainable development of rural households, improving the inclusive finance and financial well-being of vulnerable groups in rural areas [8] and strengthening the stability of rural household development resilience. Therefore, this article explores the impact and mechanism of digital payments on rural household development resilience, as well as the ways to enhance rural household development resilience to provide a reference for decision-making to enhance the ability of rural households to cope with risks and shocks.

This article uses data from the Chinese Household Finance Survey (CHFS) to conduct several empirical tests and draw some useful conclusions. First, a baseline regression is conducted, with explained and explanatory variables in degree-fixed effects. It was found that digital payments can significantly impact rural household development resilience. Second, we search for appropriate instrumental variables by decreasing estimation errors because of the endogeneity in the 2SLS instrumental variable method. Finally, we embed a robustness test in the baseline regression results by replacing the explained variables and sample sets. Our results are consistent with the basic regression results. The empirical results are consistent with the basic regression ones, which show that (1) digital payments can significantly enhance rural household development resilience, (2) relaxing liquidity constraints and promoting farmer-oriented market participation are the primary mechanisms by which digital payments affect rural household development resilience, but digital payments do not have an impact on rural household development resilience by releasing credit constraints, and (3) the heterogeneity regression analysis showed that the promotion effect of digital payments on the rural household development resilience is the strongest in the Western areas, at medium level in the Central areas, and the weakest in the Eastern areas; the influence of digital payments is not significant to rural household development resilience.

1.2. The Literature Review

(1)

Origin, definition, and measurement indicators of resilience

The concept of “resilience” originated from the psychologist Anthony. It was later applied to the ecosphere, emphasizing the capacity of the ecological environment to protect itself from ecological problems, such as natural disasters. With the extension of the term “resilience” from natural ecosystems to human social systems in the 1990s, the concept also gradually evolved into “evolutionary resilience”. According to Barrett and Constas [3], resilience is a state, standard, or threshold that the subject reaches, the ability of individuals, households, or other organizations to avoid falling into poverty when facing various pressures or multiple shocks. The subject is considered to have achieved “resilience” if, and only if, this ability is maintained at a high level over time. Zewdie Birhanu et al. [9] explained that resilience is the intrinsic ability to bear or absorb both sudden shock and a prolonged impact, cope with temporary interruptions, and recover after an event while minimizing damage and costs. From the perspective of regional economics, Liao Jingwen and Zhang Keyun [10] proposed that resilience is an evolutionary concept that highlights the active ability of a system to adjust and adapt, which includes the core performance and functions of regions in response to shocks in the short-term and long-term maintenance of economic systems. Not only is it the core function of the regions to maintain economic systems and adapt impacts by making structural adjustments, but it also encompasses the ability to achieve long-term positive economic development. Referring to the definition of resilience proposed by Barrett and Constas, Li Han and Lu Qian [11] introduced it to the field of agricultural economics. Defined as a threshold, resilience was adopted to determine the degree of resilience of rural households under several consequences. Although the definition of resilience has not yet been agreed upon, it contains the characteristics of the ability of various subjects to resist and recover and achieve sustainability when facing external risks and disturbances. The research object of this study was farming families at the micro-level, and the index of “rural household development resilience”, based on “development resilience” as defined by Barrett and Constas, was constructed to measure the sustainable development ability of rural households.

As for how to measure “developmental resilience”, Jennifer and Barrett [12] referred to the method proposed. Based on potential nonlinear path dynamics, resilience was set as a prospective and probabilistic measure of well-being. They calculated resilience using the welfare function model and selected a minimum probability threshold P for comparison. The corresponding value of development resilience over the period can be worked out by estimation, which judges the possibility of avoiding poverty for poor households under various pressures or multiple impacts. If the corresponding value remains above the standard threshold, it indicates that the higher the development resilience of poor households, the greater the possibility of avoiding poverty.

(2)

Research results on the correlation between digital payments and household welfare

① In terms of objective welfare, Munyegera [13] conducted research on 846 rural households in Uganda and found that digital payments can reduce expensive transaction fees during the financing process and improve rural households’ ability to obtain loans and accumulate savings. Yin [14] and others found that digital payments can enrich families’ social networks to provide formal and informal lending support by changing people’s attitudes toward risk. Alinaghi [15] used household data to prove that digital payments can obtain more remittances by reducing transaction costs, thus releasing credit constraints and sharing the risks. According to Rao [16], the use of digital payments makes household risk assets more readily allocated and allows for a wide expansion, which eases the liquidity constraints of households. Weisong Qiu [17] and others proposed that digital payments can increase family income, effectively alleviate the stress of low-income families and maintain the level of household welfare.

From the perspective of subjective welfare, Feinberg [18] and Chatterjee [19] stated that the spillover effect of digital payments is able to deepen the pleasure of consumption instead of the pain of paying. Yaqian Wu [20] and others explained that the development of digital payments is helpful in alleviating the chronic problem of high savings and low consumption among rural Chinese households to improve the subjective well-being of rural households, especially in vulnerable groups, such as the elderly and those with low income or low education.

② In terms of research results on the impact of financial means on resilience, surveying the beneficiaries of financial inclusion in Shimoga, Karnataka, India, between 2010 and 2015, Vighneswara [21] demonstrated that financial inclusion enabled by micro-credit promotes the resilience of the poor. Based on entropy weight TOPSIS methods, Du Jiang and Zhao Weilin [22] empirically demonstrated that China’s financial policies can significantly enhance the resilience of socioeconomic development during a pandemic. Kousky et al. [23] proposed that many developing countries have applied parametric micro-insurance to enhance the economic development resilience of low-income families. Adopting the poverty trap theory and vulnerability theory for the first time, Lokendra et al. [24] used the difference-in-differences (DID) method to demonstrate the impact of asset transfer on household development resilience. Suri et al. [25] stated that the digital age has further promoted the development of resilience, and financial technology (fintech) is a key way to improve fiscal channels and household development resilience.

In conclusion, digital payments have become a hot topic in academic discussions, and many scholars have confirmed their impact on household consumption and savings. However, research on rural household development resilience is still in its infancy. Therefore, exploring the impact and mechanism of digital payments on rural household development resilience contributes to the existing research on the use of digital payments by households to achieve sustainable development and, ultimately, common prosperity.

2. Theoretical Analysis and Research Hypotheses

Digital payment mainly refers to the use of hardware facilities, such as computers and smart devices, as well as digital technology means, such as communication technology, artificial intelligence, and information security, to cover various life scenarios, such as online shopping, offline consumption, daily consumption, and market entrepreneurship. This article drew on the classification provided by Yin Zhichao [26] and Zhao Yueqiang [27] to make Internet payments (Internet Banking and Alipay), mobile payments (mobile devices, such as Alipay, WeChat Pay, Internet Banking, and Apple Pay), and digital currency as the main payment methods, and conducted the corresponding theoretical analysis.

Digital payments overcome the drawbacks of cash and card payments with their convenience and security. Bank cards can be attached to mobile devices and terminal networks, and payments can be completed without the need to carry cash. The optimization of payment methods, such as QR code payments, facial payments, and instruction payments, meets the diverse payment needs in rural households, increases the consumption willingness of rural households, and creates a more convenient, efficient, and secure payment environment [28]. The improvement in the consumption environment reduces the probability of offenses, such as market robbery, cash damage, and counterfeit currency circulation during the transaction process, considerably protecting the fluid capital of rural households and enhancing their development resilience.

Digital payments increase transaction speed, simplify transaction processes [29], and increase the borrowing efficiency of rural households. In traditional financial services, households need to pay vast transaction costs during the process of borrowing. Third-party payment institutions using the Internet as the intermediary break through medium restrictions on transactions and provide more funding channels for rural households. Rural households, therefore, do not need to visit a bank and wait there, thus increasing the efficiency of borrowing and lending. Once the efficiency of the households’ access to funds improves [30], the confidence of rural households in accessing funds is also boosted, which encourages these families to adopt various financial behaviors to expand their income. This is a good manner to enhance rural households’ development resilience.

Digital payments improve the settlement efficiency of rural market transactions and accelerate capital return in production and operation for rural households. Digital payments enable real-time transfers and various settlements, such as T + 0 (funds on weekdays received on the same day), T + 1 (funds on weekdays received on the second weekday), D + 0 (funds on natural days received on the same day), and D + 1 (funds on natural days received on the second weekday). They accelerate the flow of funds and reduce the operating costs of market transactions. More importantly, it is easy to obtain trading information of rural households via the Internet and big data during the market process, which is beneficial to exploring and seizing more business opportunities and strengthening the demonstration of successful entrepreneurship to enhance rural households’ development resilience.

In summary, this article proposes the following hypothesis:

Hypothesis H1.

Digital payments enhance rural household development resilience.

Analysis of Digital Payments’ Impact on the Rural Households’ Development Resilience from the Perspective of Liquidity Constraints

(1)

Analysis Based on Liquidity Constraints

On the one hand, digital payments enhance the efficiency of consumption services, increase the willingness to consume, stimulate the capability of smooth and current consumption, and upgrade household consumption to enhance the development resilience of rural households. The usability and convenience of digital payments enable rural households to spend money without cash. They increase the willingness to consume [9] and alleviate household liquidity constraints, which inevitably encourages rural households to work hard and engage in productive activities to smooth the consumption ability and achieve sustainable family consumption to enhance rural household development resilience.

On the other hand, with its rapidity and covenant-lite characteristics, digital payments reduce the transaction costs of transferring funds [8], indirectly expanding the budget and family welfare of rural households. WeChat Pay, Alipay, and the digital CNY can be used at any time, allowing rural households to make payments without converting deposits into cash [31], which reduces transaction costs, such as commissions charged by financial institutions, consumers, and merchants during the transaction process. Indirectly, they promote the right shift of the budget constraint line for rural households, expand the budget space, and improve consumption utility to enhance the development resilience of rural households.

Based on the above, this article proposes the following hypothesis:

Hypothesis H2.

The development of digital payments can effectively alleviate liquidity constraints and improve rural households’ development resilience.

(2)

Analysis based on credit constraints

Internationally, narrow liquidity constraints are defined as the inability of credit demanders to obtain the required credit funds at a low cost due to factors such as information asymmetry, which means their credit demand cannot be met [32]. This article followed the approach of Yin Zhichao et al. [4] and categorized narrow liquidity constraints as credit constraints.

Digital payments improve the efficiency of fund lending and enhance the availability and convenience of credit funds for rural households to respond to emergencies and risk shocks, enhancing their development resilience. Specifically, first of all, the use of digital payments helps banks and other financial institutions to collect household credit information and assess credit qualifications, effectively breaking down information barriers between trading platforms and rural households [22]. Secondly, digital payments simplify the borrowing process and overcome the transaction hysteresis of traditional credit operations. Finally, the credit scores generated by using digital payments expand financial channels for rural households [33]. It can help them to obtain small loans provided by Alipay, Ant Credit Pay, and JD IOU. For this reason, the use of digital payments effectively reduces the implicit costs and spatiotemporal constraints in the process of signing a credit contract, promotes the lending of credit funds by financial institutions to rural areas, and keeps rural households confident in their ability to obtain, instead of rejecting, funds to effectively alleviate rural households’ credit constraints, enabling them to quickly obtain borrowed funds to manage risks or engage in other life and production activities. Thus, they enhance their development resilience.

Based on the above, this article proposes the following hypothesis:

Hypothesis H3.

The development of digital payments can effectively alleviate credit constraints to enhance rural household development resilience.

(3)

Analysis-based market participation

On the one hand, digital payments accelerate the settlement speed of market transactions, enhancing the working operational efficiency of rural households to enhance their development resilience. Rural households can speed up fund turnover and the efficiency of operations by conducting payments through e-commerce platforms, real-time transfers, and the timely settlement of digital payments; the function of peer-to-peer payments also means that offline transactions no longer rely on the traditional physical form for transfers. This expands the profit-making space of rural household production and operation and motivates rural households to participate in output markets to benefit from innovation. On the other hand, digital payments provide an effective channel for information dissemination during transactions. They contribute to rural households’ exploration of opportunities, enhancing their working output to improve their development resilience. Goods or services can be sold worldwide through digital payments. In this process, the client base and elements are expanded, and isolated information islands among production, supply, and marketing are connected to promote an industrial cluster effect, which optimizes the allocation of resources and drives high-quality coupling development, increasing incomes to enhance the development resilience of rural households.

Based on the above, this article proposes the following hypothesis:

Hypothesis H4.

The development of digital payments can increase the market participation of rural households to enhance rural households’ development resilience.

3. Research Design

3.1. Data Source

The data used in this article were obtained from two different databases across three years: (1) The China Household Finance Survey (CHFS) released by the China Household Finance Research Center at the Southwestern University of Finance and Economics. The data cover detailed information on individual, household, and comprehensive variables, including financial knowledge, demographic characteristics (partially), and total household income, in 29 provinces (autonomous regions and municipalities) across the country, except for Tibet, Xinjiang, Hong Kong, Macao, and Taiwan; (2) The relevant indicators of the corresponding year and province (autonomous region and municipality) from the National Bureau of Statistics. After selecting and deleting some outliers of variables, a total of 3194 households were retained, and macro- and micro-data were matched as three-period-balanced panel data.

3.2. Empirical Model Setting

(1)

Benchmark Regression Model

To test whether digital payments affect the development resilience of rural households, this study employed a bidirectional fixed effects model, which has the advantage of obtaining a uniformly unbiased estimator even in the presence of missing variables that do not vary over time or individuals. The baseline model was set as follows:

$$D{R}_{ijt}^{}={\alpha }_0+{\alpha }_{1}D{P}_{jt}+{\alpha }_2{X}_{ijt}+{\delta }_i+{\phi }_t+{\mu }_{ijt}$$

(1)

In this paper, the subscripts $i$, $j$, and $t$ represent household, region, and year, respectively. The explained variable in Equation (1) represented by $D{R}_{ijt}$ is the rural household development resilience for different households, regions, and years. The explanatory variable $D{P}_{\mathrm{jt}}$ represents digital payment behavior in different regions and years. The variable $X_{ijt}$ is a set of characteristic variables at the level of household heads, households, and regions; the variable ${\delta }_i$ represents household fixed effects; the variable ${\phi }_t$ represents year fixed effects, and the variable ${\mu }_{ijt}$ represents the error term.

(2)

Mediation effect model

To further verify the mechanism of the effects of digital payments on the development resilience of rural households, this paper referred to the method proposed by Wen Zhonglin and Ye Baojuan [34] and set up the following mesomeric effect model:

$$M_{ijt}={\beta }_0+{\beta }_{1}D{P}_{jt}+{\beta }_2{X}_{ijt}+{\delta }_i+{\phi }_t+{\mu }_{ijt}$$

(2)

$$D{R}_{ijt}^{}={\gamma }_0+{\gamma }_{1}D{P}_{jt}+{\gamma }_2{M}_{ijt}+{\gamma }_3{X}_{ijt}+{\delta }_i+{\phi }_t+{\mu }_{ijt}$$

(3)

In the above equation, $M_{ijt}$ represents the mediator variable, while the other variables are consistent with the baseline regression model. The intermediate test is as follows: when both the coefficient ${\beta }_1$ in Equation (2) and the coefficient ${\gamma }_2$ in Equation (3) are significant, there is a mediating effect of $M_{ijt}$, and the $M_{ijt}$ mechanism is established; if either the coefficient ${\beta }_1$ in Equation (2) or the coefficient ${\gamma }_2$ in Equation (3) is not significant, there is no mediating effect of $M_{ijt}$, and the $M_{ijt}$ mechanism is not established.

(3)

Measurement Methods and Variable Setting

① Dependent variable: resilience in rural household development.

This article referred to the methods of Cissé and Barrett [12] and Li Han and Lu Qian [11] to measure resilience in rural household development. The specific steps were as follows:

First, construct a model with the average stochastic well-being of rural households as the explained structure and the rural polynomial lag in explained variable as the explanatory variable, as shown in Equation (4).

$$W_{ijt}={\displaystyle \sum }_{n=1}^k{}_{{\beta }_M}{}^{\wedge }g{}_M^n\left(W_{i,j,t-1},X_{ijt},{\beta }_M\right)+{\delta }_M{X}_{ijt}+{\mu }_{Mijt}$$

(4)

In Equation (4), the explained variable $W_{ijt}$ represents the random welfare mean value of rural households $i$ during period $t$. The core explanatory variable $g_M^n$ represents the polynomial function of the lagged dependent variable, which is the random welfare of rural households $W_{i,j,t-1}$. The parameter ${\beta }_M$ represents the risk that rural household $i$ faces from shocks or exposure, while ${\mu }_{\mathit{Mijt}}$ represents the random disturbance term. In this equation, the subscript m denotes the expected value equation, n denotes the moment order of higher moments, and k denotes the highest moment order of center moments. Based on Li Han and Lu Qian [11], the value of $W_{ijt}$ is the natural logarithm of the total annual consumption expenditure per capita of the household. The parameter k reflects the non-linear characteristics of the path of risk shocks encountered by farmers, and after significant testing, its value was determined to be 3.

Then, under the assumption of the zero mean of the disturbance term (E[${\mu }_{\mathit{Mijt}}$] = 0), the predictive values of the conditional expectation ${\mu }_{1\mathit{ijt}}$ and the conditional variance ${\mu }_{2\mathit{ijt}}$ of rural households were estimated separately, as shown in Equations (5) and (6).

$${\widehat{\mu }}_{1ijt}=\widehat{E}[W_{ijt}|W_{i,j,t-1},X_{ijt}]=g_M\left(W_{i,j,t-1},X_{ijt},{\widehat{\beta }}_M\right)+{\widehat{\delta }}_M{X}_{ijt}$$

(5)

$${\widehat{\mu }}_{2ijt}=\widehat{E}[W_{ijt}|W_{i,j,t-1},X_{ijt}]=g_V\left(W_{i,j,t-1},X_{ijt},{\widehat{\beta }}_V\right)+{\widehat{\delta }}_V{X}_{ijt}$$

(6)

Finally, select a specific poverty identification line $\underset{¯}{W}$ and jointly estimate the conditional welfare probability density function and the corresponding complementary cumulative density function; estimate the probability of rural households $\underset{¯}{W}$ exceeding the poverty identification line $i$ during period $t$ and calculate the resilience of household development $\stackrel{\wedge }{{\displaystyle D{R}_{ijt}}}$. The level of probability reflects the degree of resilience of rural households, as shown in Equation (7).

$$\stackrel{\wedge }{{\displaystyle {{\displaystyle D{R}_{ijt}}}_{}}}\equiv P\left({{\displaystyle W}}_{ijt}\ge \underset{¯}{W}|{{\displaystyle W}}_{i,j,t-1},{{\displaystyle X}}_{i,j,t}\right)=\overline{F_{{{\displaystyle W}}_{i,j,t}}}\left(\underset{¯}{W};\stackrel{\wedge }{{\mu }_{1i,j,t}}\left({{\displaystyle W}}_{i,j,t},{{\displaystyle X}}_{i,j,t}\right),\stackrel{\wedge }{{\mu }_{2i,j,t}}\left({{\displaystyle W}}_{i,j,t},{{\displaystyle X}}_{i,j,t}\right)\right)$$

(7)

In Equation (7), the poverty line $\underset{¯}{W}$ is based on the work of Li Han and Lu Qian (2021), which was set as USD 1.9 from the World Bank’s poverty line as the per capita daily consumption expenditure amount. The corresponding exchange rate and the CPI were converted to comparable figures from 2013. Then, use its natural logarithm.

② The core explanatory variable: digital payments.

The “digital payments” variable is characterized by the “third-party payments” questions of CHFS, which takes the payment platforms Alipay, WeChat Pay, Mobile Banking, JD Wallet, and Baidu Wallet as its research objects. In 2019, the respondents were asked whether they used third-party accounts. Those who answered “yes” indicated that they used digital payments; otherwise, it meant they did not use it. In 2017, the question was about “paying via mobile terminals, such as mobile phones and pads (Alipay, WeChat Pay, mobile banking, Apple Pay, etc.)”. The value “1” indicated that they used digital payments, while “0” showed that they did not use them. In 2015, “third-party payment platforms” was selected. The answers were the same as those in 2017.

③ Control variables.

To control the variables that may affect the development resilience of rural households other than digital payments, this paper selected control variables at three different levels: householder, family, and region. The control variables of the householder include gender, party membership, and education years. The control variables of family comprise family members, engaged in agricultural activities or not, the ownership of medical insurance, the median household assets, and the average household income. The regional control variables consist of the regional GDP, TAV, and financial development level.

Table 1. Variable definitions and descriptive statistics.

Variable Name	Variable Definition	Observations	Mean	Standard Deviation
Resilience of Rural Households	Measured according to Formulas (4) to (7)	6568	0.123	0.0236
Digital Payments	Assigned based on whether “third-party payments” are used, yes = 1, no = 0	6573	0.238	0.426
Gender	Male = 1, Female = 0	9863	0.892	0.311
Whether a Member of the Communist Party of China	A member of the Communist Party of China = 1, No = 0	9314	0.840	1.516
Years of Education	No education = 0, Primary school = 2, Junior high school = 9, Senior high school = 12, Technical secondary school/vocational school = 13, Junior college/technical college = 15, Undergraduate = 16, Master’s degree = 19, Doctoral degree = 23	9863	7.494	3.246
Total Family Population	Calculated according to total family population	9863	3.652	1.764
Engaged in Agricultural Activities	Assigned based on the nature of family members’ work, engaged in farming = 1, not engaged = 0	9863	0.751	0.969
Medical Insurance	Has urban employee basic medical insurance, urban resident basic medical insurance, new rural cooperative medical insurance, urban and rural residents’ basic medical insurance, or public medical care = 1, none of the above = 0	9863	0.964	0.187
Commercial Insurance	Has commercial life insurance, commercial health insurance, or other commercial insurance = 1, none of the above = 0	9863	0.0641	0.245
Average Total	Assets of the Family: The mean of the total assets of the family taken as the natural logarithm	9863	10.76	1.418
Regional Gross Domestic Product	The gross domestic product of each province (unit: CNY 100 million)/1000	9863	29.92	22.47
Level of Financial Development	The proportion of financial institutions’ loans and deposits to the regional gross domestic product	9863	3.318	0.790
Liquidity Constraint	Whether credit card is used, yes = 1, no = 0	9811	0.0692	0.254
Credit Constraint	Whether bank loans were obtained, yes = 1, no = 0	9863	0.0116	0.107
Market Participation	Whether engages in industrial and commercial operations, yes = 1, no = 0	9863	0.101	0.302
Number of Households	3194 households

shows the specific definitions of variables and their descriptive statistics.

4. Benchmark Regression Results and Analysis

In this section, we empirically test whether digital payments affect the resilience of rural households. Firstly, a benchmark regression was conducted with and without control variables using a fixed effects model. Secondly, suitable instrumental variables were identified, and 2SLS instrumental variable estimation was used to address potential endogeneity issues. Finally, we embedded a robustness test in the baseline regression results by replacing the explained variables and sample sets.

4.1. Benchmark Regression Results

Table 2. Benchmark regression results.

Variable	Rural Households’ Development Resilience	Rural Households’ Development Resilience
	(1)	(2)
Digital Payment	0.004 ***	0.005 ***
	(0.001)	(0.001)
Gender		0.006 ***
		(0.002)
Member of the Communist Party of China (CPC)		0.000
		(0.000)
Education level		−0.003 ***
		(0.000)
Total household population		−0.004 ***
		(0.000)
Engagement in agricultural activities		0.003 ***
		(0.000)
Medical insurance		−0.001
		(0.002)
Commercial insurance		0.005 ***
		(0.002)
Total average assets of the household		0.002 ***
		(0.000)
Regional Gross Domestic Product (GDP)		0.000
		(0.000)
Financial development level		0.002
		(0.003)
Constant term	0.120 ***	0.113 ***
	(0.000)	(0.013)
Control variables	No	Yes
Fixed effects for households	Yes	Yes
Fixed effects for years	Yes	Yes
Observations	6568	6189
R2	0.029	0.089

Robust t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

shows the baseline regression results of the impact of digital payments on the development of resilience. The first column (1) reports the regression results without the control variables, wherein the regression coefficient of digital payments is significantly positive at the 1% level, indicating that digital payments improve the development resilience of rural households. The second column (2) reports the regression results with the control variables, and the regression coefficient of digital payments is 0.005, which is significantly positive at the 1% level, showing that, for each additional unit in digital payments, the development resilience of rural households increases by 0.005 units. This is probably because digital payments offer new payment methods, leading to the development of a new generation of the digital economy represented by network technology and breaking the limitations of the original resources, which provides faster and more convenient transaction channels for rural households to improve their quality of life by enhancing their development resilience. Specifically, digital payments can meet the consumption needs of rural households in various scenarios. The information on rural households is collected by third-party institutions when trading using digital payments, which ensures the credit ability and security of rural households. In this way, information barriers are broken down, and rural households are able to enjoy a higher level of digital welfare offered by financial services, which breaks through resource scarcity. From an economic perspective, by reducing expensive service and information fees during the process of market operation, digital payments offer rural households more capital chains to participate in market activities and respond to emergencies, which helps them to accumulate funds to alleviate risks and motivate endogenous capacity.

From the perspective of the results of the regression with the control variables, (1) the regression coefficients of gender, engagement in agricultural activities, commercial insurance, and total average assets of the household are significantly positive. This indicates that men play a major role in promoting the development resilience of rural households. Engagement in agricultural activities and increasing the total average assets of the household are beneficial for improving the development resilience of rural households. Commercial insurance provides rural households with prior risk prevention and ex-post reduction in negative losses, which boosts their confidence to engage in more economic activities so as to enhance rural households’ development resilience. (2) The regression coefficient of education level is significantly negative, indicating that it has a reverse impact on rural households’ development resilience. The reason may be that, compared to cities, poor infrastructure, as well as depleted technical and knowledge-based industries, force the most advanced talent to relocate away from the countryside due to a lack of employment, entrepreneurship, and other feedback mechanisms, which are adverse to enhancing rural households’ development resilience. (3) The regression coefficient of the total household population is significantly negative, indicating that the larger the family, the less conducive it is to improving rural households’ development resilience. It is because a large family increases the total household consumption expenditure. Some unpredictable risks, such as diseases and accidents, will be increased. It will lay a considerable financial burden on families. Although the speed of fund turnover is raised by digital payments, rural households are still unable to obtain large loans because of their weak position. With the growing number of vulnerable groups in the total household population, household welfare is challenging.

4.2. Endogeneity Analysis

To mitigate the estimation error caused by endogeneity, such as omitted variables and reverse causality, in the baseline model, this article used “the Internet penetration rate” of each province in the China Statistical Yearbook as the instrumental variable of “digital payments” and estimated it using the 2SLS instrumental variable method. The reasons for the selection are as follows: Firstly, the Internet and traditional payments are the communication media of digital payments, so the Internet penetration rate is one of the main factors that determine the level of the development of digital payments. In other words, there is an instrumental variable correlation between the Internet penetration rate and digital payments. Secondly, with the continuous increase in the Internet penetration rate, the digital divide is broken, which promotes economic development. However, it is not related to the development resilience of rural households but, instead, to the exogeneity of instrumental variables.

Table 3. Instrumental variable method estimation results (2SLS).

Variable	Digital Payments	Rural Households’ Development Resilience
	(1)	(2)
Internet Penetration Rate	0.010 ***
	(0.001)
Digital Payments		0.016 ***
		(0.005)
Control Variables	Yes	Yes
First-Stage F-value		132.89 ***
Observations	6189	6189
R2	0.191	0.182

Robust t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

reports the estimates of the instrumental variables. Column (1) reports the regression results of the first stage of the instrumental variables. The regression coefficient of the Internet penetration rate is 0.010, which is significant at the 1% level, indicating that for each unit of the Internet penetration rate, digital payments increase by 0.010 units. In addition, the F-value in the first stage is significant at the 1% level, at 132.89, indicating that the Internet penetration rate is an appropriate instrumental variable for digital payments through weak instrumental variable testing. Column (2) reports the regression results of the second stage. The regression coefficient of digital payments is significant at the 1% level, indicating that rural households’ development resilience is steady. Compared with the baseline regression coefficient, the regression coefficient approximately tripled, indicating that the baseline regression suffers from estimation bias due to endogeneity issues and underestimates the level of the impact of digital payments on rural households’ development resilience.

4.3. Robustness Check

To test the robustness of the baseline regression results and how digital payments affect rural households’ development resilience, this paper performed two types of robustness checks, replacing the dependent variable and changing the sample set. On the one hand, this paper selected the 2015 poverty line standard of CNY 2800 in China and used the daily average and natural logarithm to recalculate rural households’ development resilience to replace the original dependent variable. The regression results are shown in column (1) of

Table 4. Robustness regression results.

Variable	Replacing the Dependent Variable	Changing the Sample Set
	(1)	(2)
Digital Payments	0.005 ***	0.005 ***
	(0.001)	(0.001)
Constant	0.115 ***	0.104 ***
	(0.014)	(0.014)
Control Variables	Yes	Yes
Observations	6189	6007
R2	0.089	0.093

Robust t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

, where the regression coefficient of digital payments is significantly positive at the 1% level. On the other hand, since digital payments originated mainly in Hangzhou, this paper referred to the relevant literature by Yin Zhentao et al. [35] and deleted the rural household sample located in Zhejiang Province for re-regression. Column (2) of Table 4 reports the regression results after changing the sample set, where the regression coefficient of digital payments is still significantly positive at the 1% level. Comparing the above two robustness check results with the baseline regression results, we found that the three regression results are basically consistent, indicating that the conclusion that digital payments improve rural households’ development resilience is reliable.

5. Regression Results and Analysis

Digital payments have certain mechanisms for improving rural households’ development resilience. This section verifies these mechanisms from three perspectives: liquidity constraints; credit constraints; and market participation.

5.1. Liquidity Constraints

In this study, the variable “whether to use credit cards” in the CHFS database was used to represent the mechanism variable of “liquidity constraints”, with a value of 1 for “yes” and 0 for “no”. Following the approach of Yin Zhichao et al. [4], column (1) of

Table 5. Results of testing for liquidity constraints.

Variable	Liquidity Constraints	Rural Households’ Development Resilience
	(1)	(2)
Digital Payments	0.071 ***	0.005 ***
	(0.014)	(0.001)
Liquidity Constraints		0.006 ***
		(0.002)
Constant	−0.767 **	0.115 ***
	(0.342)	(0.013)
Control Variables	Yes	Yes
Observations	6157	6154
R2	0.024	0.094

Robust t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

reports the regression results of digital payments’ alleviation of liquidity constraints, and the regression coefficient is significant at the 1% level, indicating that digital payments can alleviate liquidity constraints for rural households. Column (2) reports the regression results of liquidity constraints on rural households’ development resilience, with a regression coefficient of 0.006, significant at the 1% level, indicating that for every unit of alleviated liquidity constraints, the level of rural households’ development resilience increases by 0.006 units. According to the mediating effect test, both regression coefficients are significant and positive, indicating that digital payments enhance rural households’ development resilience by alleviating liquidity constraints. The reason may be that digital payments overcome the limitations of cross-temporal and spatial consumption, help rural households to expand their consumption willingness, alleviate the impact of liquidity constraints, and smooth consumption through intertemporal income. When rural households face expected income declines or need funds to purchase urgently needed goods, the alleviation of liquidity constraints improves the consumption utility of rural households, forcing them to actively work to obtain high returns to smooth consumption and enhance their development resilience.

5.2. Credit Constraints

In this study, the variable “whether to obtain a bank loan” in the CHFS database was used to represent the mechanism variable of “credit constraints”. The fact that rural households obtain bank loans indicates that there are no credit constraints, and vice versa. Column (1) of

Table 6. Results of testing for credit constraints.

Variable	Credit Constraints	Rural Households’ Development Resilience
	(1)	(2)
Digital Payments	0.013 **	0.005 ***
	(0.006)	(0.001)
Credit constraints		0.006
		(0.004)
Constant	−0.033	0.113 ***
	(0.046)	(0.013)
Control Variables	Yes	Yes
Observations	6192	6189
R2	0.011	0.089

Robust t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1

reports the regression results of digital payments relieving credit constraints, and the regression coefficient is significant at the 1% level, indicating that digital payments can alleviate credit constraints for rural households. Column (2) reports the regression results of credit constraints on rural households’ development resilience with a non-significant regression coefficient. According to the mesomeric effect, digital payments do not affect rural households’ development resilience by alleviating credit constraints. A possible explanation for this is that, compared to traditional monetary payments and the traditional financial system, digital payments are embedded in the financial system, deepening and innovating the means of financial services, lowering the entry barrier for farmers to access credit, simplifying the cumbersome procedures in the transaction process, and easing the credit constraints of farmers. In this sample, the factors for digital payments being unable to alleviate credit constraints are the lower permeability of rural financial sectors and the entrenchment of financial self-exclusion. Deep-rooted tradition means that they are not good at applying for loans to solve financial issues. Faced with the risk of increased expenditure led by education, training, and accidents, they are likely to resort to transfer payments or private lending.

5.3. Market Participation

In this study, the variable “whether engaged in commercial or industrial operations” in the CHFS database was used to represent the variable “market participation mechanism”. They are made of the individual business, tenancy, transportation, online stores, WeChat Business, daigou agents, family farms, cooperatives, agricultural enterprises, and companies. The value of 1 is assigned for “yes” and 0 for “no”. Following the approach of Li Han et al., column (1) of

Table 7. Results of testing for market participation.

Variable	Market Participation	Rural Households’ Development Resilience
	(1)	(2)
Digital Payments	0.043 ***	0.005 ***
	(0.014)	(0.001)
Market participation		0.004 **
		(0.002)
Constant	−0.256 **	0.114 ***
	(0.120)	(0.013)
Control Variables	yes	yes
Observations	6192	6189
R2	0.024	0.090

Robust t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

reports the regression results of digital payments on market participation, with a regression coefficient of 0.043, significant at the 1% level, indicating that for each unit in digital payments, the market participation of rural households increases by 0.043 units; i.e., digital payments enhance the market participation of rural households. Column (2) reports the regression results of market participation on rural households’ development resilience, with a significant positive regression coefficient at the 1% level, indicating that the higher the rural households’ market participation, the better their ability to enhance their development resilience. According to the mesomeric effect testing, digital payments can enhance rural households’ development resilience by increasing their market participation. A possible explanation is that digital payments can not only alleviate rural households’ budget constraints but also provide an effective channel for information dissemination. With the rapid development of technologies, such as artificial intelligence, big data, and cloud computing, more business opportunities can be explored and grasped in rural households’ production and operation. Rural households expand their client base and elements through online e-commerce platforms to increase their income from production and operation and thus enhance their development resilience.

6. Heterogeneity Analysis

This section analyzes the heterogeneity of the impact of digital payments on rural households’ development resilience from two perspectives: “a targeted poverty alleviation family or not” and “geographic location”.

6.1. Regional Heterogeneity

In this study, the “geographic region to which the sample household belongs” was divided into three groups, “Eastern”, “Central”, and “Western”, for the heterogeneity analysis. Columns (1), (2), and (3) of

Table 8. Regression results of geographical location heterogeneity.

Variable	Eastern Region	Central Region	Western Region
	(1)	(2)	(3)
Digital Payments	0.003 *	0.006 ***	0.004 **
	(0.002)	(0.002)	(0.002)
Constant	0.163 ***	0.137 *	0.063 *
	(0.020)	(0.075)	(0.034)
Control Variables	Yes	Yes	Yes
Observations	2180	2275	1699
R2	0.102	0.120	0.090

Robust t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

, respectively, show the regression results of digital payments on the resilience of rural households in the Eastern, Central, and Western regions. The regression coefficients of all three regions are significantly positive, indicating that digital payments in all regions have a positive, promoting effect on rural households’ development resilience. However, this promoting effect decreases from strong to weak in the order of Western, Central, and Eastern regions. A possible explanation is that the financial system in the Central and Western regions lags behind, being limited by weak infrastructure and a low economy, contrasting to the Eastern region with a higher economic level and a more mature financial system. Digital payments can not only surpass the limits of time and space to improve economic and financial regressions in the Central and Western regions but also alleviate liquidity and credit constraints and increase their market participation, which plays an important role in enhancing rural households’ development resilience.

6.2. Poverty Level Heterogeneity

In this paper, “whether your family is a subsistence allowance household/poor household” in the CHFS database was used to divide the sample into “subsistence allowance households” and “non-subsistence allowance households”.

According to the national poverty standards, a household with subsistence allowance mainly refers to rural households whose annual per capita net income is lower than the local minimum living guarantee standard. Column (1) of

Table 9. Regression results of poverty level heterogeneity.

Variable	Households with Subsistence Allowance	Households without Subsistence Allowance
	(1)	(2)
Digital Payments	0.008	0.004 ***
	(0.008)	(0.001)
Constant	0.118 *	0.128 ***
	(0.064)	(0.014)
Control Variables	Yes	Yes
Observations	895	5196
R2	0.084	0.107

Robust t-statistics in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

reports the regression results of the impact of digital payments on the resilience of households with subsistence allowance, where the regression coefficient is not significant, indicating that digital payments do not affect rural households’ development resilience. A possible explanation is that the use of digital payments has certain requirements for age, education level, and family infrastructure conditions.

Households with subsistence allowance are mainly those who face long-term difficulties in life due to illness, disability, old age, physical weakness, and poor living conditions. They are not exposed to digital payments and are unable to cultivate the ability to use digital payment tools to increase their development. Column (2) reports the regression results of the impact of digital payments on the resilience of households without subsistence allowance, where the regression coefficient is significantly positive at the 1% level, indicating that digitally inclusive finance has a positive, promoting effect on the resilience of households without subsistence allowance. Compared to households with subsistence allowance, households without subsistence allowance have better living conditions, such as owning smartphones and broadband access, which promote their use of digital payments. At the same time, households without subsistence allowance have a stronger awareness of poverty alleviation and self-development and are more capable of using digital payments to obtain the capital of existence and evolution. This enhances their development resilience.

7. Discussion

Aiming at completing the research that previously only analyzed short-term effects by examining digital payments on the nonlinear nature of event development dynamics, this paper discussed the impact of digital payments on the development resilience and its enforcement mechanism of rural households in China. It provided a theoretical basis for future research on the relationship between digital payments and development resilience.

Firstly, this paper defined the concept of “development resilience” and its measurement to assess the level of rural household development resilience in China. Based on poverty trap and nonlinear path dynamics theories, this paper considered the evolution of poverty dynamics and used the Markov method to set up a family welfare model to estimate development resilience as the conditional probability of meeting a certain welfare standard, which is able to better predict and monitor the future long-term trend of poverty alleviation among rural households in China.

Secondly, this paper examined the impact of digital payments on the development resilience of rural households in China. It can be found that digital payments can enhance rural households’ development resilience, which is consistent with the findings of Suri et al. [25]. They argued that the digital era further promotes family resilience, and digital finance is more conducive to improving rural household development resilience in response to negative shocks, which is a key way to improve fiscal channels and family resilience [25]. This is basically consistent with our findings, although the measurements in this paper differed from those of Suri et al. [25]. Digital finance cannot represent the overall level of the digital economy.

At present, there is relatively little literature on the impact of digital payments on rural household development resilience, in which there is no direct evidence to indicate the relationship between digital payments and rural household development resilience. In particular, the relationship and internal mechanism between digital payments and rural household development resilience are deduced and discussed from the available literature, and the relevant results of such research are expanded from the perspective of the sustainable development of rural households. To better verify the reliability of these inferences, this paper used the China Household Finance Survey (CHFS) data compiled by the Southwest University of Finance and Economics to select samples whose household registration is “rural household”, which reflects the research object of this paper as rural households, and test the impact of digital payments on the development resilience and its mechanism of rural households in the 2SLS instrumental variable method. It is also used to mitigate the estimation error caused by endogeneity. We embedded a robustness test in the baseline regression results by replacing the explained variables and sample sets. The final conclusion is consistent with the baseline regression results, which indicates that the results are reliable.

It was also found that digital payments are a major mechanism affecting rural household development resilience by easing liquidity constraints and promoting market participation. They provide an important reference for the formulation of relevant policies to enhance rural household development resilience. Digital payments do not have an impact on rural household development resilience by alleviating credit constraints. This research is useful for discovering the existing problems in the development of digital payments, creating relevant measures to optimize the development of digital payments, and releasing credit constraints to effectively enhance the development resilience of rural households.

Finally, to examine the differences in research fields, this paper analyzed the heterogeneity of the impact of digital payments on the development resilience of rural households from two perspectives, “a poverty-alleviated family or not” and “geographic location”. The results of the heterogeneity analysis show that the promotion effect is the strongest in the Western area, medium level in Central areas, and the weakest in the Eastern areas. This indicates that digital payments are an effective way to enhance rural household development resilience in Western and Central China. At the same time, the impact of digital payments on the development resilience of low-income households is not significant, signifying that the effective role of digital payments depends on material conditions. This is helpful in understanding the pros and cons of digital payments, which actively promotes digital payments in the future.

Limitations of this Study

Firstly, the availability of relevant data on rural households is relatively low because of the accessibility of public data, although the data are genuine and reliable. The panel data spanning three periods can have some abnormal or missing values or lack some variables related to rural household development resilience. Thus, it can fail to reflect the key factors that affect rural household development resilience. Secondly, digital payments can enhance rural household development resilience by alleviating liquidity constraints and promoting participation in the market mechanism. However, it is still necessary to explore how to establish a long-term mechanism to enhance rural household development resilience through digital payments under more complex, multidimensional, and persistent relative poverty. Finally, there is relatively little literature on rural household development resilience. So, the theoretical basis on which this paper relies is limited. This led to a lack of objective theoretical basis for some mechanism deduction and empirical analysis, which is somewhat missing.

Overall, in a more complex socioeconomic environment, research on the relationship between digital payments and rural household development resilience helps to evaluate the welfare status of rural households and the long-term state facing multidimensional relative poverty. It can promptly detect poverty return or new poverty and control these problems as soon as possible to achieve sustainable poverty alleviation, poverty reduction, and sustainable development.

8. Conclusions and Countermeasures

8.1. Conclusions

Taking rural China as the research area and rural households in China as the research object, this study used data from three waves (2015, 2017, and 2019) of the CHFS to empirically investigate the impact of digital payments on rural household development resilience. The main conclusions are as follows:

Firstly, the baseline regression results indicate that digital payments significantly enhance rural household development resilience. It was verified that Hypothesis 1 in the previous study is valid. This conclusion still holds after the endogeneity and robustness tests. This reflects that digital payments are an effective way to enhance rural household development resilience;

Secondly, the mechanism regression results indicate that alleviating liquidity constraints and promoting rural households’ market participation are the main mechanisms through which digital payments affect rural household development resilience. It was verified that Hypotheses 2 and 4 are valid. However, digital payments do not affect rural households’ development resilience by alleviating credit constraints, which means Hypothesis 3 is not supported. This provides a theoretical reference for relevant entities to formulate relevant policies;

Thirdly, the heterogeneity regression results show that the promotion of rural household development resilience by digital payments varies from strong to weak in the order of the Western, Central, and Eastern regions. Digital payments have a significant impact on the development resilience of non-low-income households instead of low-income ones.

8.2. Countermeasures

In order to enhance rural household development resilience, achieve the sustainability of rural poverty alleviation and sustainable development, consolidate the achievements of poverty alleviation of the digital ecosystem, and comprehensively promote rural revitalization, this paper proposes the following policy suggestions.

Firstly, the government should improve the digital ecosystem and enhance the role of digital payments in rural construction. The empirical results of this paper show that the development of digital payments effectively improves rural household development resilience, but it depends on the government-driven settlement of digital payments in rural areas. It should support the promotion of new digital professions in rural areas, increase digital employment opportunities and digital entrepreneurship projects, and enable households to find employment and start businesses in these areas. In this way, the channels for increasing incomes are broadened, household participation in the market is promoted to accumulate livelihood and development capital, and rural development is encouraged.

Secondly, the government should strengthen education and training and establish a digital payment system. This study found that it has great potential to promote financial technology in shaping “resilient farmers” and realizing the harmonious development of urban and rural areas. However, some households are still subject to credit constraints due to a psychological reluctance to borrow capital. On the one hand, efforts should be made to increase the training in digital financial technology in rural areas, which will gradually narrow the gap in the use of digitized products between urban and rural areas. On the other hand, financial institutions should further innovate and launch financial products to meet the needs of rural households, enabling them to improve credit availability through digital payment channels to enhance their development resilience. This will affect the “long-tail effect” of the financial market well.

Thirdly, the research results show that the use of mobile payments increases the likelihood of rural households’ engagement in innovative activities. As digital payments create more favorable conditions for households to start businesses, they enhance development resilience. They should enrich financing channels, support the development of Internet finance, strengthen the supervision of online lending platforms, provide better financial support for rural households, simplify the procedures for accessing the entrepreneurial market, and improve the efficiency and transparency of information for startups. This will help rural households to complete high-quality entrepreneurship and improve the sustainable development capacity of rural households.