A Study of the Impact of Mobile Payment on the Enhancement of Consumption Structure and Pattern of Chinese Rural Households

Abstract

Amidst China’s economic transition towards high-quality development, the latent potential of the rural consumer market has been significantly unlocked, rendering the evolution of rural household consumption patterns a pivotal area of research. In this paper, we empirically investigate the influence of mobile payment on the transformation of consumption patterns within Chinese rural households, utilizing data from the China Household Finance Survey 2017 and China Household Finance Survey 2019. This study’s findings reveal the following key points: Firstly, mobile payment significantly contributes to the enhancement of the consumption structure within rural households. Secondly, the mechanistic analysis demonstrates that mobile payment plays a pivotal role in alleviating mobility constraints and optimizing the consumption environment, thereby facilitating the improvement of rural households’ consumption patterns. Lastly, the heterogeneity analysis indicates that mobile payment has a more pronounced effect on the upgrading of consumption structures among rural households with older household heads and higher education levels. In light of these findings, this paper suggests advancing the development of mobile payment infrastructure in rural areas, enhancing the proliferation of smartphones in rural regions, and bolstering financial education initiatives within rural communities.

1. Introduction

Since the inception of economic reforms and opening up, China has experienced robust economic growth; however, the expansion of domestic demand has not kept pace with this development. Over an extended period, China has adhered to an economic growth model characterized by investment-driven and export-oriented strategies [1]. While this model has achieved some success, it has also given rise to several challenges. Among the most conspicuous challenges is the sluggish growth in consumption, resulting in an imbalance in the economic composition and a dearth of domestic demand [2], thereby significantly constraining the sustainable development of the national economy. Especially in the present circumstances, considering the ongoing international financial turmoil, the downturn in global markets, the domestic economic growth slowdown, and mounting inflationary pressures, it is no longer tenable to persist in relying on investment and exports as the primary drivers of economic growth. In response to this challenge, the Chinese government has implemented a range of measures aimed at fostering consumption to advance high-quality economic development. In 2019, ten ministries and commissions, among them the Development and Reform Commission, collaborated to release the “Implementation Programme for Further Enhancing Supply to Foster Stable Consumption Growth and Encourage the Establishment of a Robust Domestic Market (2019).” This document explicitly emphasized the imperative of enhancing product quality, refining the consumption environment, and bolstering consumption capacity to facilitate the development of a robust domestic market [3]. Subsequently, a cascade of policy documents, including the “Opinions of the State Council of the Central Committee of the Communist Party of China on Enhancing the Consumer Promotion System to Further Unleash Population Consumption Potential” and the “Implementation Plan for Enhancing Institutional Mechanisms to Encourage Consumption,” were sequentially released. These documents played a pivotal role in providing additional clarity on strategic objectives and policy measures aimed at promoting consumption. Amidst China’s shift toward high-quality economic development, consumption is recognized as a pivotal driver of economic growth, with the expansion of domestic demand reemerging as a central pillar of China’s macroeconomic policy. Consequently, expediting the shift in the economic development paradigm, particularly by firmly establishing a strategic foundation for expanding domestic demand, has gained paramount importance [4]. Hence, to expedite economic transformation, particularly the objective of expanding domestic demand, it is imperative to enhance the consumption capacity of the Chinese populace. This augmentation will propel the nation’s economic growth.

At the Fifth Plenary Session of the 19th Central Committee of the CPC Central Committee, the Chinese government clearly defined the important strategic tasks for the 14th Five-Year Plan period as “forming a strong domestic market and building a new development pattern” and “giving priority to the development of agriculture and rural areas, and promoting the revitalisation of the countryside in an all-round way.” “In particular, the importance of “developing the urban and rural consumer markets” was emphasized. This strategic objective reflects the substantial importance attributed to the consumption potential of rural residents and the pressing necessity to construct a novel economic development framework [5]. Currently, approximately 40% of China’s total population resides in rural areas, and rural consumption demonstrates features of both quantity expansion and structural enhancement. In 2021, the per capita consumption expenditure and disposable income of rural residents increased by 15.3% and 9.7%, respectively, surpassing those of urban residents by 4.2 percentage points and 2.6 percentage points, respectively (Data source: Website of the National Bureau of Statistics of China). Amidst China’s enduring consumption growth, the rural consumption market has seen a significant unleashing of its potential, leading to a continuous expansion in the rural consumption scale and a rapid increase in consumption expenditure [6]; this not only enhances their living standards but also provides a significant boost to national economic growth, thereby fostering the establishment of a robust domestic market. Nevertheless, it is evident that China’s urban–rural dual economic structure persists at present, resulting in notable disparities in the quality of life between urban and rural residents, particularly in terms of consumption levels and consumption patterns [7]. In contrast to urban areas, rural residents exhibit lower consumption rates, encounter insufficient availability of service-oriented consumption options, and contend with high household indebtedness rates, among other challenges. These factors undermine the expansion of the domestic market’s development. In the milieu of urban–rural integration development, addressing the imperative of enhancing rural residents’ consumption patterns and ameliorating the urban–rural consumption divide emerges as a pressing concern.

Mobile payment originates from the convergence of information and communication technology (ICT) and financial services. With the rapid development of Internet technology and e-commerce, mobile payment has become an important part of our daily lives, providing an opportunity to upgrade the consumption structure of rural residents in China [8,9]. Mobile payment, as a cost-effective and efficient payment method, has seamlessly integrated with the Internet, terminal devices, and financial institutions, culminating in the creation of a novel payment ecosystem that is now an integral component of the contemporary commodity retail market [10]. As mobile payment has gained widespread popularity, the number of mobile payment users in China has steadily risen from 187 million in 2011 to 904 million in 2022, reflecting a remarkable growth trajectory. By the close of 2021, mobile payment users constituted 87.6% of the entire Internet user base. Survey results indicate that nearly 98% of respondents identified mobile payment as their primary mode of payment, with individuals utilizing mobile payment services an average of three times per day, with QR code payment representing a substantial 85% of these transactions. The advent of mobile payment has empowered rural residents with a convenient payment method, enhancing their purchasing capacity and stimulating increased participation in market transactions. This phenomenon has catalyzed a transformation in consumption patterns, enabling rural residents to access a wider array of high-quality and diverse goods and services, thereby facilitating the elevation of consumption standards (Data source: Website of the National Bureau of Statistics of China). Consequently, this paper will investigate the influence of mobile payment on the consumption structure of rural residents, with a specific focus on its impact on consumption behavior and its role in driving consumption improvement, ultimately steering China’s economy towards high-quality development.

The subsequent sections of this paper are meticulously structured as follows: The second segment delves into an extensive review of pertinent literature concerning mobile payment and the evolution of consumption patterns. Here, research hypotheses are formulated. The third section meticulously delineates the data, variables, and model configurations employed in this study. Moving forward, the fourth part delves into a comprehensive analysis comprising benchmark results, robustness tests, mechanism analysis, and heterogeneity examination. The fifth part is a discussion. Lastly, the sixth section encapsulates the conclusive findings and pertinent policy recommendations.

2. Literature Review and Research Hypotheses

2.1. Literature Review

The upgrading of consumption structure and optimization of production structure serve as the foundational drivers of sustained economic growth, reflecting a nation’s economic development quality and potential [11]. Currently, numerous scholars, both domestically and internationally, have conducted pertinent research on consumption structure upgrading. Several scholars have offered definitions for consumption structure upgrading, with William J. Baumol [12] characterizing it as a dynamic adaptive process wherein a region or nation shifts its consumption framework from one centered on survival to one emphasizing development and enjoyment. In tandem with economic advancement, the makeup of residents’ expenditure undergoes a transformation: the share dedicated to basic survival expenses diminishes gradually, while the allocation towards enjoyment and developmental expenditures experiences a corresponding augmentation. This evolution constitutes the essence of consumption structure upgrading [13]. Throughout this progression, individuals increasingly prioritize health, nutrition, leisure, and entertainment, all the while cultivating a penchant for elevated consumption quality and refined tastes [14]. Considering the growing importance and popularity of mobile payments, researchers investigated the factors that influence consumers’ willingness and intention to adopt mobile payments [15]. This examination predominantly falls into two categories: one focuses on identifying the factors that drive consumption structure improvement. Zhang et al. [16] have identified wage income and transfer income as crucial drivers for advancing the enhancement of China’s consumption structure. Additionally, the study conducted by Wang and Zhang [17] has demonstrated the effective impact of financial support on elevating the consumption standards of rural residents. Another strand of research investigates the factors that hinder the upgrading of the consumption structure. Trutsch [18] contends that the low-income level and significant income disparity represent the primary factors constraining rural residents’ consumption. In tandem with economic development and income-related influences, Shi et al. [19] assert that deficiencies in the social security system, coupled with shortcomings in the consumption environment, contribute to the low consumption levels observed among rural residents. Particularly, the rural healthcare facilities and management system are imperfect. Some rural residents lack access to a reliable healthcare system, facing shortages in healthcare facilities and a shortage of highly skilled medical personnel. These factors directly impede rural residents’ ability to enhance their consumption quality and contribute to the widening gap between urban and rural consumption levels. Hence, there is a need to facilitate the enhancement and alignment of consumption patterns among urban and rural residents through the reduction of income disparities [6].

The relentless advancement of communication and information technology, interwoven with consumer behaviors, has ushered in the era of mobile payment. Rooted in the evolution of internet technologies like big data and cloud computing, mobile payment has solidified its place as one of China’s “Four Great Inventions of the New Era.” Scholars have undertaken extensive research into this phenomenon. Dahlberg et al. [20] define mobile payments as using wireless and other communication technologies to facilitate payments for goods, services, and bills. Prelec [21] defines mobile payment as a method through which consumers utilize mobile devices to finalize transactions for acquiring goods or services. This process entails the transmission of data and information from the payer to the receiver via a third-party transaction platform or an intermediary, adhering to the pre-agreed transaction amount. Falk et al. [22] discovered that consumers exhibit a greater willingness to make payments using their mobile phones as opposed to cash or credit cards.

According to Zhang et al. [23], mobile payment wields a significant and affirmative influence on household healthcare participation and healthcare expenditures. It serves as an impetus for households to bolster investments in preventive healthcare, providing digital financial underpinning; this, in turn, addresses the challenge of shifting the healthcare gateway forward, contributing to the construction of a healthier China and offering a pathway out of poverty for impoverished households. According to Yin and Lee [24], mobile payments contribute to flexibility and self-employment, which in turn helps to establish a long-term mechanism to address relative poverty and help poor households escape from poverty. Yin et al. [25] have observed that mobile payment substantially diminishes households’ need for precautionary savings to address uncertainties encompassing health, medical matters, unemployment, and income fluctuations. In a similar vein, the studies by Jack W et al. [26] and Zhang et al. [27] have substantiated that the evolution of digital finance fosters inclusive growth by stimulating entrepreneurship within households possessing limited physical or social capital. Concurrently, the significance of digital finance in bolstering residents’ consumption has garnered widespread agreement. The digitization of payment methods mitigates both search and transaction expenses, consequently influencing residents’ spending habits. In addition, several other researchers have demonstrated the positive impact of mobile money adoption on self-employment, household access to well-paying jobs, receipt of remittances, savings, investment, ability to cope with unforeseen shocks, and firm performance [28,29,30].

A study conducted by Chatterjee et al. [31] revealed that the consumer’s choice of payment method significantly influences their perception of the product. When a purchase is made with a credit card, consumers tend to emphasize the benefits derived from the product, while payments made in cash shift their focus toward the cost incurred during the purchase. Hirschman [32] posits that credit cards, being a novel payment method, offer users a credit limit, thereby enabling the potential for overspending, and the availability of a credit line increases the probability that consumers will make a spending decision [33,34]. Consequently, payment through credit cards is associated with a noteworthy increase in residents’ consumption expenditure. Prelec [21] and others argue that the use of credit cards not only promotes an increase in consumer spending but also increases the likelihood of spending and reduces the time to make a spending decision. Pham [35] states that Alipay online payment offers a mobile phone code-sweeping payment feature, essentially turning a mobile phone into a wallet. This innovation eliminates the inconvenience of dealing with change, significantly enhancing transaction efficiency. Furthermore, mobile payment, a pivotal component of digital finance, enhances households’ capacity to diversify risk and elevate their consumption standards [36]. Flavian [15] asserts that mobile payments, e-wallets, and mobile banking, owing to their convenience, are expected to gain increasing popularity. Mobile payments have become increasingly pertinent in people’s lives, gradually supplanting cash and credit card transactions as the primary mode of payment for everyday consumption [37]. Nevertheless, certain scholars have raised doubts regarding the efficacy of mobile payment. As observed by He et al. [38], the advancement of digital finance primarily spurs urban residents’ consumption, manifested in fundamental expenditures like food and clothing. Its impact on rural residents’ consumption, on the other hand, remains limited and inconsequential. Gerpott et al. [39] explored the factors influencing consumer acceptance of mobile payment systems and found that a significant number of respondents expressed concerns regarding the security of the mobile payment transaction environment. They emphasized that the assurance of secure transactions is pivotal for the advancement of mobile payments.

Jack and Suri (2014) provide an illustrative case of a Kenyan household, showcasing how mobile money effectively mitigates consumption shocks [26]. Similarly, Munyegera and Matsumoto (2016) focus on rural Ugandan households, demonstrating that the adoption of mobile money improves household welfare, measured by real per capita consumption, by facilitating remittances [40]. Many mobile payment tools, serving as crucial financial inclusion instruments, offer consumer credit functions to users. This feature plays a pivotal role in advancing the upgrading of household consumption structures [41]. At the same time, in the era of mobile payments, merchants have the ability to influence consumers’ psychological perceptions through various e-wallets, thereby stimulating consumer demand. According to the theory of psychological accounting, consumers perceive funds stored in these e-wallets as unexpected gains or “windfalls” [42,43,44]. In comparison to cash payments, mobile payments alleviate the discomfort associated with the physical loss of cash during transactions, thereby encouraging increased consumer spending [45]. Consequently, it can be argued that the adoption of mobile payments yields positive effects on household online shopping [46].

Through the combing and summarizing of the literature, it can be found that the current research on the upgrading of the consumption structure and mobile payment has been very rich, and scholars have put forward their own viewpoints from different perspectives, which provides a rich research basis for this study. Nonetheless, the present research exhibits certain limitations. There are still several aspects of the current debate about the impact of mobile payments and the upgrading of the consumer structure that are worth exploring. Firstly, although scholars have studied the impact of mobile payment on household consumption to a certain extent, the scope of the study is relatively broad, with no definition of the scope of the boundaries of the household, i.e., the study of urban and rural households has not been defined in detail, and there is the problem of unclear research objects. Secondly, household consumption encompasses a broad array of elements. Presently, most research concentrates on examining household consumption behaviors. However, research concerning the upgrading of consumption structures, particularly within rural Chinese households, remains insufficient. While a limited number of studies have touched upon the consumption structure of rural Chinese families in the context of mobile payment, they predominantly delve into its pros and cons, lacking in-depth mechanistic analysis. Thirdly, although there are a few studies involving the impact of mobile payment on the upgrading of consumption structure, the perspective of the studies is unfolded from a macro viewpoint. The analyses are mainly based on theories, lacking empirical analyses, and the credibility of the results is controversial. Hence, this study meticulously explored the nexus between mobile payment and consumption structure enhancement by constructing a model employing data from the 2017 and 2019 China Household Finance Surveys, focusing on rural households in China. Our analysis substantiates the presence of a significant correlation between mobile payment adoption and consumption structure improvements. Additionally, we delved into the underlying mechanisms, addressing gaps in prior research. These findings yield valuable policy insights for the future trajectory of mobile payment technologies and the evolution of consumption patterns.

2.2. Research Hypothesis

The widespread adoption of mobile payment technology has instigated profound transformations within Chinese rural households, exerting significant influence on the enhancement of consumption structures. To begin, mobile payment extends an array of payment channels and financial alternatives to rural households. These families can conveniently manage payments encompassing living expenses, utility bills, online shopping, airline ticket bookings, and more via mobile platforms. This assortment of payment options facilitates diverse consumer needs, spanning daily expenditures to specialized purchases, thereby contributing to the refinement and diversification of consumption structures. Moreover, mobile payments provide enhanced financial and savings avenues. Numerous mobile payment platforms incorporate savings features, permitting households to deposit surplus funds and earn financial returns. This supplementary savings and investment potential empowers households to strategize for long-term financial objectives more effectively, such as education, homeownership, and retirement; this, in turn, enhances the sustainability and strategic nature of households’ consumption structures. Furthermore, mobile payments facilitate a digitized consumption experience. Household engagement in promotions, discounts, and reward programs via mobile apps fosters heightened consumption. Furthermore, by monitoring and analyzing consumption data, mobile payment platforms can furnish personalized expenditure guidance to assist households in more effective financial management. This personalized service is anticipated to enhance consumption decision-making, prompting households to favor high-quality goods and services with long-term value. Consequently, this encourages the advancement of consumption structures. Building upon the preceding analysis, this paper posits the Research Hypothesis that mobile payment fosters the enhancement of rural households’ consumption structure.

Rural households frequently contend with cash-flow deficits, resulting in constraints on their daily lives and consumption. Such liquidity constraints may render households incapable of procuring higher-quality goods and services, managing unforeseen expenses, and formulating long-term consumption objectives. Nevertheless, the ubiquitous accessibility of mobile payments resolves this issue by affording these households a more convenient means of accessing their finances. Through mobile payments, they gain the capability to manage their e-wallets or bank accounts, facilitating money transfers, bill payments, or fund withdrawals at their discretion, regardless of location, without the need for physical visits to a bank or ATM. Such convenience simplifies the process of rural households’ adaptation to unforeseen expenses and enhances their ability to strategize for both daily living and future expenditures. Concurrently, mobile payments are frequently linked with online credit services, carrying significant ramifications for rural households. Historically, numerous rural households encountered difficulties in securing loans or credit assistance through conventional financial channels, primarily due to insufficient liquidity. This predicament curtailed their capacity to embark on entrepreneurial ventures, invest in agriculture, or participate in other economic endeavors. Nevertheless, the widespread adoption of mobile payments has facilitated these households’ access to microcredit, affording them the opportunity to participate in economic endeavors and generate additional wealth. This supplementary financial support offers increased prospects for consumption, thereby fostering the enhancement of rural households’ consumption structure. In light of the aforementioned analysis, this paper posits research hypothesis H1:

H1. 

Mobile payments can facilitate the enhancement of rural households’ consumption structure by mitigating liquidity constraints.

The widespread adoption of mobile payments in rural areas has not only provided significant financial convenience but has also positively influenced the enhancement of rural households’ consumption structure through the optimization of the consumption environment. Consumer environment refers to the external, objective factors that consumers face in the process of survival and development that have a certain impact on them. Firstly, mobile payment extends the shopping horizons and payment alternatives for rural households. Traditionally, rural residents have faced limitations in terms of where and when they can shop, often requiring long journeys to urban or town-based stores or having access only to a limited selection of local shops. Nonetheless, the advent of mobile payments has brought about a transformation. Households now possess the capability to enter online stores and peruse and procure a vast array of goods and services, unrestricted by time or location, thanks to mobile applications and internet browsers. This multiplicity of shopping avenues broadens the spectrum of options at households’ disposal, simplifying their ability to satisfy diverse needs. For instance, they can purchase various items online, such as food, clothing, electronics, and educational courses, which leads to a transformation in their consumption pattern, making it more diverse and contemporary. Additionally, mobile payments facilitate digital interactions for consumers. Numerous mobile payment platforms provide incentives like promotions, discounts, and rewards programs, motivating households to participate more actively in consumption. Mobile applications enable families to effortlessly engage in a range of shopping activities, benefiting from reduced prices and bonus rewards. These digital promotions not only make products more cost-effective for families but also enrich their shopping experience. Furthermore, mobile payment platforms have the capacity to furnish households with tailored financial guidance through the meticulous tracking and analysis of consumption data. These suggestions are rooted in consumers’ purchase history and preferences, enabling them to effectively oversee their finances and arrive at well-informed spending choices. Household choices tend to prioritize high-quality goods and services with enduring value over immediate needs, thereby fostering an elevation of the consumption structure. In light of the preceding analysis, this paper posits research hypothesis H2:

H2. 

Mobile payment can enhance the upgrading of the consumption structure in rural households through the optimization of the consumption environment.

3. Materials and Methods

3.1. Data Sources

This study utilizes data from two CHFS survey rounds, administered in 2017 and 2019 by Southwestern University of Finance and Economics. The CHFS, conducted by the China Household Finance Survey and Research Centre (CHFSRC), is a comprehensive national survey project designed to gather pertinent micro-level data on household finances. The CHFS aligns with the National Bureau of Statistics (NBS) in various dimensions, including the age composition of the sampled population, urban and rural demographic distribution, and gender composition, rendering the data highly representative. The fourth round of the 2017 CHFS surveyed 29 provinces, 355 counties (districts and cities), and 1428 communities (villages) nationwide, collecting data from a total of 40,011 households. The fifth round of the 2019 CHFS similarly covered 29 provinces, 345 counties (districts and cities), and 1359 communities (villages) across the country, gathering information from 34,643 households. This data encompassed a wide range of details, including demographic characteristics, assets and liabilities, insurance and protection, expenditures and income, financial literacy, and household education. To mitigate the influence of outliers on the estimation outcomes, we excluded samples where the total household income equaled or fell below zero. Additionally, samples containing missing values for the variables relevant to this study were omitted. Consequently, the final dataset comprised 12,324 rural households in 2017 and 11,732 in 2019, summing up to a total of 24,056 rural households.

3.2. Variable Selection

3.2.1. Explained Variables

This study examines the influence of mobile payment on the enhancement of rural households’ consumption structure. The explanatory variable, in this case, is the improvement in rural households’ consumption structure. To bolster the robustness of our findings, we employed two distinct methods to define this enhancement in the consumption structure. The initial approach, as proposed by Zhu [47], assessed the consumption structure based on the ratio of developmental consumption to total consumption. A higher coefficient value signifies a more pronounced enhancement in the consumption structure. Cultural consumption, entertainment spending, and tourism outlays are commonly regarded as key components of developmental consumption. Therefore, the total of these expenses was employed to gauge developmental consumption. The precise calculation formula is outlined below:

$${\mathrm{C}\mathrm{u}\mathrm{s}\mathrm{t}\mathrm{r}}_1=(\mathrm{C}\mathrm{c}\mathrm{o}\mathrm{n}+\mathrm{E}\mathrm{c}\mathrm{o}\mathrm{n}+\mathrm{T}\mathrm{o}\mathrm{c}\mathrm{o}\mathrm{n})/\mathrm{T}\mathrm{c}\mathrm{o}\mathrm{n}$$

(1)

In this context, ${\mathrm{C}\mathrm{s}\mathrm{t}\mathrm{r}}_1$ denotes consumption structure, $\mathrm{C}\mathrm{c}\mathrm{o}\mathrm{n}$ denotes cultural consumption, $\mathrm{E}\mathrm{c}\mathrm{o}\mathrm{n}$ denotes entertainment consumption, and $\mathrm{T}\mathrm{o}\mathrm{c}\mathrm{o}\mathrm{n}$ denotes tourism consumption.

To comprehensively capture the evolving consumption structure across various tiers, the second method draws inspiration from Wang et al. [48] and Jing et al. [49]. It categorizes residents’ consumption spending into three levels: primary, intermediate, and advanced consumption. These levels are represented by expenditures on food, residential, and transport and communication, respectively. We selected transport and communication as representatives of residents’ advanced consumption for two main reasons. Firstly, China’s consumption upgrade since 2000 has been prominently manifested in increased spending on transport and communication. Secondly, expenditures in these categories also encompass emerging consumer trends, including tourism and communication-related expenses. In this study, various coefficients are attributed to consumption expenditures at distinct levels, with larger coefficients signifying greater significance. This indicator is positively oriented, where an increase in its value signifies a more pronounced shift toward consumption upgrading. The calculation formula is outlined as follows:

$${\mathrm{C}\mathrm{s}\mathrm{t}\mathrm{r}}_2=\mathrm{F}\mathrm{c}\mathrm{o}\mathrm{n}/\mathrm{T}\mathrm{c}\mathrm{o}\mathrm{n}\times 1+\mathrm{H}\mathrm{c}\mathrm{o}\mathrm{n}/\mathrm{T}\mathrm{c}\mathrm{o}\mathrm{n}\times 2+\mathrm{C}\mathrm{o}\mathrm{c}\mathrm{o}\mathrm{n}/\mathrm{T}\mathrm{c}\mathrm{o}\mathrm{n}\times 3$$

(2)

In these equations, ${\mathrm{C}\mathrm{s}\mathrm{t}\mathrm{r}}_2$ denotes consumption structure, $\mathrm{F}\mathrm{c}\mathrm{o}\mathrm{n}$ denotes food consumption, $\mathrm{H}\mathrm{c}\mathrm{o}\mathrm{n}$ denotes residential consumption, $\mathrm{C}\mathrm{o}\mathrm{c}\mathrm{o}\mathrm{n}$ denotes transport and communication consumption, and $\mathrm{T}\mathrm{c}\mathrm{o}\mathrm{n}$ denotes total consumption.

3.2.2. Core Explanatory Variables

The primary explanatory variable in this study is mobile payment. Following the approach of Yin et al. [50] and based on the 2017 China Household Finance Survey questionnaire, households that use mobile devices like smartphones and iPads for payments (including platforms such as Alipay, WeChat, mobile banking, Apple Pay, etc.) are coded as 1, while others are coded as 0. It is worth noting that there was a modification to the questionnaire in 2019. In this version, households with accounts on payment platforms like Alipay, WeChat Payment, Jingdong Netbanking Wallet, Baidu Wallet, and other third-party payment services are coded as 1, with the rest coded as 0.

3.2.3. Control Variables

In line with established practices in the literature [51,52], this study incorporates various control variables to enhance model accuracy. These variables account for potential influences on the consumption structure of rural households, reducing the risk of estimation errors due to omitted factors. The control variables encompass the personal attributes of the household head, including age, age squared divided by 100, gender, health status, marital status, and educational attainment. Additionally, household-related characteristics, such as household size, per capita savings, and per capita income, are included. Total household assets include financial and non-financial assets. The calculation of financial assets in this context involves several components, including the balance of social security accounts, deposit account balances, cash reserves, market values of equities, bonds, funds, and financial derivatives. Additionally, it encompasses the valuation of financial products, non-RMB assets, gold holdings, other financial assets, and lending activities. Non-financial assets, as defined in this study, comprise a variety of categories, including agricultural assets, commercial and industrial assets, land holdings, housing properties, commercial establishments, vehicle assets, and other non-financial holdings. In the provided dataset, the balances of social security accounts are sourced from individual-level data. This data is aggregated at the household level by summing the individual social security account balances. All other information used in this study is derived from the household dataset. The calculation for assets per capita involves dividing the total household assets by the number of individuals in the household. The application of a logarithmic transformation to both per capita assets and per capita income is imperative for two main reasons. Firstly, there exists a substantial disparity in order of magnitude between the data pertaining to explanatory variables and control variables for per capita assets and per capita income. Secondly, the variation in the values of per capita assets and per capita income is exceptionally wide; employing a logarithmic scale mitigates this discrepancy and reduces both the differences in values and heteroskedasticity.

Table 1. Variable Definitions and Descriptive Statistics.

	Variable Name	Definition	2017		2019
			Average Value	Standard Deviation	Average Value	Standard Deviation
Variable type	Consumption structure 1	Developmental consumption as a percentage of total consumption: computed using Equation (1),	0.007	0.031	0.013	0.040
	Consumption structure 2	Consumption expenditures at various levels: computed using Equation (2),	0.967	0.370	0.981	0.390
Core explanatory variables	Mobile payment	In 2017, households were assigned a value of 1 if they made purchases using mobile phones, tablets, or other mobile terminal payments; otherwise, they received a value of 0. In 2019, households received a value of 1 if they had a third-party account and 0 otherwise.	0.105	0.307	0.220	0.414
Control variables	Age	Age of household head.	57.035	12.306	57.588	12.082
	Age sq./100	Square of the age of household head divided by 100.	34.044	14.193	34.624	13.991
	Gender	Assign 1 for male and 0 for female.	0.889	0.314	0.868	0.338
	Health status	Health status: 1 = very good; 2 = good; 3 = fair; 4 = bad; and 5 = very bad.	2.858	1.063	2.893	1.055
	Marital status	Assign 1 for married, 0 otherwise.	0.869	0.342	0.864	0.347
	Educational attainment	Education level: 1–9 (ranging from no schooling to PhD).	2.499	0.988	2.517	0.990
	Family Size	Household size.	3.536	1.756	3.411	1.727
	Per capita income	Logarithmic per capita household income.	8.973	1.289	8.974	1.264
	Assets per capita	Logarithmic per capita household assets.	10.641	1.530	10.760	1.488

presents the variable definitions and descriptive statistics for each year. First, it is evident that the consumption structure of rural households in 2019 exhibited an improvement when compared to 2017. Second, there is a noticeable upward trend in the adoption of mobile payment methods. The descriptive statistics underscore a correlation between the increasing popularity of mobile payments and the enhancement of the consumption structure.

Table 2. Mobile payments and consumption structure comparison by subgroup.

Variables	Rural Households Utilizing Mobile Payments	Rural Households Not Utilizing Mobile Payments	Mean Difference
Consumption structure 1	0.023	0.010	0.013 ***
Consumption structure 2	1.077	0.954	0.123 ***

Note: *** denotes significance at the 1% level.

presents a division of rural households into two distinct groups, depending on their usage of mobile payments. The analysis revealed that the mean consumption structure of the group utilizing mobile payments surpassed that of the group abstaining from such transactions. Notably, the t-value derived from the test of difference in means proved to be statistically significant at the 1 percent level. This significance underscores that households adopting mobile payment methods exhibited an improved consumption structure compared to those not utilizing such technology.

3.3. Model Construction

3.3.1. Model Specification

To assess the influence of mobile payment on the consumption structure of rural households, the paper formulates the subsequent linear probability model:

$${\mathrm{C}\mathrm{s}\mathrm{t}\mathrm{r}}_{\mathrm{i}}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1{\mathrm{P}\mathrm{a}\mathrm{y}}_{\mathrm{i}}+{\mathsf{\alpha }}_2{\mathrm{X}}_{\mathrm{i}}+{\mathsf{\mu }}_{\mathrm{i}}$$

(3)

where subscript $\mathrm{i}$ represents households, and ${\mathrm{C}\mathrm{s}\mathrm{t}\mathrm{r}}_{\mathrm{i}}$ represents the consumption structure of the first rural household. ${\mathrm{P}\mathrm{a}\mathrm{y}}_{\mathrm{i}}$ represents the use of mobile payments by rural households, ${\mathrm{P}\mathrm{a}\mathrm{y}}_{\mathrm{i}}=1$ represents that the rural household is using mobile payments, ${\mathrm{P}\mathrm{a}\mathrm{y}}_{\mathrm{i}}=0$ represents that the rural household is not using mobile payments, and ${\mathsf{\mu }}_{\mathrm{i}}$ denotes the random perturbation term.

In order to eliminate the influence of variables that do not vary over time on the estimation results, the following two-way fixed-effects model was developed:

$${\mathrm{C}\mathrm{s}\mathrm{t}\mathrm{r}}_{\mathrm{i}\mathrm{t}}={\mathsf{\beta }}_0+{\mathsf{\beta }}_1{\mathrm{P}\mathrm{a}\mathrm{y}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\beta }}_2{\mathrm{X}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\mu }}_{\mathrm{i}}+{\mathsf{\phi }}_{\mathrm{t}}+{\mathrm{e}}_{\mathrm{i}\mathrm{t}}$$

(4)

Here, subscript $\mathrm{i}$ represents households, and $\mathrm{t}$ signifies time. ${\mathrm{C}\mathrm{s}\mathrm{t}\mathrm{r}}_{\mathrm{i}\mathrm{t}}$ represents the consumption profile of the rural household at time $\mathrm{t}$. ${\mathrm{P}\mathrm{a}\mathrm{y}}_{\mathrm{i}\mathrm{t}}$ represents the mobile payment usage of the rural household during time $\mathrm{t}$. ${\mathrm{P}\mathrm{a}\mathrm{y}}_{\mathrm{i}\mathrm{t}}=1$ indicates that the rural household used mobile payments during time t, while ${\mathrm{P}\mathrm{a}\mathrm{y}}_{\mathrm{i}\mathrm{t}}=0$ indicates that the rural household did not use mobile payments at time $\mathrm{t}$. ${\mathrm{X}}_{\mathrm{i}\mathrm{t}}$ serves as a control variable while ${\mathsf{\mu }}_{\mathrm{i}}$ and ${\mathsf{\phi }}_{\mathrm{t}}$ denote household-specific fixed effects and year-specific fixed effects, respectively, and ${\mathrm{e}}_{\mathrm{i}\mathrm{t}}~\mathrm{N}(0,{\mathsf{\epsilon }}^2)$ represents the disturbance term.

In addition, the following econometric model is developed to examine the channels through which mobile payments act on the consumption structure of rural households:

$${\mathrm{C}\mathrm{s}\mathrm{t}\mathrm{r}}_{\mathrm{i}\mathrm{t}}={\mathsf{\gamma }}_0+{\mathsf{\gamma }}_1{\mathrm{P}\mathrm{a}\mathrm{y}\mathrm{t}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\gamma }}_2{\mathrm{P}\mathrm{a}\mathrm{y}}_{\mathrm{i}\mathrm{t}}{\mathrm{Y}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\gamma }}_3{\mathrm{Y}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\gamma }}_4{\mathrm{X}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\mu }}_{\mathrm{i}}+{\mathsf{\phi }}_{\mathrm{t}}+{\mathrm{e}}_{\mathrm{i}\mathrm{t}}$$

(5)

where ${\mathrm{Y}}_{\mathrm{i}\mathrm{t}}$ represents the mediating variable for liquidity constraints and online consumption. ${\mathrm{C}\mathrm{s}\mathrm{t}\mathrm{r}}_{\mathrm{i}\mathrm{t}}$ represents the consumption profile of the rural household at time $\mathrm{t}$. ${\mathrm{X}}_{\mathrm{i}\mathrm{t}}$ serves as a control variable while ${\mathsf{\mu }}_{\mathrm{i}}$ and ${\mathsf{\phi }}_{\mathrm{t}}$ denote household-specific fixed effects and year-specific fixed effects, respectively, and ${\mathrm{e}}_{\mathrm{i}\mathrm{t}}~\mathrm{N}(0,{\mathsf{\epsilon }}^2)$ represents the disturbance term.

3.3.2. Endogeneity Analysis

This paper tackles the concern of time-invariant omitted variables through the implementation of a two-way fixed effects model. However, there may still be endogeneity problems in model (5). Firstly, there exists an omitted variable dilemma. Unobservable time-varying factors, including environmental influences, knowledge levels, and individual preferences, may concurrently influence both the adoption of mobile payments by rural households and their consumption patterns, resulting in endogeneity; for instance, the payment habits of households and their willingness to embrace innovation. In contemporary times, there is a growing inclination towards a cashless society driven by payment habits and the willingness to embrace innovation. Some consumers may not be familiar with mobile payments, while others may be constrained by their knowledge and skill levels. Both factors can significantly influence the consumption patterns of rural households, potentially leading to endogeneity issues. On the flip side, there is the issue of measurement error. Micro-survey data is frequently impacted by measurement errors due to limitations in respondents’ comprehension of the questions posed. These errors can, in turn, impact the measurement of mobile payments and give rise to endogeneity.

To address the issue of endogeneity, this paper draws upon the research methodologies proposed by Yin et al. [50], employing the instrumental variable method to mitigate potential endogeneity concerns. Initially, rural households are categorized based on their respective communities. Subsequently, the instrumental variable for mobile payment is derived by calculating the proportion of households within each group utilizing mobile payments, excluding their own. The rationale for choosing this instrumental variable is twofold: (1) Rural areas epitomize a traditional society characterized by interpersonal familiarity, where one rural household’s choices can indeed influence the behaviors of others. In such a context, a household’s decision to adopt mobile payment is likely influenced by the mobile payment usage of neighboring households, thus meeting the correlation criterion for instrumental variable selection. (2) The utilization of mobile payment by neighboring households does not influence the consumption structure of the target household. Furthermore, the mobile payment behavior of other households lies outside the control of the interviewed households, thus meeting the exogeneity requirement for selecting instrumental variables.

4. Results

4.1. Analysis of Baseline Results

Initially, this study estimates the influence of mobile payments on the consumption patterns of rural households. The findings from the baseline regression models are provided in

Table 3. Benchmark regression results.

Explanatory Variable	Consumption Structure One
	(1) LPM	(2) FE	(3) 2SLS	(4) FE-IV
Mobile payment	0.006 *** (0.001)	0.010 *** (0.001)	0.035 *** (0.002)	0.036 *** (0.003)
Age	−0.000 (0.000)	−0.000 (0.000)	0.001 *** (0.000)	−0.001 * (0.000)
Age sq./100	0.000 * (0.000)	0.001 (0.000)	−0.000 (0.000)	0.001 *** (0.000)
Gender	−0.002 ** (0.001)	−0.005 *** (0.002)	−0.001 * (0.001)	−0.003 * (0.002)
Health status	−0.001 *** (0.000)	−0.002 *** (0.001)	−0.001 ** (0.000)	−0.002 ** (0.001)
Marital status	0.001 (0.001)	0.001 (0.001)	0.000 (0.001)	0.001 (0.001)
Educational attainment	0.002 *** (0.000)	0.002 ** (0.001)	0.001 *** (0.000)	0.001 (0.001)
Family size	0.000 (0.000)	−0.000 (0.000)	−0.001 *** (0.000)	−0.001 ** (0.001)
Per capita income	0.002 *** (0.000)	0.001 (0.000)	0.001 *** (0.000)	0.000 (0.000)
Assets per capita	0.002 *** (0.000)	0.002 *** (0.000)	0.001 *** (0.000)	0.001 (0.000)
Constant term	−0.041 *** (0.005)	−0.013 (0.014)	−0.042 *** (0.005)	0.018 (0.015)
Household fixed effects	No	Yes	No	Yes
Year fixed effects	Yes	Yes	Yes	Yes
N	24,056	24,056	24,056	24,056
${\mathrm{R}}^2$	0.052	0.022
First stage F-value			897.12	189.16
Instrumental variable T-value			47.00	39.73

Note: ***, **, and * indicate significant at the 1%, 5%, and 10% levels, respectively; () are heteroskedasticity robust standard errors.

. Both the linear probability model (LPM) in column (1) and the two-way fixed-effects model (FE) in column (2) demonstrate a substantial increase in the share of developmental consumption among rural households due to mobile payment usage, significant at the 1% level; this indicates that mobile payment significantly enhances the proportion of developmental consumption within rural households’ total expenditure, thereby fostering the advancement of consumption patterns. The estimation results from the linear probability model (LPM) in column (1) indicate that the adoption of mobile payments leads to a statistically significant increase of 0.6 percentage points in the portion of developmental consumption within the overall consumption of rural households. This significance is observed at the one percent level. The findings derived from the estimation using the two-way fixed effects model (FE) in column (2) indicate that mobile payment usage leads to a statistically significant increase of one percentage point in the proportion of developmental consumption within the overall consumption of rural households. This significance is observed at the one percent level. Evidently, mobile payments play a substantial role in augmenting the share of developmental consumption within rural households and facilitating the enhancement of consumption structure. Additionally, the estimation results for control variables reveal that having a male household head is not conducive to such structural improvement. The poorer the self-assessed health status of the household head, the heavier the financial burden of medical expenses on the household and the lower the likelihood of the household upgrading its consumption structure. As the household head’s years of education increase, so does the household’s level of human capital. Consequently, there is a likelihood of higher wages, increased per capita household income, and an increased probability of upgrading the household’s consumption structure.

While this paper accounts for variables influencing the consumption structure of rural households and the diverse characteristics of households in model (5), the model may still be susceptible to time-varying unobservable factors impacting both the consumption structure of rural households and the adoption of mobile payments. Such factors have the potential to introduce estimation bias. In order to mitigate the endogeneity issue arising from time-varying omitted variables and measurement errors, this study employs the proportion of households within the same community utilizing mobile payments, distinct from their own, as an instrumental variable for estimating mobile payment adoption. Columns (3) and (4) present the estimation results obtained through a two-stage least squares (2SLS) approach and a two-way fixed effects model (FE), which combines instrumental variables methodology. It is established that when the F-value exceeds a critical threshold of 16.38 at a 10 percent bias level, it rejects the null hypothesis of weak instrumental variables, signifying the absence of a weak instrumental variables issue. The t-values of the instrumental variables have successfully cleared the 1% significance test, signifying that these variables fulfill the correlation requirement. The results of the estimation in column (3) indicate that the utilization of mobile payments leads to a statistically significant 3.5 percent increment in the proportion of developmental consumption within the total expenditure of rural households, significant at the one percent level. Column (4) presents the results of the Fixed Effects Instrumental Variable (FE-IV) estimation, indicating that the adoption of mobile payments leads to a significant 3.6 percent rise in the proportion of developmental consumption within the total expenditures of rural households, significant at the one percent level. The aforementioned findings additionally indicate that the adoption of mobile payments by rural households encourages an upgrade in their consumption structure.

4.2. Robustness Tests

4.2.1. Robustness Test I: Probit Model Estimation

To assess the robustness of our findings, we employed the Probit model to analyze the influence of mobile payments on the consumption structure of rural households. The results of this analysis are presented in

Table 4. Robustness test estimation results.

Explanatory Variable	Consumption Structure One		Consumption Structure Two
	(1) Probit	(2) Ivprobit	(3) LPM	(4) FE	(5) 2SLS	(6) FE-IV
Mobile payment	0.307 *** (0.024)	2.270 *** (0.032)	0.067 *** (0.007)	0.065 *** (0.012)	0.140 *** (0.022)	0.090 *** (0.026)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes
Household fixed effects	No	No	No	Yes	No	Yes
Year fixed effects	Yes	Yes	Yes	Yes	Yes	Yes
N	24,056	24,056	24,056	24,056	24,056	24,056

Note: *** denotes significance at the 1% level.; () are heteroskedasticity robust standard errors.

, with column (1) indicating a substantial 30.7 percent increase in the proportion of developmental consumption within the total expenditures of rural households, a significance level of one percent. Further analysis, employing the proportion of households within the same community using mobile payments other than their own as an instrumental variable for mobile payment, reveals in column (2) that mobile payments substantially enhance the proportion of developmental consumption within the total expenditures of rural households. This effect is statistically significant at the 1% level. These findings support the notion that mobile payments play a role in elevating the consumption structure of rural households. The Probit model estimations reaffirm the robustness of the preceding results.

4.2.2. Robustness Test II: Redefining the Concept of Consumption Structure

Building upon our prior analysis and taking inspiration from Wang et al. [48] and Jing et al. [49], we redefined the concept of consumption structure to capture changes more comprehensively at various levels. Instead of relying solely on the share of developmental consumption in total expenditure, we now define consumption structure using expenditure profiles at different levels. The estimation results from the linear probability model (LPM) presented in Table 4, column (3) indicate that the utilization of mobile payments enhances the consumption structure of rural households by 6.7%, a significance level of 1%. In column (4), the results from the two-way fixed effects model (FE) estimation demonstrate that mobile payments enhance the consumption structure of rural households by 6.5%, a significance level of 1%. Columns (5) and (6) present the estimation results from two-stage least squares (2SLS) and two-way fixed effects model (FE) approaches, respectively, which employ an instrumental variables methodology. In column (5), the results of the estimation reveal that mobile payments enhance the consumption structure of rural households by 14%, a significance level of 1%. In column (6), the FE-IV estimation results indicate that mobile payments enhance the consumption structure of rural households by 9%, a significance level of 1%. These findings confirm the robustness of the results, even after redefining the explanatory variables.

4.2.3. Robustness Test III: Propensity Score Matching

The adoption of mobile payments by rural households presents a self-selection challenge and deviates from the principles of random sampling. Consequently, this study employs the Propensity Score Matching (PSM) method to rectify potential selectivity bias.

The utilization of mobile payments among rural households presents a self-selection challenge, deviating from the principles of random sampling. To mitigate sample selection bias, this study employs the Propensity Score Matching (PSM) method to create a “counterfactual” scenario for analysis. The Propensity Score Matching method helps mitigate estimation bias resulting from sample self-selection, thereby facilitating a more effective reflection of an individual’s implementation of a specific item. In this study, we classified samples into two groups: one using mobile payments (the experimental group) and the other not using mobile payments (the control group). We estimated the likelihood of using mobile payments using the Logit model, with the fitted conditional probability being referred to as the propensity score. Subsequently, the estimated propensity scores were employed for nearest neighbor matching (k = 1), radius matching (caliper of 0.05), and kernel matching, respectively.

Regression analysis is conducted using the matched samples, and the results are presented in

Table 5. Robustness test estimates—Propensity Score Matching.

Variable	Explanatory Variable: Consumption Structure One
	Nearest Neighbor Matching	Radius Matching	Nuclear Matching
Experimental group	0.023	0.023	0.023
Control group	0.014	0.015	0.015
ATT	0.009	0.008	0.008
Standard error	0.001	0.001	0.001
T-value	6.20	8.76	8.73

Note: To calculate the average processing effect of mobile payments, the following steps are undertaken: firstly, a Logit regression is executed, incorporating variables such as the household head’s age and its squared term divided by 100, the household head’s gender, the household head’s health, the household head’s marital status, the household head’s education, household size, household per capita income, household per capita assets, and year dummy variables to estimate the propensity score; subsequently, one-to-one nearest neighbor matching, radius matching, and kernel matching are employed.

. It is evident that, among the successfully matched samples, mobile payments consistently and significantly elevate the proportion of developmental consumption within rural households’ total expenditure, thereby facilitating the enhancement of the consumption structure. Notably, the results obtained through the application of the three matching methods exhibit coherence, with the estimated coefficients closely aligned. This coherence reinforces the robustness and credibility of the estimation results.

4.3. Mechanism Analyses

4.3.1. Mechanism Analysis I: Liquidity Constraints

Mobile payments enhance financial institutions’ understanding of individuals’ credit status, mitigating information asymmetry and bolstering rural households’ access to credit. Furthermore, the introduction of credit services by platforms like Alipay, alongside internet-based lending products, such as microloans and payment-linked borrowing, tailored to meet customers’ online, real-time, and fragmented requirements, has further eased rural households’ access to credit. Consequently, mobile payments can assist households in alleviating liquidity constraints by expanding credit accessibility, as corroborated by the empirical findings presented in this study.

Following the definition by Zeldes [53], this paper categorizes rural households with financial assets totaling less than two months of permanent income as liquidity-constrained households, assigning a binary value of 1 for those meeting this criterion and 0 otherwise. To investigate whether mobile payments can enhance consumption structure by alleviating the liquidity constraints of rural households, this study introduces an interaction term between mobile payments and liquidity constraints.

Table 6. Estimated results of mechanism analysis—liquidity constraints.

Explanatory Variable	Consumption Structure One				Consumption Structure Two
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	LPM	FE	2SLS	FE-IV	LPM	FE	2SLS	FE-IV
Mobile payment	0.022 *** (0.001)	0.025 *** (0.002)	0.118 *** (0.009)	0.116 *** (0.010)	0.167 *** (0.013)	0.239 *** (0.023)	0.423 *** (0.079)	0.303 *** (0.087)
Mobile payment × liquidity constraints	0.041 *** (0.003)	0.039 *** (0.005)	0.216 *** (0.018)	0.215 *** (0.020)	0.261 *** (0.028)	0.464 *** (0.051)	0.741 *** (0.150)	0.587 *** (0.170)
Liquidity constraints	−0.003 *** (0.001)	−0.001 (0.001)	0.000 (0.001)	0.002 * (0.001)	−0.021 *** (0.005)	−0.027 *** (0.011)	−0.013 ** (0.006)	−0.025 ** (0.011)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Household fixed effects	No	Yes	No	Yes	No	Yes	No	Yes
Year fixed effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	24,056	24,056	24,056	24,056	24,056	24,056	24,056	24,056

Note: ***, **, and * indicate significant at the 1%, 5%, and 10% levels, respectively; () are heteroskedasticity robust standard errors.

presents the estimation results for Consumption Structure One and Consumption Structure Two. The linear probability model (LPM) estimates reveal that the coefficient of the interaction term between mobile payments and liquidity constraints is both positive and highly significant at the 1% level. This finding substantiates that mobile payments play a more substantial role in improving the consumption structure of households facing liquidity constraints. Furthermore, to mitigate the influence of time-invariant variables on the outcomes, the Fixed Effects (FE) estimation results confirm a significantly positive coefficient on the interaction term between mobile payments and liquidity constraints, aligning with the earlier findings. Additionally, instrumental variables are employed to address the endogeneity concern in estimation. The Fixed Effects Instrumental Variables (FE-IV) estimation results reveal that mobile payments significantly facilitate the improvement of consumption structure among households facing liquidity constraints. These findings substantiate that mobile payments can enhance the consumption structure of rural households by alleviating liquidity constraints, thereby confirming Hypothesis 1.

4.3.2. Mechanism Analysis II: Enhancing the Consumer Environment

The advent of mobile payments has empowered rural households to engage in online consumption, significantly enhancing the variety and accessibility of products. This improvement in the consumption environment necessitates the use of a mediator variable called “whether the household has prior online shopping experience.” It is assigned a value of 1 if the household has engaged in online shopping and 0 if it has not.

Table 7. Estimated results of the institutional analysis—enhancing the consumer environment.

Explanatory Variable	Consumption Structure One				Consumption Structure Two
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	LPM	FE	2SLS	FE-IV	LPM	FE	2SLS	FE-IV
Mobile payment	0.001 (0.001)	0.007 *** (0.002)	0.053 *** (0.004)	0.055 *** (0.005)	0.053 *** (0.010)	0.047 *** (0.016)	0.192 *** (0.039)	0.097 ** (0.046)
Mobile payment × online shopping	0.005 *** (0.001)	0.004 (0.003)	0.041 *** (0.004)	0.046 *** (0.006)	0.001 (0.014)	0.034 (0.025)	0.133 *** (0.040)	0.018 (0.051)
Online shopping	0.005 *** (0.001)	0.002 (0.002)	0.013 *** (0.001)	0.014 *** (0.002)	0.038 *** (0.009)	0.001 (0.017)	0.060 *** (0.011)	0.011 (0.020)
Control variables	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Household fixed effects	No	Yes	No	Yes	No	Yes	No	Yes
Year fixed effects	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	24,056	24,056	24,056	24,056	24,056	24,056	24,056	24,056

Note: *** and ** indicate significance at the 1% and 5% levels, respectively; () are heteroskedasticity robust standard errors.

presents the estimation results for Consumption Structure One and Consumption Structure Two. The linear probability model (LPM) estimates reveal that the coefficient of the interaction term between mobile payments and online shopping is not only positive but also highly significant at the 1% level. This finding corroborates that mobile payments play a more substantial role in improving the consumption structure of households engaged in online purchases. The Fixed Effects Instrumental Variables (FE-IV) estimation results demonstrate that mobile payments, when used by households engaged in online shopping, substantially contribute to the enhancement of consumption structure. These findings affirm that mobile payment fosters the improvement of the consumption structure of rural households by enhancing the consumer environment, thus confirming Hypothesis 2.

4.4. Analysis of Heterogeneity

The age and education level of the household head serve as crucial indicators of the human capital endowment within rural households [54]. Building upon prior research, this study examines potential heterogeneity in the impact of mobile payments on the consumption structure of rural households, with a primary focus on the age and education level of the household head. In this study, we categorized the sampled households into two age groups based on the age of the household head (“<50” and “≥50 years old”). Additionally, they were classified into low- and high-education groups, depending on whether the head of the household had completed junior high school education. Subsequently, we conducted separate group regressions, and the results are presented in

Table 8. The regression results based on age groups.

Explanatory Variable	High Age Group		Low Age Group
	Consumption Structure One	Consumption Structure Two	Consumption Structure One	Consumption Structure Two
Mobile payment	0.0080 *** (0.0019)	0.0774 *** (0.0165)	0.0076 *** (0.0023)	0.0507 ** (0.0239)
Control variables	Yes	Yes	Yes	Yes
Household fixed effects	Yes	Yes	Yes	Yes
Year fixed effects	Yes	Yes	Yes	Yes
N	17,693	17,693	6363	6363

Note: *** and ** indicate significance at the 1% and 5%levels, respectively; () are heteroskedasticity robust standard errors.

and

Table 9. Regression results for educational subgroups.

Explanatory Variable	High Age Group		Low Age Group
	Consumption Structure One	Consumption Structure Two	Consumption Structure One	Consumption Structure Two
Mobile payment	0.0088 *** (0.0041)	0.1293 *** (0.0379)	0.0085 *** (0.0013)	0.0573 *** (0.0128)
Control variables	Yes	Yes	Yes	Yes
Household fixed effects	Yes	Yes	Yes	Yes
Year fixed effects	Yes	Yes	Yes	Yes
N	2908	2908	21,148	21,148

Note: *** indicates significance at the 1% level; () are heteroskedasticity robust standard errors.

. Examining the age-based segmentation in Table 8, it becomes evident that mobile payment plays a substantial role in enhancing the consumption structure of rural households across both higher and lower age groups. However, delving into the precise coefficients reveals a more pronounced impact of mobile payment within the higher age groups. This observation aligns with expectations, as older household heads typically possess greater leisure time, facilitating increased engagement in travel and recreational activities.

Examining the educational subgroups in Table 9, it becomes evident that mobile payment substantially enhances the consumption structure of rural households in both high and low-education groups. However, delving into the precise coefficients reveals a more pronounced impact of mobile payment within the high-education group. This distinction arises from the fact that households with highly educated heads exhibit a greater inclination toward cultural consumption, an area where mobile payment offers a convenient, diverse, and socially integrated means to fulfill their cultural needs. Consequently, they are more likely to derive benefits from mobile payment’s facilitative role in cultural consumption. Highly educated household heads tend to show a greater propensity for engaging in cultural activities, such as art exhibitions, museum tours, concerts, and theater performances. Mobile payments offer a convenient avenue for acquiring tickets to these cultural events. Furthermore, individuals with higher levels of education often prioritize self-improvement and education, leading them to invest in educational and culturally enriching products, such as online courses, e-books, academic journals, and cultural magazines. Mobile payments offer a convenient avenue for acquiring these resources online, facilitating access to knowledge and cultural information. This drive for self-improvement often prompts increased investment in cultural consumption, encompassing the acquisition of learning materials and cultural resources. Consequently, household heads with higher levels of education are more inclined to allocate funds to cultural expenditures through mobile payments, thereby bolstering the enhancement of the consumption structure.

5. Discussion

5.1. Integration with Previous Studies

Presently, scholarly inquiry into the influence of mobile payments on consumption primarily centers on their impact on overall expenditure. Certain researchers have utilized cross-sectional data from the 2017 China Household Finance Survey, revealing a notable disparity in total consumption between households employing mobile payments and those abstaining. Additionally, these studies identify consumer credit as an intermediary factor and highlight the moderating role of financial literacy. Moreover, a subset of scholars has delved into the impact of mobile payments on consumption structure. Utilizing data from the 2017 China Household Finance Survey, these researchers have demonstrated a significant augmentation in the proportion of development and enjoyment-related expenditures within total consumption. This phenomenon has been instrumental in advancing the evolution of consumption patterns.

Nevertheless, the analyses conducted by these scholars often lack empirical support, relying heavily on theoretical foundations. Therefore, this study enriches this area of research in three key ways. Firstly, leveraging data from the 2017 and 2019 China Financial Surveys, this paper employs a two-way fixed-effects model. This approach not only compares the consumption structures of rural households utilizing mobile payments with those abstaining but also delves into the dynamic impact of mobile payments on the consumption patterns of rural households.

Additionally, this study enhances its methodological rigor by employing the average mobile payment usage of households within the same community as an instrumental variable, thereby bolstering result reliability. Lastly, the paper delves deeply into the mechanisms through which mobile payments influence the consumption structure of rural households. This exploration encompasses an in-depth analysis of how mobile payments alleviate liquidity constraints and optimize consumption environments. This study also incorporates heterogeneity analyses based on the age and education levels of household heads, coupled with empirical tests. These meticulous analyses enrich the existing research in this domain.

5.2. Comparison with Relevant International Studies

Mobile payments have sparked a significant increase in digital consumption on a global scale. Whether in China or other parts of the world, individuals increasingly opt for mobile applications to shop and settle bills, leading to a surge in online retail activity. Simultaneously, the widespread adoption of mobile payments has expanded access to financial services, particularly in developing nations. Even individuals without traditional banking services can now conveniently engage in economic activities using their smartphones. This heightened convenience not only alleviates temporal and spatial constraints associated with shopping and transactions but also encourages consumers to make frequent small purchases. Consequently, this trend has contributed to a noteworthy shift in global consumption patterns.

Nevertheless, the influence of mobile payments on consumption in China stands as a distinctive phenomenon in comparison to other nations and regions. Within China, mobile payments have experienced a swift and unparalleled surge in popularity, establishing themselves as the primary mode of payment. Leading mobile payment applications like Alipay and WeChat Pay have monopolized the Chinese market. Citizens can seamlessly utilize their smartphones for payments across myriad everyday situations, whether grocery shopping, dining at restaurants, or engaging in online shopping [55,56]. This seamless and expeditious payment process characterizes the Chinese consumer experience. The observed trend finds its roots in the rapid evolution of China’s mobile payment market. Propelled by the pervasive adoption of smartphones and the relentless advancement in mobile payment technologies, Chinese consumers exhibit a pronounced preference for mobile-based transactions over traditional credit card methods. In contrast, developed nations, such as the United States and Canada, predominantly rely on credit card payments [57,58,59], a practice widely embraced by both consumers and merchants. These divergent consumer cultures and payment habits underscore the distinctiveness of China’s mobile payment advancement, a phenomenon that has garnered significant global interest.

Differences in policy support and regulatory frameworks exist among various countries and regions [60]. In its endeavor to foster consumer development through mobile payments, China has implemented multifaceted policy initiatives. Financial regulation, as a primary focus, has been bolstered by the Chinese government, which has enhanced oversight of the payment industry through both the central bank and other financial regulatory bodies. This meticulous approach ensures the security and stability of mobile payment systems. Moreover, government entities in China have proactively championed mobile payment applications. These efforts span diverse sectors, including public transportation and social welfare, thereby significantly contributing to the widespread adoption of mobile payment methods. Simultaneously, the Chinese government actively fosters fintech innovation, offering substantial policy support and regulatory flexibility. This proactive stance significantly propels the ongoing advancements in mobile payment technology within the nation. In contrast, the United States operates within a framework dominated by market dynamics and collaboration. Within the U.S. mobile payment sector, a notable degree of freedom characterizes the market [61]. Government initiatives primarily focus on stimulating market competition and incentivizing enterprises to innovate payment methodologies. Regarding partnerships, the U.S. government collaborates with prominent corporations and financial institutions, urging them to offer enhanced payment services. However, the government does not prioritize specific mobile payment policies in its initiatives. In Europe, policies have primarily centered on consumer protection and the facilitation of internal market integration. European nations emphasize stringent regulations for mobile payment service providers, ensuring compliance to safeguard user information’s security and privacy [62]. Simultaneously, the European Union (EU) actively fosters cross-border payments among member states, striving to enhance the integration of the European internal market and establish a more convenient international payment system through the promotion of mobile payments. These policy initiatives have led to the creation of unique mobile payment ecosystems in various member countries.

5.3. Research Limitations and Future Research Directions

This study empirically investigates the influence of mobile payment on the enhancement of rural households’ consumption structure, utilizing unbalanced panel data extracted from the China Financial Survey of 2017 and 2019. The research also encompasses robustness tests, mechanism analyses, and heterogeneity analyses. Despite the extensive efforts undertaken in this study, certain limitations have been identified in the acquisition and utilization of the data.

A limitation of this study lies in the utilization of unbalanced panel data solely from the China Financial Survey conducted in 2017 and 2019. Unfortunately, data for the year 2021 has not been published yet. Given the rapid surge in mobile payments in China between 2019 and 2021, the inclusion of data from this period would significantly enhance the persuasiveness of the article.

Secondly, introducing a logarithmic transformation to variables that can logically accommodate negative values can introduce selectivity bias into the sample. In this study, applying a logarithmic transformation to per capita household income excludes rural households with incomes equal to or less than zero. Nevertheless, situations arise where total household income may fall below zero due to business losses. Utilizing the logarithmic transformation in such cases leads to the exclusion of samples with negative total household incomes, introducing selective bias into the sample.

Thirdly, in the realm of mechanism analysis, mobile payment serves a multifaceted purpose. Not only does it alleviate liquidity constraints and optimize the consumption environment, but it also mitigates the discomfort associated with payments. The inherent characteristics of mobile payment methods, characterized by low transparency and limited connectivity, reduce consumers’ discomfort during transactions and heighten the pleasure derived from consumption. This phenomenon acts as a catalyst, encouraging rural residents to engage in seemingly irrational high-end expenditures, thereby facilitating the enhancement of consumption structures. It is imperative to note, however, that due to the absence of specific inquiries about payment discomfort and substitution in the Chinese financial questionnaire, this paper does not explore this particular aspect of the mechanism.

In future research, we propose two key approaches. Firstly, a continued focus on the data from the China Household Finance Survey is crucial. Utilizing the latest available data from 2021 will enable us to conduct up-to-date research. Secondly, employing meticulously designed questionnaires and involving team members in the research process will yield significant benefits. This methodology not only eliminates the impact of outliers on the model but also allows for a more nuanced analysis of the role of mobile payment in shaping the consumption structure of rural households.

6. Conclusions and Policy Recommendations

Currently, China’s economy is undergoing a transition towards higher quality, emphasizing the crucial need to enhance the consumption capacity of rural residents. The proliferation of the Internet and e-commerce has opened avenues for rural residents to enhance their purchasing capabilities. Particularly, the emergence of mobile payment methods has streamlined transactions for rural consumers, thereby facilitating the elevation of their consumption patterns. In this context, this study delves into the influence of mobile payment on the enhancement of consumption structures within Chinese rural households and presents pertinent policy recommendations.

6.1. Conclusions

Enhancing the consumption structure in rural areas holds paramount importance in driving rural economic and social progress, elevating the quality of life for farmers, fostering the modernization and revitalization of rural industries, and achieving long-term sustainability. This study conducts an empirical analysis of the influence of mobile payment on the enhancement of consumption patterns among Chinese rural households, leveraging data from CHFS2017 and CHFS2019. This study’s findings indicate three key outcomes: firstly, mobile payment significantly drives the enhancement of consumption structures in rural households. Secondly, mechanism analysis underscores mobile payment’s vital role in mitigating mobility constraints and enhancing the consumption environment, subsequently contributing to the betterment of rural households’ consumption structures. Lastly, the heterogeneity analysis reveals that mobile payment exerts a more pronounced influence on the consumption structure improvement among rural households with older household heads and higher levels of education.

6.2. Policy Recommendations

Building upon the aforementioned findings, this paper provides the following policy recommendations: First, foster rural mobile payment infrastructure development. On the one hand, the government should boost investments in rural mobile payment infrastructure, encompassing mobile network base stations and fiber-optic network coverage, to ensure consistent network connectivity across rural regions. On the other hand, there should be incentives for technology companies to intensify their research and development efforts in mobile payment technologies; this would encourage the advancement of innovative payment technologies and service models tailored to the specific needs of rural residents. Secondly, there is a need to enhance smartphone penetration in rural regions. The government should alleviate the financial strain on farmers by implementing policies like subsidies and tax reductions for smartphone purchases. Additionally, there should be incentives for the development of rural-specific apps tailored to farmers’ requirements. Offering services such as agricultural product trading, agricultural technology information dissemination, and agricultural insurance through smartphones can entice farmers to adopt this technology. Moreover, comprehensive educational programs should be conducted to disseminate smartphone knowledge in rural areas. These programs should encourage farmers to engage in e-commerce activities through smartphones and provide training on fundamental smartphone operations and app usage. This will increase their awareness of smartphones and drive up demand for these devices. Thirdly, there is a need to enhance rural financial literacy. Financial institutions should play an active role in rural financial education. They can accomplish this by offering professional financial knowledge lectures and training, highlighting successful rural financial case studies, sharing rural success stories in the financial sector, and fostering farmers’ enthusiasm for learning about financial concepts; this will enable farmers to gain a deeper understanding of financial products and services, ultimately alleviating the constraints related to liquidity that hinder the improvement of consumption patterns.