Effect of Digital Inclusive Finance on the Upgrading of the Consumption Structure of Rural Residents in China

Abstract

In order to study whether digital inclusive finance has played an active role in promoting rural residents’ consumption and optimizing consumption structure under the background of implementing rural revitalization strategy in China. This study uses macro data from 31 provincial regions in China from 2011 to 2021, and use a fixed-effect model to empirically study the impact of China’s digital inclusive finance development on the scale and structure of rural residents’ consumption, as well as the mechanism and effect differences. The study finds that the development of digital inclusive finance in China can drive the consumption level of rural residents to increase and optimize the consumption structure of rural residents, thus achieving the consumption upgrading of rural residents. At the same time, based on the test of mediating effect, it is found that the development of digital inclusive finance in China can have a positive impact on the consumption upgrading of rural residents by promoting the development of e-commerce. In addition, for different levels of consumption level of rural residents, the development of digital inclusive finance has a positive impact on it, and there is no significant marginal increase or decrease effect. However, for different levels of consumption structure of rural residents, the development of digital inclusive finance has a more obvious promoting effect at its low level. When the development of digital inclusive finance reaches a certain height, its effect on the consumption level of rural residents will weaken, but it will be more significant in optimizing and improving the consumption structure of rural residents in China. There are also differences in the effects of digital inclusive finance on the consumption upgrading of rural residents in different regions, incomes and industrial structures.

Plain language summary

The influence of the digital inclusive finance on the consumption structure upgrading of rural residents in China

Purpose: Under the background of China’s rural revitalization strategy, this study explores the role of digital inclusive finance in boosting rural residents’ consumption and optimizing consumption structure, so as to provide guidance for economic growth and consumption quality in rural areas. Methods: This study uses macro data from 31 provincial regions in China from 2011 to 2021, and uses a fixed-effect model to empirically study the impact of China’s digital inclusive finance development on the scale and structure of rural residents’ consumption, as well as the mechanism and effect differences. Conclusions: The study finds that the development of digital inclusive finance in China can drive the consumption level of rural residents to increase and optimize the consumption structure of rural residents in structural characteristics, mechanism identification and have effect differences. Implications: Efforts will be made to improve the technical level of digital inclusive finance and continue to promote the basic construction of digital inclusive finance. At the same time, vigorously cultivate digital talents in rural areas, which is very important for implementing digital inclusive finance and really benefiting rural residents. Limitations: The core explanatory variable of this paper is Peking University Digital Inclusive Finance Index, but the data source of the index is mainly internet financial institutions such as Ali-pay, which mainly serve online users, and the demand for groups without internet access and offline financial services may not be fully covered; In addition, the digital inclusive finance index uses a number of indicators to reflect the development level of digital inclusive finance. With the rapid development and changes of digital inclusive finance, the weights of some indicators may need to be adjusted in time.

Introduction

The report of the 20th National Congress of the Communist Party of China clearly pointed out that achieving common prosperity is the essential requirement of Chinese path to modernization, and Chinese path to modernization is the modernization of common prosperity for all people. The key to achieving common prosperity lies in increasing farmers’ income, expanding channels for farmers to increase their income, and thus narrowing the income gap between urban and rural areas (K. Liang & Liu, 2022). Against the backdrop of the shift from poverty alleviation to rural revitalization, promoting the upgrading of rural residents’ consumption has become an important engine for driving sustained economic growth in China. The consumption of rural residents plays a very important role in rural revitalization. The No. 1 central document of the Central Government in 2022 clearly pointed out that rural consumption should be improved, expanded and upgraded. However, the current consumption structure of rural residents in China is still relatively lagging behind, which not only limits the optimization and upgrading of the economic structure, but also affects the high-quality development of the overall economy. Therefore, actively exploring paths that can effectively promote the upgrading of rural residents’ consumption is of great significance for promoting China’s economy to move toward a higher level of development.

In the process of steadily advancing the rural revitalization strategy, the economy in rural areas of China has developed rapidly, and the consumption market hidden behind it has enormous space and potential, which has become a “blue ocean” driving domestic consumption growth. On the one hand, the huge base of rural residents has enormous potential in consumption potential. In 2022, the resident population in rural areas reached 490 million, which can become a key component of a large-scale market. At the same time, the per capita disposable income of rural residents in 2022 only accounted for 40.9% of urban residents, and the per capita consumption expenditure only accounted for 54.7% of urban residents. It can be seen that the consumption potential has not been fully released and needs to be further explored. On the other hand, new consumption models brought about by the development of information technology are expected to bring new consumption growth opportunities to rural areas. Based on inclusive finance, digital inclusive finance generated by using digital technologies such as big data and cloud computing has broadened the scope of financial services and reduced the cost of financial services, which is of great significance in narrowing the gap between urban and rural areas and promoting consumption of rural residents.

The economic effect of China digital inclusive finance has naturally been widely discussed at home and abroad. Digital inclusive finance is a financial service model based on Internet and big data technology and aimed at improving the accessibility of financial services, reducing the cost of financial services and improving the quality of financial services. China digital inclusive finance has innovated the traditional credit evaluation system through digital technology, enabling residents who could not get loans under the original conditions to get loans, reducing their uncertainty about the future and reviving consumer confidence (Song et al., 2020). From the microscopic point of view, Xie et al. (2018) and B. Liang and Zhang (2019) believe that the development of digital inclusive finance in China plays a important role in the practice of “mass entrepreneurship and innovation.” In addition, Lai et al. (2020) after in-depth study of the relationship between household consumption and digital inclusive finance in China, think that under the background of the continuous development of digital inclusive finance, the influence of temporary impact on household consumption is weakened. Wan et al. (2020) further discovered this important influence channel to ease financing constraints on the basis of the above. Tang et al. (2020) conducted a more in-depth study, his research conclusion shows that digital inclusive finance can better drive enterprise innovation in areas with poor financial endowment, and the driving effect is more significant under certain financial supervision constraints.

China digital inclusive finance occupies a decisive position in the consumption environment. J. Li and Li (2017) found that the popularity of credit cards has promoted residents’ consumption of high-level consumer goods (such as cars, black and white household appliances, etc.), which shows that innovative financial services can promote the improvement of residents’ consumption quality to some extent. Taking Chinese households holding financial assets as the research sample, Peng and Liu (2024) discover that the development of digital inclusive finance has positive influence on household financial resilience. Ge et al. (2022) found that the development of digital inclusive finance can reduce the financing cost of residents and improve the entrepreneurial activity of self-employed, small and micro enterprises and labor-intensive industries; It can expand farmers’ credit supply, promote agricultural and rural employment, improve farmers’ wage and operational income, and thus narrow the income gap between urban and rural areas. Sharmila (2019) believes that digital inclusive finance has expanded the channels of obtaining credit, optimized the allocation of resources and provided new opportunities for economic development. Zhao and Gai (2020) found that credit supply not only meets the basic living needs of residents, but also provides more financial support for their entertainment consumption. From a micro perspective, Yi and Zhou (2018) found that the popularization of digital inclusive finance in China provides convenient and equal financial services for more people by using digital technology, which promotes the activity and development of the consumer market, and further confirms the two important channels of influence: easing liquidity constraints and improving payment convenience. However, X. Zhang et al. (2019, 2020) questioned this channel to ease the liquidity constraint, arguing that this channel is not an effective way to enhance residents’ consumption in digital inclusive finance. In addition, the research of He and Song (2020) supplemented the uncertainty mechanism of China digital inclusive finance’s influence on residents’ consumption, and further consolidated the conclusion that digital inclusive finance promoted the consumption level.

Throughout the existing literature, this study finds that it is mainly discussed from the theoretical and empirical aspects. From the theoretical research, many scholars mainly focus on how digital inclusive finance can promote residents’ consumption upgrading by providing more extensive financial services, more effective payment methods and more optimized consumption structure. For example, digital inclusive finance can ease liquidity constraints, increase residents’ property income and optimize the consumption payment environment, thus promoting the upgrading of residents’ consumption. From the empirical research, scholars found that the development of digital inclusive financial system has improved the purchasing power of consumers, promoted the growth of consumer market, significantly improved the consumption level of residents, and also played a significant role in optimizing consumption structure.

Since the concept of digital inclusive finance was first put forward at the G20 Hangzhou Summit in 2016, many scholars have done a lot of research on its influence on residents’ consumption. However, there are few literature on the influence of rural residents’ consumption ability, and rural residents have their own consumption characteristics, so the research conclusions are often different.

Based on this, the study is based on China’s provincial-level agricultural economic data from 20011 to 2021, employing a fixed-effect model to accurately measure the impact of China’s digital inclusive finance development on the scale and structure of rural residents’ consumption. Then, the mediation effect model is used to further study the intermediary effect e-commerce in upgrading rural residents’ consumption. In addition, heterogeneity was used to test the differences in the consumption upgrading of Chinese rural residents in different regions, incomes, and industrial structures through digital inclusive finance, providing empirical references for further promoting digital inclusive finance reform and better promoting the upgrading of rural residents’ consumption structure.

In view of this, this study attempts to make useful additions to the existing research and make the following contributions. First, as for the research object, the rural residents are selected as the research object, and the analysis is made from two angles: the consumption level and the consumption structure. Compared with the previous literature, this research angle is comprehensive, which has contributed to enriching the empirical evidence. Second, regarding the indicators for measuring the consumption structure of rural residents, we construct the index of “the consumption structure upgrading coefficient of rural residents,” which combines the method of defining high/low-level consumption by W. Yang et al. (2021) and the method of constructing industrial structure upgrading coefficient by Xu and Jiang (2015). In addition, in the aspect of mechanism research, this paper empirically studies the influence path that the development of digital inclusive finance in China promotes the consumption level and consumption structure of rural residents by influencing the development of e-commerce, which can supplement the existing theoretical research. At the same time, this paper uses quantile model and threshold effect test to further explore the relationship between the development of digital inclusive finance in China and the consumption upgrading of rural residents, further enriching the existing research.

Conceptual Review and Hypotheses Development

According to the previous literature review, the development of digital inclusive finance in China can promote consumption expansion and improve consumption quality, which has an important impact on the consumption upgrading of rural residents in China. Therefore, the following hypothesis is put forward:

• H1. The development of digital inclusive finance in China has played a positive role in promoting the consumption level and optimizing the consumption structure of rural residents.

The development of e-commerce has promoted the change of rural residents’ consumption habits. In the past, rural residents tended to shop in physical stores, and there were still some doubts about online shopping. However, with the popularity of rural e-commerce, more and more rural residents began to try and gradually like online shopping. Online shopping is not only convenient and fast, but also allows rural residents to enjoy more discounts, thus changing their consumption habits (Fu, 2023). In addition, Zhen (2021) found that the development of digital payment is the driving force and engine for the popularization of e-commerce in rural areas, and it also provides a driving force for the consumption upgrading of rural residents. Y. Wang et al. (2023) have proved that the development of rural e-commerce is conducive to increasing the income of rural residents and it is an important engine to promote rural revitalization. Therefore, this paper takes the e-commerce index as an intermediary variable to further analyze the influence path of the influence of digital inclusive finance on rural residents’ consumption, so the following hypothesis is put forward:

• H2. China digital inclusive finance can provide more choices of goods and services for rural residents by promoting the development of e-commerce, thus promoting the consumption upgrade of local residents.

According to the theory of relative income, when the income increases to a certain extent, the increase of consumption will gradually decrease, while when the income decreases, the decrease of consumption will gradually increase. The empirical results of An et al. (2023) also show that digital inclusive finance has different effects under different household consumption classes. Therefore, this paper holds that the influence of China digital inclusive finance on rural residents’ consumption may be different under different consumption levels and consumption structures. For rural residents with low consumption level and quality, digital inclusive finance can increase their disposable income by providing more convenient financial services, thus promoting consumption. Digital inclusive finance can lower the transaction cost and threshold of rural residents, and provide more diversified and convenient financial services, so as to better meet their consumption demand and improve their consumption level. Therefore, the following hypothesis is put forward:

• H3. China digital inclusive finance has a more obvious effect on consumption upgrading when the consumption rank of residents is low.

L. Li and Li (2022) believe that there is a non-linear relationship between digital inclusive finance and economic development. Similarly, this paper believes that there may be a non-linear relationship between digital inclusive finance and rural residents’ consumption level and consumption structure, that is, there is a critical point. In the interval above or below this critical point, the relationship between digital inclusive finance index and rural residents ' consumption will change. Therefore, the following hypothesis is put forward:

• H4. There is a threshold effect on the impact of China digital inclusive finance on the consumption upgrading of rural residents.

According to the theory of financial exclusion, capital demanders in remote areas are often outside the scope of financial services, unable to obtain the necessary financial resources, and have limitations in information exchange, time and space. In addition, Yan and Feng (2021) believe that rural residents in different regions may have differences in consumption habits and economic conditions. For example, rural residents in some regions may be more inclined to save, while in other regions may be more willing to consume. The development level of different regions is different, and the corresponding digital inclusive financial supporting system will also be different. These factors may affect the impact of digital inclusive finance on rural residents’ consumption.

The income level directly determines the consumption ability of rural residents. Pan and Yang (2012), B. Yang et al. (2012), and J. Zhang (2013) have all empirically proved the significant impact of rural residents’ income on rural consumption. When the income of rural residents increases, their consumption capacity will increase accordingly, so that they can buy more goods and services. In addition, there are also differences in the consumption structure of rural residents with different income levels. With the increase of income, the importance of basic consumption expenditure such as food in life is gradually decreasing, while the proportion of high-level consumption expenditure such as education, culture and entertainment in total consumption expenditure is increasing.

There are differences in the consumption structure of regions with different industrial structures. In areas dominated by agriculture, the consumption concept of rural residents may be more traditional and conservative, paying more attention to savings and family security. In areas with developed industries and service industries, rural residents ' consumption concepts may be more open and advanced, and pay more attention to consumption quality and consumption experience (Yu & Dong, 2022). From the above discussion, the following hypotheses are put forward:

• H5a. The impact of China’s digital inclusive financial development on the consumption upgrading of rural residents in different regions varies, with a more significant promotion effect on the eastern region.

• H5b. The development of digital inclusive finance in China has different effects on consumption upgrading of rural groups with different incomes, and the promotion effect on low-income groups is more significant.

• H5c. The development of digital inclusive finance in China will be influenced by the developed degree of local service industry, which will lead to different degrees of influence on the consumption upgrading of local residents, and the promotion effect on the groups in the developed areas of tertiary industry is more significant.

Figure 1 shows the theoretical mechanism of the digital inclusive finance on the consumption structure upgrading of rural residents.

Figure 1.

Framework diagram of the theoretical mechanism.

Research Design

Variable Definitions

Explained Variable

Promoting rural residents’ consumption is mainly reflected in two aspects: one is the improvement of consumption level, and the other is the improvement of consumption quality.

For the consumption level, this paper chooses the consumption level of rural residents to measure.

As for the quality of consumption, this paper chooses to construct “consumption structure upgrading coefficient of rural residents” to reflect the optimization degree of rural residents’ consumption content. The construction process is as follows: According to the composition of rural residents’ consumption expenditure in China given by the National Bureau of Statistics and referring to the method of W. Yang et al. (2021), this study chooses to analyze the elasticity of income demand by using the Extend Linear Expenditure System (ELES) model, and distinguish high-level consumption from low-level consumption according to the demand elasticity calculated by different composition of expenditure, and finally construct the consumption structure upgrading coefficient of rural residents accordingly. ELES model and the calculation formula of demand elasticity are as follows:

$$O_i=b_i+a_{i}Y+{\epsilon }_i$$ (1)

$${\rho }_i=a_i\frac{Y}{O_i}$$ (2)

Among them, $O_i$ represents the per capita expenditure of the i-type commodity and service of rural residents, $Y$ represents the per capita disposable income in rural areas, $a_i$ represents the parameter to be estimated, ${\rho }_i$ represents the demand elasticity for serving the i-type commodity, and ${\epsilon }_i$ is the random disturbance term. Based on the consumption data of eight categories of goods and services of rural residents in China (this category includes residents’ food, clothing, housing, daily necessities and services, health care, transportation and communication, culture, entertainment and education, other supplies and expenditure on services), this paper substitutes formula 1 and formula 2, and the calculated results are shown in Table 1 below. The demand elasticity of the eight categories of goods and services is positive, and the average value of the overall demand elasticity is 1.0873.

Table 1.

Elasticity of Consumption Demand.

Year	Food	Clothing	Housing	Life	Medical	Transportation	Culture	Other
2011	0.7847	0.7768	1.4586	1.0842	1.8370	1.5052	1.8482	0.6291
2012	0.8074	0.7590	1.4648	1.1121	1.7720	1.4311	1.8660	0.5901
2013	0.8449	0.7713	1.4494	1.1029	1.6655	1.3187	1.9224	0.5596
2014	0.8331	0.7365	1.1279	0.9375	1.5090	1.1527	1.2084	0.6677
2015	0.8376	0.7442	1.1242	0.9470	1.4643	1.0929	1.1672	0.6810
2016	0.8971	0.8170	1.1574	0.9957	1.5304	1.0727	1.2132	0.7311
2017	0.9321	0.8346	1.1472	1.0169	1.4590	1.0503	1.2044	0.7355
2018	0.9501	0.8572	1.1041	0.9745	0.9946	1.3910	1.1788	0.7376
2019	0.9497	0.8538	1.1216	1.0065	1.0029	1.3304	1.1351	0.7312
2020	0.9065	0.9130	1.1625	1.0707	1.3893	1.0980	1.3741	0.8413
2021	0.8628	0.8375	1.1479	1.0097	1.3779	1.0478	1.2084	0.7332
Mean	0.8733	0.8092	1.2241	1.0234	1.4547	1.2264	1.3933	0.6943

This paper defines the expenditure on goods and services whose demand elasticity is higher than the average of overall demand elasticity as high-level consumption, and the expenditure on goods and services whose demand elasticity is lower than the average of overall demand elasticity as low-level consumption. Therefore, high-level consumption includes housing, transportation, culture and medical, and low-level consumption includes other, clothing, food and life.

This paper refers to the method of constructing industrial structure upgrading coefficient in Xu and Jiang (2015), and constructs “the consumption structure upgrading coefficient of rural residents.” Based on the consumption levels divided above, the low-level consumption and high-level consumption are given different weights subjectively, and the consumption structure upgrading coefficient of rural residents representing is obtained after calculation. The calculation formula is as follows:

$$\mathrm{up}\_\mathrm{cons}=1*\mathrm{low}\_\mathrm{cons}\%+2*\mathrm{high}\_\mathrm{cons}\%$$ (3)

Among them, up_cons represents the consumption structure upgrading coefficient of rural residents, low_cons% represents the percentage of low-level consumption in total consumption expenditure, and high_cons% represents the percentage of high-level consumption in total consumption expenditure. Thus, the consumption structure upgrading coefficient of rural residents is obtained.

Core Explanatory Variable

In recent years, China’s digital finance has achieved rapid development, and has had a greater impact on a global scale. The research group of Digital Finance Research Center of Peking University started to build a set of indicators on the development of digital inclusive finance in China in 2016, and updated the indicators in the following years. The birth of Digital Inclusive Finance Index has had a great impact in the industry and academia, many academic papers and industry research reports on the direction of digital finance have used this index.

Considering that this indicator can truly reflect the changes of digital finance development in China, it is selected as the core explanatory variable in this paper.

Control Variable

There are many factors affecting the rural residents’ consumption upgrading. Based on the relevant literature and comprehensive consideration, this paper selects the following control variables: Structure: the proportion of non-wage income of rural residents in disposable income. Government: social security and employment expenditure of local governments. Urbanization: the proportion of urban population in each place to the total local population. Dependency : the total population of children and adolescents aged 0 to 14 and elderly people over 65 accounts for the proportion of the local population aged 15 to 64. Road mileage: the ratio of road mileage length to local population. Price index: the relative number of the changing trend and amplitude of the prices of daily necessities and service items purchased by rural households, which can reflect the actual changes in rural residents’ living standards. Investment: the investment in fixed assets of the whole society in agriculture, forestry, animal husbandry, and fishery industries, which can reflect the state’s support for agricultural and rural development. Traditional finance: the ratio of deposit and loan balance of local financial institutions to local GDP.

Instrumental Variable

Mobile phones have increasingly become artificial intelligence assistants and the core carrier of financial services. It leads the development of mobile financial services in a more intelligent direction through security, open capabilities and intelligent experience. Therefore, referring to the practice of W. Wang and Gu (2022), this paper selects Mobile (the number of mobile phones per 100 households of rural residents at the provincial level) as a tool variable.

Mediator Variable

At present, the digital economy is developing rapidly. The data show that the annual e-commerce transaction volume in 2022 exceeded 43.82 billion yuan, an increase of 3.5% compared with the previous year. This shows that the new retail business model represented by e-commerce and the new Internet format are gradually becoming an important driving force for rural residents’ consumption upgrading. The development of digital inclusive finance may have an important impact on rural residents’ consumption upgrading by affecting the development of e-commerce. Considering the availability of data, this paper chooses the E-Commerce Index published by Ali Research Institute as the intermediary variable to analyze the intermediary effect.

Regulated Variable

H. Zhang and Huang (2022) found that the upgrading of industrial structure can significantly promote the improvement of residents’ consumption level and consumption structure. The development of industry provides more channels for rural residents’ employment, increases rural residents’ income sources, and provides a material basis for rural residents’ consumption, thereby improving rural residents’ consumption level and driving rural residents’ consumption structure to be advanced.

Referring to the views of Sun and Xu (2018), this paper holds that the development of tertiary industry can promote the upgrading of industrial structure, make residents no longer confined to the consumption of low-level products, improve product quality from the supply side, and play a positive regulatory role in promoting the consumption upgrading of rural residents. Therefore, this paper represents the development of the tertiary industry with the proportion of the tertiary industry GDP to the total GDP, and takes the development of the tertiary industry as a regulating variable.

Model Specification

In the selection method of the model of panel data, F test is usually used to decide whether to choose the POOL (pooled regression) model or fixed effect (FE) model, Breusch-Pagan test is used to decide whether to choose random effect (RE) model or POOL model, and Hausman test is used to decide whether to choose RE model or FE model. In this paper, the above inspection methods are used, and the inspection results are shown in Table 2.

Table 2.

Model Select.

Explained variable	Level_cons	Up_cons
F test	57.598***	22.741***
Breusch-Pagan test	715.070***	531.049***
Hausman test	65.497***	27.47***

***

p < .01.

By analyzing Table 2, it can be seen that according to the F test, the significance p value is .000, which is significant at 1% level. The original hypothesis is rejected and the FE model is selected. According to the Breusch-Pagan test, the significance p value is .000, which is significant at the level of 1%. The original hypothesis is rejected and the RE model is selected. According to the Hausman test, the significance p value is .001, which is significant at the level of 1%. The original hypothesis is rejected and the FE model is finally selected.

To sum up, this paper constructs the basic regression model as follows:

$${\mathrm{level}\_\mathrm{cons}}_{\mathrm{it}}={\alpha }_0+\alpha *{\mathrm{digital}\_\mathrm{finance}}_{\mathrm{it}-1}+X*{\mathrm{Controls}}_{\mathrm{it}-1}+{\eta }_i+{\epsilon }_{\mathrm{it}-1}$$ (4)

$${\mathrm{up}\_\mathrm{cons}}_{\mathrm{it}}={\alpha }_0+\alpha *{\mathrm{digital}\_\mathrm{finance}}_{\mathrm{it}-1}+X*{\mathrm{Controls}}_{\mathrm{it}-1}+{\eta }_i+{\epsilon }_{\mathrm{it}-1}$$ (5)

Among them, level_cons is the consumption level of rural residents, up_cons is the consumption structure upgrading coefficient of rural residents, digital_finance is the core explanatory variable, and Controls is the control variable. Subscript “i” stands for different provinces in China, and subscript “t” stands for year.

In order to weaken the reverse causality problem, this paper adopts the first-order lag term for the core explanatory variables such as digital inclusive finance index and related control variables.

Empirical Results

Benchmark Regression Analysis

Table 3 below shows the empirical results based on the regression models of formula 4 and formula 5.

Table 3.

Benchmark Regression.

Explained variable	Level_cons		Up_cons
	Model 1	Model 2	Model 3	Model 4
Digital inclusive finance index	0.401*** (0.012)	0.112*** (0.018)	0.052*** (0.002)	0.017*** (0.007)
Structure		0.289*** (0.107)		−0.043 (0.038)
Government		0.430*** (0.032)		0.040*** (0.011)
Urbanization		1.669*** (0.302)		−0.017 (0.109)
Dependency		0.278 (0.208)		−0.186** (0.075)
Road mileage		0.013** (0.006)		0.006*** (0.002)
Price index		0.003 (0.005)		−0.008*** (0.002)
Investment		0.009 (0.011)		0.013*** (0.004)
Traditional finance		−0.035** (0.015)		0.004 (0.005)
Individual fixation	Y	Y	Y	Y
N	310	310	310	310
R 2	.810	.956	.667	.736

Note. () is the standard deviation.

**

p < .05. ***p < .01.

In Model 1 and Model 3, only the digital inclusive finance index, the core explanatory variable, is added. The results show that the digital inclusive finance index and the coefficients of the two explained variables are significantly positive at the level of 1%.

In order to avoid other variables affecting this conclusion, this paper adds the control variables in Model 2 and Model 4. By analyzing Table 3, after adding a series of control variables, the influence coefficient is still significantly positive at the level of 1%, which shows that the digital inclusive finance has provided a positive role in promoting the consumption level of rural residents and optimizing the consumption structure of rural residents, thus H1 is verified.

In terms of control variables, the empirical results of Model 2 show that the optimization of rural per capita disposable income structure can promote the improvement of rural residents’ consumption level, because the increase of rural residents’ property sources disperses the risk of a huge decline in income caused by the disappearance of a single property source, thus boosting rural residents’ consumption confidence. The government’s security level is also positively related to the consumption level of rural residents, which shows that if the government wants to improve the consumption level of rural residents and promote economic development, it needs to introduce more policies that are beneficial to the grassroots and provide practical living security for rural residents. The improvement of urbanization level has also significantly improved the consumption level of rural residents, which may be because with the advancement of urbanization process and the improvement of urbanization rate, the economic development of the whole region has been promoted, and the income security and related welfare of rural residents have also been improved accordingly, thus driving consumption. The per capita road mileage represents the convenience of local transportation. Since ancient times, the areas with rapid economic development are often areas with convenient transportation. The transportation extending in all directions can bring many resources to the local area, strengthen the close ties between regions, and then stimulate the consumption demand of rural residents. There is a significant negative correlation between the development level of traditional finance and the consumption level of rural residents, which may be due to the fact that per capita disposable income’s nonlinear role is not taken into account. The development level of traditional finance and the consumption level of rural residents may have a “U” relationship, that is, the development level of traditional finance may first inhibit and then promote the consumption level of rural residents.

The empirical results of Model 4 show that there is a significant positive correlation between the level of government security and the consumption structure upgrading coefficient of rural residents, because rural residents will be willing to spend their time, money, and energy on other high-quality goods and services only after their basic life is fully guaranteed. The results show that the total dependency ratio and the consumption structure upgrading coefficient of rural residents are significantly negative. Because of the need to support the elderly and children, residents’ consumption and shopping tend to be more about goods and services that guarantee the lives of the old and the young, thus inhibiting the transformation and upgrading of consumption structure. The per capita road mileage has significantly promoted the consumption upgrading of rural residents, because the improvement of transportation infrastructure can improve the circulation efficiency of goods and services, reduce logistics costs, and make it easier for consumers to obtain high-quality goods and services, such as imported goods and high-end brands, which will help upgrade the consumption structure. The consumption structure upgrading coefficient of rural consumer price index and rural residents is significantly negative, because the numerical increase of rural consumer price index directly reflects the increase of rural residents’ living burden, which directly affects rural residents’ consumption desire for high-level goods and services, and then inhibits the upgrading of rural residents’ consumption structure. Investment in agricultural fixed assets reflects the state’s support for agricultural and rural development, which has a significant positive impact on the consumption structure upgrading coefficient of rural residents, indicating that the state should increase investment in rural infrastructure and support for agricultural and rural development, so as to make farmers willing to consume goods at a higher level.

Robustness Test

In order to verify the reliability of the above regression results, this paper refers to the practice of W. Wang and Gu (2022), taking the number of mobile phones per 100 households in rural areas as a tool variable, and carries out an endogenous test based on formulas 4 and 5. The test results (Table 4) show that when the explained variable is the consumption level of rural residents, the significance p value is .632, which is not significant and cannot be. When the explained variable is the consumption structure upgrading coefficient of rural residents, the significance p value is .000, which is significant at the level of 1%. The original hypothesis is rejected, and the selected endogenous variables are endogenous, which can be fitted by two-stage least square regression.

Table 4.

Endogenous Test.

Explained variable	Statistic	p
Level_cons	0.230	.632
Up_cons	19.106	.000

Table 5 based on formula 5, the instrumental variables are regressed in two stages. Through analysis, it can be found that in the first stage of regression, the number of mobile phones per 100 households in rural areas has a significant and positive impact on the digital inclusive finance index, and the F value is 60.387, which is greater than 10, which meets the requirements for instrumental variables. In the second stage of regression results, the coefficient of digital inclusive finance index to rural residents’ consumption structure upgrading coefficient is 0.096, which is still significantly positive.

Table 5.

Two-Stage Regression.

Explained variable	Stage I	Stage II
	Digital inclusive finance index	Up_cons
Mobile	0.005*** (0.001)
Digital inclusive finance index		0.096*** (0.021)
Control variable	Y	Y
Individual fixation	Y	Y
N	310	310
F	60.387
R 2	.829	.433

Note. () is the standard deviation.

***

p < .01.

Based on the above analysis, it can be concluded that the results of basic regression are robust.

Intermediary Effect Test

Taking the development index of e-commerce as an intermediary variable, this paper explores the transmission mechanism of the development of digital inclusive finance to rural residents’ consumption by using the bootstrap method of sampling 1,000 times. The results are shown in Table 6.

Table 6.

Bootstrap Test.

Parameter		Level_cons	Up_cons
Coefficient		0.021***(0.007)	0.004*(0.002)
Z		2.9	1.84
95% Confidence interval	Upper limit	0.037	0.011
	Lower limit	0.009	0.001
Intermediary effect proportion		18.45%	23.53%

Note. () is the standard deviation.

*

p < .1. ***p < .01.

As can be seen from the table, when the explained variable is level_cons, the Z value with e-commerce index as the intermediary variable is 2.9, which is significant at the level of 1%, and the upper and lower limits of 95% confidence interval are 0.037 and 0.009 respectively, excluding 0. When the explained variable is up_cons, the Z value mediated by rural e-commerce index is 1.84, which is significant at the level of 10%, and the upper and lower limits of 95% confidence interval are 0.011 and 0.001 respectively, excluding 0. To sum up, the intermediary effect model with e-commerce index as the intermediary variable has been verified, and there are some intermediary effects. Further calculation shows that the intermediary effect of e-commerce development in promoting the consumption level of rural residents accounts for 18.75% of the total effect, and the intermediary effect in promoting the consumption level of rural residents accounts for 23.53%. Therefore, this paper thinks that the development of digital inclusive finance will promote the development of e-commerce in rural areas, and then promote the consumption upgrade of local residents, thus H2 is verified.

Quantile Model Regression Analysis

In order to more comprehensively reflect the distribution law of digital inclusive finance’s influence on rural residents’ consumption at different consumption levels, this paper adopts the method of panel quantile regression for further analysis, with quantiles of (0.1, 0.5, 0.8), and the regression results are shown in Table 7 below.

Table 7.

Quantile Regression.

Explained variable	Level_cons			Up_cons
Quantile	0.1	0.5	0.8	0.1	0.5	0.8
Digital inclusive finance index	0.312*** (0.000)	0.225*** (0.000)	0.225*** (0.000)	0.041*** (0.006)	0.024*** (0.001)	0.010 (0.114)
Control variable	Y	Y	Y	Y	Y	Y
Individual fixation	Y	Y	Y	Y	Y	Y
N	310	310	310	310	310	310
R 2	.668	.677	.693	.437	.336	.298

Note. () is the standard deviation.

***

p < .01.

When the explained variable is level_cons, the coefficients of the digital inclusive finance Index are all positively significant at the 1% level in the quantiles of 0.1, 0.5, and 0.8, indicating that with the improvement of rural residents’ consumption level, there is no obvious marginal increasing or decreasing effect on the digital inclusive finance. When the explained variable is up_cons, the estimation coefficient of digital inclusive finance is significantly positive on the quantile of 0.1 and 0.5, but the coefficient size is reduced, but it is not significant on the quantile of 0.8, and the coefficient size is further reduced. This shows that with the upgrading of rural residents’ consumption structure, the marginal promotion effect of digital inclusive finance on rural residents’ consumption structure upgrading coefficient is weakened. The reason for this situation may be that under the highly perfect consumption structure, the ability of digital inclusive finance to explore underutilized resources has reached saturation, and the space for creating wealth by using resources has gradually become crowded, thus weakening the promotion of consumption upgrading. To sum up, when the consumption rank of residents is low, the promotion effect of digital inclusive finance on consumption upgrading is more obvious. Thus H3 is verified.

Threshold Effect Test

There may be a nonlinear relationship between digital inclusive finance and rural residents’ consumption level and consumption structure, so this paper sets 300 bootstrap times by self-sampling method to determine the threshold number for threshold effect test. The final results are shown in Tables 8 and 9.

Table 8.

Threshold Effect Test (level_cons).

Explained variable	Level_cons
Threshold	Threshold effect test	Triple threshold
Single threshold	48.540***
Double threshold	23.360*
Triple threshold	27.730*
Digital inclusive finance index (X < 3.215)		0.463***(0.079)
Digital inclusive finance index (4.215 < X < 3.735)		0.404*** (0.066)
Digital inclusive finance index (4.735 < X < 4.607)		0.339*** (0.057)
Digital inclusive finance index (X > 4.607)		0.320*** (0.052)
Control variable		Y
Individual fixation		Y
N		310
R 2		.968

Note. () is the standard deviation.

*

p < .1. ***p < .01.

Table 9.

Threshold Effect Test (up_cons).

Explained variable	Up_cons
Threshold	Threshold effect test	Double threshold
Single threshold	44.380***
Double threshold	25.710**
Triple threshold	9.580
Digital inclusive finance index (X < 4.855)		0.002 (0.005)
Digital inclusive finance index (4.855 < X<5.376)		0.010** (0.005)
Digital inclusive finance index (X > 5.376)		0.013*** (0.005)
Control variable		Y
Individual fixation		Y
N		310
R 2		.788

Note. () is the standard deviation.

**

p < .05. ***p < .01.

As shown in Table 8, when the explained variable is level_cons, the triple threshold with the digital inclusive finance index as the threshold variable is significant at the level of 10%, which indicates that the digital inclusive finance has a triple threshold effect on the development of the real economy. Through the empirical test of the triple threshold model, when the digital inclusive finance index does not reach the first threshold value of 3.215, the influence coefficient is 0.463, which is significant at 1%. When the digital inclusive finance index exceeds the first threshold value of 3.215 but does not reach the second threshold value of 3.735, the influence coefficient of digital inclusive finance on the consumption level of rural residents is 0.404, which is also 1% significant. When the digital inclusive finance index exceeds the second threshold value of 3.735 but does not reach the third threshold value of 4.607, the influence coefficient of digital inclusive finance on the consumption level of rural residents is 0.339, which is 1% significant. When the digital inclusive finance index exceeds the third threshold value of 4.607, the influence coefficient of digital inclusive finance on the consumption level of rural residents is 0.320, which is also 1% significant. It is concluded that with the development of digital inclusive finance, the consumption level of rural residents has also increased, with the characteristics of “marginal decline,” which shows that the promotion of digital inclusive finance to the consumption level of rural residents in China will be weakened when it reaches a certain height.

As shown in Table 9, when the explained variable is up_cons, the triple threshold with the digital inclusive finance index as the threshold variable fails to pass the test, but it is significant at the level of 5% under the double threshold, indicating that the influence of digital inclusive finance on the development of the real economy has a double threshold effect. Through the empirical test of triple threshold model, when the digital inclusive finance index is lower than the first threshold value of 4.855, the influence coefficient of digital inclusive finance on rural residents’ consumption upgrading is 0.002, which is not significant, which may indicate that the relationship of trigger effect is not linear in this interval, that is, there may be other complex factors or nonlinear mechanisms at work. When the digital inclusive finance index is greater than the first threshold value of 4.855 and less than the second threshold value of 5.376, the influence coefficient of digital inclusive finance on the consumption structure upgrading coefficient of rural residents is significant at the level of 5%, with a coefficient of 0.01. When the digital inclusive finance index is greater than the second threshold value of 5.376, the influence coefficient of digital inclusive finance on the consumption structure upgrading coefficient of rural residents is 0.013, which is significant at the level of 1%. This shows that the consumption structure of rural residents in China will be improved with the development of digital inclusive finance, and it is characterized by “marginal increase.” Thus H4 is verified.

Heterogeneity Analysis

Regional Heterogeneity

In order to further explore the differences in the impact of the development of digital inclusive finance on the consumption upgrading of rural residents in different regions, this paper divides 31 provinces of China into three regions: East, Middle and West. The regression results are shown in Table 10.

Table 10.

Regional Heterogeneity.

Explained variable	Level_cons			Up_cons
	East	Middle	West	East	Middle	West
Digital inclusive finance index	0.204*** (0.040)	−0.049 (0.037)	0.123*** (0.020)	0.041** (0.017)	0.022 (0.014)	0.011 (0.008)
Control variable	Y	Y	Y	Y	Y	Y
Individual fixation	Y	Y	Y	Y	Y	Y
N	110	80	120	110	80	120
R 2	.969	.977	.974	.699	.825	.793

Note. () is the standard deviation.

**

p < .05. ***p < .01.

From the perspective of consumption level, digital inclusive finance has a positive impact on rural groups in eastern and western regions, and the promotion effect in eastern region is better than that in western region. The possible reason is that the eastern region has a higher level of economic development and the infrastructure of related financial services is more perfect, and the depth and breadth of digital inclusive finance are better than other regions, so it can better meet the consumption needs of local rural residents.

From the perspective of consumption structure, digital inclusive finance has a significant positive impact only on the eastern rural areas. The reason may be that the infrastructure of digital inclusive finance in central and western areas is still relatively weak, and the circulation of various financial services, products and information is not smooth enough, and there is still a significant gap with the eastern region. Especially in the western region, many places in this region are remote, which brings great difficulties to the related infrastructure construction of digital inclusive finance. As a result, local rural residents can’t make good use of the information and resources brought by digital inclusive finance and turn them into the driving force for high-level consumption of goods and services. On the other hand, because there are a large number of backward and remote areas in the central and western regions, the economy is relatively backward, and the income level of local residents is not high, which leads local rural residents to tend to more basic goods and services, thus inhibiting the optimal transformation of rural residents’ consumption structure.

To sum up, the influence of digital inclusive finance in different regions on rural residents’ consumption upgrading is different, and the promotion effect on groups in the eastern region is more obvious, thus H5a is verified.

Income Level Heterogeneity

In this paper, the average value of rural per capita disposable income is taken as the dividing line, and the total sample is divided into high-income group and low-income group, and the two groups are respectively subjected to benchmark regression. The results are shown in Table 11 below.

Table 11.

Income Level Heterogeneity.

Explained variable	Level_cons		Up_cons
	Low level	High level	Low level	High level
Digital inclusive finance index	0.087*** (0.018)	0.260*** (0.044)	0.027*** (0.007)	0.010(0.018)
Control variable	Y	Y	Y	Y
Individual fixation	Y	Y	Y	Y
N	179	131	179	131
R 2	.961	.948	.800	.600

Note. () is the standard deviation.

***

p < .01.

The regression results show that: on the one hand, the development of digital inclusive finance has a positive effect on rural groups with different income levels, indicating that the income difference has no obvious influence on the consumption level of local residents. On the other hand, the influence of digital inclusive finance on the consumption structure of low-income rural groups is significant and positive, while the influence on high-income rural groups is not significant, which shows that digital inclusive finance can better meet the financial service needs of low-income groups and promote the consumption upgrading of low-income rural residents. Thus H5b is verified.

Industrial Structure Heterogeneity

In order to verify the heterogeneity of industrial structure, this study uses the moderating effect model to explore the moderating effect of tertiary industry development as a moderating variable. Specifically, this paper represents the development of the tertiary industry with the proportion of the GDP of the tertiary industry to the total GDP, and introduces the development of the tertiary industry and the interaction between digital inclusive finance and the development of the tertiary industry to further explore whether the development of the tertiary industry can promote the positive impact of digital inclusive finance on the consumption upgrading of rural residents. In this paper, on the basis of formulas 4 and 5, the moderating effect model is constructed as follows:

$$\begin{array}{c}\multicolumn{1}{c}{{\mathrm{level}\_\mathrm{cons}}_{\mathrm{it}}={\theta }_0+\theta *{\mathrm{digital}\_\mathrm{finance}}_{\mathrm{it}-1}+\nu *{\mathrm{service}}_{\mathrm{it}-1}}\\ \multicolumn{1}{c}{+\mu *({\mathrm{digital}\_\mathrm{finance}}_{\mathrm{it}-1}}\\ \multicolumn{1}{c}{-\overline{\mathrm{digital}\_\mathrm{finance}})*({\mathrm{service}}_{\mathrm{it}-1}-\overline{\mathrm{service}})}\\ \multicolumn{1}{c}{+\sum X*{\mathrm{Controls}}_{\mathrm{it}-1}+{\eta }_i+{\epsilon }_{\mathrm{it}-1}}\end{array}$$ (6)

$$\begin{array}{c}\multicolumn{1}{c}{{\mathrm{up}\_\mathrm{cons}}_{\mathrm{it}}={\theta }_0+\theta *{\mathrm{digital}\_\mathrm{finance}}_{\mathrm{it}-1}+\nu *{\mathrm{service}}_{\mathrm{it}-1}}\\ \multicolumn{1}{c}{+\mu *({\mathrm{digital}\_\mathrm{finance}}_{\mathrm{it}-1}}\\ \multicolumn{1}{c}{-\overline{\mathrm{digital}\_\mathrm{finance}})*({\mathrm{service}}_{\mathrm{it}-1}-\overline{\mathrm{service}})}\\ \multicolumn{1}{c}{+\sum X*{\mathrm{Controls}}_{\mathrm{it}-1}+{\eta }_i+{\epsilon }_{\mathrm{it}-1}}\end{array}$$ (7)

Among them, “service” represents the development of the tertiary industry, and the interaction items are centralized in the model. The test results are shown in Table 12 below.

Table 12.

Regulative Effect Regression.

Parameter	Level_cons	Up_cons
θ	0.125*** (0.017)	0.019*** (0.007)
µ	0.500*** (0.08)	0.051* (0.031)
v	0.960*** (0.208)	0.082 (0.081)
Control variable	Y	Y
Individual fixation	Y	Y
N	310	310
R 2	.965	.739

Note. () is the standard deviation.

*

p < .1. ***p < .01.

By analyzing the data in the table, when the explained variable is level_cons, the values of θ and μ are 0.125 and 0.5 respectively, and both of them are positively significant at the level of 1%, which shows that the adjustment variable can strengthen the main effect, and the adjustment effect with the development of tertiary industry as the adjustment variable is significant, which shows that the development of tertiary industry plays a positive role in promoting the relationship between digital inclusive finance and the consumption level of rural residents.

When the explained variable is up_cons, the value of θ is 0.019, which is significant at the level of 1%, the value of μ is 0.051, which is significant at the level of 10%, and the sum of θ and μ are positive, which shows that the adjustment variable can strengthen the main effect, and the adjustment effect with the development of tertiary industry as the adjustment variable is significant, that is, the development of tertiary industry can promote the upgrading of rural residents’ consumption structure by digital inclusive finance.

To sum up, the development of digital inclusive finance will be influenced by the developed degree of local service industry, which will lead to different degrees of influence on the consumption upgrading of local residents, and the promotion effect on the groups in the developed areas of tertiary industry is more significant. Thus H5c is verified.

Discussion

Major Findings

In the context of implementing the rural revitalization strategy in China, digital inclusive finance plays a positive role in boosting rural residents’ consumption and optimizing their consumption structure, injecting new vitality into the economic growth and consumption quality improvement of rural areas. This study used macro data from 31 provincial regions in China from 2011 to 2021, and used a fixed-effect model to empirically study the impact of China’s digital inclusive finance development on the scale and structure of rural residents’ consumption, as well as the mechanism and effect differences. The research findings are as follows: First, consistent with previous studies, the results of this work confirm that the development of digital inclusive finance in China can drive the improvement of rural residents’ consumption level, optimize their consumption structure, and thus achieve the upgrading of rural residents’ consumption (He & Song, 2020; Yi & Zhou, 2018). This discovery emphasizes the important role of digital inclusive finance in rural residents’ consumption entities. This means that in the process of upgrading the consumption structure of rural residents, digital inclusive finance not only meets their basic living needs through credit allocation, but also provides more financial support for their recreational consumption (Zhao & Gai, 2020). Second, previous literature (Y. Wang et al., 2023; Zhen, 2021) has pointed out that the development of e-commerce can promote changes in the consumption level of rural residents. Online shopping is not only convenient and fast, but also allows rural residents to enjoy more discounts and promotions, thus changing their consumption habits (Fu, 2023). As predicted, current research results indicate that the development of e-commerce can serve as a medium between digital inclusive finance and rural residents’ consumption, transmitting the achievements of digital inclusive finance development to residents’ lives and effectively promoting the upgrading of rural residents’ consumption. Third, the impact of digital inclusive finance varies among different consumer classes in households. These findings are consistent with previous research results (An et al., 2023). Research shows that China’s digital inclusive finance has a more significant promoting effect on consumption upgrading when residents have lower consumption levels. According to the theory of relative income, when income increases to a certain extent, the increase in consumption will gradually decrease, while when income decreases, the decrease in consumption will gradually increase. Fourth, this study verifies the threshold effect of digital inclusive finance on the consumption structure of rural residents. L. Li and Li (2022) believe that there is a non-linear relationship between digital inclusive finance and economic development. Similarly, this article argues that there is a non-linear relationship between digital inclusive finance and the consumption level and structure of rural residents. The research results indicate that when digital inclusive finance develops to a certain extent in rural areas of China, its effect on improving the consumption level of local residents may weaken to a certain extent. However, its optimization effect on the consumption structure of local residents will be more significant. Finally, consistent with previous research findings (Pan & Yang, 2012; Yan & Feng, 2021; B. Yang et al., 2012; Yu & Dong, 2022; J. Zhang, 2013), this study verifies the heterogeneity of the impact of digital inclusive finance on rural residents’ consumption structure in terms of region, residents’ income level, and local real estate industry structure. The research results indicate that there are differences in the impact of China’s digital inclusive finance development on rural residents’ consumption upgrading in different regions, with a more significant promotion for the eastern region group; The development of digital inclusive finance in China has varying degrees of impact on consumption upgrading for rural groups with different incomes, and the promotion effect on low-income groups is more significant; The development of digital inclusive finance in China will be influenced by the level of development of the local service industry, resulting in varying degrees of differences in its impact on the upgrading of local residents’ consumption, and the promotion effect on groups in areas with developed tertiary industries is more significant.

Theoretical and Practical Implications

This study provides theoretical and practical implications for promoting rural residents’ consumption and optimizing their consumption structure through digital inclusive finance. First, efforts to improve the technical level of digital inclusive finance. The development of digital inclusive finance is inseparable from the support of advanced technology. On the one hand, relevant government departments should increase investment in the research and development of digital inclusive finance technology, encourage financial institutions to actively explore and innovate, and promote the development of financial technology. Encourage financial institutions to actively apply new technologies, innovate financial products and services, and improve the level of intelligence and personalization of financial services. On the other hand, relevant financial institutions and technology enterprises themselves should also increase investment in science and technology, strengthen technology research and development, promote independent innovation of key technologies, and promote digital transformation. Second, continue to promote the basic construction of digital inclusive finance. Consolidating and developing infrastructure is the key link to promote the development of digital inclusive finance. It is of great significance to continue to promote the basic construction of digital inclusive finance, especially the basic construction of digital inclusive finance in remote rural areas, to reduce regional disparities and promote fairness and justice. It is necessary to strengthen infrastructure construction such as information and communication, payment and settlement, and credit system, and improve the coverage and service quality of digital inclusive finance services. Especially in rural areas and poverty-stricken areas, it is necessary to increase investment in infrastructure, improve the accessibility and convenience of digital inclusive finance services, and let more people enjoy the convenience brought by digital inclusive finance. Third, vigorously cultivate digital talents in rural areas. Cultivating digital talents in rural areas is an important link to promote the development of digital inclusive finance. Digital inclusive finance is just a tool, and it is always people who use tools well. Therefore, it is very important to strengthen the cultivation and introduction of rural digital talents and improve the quality and ability of rural digital practitioners for implementing digital inclusive finance and really benefiting rural residents.

Conclusion

This study explores the effect of China’s digital inclusive finance development on the consumption scale and structure of rural residents through a fixed effects model. Due to the scarcity of literature exploring the impact of digital finance on the upgrading of consumption structure among rural residents, this study attempts to fill this gap through a systematic approach. The research results are as follows:

• The development of digital inclusive finance in China can promote the consumption level of rural residents and optimize the consumption structure of rural residents, thus realizing the consumption upgrading of rural residents.

• The development of e-commerce can be used as a medium between digital inclusive finance and rural residents’ consumption, and the achievements of the development of digital inclusive finance can be passed on to residents’ lives, so as to effectively promote the consumption upgrading of rural residents.

• When the consumption rank of China digital inclusive finance is low, the promotion effect on consumption upgrading is more obvious.

• China digital inclusive finance has a threshold effect on rural residents’ consumption upgrading. Specifically, when digital inclusive finance develops to a certain extent in rural areas of China, its effect on improving the consumption level of local residents may be weakened to some extent, however, its effect on optimizing the consumption structure of local residents will be more significant.

• The impact of China’s digital inclusive financial development on the consumption upgrading of rural residents in different regions varies, with a more significant promotion effect on the eastern region. The development of digital inclusive finance in China has different effects on consumption upgrading of rural groups with different incomes, and the promotion effect on low-income groups is more significant. The development of digital inclusive finance in China will be influenced by the developed degree of local service industry, which will lead to different degrees of influence on the consumption upgrading of local residents, and the promotion effect on the groups in the developed areas of tertiary industry is more significant.

Limitations

The core explanatory variable of this paper is Peking University Digital Inclusive Finance Index, but the data source of the index is mainly internet financial institutions such as Ali-pay, which mainly serve online users, and the demand for groups without internet access and offline financial services may not be fully covered; In addition, the digital inclusive finance index uses a number of indicators to reflect the development level of digital inclusive finance. With the rapid development and changes of digital inclusive finance, the weights of some indicators may need to be adjusted in time.