How Digital Financial Inclusion Based on Spatial Durbin Modeling Affects the Rural-Urban Gap
Publié en ligne: 19 mars 2025
Reçu: 15 oct. 2024
Accepté: 03 févr. 2025
DOI: https://doi.org/10.2478/amns-2025-0509
Mots clés
© 2025 Sanbao Zhang et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Digital financial inclusion refers to the use of advanced digital technology to provide convenient, low-cost and safe financial services to residents in urban and rural areas, aiming to solve the problem of “financial exclusion” that cannot be covered by the traditional financial system, and to promote financial inclusion and sustainable development [1–4]. With the popularization of the Internet and mobile payment technology, digital financial inclusion has made great progress in China. However, in the real application, digital inclusive finance still faces many problems and challenges [5–7]. Therefore, it is of great theoretical significance and practical value to study the relationship between digital inclusive finance and urban-rural income gap, and to explore the specific roles and countermeasures of digital inclusive finance in promoting the development of urban-rural integration, and the spatial Durbin model can play an important role in this process [8–11].
The study of digital financial inclusion has important practical significance and far-reaching theoretical significance on urban-rural income gap. On the one hand, as China’s urban-rural development imbalance and income gap gradually widen, digital inclusive finance can be an important way to make up for the urban-rural gap and promote the narrowing of the wealth gap nationwide [12–15]. Exploring the mechanism and spatial spillover effects of digital inclusive finance on the urban-rural income gap can help to deeply understand the relationship between China’s financial development and urban-rural economic development, and further improve China’s financial policies and urban-rural development strategies [16–19].
This paper examines the existing research results on the impact of digital financial inclusion on the urban-rural gap, on the basis of which a number of indicators such as the level of urbanization and the degree of openness to the outside world are chosen as control variables to analyze the impact of digital financial inclusion on the urban-rural gap. Municipal economic panel data of 160 prefecture-level cities in China from 2012 to 2019 are used, and an extremely rich data skeleton is employed to construct the spatial econometric model. The spatial autocorrelation test of digital financial inclusion and urban-rural income gap is conducted by global Moran index and local Moran index, and the empirical analysis of spatial effects is conducted by using the spatial Durbin model that has passed the LM test and robustness test.
Kuznets’ theory of the “inverted U” curve [20], also known as the Kuznets curve. This theory expresses a tendency that the level of inequality in income distribution may change at different stages of a country’s economic development. At the beginning of a country’s economic development, the distribution of income may be unequal and concentrated mainly in the hands of a few people. As the economy grows and develops, this inequality may increase. Based on the above views, the theory is divided into the following three main stages: In the early stages of economic growth. In the early stage of economic growth, the Kuznets curve shows an upward trend and as some people have more productive capital and resources, they are able to benefit more from economic growth, leading to an increase in income inequality. The middle stage of economic development. As the economy continues to grow, the Kuznets curve reaches its peak, the high point of the “inverted U” curve. At this stage, economic development may begin to bring benefits to a wider range of people, leading to an improvement in income distribution. Later stages of economic development. However, when economic development reaches a certain stage, the Kuznets curve falls, suggesting that the distribution of income may become unequal again. This may be due to the fact that some people are better able to cope with change, thus further entrenching their position in the economy.
The Lewis theory of dual economic structure, which focuses on labor transfer and capital accumulation between agriculture and industry, and the impact of digital inclusive finance on the urban-rural income gap can be analyzed through this theory: under the framework of the Lewis model, the urban industrial sector tends to attract a large number of laborers originating from the countryside, which leads to the problem of labor shortage in rural areas. With the rise of digital inclusion, rural laborers may migrate more to cities in search of job opportunities in the digital sector. This labor mobility could lead to rising wages in cities and an insufficient supply of human resources in rural areas, resulting in a relative lag in rural income levels. Digital inclusion typically has a higher demand for highly skilled labor. Urban residents are more likely to have access to training and education related to digital technologies and are therefore more likely to find employment in the digital sector. In contrast, residents in rural areas may face a digital skills deficit, making them less competitive for employment in the era of digital financial inclusion and thus affecting their income levels. Digital financial inclusion typically requires large capital investments, which may attract more capital flows to cities, further increasing the gap between urban and rural capital accumulation. The rapid growth of digital industries in cities may make cities more of a capital magnet, while rural areas lag behind relatively due to lack of investment and infrastructure development. This may limit the opportunities for rural residents to integrate into digital financial inclusion, leading to a further widening of the income gap.
Information asymmetry theory is an important theoretical framework in economics for analyzing market behavior and economic interactions under conditions of information asymmetry. The theory is concerned with the uneven distribution of information among market participants in the course of a transaction, i.e., one party has more or better information in a transaction, while the other party has relatively less or low quality information. The main contribution of the information asymmetry theory is to reveal the impact of information differences on market efficiency and resource allocation, and how it leads to a range of market failures. In information asymmetry theory, there are two main players: the information holder (usually the seller) and the party with relatively little information (usually the buyer). The information holder may use its information advantage for personal gain, while the party subject to insufficient information may be exposed to risk and loss. The main elements of information asymmetry theory include hidden information and hidden actions, moral hazard and adverse selection, signaling and shielding, and the application of game theory. In the era of digital financial inclusion, the problem of information asymmetry is more prominent, thanks to the rapid progress of digital technology, the acquisition and transmission of information has become simpler. In transactions, the flow of digitized information may be uneven, resulting in one party having easier access to key information, while the other party may fall into an information disadvantage. This poses challenges to both decision-making by economic participants and market efficiency.
The advancement of the Internet, big data, blockchain and other technologies is the driving force behind the development of digital inclusive finance, the ultimate goal of which is to apply advanced digital technologies to people’s daily production and life. The development of digital inclusive finance can, on the one hand, provide a good employment environment for rural residents through the employment effect, on the one hand, help rural residents alleviate the problem of production capital shortage through the financing effect, and on the other hand, enhance the enthusiasm of rural residents’ market participation through the market participation effect, which in turn promotes the improvement of the income level of rural residents and reduces the urban-rural income gap.
The development of digital inclusive finance can promote the employment of rural residents from the three aspects of providing employment information, creating jobs and improving the level of human capital, improve the income level of rural residents and narrow the income gap between urban and rural areas.
The application of digital inclusive finance in the financial sector has well eased the credit constraints of rural residents, providing them with more funds for investment and expanding the scale of agricultural production, thus increasing the income of rural residents and narrowing the income gap between urban and rural areas.
The development of digital inclusive finance can help rural residents better participate in market transactions. Under the traditional economic transaction model, buyers and sellers must be in a designated location in order to complete the transaction. Factors such as underdeveloped information in offline markets in rural areas and the distance between urban and rural areas can have a negative impact on the trading of agricultural products, making it difficult for rural residents to fully participate in market transactions, which in turn leads to a decline in the income level of rural residents. Digital inclusive finance can alleviate this problem, digital inclusive finance can enhance the degree of market information transparency, so that rural residents can obtain real market information in a timely manner, to understand the actual needs of the market, to enhance the degree of participation of rural residents in the market, to help rural residents to pre-determine the range of production and price range of agricultural products, as far as possible to avoid losses due to over-production, to increase the income of rural residents, and to reduce the income gap between urban and rural areas. Reduce the income gap between urban and rural areas.
The development of digital inclusive finance can lead to the emergence of an urban-rural digital divide, thereby widening the urban-rural income gap. The urban-rural digital divide refers to the large gap between urban and rural areas in terms of the degree of digitization, which mainly includes the following two aspects: first, the urban-rural digital access divide, i.e., the gap between urban and rural areas in terms of the construction of digital infrastructure such as the Internet, computers, cell phones and 5G base stations. The second is the urban-rural digital use divide, i.e. the gap between urban and rural residents in terms of proficiency and flexibility in the use of digital devices and digital technologies due to differences in digital literacy, digital skills and education levels, and the irrational allocation of resources. This makes the development process of digital inclusive finance is characterized by urban priority, the corresponding resources and policies tend to give greater support to the city, thus leading to the emergence of urban-rural digital divide. And this phenomenon will be exacerbated as the degree of regional economic development decreases, i.e., the poorer the region in the development of digital inclusive finance will produce a more obvious urban-rural digital divide.
Spatial correlation refers to the interconnectedness or interdependence of data values that exist in geographic space. This correlation implies that geospatially close locations may have similar or related characteristics, while distant locations may have different characteristics. There are three main types of spatial relationships. The first is a positive correlation, meaning that if one location has a higher value, then neighboring locations may also have higher values, or vice versa. The second type of relationship is a negative correlation, meaning that if one location has a higher value, then neighboring locations may have a lower value, or vice versa. The third type of relationship is a stochastic relationship, meaning that the relationship between the data on the geospatial is random, with no obvious positive or negative correlation. Global spatial autocorrelation This paper takes the test of global spatial autocorrelation, calculated the correlation between the geospatial data values of the Moran index test, he generates a Moran statistical value, which is compared with the null hypothesis (i.e., there is no global spatial autocorrelation) in order to determine whether there is spatial correlation. The Moran index is categorized into global Moran index [21] and local Moran index. The formula for the global Moran index is shown below:
The value of global Moran index I is [-1, 1], when Local spatial autocorrelation The global Moran index measures whether there is spatial correlation in the whole spatial data set, but it cannot measure whether there is spatial correlation between each spatial unit and its neighboring units. In this case, we need to use the local Moran index as shown below.
A spatial weight matrix, which is used to describe the interrelationships between different locations in geographic space. A spatial weight matrix is usually expressed as
There are three main ways to calculate the spatial weight matrix as follows. 0-1 weight matrix matrix In this kind of matrix, if there is a connection or association between two locations, it is represented by 1; if there is no connection between them, it is represented by 0. Its calculation formula is shown below.
Geographic distance weight matrix The geographic distance weight matrix is a matrix used to represent the geographic location relationship between a group of spatial units (e.g., cities, regions, countries, etc.). Its elements are the inverse of the geographic distance between the two spatial units, the geographic distance between the two spaces Economic Distance Matrix To assess the level of development of the regional economy by calculating GDP, we take the GDP values of spatial units
Spatial lag model (SLM). This model takes into account spatial dependence, i.e., observations in a given region are influenced not only by factors within the region but also by observations in neighboring regions. In this framework, the introduction of a spatial lag term captures the interactions and dependencies between spatial units and is usually reflected in the use of the spatial weight matrix
In the above equation,
Spatial error model (SEM). It is a statistical model used in the field of spatial econometrics to deal with the problem of spatial autocorrelation. Unlike the spatial lag model (SLM) which focuses on the spatial dependence of the explanatory variables, the spatial error model focuses on the spatial correlation of the error terms. The model assumes that the error terms may be spatially correlated, which, if not taken into account, may lead to biased and inaccurate model estimates.
In the above equation,
Spatial Durbin Model (SDM) [22] is a regression model used to analyze spatial data, which assumes that the dependent variable is not only affected by the independent variable, but also by the dependent and independent variables of other spatial units. The spatial Durbin model can simultaneously consider spatial correlation and spatial spillover effects between spatial units, thus improving the accuracy and interpretation of regression results.
The above equation adds a term
The explanatory variable is the urban-rural income gap (gap), which is measured by the Theil index (Theil), which is widely used as an indicator of regional income and disparity, and whose value is positively correlated with the gap in the measure, i.e., the larger the value, the larger the gap, and the measure is intuitive. The main explanatory index in this paper is digital inclusive finance (difi). For the selection of control variables, this paper refers to previous scholars to select the urban-rural income gap control variables, resulting in the following control variables in this paper: the level of urbanization (urban), the degree of opening up to the outside world (open), the level of economic development (rgdp), the level of financial expenditure (fe), the level of industrial structure (is) and the level of financial support for agriculture (afe), the degree of education (edu), the structure of employment (work).
Table 1 shows the global Moran’s index for both explanatory and interpreted variables of the studied variables. In terms of the significance level, in the eight years of panel data from 2012-2019, both are significant at the 1% significance level, and at the same time, the Moran index of both is positive in each year, indicating that there is a positive autocorrelation in the space of the studied prefecture-level cities, i.e., if the income gap between the urban and rural residents in one prefecture-level city is larger (smaller), the income gap between the urban and rural residents in the neighboring prefecture-level cities will also be larger (smaller), and the digital inclusion financial development level is the same. Note: ***, **, * denote significant levels of 1%, 5%, and 10%, respectively.
The classification of the gap between urban and rural in 160 cities
| Year | Urban and rural income gap | difi | ||
|---|---|---|---|---|
| Moran’sI | z | Moran’sI | z | |
| 2012 | 0.458*** | 8.351 | 0.539*** | 9.689 |
| 2013 | 0.447*** | 8.052 | 0.535*** | 9.678 |
| 2014 | 0.442*** | 8.033 | 0.556*** | 10.019 |
| 2015 | 0.436*** | 7.886 | 0.489*** | 8.795 |
| 2016 | 0.406*** | 7.306 | 0.536*** | 9.688 |
| 2017 | 0.365*** | 6.584 | 0.515*** | 9.325 |
| 2018 | 0.361*** | 6.579 | 0.529*** | 9.536 |
| 2019 | 0.398*** | 7.116 | 0.584*** | 10.548 |
In this paper, the analysis of local spatial autocorrelation follows the practice of most scholars by making a Moran scatterplot. The scatterplot consists of four quadrants: when the data are located in the first quadrant, it indicates that the explanatory and interpreted variables under study are characterized by the clustering of high values and high values. The second quadrant is characterized by the aggregation of low and high values at the prefecture level; the third quadrant is characterized by the aggregation of low and low values at the prefecture level. The fourth quadrant is the prefecture-level city where high and low values are clustered. Therefore, the variables in the first three quadrants have the aggregation characteristics of positive spatial correlation, and vice versa, the variables in the second four quadrants have the aggregation characteristics of negative spatial correlation.
In this paper, the Moran scatterplots of 160 prefecture-level cities in China in 2012 and 2019 are studied for the urban-rural residents’ income gap, and 30 representative prefecture-level cities are selected and numbered, and the Moran scatterplots of the 30 representative prefecture-level cities are displayed, and the scatterplots of the urban-rural residents’ income gap of some cities in 2012 and 2019 are shown in Fig. 1 and Fig. 2, respectively. Selecting the initial and termination years of the study year can better see the development of the variables. The intuitive results show that the overall distribution of prefecture-level cities has a reduced slope toward the in-situ aggregation objective, from 0.457 to 0.389. The slope of the fitted line is positive across one or three quadrants, and Moran’sI in the graph is greater than 0, which indicates that there are the characteristics of high high aggregation and low low aggregation of urban and rural residents’ income difference in the spatial distribution of the 160 selected prefecture-level cities, and at the same time, there are also some locations of the prefecture-level cities which are carried out with quadrant shifted, which may be caused by dry digital financial inclusion.
In 2012, the consumption gap of urban and rural residents in city
In 2019, the consumption gap of urban and rural residents in city
The Moran scatter plots of digital financial inclusion in 2012 and 2019 for the 160 prefecture-level cities under study in this paper are shown in Figures 3 and 4, respectively.In the Moran scatter plots of the two years, the slopes of the fitted lines are positive, crossing the one-third quadrant, indicating that the variables mainly have positive neighboring relationships, i.e., the high values are clustered with the high values and the low values are clustered with the low values. At the same time, the points representing prefecture-level cities in the two scatter plots have also changed in position, also showing an overall tightening of aggregation toward the origin, but the slope has become smaller, from 0.589 to 0.535. And comparing Figures 1, 2 and 3 and 4, it can be found that the points of prefecture-level cities, which are located in the third quadrant of the local scatter plots of the explanatory variables, instead appear in the third quadrant of the local scatter plots of the explanatory variables, which suggests that the that prefecture-level cities whose urban-rural residents’ income gap performance exhibits the aggregation of low values with low values, on the contrary, their digital financial inclusion performance is characterized by the aggregation of high values with high values, and vice versa. Therefore, through the scatter plot, it can be hypothesized that prefecture-level cities with a high level of digital financial inclusion development can increase their income willingness and acquire long-tail customers by reaching financial services to rural areas through relevant technologies. However, the correctness of this conclusion still needs to be proved by subsequent empirical research, after all, the scatterplot is only used to observe the correlation or not, and cannot judge the causality or not.
In 2012, some local digital market Numbers, puhui financial development
In 2019, some local digital market Numbers, puhui financial development
In this paper, when studying the spatial effect of digital financial inclusion development on urbanrural income gap, the first step is to conduct LM test and robust LM test. The test results are shown in Table 2.The LM test and robust LM test of the SAR and SEM models are highly significant, indicating that the spatial econometric model built on this basis is robust.
LM test results
| Test type | Statistic | P value |
|---|---|---|
| LM-lag | 15.411 | 0.000 |
| RobustLM-lag | 5.234 | 0.026 |
| LM-error | 25.783 | 0.000 |
| RobustLM-error | 15.615 | 0.000 |
From the above analysis, it can be seen that in the analysis of spatial effects, both SAR and SEM models can be applied, which requires us to conduct a deeper study of the SDM model, i.e., to study whether it can be degraded to the two models mentioned above, where we have to use the Wald and LR tests to determine. The test results are shown in Table 3.
Wald, LR test results
| Test type | Statistic | P value |
|---|---|---|
| Wald-spacial-lag | 6.85 | 0.0095 |
| Wald-spacial-error | 9.69 | 0.0023 |
| LR-spacial-lag | 64.55 | 0.0000 |
| LR-spacial-error | 91.26 | 0.0000 |
From the results of Wald test and LR test, it can be seen that both Wald and LR tests are highly significant, which indicates that the SDM model is not degradable, so the SDM model is chosen. Next, the Hausman test discerns that the spatial panel model should be constructed by choosing random effects.
Based on the above series of analyses, the spatial Durbin model (SDM model) with random effects is selected in this paper to study the impact of digital financial inclusion on the urban-rural income gap in the spatial dimension. Table 4 shows the regression results of the spatial Durbin model (SDM model). ***p<0.01, **p<0.05, *p<0.1.
SDM model regression results
| Variable | Regression coefficient | P value |
|---|---|---|
| ρ | 0.6444*** | 0.000 |
| difi | -0.0186*** | 0.000 |
| urban | -0.1009*** | 0.002 |
| is | 0.0892** | 0.014 |
| open | -0.0467*** | 0.000 |
| fe | 0.0365** | 0.013 |
| edu | 0.0046** | 0.032 |
| rgdp | 0.0018** | 0.011 |
| afe | -0.0023 | 0.956 |
| work | -0.0498** | 0.012 |
| W*difi | 0.0175** | 0.000 |
| W*urban | -0.1115** | 0.047 |
| W*is | -0.0369 | 0.165 |
| W*open | 0.0368*** | 0.005 |
| W*fe | 0.0295 | 0.326 |
| W*edu | -0.0018 | 0.679 |
| W*rgdp | 0.0013 | 0.467 |
| W*afe | 0.1118* | 0.084 |
| W*work | 0.0385 | 0.204 |
| R-sq | 0.7397 | |
| Log-L | 1053.7568 |
The spatial lag coefficient ρ of the urban-rural income gap (gap) is significantly positive at the 1% level, which proves that the SDM model can well overcome the shortcomings of the conventional panel model and incorporate the spatial correlation among variables, and reflects the positive spatial spillover effect of the urban-rural income gap in each region, i.e., an increase in the urban-rural income gap in the region will widen the income gap of the neighboring cities, towns and villages. Income gap between urban and rural residents in neighboring regions. The core explanatory variable of this paper, digital financial inclusion (difi), is significantly negative at the 1% significance level, and the coefficient of its spatial interaction term W*difi is positive and significant at the 1% level. This suggests that the development of digital financial inclusion can effectively improve the income gap between urban and rural residents in this provincial region, but at the same time it can further widen the income gap in neighboring provinces. The root cause of this may lie in the fact that the development of inclusive digital finance has led to fierce competition among borrowing capitals between neighboring provinces and cities, resulting in the inflow of capitals from economically weaker regions to developed regions, and the income gap between urban and rural residents is further widened. The coefficient of urbanization rate (urban) is significantly negatively correlated with both the level of urban and the spatial lag W*urban, which indicates that the increase of urbanization can effectively improve the gap between the income levels of urban and rural residents in China. The coefficients of the spatial lag of openness to the outside world (open) are significantly positively correlated, while the coefficients of the horizontal term are significantly negatively correlated, indicating that openness to the outside world plays a positive role in reducing the income gap between urban and rural residents in this province, whereas it has the opposite impact on the surrounding neighboring provinces and regions. The coefficient of the spatial lag term of financial support to agriculture (afe) shows a significant positive correlation, indicating that an increase in the proportion of financial support to agriculture will have a positive spatial spillover effect on the urban-rural income gap in neighboring provinces. In addition, the coefficient of the horizontal term of employment structure (work) is significantly negative at the 5% level, which suggests that the adjustment of employment structure can in some sense reduce the income gap between urban and rural residents, while the effects of financial expenditure (fe), industrial structure (is), and education level (edu) are opposite.
On this basis, this paper utilizes the partial differential method to study the model more deeply, which is divided into direct effect, indirect effect and total effect. The effect decomposition of the Spatial Durbin Model (SDM) is shown in Table 5. First of all, analyzing from the perspective of direct effect, digital financial inclusion (difi), urbanization rate (urban), degree of openness to the outside world (open), and employment structure (work) can produce a significantly negative direct effect, that is to say, the increase of the above economic variables has a significant impact on the reduction of the income gap between urban and rural residents. The industrial structure (is), the level of financial expenditure (fe) and the level of education (edu) can produce a significant positive direct effect. That is, an increase in the level of these economic variables has a significant widening effect on the local urban-rural income gap. The directions of the direct effects of the above economic variables are all consistent with the findings above.
The effect of the dubin model (SD,) decomposition
| Variable | Direct effect | P value | Indirect effect | P value | Total effect | P value |
|---|---|---|---|---|---|---|
| difi | -0.0175*** | 0.000 | 0.0135** | 0.025 | -0.0037** | 0.028 |
| urban | -0.1396*** | 0.000 | -0.4478*** | 0.006 | -0.5876*** | 0.000 |
| is | 0.0915** | 0.021 | -0.0176 | 0.887 | 0.0741 | 0.612 |
| open | -0.04656*** | 0.000 | 0.0178 | 0.598 | -0.0278 | 0.456 |
| fe | 0.0478*** | 0.004 | 0.1345* | 0.075 | 0.1815** | 0.027 |
| edu | 0.0049** | 0.035 | 0.0033 | 0.748 | 0.0078 | 0.479 |
| rgdp | 0.0023*** | 0.012 | 0.0056 | 0.145 | 0.0075* | 0.085 |
| afe | 0.0206 | 0.598 | 0.2836* | 0.086 | 0.3038* | 0.098 |
| work | -0.0476** | 0.028 | 0.0137 | 0.867 | -0.0339 | 0.716 |
Secondly, analyzing from the perspective of indirect effects, the indirect effects of digital inclusive finance (difi), fiscal expenditure level (fe), and financial support for agriculture (afe) are all significantly positive, that is to say, the spatial spillover effect is positive, while the indirect effect of urbanization rate (urban) is significantly negative, that is to say, the spatial spillover effect is negative. It can be seen that the direction of the indirect effects of the above economic variables are all consistent with the results of the study above. It indicates that with the increase in the level of digital inclusive financial development, the proportion of fiscal expenditure and the proportion of fiscal support for agriculture in the region, thus further widening the income gap between urban and rural residents in neighboring regions. And the promotion of urbanization can reduce the income gap between urban and rural residents in the surrounding areas to a certain extent.
Finally, analyzing from the perspective of the total effect, the total effect of digital financial inclusion (difi) is significantly negatively correlated, which indicates that, as the degree of China’s digital financial inclusion development continues to increase, the income gap between urban and rural residents will be significantly narrowed, and although the spatial spillovers of digital financial inclusion are smaller than the direct effect shown, the spatial contribution of its spatial spillover effect in the total effect is It can be seen that the spatial spillover effect of digital inclusive finance will have a great impact on the income gap between urban and rural residents, and the spatial spillover effect of the urbanization rate (urban), the level of financial expenditure (fe) and financial support for agriculture (afe) are the main influencing factors, which have a greater degree of influence than the direct effect and ultimately affect the direction of the overall effect, which shows that the spatial spillover effect has a great influence on the urban-rural income gap has a great impact, fully reflecting the significance and value of the research in this paper. Through the analysis of regional spatial effects, we can realize the synergistic development of localities and neighboring provinces and cities, and realize the synergistic development among regions, so as to achieve the common purpose of narrowing the income gap between urban and rural residents in China.
Based on the existing theoretical framework, this study used the spatial Durbin model to explore the mechanism of the development of digital inclusive finance on the urban-rural gap.
The test of spatial autocorrelation shows that there is a strong positive spatial correlation between digital financial inclusion and the income gap between urban and rural residents during the period of 2012-2019, so it can be preliminarily inferred that the urban areas with a higher level of digital financial inclusion development can reach the rural areas with financial services, which improves the income willingness of rural residents.
From the overall spatial Durbin model regression results, digital financial inclusion development effectively improves the income gap between urban and rural residents in provincial areas. It can be seen that digital financial inclusion development has promoted China’s urban-rural integration development to a certain extent.
Analyzed from the perspective of direct effect, digital financial inclusion development, urbanization rate, degree of opening up to the outside world, and employment structure will make the income gap between urban and rural residents in the study area show a significant reduction, and the income distribution will be more balanced.
Analyzing from the perspective of indirect effect, the increase of digital inclusive financial development, the proportion of fiscal expenditure, and the proportion of fiscal support to agriculture further widens the income gap between urban and rural residents, while the promotion of urbanization narrows the income gap between urban and rural residents to a certain extent.
Analyzed from the perspective of the total effect, as the degree of digital financial inclusion development continues to increase, the income gap between urban and rural residents will be significantly narrowed, significantly improving the income gap between urban and rural residents.