Study of the Digital Inclusive Finance, Entrepreneurial Activism and Rural Revitalization Based on Provincial Panel Data in China

. According to the Central Document No. 1 of 2022, digital technology-driven rural financial inclusion will continue to take root across the country, and innovative applications of digitalization in financial services will continue to emerge, providing innovative solutions for rural revitalization. On the basis of the rural revitalization index measured by the entropy weight method, mediation effect model and spatial Durbin model are empirically tested to examine the inherent mechanism. Firstly, the findings indicate that digital inclusive finance significantly contributes to rural revitalization, even after modifying the method of robustness testing to employ the entropy TOPSIS method and selecting historical data as an instrumental variable. Second, stimulating mass entrepreneurship is an effective mechanism for digital inclusive finance to promote rural revitalization. Third, The positive impact of digital inclusive finance is accompanied by spatial spillovers, as demonstrated by spatial autoregression models and spatial Durbin models.


Introduction
Rural revitalization is not only one of the core objectives of the 14th Five-Year Plan, but also a new challenge for building a strong socialist modern state and rejuvenating Chinese society as a whole. As a capital relief channel, rural finance is of utmost importance in rural revitalization strategies, because it is capable of supporting rural revitalization through individual behavior , industry-driven , environmental improvement , and social norm (Ye X.,2018). With the decrease in the threshold of financial services, reduced physical outlet restrictions, and an expansion of business hours, digital inclusive finance is able to facilitate the solution of the "last mile" feature of financial services and offers new opportunities for rural revitalization.
Therefore, how does digital inclusion finance contribute to rural revitalization? And whether to consider spatial transfer effects? It is imperative to address these theoretical and practical issues as soon as possible. Existing studies on the indirect effect of digital inclusive finance on rural revitalization are not extensive and rarely involve spatial factors, which may lead to some bias in the conclusions drawn. The paper also has the potential to contribute marginally.
According to related studies, First, digital inclusive finance reduces rural poverty. By taking advantage of channels and information, digital inclusive finance can decrease the cost of financial services, increase efficiency, and widen coverage, which can increase consumption, raise income levels, and reduce rural poverty (Park C and Mercado R., 2015 ).Second, digital inclusive finance can narrow the income gap between urban and rural areas. The urban-rural income gap can be effectively closed by innovating and developing the basic function of inclusive digital economy (Zhou L et al., 2020).Thirdly, In terms of industrial development, digital inclusive finance provides a set of financial services aimed at integrating rural industries by reducing geographic segmentation in traditional finance, which is critical for rationalizing and advancing industries (Li Z G et al., 2021). Therefore, digital inclusive finance can stabilize agricultural production and promote industrial upgrading, resulting in a comprehensive rural renaissance. Finally, In terms of ecological civilization, as inclusive digital finance increases green total factor productivity ,it also reflects "green water and green mountains are golden mountains", (He Z et al., 2022). Through innovative financial products and services, digital inclusive finance can effectively reduce the transfer of resources to highly polluting industries, increase capital flows to green industries, and reduce the transaction costs of green financial products (Liu Jet al., 2022), thereby increasing the contribution of regional entrepreneurship to green total factor productivity.
Specifically, The marginal contributions are as follows. (1)Research perspective. little research is being conducted on rural revitalization from the perspective of entrepreneurial activity, so this article contributes to the advancement of research on rural revitalization development. (2)research methodology, the article measures the rural revitalization development index through the entropy method, and also systematically investigates the impact using intermediary effect model, spatial panel model and instrumental variable method, in order to provide deeper theoretical explanations and empirical evidence for China's rural revitalization.

Mechanisms of digital inclusive finance on rural revitalization
The process of entrepreneurship involves integrating different resources to create value, which relies inevitably on financial support. Rural enterprises face serious financial constraints due to a lack of financial institutions in rural areas, low credit ratings and unstable incomes of farmers(Paulson A L and Townsend R., 2004).Digital Inclusive Finance promotes farmer entrepreneurship in two ways. First, financial services such as remittances, payments, loans and investments can now be delivered online for farmers, reducing service constraints, transaction costs and bringing convenience, which overcome time and space constraints and reduce operational barriers (He J and Li Q H., 2019).Second, digital inclusive finance optimizes traditional financial information collection and processing by collecting information on taxation, industry and trade, subsidies, and other relevant factors to fully understand farmers' creditworthiness and repayment ability (Xie W et al., 2020).
At the same time, Rural revitalization has also been facilitated by increased entrepreneurial activity. On the one hand, the key to revitalizing the countryside is to develop industries. Supporting migrant workers start their own businesses contributing to the revitalization of rural areas, which promotes more talent, technologies, capital, and other factors to gather in rural areas and opens up new employment channels. On the other hand, entrepreneurship leads to the affluence of farmers' lives and increases their income by promoting employment and rural economic growth (Crépon B et al., 2015).As a result, entrepreneurs' income increases, which is indirectly transmitted to consumers' consumption and stimulates economic growth even further. It has been shown that successful entrepreneurship boosts entrepreneurial vitality in rural areas, creating a competitive advantage and promoting rural economic development (Sato Y., 2000).
In conclusion, entrepreneurial activities can create employment opportunities, increase the incomes of rural residents, stimulate consumption, and contribute to the revitalization of the countryside. Therefore, a research hypothesis is proposed.
Hypothesis 1: Digital financial inclusion can have a positive effect on rural revitalization development by increasing entrepreneurial activity.

Spatial impact of digital inclusive finance on rural revitalization
With advanced digital information technology, time and space barriers will be broken down, as well as the flow and transaction costs of resources, which accelerates the process of financial agglomeration and diffusion (Keller W., 2002).Firstly, digital finance can expand the supply of financial services, assist in the development of technological innovation activities of enterprises in neighboring regions, and therefore have a positive impact on the transformation of their industrial structures through big data technology (Huang Q et al., 2019).Secondly, the "demonstration effect" enhances digital finance and indirectly increase the export of financial factors from traditional financial institutions to the surrounding areas(Jiao S and Sun Q., 2021). Thirdly, due to the free flow BCP Business & Management

EMEHSS 2022
Volume 25 (2022) 572 of capital, technology, manpower, and natural resources between regions over time, competition or cooperation will emerge between them. As rural revitalization spreads across different regions, the "trickle-down effect" may result in the development of lagging regions, such as those in which the first rich bring the last rich to promote the development of the lagging regions. Therefore, a research hypothesis is proposed(Aghion P and Bolton P., 1997).
Hypothesis 2: Rural revitalization in neighboring areas can be influenced by spatial spillover effects of digital financial inclusion.

Intermediary model
In order to discuss a possible mediating variable of digital inclusive finance, a mediating effect model is constructed. The specific form of the above regression model was set as follows.
On the basis of the regression coefficient 1 α of the effect of digital inclusive finance , LNDIFI i t on rural revitalization , LNRURAL i t in the baseline linear regression model (1)

Spatial econometric models
The spatial impact of digital inclusive finance on rural revitalization is further investigated, and spatial interaction terms are introduced in equation (1), which is further expanded into a spatial Durbin model (SDM).
Where , i t W is an N × N dimensional spatial weighting matrix, The geographical distance matrix is represented by the inverse distance matrix is the distance of a and b, and a larger distance between regions means a smaller spatial weight. Adjacency weight matrix l W , 1 for region adjacency, 0 for non-adjacency.The economic distance matrix is

rural revitalization
Based on the National Strategic Plan for Rural Revitalization (2018-2022) and Zhang T et al.'s (2018) study, this paper selects indicators scientifically. Ultimately, this paper outlines an evaluation system of "two highs, three degrees, three winds, three governance, and three riches" (see Table 1). A weighted average of the indicators in Table 1 is calculated, along with the indices for the five primary indicators, using the information entropy method. Empirical measurements are taken of rural revitalization in 30 province-level regions (except Tibet) between 2011 and 2020.

Digital Inclusive Finance
A provincial Digital Inclusive Finance Index (LNDIFI) is calculated based on the Peking University Digital Inclusive Finance Index.

3.2.3Entrepreneurship in rural areas
According to Zhao T et al. (2020), the sum of private and self-employment as a percentage of rural population is used as a proxy variable for rural entrepreneurial dynamism (ENTEP).

Control variables
Fiscal decentralization (LNFISD) measures the proportion of local fiscal expenditures to central fiscal expenditures; the proportion of fiscal expenditure on agriculture, forestry and water to total expenditure is used to represent the level of fiscal support for agriculture (LNAFE); the level of information technology (LNTEC) is expressed as the logarithm of the number of website registrations of enterprises in each province. The old-age dependency ratio (LNODR) is calculated as the number of non-working-age elderly people in the total population of each province, region and city as a proportion of the working-age population; A secondary sector's contribution to GDP is measured by the industrial structure (CJ) .
An analysis of 10 years of data from 2011 to 2020 for 30 provinces in China is presented in this paper. For this study, data is obtained from the China Statistical Yearbook, the China Rural Statistical Yearbook, and selected prefecture-level city statistical annual reports as well as the Wind Information

EMEHSS 2022
Volume 25 (2022) 574 Database. For some missing data, linear interpolation is used to fill in the gaps. To remove the effect of outliers, this paper applies a two-sided 0.5% tail reduction to the control variables. Table 2 shows descriptive statistics.

Empirical testing 4.1 Analysis of benchmark regression results
Item (1) and (2)in Table 3 shows the regression result , the estimated coefficients of Digital Inclusive Finance (LNDIFI) are significantly positive which indicates digital inclusive finance promotes the development of rural revitalization. Furthermore, in model (2), fiscal decentralization (FISD), industrial structure (IS), agricultural support effort (AFE) all have significant positive effects on rural revitalization, demonstrating that autonomy in local decision-making, enhanced rural industrial structure, and increased government support for agriculture will significantly contribute to rural revitalization. The level of information technology (TEC) is significantly positive, proving that the Internet contributes to rural regeneration and the overall development of the region through the advancement of information technology. The significant negative effect of old age dependency level (ODR) on rural revitalization can be attributed to the fact that in an aging society there will be less workable people, leading to a lack of labor factors, which are detrimental to industrial development and improving living standards.
After using the stepwise regression method for analysis, the coefficient of digital inclusive finance for rural revitalization in model (4) decreases compared to model (2), indicating that increased entrepreneurial activity is the mechanism and this empirical result supports hypothesis 1. As a result, the value of the mediating effect β1-γ2 / α1 is 15. 93%.

Analysis of spatial spillover effects
The first step in selecting the spatial econometric model is to perform a Lagrange multiplier test (LM test). In table 4, the SDM is used when both error terms of the LM test are significant. Further validation of the SDM model is conducted using the likelihood ratio test (LR test) and the Wald test. According to Table 4, both the Wald and LR tests reject the original hypothesis at the 1% level, indicating that the SDM model does not degenerate into a SEM or SAR. A spatial Durbin's fixed effect model is employed in this study most appropriately.  Notes: Standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p<0.1 Table 5 reports the results of the spatial regression model of digital inclusive finance on rural revitalization under three different spatial weight matrices. This paper identifies the SDM model with dual space-time fixed effects as the optimal choice. Then spatial lagged model (SAR) with double fixed effects in space and time is also presented in the paper for the purpose of comparing the BCP Business & Management

EMEHSS 2022
Volume 25 (2022) 576 robustness of the estimates. Referring to the partial differential approach proposed by LeSage J et al. (2009), the coefficients of the effects of the respective variables on rural revitalization are decomposed into direct effects, indirect effects, and total effects. Taking the geographical matrix as an example, the coefficient of digital inclusive finance on rural revitalization is 0.101 in direct effect and passes the significance test. In indirect effects, the coefficient (0.015) is significant. The coefficient in the total effect is 0.115 and significant. The increase in digital inclusive finance in the local area has a positive effect on the growth in rural revitalization in regional and territorial provinces, validating hypothesis2.

Endogenetic treatment
On the one hand, despite controlling for variables affecting the development of rural revitalization to the greatest extent possible, there may be reverse causality. This paper mitigates the possible endogeneity problem by using the explanatory variable lagged by one period as the instrumental variable. On the other hand, static spatial panel model tends to ignore the influence of temporal factors beyond the explanatory variables and spatial influences. To address this issue, this paper develops a dynamic spatial panel model . In column (1) and (2) of Table 6, The coefficients of the impact of digital inclusive finance on rural revitalization are significantly positive, which is consistent with the previous benchmark regression.

Robustness test
Firstly, replace the explanatory variables. The TOPSIS method is applied to re-measure the rural revitalization index. Secondly, Due to the special status and policy bias, four municipalities, Beijing, Shanghai, Tianjin, and Chongqing, are excluded and re-estimated among the provincial administrative units. In Table 6,the results is still robust.

Conclusions and policy implications
On the basis of the rural revitalization index measured by the entropy weight method, mediation effect model and spatial Durbin model are empirically tested to examine the inherent mechanism. The findings are as follows: Firstly, digital inclusive finance directly improve rural revitalization. Various robustness tests, such as the endogeneity test, the exclusion of municipalities under the central government, and the replacement of explanatory variables, have confirmed that the above statements are valid; secondly, entrepreneurial activity can indirectly promote rural revitalization, indicating that the two-wheel drive phenomenon formed by "mass entrepreneurship and innovation" is of great significance in realizing China's rural economic development. Thirdly, the spatial spillover effects have also been confirmed, indicating the spatial spillover effects between regions.
This study proposes the following policy suggestions. Firstly, it is necessary to foster farmers' entrepreneurship to improve the employment opportunities. On the one hand, it is imperative to accelerate the integration of digital technology and financial products, develop diversified and humane financial products and services to meet the needs of people in a variety of ways. On the other hand，governments can introduce preferential policies in order to stimulate farmers' business startups and create more employment opportunities. Secondly, pay attention to the spatial spillover effects of digital inclusive finance and rural revitalization from a regional perspective. It is essential to coordinate the development of neighboring provinces, avoid negative spatial externalities caused by excessive competition, allocate financial resources rationally.