Exploring the Relationship between Digital Inclusive Finance and Traditional Finance in China

,


Introduction
Although digital inclusive finance (DIF) is a product of information technology, its realization relies heavily on government promotion and advocacy.Conventional formal credit access is biased toward large enterprises with substantial assets, as it relies on a credit model that favors businesses with significant assets.Consequently, this restricts the opportunities for small and medium-sized enterprises (SMEs) or businesses with asset-light models Provinces also issued "Implementation Opinions on Promoting Inclusive Financial Development" to comply with notification requirements.Since 2017, large and medium-sized commercial banks have established dedicated departments for inclusive finance.The period since 2016 has witnessed significant policy support for the advancement of DIF.According to the "Report on Financial Inclusion Innovation in China ( 2020)", digital finance has emerged as the predominant means of promoting financial inclusion in China (Zeng et al., 2020;Liu et al., 2021).
The implementation of various policies since 2016 not only facilitates the sustainable development of finance but also provides an opportunity for this study to investigate the relationship between DIF and traditional finance.While formal financial institutions are primarily driven by policies, achieving success in the digital transformation of traditional financial institutions requires a substantial investment of time.Financial technology (FinTech) companies have played a pioneering role in advancing digital financial inclusion (Wu, 2015;Xie et al., 2015;Bollaert et al., 2021 ).Consequently, these policies have created a favorable policy and market environment for FinTech companies.Accordingly, we can define "digital inclusive finance" as the financial development propelled by FinTech companies, and "traditional finance" as the financial development dominated by conventional financial institutions during the initial stages of policy implementation (Huang and Huang, 2018;Song et al., 2022).This paper examines the impact of multiple policies on DIF as exogenous factors by employing methodologies proposed by Greenland et al. (2019) and Li et al. (2022).Generalized difference-in-difference (generalized DID) is utilized to examine the impact of policies in regions with varying levels of traditional finance.This approach assumes that the nationwide implementation of a policy will uniformly impact all regions of the country, resulting in a treatment group without a control group.However, considering variations in the degree of policy effects across regions, the DID method can still be employed.Although there are no discernible regional characteristics, the impact of national policies may vary based on varying traditional fiscal levels across different regions.Therefore, this paper incorporates the pre-policy implementation level of traditional finance as a variable to measure regional impacts and assesses changes in DIF before and after policy introduction.This process also clarifies the relationship between DIF and traditional finance.
The main contributions of this paper are as follows: Firstly, the emergence of DIF has ignited extensive discourse in both practice and academia regarding its relationship with traditional finance.This study organizes existing perspectives and conducts further analysis to enhance theoretical research on their relationship.Secondly, this article utilizes the introduction of diverse policies that promote inclusive finance since 2016 as a quasi-natural experiment to study its impact on the regional development of DIF and effectively evaluate the effects of these policies.Thirdly, the discussion on the relationship between DIF and traditional finance mainly remains at a theoretical level, with limited empirical research.Drawing upon panel data from prefecture-level cities in China spanning from 2011 to 2019, this paper employs generalized DID estimation to empirically investigate the extent to which traditional finance influences DIF development, thereby providing empirical evidence for exploring their relationship.

Literature Review
Before evaluating the implementation impact of a series of policies on digital inclusive finance, it is imperative to clarify the theoretical relationship between digital finance and traditional finance.On one hand, the formulation of policies regarding DIF should consider the regional level of traditional finance.On the other hand, existing research has presented diverse perspectives on the "substitution relationship" and "complementary relationship" between DIF and traditional finance, but a consensus regarding their interconnection remains elusive.
In China, the view of "substitution relationship" primarily emerged during the era of Internet finance i .Since the inception of Yu'e Bao ii , there has been a proliferation of Internet financial products, resulting in a substantial outflow of commercial bank deposits to Internet-based financial enterprises (Wang and Zhang, 2015).Gopal and Schnabl (2022) also observed that following the financial crisis, there was a notable surge in loan volume for FinTech enterprises in the United States, while traditional bank credit to small businesses experienced a decline; however, these two shifts in credit nearly balanced each other out.This implies an inverse relationship between digital finance and traditional credit.According to Wolfe and Yoo (2017), peer-to-peer lending has resulted in a decrease in loan volume for small commercial banks, particularly in terms of personal loans.However, Balyuk et al. (2020) suggested that newly established FinTech firms often displace loans from large banks rather than those from small banks.Leveraging information technology and Internet platforms, Internet finance can effectively reduce transaction costs, enhance financial efficiency, and mitigate information asymmetry between transacting parties (Li et al., 2021).In addition to serving as a direct trading platform for capital supply and demand, Internet finance also effectively mobilizes idle funds to meet small-scale financing needs.While competing with banks for high-value customers, it also possesses an inherent advantage in serving the traditionally excluded "long-tail group" (i.e., the large low-income group) (Berg et al. 2019;Liu et al., 2021;Li et al., 2021), thereby presenting significant challenges to traditional finance.Xie and Zou (2012), who first introduced the concept of "Internet finance" in China, contend that modern information technology, epitomized by the Internet, has had a disruptive impact on traditional financial models.They predicted that Internet finance would become an outstanding financial paradigm in the future.Xie (2015), Tang (2019), andBollaert et al. (2021) argued that the Internet has the potential to expand the boundaries of trading, and in the future, Internet finance could facilitate financial transactions without relying on traditional intermediaries.This may result in displacing traditional banks and capital markets, leading to a situation akin to Walras's general equilibrium where financial intermediation becomes redundant.Therefore, following the substitution theory, Internet finance is considered a disruptive financial paradigm that presents significant challenges to traditional models and competes with them for market dominance (Xie, 2015;Wolfe &Yoo, 2017;Erel & Liebersohn, 2020).
A growing number of scholars support the notion of a "complementary relationship," suggesting that the development of DIF is dependent on traditional finance and that the pace of digital finance's advancement corresponds to the initial progress level of traditional finance (Cornelli et al., 2020;Wang et al., 2021).Digital finance represents an innovative paradigm, encompassing advancements and enhancements of financial products and services through the integration of digital technology.It is important to note that DIF operates within the framework of conventional finance without seeking to undermine or supplant it (Jagtiani & Lemieux, 2019).The primary role of FinTech banks is to enhance the supply of credit and address the inherent limitations of traditional finance, particularly in regions facing intense competition from conventional financial institutions (Jagtiani & Lemieux, 2018).Yao and Shi (2017) contended that digital finance represents a gradual evolution of traditional finance rather than a disruptive force, thereby exhibiting path dependence on the latter.From the perspective of financial knowledge spillover, traditional finance can serve as a reservoir of talent and expertise to facilitate financial innovation.The study conducted by Wang et al. (2021) underscores the pivotal role of traditional finance in fostering DIF, encompassing four fundamental dimensions: capital provision, dissemination of financial knowledge, enhancement of infrastructure, and stimulation of financial demand.The provision of financial resources by traditional finance is essential.In well-established regions, a diverse range of financing channels can be offered to facilitate the growth of FinTech enterprises and establish an enabling financial ecosystem (Haddad & Hornuf, 2019).Generally, regions with robust traditional financial systems have greater financial needs, which provide a broader market space for the growth of DIF (Cornelli et al., 2020).According to Xu et al. (2022), DIF has effectively complemented traditional finance in alleviating credit rationing and liquidity constraints for rural residents, based on their research on China's rural financial market.Moreover, it has contributed to narrowing the urban-rural income gap in conjunction with traditional finance (Song et al., 2022), demonstrating complementary functions in financial services.
The perspective that lies between the "substitution relationship" and the "complementary relationship" posits that the primary function of Internet finance development is to alleviate financial risks arising from information asymmetry and enhance financial efficiency, rather than undermining traditional finance (Wang & Zhang, 2015).The rapid expansion of China's Internet finance sector can be attributed to the regulatory arbitrage strategies employed by Chinese Internet finance companies, capitalizing on existing deficiencies in China's financial system such as interest rate controls, credit allocation mechanisms, and discriminatory lending practices toward NSOEs.Moreover, the informal financial sector in China lacks comprehensive regulatory measures (Buchak et al., 2017;Huang & Huang, 2018;Cornelli et al., 2020).Although financial technology enterprises can cater to the financial needs of rural residents and small and micro-enterprises, bridging the gaps left by formal finance, Internet finance may encounter challenges and struggle to establish itself as a mainstream form of finance in the future amidst the deepening of financial system reform and gradual implementation of regulatory measures.

Theoretical Analysis and Research Hypothesis
The substitution view characterizes digital finance as a disruptive financial innovation model that differs significantly from traditional finance.While traditional finance relies on financial institutions or markets for mediation, digital finance is expected to evolve into a financial model that operates without the need for intermediaries or markets, potentially replacing traditional finance in the future.However, this perspective fails to acknowledge the indispensability of financial regulation and the fundamental role played by financial intermediation.
In China, the primary objective of financial regulation is to ensure the maintenance of financial stability and prevent systemic financial risks.To achieve this goal, traditional finance is subjected to stringent regulations that restrict the allocation of financial resources and impact the enhancement of financial efficiency.FinTech enterprises, such as Ant Financial, are quasi-financial institutions that operate without financial licenses and are not directly supervised by regulatory authorities.The rapid emergence of these novel financial institutions can be attributed to the exploitation of regulatory loopholes (Huang & Huang, 2018;Huang & Tao, 2019;Cornelli et al., 2020).However, as their scale expands, ensuring financial security becomes an increasingly pressing concern.This will inevitably lead to the reinforcement of regulatory measures and industry rectification, ultimately undermining their advantage in exploiting regulatory loopholes (Wang & Zhang, 2015).Since 2016, the Chinese government has gradually implemented policies to regulate the development of Internet finance; however, owing to technological and other factors, the regulatory framework for it remains considerably less stringent compared to that of traditional finance (Zetzsche et al., 2019;Bollaert et al., 2021).
In terms of the fundamental functions of financial intermediation, China's traditional financial institutions are predominantly government-controlled, and government endorsement significantly enhances their credit intermediation function.However, in digital finance, FinTech enterprises serve as the primary players with Internet platforms as carriers.While digital technology reduces information asymmetry between lenders and borrowers, it does not necessarily increase trust between them; in fact, it may even lower the threshold for fraud (Bollaert et al., 2017).Therefore, FinTech enterprises can only act as "information intermediaries" rather than "credit intermediaries" (Huang & Tao, 2019).This also implies that digital finance cannot entirely replace traditional finance.
Although the conclusion that the disappearance of regulatory arbitrage advantages will limit the development of digital finance contradicts the substitution view, it fails to acknowledge the integration trend between traditional finance and DIF, thereby overlooking their interdependence.In contrast, the complementarity view emphasizes the relationship between inheritance and development within both digital finance and traditional finance, aligning more closely with current developmental realities and future trends in this domain.
We contend that DIF has been rooted in conventional finance since its inception.First of all, the development of digital technology has been facilitated by the sustained support of traditional finance for science and technology in China.Guided by the principle that "Science and technology constitute a primary productive force," China has consistently increased financial institutions' backing for scientific research, gradually establishing a long-term mechanism for the banking industry to support technological enterprises (Hao, 2017).The emergence and growth of DIF are direct results of the feedback from science and technology in finance.Secondly, digital finance embodies an innovative financial paradigm, necessitating specific prerequisites for its market acceptance (Sandhu & Arora, 2022).In terms of financial infrastructure, the rapid economic growth in China has resulted in a swift expansion of the banking network.The physical branches dispersed throughout the nation have not only enhanced residents' accessibility to financial services (Leyshon & Thrift, 1993), but also progressively enriched their financial knowledge, improved their financial literacy, and fostered positive financial habits.These factors are pivotal in bolstering public acceptance of financial innovation.For instance, as emphasized by Moorthy (2020), customers' perception of security significantly influences their inclination toward adopting mobile payment services.Consequently, this can diminish households' self-exclusion and stimulate their demand for financial products and services (Zhang & Yin, 2016;Guo & Wang, 2020;Song et al., 2022).In terms of information technology, digital finance has emerged based on the achievements made in the electronization and informatization of traditional financial institutions (Shahrokhi, 2008;Sandhu & Arora, 2022).Significant advancements have been achieved in the realm of e-banking services, including online banking and mobile banking, prior to the emergence of Internet finance businesses.Although banks still rely on offline channels for customer acquisition, the primary goal of developing e-banking was to alleviate pressure on bank branches rather than digitize operations.However, e-banking has enabled financial services to withdraw from physical channels and has laid a foundation for the widespread acceptance of digital financial business by the public.
Given the profound integration of finance and technology, the digital transformation of financial institutions has emerged as an overarching trend.To capitalize on this opportunity and gain a competitive edge, traditional financial intermediaries are actively embracing FinTech and leveraging data and digital technology to reshape their business models (Gomber et al., 2017;Drasch et al., 2018;Jagtiani & Lemieux, 2018).Yang (2020) highlighted that bank digital transformation involves two levels: "comprehensive transformation," transitioning from partial breakthroughs to overall development, and "in-depth transformation," shifting from general fields to specific domains.The former refers to the digital transformation of all aspects of business, including its form, management practices, and operational models.The latter focuses on enhancing value by shifting service scenarios from traditional financial settings to "finance + life" contexts and broadening the customer base to include both bank clients and online users.The transformation will take time; however, traditional financial institutions will remain crucial in driving digital finance advancement (Hao, 2017).As stated in the "Report on the Development of China's FinTech and Digital Inclusive Finance (2020)", China's future digital financial system will primarily rely on banking financial institutions, with support from Internet enterprises and complementation from non-bank financial institutions.In 2022, the China Banking and Insurance Regulatory Commission (CBRC) issued "Guiding Opinions on the Digital Transformation of Banking and Insurance Industries", marking it the first authoritative document specifically targeting digital reform in banking.This document effectively presents principles, an operational framework, and guidelines for digitization across financial institutions.Therefore, the substitute view suggests that FinTech companies will play a crucial role in the future development of DIF, disregarding the potential transformative impact on traditional financial institutions and asserting that digital finance and traditional finance are inherently distinct and incompatible.However, this perspective lacks foresight.
Irrespective of perspectives, it is widely acknowledged that the development of regional DIF is closely intertwined with the initial state of traditional finance.Consequently, the efficacy of policies pertaining to DIF will vary across regions based on different levels of traditional finance.Building upon the aforementioned analysis, this study suggests a complementary association between DIF and traditional finance.Given this proposition, the following research hypothesis is proposed.
H1: Where the initial level of traditional finance is higher, the development of digital inclusive finance will be faster, and the impact of digital inclusive finance policies will be more significant.

Sample Selection and Data Sources
The prefecture-level data used in this paper are mainly from the China City Statistical Yearbook, the provincial-level data are from the China Statistical Yearbook, and the digital inclusive finance index is obtained from the Institute of Digital Finance, Peking University.After removing the missing values of the samples, a total of 279 prefecture-level cities are included iii .Since the data on digital inclusive finance started in 2011, the sample period for this paper is from 2011 to 2019.

Construction of Benchmark Model
The issuance of Notice, Principles, and a series of policies since 2016 has become an external driving force for the rapid expansion of digital inclusive finance.Despite the challenges in distinguishing between the treatment group and control group, regional disparities in traditional finance can lead to variations in policy outcomes.Therefore, this study employs the methodologies proposed by Greenland et al. (2019) and Li et al. (2022) to investigate the impact of relevant policies on the development of DIF in regions with varying levels of traditional finance using the generalized DID approach.The model is set as follows: In model (1), the dependent variable digital_fin represents the level of DIF, while the independent variable post×loan_m is an interaction term that measures the policy effects of digital inclusive finance-related policies on prefecture-level cities with varying levels of traditional finance.X serves as a set of control variables.By controlling for fixed effects at the city and time levels, we can account for differences between cities and changes caused by other macro factors before and after policy implementation.This model can not only examine the policy effects but also capture the dependence of DIF on traditional finance.

Variable Selection
1) Explained variable.Digital inclusive finance (digital_fin) is a complex concept that cannot be fully captured through a singular perspective.The Peking University Digital Financial Inclusion Index of China (PKU-DFIIC), compiled by the Institute of Digital Finance, Peking University, utilizes transaction data from Ant Financial Services Group (Ant Financial) iv to establish an index based on three dimensions: breadth of coverage (coverage), depth of adoption (usage), and level of digitalization (digitization).Due to the challenges in obtaining data and matching data from various institutions, only one representative FinTech institution is used as the data source for this index.Nevertheless, it still reflects the development trends and regional differences among regions, making it the most widely utilized digital finance index in China (Guo et al., 2020;Luo et al., 2022).
2) Explanatory variable.The explanatory variable is the interaction term post×loan_m, which is derived as follows: Firstly, considering China's bank credit-dominated financial structure, we utilize "the ratio of urban financial institutions' loan balance to GDP at year-end" as a measure of each region's level of traditional finance.We calculate the average value of this measure (loan_m) before the implementation of policies (2011)(2012)(2013)(2014)(2015).A higher loan_m indicates a higher initial level of traditional finance.Secondly, 2016 marks a significant turning point and the variable post is utilized to indicate the temporal occurrence of this event.Specifically, post=0 denotes the period prior to 2016, while post=1 represents the period from 2016 onwards (including 2016).Thirdly, we utilize the interaction term post×loan_m to quantify the environmental changes encountered by DIF in regions with varying degrees of traditional finance before and after the implementation of policies.Notably, despite the significant growth of China's direct financing market in recent years, due to the dominant position of bank credit in its financial structure and practical limitations on the segmentation of regional stock and bond markets, when evaluating traditional finance at the city level, bank credit still serves as a proxy indicator (Li et al., 2013).
3) Control variables.X represents a set of control variables, and to mitigate the influence of policy changes on these variables, their baseline is established as the initial value in 2011.The control variables consist of six factors: (a) level of economic development (lngdp), which is measured by the logarithm of each city's real GDP deflated by its GDP deflator to eliminate inflation.This variable reflects the principle that financial services serve the real economy and indicates that regions with higher levels of economic development have a stronger demand for finance.(b) Industrial structure (structure), represented by the proportion of tertiary industry in GDP, also affects the financial demand of a region.(c) Regional openness (open) is calculated by dividing the total import and export volume of the province where the city is located by its GDP, due to the unavailability of city-level import and export data.This variable reflects how regional openness influences financial development.(d) The degree of government intervention (fiscal), expressed as budgetary fiscal expenditure divided by GDP, plays an important role in DIF development.Additionally, (e) the mobile payment penetration rate (mobile) is represented by the ratio of mobile phone users to the total population in a given region while (f) the level of Internet development (tele) is measured by the ratio of telecommunications service revenue to GDP.

Descriptive Statistics
The descriptive statistics of the main variables are presented in Table 1.The DIF variable, digital_fin, exhibits a mean value of 165.2668 with a range from 17.02 to 321.6457.Specifically, the mean values of the sub-dimensions coverage, usage, and digitization of the PKU-DFIIC are 155.5299,163.0097, and 201.5243 respectively, which align with the total index but exhibit significant variations among samples.It is evident that the mean values have significantly increased from their initial values (118.0597, 113.3403, 110.8546, and 146.7338) to their subsequent values (224.1483, 208.1531, 228.0628, and 269.8646) with the introduction of relevant policies.This indicates a rapid development in DIF across various dimensions.The variable loan has an average value of 1.3173 during the sample period, with a minimum value of 0.3711 and a maximum value of 9.6221.This suggests significant heterogeneity in the level of traditional finance across regions and implies that regional differences may impact the effectiveness of implementing DIF policies.

Empirical Results
Table 2 presents the regression results of Model (1).Column (1) depicts the impact of policy implementation on DIF development in regions with varying levels of traditional finance.The interaction term is scaled by a factor of 100 to standardize its magnitude and facilitate the interpretation of the regression results.Our findings suggest that, after controlling for other covariates, the coefficient associated with the interaction term post×loan_m exhibits a positive and statistically significant relationship with traditional finance.This implies that the impact of regional policies becomes increasingly significant as the level of development in traditional finance rises.Since 2016, DIF has experienced robust growth due to improvements in policy and market environments.In regions with higher levels of formal credit, the growth of DIF is more rapid, indicating that a city's initial level of formal credit serves as a solid foundation for financial innovation and development.Among the control variables, the level of urban economic development and the degree of regional openness significantly contribute to fostering DIF.
The regression results for the explanatory variable and control variables on each dimension of DIF are presented in columns ( 2), (3), and ( 4) respectively.The coefficient of the interaction term post×loan_m is significantly positive at a 1% confidence level, indicating that an improved development environment greatly enhances the reach of DIF.This effect is particularly pronounced in regions with robust traditional financial systems, where the growth rate of DIF becomes even more substantial.Similarly, Column (3) presents the results for the depth of DIF adoption.The coefficient of post×loan_m in this column is also significantly positive at a 1% confidence level, suggesting that users in regions with higher levels of traditional financial services have exhibited greater acceptance and adoption of DIF since 2016.However, in Column (4), the coefficient of the interaction term is not statistically significant.This suggests that there was no discernible disparity in the level of DIF development among regions after 2016.

Analysis of Results
The breadth of coverage includes the number of Alipay accounts per 10,000 people, the proportion of Alipay users who link their Alipay account to bank cards, and the average number of bank cards linked to each Alipay account.To open an Alipay account, it is necessary to possess a bank account because Chinese FinTech companies currently cannot accept deposits or conduct large-scale transfers.The DIF business aims to offer customers innovative digital financial products and services by leveraging the existing customer bases of banks, necessitating collaboration with traditional financial institutions and relying on the advancement of traditional finance.
The depth of DIF adoption encompasses the frequency of utilizing diverse services, including payment, money market funds, loans, insurance, investment, and credit investigation.Acquiring these services requires not only economic resources but also necessary financial literacy and habits.According to the 2014 World Bank survey, a lack of financial knowledge was identified by 78% of respondents as the primary barrier for households to access financial accounts.In regions with robust traditional financial institutions, users not only gain access to a diverse range of financial accounts and services but also have the opportunity to acquire extensive financial knowledge and gradually cultivate investment habits.Moreover, innovative products and services based on digital technology are more readily embraced by users in these regions.
The degree of digitalization in inclusive finance refers to the mobile, affordable, credit and facilitation of financial services with a primary focus on their reliance on digital technology.The statistical insignificance of this dimension suggests that digitalization in inclusive finance is not dependent on traditional finance.While traditional finance can provide financial support for the development of digital technology, the extent of digitalization in inclusive finance depends more on the integration between information technology and finance rather than being limited by traditional credit levels.In regions with limited growth in digital technology, the inherent influence of traditional finance can actually pose significant obstacles to the establishment of digitally inclusive financial institutions and hinder banks' digital transformation (Tang, 2019).
Based on empirical evidence, a positive correlation has been observed between the development of DIF and traditional finance.In regions with higher initial levels of traditional finance, policy-driven promotion of DIF leads to faster development.The empirical findings are consistent with our theoretical analysis, indicating that regions with higher initial levels of traditional finance will experience accelerated growth in DIF, resulting in a more pronounced effect of DIF policies.This confirms hypothesis H1 and supports the complementary perspective.

Robustness Test
In the empirical section, this paper employs "the ratio of financial institutions' credit balance to GDP" as a metric for assessing the level of traditional finance.To further evaluate China's traditional finance development from various perspectives and test the robustness of regression results by substituting explanatory variables, this paper also utilizes indicators such as "financial correlation rate" and "credit ratio of non-state-owned economic sectors.
The regression results presented in Table 3 demonstrate the utilization of post×fin_m as an explanatory variable, where fin_m represents the average financial correlation rates before 2016 (2011)(2012)(2013)(2014)(2015) in each region.Financial correlation rates are measured by "the ratio of total deposits and loans of financial institutions to GDP at year-end".The interaction term post×fin_m captures variations in policy and market environments between regions with distinct pre-existing financial correlation rates before and after the introduction of policies in 2016.The results for DIF and its dimensions in columns (1) to (4) reveal that the interaction term has a significant positive impact on the overall index as well as its two dimensions: coverage breadth and adoption depth.However, there is no significant relationship between the degree of digitization in DIF and the financial correlation rate.These findings align with previously reported empirical results.In the robustness test results presented in Table 4, this paper employs the ratio of credit to GDP of non-state-owned economic sectors as an indicator of the development level of traditional finance.Following research methodologies utilized by Aziz and Duenwald (2002) and Li et al. (2013), it assumes that the bank credit is allocated proportionally between the state-owned and non-state-owned sectors, with both sectors contributing equally to overall output.The allocation proportion can be estimated based on the correlation between bank credit and the share of assets held by SOEs.The explanatory variable is "the proportion of assets in state-owned economic sectors" (Soe).Due to data availability, we use a proxy variable that represents "the proportion of assets in state-owned and holding industrial enterprises within total industrial enterprise assets" for provinces where prefecture-level cities are located from 2010 to 2019.The explained variable is the ratio of credit balance to GDP in various prefecture-level cities while controlling for fixed effects specific to each city.
To address the issue of fluctuating levels of bank credit resulting from different financial deleveraging policies during the sample period, time-fixed effects are also incorporated into the model.The estimation equation is presented in Formula (2).To address the issue of serial correlation, the formula incorporates a first-order autoregression of the error term.
Among them, βSoe can measure the proportion of credit allocated to the state-owned economic sector in relation to GDP.The ratio of credit allocation to the non-state sector of the economy as a share of GDP denoted as ploan, is constructed using a constant term, regional and time dummy variables, and an error term.
Table 4 presents the regression results using post×ploan_m as an explanatory variable, where ploan_m represents the average credit value of non-state-owned economic sectors in each region before 2016 (2011-2015).
Based on the findings presented in Table 4, it is evident that significant differences exist in the development of the DIF index and its three dimensions among regions with varying proportions of credit for non-state-owned economic sectors prior to and after a policy change in 2016.In contrast to previous findings, regions with a higher proportion of credit in non-state-owned economic sectors have witnessed not only more rapid development in the breadth and depth of adoption but also accelerated progress in digitalization.This can be attributed to the dependence of digitization on information technology advancement.As mentioned in the theoretical section, given that China's FinTech enterprises are predominantly privately owned, an increase in the credit ratio of non-state-owned economic entities can provide significant financial support for private enterprises, especially those operating in FinTech.
Table 4.The interaction term for the share of credit to the non-state sector of the economy is used

Financial Environment
The benchmark regression results reveal significant regional disparities in the policy impact of DIF.Regions with higher levels of traditional finance exhibit a more rapid advancement of DIF, indicating a complementary relationship between the two and highlighting the reliance of DIF on traditional finance for its growth.To enhance our understanding of the policy effects across regions, this study incorporates regional environmental variables and investigates how the regional environment moderates the supportive role of traditional finance in DIF.Given that DIF encompasses both financial and technological dimensions, this study examines how regional financial and technological contexts influence the effectiveness of DIF by establishing Model (3) as an extension of the benchmark model.
The variable envir represents the environment of DIF in prefecture-level cities.The financial environment includes indicators such as the level of social credit (credit), degree of marketization (market), and level of banking competition (compete).Meanwhile, the technological environment includes factors like the level of technological innovation (innovate), Internet penetration rate (netuser), and mobile payment penetration rate (mobile).A detailed description of these variables is provided in Table 5.Firstly, credit serves as a fundamental prerequisite for the execution of all contracts, and the development of finance is inseparable from its driving force, particularly in inclusive finance (Guiso, 2012).Compared to traditional finance, digital finance faces greater financial risks and encounters more challenging regulatory obstacles.Therefore, the progress of regional inclusive digital finance is more sensitive to the credit environment.
Secondly, the level of regional marketization exerts an influence on the extent of support from traditional financial institutions toward DIF, thereby impacting the efficacy of policy.The advancement of financial systems is closely intertwined with economic development, and regions characterized by a high degree of marketization demand elevated standards for financial services, thus fostering an environment conducive to nurturing innovative financial intermediaries (Bos et al., 2013;Mao, 2021).Moreover, regions exhibiting a heightened level of marketization typically possess more transparent and open market mechanisms that facilitate the growth of digital finance (Wang et al., 2021).Thirdly, the impact of competition within the regional banking industry should also be taken into consideration.On one hand, a higher degree of banking competition signifies a robust financial foundation and a relatively balanced financial market structure in a region, thereby fostering an environment conducive to financial innovation (Jagtiani & Lemieux, 2018;Song et al., 2022).On the other hand, intensified competition among commercial banks results in the growth of traditional finance and consolidation of monopoly power, which hinders the survival and expansion of innovative financial models (Bos et al., 2013).
The coefficient value of the interaction term post×loan_m×credit in Column (1) of Table 6 exhibits a significantly positive relationship at the 1% significance level.This finding suggests that regions characterized by higher levels of social credit exhibit accelerated development in DIF, thereby demonstrating the enhanced efficacy of policies.The empirical results in Column (2) demonstrate that the coefficient value of the interaction term post×loan_m×market is positively and significantly associated at the 1% level, indicating that as regional marketization levels increase, traditional finance provides stronger support for DIF, thereby amplifying policy effects.According to the empirical findings presented in Column (3), there is a positive trend observed for the coefficient of post×loan_m×compete, suggesting that regions with weaker bank competition tend to receive stronger support from traditional finance in facilitating digital finance.However, this result lacks statistical significance due to counterbalancing dual effects stemming from bank competition.Successful implementation of digital finance policies is more likely to occur within a financial environment promoting regional trust and inclusiveness., 1983;Hannon & McDowell, 1984).The integration of digital technology into financial services not only substantially decreases information and transaction expenses but also empowers financial institutions to transcend geographical and temporal limitations via Internet platforms.Particularly, the proliferation of smartphones has facilitated financial service providers to reach an unprecedented number of customers, thereby expanding the scope of financial services extensively (Moorthy et al., 2020).In China, mobile payment has emerged as a prominent digital financial business (Huang & Tao, 2019).Building upon this premise, this study investigates the influence of regional levels of scientific and technological innovation, Internet penetration rate, and mobile payment level on the policy efficacy of DIF.
Based on the findings presented in Table 7, specifically in Column (1), the coefficient estimate of the interaction term post×loan_m×invent exhibits a significantly positive association at a significance level of 1%.This implies that regions with higher technological innovation will receive increased support from traditional financial institutions to promote DIF, thus facilitating the more effective implementation of DIF policies.The coefficient value of post×loan_m×netuser in Column (2) also shows a significantly positive association at the 1% level, indicating that regions with higher Internet penetration rates demonstrate greater efficacy in implementing DIF policies and receive stronger support from traditional financial institutions toward DIF.In addition, the interaction term post×loan_m×mobile is found to be significantly positive at the 10% level in Column (3), suggesting that regions with higher levels of mobile payment exhibit superior implementation outcomes for DIF policies.These findings suggest that a favorable regional science and technology environment can have a positive impact on the effectiveness of implementing digital inclusive financial policies.

Conclusions
The relationship between DIF and traditional finance is currently a prominent issue that has garnered significant attention in the academic field.Since 2016, the development of DIF has been included in national strategic plan, contributing significantly to enhancing financial sustainability through relevant policies.This paper employs research methods from Greenland et al. (2019) and Li et al. (2022) to examine changes in the policy environment around 2016, testing variations in the level of DIF across regions with varying levels of traditional finance.
Based on the empirical findings, this study draws the following conclusions: Firstly, the results of this study demonstrate that the initial level of traditional finance in a region plays a supportive role in fostering DIF, thereby substantiating the complementary perspective.Secondly, while traditional finance significantly influences the breadth of coverage and depth of adoption of DIF, its impact on the level of digitization is limited.This may be attributed to the fact that the digitization level relies more on local advancements in financial technology and integration between technology and finance, rather than traditional finance itself.Thirdly, the effectiveness of implementing regional policies on DIF is contingent upon the local financial and technological milieu.The traditional finance plays an important role in enhancing DIF and achieving better policy outcomes in regions with favorable financial and technological conditions.The research findings of this article are consistent with the arguments put forth by Jagtiani and Lemieux (2019), Cornelli et al. (2020), and Wang et al. (2021), thereby providing support for the view of a "complementary relationship."

Implications
The focus of China's financial reform lies in "inclusive finance", with DIF emerging as the predominant form in its current development.This study confirms that the growth of regional DIF is dependent on the level of traditional finance and is influenced by financial and technological environments.This brings us to the following policy implications.In areas where traditional finance is well-established, the government should strategically leverage existing advantages and actively guide traditional financial institutions toward digital transformation.Additionally, it should enhance local credit systems and market mechanisms while continuing to strengthen regional science and technology information infrastructure.In areas lacking adequate traditional financial services, local governments must initiate financial reforms, establish robust financial infrastructure, and create an enabling environment for FinTech advancements.Additionally, they should actively leverage the enthusiasm and drive of financial institutions to develop a comprehensive range of innovative financial services.

Limitations of the Study and Recommendations for Future Research
Due to the challenges in measuring the developmental level of DIF from a financial institution perspective, this paper utilizes a regional-level DIF index to assess its development level.Additionally, we incorporate policy effects to examine the relationship between DIF and traditional finance.While this approach provides valuable insights into their interplay at a regional level, it falls short in uncovering the direct impact and micro mechanisms through which traditional finance influences DIF.In future research endeavors, conducting comparative analyses of these two models of finance at an enterprise level would yield more compelling findings.
In addition, despite uncovering the complementary relationship between traditional finance and DIF, this paper lacks in-depth empirical research on how traditional finance supports the development of DIF.The impact mechanism of traditional finance on DIF remains unclear, necessitating future research to provide more empirical support and delve into mechanistic studies.

Notes
i Note 1.The rapid development and extensive application of digital technology in the financial sector have given rise to various concepts, including Internet finance, FinTech, and digital finance.FinTech emphasizes the technical aspects of financial innovation, while Internet finance focuses on the financial activities of Internet enterprises; conversely, digital finance places greater emphasis on leveraging digital technology for financial applications (Huang & Huang, 2018;Liu et al, 2021).Despite their subtle distinctions, these three definitions share robust inclusive financial characteristics.Recognizing the inherent connection between digital finance and inclusive finance functions, a form of inclusive finance known as "digital inclusive finance" has gradually emerged.
ii Note 2. Yu 'e Bao, an Internet money fund introduced by Ant Financial Service Group in 2013, has fundamentally transformed Chinese people's perception of Internet-based financial management.It has

Table 1 .
Descriptive statistics of main variables

Table 2 .
The impact of traditional finance on digital inclusive finance

Table 3 .
The interaction term for financial correlation rates is used

Table 5 .
Description of environmental variables

Table 6 .
Impact of financial environment on policy effects

Table 7 .
The impact of the technological environment on policy effects