Digital Inclusive Finance and Firm Entry: Dual Mechanism of Alleviating Informational and Financing Constraints

Abstract

This study investigates how digital inclusive finance facilitates firm entry by alleviating informational and financial constraints. We first develop a theoretical model of entrepreneurial decision-making that incorporates both informational and financing constraints. Using panel data for 285 Chinese prefecture-level cities from 2011 to 2019, we estimate city- and year-fixed-effects models and implement two-stage least-squares, treating the interaction between each city’s distance to Hangzhou and the national number of mobile phone users as an instrumental variable for digital inclusive finance to mitigate endogeneity. The empirical results demonstrate that digital inclusive finance significantly promotes firm entry through a dual mechanism: reducing both informational barriers and financing constraints. The effects are particularly pronounced for micro, small, and private enterprises, as well as in the tertiary sector and productive service industries. Our findings suggest policymakers should strengthen digital finance in underdeveloped regions and provide targeted support to resource-constrained entrepreneurs, fostering economic vitality and sustainable development.

JEL Classification: G21, L26, O33.

Plain language summary

Digital inclusive finance and firm entry

This paper constructs a theoretical model to explore the mechanism by which digital inclusive finance promotes firm entry by alleviating informational and financing constraints, and empirically tests it based on industrial and commercial registration data from China’s prefecture-level cities from 2011 to 2019. The study finds that, first, the development of digital inclusive finance significantly promotes firm entry, consistently using the instrumental variable method in regression analysis and conducting multiple robustness tests to ensure the reliability of the conclusions. Second, digital inclusive finance fosters firm entry by mitigating information and financing constraints. Finally, heterogeneity analysis indicates that digital inclusive finance significantly supports firm entry among micro, small, and private enterprises, while also noticeably promoting the development of the tertiary industry and productive service sectors.

Introduction

Entrepreneurial activities, particularly new firm entry, are essential sources of economic growth, employment, and sustainable development (Curteis, 1997; Decker et al., 2017; Filser et al., 2019; Glaeser et al., 2015; Haltiwanger et al., 2013; Meglio & Di Paola, 2021; Stoica et al., 2020). In recent years, governments worldwide have increasingly viewed firm entry not only as a way to boost economic vitality but also as a core strategy for achieving high-quality, sustainable development. For instance, the European Union emphasized the vital role of entrepreneurial spirit in its “Europe 2020 Strategy,” the United States launched the “Startup America” initiative, and Singapore actively nurtures an innovation ecosystem through its “Smart Nation” strategy. As the world’s second-largest economy, China has also enacted policies, such as “Mass Entrepreneurship and Mass Innovation,” to facilitate economic restructuring and high-quality growth.

According to China’s industrial and commercial registration data in Figure 1, the number of newly registered enterprises has generally risen since the beginning of the 21st century, with the share of new entrants among existing enterprises fluctuating between 10% and 25%. By the end of 2020, the number of existing enterprises surpassed 50 million, reflecting the sustained vitality of China’s business environment. However, more recent declines in both the number and rate of new enterprise entries raise questions about how to further lower entry barriers and strengthen economic vitality—an issue that remains crucial for policymakers and researchers.

Figure 1.

The number of startup enterprises and existing enterprises in China over the years.

The rapid emergence of entrepreneurial entries in China coincides with the widespread adoption of digital technologies, suggesting a possible connection between technological advancements and entrepreneurial activities. One key driver in this regard is the development of digital inclusive finance, which plays a significant role in promoting entrepreneurship by easing financing constraints (Chai et al., 2023; Xie et al., 2018). According to the “China Entrepreneurial Enterprise Survey (2018),” 46.08% of entrepreneurs identified “funding constraints” as a key barrier, highlighting the critical need for accessible financial support to foster firm entry.

In addition to financing issues, entrepreneurs must also transform innovative ideas into commercial ventures, making informational constraints a major challenge (Fiet, 1996; Landström et al., 2012; Townsend et al., 2018). Recognizing this, the State Council’s “Opinions on Building a More Robust Market-Oriented Allocation Mechanism for Production Factors” (2020) highlighted data alongside traditional factors such as land, labor, capital, and technology, emphasizing the urgency of “accelerating the cultivation of the data factor market.” This policy focus acknowledges that data and information have become essential production factors for businesses. The Global Entrepreneurship Monitor 2017/2018 Global Report also emphasized that digital information technology strongly supports entrepreneurial activities globally by increasing the depth and breadth of data availability, lowering entry barriers, and fostering entrepreneurial ecosystems.

Within this context, the growth of digital inclusive finance has enhanced credit service accessibility while providing entrepreneurs with extensive, high-quality information resources. According to the “Peking University Digital Inclusive Finance Index (2011–2020)” published by the Peking University Digital Finance Research Center (Guo et al., 2020), the average provincial digital inclusive finance index rose from 40 in 2011 to 341 in 2020, reflecting an approximately eightfold increase. Such strong growth highlights the growing importance of digital inclusive finance in China’s financial system.

Although existing studies have mostly examined how digital inclusive finance promotes entrepreneurship by mitigating financing constraints (Gao et al., 2025; Li et al., 2022; Liu et al., 2022; Sun & Xie, 2024; Xie et al., 2018; Yang et al., 2022), its potential to overcome informational barriers remains underexplored. Since firm entry requires not only funding but also diverse information—from identifying market opportunities to managing operational risks—the dual role of digital inclusive finance in addressing both constraints is still not fully understood.

To fill this gap, we develop a theoretical model of entrepreneurial decision-making that incorporates both informational and financing constraints to illustrate how digital inclusive finance promotes firm entry. We then empirically test the model using industrial and commercial registration data from 285 prefecture-level cities in China between 2011 and 2019, applying instrumental variable regressions to handle potential endogeneity. Our empirical results show that digital inclusive finance significantly promotes new firm entry, and multiple robustness checks confirm this relationship. We further find that digital inclusive finance reduces both informational and financing constraints, highlighting its dual-channel mechanism. Given China’s diverse policy environment and industry structures, we also examine heterogeneity: digital inclusive finance exerts a stronger effect on micro, small, and private enterprises, as well as on tertiary and productive service sectors.

The marginal contributions of this paper are as follows. First, we develop a theoretical entrepreneurial decision-making model that combines the digital attributes and financing features of inclusive finance into a unified framework, addressing a gap left by previous empirically oriented studies (Li et al., 2022; Liu et al., 2022; Sun & Xie, 2024; Xie et al., 2018; Yang et al., 2022). Second, we identify and empirically validate a dual mechanism that mitigates both informational and financing constraints, providing a more comprehensive explanation of how digital tools facilitate entrepreneurs’ access to vital resources (Liu & Liu, 2024; Nambisan, 2017). Finally, we conduct an in-depth heterogeneity analysis, revealing that digital inclusive finance benefits micro, small, and private enterprises, as well as the tertiary sector and productive service industries.

The remainder of the paper is organized as follows. Section “Literature Review” describes the literature review. Section “Theoretical Framework and Research Hypothesis” presents the theoretical framework and research hypothesis. Section “Data and Methodology” describes the data and methodology. Section “Empirical Results” shows empirical results. Section “Mechanism Analysis and Heterogeneity Analysis” discusses the mechanism analysis and heterogeneity analysis. Section “Conclusions, Policy Implications, Limitations, and Future Directions” concludes the paper.

Literature Review

Impact of Digital Inclusive Finance

Digital inclusive finance combines “inclusive finance” and digital technology to create an innovative financial model that expands financial service coverage through digital innovation (Guo et al., 2020). Its rapid growth reflects not only the limitations of traditional financial systems, which often struggle to meet diverse needs, but also aligns with the policy goal of promoting high-quality development (Jin et al., 2023). Meanwhile, scholars emphasize its wide-ranging effects on economic growth and social welfare, including mitigating financing imbalances, fostering regional entrepreneurship, and enhancing economic vitality (Aghion et al., 2007; Welter & Smallbone, 2014). Moreover, digital inclusive finance can optimize resource allocation and reduce transaction costs, thereby supporting green finance and environmental governance through carbon mitigation and pollution reduction (He et al., 2023).

Determinants of Firm Entry

Firm entry serves as a key indicator of economic dynamism and depends on multiple factors, such as financing constraints, informational barriers, and the policy environment (Feldman, 2001). Under traditional financial systems, limited service reach and pronounced information asymmetry often result in restricted access to finance and elevated borrowing costs, which inhibit entrepreneurial activities across regions (Aghion et al., 2007; Welter & Smallbone, 2014). In recent years, digital finance has provided novel approaches to mitigating these constraints, although some scholars raise concerns about resource misallocation and investment inefficiencies (Xue et al., 2024). Nevertheless, research suggests that effective risk governance and robust policy and technological support can help digital financial tools alleviate financing and informational constraints without introducing negative shocks to resource allocation (Yi et al., 2024), thus promoting healthier enterprise development (Wang, 2015).

Digital Inclusive Finance and Firm Entry

Digital inclusive finance targets both financing and informational constraints through its dual focus on “inclusiveness” and “digitalization.” First, its inclusiveness reduces financing thresholds and increases credit availability. Compared with traditional financial systems, digital inclusive finance leverages big data, artificial intelligence, and cloud computing to strengthen risk management (Duarte et al., 2012; Huang & Huang, 2018) and mitigate credit mismatches caused by information asymmetry (Wang, 2015), especially in underdeveloped regions and among small firms (Xie et al., 2018). Second, its digital technologies help firms lower marketing, investment, and operating costs (Fairlie, 2006), while online social platforms facilitate information sharing and entrepreneurial learning (Hauptman, 2003; Nambisan & Sawhney, 2011; Wan et al., 2024), thereby creating more entrepreneurial opportunities (Fischer & Reuber, 2011; Huang et al., 2021). Studies also show that digital financial services, such as mobile payments, can reduce startup costs and alleviate financing constraints, ultimately encouraging firm entry (Yin et al., 2019). However, the specific role of digital inclusive finance in alleviating informational barriers for firm entry remains unclear (Le Dinh et al., 2018; Peng et al., 2022), highlighting the need for further empirical investigation with more comprehensive data.

Theoretical Framework and Research Hypothesis

Theoretical Framework

To investigate how digital inclusive finance influences firm entry, we develop an entrepreneurial decision-making model. Building on the research of Yin et al. (2019), we innovatively apply Lucas’s (1978) approach, which separately considers management skills and enterprise production technology. In line with institutional economics and complex systems theory, we consider that digital inclusive finance and traditional finance may both compete and complement each other, shaped by local policies and institutional settings. By introducing firms’ informational inputs into managerial technology as a production factor, the model captures both informational and financing constraints. This approach offers a comprehensive view of how these constraints affect entrepreneurial decisions and highlights the role of digital inclusive finance in mitigating them.

Assume an individual has an initial endowment z, informational input $i\left(m\right)$ , and entrepreneurial ability θ. At the beginning of the period, the individual decides whether to be employed by someone else or start a firm. Working for someone else yields wage income w plus interest on the initial endowment at rate r, totaling $(1+r)z$ . If the individual becomes an entrepreneur, they earn operating income:

$$\begin{array}{c}y={\left[i\left(m\right)\right]}^{\alpha }{\theta }^{\mathrm{\delta }}{k}^{\beta }v\#\hfill \end{array}$$

(1)

Here, k is the initial capital investment for firm entry; α, δ, and β are parameters in $(0,1)$ satisfying $\alpha +\delta +\beta =1$ ; and v is a random disturbance term. The informational input $i\left(m\right)$ is an increasing function of the level of development of digital inclusive finance m, implying that $\partial i\left(m\right)/\partial m>0$ . This reflects the role of digital inclusive finance in alleviating informational constraints faced by firms. The entrepreneur’s net income at the end of the period is:

$$\begin{array}{c}{Y}^e=y+(1+r)(z-k)\#\hfill \end{array}$$

(2)

Prior evidence suggests that small and microenterprises often lack sufficient collateral or credit history to secure substantial loans. By definition, $(k-z)$ represents borrowed funds. We assume that the financing amount depends on factors such as the entrepreneur’s creditworthiness and the level of digital inclusive finance development, rather than being a fixed proportion of the initial endowment. Entrepreneurs generally access two sources of external financing: (1) traditional financial institutions, mostly commercial banks, denoted $h(z,m)$ ; and (2) digital inclusive finance loans, denoted $f(z,m)$ . These must satisfy:

$$\begin{array}{c}h(z,m)+f(z,m)\le \overline{k}(z,m)\#\hfill \end{array}$$

(3)

where $\overline{k}(z,m)$ reflects maximum financing capacity based on the individual’s initial endowment, creditworthiness, and the local development of digital inclusive finance.

Due to loan costs and accessibility issues, entrepreneurs consider interest rates, collateral requirements, and convenience when deciding between traditional and digital finance options. Digital inclusive finance often offers more favorable terms for those underserved by traditional financial institutions. Hence, the total available capital is:

$$\begin{array}{c}0\le k\le z+h(z,m)+f(z,m)\#\hfill \end{array}$$

(4)

The entrepreneur’s net-income maximization problem is:

$$\begin{array}{c}\underset{k\in [0,z+h(z,m)+f(z,m)]}{\mathrm{Max}}\{{\left[i\left(m\right)\right]}^{\alpha }{\theta }^{\mathrm{\delta }}{k}^{\beta }+(1+r)(z-k)\}\#\hfill \end{array}$$

(5)

which yields the optimal capital $k^{*}$ from the first-order condition:

$$\begin{array}{c}d{Y}^e/\mathrm{dk}={\left[i\left(m\right)\right]}^{\alpha }{\theta }^{\mathrm{\delta }}{k}^{\beta -1}-(1+r)=0\#\hfill \end{array}$$

(6)

Solving for $k^{*}$ :

$$\begin{array}{c}{k}^{*}={\left(\frac{{\left[i\left(m\right)\right]}^{\alpha }{\theta }^{\mathrm{\delta }}\beta }{1+r}\right)}^{\frac{1}{1-\beta }}\#\hfill \end{array}$$

(7)

Distinguishing Between Financing Constraint Scenarios

No Financing Constraints

If the entrepreneur can raise enough capital to meet $k^{*}$ :

$$\begin{array}{c}{k}^{*}\le z+h(z,m)+f(z,m)\#\hfill \end{array}$$

(8)

Financing constraints do not bind, and the entrepreneur achieves the optimal production level. The condition for θ becomes:

$$\begin{array}{c}\theta \ge {\theta }_{\min }={\left(\frac{1}{{\left[i\left(m\right)\right]}^{\alpha }}{w}^{1-\beta }{\left(\frac{1+r}{\beta }\right)}^{\beta }{(1-\beta )}^{\beta -1}\right)}^{\frac{1}{\mathrm{\delta }}}\#\hfill \end{array}$$

(9)

which defines the minimum entrepreneurial ability ${\theta }_{\min }$ needed to surpass wage employment. As $i\left(m\right)$ increases with digital inclusive finance, ${\theta }_{\min }$ decreases, lowering the entry barriers.

With Financing Constraints

If the entrepreneur cannot obtain enough capital to reach $k^{*}$ :

$$\begin{array}{c}{k}^{*}>z+h(z,m)+f(z,m)\#\hfill \end{array}$$

(10)

they are constrained to:

$$\begin{array}{c}k=z+h(z,m)+f(z,m)\#\hfill \end{array}$$

(11)

Operating income then becomes:

$$\begin{array}{c}\mathrm{y}={\left[i\left(m\right)\right]}^{\alpha }{\theta }^{\mathrm{\delta }}{(z+h(z,m)+f(z,m))}^{\beta }v\#\hfill \end{array}$$

(12)

Entrepreneurial ability θ must satisfy two conditions:

A lower bound ensuring entrepreneurial income exceeds wage income:

$$\begin{array}{c}\theta \ge {\theta }_{\min }^{\text{'}}={\left(\frac{w}{{\left[i\left(m\right)\right]}^{\alpha }{[z+h(z,m)+f(z,m)]}^{\beta }}\right)}^{\frac{1}{\mathrm{\delta }}}\#\hfill \end{array}$$

(13)

An upper bound due to financing constraints:

$$\begin{array}{c}\theta \le {\theta }_{\max }={\left(\frac{(1+r)}{{\left[i\left(m\right)\right]}^{\alpha }\beta }{[z+h(z,m)+f(z,m)]}^{1-\beta }\right)}^{\frac{1}{\mathrm{\delta }}}\#\hfill \end{array}$$

(14)

Hence, individuals with $\theta \in [{\theta }_{\min }^{\text{'}},{\theta }_{\max }]$ can pursue entrepreneurship under these financing constraints. Figure 2 illustrates this theoretical framework.

Figure 2.

Theoretical framework: digital inclusive finance and firm entry.

Implications of Digital Inclusive Finance

Alleviating Informational Constraints

An increase in $i\left(m\right)$ boosts productivity and reduces ${\theta }_{\min }$ and ${\theta }_{\min }^{\text{'}}$ , thereby lowering the entry threshold for entrepreneurs.

Alleviating Financing Constraints

As digital inclusive finance develops, $f(z,m)$ rises, which increases total funds $z+h(z,m)+f(z,m)$ . This broader financing capacity expands feasible investments k, raises ${\theta }_{\max }$ by enabling higher-ability entrepreneurs to invest closer to $k^{*}$ , and reduces ${\theta }_{\min }^{\text{'}}$ .

Overall, this framework emphasizes the significant role of digital inclusive finance in reducing entry barriers by alleviating both informational and financing constraints.

Research Hypotheses

The model above provides a comprehensive framework for analyzing the relationship between digital inclusive finance and firm entry. It demonstrates that digital inclusive finance eases informational and financing constraints, thereby stimulating firm entry. Based on this, we propose the following hypotheses:

• Hypothesis 1: The development of digital inclusive finance can promote firm entry.

• Hypothesis 2: Digital inclusive finance promotes firm entry by alleviating informational constraints.

• Hypothesis 3: Digital inclusive finance promotes firm entry by alleviating financing constraints.

Data and Methodology

Data Sources

We use the enterprise registration database from the State Administration for Market Regulation, focusing on the 2011 to 2019 period to avoid distortions linked to the COVID-19 pandemic. The database covers approximately 50 million enterprises across 31 provinces in mainland China, encompassing 19 national economic industry categories and excluding international organizations. We cross-checked the data and removed observations with missing key variables, aggregating them to the city level. Enterprises are categorized by industry according to the “2017 National Economic Industry Classification” and classified as state-owned, private, or foreign-invested based on their shareholder information. Because this database is comprehensive and frequently updated, it offers an ideal foundation for analyzing firm entry.

To measure the development of digital inclusive finance, we rely on the China Digital Inclusive Finance Index compiled by the Peking University Digital Finance Research Center. This index covers sub-indicators of digital payments, online lending, and mobile financial services, weighted by their relative importance in regional financial development. We also measure banking-sector competition using license data manually collected from the China Banking and Insurance Regulatory Commission (CBIRC). For non-bank financial institutions, we reference microloan data from the People’s Bank of China. Finally, we obtain control variables from the China City Statistical Yearbook, various prefecture-level city statistical yearbooks, and the Wind database.

After excluding samples with missing data, the final panel dataset comprises 2,565 observations from 285 cities in China for the period 2011 to 2019. Table 1 presents the descriptive statistics of the main variables used in this study.

Table 1.

Descriptive Statistics of Main Variables.

Name	Symbol	Obs.	Mean	Median	Std. Dev.
Firm entry rate	Entry	2,565	18.931	18.324	4.647
Digital inclusive finance index	DIF	2,565	165.6	171.9	65.42
Economic development level	Ln(Pgdp)	2,563	10.69	10.66	0.576
Urbanization level	Urban	2,552	0.364	0.304	0.235
Industrial structure	Strc	2,564	0.984	0.854	0.546
Traditional financial development level	Finance	2,564	0.997	0.813	0.643
Foreign investment level	FDI	2,565	0.017	0.012	0.018
Fiscal expenditure ratio	Fiscexp	2,564	0.204	0.175	0.127
Population size	Ln(Pop)	2,564	5.883	5.936	0.703
Real interest rate	Rate	2,565	0.051	0.049	0.008

Variable Definitions

Dependent Variable

Firm Entry Level (Entry): The ratio of newly registered enterprises to existing enterprises in each region. We also use the per capita number of newly registered enterprises as an alternative measure for robustness checks.

Independent Variable

Digital Inclusive Finance (DIF): The digital inclusive finance index for prefecture-level cities from 2011 to 2019.

Control Variables

The control variables used in this study include the following: economic development level (Ln(Pgdp)), the logarithm of per capita GDP for each city; urbanization level (Urban), the ratio of the population in city districts at the end of the year to the total city population at year-end; industrial structure (Strc), the ratio of value-added from the tertiary industry to that from the secondary industry; traditional financial development level (Finance), the ratio of the balance of loans from financial institutions at year-end to regional GDP; foreign direct investment (FDI), the ratio of actual utilized foreign investment to regional GDP for the year; fiscal expenditure level (Fiscexp), the ratio of general budget fiscal expenditure to regional GDP; population size (Ln(Pop)), the logarithm of the total city population at year-end; and real interest rate (Rate), the Loan Prime Rate (LPR) data for six-month to one-year loans published by the National Interbank Funding Center and adjusted to an annualized rate using the number of days as a weight, with the real interest rate obtained by dividing the annualized rate by the provincial consumer price index (CPI).

Mechanism Variables

Bank Competition Level (HHI): Following Degryse and Ongena (2007), Chong et al. (2013), and Jiang et al. (2019), we use financial license information of commercial bank branches from the CBIRC to calculate the number of branches each bank has in each city per year. The Herfindahl-Hirschman Index is then calculated as $\mathrm{HH}{I}_{i,t}=\sum _{j=1}^J{\left(\frac{\mathrm{branc}{h}_{\mathrm{jit}}}{\sum _{j=1}^n\mathrm{branc}{h}_{\mathrm{jit}}}\right)}^2$ , where $\mathrm{branc}{h}_{\mathrm{jit}}$ is the number of branches of the j-th commercial bank in city i in year t, and n is the total number of banks in that city and year. Assuming all branches have the same efficiency, an HHI closer to zero indicates higher competition among banks; we use the average HHI from 2011 to 2019 for prefecture-level cities to measure bank competition.

Microloan Density (Density1, Density2): Referring to Beck et al. (2007) and Mookerjee and Kalipioni (2010), we estimate non-bank financial competition using the number of microloan institutions and personnel. Due to data limitations at the prefecture level, we follow Fan and Zhao (2019) by weighting the provincial number of microloan companies (Density1) and personnel (Density2) with the ratio of city-level fiscal expenditure to provincial fiscal expenditure, then taking the logarithm.

Empirical Model

To test hypotheses about digital inclusive finance’s impact on firm entry, we specify the following baseline model:

$$\begin{array}{c}\mathrm{Entr}{y}_{i,t}={\alpha }_0+{\alpha }_1\mathrm{Inde}{x}_{i,t}+{\alpha }_c{Z}_{i,t}+{\mu }_i+{\delta }_t+{\epsilon }_{i,t}\#\hfill \end{array}$$

(15)

where ${\mathrm{Entry}}_{i,t}$ is the firm entry level in city i during period t, $\mathrm{Inde}{x}_{i,t}$ denotes digital inclusive finance development in city i during period t, $Z_{i,t}$ encompasses a series of control variables, ${\mu }_i$ captures time-invariant individual fixed effects, ${\delta }_t$ represents time fixed effects, and ${\epsilon }_{i,t}$ is the disturbance term.

Endogeneity Issue: Instrumental Variable Approach

To address potential endogeneity, we adopt an instrumental variable (IV) strategy similar to Zhang et al. (2020). Specifically, we use each city’s distance from Hangzhou—the birthplace of Alipay—as an instrument for digital inclusive finance. Because Alipay originated in Hangzhou, digital inclusive finance across China correlates strongly with distance from Hangzhou, satisfying relevance. Moreover, geographic proximity to Hangzhou—a developed provincial capital—primarily satisfies the exclusion restriction as spatial distance does not directly determine entrepreneurship through omitted channels, thereby maintaining the instrumental variable’s exogeneity.

However, this cross-sectional instrument does not vary over time, so we follow Nunn and Qian (2011) by interacting distance with the national number of mobile phone users, creating a time-varying instrumental variable (Iv_distance). This approach accounts for the growing importance of mobile-based digital finance, enabling us to better identify cross-city differences in digital inclusive finance levels.

Empirical Results

Baseline Results

Table 2 presents the full-sample regression results based on model (15). Columns (2) and (4) show that the coefficients of digital inclusive finance are positive and significant at the 1% level, indicating a significant positive effect on firm entry. Regarding the control variables, industrial structure has a significant negative impact on firm entry, suggesting that when the tertiary sector comprises a larger share of the economy, it exerts a stronger constraint on new firm entry. The real interest rate also negatively and significantly affects firm entry, consistent with findings by Xie et al. (2018).

Table 2.

Digital Inclusive Finance and Firm Entry.

Variable	(1)	(2)	(3)	(4)	(5)
	First stage	FE	IV	FE	IV
	DIF	Entry	Entry	Entry	Entry
DIF		0.050*** (0.014)	0.136*** (0.041)	0.036*** (0.013)	0.099** (0.045)
Iv_distance	0.651*** (0.106)
Ln(Pgdp)				0.984 (0.598)	0.395 (0.72)
Urban				−0.756 (1.418)	−1.474 (1.5)
Strc				−1.012** (0.468)	−0.898* (0.473)
Finance				−0.302 (0.205)	−0.311 (0.193)
FDI				−2.55 (6.866)	−2.76 (6.865)
Fiscexp				1.415 (1.128)	1.589 (1.303)
Ln(Pop)				−0.267 (2.022)	−1.03 (2.378)
Rate				−1,495.172*** (374.173)	−1,498.373*** (376.114)
Constant		10.687*** (2.395)	−1.311 (3.989)	81.180*** (24.624)	100.345*** (31.222)
Kleibergen-Paap rk LM Statistics			32.18 [0.0000]		26.491 [0.0000]
Kleibergen-Paap Wald rk F Statistics			38.004 {16.38}		33.473 {16.38}
First stage F-statistic	38.000
City FE	Y	Y	Y	Y	Y
Year FE	Y	Y	Y	Y	Y
Adj. R2		.733	.725	.739	.735
Obs.	2,565	2,565	2,565	2,551	2,551

Note. Standard errors (in parentheses) are clustered at the city level. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. [ ] shows the p-value for the Kleibergen-Paap rk LM statistic, and { } indicates the 10% critical value in the Stock-Yogo weak identification test. The same notations apply to subsequent tables.

To address omitted variable bias and reverse causality, we adopt an instrumental variable (IV) approach. Columns (3) and (5) of Table 2 use the inverse distance of each prefecture-level city from Hangzhou, interacted with the national number of mobile phone users, as an instrument in a two-stage regression. The IV results are consistent with the fixed effects results. Column (1) of Table 2 displays the first-stage regression, showing that the instrument significantly and positively affects digital inclusive finance, with an F-statistic of 38.000. Additionally, the Kleibergen-Paap rk LM statistic rejects the null hypothesis of underidentification with a p-value of .000, and the Kleibergen-Paap Wald rk F-statistic exceeds the 10% critical value, indicating no weak identification problem. In summary, the instrument is valid, further supporting the positive influence of digital inclusive finance on firm entry and confirming Hypothesis 1.

According to the regression analysis results, when DIF increases by 1 unit, Entry increases by approximately 0.19% relative to its average value in the fixed effects regression shown in Table 2, column (4), or 0.52% in the instrumental variable regression shown in Table 2, column (5). Considering that the mean value of Entry is 0.18931, and Entry represents the ratio of new entering firms to existing firms, with approximately 50 million existing firms. It can be calculated that when DIF increases by 1 unit, the number of new entering firms increases by approximately 9,500 to 26,000.

Robustness Tests

Replacing the Dependent Variable

As an alternative measure of firm entry, we use Entry_per, which is defined as the per capita number of newly registered industrial and commercial enterprises. This measure accounts for population differences across regions and provides a more standardized metric for firm entry intensity. As shown in columns (1) and (2) of Table 3, the results remain robust after replacing our main firm entry variable with Entry_per.

Table 3.

Robustness Tests.

Variable	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	FE	IV	FE	IV	FE	IV	FE	IV
	Entry_per	Entry_per	Entry	Entry	Entry	Entry	Entry	Entry
DIF	1.075*** (0.183)	2.139*** (0.654)	0.040*** (0.014)	0.101** (0.045)	0.040*** (0.014)	0.111** (0.052)	0.043*** (0.013)	0.127** (0.057)
Kleibergen-Paap rk LM Statistic		26.491 [0.0000]		29.536 [0.0000]		34.270 [0.0000]		40.390 [0.0000]
Kleibergen-Paap Wald rk F Statistic		33.473 {16.38}		35.221 {16.38}		28.164 {16.38}		26.576 {16.38}
Citywide Control Variables	Y	Y	N	N	Y	Y	Y	Y
Urban District Control Variables	N	N	Y	Y	N	N	N	N
City FE	Y	Y	Y	Y	Y	Y	Y	Y
Year FE	Y	Y	Y	Y	Y	Y	Y	Y
Adj. R2	.869	.862	.740	.737	.739	.734	.745	.738
Obs.	2,551	2,551	2,480	2,480	2,506	2,506	2,551	2,551

Note. Standard errors (in parentheses) are clustered at the city level. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. [ ] shows the p-value for the Kleibergen-Paap rk LM statistic, and { } indicates the 10% critical value in the Stock-Yogo weak identification test. The same notations apply to subsequent tables.

Replacing Control Variables

Because mobile payments have a particularly strong impact on urban districts in China’s digital economy, and Alipay usage in rural areas is relatively limited, we use urban district-level data instead of city-level data for the control variables. Columns (3) and (4) in Table 3 indicate that the regression results remain robust.

Excluding Certain Cities

We exclude the four directly administered municipalities and Hangzhou to address potential first-mover advantages of digital inclusive finance in more developed areas. Columns (5) and (6) of Table 3 show that the results remain significantly positive; in fact, the coefficients increase slightly, confirming robustness.

Winsorization

To control for outliers, we apply a 1% two-sided winsorization to the dependent, independent, and control variables. Columns (7) and (8) of Table 3 report significantly positive coefficients in both the fixed-effects and IV regressions, further validating the robustness of our conclusions.

Mechanism Analysis and Heterogeneity Analysis

Mechanism Analysis

Based on theoretical analysis, we examine the mechanisms through which digital inclusive finance promotes firm entry, focusing on alleviating informational and financing constraints.

Alleviating Informational Constraints

Because firms vary greatly in informational endowments, digital inclusive finance should have a stronger effect on firms with low informational resources. Mobile applications (apps) can efficiently facilitate information sharing and data collection, so having an app indicates relatively higher informational endowment. Firms without apps typically face greater informational barriers. However, digital inclusive finance allows these low-endowment firms to adopt digital solutions, thereby reducing information circulation costs (Tang et al., 2020).

Following industry subcategories as the statistical standard, we calculate the ratio of firms with mobile apps to all existing firms in each industry using data from Qichacha. Using the median value, we classify industries as “low information-endowment industries” and “high information-endowment industries.” Columns (1) to (4) of Table 4 show that digital inclusive finance mainly influences firm entry in low-information-endowment industries, with no significant effect on high-information-endowment industries. This result supports Hypothesis 2.

Table 4.

Informational Constraints.

Variable	Low informational endowment		High informational endowment
	(1)	(2)	(3)	(4)
	FE	IV	FE	IV
	Entry	Entry	Entry	Entry
DIF	0.061*** (0.016)	0.186*** (0.053)	0.007 (0.015)	−0.012 (0.047)
Kleibergen-Paap rk LM Statistic		26.491 [0.0000]		26.491 [0.0000]
Kleibergen-Paap Wald rk F Statistic		33.473 {16.38}		33.473 {16.38}
Controls	Y	Y	Y	Y
City FE	Y	Y	Y	Y
Year FE	Y	Y	Y	Y
Adj. R2	.621	.607	.786	.786
Obs.	2,551	2,551	2,551	2,551

Note. Standard errors (in parentheses) are clustered at the city level. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. [ ] shows the p-value for the Kleibergen-Paap rk LM statistic, and { } indicates the 10% critical value in the Stock-Yogo weak identification test. The same notations apply to subsequent tables.

Mitigating Financing Constraints

In regions with well-developed traditional financial systems, intense competition among financial institutions typically lowers borrowing costs and improves credit access (Benfratello et al., 2008; Carbó-Valverde et al., 2009). By contrast, where traditional finance is less developed, digital inclusive finance bridges the gaps in financial accessibility, boosting local economic activity (Afjal, 2023). To test whether digital inclusive finance has a greater effect on firm entry in areas with weaker traditional finance, we measure credit market development in two ways: (1) the Herfindahl-Hirschman Index (HHI) for commercial banks, capturing banking competition, and (2) the log of the number of microloan institutions and practitioners (Density1, Density2), reflecting non-bank financial services.

We interact digital inclusive finance with these measures of traditional finance (HHI, Density1, Density2). Columns (1) and (2) of Table 5 show that the coefficient on DIF × HHI is significantly positive, indicating that in regions with less banking competition, digital inclusive finance plays a stronger role in promoting firm entry. Columns (3) to (6) show that the interaction coefficients are significantly negative when we use microloan density, implying that digital inclusive finance has a stronger impact where microloan services are less abundant. These findings confirm that digital inclusive finance eases financial exclusion and better supports firm entry when traditional finance is weak, consistent with Hypothesis 3.

Table 5.

Financing Constraints.

Variable	(1)	(2)	(3)	(4)	(5)	(6)
	FE	IV	FE	IV	FE	IV
	Entry	Entry	Entry	Entry	Entry	Entry
DIF	0.029** (0.013)	0.135** (0.062)	0.053*** (0.014)	0.240*** (0.053)	0.047*** (0.014)	0.195*** (0.051)
DIF × HHI	0.059** (0.025)	0.139*** (0.052)
DIF × Density1			−0.008*** (0.002)	−0.022*** (0.004)
DIF × Density2					−0.006*** (0.002)	−0.024*** (0.004)
Density1			−0.082 (0.554)	−1.286** (0.569)
Density2					−0.499 (0.503)	−1.577*** (0.528)
Kleibergen-Paap rk LM Statistic		36.783 [0.0000]		43.948 [0.0000]		41.019 [0.0000]
Kleibergen-Paap Wald rk F Statistic		9.445 {7.03}		10.633 {7.03}		12.883 {7.03}
Controls	Y	Y	Y	Y	Y	Y
City FE	Y	Y	Y	Y	Y	Y
Year FE	Y	Y	Y	Y	Y	Y
Adj. R2	.647	.560	.745	.700	.742	.699
Obs.	2,551	2,551	2,551	2,551	2,551	2,551

Note. Standard errors (in parentheses) are clustered at the city level. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. [ ] shows the p-value for the Kleibergen-Paap rk LM statistic, and { } indicates the 10% critical value in the Stock-Yogo weak identification test. The same notations apply to subsequent tables.

Dual Effects: Informational Constraints and Financing Constraints

Tables 4 and 5 show that digital inclusive finance promotes firm entry by addressing both informational and financing constraints. To verify whether the aforementioned mechanisms jointly exert influence, we add interaction terms between the digital inclusive finance index and information endowment level, as well as traditional financial indicators. Table 6 shows that digital inclusive finance simultaneously exerts a stronger effect in industries with low informational endowments (LIE) and in regions with underdeveloped traditional finance. Hence, digital inclusive finance stimulates firm entry via a dual mechanism: it eases both informational and financing constraints.

Table 6.

Informational Constraints and Financing Constraints.

Variable	(1)	(2)	(3)	(4)	(5)	(6)
	FE	IV	FE	IV	FE	IV
	Entry	Entry	Entry	Entry	Entry	Entry
DIF	0.029** (0.013)	0.135** (0.062)	0.034** (0.013)	0.187*** (0.049)	0.028** (0.013)	0.144*** (0.047)
DIF × LIE	0.029*** (0.002)	0.072*** (0.008)	0.029*** (0.002)	0.072*** (0.008)	0.029*** (0.002)	0.072*** (0.008)
LIE	1.654*** (0.118)	1.669*** (0.104)	1.654*** (0.118)	1.669*** (0.104)	1.654*** (0.118)	1.669*** (0.104)
DIF × HHI	0.059** (0.025)	0.139*** (0.052)
DIF × Density1			−0.007*** (0.002)	−0.021*** (0.004)
DIF × Density2					−0.005***	−0.022***
					(0.002)	(0.004)
Density1			0.211	−0.975*
			(0.517)	(0.548)
Density2					−0.205	−1.231**
					(0.466)	(0.496)
Kleibergen-Paap rk LM Statistic		45.524		43.948		41.019
		[0.0000]		[0.0000]		[0.0000]
Kleibergen-Paap Wald rk F Statistic		23.147		23.147		23.147
		{22.30}		{22.30}		{22.30}
Controls	Y	Y	Y	Y	Y	Y
City FE	Y	Y	Y	Y	Y	Y
Year FE	Y	Y	Y	Y	Y	Y
Adj. R2	.647	.560	.650	.543	.647	.542
Obs.	5,102	5,102	5,102	5,102	5,102	5,102

Note. Standard errors (in parentheses) are clustered at the city level. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. [ ] shows the p-value for the Kleibergen-Paap rk LM statistic, and { } indicates the 10% critical value in the Stock-Yogo weak identification test. The same notations apply to subsequent tables.

Heterogeneity Analysis

Firm Size

Digital inclusive finance plays a critical role in supporting micro and small enterprises, which often face severe informational and financing constraints. Because these firms typically have limited digital resources and encounter “scale discrimination” from traditional lenders (Cassar et al., 2015; Yan et al., 2015), digital inclusive finance uses AI-based credit scoring and mobile platforms to improve risk assessment, reducing information asymmetry and enhancing credit access (Arjunwadkar, 2018; Wang, 2015). By lowering collateral requirements and following equitable lending principles, digital inclusive finance helps micro and small firms enter the market.

To analyze whether digital inclusive finance has a stronger entrepreneurial effect for micro and small firms, we split the sample based on registered capital: 0 to 1 million (micro and small) versus above 1 million (medium and large). Because asset-size data are unavailable, we use the log of newly registered enterprises multiplied by 100 as the entry measure. Table 7 shows that digital inclusive finance significantly promotes the entry of micro and small enterprises but has no significant effect on medium and large enterprises. Hence, digital inclusive finance exerts a stronger impact on smaller firms.

Table 7.

Firm Size.

Variable	Micro and small enterprises		Medium and large enterprises
	(1)	(2)	(3)	(4)
	FE	IV	FE	IV
	Entry_no	Entry_no	Entry_no	Entry_no
DIF	0.549*** (0.170)	0.012** (0.006)	0.511*** (0.120)	0.007 (0.005)
Kleibergen-Paap rk LM Statistic		26.491 [0.0000]		26.491 [0.0000]
Kleibergen-Paap Wald rk F Statistic		33.473 {16.38}		33.473 {16.38}
Controls	Y	Y	Y	Y
City FE	Y	Y	Y	Y
Year FE	Y	Y	Y	Y
Adj. R2	.916	.915	.981	.981
Obs.	2,551	2,551	2,551	2,551

Note. Standard errors (in parentheses) are clustered at the city level. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. [ ] shows the p-value for the Kleibergen-Paap rk LM statistic, and { } indicates the 10% critical value in the Stock-Yogo weak identification test. The same notations apply to subsequent tables.

Firm Ownership Structure

In China’s diverse economy, state-owned enterprises (SOEs) benefit from preferential policies, enabling them to face fewer financing and informational constraints. Private firms, similar to micro and small enterprises, often have less access to credit due to policy bias toward SOEs (Allen et al., 2005). Foreign enterprises usually benefit from stronger technological capabilities and financial support from parent companies, reducing their information and financing barriers. Hence, private firms face the most pronounced informational and financial constraints.

The regression results in Table 8 highlight the impact of digital inclusive finance on firm entry across different ownership types. The findings reveal a significant positive relationship between digital inclusive finance and firm entry for private enterprises. Specifically, the coefficients from the fixed-effects regression (0.042) and instrumental variable regression (0.136) are notably larger than those for state-owned and foreign-funded enterprises, suggesting that digital inclusive finance plays a more significant role in facilitating firm entry among private enterprises by effectively alleviating their informational and financial constraints.

Table 8.

Firm Ownership Structure.

Variable	Private		State-owned		Foreign-funded
	(1)	(2)	(3)	(4)	(5)	(6)
	FE	IV	FE	IV	FE	IV
	Entry	Entry	Entry	Entry	Entry	Entry
DIF	0.042*** (0.014)	0.136*** (0.048)	0.022** (0.009)	−0.039 (0.027)	0.019 (0.020)	−0.057 (0.070)
Kleibergen-Paap rk LM Statistic		26.491 [0.0000]		26.491 [0.0000]		26.491 [0.0000]
Kleibergen-Paap Wald rk F Statistic		33.473 {16.38}		33.473 {16.38}		33.473 {16.38}
Controls	Y	Y	Y	Y	Y	Y
City FE	Y	Y	Y	Y	Y	Y
Year FE	Y	Y	Y	Y	Y	Y
Adj. R2	.725	.717	.380	.359	.419	.415
Obs.	2,551	2,551	2,551	2,551	2,551	2,551

Note. Standard errors (in parentheses) are clustered at the city level. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. [ ] shows the p-value for the Kleibergen-Paap rk LM statistic, and { } indicates the 10% critical value in the Stock-Yogo weak identification test. The same notations apply to subsequent tables.

Industry Type

Digital inclusive finance facilitates multidirectional communication among customers, merchants, and platforms, accelerating changes in traditional industries. To assess the impact across sectors, we compute the share of newly registered enterprises in the primary, secondary, and tertiary industries. Columns (1) and (2) of Table 9 show that the interaction term between DIF and the tertiary industry (TI) is significantly positive, indicating that digital inclusive finance increases firm entry more in the tertiary sector than in the primary or secondary sectors. Unlike the primary and secondary industries, which are often capital-intensive and have longer investment cycles, the tertiary sector benefits from smaller-scale, short-term financing. Thus, digital inclusive finance better supports labor-intensive and knowledge-driven service firms.

Table 9.

Industry Type.

Variable	Industry		Service industry
	(1)	(2)	(3)	(4)
	FE	IV	FE	IV
	Entry	Entry	Entry	Entry
DIF	0.017 (0.014)	0.010 (0.044)	−0.020 (0.014)	−0.040*** (0.015)
DIF × TI	0.033*** (0.002)	0.073*** (0.009)
TI	−0.618 (0.395)	−7.307*** (1.508)
DIF × PSI			0.023*** (0.001)	0.061*** (0.009)
PSI			−2.245*** (0.318)	−8.604*** (1.442)
Kleibergen-Paap rk LM Statistic		45.524 [0.0000]		45.524 [0.0000]
Kleibergen-Paap Wald rk F Statistic		34.735 {7.03}		34.727 {19.93}
Controls	Y	Y	Y	Y
City FE	Y	Y	Y	Y
Year FE	Y	Y	Y	Y
Adj. R2	.695	.643	.732	.680
Obs.	5,102	5,102	5,102	5,102

Note. Standard errors (in parentheses) are clustered at the city level. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. [ ] shows the p-value for the Kleibergen-Paap rk LM statistic, and { } indicates the 10% critical value in the Stock-Yogo weak identification test. The same notations apply to subsequent tables.

Following Wang et al. (2020), we further split the service sector into productive and life-oriented services. Columns (3) and (4) of Table 9 show that digital inclusive finance has a stronger positive effect on entry in productive service industries (PSI) than in life-oriented services. This suggests that digital inclusive finance drives growth in areas like R&D, digital information, modern logistics, and finance, thereby optimizing China’s industrial structure and fostering high-quality economic development.

Conclusions, Policy Implications, Limitations, and Future Directions

Conclusion

We develop an enterprise decision-making model that incorporates informational and financing constraints to explore how digital inclusive finance drives firm entry. Our findings demonstrate that digital inclusive finance has a significant impact on firm entry, as confirmed by instrumental variable regressions and multiple robustness checks. Mechanism analysis reveals that its impact is particularly strong in industries with low informational endowments, highlighting its role in reducing informational barriers. Similarly, digital inclusive finance exerts a more pronounced effect in regions where traditional finance is underdeveloped, underscoring its capacity to alleviate financing constraints. Heterogeneity analysis reveals that digital inclusive finance has a more pronounced effect on micro, small, and private enterprises, as well as on the tertiary sector, particularly in productive service industries.

Policy Implications

Based on the above research conclusions, this paper proposes the following policy recommendations:

First, strengthen policy support for digital inclusive finance in underdeveloped regions. Governments should enact policies that encourage the growth of digital inclusive finance institutions in less developed areas. These measures may include targeted tax incentives or fiscal subsidies to lower operational costs, thus improving service accessibility for entrepreneurs where traditional finance remains limited. In parallel, a comprehensive legal and regulatory framework must be established to ensure the healthy and sustainable growth of digital inclusive finance while mitigating risks.

Second, promote the deep application of digital technologies in inclusive finance to enhance firms’ informational input. Financial institutions should use big data, cloud computing, AI, and blockchain to develop robust information-sharing platforms and credit assessment systems, reducing information asymmetry and improving risk identification. Governments should bolster digital infrastructure and support digital ecosystem development, facilitating the widespread adoption of digital technologies in various industries. By encouraging technological innovation and talent training, society’s overall digitalization will rise, enabling more precise and efficient financial services for information-poor firms.

Third, provide differentiated support for micro, small, and private enterprises, especially in productive service sectors. As the heterogeneity analysis shows, these businesses gain the most from digital inclusive finance but still face hurdles in obtaining affordable credit and timely market information. Policymakers should streamline collateral requirements, encourage fintech-driven credit scoring, and promote specialized financial products tailored to smaller-scale service firms. Simultaneously, industry associations can offer technical support and mentorship, ensuring that these businesses effectively adopt digital solutions and grow sustainably.

Finally, these findings hold valuable implications for other countries. The role of digital inclusive finance in easing informational and financing constraints extends beyond China, offering significant insights for other nations. Amid slowing global economic growth and rising employment pressures, governments worldwide should leverage digital inclusive finance to boost firm entry and economic development. Countries should adapt digital inclusive finance to their local conditions, strengthen digital infrastructure, encourage financial technology innovation, and expand financial inclusivity. Further, international cooperation can facilitate shared best practices in digital inclusive finance, collaborating on global economic challenges to foster inclusive growth and sustainable development.

Limitations and Future Research Directions

Dynamic Perspectives and Network Effects

Although this study proposes a static model integrating informational and financial constraints, it does not fully capture the dynamic feedback loops or network effects commonly associated with digital technologies. Self-reinforcing processes, such as user-driven data accumulation, may amplify the impact of digital inclusive finance on firm entry over time. Future research could employ dynamic panel models or agent-based simulations to evaluate intertemporal risk adjustments and network externalities, including cross-country comparisons to verify these patterns in different institutional contexts.

Measurement and Generalizability of Informational Constraints

We measure informational constraints based on mobile app adoption; however, more direct indicators like firm-level IT usage, internet penetration, or data-literacy rates could provide further insight. Although this approach emphasizes information gaps in specific industries, future work might explore more detailed digital literacy or real-time platform usage data to gain deeper insights. Moreover, the empirical focus on Chinese prefecture-level cities limits external validity.

Firm Life Cycle, Exit, and Innovation

We focus on new firm entry but do not address how digital inclusive finance might impact firm exit or innovation. Both dimensions are crucial to comprehending the entire enterprise life cycle. Future research could merge exit data with patent or R&D measures to evaluate long-term outcomes of digital finance.

Data Constraints and Model Assumptions

Although the SAMR registration database and the Peking University Digital Inclusive Finance Index offer broad coverage, they lack granular firm-level indicators, such as capital structure or fintech-adoption details, that would enhance analyses of how digital finance functions. Obtaining alternative datasets, potentially linking administrative records or fintech platform transaction data, could yield more precise insights into entrepreneurs’ engagement with digital finance.

International Perspectives

While this study focuses on the Chinese context, experiences from other countries, such as India’s digital finance programs and Kenya’s mobile money, illustrate the global importance of bridging informational and financing gaps to foster entrepreneurship. Broader cross-national comparisons could illuminate how differences in governance, cultural norms, and technological infrastructure shape the effectiveness of digital inclusive finance across diverse environments. Such international case studies would offer valuable insights into best practices and potential pitfalls, thereby helping policymakers and stakeholders tailor digital finance solutions to promote inclusive growth and economic resilience.