FinTech and Corporate Innovation Sustainability: Evidence from China

Abstract

Innovation drives long-term firm development, but not all innovations create lasting technological impact. This study investigates how financial technology (FinTech) fosters innovation sustainability by employing a fixed-effects Poisson regression framework and using data from non-financial A-share listed companies in China from 2012 to 2020. Innovation sustainability is measured by forward patent citations within a three-year window, capturing the persistence and external impact of innovations. Our results show that regional FinTech development significantly enhances both the quantity and sustainability of firm innovation. Mechanism analysis reveals that FinTech promotes innovation sustainability by alleviating financing constraints, facilitating digital transformation, and optimizing human capital allocation. These findings provide empirical evidence on the role of FinTech in sustaining firms’ technological contributions and offer actionable insights for policymakers and managers aiming to support high-quality, long-term innovation.

1. Introduction

Innovation is a fundamental driver of firms’ long-term development and economic transformation [1,2]. However, innovation outcomes differ in the timing of their impact, as some innovations generate early and concentrated influence while others exhibit more delayed effects [3].

In this study, we use the term “innovation sustainability” as a descriptive label to emphasize the relative persistence of innovation impact compared with contemporaneous (year-of) citation-based measures. Building on this idea, we construct a measure based on patent forward citations within a three-year post-grant window to capture early-stage citation dynamics during the diffusion process. This measure follows the standard forward citation approach widely used in the literature to proxy innovation quality and technological value [4,5]. Prior evidence suggests that patent citations are highly concentrated in the early years after publication [6], indicating that early citation patterns contain important information about innovation impact.

In China, enterprise innovation often faces financing constraints, largely due to reliance on indirect financing channels [7]. Banks frequently hesitate to fund ventures lacking tangible assets or collateral, limiting firms’ ability to pursue high-impact and enduring innovations. These constraints not only restrict innovation quantity but also hinder firms from achieving sustained technological impact, highlighting the importance of mechanisms that can enhance innovation sustainability.

From a resource-based perspective, firms are conceptualized as bundles of heterogeneous and immobile resources, where the availability and quality of internal resources shape innovation outcomes [8]. Human, financial, and physical capital are key determinants that influence the translation of ideas into enduring value [9,10]. Consequently, high-quality, innovation with long-term impact requires both the generation of novel ideas and the effective utilization of internal resources to produce outcomes that can be continuously leveraged to support enterprise growth.

The rapid development of financial technology (FinTech) offers new opportunities to alleviate financing constraints and optimize internal resource allocation, both of which are critical for enhancing innovation sustainability. By integrating technologies such as big data, artificial intelligence, and blockchain, FinTech facilitates digital transformation, improves operational efficiency, and strengthens the allocation of financial and human capital. These capabilities enable firms to better utilize internal resources, streamline management processes, and attract innovative talent [11,12]. At the same time, FinTech expands access to external financing by reducing costs and enhancing credit assessment, particularly for firms with limited collateral or traditional credit histories [11,12,13]. Collectively, these mechanisms bolster firms’ ability to pursue innovations with persistent technological impact, thereby promoting long-term enterprise growth and sustainability.

We systematically examine how FinTech enhances innovation sustainability. Innovation sustainability is primarily measured using forward patent citations within a three-year window (excluding self-citations), which capture the persistence and external technological impact of innovations, while patent applications are included as a complementary indicator of innovation scale. To provide a clear analytical framework, we focus on two primary channels through which FinTech may influence innovation sustainability: easing firms’ financing constraints and improving the allocation of internal resources. In addition, we investigate industry- and region-level heterogeneity in FinTech’s effects on innovation sustainability, providing comprehensive empirical evidence that informs both policy formulation and theoretical development. Collectively, these analyses clarify the mechanisms through which FinTech supports firms’ sustained technological contributions and long-term growth.

2. Hypothesis Development

Innovation is influenced by a variety of factors at both the micro and macro levels. In this study, we focus on innovation sustainability, defined as the extent to which innovation outcomes generate persistent technological impact and create enduring value over time. Studies on patent forward citations as measures of innovation quality, and studies on sustainable innovation focusing on environmentally oriented technological development. Patent forward citations are widely used in the literature as a standard proxy for innovation quality and technological significance, as they reflect the extent to which an invention influences subsequent technological development [14,15,16]. However, forward citations primarily capture the cumulative magnitude of innovation impact over time and do not distinguish whether such impact is short-lived or persists across periods. Innovation sustainability focuses on the temporal persistence of innovation outcomes, capturing the extent to which innovative outputs maintain their influence over time, regardless of their environmental orientation.

Financial development has long been recognized as a key driver of firm innovation, improving the allocation of financial resources and supporting technological progress [17,18]. However, traditional financial systems are often constrained by information asymmetries and reliance on collateral, limiting their ability to sustain long-term innovation outcomes. Recent advances in financial technology (FinTech) have introduced new mechanisms that enhance financial efficiency, reduce frictions, and expand access to funding [19,20,21].

By providing a more stable financial environment, FinTech not only supports firms in generating innovations but also creates conditions for these innovations to persist and generate long-term technological impact. This perspective extends existing research by focusing on the durability of innovation outcomes, rather than just the quantity of innovations produced.

H1. 

FinTech development significantly promotes innovation sustainability.

The development of FinTech can substantially alleviate corporate financing constraints through two main channels. First, FinTech improves the efficiency of financial intermediation and enhances the accuracy of risk assessment, thereby reducing financial mismatches. This allows financial institutions to more accurately identify and support innovation-efficient firms, effectively lowering the cost of innovation financing [22]. Second, by leveraging advanced technologies such as big data analytics and artificial intelligence, FinTech reduces information asymmetries between firms and financial institutions, speeding up loan approvals and better aligning firms’ actual financing needs with available credit supply, particularly for innovation-driven enterprises [23]. By mitigating financing frictions, FinTech provides a more stable financial environment that supports sustained, high-quality innovation and enhances the long-term technological impact of firms’ R&D efforts.

Based on the above analysis, we propose the following hypothesis:

H2. 

FinTech enhances innovation sustainability by alleviating firms’ financing constraints.

FinTech also promotes innovation by optimizing internal resource allocation. On the one hand, as a technology-driven financial innovation, FinTech facilitates digital transformation, supporting the adoption of emerging technologies such as big data and artificial intelligence, accelerating firms’ digital upgrading and strengthening innovation capacity [24,25]. Digital transformation improves operational efficiency, reduces production costs, and reallocates resources toward innovation activities. On the other hand, FinTech supports the optimization of human capital allocation by enabling firms to better assess talent needs and attract highly skilled professionals, especially in R&D positions [26,27]. The enhancement of both knowledge and skills within firms reshapes the internal innovation environment, increases employees’ innovative engagement, and improves firms’ capacity to implement high-quality innovation projects. Through these channels, FinTech contributes not only to innovation output but also to innovation sustainability.

Based on the above analysis, we propose the following hypothesis:

H3. 

FinTech enhances innovation sustainability by facilitating digital transformation and optimizing internal human capital allocation.

Based on the above hypotheses, the overall research framework of this study is presented in Figure 1.

3. Materials and Methods

To empirically examine the impact of financial technology (FinTech) on firm-level innovation sustainability, this study constructs a panel data framework and estimates a series of econometric models using Stata 17. Unlike conventional approaches that focus on the quantity of innovation outputs, our analysis emphasizes the persistence and long-term technological impact of innovation outcomes.

Specifically, we model innovation sustainability as a function of FinTech development while controlling for firm characteristics and macroeconomic conditions. This framework allows us to identify the overall effect of FinTech and to further explore the mechanisms through which FinTech shapes firms’ innovation activities.

To uncover the underlying channels, we conduct mechanism analyses related to financing conditions, digital transformation, and human capital allocation. In addition, we perform heterogeneity analyses across industries and regions to examine whether the effects of FinTech vary across different economic environments.

Furthermore, to capture different dimensions of innovation sustainability, we conduct a set of extended analyses by focusing on innovations with stronger technological influence and transformative potential, including highly cited innovations, breakthrough innovations, and disruptive innovations.

3.1. Data Sources

This study uses a panel dataset of A-share non-financial listed firms in China over the period 2012–2020. Financial and firm-level data are obtained from the China Stock Market and Accounting Research (CSMAR) database, while patent application and citation data, as well as FinTech-related variables, are collected from the China National Research Data Service Platform (CNRDS).

To ensure data quality, firms in the financial industry are excluded, and observations with missing key variables are removed. Continuous variables are winsorized at the 1st and 99th percentiles to mitigate the influence of extreme values.

The sample period ends in 2020 to allow for a complete observation window of patent citations within three years after application. This ensures that the constructed measures can adequately capture the persistence and follow-on influence of innovation outcomes, which is central to our analysis of innovation sustainability.

3.2. Variables and Econometric Model Specification

• Dependent Variables

This study focuses on innovation sustainability, defined as the persistence and long-term technological impact of firms’ innovation outcomes. To capture this concept, we primarily employ patent citation-based measures.

Specifically, the main dependent variable is constructed using the number of forward citations received by a firm’s patents within a three-year window after application, excluding self-citations. Patent citations are widely used to reflect the technological influence of innovations, as they indicate the extent to which a given invention is subsequently utilized and built upon by other inventors [28,29,30]. Compared with patent counts, citation-based measures better capture the persistence and continued relevance of innovation outcomes rather than short-term outputs.

The use of a three-year citation window helps ensure comparability across cohorts of patents and mitigates truncation bias while also capturing the early diffusion and recognition of innovations. Consistent with prior evidence showing that patent citations are heavily concentrated in the early years after publication [6], the three-year window places greater emphasis on the timely and economically relevant diffusion of technological knowledge, which is more closely aligned with firms’ ongoing innovation activities and strategic decisions. In addition, focusing on a shorter window allows us to better capture the near-term technological impact of innovations, rather than long-delayed effects that may be influenced by external factors.

Excluding self-citations further reduces potential bias arising from strategic citation behavior and provides a cleaner measure of external technological impact. Therefore, this measure effectively reflects the extent to which firms generate innovations with sustained technological influence, consistent with the concept of innovation sustainability.

For completeness, we also include patent application counts as an auxiliary measure of innovation output. Patent applications reflect firms’ R&D efforts and technological accumulation, particularly in the case of invention patents, which require higher levels of novelty and practical applicability. This complementary measure allows us to distinguish between the scale of innovation and its sustainability.

2.

Explanatory Variables

The primary explanatory variable is regional FinTech development, measured as the number of FinTech companies in the region where the enterprise is located [31,32]. We take the natural logarithm of this count to capture variations in regional FinTech intensity.

This indicator reflects the overall development of the regional FinTech ecosystem, capturing the scale of FinTech-related industrial activity and the degree of FinTech agglomeration. It provides a comprehensive measure of the breadth and density of FinTech market participation within a region.

3.

Control Variables

Control variables include firm-level characteristics such as the natural logarithm of total assets, firm age, return on net assets, board size, leverage, and whether the chairman and CEO positions are combined [33,34]. Regional-level controls include GDP per capita, urbanization rate, and foreign investment levels. Firm, region, and year fixed effects are included to mitigate potential omitted variable bias.

4.

Mechanism Variables

Digital Transformation: Using the approach of Mobina Zareie et al. [35], we construct a firm-level digital transformation index based on textual analysis of firms’ annual report disclosures. Specifically, we measure the frequency of digital-related keywords (e.g., artificial intelligence, big data, and cloud computing) relative to the total word count, capturing the extent of firms’ digitalization. The final set of keywords is reported in Appendix B.

Financial Mismatch: Following Shao [36], the cost of capital—calculated as interest expenses divided by total liabilities, minus accounts payable and adjusted by industry average capital costs—is used as a proxy for financial mismatch.

Credit Availability: Measured as the ratio of bank loans (short-term and long-term) to total assets, reflecting firms’ access to external financing.

Human Capital Optimization: Assessed from two perspectives: knowledge (ratio of graduate employees to total employees) and skills (ratio of R&D personnel to total employees), capturing both the educational composition and technical capability of the workforce.

5.

Econometric Model

Given that innovation sustainability are non-negative discrete counts, traditional linear models may produce biased estimates [37]. Therefore, we employ a Poisson Pseudo-Maximum Likelihood (PPML) estimator with high-dimensional fixed effects as the baseline specification. The PPML estimator is particularly suitable for count data with a large number of zeros and potential overdispersion. Importantly, it does not rely on the equidispersion assumption and remains consistent under general forms of heteroskedasticity. This makes it widely applicable in the analysis of innovation outcomes, which are typically skewed and overdispersed.

The conditional mean of innovation outcomes is specified as:

$$\mu =E\left({\mathit{inno}}_{ijt}\mid X_{ijt}\right)=exp\left({\alpha }_1+{\beta }_1{\mathit{fintech}}_{jt}+{\gamma }_1{X}_{ijt}+\mathit{year}+\mathit{firm}+\mathit{city}+{\epsilon }_1\right)$$

where ${\mathit{inno}}_{ijt}$ denotes patent counts or citation counts for firm $i$ in industry $j$ at year $t$, ${\mathit{fintech}}_{jt}$ represents regional FinTech development, and $X_{ijt}$ includes control variables. The Poisson specification accommodates the discrete, skewed nature of patent data and provides consistent estimates for count-dependent variables.

Table 1. Descriptive Statistics.

Variable	Mean	SD	Min	Max	N
Number of patent applications	15.985	118.872	0	3765	16,029
Innovation sustainability	46.957	388.167	0	14,716	16,029
Level of economic development	11.442	0.513	9.084	13.056	16,029
Urbanization level	0.742	0.150	0.182	1.000	16,029
Level of foreign investment	0.026	0.017	0	0.198	16,029
Firm size	22.268	1.265	19.886	26.164	16,029
Asset–liability ratio	3.420	2.850	1.122	19.792	16,029
Return on total assets (ROA)	0.040	0.063	0.031	0.214	16,029
Firm age (log)	2.876	0.311	0	3765	16,029
Board size (log)	2.126	0.195	0	14,716	16,029
CEO duality	0.279	0.449	9.084	13.056	16,029

presents the descriptive statistics of the main variables used in this study. The sample consists of 16,029 firm-year observations. On average, firms apply for about 16 patents per year, while the number of patent citations varies considerably, with a mean of 46.96 and a maximum value of 14,716, indicating large differences in innovation quality across firms. The level of economic development (measured by the logarithm of regional GDP per capita) averages 11.44, and the urbanization level is 0.74, reflecting a relatively high degree of urban concentration. The average foreign investment level is 0.026, suggesting that foreign capital accounts for a small proportion of total investment.

4. Results

4.1. Benchmark Regression

Table 2. Benchmark Regression Results.

	(1)	(2)	(3)	(4)	(5)	(6)
	Patent Applications for Inventions	Patent Applications for Inventions	Patent Applications for Inventions	Innovation Sustainability	Innovation Sustainability	Innovation Sustainability
fintech	0.181 ***	−0.0754 *	0.162 ***	0.181 ***	−0.0927 **	0.151 ***
	(0.0247)	(0.0391)	(0.0518)	(0.0279)	(0.0434)	(0.0567)
GDP per capita		0.188	0.237 ***		0.0980	0.260 ***
		(0.170)	(0.0859)		(0.221)	(0.0869)
Urbanization		2.680 ***	1.626 ***		3.512 ***	1.970 ***
		(0.533)	(0.528)		(0.596)	(0.618)
Foreign investment		3.553	−3.025 **		4.147	−1.237
		(2.599)	(1.460)		(2.741)	(1.940)
Total assets		0.932 ***	0.330 ***		0.911 ***	0.251 ***
		(0.0371)	(0.0637)		(0.0359)	(0.0768)
Leverage		0.0354 ***	−0.00824		0.0364 ***	−0.0168
		(0.0113)	(0.00855)		(0.0128)	(0.0109)
Return on net assets		6.507 ***	2.884 ***		6.236 ***	2.671 ***
		(0.811)	(0.488)		(0.917)	(0.535)
Firm age		0.100	−1.332 ***		0.0847	−0.856
		(0.131)	(0.455)		(0.142)	(0.551)
Board size		1.031 ***	0.225		1.249 ***	0.250
		(0.280)	(0.193)		(0.296)	(0.205)
Dual		0.667 ***	−0.0331		0.644 ***	0.00318
		(0.103)	(0.0674)		(0.113)	(0.0867)
Constant	1.713 ***	−26.29 ***	−4.290 *	2.787 ***	−24.71 ***	−3.111
	(0.103)	(2.180)	(2.253)	(0.113)	(2.594)	(2.492)
Fixed Effects	No	No	Yes	No	No	Yes
Observations	23,210	21,574	16,029	23,210	21,574	16,029

The numbers in parentheses are robust standard errors; *, **, and *** represent significance levels of 10%, 5%, and 1%, respectively.

presents the baseline regression results examining the impact of regional financial technology (FinTech) development on firm-level innovation outcomes. The sample consists of non-financial A-share listed companies in China from 2012 to 2020. Across all specifications, we progressively introduce firm- and region-level control variables and further incorporate year, firm, and city fixed effects to ensure robustness.

Columns (1)–(3) report the results for invention patent applications, while Columns (4)–(6) report the results for patent citations, which capture the persistence and long-term technological impact of innovation outcomes.

The results consistently show that the coefficient of the FinTech variable is positive and statistically significant across all model specifications. This indicates that regional FinTech development promotes both the quantity and the sustainability of corporate innovation. In the fully specified model with comprehensive controls and fixed effects (Columns (3) and (6)), a one-unit increase in the FinTech index is associated with a 16.15% increase in patent applications and a 15.14% increase in patent citations.

By focusing on forward citations within a three-year window and excluding self-citations, we capture the sustained influence of innovations and their contribution to firms’ long-term technological value.

Overall, these benchmark results provide robust evidence that regional FinTech development enhances both the scale and persistence of firm-level innovation. These results provide support for Hypothesis 1, which states that regional FinTech development enhances firm-level innovation outcomes, including both innovation output and its sustainability. Subsequent analyses further examine heterogeneity across industries and regions, as well as the role of high-impact, breakthrough, and disruptive innovations in shaping innovation sustainability.

4.2. Robustness Tests

We construct an alternative measure of regional FinTech development using the volume of Baidu news searches. In addition, we employ a comprehensive Digital Economy Development Index, combining Internet development and the Baidu Digital Financial Inclusion Index, as a proxy for FinTech development.

As shown in

Table 3. Substituting Explanatory Variables.

	(1)	(2)	(3)	(4)
	Fintech Search Index		Fintech Comprehensive Development Index
	Patent Applications for Inventions	Innovation Sustainability	Patent Applications for Inventions	Innovation Sustainability
$fintech$	0.1116 * (0.0601)	0.1606 *** (0.0605)	2.7575 *** (0.3838)	3.0748 *** (0.4354)
Enterprise-level control variables	YES	YES	YES	YES
Regional-level control variables	YES	YES	YES	YES
Year-to-year fixed effect	YES	YES	YES	YES
Corporate fixation	YES	YES	YES	YES
Regional fixed effects	YES	YES	YES	YES
Pseudo R2	0.9148	0.9290	0.9167	0.9311
Observations	15,960	15,960	16,029	16,029

The numbers in parentheses are robust standard errors; * and *** represent significance levels of 10% and 1%, respectively.

, the results provide consistent evidence that FinTech development substantially promotes corporate innovation sustainability. Patent citations, as a key indicator of innovation sustainability, reflect the sustained utilization of firms’ intellectual outputs in academia and industry, directly supporting long-term corporate sustainability. Overall, the results provide consistent evidence that the positive impact of FinTech on both innovation scale and sustainability is robust across alternative measurement approaches.

4.3. Endogeneity

To mitigate potential endogeneity concerns, all explanatory and control variables are lagged by one period. Although the region-to-firm-level panel design already alleviates some endogeneity, we further implement a control function approach for nonlinear models. Following the procedure of Lin and Wooldridge [38], the first stage employs a fixed-effects regression to address endogeneity in the FinTech variable, using the manually aggregated average FinTech penetration in neighboring cities as an instrumental variable—a method widely adopted in FinTech research [39]. In the second stage, a fixed-effects Poisson model is estimated, with standard errors computed via the Bootstrap method to ensure robustness. This strategy effectively isolates unobserved factors affecting regional FinTech development, allowing for a cleaner identification of the causal effect of FinTech on innovation sustainability. While this instrument may not fully eliminate all concerns regarding exclusion restrictions, it has been widely adopted in the FinTech literature and provides a reasonable source of exogenous variation [32,40].

Table 4. Control Function Approach.

	(1) Patent Applications for Inventions	(2) Innovation Sustainability
$fintech$	0.2533 ** (0.1034)	0.2160 * (0.1133)
Enterprise-level control variables	YES	YES
Regional-level control variables	YES	YES
Year-to-year fixed effect	YES	YES
Corporate fixation	YES	YES
Regional fixed effects	YES	YES
Pseudo R2	0.8896	0.9109
Observations	13,931	13,792

The numbers in parentheses are robust standard errors; *, and ** represent significance levels of 10% and 5%, respectively.

reports the regression results after addressing endogeneity. Columns (1) and (2) present the impact of FinTech development on invention patent applications and patent citations, respectively. After applying the two-stage control function method combined with fixed effects Poisson regression, the coefficients of the core explanatory variable remain significantly positive, 0.2533 and 0.2160, significant at the 5% and 10% levels.

These findings indicate that even after correcting for endogeneity, FinTech development continues to exert a statistically and economically meaningful positive effect on both the sustainability and quality of firm-level innovation. Notably, the increase in patent citations reflects the sustained utilization of firms’ intellectual outputs, supporting long-term corporate sustainability. Overall, the results strengthen the causal interpretation that FinTech effectively fosters substantive innovation activity, providing robust support for high-quality, sustainable enterprise development.

While the use of neighboring cities’ FinTech development as an instrument is motivated by spatial spillovers, concerns may arise regarding the exclusion restriction, as regional FinTech may be correlated with broader economic conditions.

However, such spillovers are expected to affect firms primarily through financial intermediation channels, such as easing financing constraints, rather than directly influencing firm-level innovation [41,42,43].

To address this concern, we include comprehensive fixed effects and conduct a supplementary test, showing that the instrument has no significant direct effect on innovation outcomes after controlling for local FinTech development (see Appendix A).

A potential concern is reverse causality, whereby regions with stronger innovation capacity may attract FinTech agglomeration, leading to a spurious correlation between FinTech development and firm innovation. To address this issue, we explicitly examine the dynamic relationship between FinTech development and innovation outcomes by incorporating both lagged and lead values of FinTech in the baseline specification. This lead–lag specification allows us to test whether future FinTech development predicts current firm innovation, which would indicate potential reverse causality or simultaneity bias.

Table 5. Lead–lag Analysis and Reverse Causality Test.

	(1) Patent Applications for Inventions	(2) Innovation Sustainability
Lagged fintech	0.174 *** (0.0522)	0.153 *** (0.0590)
Future fintech	0.0803 (0.0517)	0.0920 (0.0567)
Enterprise-level control variables	YES	YES
Regional-level control variables	YES	YES
Year-to-year fixed effect	YES	YES
Corporate fixation	YES	YES
Regional fixed effects	YES	YES
Observations	13,030	YES

The numbers in parentheses are robust standard errors; *** represents significance levels of 1%, respectively.

reports lead–lag regression results examining potential reverse causality between FinTech development and firm innovation. The results show that while lagged FinTech is positively associated with firm innovation, the coefficients on the lead terms are statistically insignificant. This suggests that future FinTech development does not predict current innovation outcomes, alleviating concerns that reverse causality drives our baseline results.

Overall, these findings support the temporal ordering assumption underlying our identification strategy.

4.4. Mechanism of FinTech Impact on Innovation Sustainability

To further explore the potential channels through which FinTech development is associated with corporate innovation sustainability, we examine several mechanisms along two dimensions: the alleviation of financing constraints and the optimization of internal resource allocation. Existing studies suggest that FinTech development can reduce information asymmetries [44] in financial markets, improve the efficiency of credit allocation [45,46], and facilitate firms’ adoption of digital technologies [47,48], and reshaping firms’ internal resource structure [49,50]. Accordingly, we examine digital transformation, financial mismatch, credit availability, and human capital optimization as key mechanisms through which FinTech may influence corporate innovation outcomes.

Building on the theoretical framework, we conduct a set of mechanism analyses to assess whether the empirical patterns are consistent with these proposed channels. Specifically, digital transformation and human capital optimization are used to capture internal resource allocation, while financial mismatch and credit availability proxy for financing constraints. These analyses allow us to provide suggestive evidence on the channels through which FinTech may influence firm innovation outcomes.

According to Column (1) of

Table 6. Mechanism Analysis.

	(1)	(2)	(3)	(4)
	Digital Transformation	Financial Mismatches	Financing Optimization	Human Capital Optimization
$fintech$	0.0278 * (0.0145)	−0.0447 *** (0.0107)	0.0038 ** (0.0019)	0.0129 *** (0.0013)
Enterprise-level control variables	YES	YES	YES	YES
Regional-level control variables	YES	YES	YES	YES
Year-to-year fixed effect	YES	YES	YES	YES
Corporate fixation	YES	YES	YES	YES
Regional fixed effects	YES	YES	YES	YES
Adj R2	0.7898	0.3069	0.7254	0.7525
Observations	21,475	21,464	15,342	21,462

The numbers in parentheses are robust standard errors; *, **, and *** represent significance levels of 10%, 5%, and 1%, respectively.

, regional FinTech development is positively associated with firm-level digital transformation, suggesting that FinTech facilitates firms’ adoption of digital technologies. This finding is consistent with the view that digital transformation enhances production efficiency and data utilization, thereby improving firms’ capacity for sustained innovation. Existing studies further show that digital transformation is closely linked to improved innovation performance by reducing information frictions and enabling more efficient resource allocation [51,52].

Columns (2) and (3) indicate that FinTech development is positively associated with firms’ credit availability, suggesting that FinTech helps alleviate financing constraints, particularly for innovation-intensive firms that require substantial external funding. Prior literature documents that improved access to external finance significantly enhances firms’ innovation investment and long-term technological development, as it reduces funding uncertainty and supports continuous R&D activities [53,54,55].

Finally, Column (4) shows that FinTech development is positively associated with improvements in human capital within firms. This finding suggests that FinTech may facilitate more efficient allocation and upgrading of human resources. Existing studies indicate that higher-quality human capital significantly enhances firms’ innovation capability by improving knowledge accumulation, absorptive capacity, and the efficiency of innovation processes [56,57].

Overall, these results provide empirical evidence for the proposed mechanisms through which FinTech affects innovation sustainability, namely digital transformation, financing constraints alleviation, and human capital optimization.

4.5. Heterogeneity Analysis

Innovation activities vary substantially across industries, reflecting differences in technological challenges and dependence on continuous innovation. Strategic emerging industries, characterized by high technology intensity and rapid product evolution, require firms to engage in independent innovation to survive in competitive markets. In contrast, non-strategic emerging industries are less reliant on technological innovation for firm survival [58]. To examine how the impact of FinTech on innovation differs by industry, we categorize firms into strategic and non-strategic emerging industries. Given their higher innovation intensity and greater dependence on external financing, firms in strategic emerging industries are expected to benefit more from FinTech development [59,60], as FinTech can alleviate financing constraints and improve the efficiency of resource allocation in innovation activities.

Geographical disparities are another salient feature of China’s economy. Regional differences in economic structure, industrial composition, and technological infrastructure may lead to heterogeneity in the effect of FinTech. Accordingly, we divide regions into eastern and non-eastern areas to explore geographical heterogeneity [60,61]. Due to more developed financial markets, stronger digital infrastructure, and higher adoption of financial technologies, firms in eastern regions are expected to experience a stronger positive impact of FinTech on innovation sustainability.

As shown in

Table 7. Industry Heterogeneity.

	(1)	(2)	(3)	(4)
	Strategic Emerging Industries		Non-Strategic Emerging Industries
$fitech$	0.2248 *** (0.0597)	0.2143 *** (0.0605)	−0.0010 (0.0672)	0.0013 (0.0778)
Enterprise-level control variables	YES	YES	YES	YES
Regional-level control variables	YES	YES	YES	YES
Year-to-year fixed effect	YES	YES	YES	YES
Corporate fixation	YES	YES	YES	YES
Regional fixed effects	YES	YES	YES	YES
Pseudo R2	0.9319	0.9412	0.8287	0.8555
Observations	9156	9156	6872	6872

The numbers in parentheses are robust standard errors; *** represents significance levels of 1%, respectively.

, FinTech significantly increases both patent applications and citations for firms in strategic emerging industries, with coefficients of 0.2248 and 0.2143, respectively, both significant at the 1% level. This indicates that in technology-intensive sectors where continuous innovation is critical for maintaining a competitive advantage, FinTech effectively promotes not only the generation of innovation outputs but also the long-term technological impact, enhancing innovation sustainability. In contrast, for firms in non-strategic emerging industries, the coefficients are statistically insignificant, suggesting that the effect of FinTech on innovation sustainability is relatively limited in industries less dependent on continuous technological advancement, highlighting the heightened sensitivity and reliance of high-tech sectors on FinTech-enabled innovation support.

Table 8. Regional Heterogeneity.

	(1)	(2)	(3)	(4)
	Eastern Region		Non-Eastern Regions
$fitech$	0.3077 *** (0.0677)	0.2802 *** (0.0729)	−0.0855 (0.0700)	−0.0495 (0.0687)
Enterprise-level control variables	YES	YES	YES	YES
Regional-level control variables	YES	YES	YES	YES
Year-to-year fixed effect	YES	YES	YES	YES
Corporate fixation	YES	YES	YES	YES
Regional fixed effects	YES	YES	YES	YES
Pseudo R2	0.9273	0.9374	0.8310	0.8692
Observations	11,270	11,270	4743	4743

The numbers in parentheses are robust standard errors; *** represents significance levels of 1%, respectively.

shows that in the eastern region, FinTech significantly enhances both patent applications and citations, with coefficients of 0.3077 and 0.2802, respectively, both significant at the 1% level. Conversely, in non-eastern regions, the effects are statistically insignificant, suggesting that the innovation-enhancing effect of FinTech is more pronounced in regions with more advanced financial infrastructure, higher adoption of digital technologies, and a business environment conducive to innovation.

Overall, firms are the primary drivers of technological progress and play a central role in industrial upgrading and long-term economic growth. Our empirical findings indicate that FinTech not only increases the quantity of invention patents but also strengthens their long-term impact, thereby enhancing innovation sustainability. Mechanism analyses show that FinTech promotes sustainable, high-quality innovation by facilitating digital transformation, alleviating financial mismatches, improving credit access, and optimizing human capital allocation. Heterogeneity analyses further reveal that these effects are strongest in strategic emerging industries and in the eastern region, emphasizing the critical role of both industry characteristics and regional contexts in shaping the innovation-promoting effects of FinTech.

5. Conclusions and Discussion

This study, using data from A-share non-financial listed companies from 2012 to 2020, examines the impact of FinTech on innovation sustainability at the firm level. The results show that regional FinTech development significantly increases both patent applications and forward citations, indicating that FinTech not only expands innovation output but also enhances the persistence and technological influence of innovations.

Mechanism analysis suggests that FinTech promotes innovation sustainability through two main channels: alleviating financing constraints and improving internal resource allocation. By enhancing financial intermediation efficiency and risk assessment, FinTech reduces financial mismatches and expands access to credit. At the same time, it facilitates digital transformation and optimizes human capital deployment, supporting firms in undertaking long-term and high-value innovation activities.

Further analysis shows that FinTech has a stronger effect on exploratory and breakthrough innovations, and its impact is more pronounced in strategic emerging industries and eastern regions. These findings suggest that FinTech plays a key role not only in increasing innovation quantity but also in shaping the direction and persistence of technological progress.

This study also contributes to the literature on citation-based measures of innovation quality by introducing a temporal dimension that captures the persistence of innovation outcomes. While existing studies widely use forward patent citations as a proxy for innovation quality and technological impact, they typically emphasize the cumulative magnitude of citations without explicitly distinguishing the timing and persistence of such impacts [14,15,16]. By focusing on citation outcomes within a fixed post-application window, this paper provides a complementary perspective that highlights the durability of innovation effects over time. This temporal perspective enriches the understanding of innovation quality by shifting attention from purely aggregate impact to the persistence structure of technological influence.

These findings provide mechanism-based policy implications. The results suggest that FinTech may enhance firms’ innovation outcomes through alleviating financing constraints, facilitating digital transformation, and improving human capital allocation. Accordingly, supporting the development of digital financial infrastructure and FinTech ecosystems may help strengthen firms’ innovation capacity. The evidence of regional heterogeneity further indicates that the effectiveness of FinTech depends on underlying financial and digital infrastructure, implying that complementary improvements in financial and digital environments are important for fully realizing its innovation-enhancing effects. Importantly, these implications should be interpreted as contingent on similar structural and institutional conditions rather than as universal policy prescriptions.

While this study provides robust evidence, several limitations remain. Patent data are limited through 2020, and measures of innovation sustainability rely on publicly available information, potentially overlooking certain innovative activities. In addition, although patent citations are widely used to proxy the persistence and technological influence of innovation, they may not fully capture other important dimensions of innovation sustainability, such as economic value, commercialization outcomes, and broader social impact. Future research could therefore develop more comprehensive measures by integrating multiple data sources and alternative indicators. Although robustness checks, instrumental variable approaches, and control function methods were employed, unobserved factors may still influence the results. Future research could extend the sample period, examine emerging FinTech tools such as blockchain and AI-driven financial services, explore the heterogeneous effects of different types of FinTech on innovation, and conduct cross-country comparisons to better understand how institutional and industrial contexts shape FinTech’s impact on corporate innovation sustainability.