Digital Financial Inclusion to Corporation Value: The Mediating Effect of Ambidextrous Innovation

Abstract

Corporate innovation can be subdivided, according to its approach and novelty, into exploitative innovation and exploratory innovation, i.e., ambidextrous innovation. Defined as actions to promote financial inclusion through digital financial services, digital financial inclusion brings new opportunities for the implementation of corporate innovation projects and the improvement of corporation value. Based on the annual reports (2012–2020) released by 1604 listed SMEs in China and the index of digital financial inclusion from Peking University, this paper explores the way that digital financial inclusion affects the corporation value of SMEs, with some moderating factors such as the financial flexibility, corporate social responsibility and product market competition in ambidextrous innovation. The study shows, in SMEs: (1) digital financial inclusion has a significant positive impact on exploitative innovation, but has less effect on exploratory innovation with a time lag; (2) ambidextrous innovation plays a partial intermediary role in the effect of digital financial inclusion on corporation value; (3) financial flexibility of the enterprise positively moderates the relationship between digital financial inclusion and corporate value. In the short term, corporate social responsibility negatively moderates the relationship between digital financial inclusion and corporate value; however, in the long term, it does contribute to the growth of corporate value. (4) Product market competition positively moderates the relationship between digital financial inclusion and exploitative innovation, but does not positively moderate the relationship between digital financial inclusion and exploratory innovation.

1. Introduction

With the recent rapid development of information technologies, digital technologies such as big data and cloud computing continue to be integrated deeply with financial inclusion [1]. Thus, the digital financial inclusion (hereafter, DFI) has experienced a great leap forward in China. Supplementing the traditional financial service system, DFI has a significant impact on the production and operation of enterprises, which has become one of the important directions of the global financial reform, according to the white paper of global partnership for financial inclusion (GPFI) released at the G20 Hangzhou summit [2]. The G20 DFI Advanced Principles state that DFI has become one of the most important directions for global financial reforms. As one of the stability factors in China, DFI narrows the income gap between the urban and rural areas [3], and plays a catalytic role in promoting the residents’ consumption [4]. Moreover, DFI is generally considered to break the Pareto Principle of traditional finance mode, and to alleviate the relative poverty [5], with remarkable results in promoting high-quality economic development.

In the traditional financial market, which is dominated by indirect financing, financial institutions often face strong financing constraints for innovative SME activities, based on a combination of risk and profitability considerations. As an important financial innovation in the digital economy, digital inclusive finance must demonstrate its inclusive nature and provide affordable and formal financial support to SMEs. Digital inclusive finance can effectively absorb idle social capital into financial supply and enhance the scale and efficiency of financial market capital supply, which is both a stock optimization and an important incremental supplement to financial resources. The use of digital technology in the area of inclusive finance promotes innovation in financial products and services, and provides more diversified financial support for SMEs to carry out innovative activities.

Similar to financial inclusion, researchers on DFI pay more attention to its constraints on corporate financing [6,7,8], corporate innovation (e.g., [9,10]) and investment efficiency [11,12]. These studies provide important references for this paper, but the corporate activities mainly aim to maximize the corporation value. However, there is little discussion of the relationship between DFI and corporation value. On one hand, DFI can expand the financing channels to meet the diversified service needs of enterprises [13] and provide economic benefits for their value growth [14], which effectively overcomes the distance limitations of traditional finance [15]. On the other hand, DFI can improve the transparency of business-related information [16] to reduce the financial mismatch risks and friction costs [17], which alleviates the information asymmetry between investors and firms [18] in the big data traceability and anti-fraud fields.

However, there are obvious drawbacks to DFI. It may increase financial risks due to the complexity and chain reaction of digital technology [19], and some scholars considers that the digital technology cannot change the invisibility, suddenness or contagion of financial risks [20]. Especially in the contagion of the network and cross-market, it is easy to exacerbate the financial market turmoil [21], which can ultimately lead to unreasonable investment psychology and enterprise vitality in micro-economics [22]. The management ability of the enterprise is greatly important regarding the opportunities and risk brought by DFI [23], which mainly refers to financial flexibility [24,25], including the financial resources [26], investment opportunities [27] and maximization of enterprise [28,29]. In the crisis-ridden environment of DFI, companies with higher financial flexibility could deal with the financial risks and external uncertainties, thus capturing the opportunities of DFI. Moreover, corporate social responsibility (hereafter, CSR) is closely related to the social environment in the external environment of the company. Thus, the companies that actively disclose social responsibility reports tend to have a better corporate image and more investment confidence [30,31] in the development of DFI. It can be seen that the influencing factors, direction and effect of DFI on the value of SMEs are still uncertain; therefore, it is necessary to study the further theoretical and logical derivation of their relationship using empirical tests. In the digital economy, DFI can be used to improve the quality and efficiency of financial services, which makes a great contribution to and has practical value in high-quality economic development.

With the implementation of the Innovative National Strategy in China, the independent innovation capability is particularly important to the developing strategy of enterprise, and the emerging factors, such as technology and information, are the driving force for the company’s growth. Previous studies have shown that DFI can improve the technological innovation capabilities of companies through various transmission mechanisms [7,13,32]. For the innovation behavior of the enterprise, exploitative innovation focuses on the existing products and services to meet customer and market needs [33], while exploratory innovation seeks to identify potential opportunities for future business activities [34]. In contrast, the latter faces longer risk cycles and more uncertainty, and both are important ways to affect corporate performance, which need to be taken into account. However, exploitative innovation and exploratory innovation are opposite behaviors in a continuous activity within the enterprise, and some factors will influence the choice of innovation methods [35], such as enterprise goal, internal and external environment, etc. In addition, why do SMEs of the same size, function and structure still differ in creating corporation value, even if they have the same indicators of DFI? Therefore, the impacts of DFI on the value of SMEs in different situations need to be discussed in detail.

2. Theory and Assumptions

2.1. Development of DFI and Corporation Value

Enterprises are the main driving force of social and economic growth, and financial development is conducive to enhancing corporation value, which has been verified in some research about their relationship with the aspects of financing constraints, corporate ownership structure, innovation ability and internal governance ability [35]. However, there are some problems with the traditional financial structure, such as space limitations [15], asymmetry of qualification guarantee and information [6], which will slow down the efficiency of financial resource allocation, and then result in an imbalance of market capital supply and demand [14]. With the development of technology, the combination of digital technology represented by the Internet and finance extends financial services to areas that are difficult to reach with traditional finance [3], providing financial services to a wider range of groups, and thus, DFI has emerged [4]. DFI became a hot topic at the G20 Summit, held in Hangzhou, China in September 2016, when the G20 High Level Principles on Digital Financial Inclusion were officially released. Taking advantage of digital technology, DFI can enable financial services to reduce transaction costs and expand the coverage of financial services [36], thereby enhancing the efficiency of regional allocation of innovation factors and alleviating the dilemma of financing constraints faced by SMEs in implementing innovation [2]. In the external environment of enterprises, DFI can significantly enhance the SMEs value of the listed companies in long-term and vertical extension, especially for the disclosure of financial structure and quality of information [16,18]. In order to adapt to the changes in the external financial and digital environment, SME will inevitably optimize its own organizational structure and improve the management in the internal incentive and control mechanism [37]. Based on the above analysis, this paper proposes the following hypothesis:

Hypothesis 1 (H1).

The development of DFI has a significant positive impact on corporation value.

2.2. Ambidextrous Innovation and Corporation Value

The choice of innovation model is crucial to the innovation output and sustainability of a firm [38]. Based on the mode of innovation and the degree of novelty, corporate innovation can be subdivided into exploitative innovation and exploratory innovation, i.e., ambidextrous innovation. Specifically, exploitative innovation is an ongoing, small-scale innovation activity that emphasizes the improvement, adaptation or extension of existing products, services and technologies, and is characterized by low risk and quick results, but low returns. Exploratory innovation is essentially a transformative, large-scale innovation activity, which emphasizes the complete transformation or development of products, services, processes, etc., and is characterized by high risks and slow results, but high benefits [39]. As China’s economy enters the era of Industry 4.0, the complex and changing external environment places higher demands on innovation, and both of these innovation models are indispensable for enterprises. This is because if the former is overemphasized, enterprises may become overly dependent on their own technology and resources, and may fall into the dilemma of “path dependence”, which will eventually lead to a decline in their own innovation capacity. If the latter is overemphasized, enterprises may become obsessed with developing new products and markets, and may fall into the cycle of “innovation–failure–re-innovation”, which eventually leads to innovation failure and a sudden increase in risk [40]. This means that in order to gain a sustainable competitive advantage, companies need to balance the contradictions between exploitative and exploratory innovation, and promote the synergy and balanced development of both. Currently, many scholars have analyzed the factors influencing ambidextrous innovation from various perspectives, such as managerial characteristics [41], leadership styles [42] and technology habitual domain [43].

Ambidextrous innovation can effectively cultivate the dynamic capabilities of enterprises and create corporation value with the competitive advantages for sustainable development [44]. Exploitative innovation is important for improving short-term financial performance, and exploratory innovation is critical for a long-term competitive advantage. Exploitative innovation helps the company to provide better product quality and service at a low cost, which is effective in resource utilization and conversion rates, and promotes short-term financial performance [45]. Exploratory innovation is conducive to the research and development of new products, with more opportunities in emerging markets and stronger core competitiveness of enterprises [33]. In the long term, it could help the enterprise to gradually master the right to speak in the industry with a strong competitive advantage, and even change the competition model of the industrial market with technological breakthroughs. Accordingly, this paper proposes the following assumptions:

Hypothesis 2 (H2).

Ambidextrous innovation has a significant positive impact on corporation value.

Hypothesis 2a (H2a).

Exploitative innovation has a significant positive impact on corporation value.

Hypothesis 2b (H2b).

Exploratory innovation has a significant positive impact on corporation value.

2.3. DFI and Ambidextrous Innovation

MM theory shows that in a perfect capital market, there is no difference between internal and external funds for corporate decision makers [46]. However, the research and development (hereafter, R&D) not only requires huge, continuous capital investment, but also has high uncertainty, a long cycle and irreversible investment, so enterprise managers often lack incentives for technological innovation [47]. In addition, the information asymmetry between enterprises and innovative investors can easily lead to potential moral hazards and financing constraints.

Since March (1991) [48] defined “exploitation” and “exploration” as different organizational learning paradigms and suggested that they compete with each other for scarce resources within the firm, the debate on dual innovation has begun. Most scholars believe that “exploitative innovation” is the redevelopment of existing knowledge or capabilities, and is characterized by low risk, low cost and incrementalism; “exploratory innovation” is the search for new knowledge or the development of new products, and is often characterized by high risk, high cost and disruptiveness. He and Wong (2004) [49] make a significant distinction between exploitative and exploratory innovation, arguing that exploitative innovation refers to ex ante strategic goals, while exploratory innovation refers to ex post outcomes. Exploitative innovation in ambidextrous innovation is a small-scale incremental behavior that improves short-term financial performance of a company through product optimization [34], while exploratory innovation is a large-scale R&D activity with breakthrough, which can create long-term competitive advantages for companies through the innovative development of knowledge and technology [34]. Based on ambidextrous innovation theory, exploitative innovation can reduce costs and bring expected returns in less time than exploratory innovation. The latter also requires technical expertise with strong financing constraints, which depends on internal funds of enterprises and leads to information asymmetry as well as uncertain investment returns.

As a new financial type of business model, DFI expands the scope of financial services, which could meet the financing needs (small amount, high frequency and continuous financing) of SMEs’ risky technological innovation activities [50]. With the help of digital technology, DFI can build the information process, the monitoring system and the risk control system, which are conducive to the efficient and rapid collection and mining of credit data [18]. It could then comprehensively define the credit risk level and operation conditions of enterprises. On this basis, alleviating the information asymmetry with less size discrimination, DFI reduces the difficulty and cost of financing with more allocation efficiency of credit funds, which promotes the innovation resources and capabilities of enterprises in exploitative innovation and exploratory innovation. Based on this, the following assumption is made:

Hypothesis 3 (H3).

DFI will have a positive effect on ambidextrous innovation investment of SMEs, and this effect will be stronger on exploitative innovation.

2.4. The Mediating Role of Ambidextrous Innovation

Ambidextrous innovation is manifested in the integration and creation of knowledge resources, through which DFI can realize corporation value. On the one hand, with the support of digital technologies, DFI can broaden the financing channels of enterprises with low cost, and increase investment in R&D and human capital, thereby promoting the level of exploratory innovation. Moreover, the technology and knowledge created by innovation can drive a higher level and better capability of exploratory innovation in the entire industry [51]. These factors are crucial to the sustainable competitive advantage of an enterprise in the breakthroughs of corporation value. On the other hand, DFI can point out the innovation opportunity for and direction of transforming innovation achievements to enterprises with more efficiency [52]. Therefore, DFI provides the enterprises with high-quality information tools for technology analysis [18], which helps to identify the best path of technological succession with a better matching path and price mechanism [26], and finally realizes the improvement of corporation value [14]. Accordingly, this paper proposes the following assumptions:

Hypothesis 4a (H4a).

Exploitative innovation has a mediating effect on DFI and corporation value.

Hypothesis 4b (H4b).

Exploratory innovation has a mediating effect on DFI and corporation value.

2.5. The Moderating Effect of Financial Flexibility

The definition of financial flexibility is made from the perspective of financing. Many scholars argue that financial flexibility is the ability of a firm to finance its activities at lower transaction and opportunity costs [53]. Enterprises with high financial flexibility are able to mobilize financial resources in a timely and cost-effective manner, and respond positively to the dynamic financial environment [24,25]. Graham et al. (2001) [54] argue that financial flexibility is an economic operation that allows a firm to prepare for the ability to finance its liabilities in order to improve its competitiveness in future operations. According to Gamba and Triantis (2008) [28], financial flexibility is the ability of a firm to mobilize relevant financial resources sufficiently effectively to defend itself against many unknowns faced at a later stage, thus better achieving the goal of maximizing corporation value. DeAngelo and DeAngelo (2007) [53] argue that financial flexibility is the ability to enable firms to access capital markets more easily in order to meet unexpected future events. They also define financial flexibility as the ability to make it easier for firms to access capital markets in order to meet future capital requirements arising from unexpected declines in surplus and profitable investment opportunities. Based on the literature on financial flexibility, this paper argues that financial flexibility is the ability of a firm to mobilize relevant financial resources at the desired price or at the lowest possible cost, in particular to coordinate the flow and structure of funds to maximize corporation value.

Under the development of DFI in China, the rapid changes in the capital markets and the high level of environmental uncertainty have made financial flexibility appear to play a pivotal role in corporate development. DFI is currently at a crossroads of risk and opportunity in China. While it brings opportunities for enterprises, it also presents problems such as frequent financial fraud and disclosure of user information. Coupled with the regulatory policies to be improved in China, the digital divide has led to technological financial exclusion, amplifying technological risks, which further increases the instability of financial development. However, financial flexibility management has a sensitive response to environmental uncertainty [52]. An enterprise with high financial flexibility has a stronger ability to raise and mobilize funds [23], and thus, the stronger the uncertainty environment is, the more significantly financial flexibility improves corporation value [29,53]. Based on the above analysis, this paper proposes the following hypothesis:

Hypothesis 5 (H5).

Financial flexibility can positively adjust the promoting effect of DFI on corporation value.

2.6. The Moderating Role of CSR

The impact of CSR on corporation value is highly debated, and some scholars believe that companies will waste capital and other resources with more social responsibilities, and thus create a disadvantage in competition [55]. Moreover, CSR should be limited to the laws and regulations, and the businesses run ethically without taking on additional social responsibilities [31]. However, there are still many opposite views holding that CSR is conducive to enhancing corporation value. Firstly, according to the reputation mechanism, enterprises could attract the attention of the government by fulfilling their social responsibilities actively [56], and expand the credit channels with the help of the government [31,57]. Secondly, in the case of information asymmetry, the information disclosed in a timely manner about CSR can improve the external financing ability of the company with a higher credit rating [58]. Considering the risk control, creditors will evaluate the enterprises that fulfill social responsibilities with lower risks, and issue the loans at lower interest rates, which is more conducive to reducing financing costs. Finally, based on the stakeholder theory, CSR means that the company is willing to take responsibility for its stakeholders, who are the owners of the company, and invest funds into the company with the expectation of investment returns in the future [59]. The company carries out daily business with shareholders’ funds, and has goals to create profits for shareholders. Therefore, shareholders should be the primary consideration of CSR, and if the interests of shareholders cannot be guaranteed, others’ benefits will not be guaranteed either.

When the external financing environment is not quite perfect, informal institutions largely compensate for the absence of formal institutions, and have a significant impact on the behavioral decisions of micro individuals [60]. For example, the disclosure of financial and non-financial information is considered by many scholars as a signaling behavior to improve the public recognition of the enterprise and to become a “quality” enterprise different from the general enterprise, so as to achieve the purpose of reducing information asymmetry and agency costs. In recent years, CSR disclosure has taken an important place in non-financial information in China. Companies with more social responsibility undertakings, i.e., a higher quality of information disclosure, are more willing to fulfill the information disclosure. So, in a DFI ecosystem with a greater openness of information, do the traditional ways of decreasing information asymmetry levels still play a significant role? First, companies with more social responsibility have higher ethical pursuits, and such companies place more emphasis on social welfare contributions. This is in line with the purpose of DFI, which highlights social equity, breaks the lending inertia of the traditional financial industry and benefits all segments of society in an all-around way. From this dimension, CSR has the potential to promote the maximum effectiveness of DFI. Second, CSR fulfillment is a good signal for CSR risk management [61]. This is because in the course of business operations, the various social and environmental risks can occur, to some degree, in conjunction with each other. However, CSR helps to ensure a stable contractual relationship between stakeholders, thereby reducing the likelihood of potential “accidents”. One of the purposes of developing DFI is to complement China’s credit system, which is not yet complete. In a more optimized financial ecosystem, companies with relatively higher “credit” scores have these due to their greater social responsibility. From this point of view, it is clear that there is a strong possibility that the signaling role of CSR will be further enhanced. Finally, the fulfillment of social responsibility is in line with society’s expectations of corporate behavior, as the behavior meets the needs of shareholders, creditors, governments and stakeholders to a certain extent. This reflects the high level of trust that the company itself has in the community, as well as the trust it has gained from all sectors of society. There is no doubt about the important role of trust mechanisms in the construction of the DFI system. An increase in the level of trust of micro-individuals is conducive to improving the chances of benefiting from DFI and its effectiveness. Each of these potential mechanisms may enable CSR to play a role as a moderating variable in the process by which DFI affects corporate value. In other words, there are differences in the degree of positive impact that different companies enjoy from the development of DFI, with companies that are more socially responsible being more likely to receive support from DFI in terms of increased corporate value.

Based on the above analysis, this paper proposes the following hypothesis:

Hypothesis 6 (H6).

The level of CSR has a positive moderating effect on corporation value in DFI. For enterprises with a higher level of CSR, DFI has a greater effect on corporation value.

2.7. The Regulatory Role of Product Market Competition

With the changes in market competition, enterprises usually create corresponding strategies according to the product management and target orientation, which deal with the changes in technology, customer preferences and fluctuations of product supply and demand. Fierce competition may reduce the economic return, and the survival and development of enterprises depends on the dynamic ability to integrate, cultivate and reallocate internal and external resources. Eisenhardt and Martin (2000) proposed that dynamic capabilities include two types of strategic logic: the logic of exploration and the logic of utilization. The former refers to exploratory innovation, which cultivates new knowledge to acquire new resources and skills, while the latter focuses on exploitative innovation, which refines new applications based on existing knowledge, resources and skills [62].

The contribution of DFI to ambidextrous innovation may also be influenced by the external market environment for enterprises. The degree of product market competition (hereafter, PMC) affects the demand for innovation by enterprises. If a company invests boldly into innovative projects in the face of fierce PMC, it may be able to gain the upper hand in the product market because of the new products or technologies it has acquired through its innovative research and development [63]. This state of competition will motivate other companies to actively innovate, which will ultimately contribute to the increase in productivity of society and to the improvement of the technological innovation capacity of the entire society [64]. From the perspective of enterprise resource theory, the heterogeneous resources of an enterprise, as the brand and facade of the enterprise, represent the uniqueness and irreplaceability of the enterprise [65]. By attracting customers with distinctive products, they create their own brand and reputation exclusively among customers, and achieve a leading position in the industry competition [66]. This heterogeneous resource is precisely the result of the research and development of the enterprise through R&D. Therefore, under the development of DFI in China, enterprises in more competitive product markets have a stronger need to innovate and to carry out innovative activities. That is, the enterprises facing more intense PMC help to amplify the role of DFI in promoting innovation, providing incentives for enterprises to increase their R&D investment and achieve exploitative and exploratory innovation. Conversely, DFI has a more limited effect on the innovation of enterprises in less competitive product markets. Under the evolution of DFI, the ability of enterprises to achieve exploitative and exploratory innovation also depends on the differences in the external competitive market environments in which they operate.

According to the ambidextrous innovation theory, exploitative and exploratory innovation activities are different in terms of return on investment and investment risk [66], and, therefore, the effects of PMC on both are different. In the context of fierce PMC, ambidextrous innovation enables enterprises to enter new product market areas while improving the market position of existing products. On the one hand, facing the challenges from competitors, enterprises need to carry out exploitative innovation to continuously expand existing skills and technologies, as well as bring about incremental improvements in products and processes with cost reduction and high quality in order to obtain stable customer orders. On the other hand, Popadić and Černe (2016) [67] proposed that incremental improvements brought about by exploitative innovation cannot ensure good performance, as they only focus on current customers in a turbulent market, while exploratory innovation is necessary for companies to explore new markets and customers, as well as to cultivate new capabilities, such as product update and sale. Managers are aware that PMC can lead to crises such as business bankruptcy and loss of jobs. Therefore, maintaining the balance of exploitative innovation and exploratory innovation at the same time, ambidextrous innovation enables enterprises to achieve efficiency and adaptability with continuity in order to overcome the challenges in PMC. Based on the above analysis, this paper proposes the following assumptions.

Hypothesis 7 (H7).

The more intense the market competition is, the more able the company is to promote ambidextrous innovation through DFI.

Hypothesis 7a (H7a).

The more intense the market competition is, the more able the company is to promote exploitative innovation through DFI.

Hypothesis 7b (H7b).

The more intense the market competition is, the more able the company is to promote exploratory innovation through DFI.

The conceptual model of this paper is shown in Figure 1.

Figure 1 Alt Text: The hypothesized relationship between the variables. Shows the effect of moderating and mediating effects of relevant variables in the relationship between DFI and corporation value.

Figure 1 Long Description: The development of DFI has positive effects on corporation value; ambidextrous innovation has positive effects on corporation value; financial flexibility positively moderates the relationship between DFI and corporation value; CSR positively moderates the relationship between DFI and corporation value; PMC positively moderates the relationship between DFI and ambidextrous innovation.

3. Methods

3.1. Sample Selection and Data Sources

In order to study the relationship between DFl, ambidextrous innovation and corporation value, the city-level DFI index was matched with micro-level financial indicators of SMEs. Drawing on the common practice of existing studies, the DFI was measured using the Beijing University DFI index (divided by 100). Based on the large-scale data and comprehensive evaluation conducted by DFI Research Center of Peking University and Ant Financial Services Group, this index has been widely applied by scientists from all over the world with reliable data support for DFI research. The research samples selected in this paper were from Chinese listed SMEs from 2012 to 2020, and the original data were integrated and analyzed according to the research needs. After screening, a total of 1604 listed SMEs for the period of 2012–2020 were finally obtained as regression samples. The following screening criteria were applied: (1) Financial companies were excluded. (2) Companies with missing core data were excluded. Continuous variables were Winsorized at a 1% level.

Measurement methods of the variables involved in this study are shown in

Table 1. Measurement methods of variables.

Variables	Acronyms	Calculation	References
Corporate value	TobinsQ	TobinsQ = (Equity Market Value + Net debt market value)/Total assets at period end	McConaughy et al. (1998) [68]; Kang et al. (2019) [69];
Digital finance inclusive	DFI	The Peking University Digital Financial Inclusion Index of China (PKU-DFIIC)	Yang and Zhang (2020) [2]; Yang et al. (2022) [70]; Sun and Tang (2022) [71]; Lee and Wang (2022) [72]; Li et al. (2022) [73]
Exploitative innovation	Incremental	Capitalized expenditure on R&D investments/Operating revenue	Kang et al. (2019) [69]; Liu and Han (2019) [74]
Exploratory innovation	Radical	Expensed expenditure on R&D investments/Operating revenue
Financial flexibility	FF	Financial Flexibility = (Corporate cash holding ratio-Industry average cash holding ratio) +Max (Industry average gearing ratio-Corporate gearing ratio, 0)	DeAngelo, H., and DeAngelo, L.E. (2007) [53]
Corporate social responsibility	CSR	Hexun Social Responsibility Report Professional Review	Wang et al. (2021) [75]
Product market competition	PMC	Product market competition = (Total profit + total tax + total interest)/Average total assets	Peress J. (2010) [76]; Kale and Loon (2010) [77]; Wang and Zhou (2020) [78]
Enterprise size	Size	Ln (Assets at end of period + 1)	Chen et al. (2021) [22]
Enterprise age	Age	Year of statistics–Year of Establishment	Li et al. (2016) [41]
Enterprise growth	Gro	(Total operating income for the period–Total operating income for the previous period)/Total operating income for the previous period	Cheung (2016) [57]
Leverage	Lev	Total liabilities/Total Assets	Dhaliwal et al. (2011) [31]
Cash ratio	CR	Cash and cash equivalents/Current liabilities	Yang and Zhang (2020) [2]
Size of board of directors	Bod	Number of Board of Directors	Chen et al. (2021) [22]
Industry	Industry	Industry dummy variable
Year	Year	Year dummy variable

.

3.2. Model Setting and Variable Definition

Firstly, this paper studies the impact of DFI on corporation value in a fixed-effect model, and the benchmark econometric regression model is expressed as (1):

TobinsQ_it = α₀ + α₁DFI_it + α₂Controls + ΣYear + ΣIndustry + ε_it

(1)

This paper uses Tobin’s Q as a metric of corporation value; DFI_it measures the development of DFI in the city where the enterprise is located; α0 represents the constant term. Controls represent the control variables that may have impacts, which include six variables: enterprise size, enterprise age, enterprise growth, leverage, cash ratio and size of board of directors. Enterprise size is the natural logarithm of total assets; enterprise age is the number of years since the establishment of the enterprise. To control the growth rate of the company, enterprise growth, which represents the growth rate of sales, was used. Leverage is the total debt divided by total assets; cash ratio was measured by the ratio of cash and equivalent to current liabilities; size of board of directors was measured by number of board of directors.

Year represents the year dummy variable; Industry represents the industry dummy variable; i and t represent the province and year, respectively; and ε_it is the random disturbance term. The measurements of specific variables are shown in Table 1.

Secondly, the impact of DFI on ambidextrous innovation in enterprises was further examined, and (2) and (3) are the specific models. Referring to the practice of Kang et al. (2019) [69], R&D activities are divided into two stages: research and development. Thus, the investment in exploitative innovation is measured by the expenditure of R&D activities, and that in exploratory innovation is measured by the capitalized expenditure.

TobinsQ_it = α₀ + α₁Incremental + α₂Controls + ΣYear + ΣIndustry + ε_i

(2)

TobinsQ_it = α₀ + α₁Radical + α₂Controls + ΣYear + ΣIndustry + ε_it

(3)

In (2) and (3), Incremental represents the level of exploitative innovation in the enterprise, Radical represents the level of exploratory innovation in the enterprise, and the definitions of other variables are consistent with (1).

(4) and (5) mainly verify Hypothesis 3, using different innovation behaviors as dependent variables to reflect the impact of DFI on ambidextrous innovation:

Incremental = β₀ + β₁DFI_it + β₂Controls + ΣYear + ΣIndustry + ε_it

(4)

Radical = β₀ + β₁DFI_it + β₂Controls + ΣYear + ΣIndustry + ε_it

(5)

(4) and (5) examine the impact of DFI on ambidextrous innovation in enterprise, but this is not enough to clarify the specific mechanism by which DFI affects corporation value. In order to verify whether ambidextrous innovation is an internal mechanism for DFI to affect corporation value or not, the paper uses the mediation effect test procedure of Wen et al. (2014) [79] to distinguish between a partial and full mediation effect, and ensures that the first and second types of errors are at a low level. The specific models are as follows:

TobinsQ_it = γ_0,1 + γ_1,1DFI_it + γ_2,1Incremental + γ_3,1Controls + ΣYear + ΣIndustry + ε_it

(6)

TobinsQ_it = γ_0,2 + γ_1,2DFI_it + γ_2,2Radical + γ_3,2Controls + ΣYear + ΣIndustry + ε_it

(7)

Among the models, the coefficient α₁ represents the size of the effect; the coefficient β₁ in (4) and (5) reflects the impact of DFI on the intermediary variables, that is, exploitative innovation or exploratory innovation. The coefficient γ₁ in (6) and (7) reflects the direct effect of DFI on corporation value.

In addition, there are other factors in the enterprise, such as financial flexibility, CSR, etc. Do these contribute to the above differences in some degree? That is, do they play a moderating role in the impact of DFI on corporation value? To answer the above questions, financial flexibility, CSR and their interaction with DFI were added to (1) for testing. The specific models are as follows:

TobinsQ_it = α₀ + α₁DFI_it + α₂FF + α₃FF×DFI_it + α₄Controls + ΣYear + ΣIndustry + ε_it

(8)

TobinsQ_it = α₀ + α₁DFI_it + α₂CSR + α₃CSR×DFI_it + α₄Controls + ΣYear + ΣIndustry + ε_it

(9)

Since the corporate equity financing is subject to strict supervision in China, resulting in low financial flexibility, financial flexibility is thus measured by cash flexibility and debt financing flexibility according to the practice of DeAngelo and DeAngelo (2007) [53]. CSR is measured by the evaluation data of CSR from Hexun.com, which objectively reflects the degree of CSR in China [80]. As an important basis for domestic scholars to measure the behavior of CSR, the score is authoritative and reasonable, and the higher the score is, the greater the CSR it represents. The definitions of the other variables are the same as above.

In addition, in order to ascertain the moderating role of market competition in the impact of DFI on ambidextrous innovation, the market competition and its interaction with DFI were added to (4) and (5) to attempt to reflect the possible moderating effect. For the calculation of PMC, the Herfindahl index is currently used mainly to measure the competitiveness of a company’s products, but Chinese companies have a wide range of products in circulation and are ot significantly affected by regional restrictions, so the Herfindahl Index is ineffective to truly measure the competitiveness of products in the actual market in China. Therefore, drawing on the existing literature [76,77], this paper uses the Lerner Index (PMC) to measure the position of firms in product markets competition. PMC is calculated as follow:

(Total profit + total tax + total interest)/Average total assets

The higher the PMC is, the higher the competitive position of the corporation in the industry and the stronger the pricing ability. The definitions of other variables are the same as above. The specific models are as follows:

Incremental = δ_0,1 + δ_1,1DFI_it + δ_2,1PMC + δ_3,1Controls + ΣYear + ΣIndustry + ε_it

(10)

Radical = δ_0,2 + δ_1,2DFI_it + δ_2,2PMC + δ_3,2Controls + ΣYear + ΣIndustry + ε_it

(11)

Incremental = ξ_0,1 + ξ_1,1DFI_it + ξ_2,1PMC + ξ_3,1PMC×DFI_it + ξ_4,1Controls + ΣYear + ΣIndustry + ε_it

(12)

Radical = ξ_0,2 + ξ_1,2DFI_it + ξ_2,2PMC + ξ_3,2PMC×DFI_it + ξ_4,2Controls + ΣYear + ΣIndustry + ε_it

(13)

3.3. Descriptive Analyses

In this study, Stata 15.0 was used to provide descriptive statistics for each variable, and the results are shown in

Table 2. Descriptive statistical results of each variable.

Variable	Mean	Standard Deviation	Min	Max
TobinsQ	1.412	1.535	0.811	6.542
DFI	291.062	80.672	75.870	431.928
Incremental	0.124	1.485	0.000	4.865
Radical	0.076	1.504	0.015	1.000
FF	0.124	0.215	−0.279	0.982
CSR	20.675	10.767	−18.450	85.240
PMC	0.158	0.179	−2.910	4.730
Size	21.400	0.838	19.290	25.777
Age	15.484	4.886	3.000	25.000
Gro	0.505	0.365	−0.683	1.294
Lev	0.315	0.183	0.011	1.687
CR	1.589	3.641	0.001	2.931
Bod	7.944	1.432	4.000	15.000

.

Correspondingly, the mean values of Incremental and Radical in the sample period were 0.124 and 0.076, respectively, indicating that most companies are more inclined to exploitative innovation at present. The mean and standard deviation of the variable TobinsQ were 1.412 and 1.535, respectively, indicating that the market value of the sample companies was relatively scattered, and the market valuations of different companies are quite different. Secondly, there were obvious differences in the development of DFI in various cities in China, as shown in Table 2. From the data, DFI is characterized by polarization in China, and the level of DFI in various regions is quite different, which is consistent with the existing research results. Overall, the average value of DFI is on the rise, indicating that the application of DFI is becoming more and more extensive.

4. Empirical Test and Results Analysis

4.1. Test of the Impact of DFI on Corporation Value

(1) was used to estimate the impact of DFI on corporation value, which was tested by the Hausman, and the corresponding p value was 0.000. Thus, a fixed-effect model was selected to study the impact of DFI on corporate financing constraints. As shown in

Table 3. The influence of digital financial inclusion on enterprise value.

	(1)	(2)	(3)	(4)	(5)
Variable	TobinsQ			Incremental (1 Period Lagged)	Radical (1 Period Lagged)	VIF	Radical (2 Period Lagged)
DFI	0.005 **			0.614 *	0.242	1.260	0.360 ***
	(0.002)			(0.338)	(0.235)		(0.134)
Incremental		2.997 **
		(1.382)
Radical
			(1.429)
	−0.250 ***	−0.173 **	−0.968 ***	28.658 ***	−9.271 **	1.160	−5.174
	(0.072)	(0.072)	(0.091)	(9.334)	(3.639)		(4.531)
Age	−0.111	0.021	−0.064 ***	−23.919 **	−9.715 **	1.390	−8.871 **
	(0.072)	(0.016)	(0.021)	(10.041)	(4.014)		(4.499)
Gro	0.094 ***	0.074 ***	0.251 ***	4.532	−0.002	1.050	−0.648
	(0.024)	(0.024)	(0.057)	(2.916)	(0.058)		(1.382)
Bod	−0.036	−0.054 *	−0.044	6.988 *	−4.373 **	1.120	−4.166 **
	(0.033)	(0.032)	(0.040)	(4.246)	(1.700)		(1.927)
CR	−0.064 ***	−0.067 ***	−0.062 ***	4.392 ***	0.348	1.320	0.613
	(0.009)	(0.008)	(0.010)	(1.196)	(0.398)		(0.414)
Lev	−0.798 ***	−0.447	−0.222	11.053	4.113	1.180	20.125
	(0.289)	(0.284)	(0.351)	(36.756)	(14.928)		(17.711)
Industry	Yes	Yes	Yes	Yes	Yes		Yes
Year	Yes	Yes	Yes	Yes	Yes		Yes
Constant	9.007 ***	6.755 ***	25.387 ***	−426.167 **	213.778 ***		232.461 ***
	(1.453)	(1.364)	(1.718)	(185.069)	(73.476)		(85.584)

Note: standard error in brackets, * p < 0.1, ** p < 0.05, *** p < 0.001.

, DFI was positively correlated with corporation value, which indicated that the development of DFI could promote the appreciation of corporation value. Thus, Hypothesis H1 is verified.

(2) verified the impact of ambidextrous innovation on corporation value. According to the estimation results in Table 3, the estimated coefficients of the variables Incremental and Radical were both positive and significant, indicating that both ambidextrous innovation and exploratory innovation were significantly positively related to corporation value; that is, ambidextrous innovation can significantly improve corporation value. Therefore, H2a and H2b were verified, and H2 was verified as a whole.

(4) and (5) verified the impact of DFI on ambidextrous innovation. R&D investment cannot be immediately transformed into innovation results in the current period, and mainly relies on the accumulation and transformation of previous results. Thus, there is a certain time lag in corporate innovation, and all explanatory variables are lagged by one period. In Table 3, DFI and exploitative innovation are shown to have a significant positive relationship at the 10% level, indicating that DFI has a significant positive impact on the level of exploitative innovation in enterprises.

The results in Table 3 show that the coefficient between DFI and exploratory innovation was positive, but it was not significant. In order to exclude the possibility of multi-collinearity, the paper used the VIF test, and the test results are shown in Table 3. The VIF value of the variance expansion coefficient was between 1.050 and 1.390, indicating that there was no multi-collinearity. Exploratory innovation is a fundamental and radical innovation activity on a large scale, which aims to explore new knowledge and technologies and often focuses on emerging and unmet needs in the market. Moreover, compared with exploitative innovation, exploratory innovation requires more pre-knowledge accumulation and slower transformation [75]. It is also a complex process of interaction between endogenous factors and exogenous factors [81], and, thus, may be influenced by some resource constraints of enterprises, such as knowledge composition, capability structure, etc. [82]. In light of the above situation, the explanatory variables in (5) were estimated with a lag of two periods. In Table 3, the estimated coefficient of DFI was 0.360, which passed the significance test, indicating that DFI had a significant impact on exploratory innovation with a certain time lag. The coefficient between DFI and exploitative innovation was higher than that with exploratory innovation, in both significance and coefficient size. Therefore, DFI will have a certain promoting effect on the investment of ambidextrous innovation in enterprises, especially on exploitative innovation. Thus, Hypothesis H3 was verified.

4.2. Test in the Mediating Role of Ambidextrous Innovation

As the results of (1) in Table 3 show, DFI had a significant positive impact on corporation value (α₁ = 0.005, p < 0.05), and the results of (4) show that DFI also had a significant positive effect on exploitative innovation (β_1,1 = 0.614, p < 0.1). As shown in (6) in

Table 4. The intermediary effect of dual innovation.

	(6)	(7)
Variable	TobinsQ
Incremental/Radical	0.001 **	0.000 **
	(0.000)	(0.000)
DFI	0.016 ***	0.013 ***
	(0.003)	(0.003)
L.Size	−0.741 ***	−0.733 ***
	(0.086)	(0.085)
L.Age	0.448 ***	−0.400 ***
	(0.089)	(0.089)
L.Gro	0.083 ***	0.053 **
	(0.027)	(0.026)
L.Lev	−0.030	−0.021
	(0.038)	(0.037)
L.CR	0.482	0.590 *
	(0.336)	(0.333)
Industry	Yes	Yes
Year	Yes	Yes
Constant	16.676 ***	21.147 ***
	(1.699)	(1.673)
Sobel	α1, β1,1 and γ1,1 are significantly	Z = 2.779 > 0.97

Note: standard error in brackets, * p < 0.1, ** p < 0.05, *** p < 0.001.

, the correlation between corporation value and the variable Incremental was significantly positive (γ_2,1 = 0.001, p < 0.05), and the correlation between corporation value and DFI was significantly positive (γ_1,1 = 0.168, p < 0.05), so exploratory innovation was a partial mediator of the relationship between DFI and corporation value. Thus, Hypothesis H4a was verified.

Similarly, a regression test was conducted on exploratory innovation, and the regression results showed that DFI had a significant positive impact on corporation value (α₁ = 0.005, p < 0.05). There was a significant positive correlation between corporation value and variable radical (γ_2,2 = 0.000, p < 0.05), while the positive correlation was not obvious between DFI and exploratory innovation (β_1,2 = 0.242, p > 0.1).

In order to reduce the probability of error in the second type, the Sobel test was performed on the samples. From Table 4, the test results showed significantly that exploratory innovation was a partial mediator of the relationship between DFI and corporation value. Thus, Hypothesis H4b was verified.

4.3. Test of the Regulatory Role in Financial Flexibility and CSR

According to the financial flexibility of different enterprises, the paper tests the different impacts of DFI on corporation value. From

Table 5. The moderating effects of financial flexibility and corporate social responsibility.

	(8)	(9)	(9)
Variable	TobinsQ		TobinsQ (1 Period Lagged)
DFI	0.001 *	0.003 ***	0.002 ***
	(0.001)	(0.001)	(0.001)
CSR		−0.029 ***	0.026 ***
		(0.010)	(0.004)
CSR×DFI		−0.000 ***	0.000 ***
		(0.000)	(0.000)
FF	3.899 ***
	(0.607)
FF×DFI	0.015 ***
	(0.002)
Size	−0.528 ***	−0.673 ***	−0.746 ***
	(0.047)	(0.046)	(0.051)
Age	−0.029 ***	−0.023 ***	−0.016 *
	(0.008)	(0.008)	(0.008)
Gro	0.065 **	0.067 **	0.044
	(0.029)	(0.029)	(0.031)
Bod	−0.053 **	−0.063 ***	−0.079 ***
	(0.025)	(0.024)	(0.026)
CR	−1.918 ***	−1.426 ***	−1.008 ***
	(0.288)	(0.232)	(0.257)
Lev	0.009	0.000	0.040 ***
	(0.012)	(0.010)	(0.010)
Industry	Yes	Yes	Yes
Year	Yes	Yes	Yes
Constant	16.405 ***	19.184 ***	20.075 ***
	(0.966)	(0.959)	(1.047)

Note: standard error in brackets, * p < 0.1, ** p < 0.05, *** p < 0.001.

, the coefficients of financial flexibility, as well as the interaction term (FF×DFI) of DFI and financial flexibility, were all significantly positive at the 1% level, indicating that financial flexibility positively moderates the relationship between DFI and corporation value. Thus, Hypothesis H5 was verified.

It can be seen from Table 5 that the coefficient of interaction term (CSR×DFI) between the DFI and CSR was significantly negative at the 1% level, which means that in the short term, there is an inverse relationship between CSR and corporation value. This may be due to the fact that it takes some time for CSR to contribute to DFI and corporate value. Additionally, in the short term, CSR not only fails to deliver timely economic benefits for enterprises, but also has the potential to deplete corporate resources, thus playing a negative role in DFI and corporate value. Specifically, in the short term, CSR consumes the resources of shareholders, although the objective is to maintain and manage relationships with stakeholders, and to maximize their benefits (i.e., investors, suppliers, customers, employees, communities, government) [83]. This reflects the agency problems of the company’s management, and, therefore, may have a negative value correlation in the short term [84]. As shown in Figure 2, the mismatch between costs and benefits has led to shareholders viewing CSR investment unfavorably in the short term, seeing it as a drain on resources. Therefore, CSR may theoretically reduce shareholder wealth in the short term.

The coefficient of the interaction term (L.CSR×DFI) with a lag period changes from negative to positive, and is significant at the 1% level, indicating that the impact of CSR on corporation value has a lagging effect. In the long run, corporations that take on more social responsibility are better able to benefit from the growth of DFI. This is due to the fact that in the long term, the objective of a company in the traditional sense is to maximize the interests of shareholders. Then, CSR can be seen as a long-term investment. CSR generates transactional, financial and ethical reputational capital to corporate stakeholders that not only sustains long-term value, but also generates long-term value.

4.4. Test of the Regulatory Role of Product Market Competition

The results of (10) in

Table 6. The moderating effects of product market competition.

	(10)	(11)	(12)	(13)
Variable	Incremental	Radical	Incremental	Radical
DFI	1.595 ***	0.249 **	1.577 ***	0.189 *
	(0.273)	(0.113)	(0.273)	(0.115)
PMC	148.134 ***	−51.802	125.303 ***	96.628 ***
	(27.865)	(48.969)	(28.556)	(32.965)
PMC×DFI			46.278 ***	0.328 ***
			(8.981)	(0.101)
Size	−40.328 ***	−11.040 ***	−52.984 ***	−10.180 ***
	(8.863)	(3.598)	(8.450)	(3.605)
Age	−51.599 ***	9.504 ***	−20.044 ***	9.060 ***
	(8.452)	(3.503)	(5.686)	(3.503)
Gro	5.050 *	0.478	3.410	0.134
	(2.854)	(1.167)	(3.907)	(0.902)
Bod	3.078	3.164 **	−25.567	3.154 *
	(3.910)	(1.611)	(34.854)	(1.609)
CR	−36.452	14.081	−0.537	6.776
	(34.800)	(14.163)	(1.014)	(14.322)
Lev	−0.042	−0.026	1.577 ***	−0.108
	(1.010)	(0.420)	(0.273)	(0.420)
Industry	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
Constant	−434.489 **	256.031 ***	−536.664 ***	0.122
	(176.928)	(71.987)	(178.713)	(0.434)

Note: standard error in brackets, * p < 0.1, ** p < 0.05, *** p < 0.001.

show that the correlation between variable increment and PMC was significantly positive (δ_2,1 = 148.134, p < 0.001), and the results of (12) show that the correlation between variable Incremental and PMC was significant (δ_2,1 = 125.303, p < 0.001). The coefficient of the interaction term between PMC and DFI (PMC×DFI) was significantly positive (ξ_3,1 = 46.278, p < 0.001), demonstrating that PMC positively moderates the relationship between DFI and exploitative innovation. Therefore, with the intensification of PMC, DFI will further promote the behavior of exploitative innovation in the enterprise. Thus, Hypothesis H7a was verified. The data are consistent with the results of related research: with the intensification of PMC, if the product is not actively improved, it will be difficult for enterprises to ensure stable revenue [85], while exploitative innovation can effectively reduce costs and increase market share by introducing new products [82].

In Table 6, the results of (11) and(13) show that PMC did not positively moderate the relationship between digital financial inclusion and exploratory innovation, and thus, Hypothesis H7b was not verified. The possible explanation for this is that, with the intensification of PMC, it is difficult for enterprises to maintain profit growth; thus, they are more concerned about the current economic benefits. In addition, intense competition reduces resources for exploratory innovation, which is a costly and risky type of innovation that has a significant impact on firm profitability and viability [75,81]. However, at this time, companies that focus on exploitative innovation can maintain profitability in a competitive environment by improving current products for existing customers.

4.5. Robustness Test

4.5.1. Replacement of Key Variables

In order to increase the reliability of empirical results, the key variables are replaced for robustness tests. Drawing on the method of Li et al. (2020) [86] and Wu and Zhang (2021) [87] different types of patent applications were used to measure different innovation methods: exploratory innovation was measured by the number of invention patents (R) applied for by enterprises in the year, and exploitative innovation was measured by the number of utility model and appearance patents (D). Then, we added one to the data respectively, and we took the logarithm. Return on equity (ROE) represents corporation value. After the variables were substituted, the sample size of enterprises was reduced, but there was no significant difference as a whole obtained through testing. Hypotheses H1-H6 were tested again with new variables, and the results are shown in

Table 7. Robustness test results for replaced key variables.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
Variable	ROE	ROE	ROE	D	R	ROE	ROE	ROE	ROE	ROE (1 Period Lagged
D		0.210 **				0.002 **
		(0.032)				(0.001)
R							0.002 **
			0.102 ***				(0.001)
DFI	0.000 *		(0.001)	0.014 **	0.024 ***	0.002 **	0.002 **	0.001 ***	0.002 ***	0.001 **
	(0.000)			(0.006)	(0.005)	(0.001)	(0.001)	(0.000)	(0.000)	(0.001)
CSR									−0.017 ***	0.015 ***
									(0.003)	(0.001)
CSR×DFI									0.000 ***	0.000 *
									(0.000)	(0.000)
FF								0.105 ***
								(0.009)
FF×DFI								0.002 ***
								(0.001)
Size	0.078 ***	−0.004	−0.298 ***	0.094	0.009	−0.305 ***	−0.305 ***	0.054 ***	0.046 ***	0.036 *
	(0.013)	(0.005)	(0.029)	(0.517)	(0.492)	(0.029)	(0.029)	(0.013)	(0.013)	(0.020)
Age	−0.001	0.028	−0.016 ***	−0.021	0.056	−0.016 ***	−0.016 ***	−0.000	−0.000	0.001
	(0.002)	(0.020)	(0.005)	(0.086)	(0.082)	(0.005)	(0.005)	(0.002)	(0.002)	(0.003)
Gro	0.026 ***	0.051 ***	−0.002	−0.083	−0.271	0.032 **	0.032 **	0.025 ***	0.020 **	−0.003
	(0.007)	(0.017)	(0.018)	(0.279)	(0.309)	(0.016)	(0.016)	(0.008)	(0.008)	(0.012)
Bod	0.004	0.016 **	−0.020	0.136	0.093	−0.017	−0.017	−0.000	0.003	0.003
	(0.007)	(0.007)	(0.015)	(0.270)	(0.257)	(0.015)	(0.015)	(0.007)	(0.007)	(0.010)
CR	0.012 ***	1.378 ***	0.004	0.064	−0.001	0.004	0.004	0.010 ***	−0.010 ***	0.001
	(0.003)	(0.155)	(0.007)	(0.116)	(0.111)	(0.007)	(0.007)	(0.003)	(0.003)	(0.004)
Lev	−1.077 ***	−0.003 ***	−1.384 ***	0.644	−4.383 *	−1.374 ***	−1.383 ***	−0.841 ***	−0.770 ***	0.076
	(0.064)	(0.001)	(0.141)	(2.509)	(2.385)	(0.141)	(0.141)	(0.066)	(0.066)	(0.099)
Industry	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Constant	−1.240 ***	6.820 ***	8.652 ***	5.011	2.123	8.205 ***	8.209 ***	−0.748 ***	−0.294	0.719
	(0.270)	(0.652)	(0.605)	(10.514)	(10.003)	(0.634)	(0.634)	(0.269)	(0.292)	(0.445)

Note: standard error in brackets, * p < 0.1, ** p < 0.05, *** p < 0.001.

. In (1), the coefficient of the impact of DFI on corporation value was 0.002, which is significant at the 5% level, indicating that with the development of DFI, its role in promoting corporation value will become more and more important. The coefficients of ambidextrous innovation in (2) and (3) were all positive and significant at the 1% level and 5% level, respectively, which indicates that ambidextrous innovation can significantly enhance corporation value. In (4) and (5), the coefficients of DFI on exploitative innovation and exploratory innovation were 0.014 and 0.024, which are significant at the 5% level and 1% level, respectively. The positive and negative coefficient, coefficient size and significance level are consistent with the above regression results, indicating that the results are robust. Therefore, Hypothesis H1 and H2 were further verified.

In (6) and (7), the positive and negative coefficient of exploitative innovation and exploratory innovation were consistent with the above results, as well as the significance level. The coefficients of DFI on corporation value with ambidextrous innovation were all significantly positive, indicating the existence of a mediating effect. The coefficients of the interaction terms in (8) and (9) were both significantly positive at the level of 1%; that is, both CSR and financial flexibility have a positive moderating effect on the relationship between DFI and hybrid innovation. Hypotheses H5 and H6 demonstrate the robustness of the results.

4.5.2. Endogeneity

The research on the impact of DFI on corporation value may be interfered with by endogenous problems. Although the main form of implementation of digital finance is online, the extent of its development is still influenced by geographical and spatial factors, and in China, the degree of development of digital finance is characterized by a greater difficulty of diffusion further away from Hangzhou. This paper used the relatively exogenous variable, the air route distance from Hangzhou to the selected city, as an instrumental variable for the DFI index, and retested the model using 2SLS. The distance to Hangzhou is closely related to the development level of DFI in sample city, and the variables related to enterprise operation are controlled. Since the distance to Hangzhou has no direct impact path on business operation, it can be used as a suitable instrument variable for DFI.

The results in columns 1 and 2 of

Table 8. Regression of instrumental variables.

	Phase I Returns		Phase II Returns
Variable	DFI	DFI	ROE	ROE
distance	−0.018 ***	−0.019 ***
	(0.001)	(0.001)
DFI			0.002 **	0.002 **
			(0.001)	(0.001)
Age		3.811 ***		−0.305 ***
		(0.543)		(0.029)
Gro		0.166 *		−0.016 ***
		(0.091)		(0.005)
Bod		−0.348		0.032 **
		(0.291)		(0.016)
Size		−1.531 ***		−0.017
		(0.279)		(0.015)
CR		0.259 **		0.004
		(0.120)		(0.007)
Lev		−7.121 ***		−1.383 ***
		(2.634)		(0.141)
Constant	3.049 ***	2.354 ***	0.953 ***	8.214 ***
	(0.097)	(0.285)	(0.218)	(0.634)

Note: standard error in brackets, * p < 0.1, ** p < 0.05, *** p < 0.001.

show that the coefficient of instrumental variables was significant below the 1% level, meeting the correlation requirements, and the F statistics were all greater than 700, which excludes the problem of weak instrumental variables. The results in column 3 and 4 show that the core explanatory variables were significantly positive at the 1% level. Therefore, the conclusion that the development of DFI has a positive effect on corporation value is robust.

5. Conclusions and Discussion

Nowadays, DFI is widely given importance in the international community. In the last decade in China, the views of governments and companies on DFI have both changed. This article analyzes the mediating effect of ambidextrous innovation with regard to the impact of DFI on the value of SMEs. Based on sample data of listed Chinese SMEs from 2012 to 2020, it introduces other internal and external factors, such as CSR, financial flexibility and PMC, for further analysis. The study found:

(1)

DFI has a significant positive impact on corporation value and ambidextrous innovation. Present researchers believe that DFI can promote corporate innovation, but have not explored the differences in the impact of DFI on different innovation methods. It has been found that DFI plays a significant role in promoting exploitative innovation, but there is a time lag in the role of exploratory innovation, and its promoting effect on exploratory innovation is weaker than that on exploitative innovation.

(2)

By examining the mediation effect test, we find that ambidextrous innovation partially mediates the relationship between DFI and corporation value; that is, DFI promotes the development of corporation value by affecting the ambidextrous innovation.

(3)

Inside of the enterprise, financial flexibility positively moderates the relationship between DFI and exploitative innovation. In the external environment of the corporation, CSR negatively moderates the relationship between digital inclusive finance and corporate value in the short term. However, in the long term, it does contribute to the growth of corporate value. This indicates that corporations need to strengthen financial flexibility internally, and improve CSR externally, in order to leverage DFI to create and deliver corporation value.

(4)

The more intense PMC is, the stronger the ability of enterprises to promote exploitative innovation through DFI becomes. However, product market competition does not have a significant impact on DFI with regard to exploratory innovation, because exploratory innovation will lead to more serious information asymmetry compared with exploitative innovation, which requires higher financing constraints and adjustment costs.

Based on the conclusions of the study, the following recommendations can be made:

Firstly, as China’s economy moves from a phase of high speed to a phase of high-quality development, Chinese companies urgently need to master core technologies and achieve a leap in the value chain. This makes it crucial to examine how to promote innovation, especially exploratory innovation, in Chinese enterprises. The findings of this paper provide new empirical evidence that DFI enhances the value of SMEs by promoting ambidextrous innovation. Therefore, the government should further strengthen the development of DFI and effectively regulate it in order to create a healthy and ordered financing environment for enterprises. This is necessary to help enterprises to better and more quickly master the core technology with independent intellectual property rights, as well as to take the lead in industrial development.

Secondly, SMEs should improve the role of ambidextrous innovation between DFI and corporation value. Attention should be paid to the matching selection of ambidextrous innovation, since the balance effect of ambidextrous innovation can prevent SMEs from innovation traps, so as to reduce business risks and improve the value of SMEs. In addition, the innovation methods should also be different when an enterprise faces different internal and external environments. For example, a company with a turbulent internal and external environment needs stable short-term financial performance, and it may focus on exploitative innovation at this time. While a company in a stable internal and external environment already has a certain economic foundation and resource advantage, it could increase the investment on exploratory innovation to enhance the sustainable competitive advantage. Therefore, Chinese SMEs should respond to the needs of China’s high-quality economic development and allocate innovation resources appropriately.

Thirdly, the dividends induced by DFI are influenced by the degree of PMC. The SMEs with higher degree of PMC are more able to exploit the role of DFI in enhancing exploitative innovation, while DFI has a more limited role in enhancing innovation in SMEs with a lower degree of PMC. Therefore, in order to better enhance the effectiveness of DFI in promoting innovation in SMEs, the government can adopt a phased, focused and differentiated policy of supporting innovation for SMEs with different innovation needs.

Fourthly, the government needs to guide SMEs to allocate resources rationally, and actively undertake social responsibilities. At this stage, CSR has attracted more and more attention, and even becomes an advantage in corporate competition. SMEs should allocate resources rationally, and evaluate innovation investment carefully, before they have a stable internal cash flow. From a long-term development point of view, companies should take social responsibility, as this contributes to corporate value.

Finally, relevant institutions need to improve the supervision system of DFI, since technological development is a double-edged sword. DFI not only brings vitality to economic development, but also significant risks in information security, agency and supervision, all of which may have an impact on market order. Therefore, while strengthening the publicity of DFI, it is necessary to improve its supervision system and promote the healthy and sustainable development of DFI.

There are still imperfections in the research in this paper.

First, this paper has not yet examined the pathways of the influence of DFI on ambidextrous innovation. The results of the existing research suggest that DFI can promote enterprise innovation by alleviating financing constraints, reducing agency costs, influencing sales and management activities, etc. Therefore, it is important to explore the mechanism of action of DFI in influencing ambidextrous innovation in future studies.

Secondly, in measuring the DFI of the region where the listed enterprises are located, taking into account the attributes of the Internet platform and the spillover effect of information technology, this paper adopts provincial statistics to measure DFI index in the province where the listed enterprises have their registered addresses, instead of using the DFI development indicators of the cities where the enterprises are located. In reality, the provincial DFI Index gives an approximate overview of the development of DFI in a region. Even if spillover effects exist, the development of DFI can vary considerably between cities within the same province, taking into account the realities of infrastructure development, staffing, regional development, etc. The provincial DFI index, as a mean figure, does not necessarily directly reflect the development of DFI in the city where the company is located, and may, therefore, have limitations. In future works, it is hoped that the DFI will be broken down to the city where the business is located, taking into account the reality of the situation and identifying the appropriate indicators to measure DFI.