Digital Finance, Financing Constraints, and Green Innovation in Chinese Firms: The Roles of Management Power and CSR

Abstract

With the increasing global emphasis on sustainable development goals, and in the context of pursuing high-quality sustainable development of the economy and enterprises, this study empirically examines the effect of digital finance on corporate financing constraints and the impact on corporate green innovation with a sample of China’s A-share-listed companies in the period of 2011–2020 and explores the issue from the perspectives of management power and corporate social responsibility (CSR) at the micro level of enterprises. The empirical results show that digital finance can indeed alleviate corporate financing constraints. Still, the synergistic effect of the two on corporate green innovation produces a “quantitative and qualitative separation” effect, which only promotes the enhancement of iconic green innovation, and the effect on substantive green innovation is not obvious. The power of management and CSR performanceshave different moderating roles in the alleviation of financing constraints by the empowerment of digital finance. Management power and corporate social responsibility have different moderating effects on digital financial empowerment to alleviate financing constraints. The findings of this study enrich the research in related fields and provide more basis for the promotion of digital financial policies and more solutions for the high-quality development of enterprises.

1. Introduction

As the world places increasing emphasis on sustainable development goals, China is actively promoting “dual-carbon” initiatives, the Chinese economy is entering a new phase of high-quality development, and traditional industries are facing urgent transformation and upgrading requirements. Within the framework of China’s pursuit of a green, low-carbon economic trajectory, the significance of green practices becomes paramount. Green innovation is a key method to revitalise the ecosystem and provide an indispensable technological foundation for sustainable economic development [1]. Enterprises, as well as market participants, need to continuously increase their investment in capital, technology, and talent, which will cause enterprises to continuously grow their financing needs; therefore, alleviating enterprise financing constraints, improving financing efficiency, and broadening the financing channels have a key and bridging role in the sustainable development of the real economy nowadays [2]. Exploring the triggers of enterprise financing constraints in China, it can be understood that information asymmetry, property rights discrimination, and other issues are obstacles to enterprise financing [3]. The cost of capital, scale effects, regional economic effects, and external macroeconomic policies also affect enterprise financing constraints to a certain extent [4].

In recent years, internet information technology, represented by big data, blockchain, cloud computing, and artificial intelligence, has gradually been applied to the traditional financial industry. The in-depth application of digital technology can help realise the interconnection of everything and the interoperability of information, solve the problem of the complex flow of information between market players, and optimise information value [5]. Therefore, digital finance provides a strong foundation for alleviating information asymmetry, innovating financing modes, and easing corporate financing constraints [6]. However, digital finance essentially belongs to the finance category, with the characteristic attributes of financial risk, such as contagiousness, uncertainty, relevance, and other basic features; digital finance has not changed the above risk characteristics and even has the possibility of exacerbation. Digital finance in the development process will likely compound these risk types [7]. It has also been shown that digital financial inclusion can address the financing constraints of micro and small enterprises. However, in a sample of listed companies, it could not play a similar role [8]. Moreover, since regulation always follows behaviour, at this stage, the regulatory system and tools that match the rapid development of big data have not yet been fully perfected. Some studies have found that the development of digital finance negatively affects bank liquidity creation, and that bank risk-taking mediates this relationship [9]. Therefore, is digital finance capable of alleviating the financing difficulties faced by enterprises in their transformation and upgrading and thus helping to enhance their ability in the green direction and ultimately assisting them in high-quality and sustainable development? The above questions are worth exploring further.

Based on this, this paper selects A-share, non-financial listed companies from 2011 to 2020 as research samples, empirically examines whether digital finance can alleviate corporate financing constraints and help the green development of the enterprises, and, based on the management’s power and the perspective of corporate social responsibility, examines the differential impacts of digital finance on corporate financing constraints and the economic consequences of the study in different contexts.

The contributions of this paper lie in the following: First, existing studies discussing the impact of digital finance at the firm level have addressed aspects such as SME leverage [10], corporate ESG performance [11,12,13], corporate insolvency risk [14], corporate financing [15], and there is an increasing focus on the sustainable development of enterprises. However, few studies have focused on whether financing constraints can be alleviated in order to enhance the green innovation aspect of enterprises, and even though there is a focus on green transformation, green technology [16], or low-carbon energy transformation [17], there is currently no study that focuses on the impact of digital finance on corporate green innovation in terms of both iconic green innovation and substantive green innovation and which further discovers the “quantitative and qualitative separation” effect in the test. That is, the final test finds that digital finance indeed promotes corporate green innovation, but only in terms of iconic green innovation, which is only promoted by digital finance. However, it only promotes iconic green innovations and has no significant impact on substantive green innovations. Therefore, this paper enriches the research on the economic consequences of digital finance. Second, enterprise sustainable development, or green development, enterprise risk, and other aspects have become a new focus of digital finance research, but the path relies mainly on the improvement of the external financing environment and financing links [17,18], and less focus is placed on the internal links of the enterprise. Third, this paper focuses in on the enterprise level based on the perspective of management power and corporate social responsibility to examine the impact of digital finance on enterprise financing, which enriches the perspective and research on enterprise financing constraints of digital finance. In the current era of rapid development in digital finance, regulation has not yet been fully perfected, and this new perspective can help the enterprise level to give more effective suggestions on alleviating financing constraints. Finally, in the heterogeneity analysis, we find that the financing constraints of enterprises in the digital finance chain are not property rights discriminatory and that digital finance is effective in empowering various industries, including industry, agriculture, and the service industry, which provides an effective basis for further promoting digital finance policies. We also add a heterogeneity analysis of high-carbon and low-carbon industries and find that the role of financing constraints in the empowerment chain is stronger for digital finance in low-carbon industries. This detail also provides relevant insights for enterprises that are actively promoting low-carbon operations. The economic consequences of “quantitative and qualitative separation” effect green innovations found in the economic consequences test and suggest that firms should avoid merely improving “superficial” green innovations, which are not really green innovations. These findings provide empirical evidence and a theoretical basis for policymakers and business managers to provide more options for high-quality business development.

2. Literature Review and Research Hypothesis

It is widely recognised in academia that information asymmetry is one of the main reasons firms face financing constraints [19]. Traditional market outcomes are affected by information asymmetry [20], agency problems [21,22], and difficulties in allocating residual control through contracts [23]. Therefore, the emergence of digital finance has laid favorable basic conditions for reducing information asymmetry and alleviating financing constraints.

Didier et al. (2021) found that the lack of external financing is an important constraint to business development [24]. The Chinese Business Operators Questionnaire Survey Report (2021) also points out that financing constraints are an important obstacle to the development of Chinese enterprises, especially private small or medium-sized enterprises (SMEs). In addition, due to the late start of China’s financial market, the imperfect construction of the system, the low degree of market improvement, and the prevalence of information asymmetry, the financial system can not fully meet the financing needs of private SMEs, resulting in the “Macmillan Gap”. In addition, China’s financial market exists in conditions of size discrimination and ownership discrimination, which leads to small and medium-sized enterprises and private enterprises facing more financing constraints. In addition, the size and ownership discrimination in China’s financial market has led to more financing constraints for SMEs and private enterprises [3,10]. In the context of enterprises’ pursuit of high-quality and sustainable development, the transformation and upgrading of enterprises are also facing greater demand for financial and technical support.

Existing literature suggests that the development and application of digitally inclusive financial technologies have attracted widespread attention globally. Digital inclusive finance is a product of the fusion of digital technologies, such as the Internet, and inclusive finance, providing financial products and services with wide coverage and variety, low cost, and high convenience, representing the digital transformation of the entire financial system and emphasising the inclusive character of finance [25].

First, digital finance, as a kind of financial spillover, can drive the reshaping of the traditional financial system, improve the conventional credit pricing model through the transparency and informatization of credit, and mine the massive amount of standardised and non-standardised data, which will reduce the degree of information asymmetry between the supply and demand sides of the funds, and based on the theory of information asymmetry, the adverse selection and moral hazards can be effectively avoided [26].

Second, digital finance makes up for the limitations of the traditional financing model, reduces the cost of enterprise financing by optimising the financial structure and other ways, and reduces the transaction cost by opening up the information transmission channels and reducing the information asymmetry in the borrowing and lending process [26].

Thirdly, digital finance improves access to finance and enhances firms’ financing opportunities. Emerging technologies have effectively boosted the banking sector’s credit supply to SMEs [10,27]. Digital finance breaks down the geographical limitations of traditional financial institutions, providing a basis for realising broad coverage of financial services, and broadens and improves the physical limitations, providing great inclusiveness, enhancing and expanding access to corporate finance [27]. The digitalisation of finance offers new technologies that improve businesses’ access to finance. Based on the above analysis, this paper proposes the following hypothesis:

Hypothesis 1:

Digital finance can significantly alleviate corporate financing constraints.

The rapid development of digital finance in China has brought challenges to regulation, and the corresponding regulatory system has not yet been developed and perfected [28]. So as to alleviate the difficulties of enterprise financing, is it possible to seek solutionsfrom the enterprises’ “internal demand”, and realise the “internal and external” combination path to provide more financial support for the transformation and upgrading of enterprises, thus assisting the high-quality and sustainable development of enterprises. The combination of the “internal and external” path for enterprise transformation, includes upgrading to provide more financial support, so as to help enterprises with high-quality, sustainable development. Based on this, this paper is focused on the enterprise level and tries to explore the influence of digital finance on the role of corporate finance constraints from the perspectives of management power and corporate social responsibility. The main reason for choosing these two perspectives is that the internal core problem of enterprise management, operation, and development is the principal-agent problem, which is a management problem that cannot be avoided by enterprises [21,22] and in which the magnitude of management’s power to a certain extent describes this contradiction [29]. Corporate social responsibility (CSR), on the other hand, in the context of the concept of sustainable development, expresses not the results of the business performance of the enterprise but the embodiment of the responsibility of the enterprise to society in the development of the enterprise, which is a necessary need that cannot be circumvented by the development of the enterprise at the present time [30]. Therefore, this study introduces two moderating variables, management power and CSR, at the firm level to further explore the empowerment of digital finance on corporate finance constraints.

Exploring the internal causes of enterprise financing constraints, the increase of management power is likely to be an important cause of an increase in enterprise management’s opportunism and thus lead the enterprise into financing difficulties. Bedchuk and Fried (2006) put forward the theory of management power, which states that management has the ability to influence their remuneration and use their power to seek rent [29]. The greater the power of management, the easier it is to use on-the-job consumption to meet their wishes, which becomes a means of rent-seeking, but also the more likely they are to achieve control of free cash flow within the enterprise and “seize” the enterprise’s cash flow for their benefit. Therefore, the greater the managerial power, the more serious the corruption may be. Excessive managerial power may increase the opportunistic behaviours of management, including paying excessive bonuses to executives, increasing on-the-job spending, maintaining personal relationships, and over-investing, which ultimately leads to more serious corruption and may reduce the value of the firm and damage the financing environment of the firm. Experiments have shown that experiential power leads to overconfident decision making [31], and management overconfidence can explain the firm’s investment distortions, leading to overinvestment [32], which leads to a decrease in internal funds and thus internal financing constraints [33,34]. Some studies have shown that increased management power deprives the use of capital and significantly reduces green innovation in firms [33,34].

Therefore, based on the above analysis, Hypothesis 2 is proposed in this paper:

Hypothesis 2:

Management power will inversely regulate the relationship between digital finance and corporate financing constraints.

On the other hand, however, firms with better CSR performance have established a better public image and actively fulfill their social responsibilities [30], should be more likely to gain the trust of investors and upstream and downstream firms in the supply chain [30,35] and, at the same time, be more likely to obtain government support to broaden corporate financing channels and reduce the cost of equity capital [36,37], which in turn alleviates financing constraints. Some studies have shown that firms with better CSR performance face significantly lower capital constraints [37,38,39,40]. Studies have shown that better CSR performance leads to more financial support for companies, which in turn enhances their green innovation capability [37,40].

Therefore, based on the above analysis, this paper proposes Hypothesis 3:

Hypothesis 3:

Improving CSR will promote the mitigating effect of digital finance on corporate finance constraints.

3. Research Design and Sample Selection

3.1. Data and Sample Selection

Data sources for this paper: data on digital finance are from the Digital Financial Inclusion Index released by the Digital Finance Research Center of Peking University; industry category data are sourced from the GICS Global Industry Classification System in the S&P Global Market Intelligence database, with companies classified by industry using the first two digits of the industry classification code; data on social responsibility are adopted from the Rankins CSR rating and scoring measurement and other research data from the CSMAR database. This paper selected the annual report data of A-share non-financial listed companies in Shanghai and Shenzhen from 2011 to 2020 as the initial research samples, excluded ST, *ST, and missing data samples, and finally obtained 17,238 sample observations. To eliminate the influence of extreme values on the empirical results, the observations were subjected to a winsorize treatment of the upper and lower 1% quartiles.

3.2. Definition of Variables

3.2.1. Dependent Variables

As to the rigour of the article’s index selection, asthe validity of the KZ index has been questioned a lot. Based on existing research [15], this study uses the financing constraint (WW) indices to measure corporate financing constraints.

The WW index construction methodology can accurately satisfy the underlying financing constraints [41] and is highly consistent with the evaluation metrics of the firms’ other dimensions [15]. Research has found that the WW index is more effective than the KZ index and SA index in accurately and objectively describing corporate financing constraints. Furthermore, the construction method of the WW index is more consistent with the definition of financing constraints [15].

Drawing on the research methodology of Whited and Wu (2006) and others [42], the WW index is constructed:

WW = −0.091 × CF − 0.062 × DivPos + 0.021 × Lev − 0.044 × Size + 0.102 × ISG − 0.035 × SG

(1)

The calculation is based on a linear combination of corporate financial characteristics and is estimated using a regression model. The WW index is based on the Euler equation of optimal investment behaviour and uses GMM to estimate the impact of financing constraints on corporate investment decisions. The model assumes that financing constraints distort corporate investment behaviour and that financial variables can capture this distortion.

Here, CF: cash flow to total assets ratio = net cash flow from operating activities/total assets; DivPos: cash dividend payment dummy variable, one if cash dividends are paid in the current period, zero otherwise; Lev: long-term liabilities-to-assets ratio; Size: the natural logarithm of total assets; ISG: industry average sales growth rate; according to the 2012 SEC industry classification standards. According to the 2012 SEC industry classification standard, the manufacturing industry takes two codes, and other sectors take one code; SG: sales revenue growth rate.

The coefficient reflects the direction and strength of the marginal impact of different financial variables on corporate financing constraints. The specific meanings of the coefficients are as follows: The coefficient for CF is −0.091. Its specific meaning is that for every 1-unit increase in the cash flow ratio (CF), the WW index decreases by 0.091, indicating that the company’s financing constraints are intensifying. This is because companies with strong financing constraints rely more on internal cash flows (due to high external financing costs). Therefore, companies with high cash flows may actually be the ones with more severe financing constraints (cash flows become the primary source of investment). The coefficient for DIVPOS is −0.062, meaning that for companies paying dividends (DIVPOS = 1), the WW index decreases by 0.062, indicating more severe financing constraints. The coefficient for Lev is 0.021, indicating that the higher the long-term debt ratio, the lighter the financing constraints (due to stronger debt financing capacity). The coefficient for ISG is 0.102, indicating that industry growth (ISG) alleviates financing constraints (high-growth industries are more likely to secure external funding). The coefficient for SG is −0.035, indicating that corporate sales growth (SG) may increase funding requirements; if growth is too rapid and financing channels are limited, this may exacerbate constraints.

3.2.2. Economic Consequences Explanatory Variables

Green Innovation. The explanatory variable of the economic consequence variable of the study is green innovation. Since the measurement method of green innovation has not yet formed a unified and standardised international standard, scholars in different fields have scientifically quantified green innovation from various perspectives. Based on scientific considerations, we determine the number of green invention patent applications filed by enterprises based on the International Patent Classification (IPC) code in the Green List of the International Patent Classification [43]. Referring to existing research [44], two types of innovation behaviours are described from the perspective of innovation effects based on the definition of Chinese patent law and existing research literature [45]. Firms applying for “quality” green invention patents are recognised as (Gpi1) substantive green innovations, and firms applying for green utility model patents are identified as (Gpu1) signature green innovations.

3.2.3. Independent Variables

Digital Finance Development Index Group (Index). For the selection of variables based on the more recognised approach of the larger body of existing research on digital finance [10,15,46], we use the China Digital Financial Inclusion Index compiled by the Digital Finance Research Centre of Peking University to describe the development of digital finance in the region. The index establishes a multilevel index system based on a large amount of data provided by the Ant Group of Companies, which contains an overall index (Index) and further distinguishes three more refined dimensions. The breadth of digital financial coverage (Coverage) describes the coverage of DF services through indicators such as comprehensive e-account coverage and the number of tied cards. The depth of digital financial usage (Usage_depth) describes the number of users and active transactions of various DF services such as payment, credit, investment, insurance, etc., and the degree of inclusive digital financial digitisation (Digital) is used to quantify the extent of digital financial digitisation by quantifying the number of users of DF services. (Digital) is used to measure the inclusiveness of DF by quantifying the convenience of DF services, the cost of individual and MSME loans, and the degree of creditworthiness. In this way, the downscaling and decomposition of the total index can more accurately describe the development level of DF at the provincial and municipal levels in China from different aspects and provide a more objective and comprehensive index evaluation of the development status of DF in China. Referring to existing studies [10], the selected data are from the Digital Financial Inclusion Index of Chinese Municipal Cities. The Peking University Digital Financial Inclusion Index examines the level of digitally inclusive financial development in 31 provinces, 337 cities at the county level and above, and about 2800 counties in mainland China from 2011 to 2020. The index system is a dimensionless processing of 33 specific sub-indicators with corresponding weights, and it measures them in terms of breadth of coverage (Ccoverage), depth of use (Cusage_depth), degree of inclusive digital financial digitisation (Cdigital) and finally synthesizes the Digital Inclusive Finance Index at the city level. We take the logarithm of this index for empirical research. In the robustness test, the provincial DF development index will be chosen as the core explanatory variable.

3.2.4. Moderator Variables

The moderating variables in this paper are management power (Power) and corporate social responsibility (CSR).

Management power (Power): Since management authority is measured unilaterally, there may be a problem of not being able to measure it comprehensively. As a result, the research metrics integrate insights from existing research [47,48] and are based on existing research [33]. A composite indicator of management power intensity (Power) is constructed by principal component regression on the following five indicators: (1) whether the CEO and the chairman of the board of directors are two positions in one, which takes the value of 1 when the CEO is also the chairman of the board of directors of the company, otherwise it is 0; when the CEO is also the chairman of the board of directors of the company, the intensity of management power is higher. (2) The size of the board of directors is equal to the total number of board members in the current year. The larger the size of the board, the greater the intensity of management power. (3) Ratio of inside directors, equal to the ratio of the number of inside directors to the total number of board members, where the number of inside directors is equal to the total number of board members minus the number of independent directors. The greater the value, the greater the strength of management’s power. (4) Shareholding dispersion, which is the ratio of the proportion of shares held by the second to tenth largest shareholders to the proportion of shares held by the first largest shareholder. The greater the ratio, the greater the intensity of management power. (5) Management shareholding, measured by the ratio of management’s shareholding. The greater the ratio, the greater the intensity of management power. It can be seen that the larger the value of the comprehensive index of management power intensity (Power), the larger the management power intensity is.

Corporate social responsibility (CSR): This indicator is measured using the Rankins CSR Rating Score, a third-party authority widely used in current research to study CSR issues of listed companies in China [40,49]. Higher scores indicate better CSR performance.

3.2.5. Control Variables

In order to reduce parameter estimation bias due to omission of key variables, this study employs a series of control variables considering multiple dimensions. First, at the firm level, a total of 10 variables such as firm characteristics, operating performance, asset structure, and board composition are the control variables: firm size (size), operating performance (roa), financial leverage (lev), the nature of ownership (SOE), the firm’s age of establishment (FrimAge), growth capacity (growth), Tobin’s Q (TobinQ), board size (board), audit quality (big4), and percentage of independent directors (indep). Second, as the adoption of digital finance is likely to be related to the degree of digital development of firms, firm digital transformation (wdigital-trans) is added as a control variable based on the degree of digital transformation of firms measured by existing studies [50]. At the macro level, the development and operation of firms may be more effective in more developed regions; therefore, regional GDP, prefecture, and city-level GDP are added as macro-level control variables. In addition, year (YEAR) and industry (IND) fixed effects are used, and provincial clustering standard errors are used to ensure the rigour of the study. The main variables are defined and measured as shown in

Table 1. Definition table of main research variables.

Variable Type	Variable Names	Variable Symbols	Variable Definitions
Dependent variables	Financing constraints	WW	Drawing on Whited and Wu (2006) [42] and others, the construction of the WW index.
Economic consequences: explanatory variables	Substantive green innovations	Gpi1	ln (number of green inventions independently filed in the year + 1)
	Signature Green Innovations	Gpu1	ln (number of green utility models filed independently in the year + 1)
Independent variables	Digital Finance Index Municipal	Cindex	Digital finance metropolitan-level aggregate indexes are standardised
	Digital Finance Coverage Breadth Municipal	Ccoverage	Standardising the breadth of digital finance coverage at the municipal level
	Depth of digital finance use at the municipal level	Cusage_depth	Digital finance local municipalities use depth to do standardisation
	Digital Finance Digitisation Municipal	cdigital	Digital finance digitisation at the municipal level for standardisation
	digital finance	Index	Provincial digital finance aggregate indexes are standardised
	The breadth of coverage of secondary indicators on digital finance	Coverage	Digital Finance Provincial Breadth of Coverage Index Normalised
	Depth of use of secondary indicators on digital finance	Us-age_depth	Digital Finance Provincial Depth of Usage Index Normalised
	Level of digitization of secondary indicators of digital finance	Digital	Digital Finance Provincial Level of Digitization Index Normalised
Moderator variable	Management Power	Power	A composite indicator of management power intensity (Power) is constructed by principal component regression on five indicators.
	Corporate Social Responsibility	CSR	Rankin’s CSR Rating Score Measure
	Enterprise size	Size	Operating income in natural logarithms
Control variables	Business performance	Roa	Average year-end net profit/total assets
	Earnings on net assets	Roe	Net profit/average shareholder equity
	Financial leverage	Lev	Total liabilities/total assets at the end of the year
	Ownership properties	SOE	It takes the value of 1 if the actual controller is a state-owned property and vice versa.
	Growth Capacity	Growth	Growth in operating income/total operating income of the previous year
	Tobin’s Q	TobinQ	The market value of the company/replacement cost of assets
	Board Size	Board	The number of board members is taken as a natural logarithm.
	Audit Quality	Big4	Auditors from the Big Four accounting firms in China take the value of 1, and vice versa, 0.
	Percentage of Independent Directors	Indep	Number of independent directors/total number of board members
	Years of Establishment	FirmAge	ln (current year-year of establishment + 1)
	Cash flow ratios	Cashflow	Net cash flows from operations/total assets
	Degree of digitisation of the enterprise	wdigitaltrans	The text in the annual report was analysed for word frequency, counting the number of times the keywords appeared in the text.
	City GDP	Cgdp	Standardised data processing of regional GDP
	Year	Year	Year fixed effects
	Industry	Ind	Industry fixed effect

.

3.2.6. Model Construction

This paper constructs models based on theoretical analysis and research hypotheses, mainly including the following six:

First, H1 is tested with the model (2). The dependent variable ${WW}_{i,t}$ represents the degree of financing constraints faced by company i in year t. The independent variable ($Cindex$) represents the level of digital finance in the city where the company is located in year t. The control variables are shown in Table 1. ${\alpha }_0$ is the constant term, ${\epsilon }_{i,t}$is the error term, and annual fixed effects (${\tau }_t$) and industry fixed effects (${\gamma }_i$) are taken into account. The coefficient${\alpha }_1$ represents the effect of digital finance on alleviating corporate financing constraints:

$${WW}_{i,t} = {\alpha }_0 + {\alpha }_{1}Cindex +{ \alpha }_2{Controls}_{i,t} + {\gamma }_i + {\tau }_t + {\epsilon }_{i,t}$$

(2)

Second, model (3) is used to test H2, where the coefficient ${\delta }_3$indicates the moderating effect of managerial power:

$${WW}_{i,t} = {\delta }_0 +{ \delta }_{1}Cindex + {\delta }_2{Power}_{i,t} + {\delta }_{3}Cindex \times {Power}_{i,t} + {\delta }_4{Controls}_{i,t} + {\gamma }_i + {\tau }_t +{\epsilon }_{i,t}$$

(3)

Third, model (4) is used to test H3, where the coefficient ${\beta }_3$ indicates the moderating effect of CSR.

$${WW}_{i,t}={\beta }_0+{\beta }_{1}pro+{\beta }_2{CSR}_{i,t }+{\beta }_{3}Cindex\times {CSR}_{i,t}+{ \beta }_4{Controls}_{i,t}+{\gamma }_i+{\tau }_t+{\epsilon }_{i,t}$$

(4)

The economic consequences of digital financial empowerment for corporate financing constraints on firms’ green innovations are tested with models (5) and (6).

$${\mathrm{G}\mathrm{p}\mathrm{i}1}_{\mathrm{I},\mathrm{T}}={\theta }_0+{\theta }_{1}Cindex+{\theta }_2{WW}_{i,t}+{\theta }_{3}Cindex\times {WW}_{i,t}+{\theta }_4{Controls}_{i,t}+{\gamma }_i+{\tau }_t+{\epsilon }_{i,t}$$

(5)

$${Gpu1}_{i,t}={\omega }_0+{\omega }_{1}Cindex+{\omega }_2{WW}_{i,t}+{\omega }_{3}Cindex\times {WW}_{i,t}+{\omega }_4{Controls}_{i,t}+{\gamma }_i+{\tau }_t+{\epsilon }_{i,t}$$

(6)

4. Empirical Results and Analysis

4.1. Descriptive Statistics Analysis

Table 2. Descriptive statistical analysis.

Variables	MEAN	STD	MIN	Q1	MEDIAN	Q3	MAX
CINDEX	0.670	0.264	0.000	0.497	0.722	0.874	1.000
CCOVERAGE	0.551	0.237	0.000	0.371	0.572	0.712	1.000
CUSAGE_DEPTH	0.478	0.265	0.000	0.270	0.463	0.685	1.000
CDIGITAL	0.526	0.308	0.000	0.318	0.395	0.919	1.000
INDEX	0.543	0.289	0.000	0.300	0.586	0.802	1.000
COVERAGE	0.521	0.303	0.000	0.256	0.571	0.746	1.000
USAGE_DEPTH	0.566	0.249	0.000	0.375	0.582	0.798	1.000
DIGITAL	0.482	0.280	0.000	0.269	0.519	0.687	1.000
WW	−1.026	0.071	−1.258	−1.069	−1.022	−0.977	−0.795
POWER	−0.064	1.319	−5.286	−1.006	0.170	1.134	3.013
CSR	4.643	2.570	0.000	2.000	5.000	7.000	8.000
SIZE	22.189	1.291	19.563	21.248	22.003	22.929	26.398
ROA	0.043	0.067	−0.368	0.016	0.041	0.075	0.241
ROE	0.065	0.164	−4.320	0.032	0.075	0.125	2.379
LEV	0.414	0.204	0.028	0.249	0.404	0.566	0.909
GROWTH	0.163	0.394	−0.658	−0.021	0.104	0.260	4.429
BOARD	8.663	1.738	5.000	7.000	9.000	9.000	15.000
TOBINQ	1.992	1.332	0.809	1.231	1.583	2.237	17.676
INDEP	37.565	5.288	30.000	33.330	36.360	42.860	60.000
SOE	0.363	0.481	0.000	0.000	0.000	1.000	1.000
BIG4	0.056	0.231	0.000	0.000	0.000	0.000	1.000
FIRMAGE	2.908	0.329	0.693	2.708	2.944	3.135	4.143
CASHFLOW	0.047	0.074	−0.888	0.009	0.047	0.088	0.876
WDIGITALTRANS	3.725	1.518	0.000	2.639	3.497	4.595	7.170
CGDP	0.507	0.243	0.000	0.313	0.480	0.708	1.000

reports the results of descriptive statistics for each variable after eliminating the effect of extreme values. It can be seen that the mean value of firms’ financing constraints as measured by the WW index is −1.026 with a standard deviation of 0.071, a maximum value of −0.795, and a minimum value of −1.258, which suggests that the extent of financing constraints faced by firms varies significantly. Regarding other firm characteristics, the average firm in the sample is of medium size (size = 22.189) with an ROE and leverage ratio (LEV) of 0.065 and 0.414, respectively. These data suggest that, on average, the firms in the sample exhibit expansive growth potential and maintain moderate levels of leverage. These observations are consistent with the trends documented in previous studies [10]. In terms of governance characteristics, the average board size is eight, and the average proportion of independent directors on the board is 37.6%.

Figure 1 shows the correlation trend between digital finance at the city level (Cindex) and corporate financing constraints (WW) across the entire sample. We can clearly see from the figure that the two trends are opposite.

Figure 2 shows the provincial composite index for digital finance (Index) and the secondary indices for digital finance coverage (Coverage), digital finance usage depth (Usage_depth), and digital finance digitisation level (Digital), as well as trends in changes across the entire sample.

4.2. Basic Regression Analyses

Table 3. Impact of digital finance on corporate finance constraints.

Variables	(1)	(2)	(3)	(4)
Intercept	0.030 ***	0.029 ***	0.031 ***	0.028 ***
	(3.66)	(3.50)	(3.64)	(3.28)
Cindex	−0.008 ***
	(−2.81)
Ccoverage		−0.005 *
		(−2.02)
Cusage_depth			−0.009 ***
			(−4.74)
Cdigital				−0.005 **
				(−2.68)
Size	−0.047 ***	−0.047 ***	−0.047 ***	−0.047 ***
	(−124.89)	(−124.63)	(−122.53)	(−122.21)
ROA	−0.155 ***	−0.156 ***	−0.155 ***	−0.156 ***
	(−23.01)	(−23.42)	(−22.68)	(−23.92)
Lev	0.025 ***	0.026 ***	0.025 ***	0.025 ***
	(13.50)	(13.66)	(13.26)	(13.33)
Growth	−0.043 ***	−0.043 ***	−0.043 ***	−0.043 ***
	(−48.38)	(−48.34)	(−48.44)	(−48.37)
TobinQ	0.002 ***	0.002 ***	0.002 ***	0.002 ***
	(9.47)	(9.48)	(9.62)	(9.70)
Board	−0.002	−0.002	−0.002	−0.002
	(−1.30)	(−1.27)	(−1.31)	(−1.20)
Big4	−0.002 *	−0.002 *	−0.002 *	−0.002 *
	(−1.93)	(−1.92)	(−1.90)	(−1.89)
Indep	−0.001	−0.001	−0.001	−0.001
	(−1.03)	(−1.05)	(−1.07)	(−1.09)
SOE	−0.001	−0.001	−0.001	−0.001
	(−1.34)	(−1.25)	(−1.55)	(−1.22)
FirmAge	0.004 ***	0.004 ***	0.004 ***	0.004 ***
	(2.95)	(2.96)	(3.05)	(3.00)
ROE	−0.003 *	−0.003 *	−0.004 *	−0.004 *
	(−1.71)	(−1.71)	(−1.79)	(−1.73)
Cashflow	−0.098 ***	−0.098 ***	−0.097 ***	−0.098 ***
	(−32.56)	(−32.47)	(−32.27)	(−32.25)
Wdigitaltrans	0.002 ***	0.002 ***	0.002 ***	0.002 **
	(2.78)	(2.76)	(2.78)	(2.69)
Cgdp	−0.001	−0.002	<0.001	−0.003 ***
	(−0.43)	(−1.06)	(−0.17)	(−2.85)
year	YES	YES	YES	YES
ind	YES	YES	YES	YES
R2	87.12%	87.11%	87.13%	87.11%
Adj R2	87.07%	87.06%	87.08%	87.06%
N	26,381	26,381	26,381	26,381
F-value	764.382 ***	951.594 ***	731.767 ***	659.997 ***

Note: *, **, *** indicate significance at the 10%, 5%, and 1% levels, respectively.

reports the effect of digital finance on firm financing constraints (WW) after adding control variables, year, and industry fixed effects to test whether H1 holds. The results in column (1) of Table 3 show that digital finance can significantly alleviate corporate finance constraints. To further explore the effect of digital finance on corporate financing constraints, regression tests are conducted on the impact of digital finance on corporate financing constraints from the three dimensions of digital finance: breadth of digital finance coverage (Prc), depth of digital finance use (Prd), and digital finance digitisation degree (Pri) in turn. The results in columns (2) to (4) of Table 3 show that digital finance can significantly alleviate corporate financing constraints, and the depth of digital financial use (Prd) significantly alleviates corporate financing constraints. Mitigation of corporate financing constraints is the most significant, and Hypothesis 1 is valid.

4.3. An Analysis of the Moderating Role of Management Power on Digital Finance and Financing Constraints

Table 4. Test results of the relationship between digital finance, management power, and corporate finance constraints.

Variables	(1)	(2)	(3)	(4)
Intercept	0.034 ***	0.035 ***	0.034 ***	0.035 ***
	(5.73)	(5.92)	(5.74)	(5.95)
Cindex	<0.001
	(0.03)
Ccoverage		−0.006 ***
		(−6.13)
Cusage_depth			−0.001
			(−1.58)
Cdigital				−0.003 ***
				(−4.16)
Cindex*Power	0.001
	(1.23)
Ccoverage*Power		0.002 ***
		(3.72)
Cusage_depth*Power			0.001 ***
			(2.59)
Cdigital*Power				0.001 *
				(1.92)
Power	−0.002 ***	−0.003 ***	−0.002 ***	−0.002 ***
	(−4.52)	(−6.60)	(−6.54)	(−5.64)
Size	−0.047 ***	−0.047 ***	−0.047 ***	−0.047 ***
	(−234.91)	(−234.65)	(−235.13)	(−234.99)
ROA	−0.155 ***	−0.155 ***	−0.155 ***	−0.155 ***
	(−30.82)	(−30.59)	(−30.85)	(−30.89)
Lev	0.026 ***	0.025 ***	0.025 ***	0.025 ***
	(21.14)	(20.75)	(21.10)	(21.10)
Growth	−0.043 ***	−0.043 ***	−0.043 ***	−0.043 ***
	(−51.78)	(−51.75)	(−51.78)	(−51.80)
TobinQ	0.002 ***	0.002 ***	0.002 ***	0.002 ***
	(14.18)	(14.21)	(14.20)	(14.05)
Board	−0.006 ***	−0.006 ***	−0.006 ***	−0.006 ***
	(−4.48)	(−4.78)	(−4.42)	(−4.44)
Big4	−0.003 ***	−0.002 ***	−0.003 ***	−0.003 ***
	(−3.72)	(−3.33)	(−3.77)	(−3.70)
Indep	<0.001 ***	<0.001 ***	<0.001 ***	<0.001 ***
	(4.14)	(4.01)	(4.16)	(4.04)
SOE	−0.002 ***	−0.002 ***	−0.002 ***	−0.002 ***
	(−4.00)	(−4.23)	(−3.97)	(−4.32)
FirmAge	0.004 ***	0.004 ***	0.004 ***	0.004 ***
	(7.24)	(7.33)	(7.22)	(7.38)
ROE	−0.004 *	−0.004 *	−0.004 *	−0.003 *
	(−1.86)	(−1.88)	(−1.89)	(−1.86)
Cashflow	−0.098 ***	−0.098 ***	−0.098 ***	−0.097 ***
	(−34.92)	(−34.93)	(−34.90)	(−34.75)
Wdigitaltrans	<0.001	<0.001	<0.001	<0.001
	(−1.02)	(−0.91)	(−1.00)	(−0.81)
Cgdp	−0.007 ***	−0.003 ***	−0.007 ***	−0.006 ***
	(−5.42)	(−3.41)	(−8.10)	(−6.31)
year	YES	YES	YES	YES
ind	YES	YES	YES	YES
R2	87.12%	87.14%	87.12%	87.13%
Adj R2	87.07%	87.09%	87.07%	87.07%
N	26,371	26,371	26,371	26,371
F-value	17,118.588 ***	17,209.631 ***	17,112.658 ***	17,104.265 ***

Note: *, *** indicate significance at the 10% percent and 1% percent levels, respectively.

reports the results of the test of the moderating effect of management power (Power) on the relationship between digital finance and financing constraints (WW) after adding control variables, fixed year and industry effects, and using provincial clustering standard errors. The results show that management power will have a reverse moderating effect on the role of breadth of coverage, depth of use, and degree of digitisation of digital finance on financing, and Hypothesis 2 holds.

4.4. Analysing the Moderating Role of Corporate Social Responsibility on Digital Finance and Financing Constraints

Table 5. Results of the relationship between digital finance, corporate social responsibility, and corporate finance constraints.

Variables	(1)	(2)	(3)	(4)
Intercept	1.036	0.461	1.296 *	−0.208
	(1.56)	(0.64)	(1.92)	(−0.25)
Cindex	0.222 **
	(2.37)
Ccoverage		0.315 ***
		(5.30)
Cusage_depth			0.175 **
			(2.08)
Cdigital				0.410 ***
				(5.43)
Cindex*csr	−0.018 ***
	(−12.15)
Coverage*csr		−0.014 ***
		(−11.55)
Cusage_depth*csr			−0.019 ***
			(−11.34)
Cdigital*csr				−0.012 ***
				(−11.87)
Csr	0.065 ***	0.045 ***	0.069 ***	0.036 ***
	(9.59)	(8.05)	(8.90)	(7.66)
Size	−0.065	−0.065	−0.064	−0.067
	(−1.32)	(−1.31)	(−1.30)	(−1.36)
Roa	−0.006	−0.006	−0.006	−0.006
	(−1.06)	(−1.07)	(−1.06)	(−1.08)
Lev	0.035 ***	0.035 ***	0.035 ***	0.035 ***
	(5.85)	(5.83)	(5.84)	(5.83)
Growth	<0.001	<0.001	<0.001	<0.001
	(1.13)	(1.08)	(0.86)	(0.79)
SOE	0.291 ***	0.295 ***	0.286 ***	0.293 ***
	(8.80)	(8.80)	(8.68)	(8.59)
TobinQ	0.013 **	0.013 **	0.013 **	0.012 **
	(2.36)	(2.36)	(2.37)	(2.36)
Big4	−0.075	−0.081	−0.069	−0.112 **
	(−1.40)	(−1.50)	(−1.29)	(−2.00)
Board	−0.029 ***	−0.029 ***	−0.029 ***	−0.027 ***
	(−3.60)	(−3.56)	(−3.62)	(−3.37)
Indep	0.875 ***	0.889 ***	0.868 ***	0.928 ***
	(3.93)	(3.99)	(3.89)	(4.17)
FirmAge	−0.003 **	−0.003 **	−0.003 **	−0.003 **
	(−2.48)	(−2.50)	(−2.42)	(−2.31)
ROE	<0.001	<0.001	<0.001	<0.001
	(0.93)	(0.95)	(0.93)	(0.95)
wdigitaltrans	0.006 ***	0.006 ***	0.006 ***	0.006 ***
	(3.75)	(3.59)	(3.82)	(3.56)
Cashflow	−0.097 ***	−0.097 ***	−0.097 ***	−0.097 ***
	(−24.32)	(−24.37)	(−24.38)	(−24.37)
Cgdp	−0.001	−0.001	−0.001	−0.001
	(−1.44)	(−1.42)	(−1.51)	(−1.47)
Year	0.003	0.004	−0.002	−0.007 ***
Ind	(0.78)	(0.70)	(−0.51)	(−4.11)
R2	49.87%	49.75%	49.86%	49.69%
Adj R2	49.78%	49.66%	49.77%	49.60%
N	17,238	17,238	17,238	17,238
F-value	340.966 ***	333.295 ***	339.573 ***	331.206 ***

Note: *, **, *** indicate significance at the 10% percent, 5% percent, and 1% percent levels, respectively.

reports the results of the moderating effect test of corporate social responsibility (CSR) on the relationship between digital finance (Cindex) and corporate finance constraints (WW), as well as the results of the moderating effect test of CSR on the relationship between the three secondary indicators of digital finance and corporate finance constraints (WW). Moderating role test results. The results show a negative corporate social responsibility (CSR) toward moderating the relationship between digital finance and corporate finance constraints; H3 holds.

4.5. Endogeneity Test

This paper first uses the instrumental variable method for the endogeneity test.

Table 6. Endogeneity test.

Variables	1st Stage	WW		1st Stage	WW
Intercept	0.023 **	0.028 ***	Intercept	−0.616 ***	0.024 ***
	(2.34)	(3.88)		(−18.58)	(4.44)
X		−0.004 ***	X		−0.005 ***
		(−5.21)			(−4.39)
lag_var	0.975 ***		internetrate	0.016 ***
	(798.98)			(191.65)
lag_indvar	−0.035 ***
	(−4.73)
Size	<0.001	−0.047 ***		0.004 ***	−0.047 ***
	(0.15)	(−181.10)		(3.56)	(−240.59)
ROA	−0.003	−0.157 ***		0.155 ***	−0.156 ***
	(−0.41)	(−20.12)		(5.51)	(−31.02)
ROE	0.006 ***	−0.007 **		−0.025 **	−0.003 *
	(2.65)	(−2.25)		(−2.40)	(−1.83)
Lev	0.001	0.025 ***		−0.073 ***	0.025 ***
	(0.82)	(16.05)		(−10.75)	(20.82)
Growth	0.001	−0.041 ***		−0.001	−0.043 ***
	(0.94)	(−40.45)		(−0.43)	(−52.25)
TobinQ	<0.001 **	0.002 ***		−0.001	0.002 ***
	(2.29)	(10.98)		(−0.80)	(14.80)
Board	<0.001	−0.002		−0.021 ***	−0.001
	(0.04)	(−1.55)		(−3.29)	(−1.40)
Big4	0.004 ***	−0.001		0.025 ***	−0.002 ***
	(3.36)	(−1.53)		(5.23)	(−3.23)
SOE	<0.001	−0.001		−0.016 ***	−0.001 **
	(0.06)	(−1.22)		(−6.86)	(−2.43)
Indep	<0.001	<0.001		−0.002 ***	<0.001
	(−1.36)	(−0.08)		(−7.18)	(0.91)
FirmAge	0.001	0.003 ***		−0.007 **	0.004 ***
	(0.77)	(4.87)		(−2.26)	(8.02)
Cashflow	0.009 **	−0.098 ***		0.025 *	−0.097 ***
	(2.11)	(−27.05)		(1.65)	(−34.65)
wdigitaltrans	<0.001	<0.001		−0.006***	<0.001
	(−0.72)	(−1.05)		(−3.95)	(−1.32)
Cgdp	0.012 ***	−0.005 ***		0.296 ***	−0.005 ***
	(8.54)	(−4.16)		(56.69)	(−4.23)
year	YES	YES		YES	YES
ind	YES	YES		YES	YES
R2	98.45%	87.19%		61.66%	87.11%
N	15,071	15,071		17,328	17,328
F-value	11,399.175 ***	1160.012 ***		1466.440 ***	17,472.714 ***

Note: *, **, *** indicate significance at the 10% percent, 5% percent, and 1% percent levels, respectively.

shows the results. The lagged one-period explanatory variable (lag_var) and the lagged one-period peer mean (lag_indvar) are selected as instrumental variables and tested using the two-stage least squares (LS) method. In the first stage test, the coefficients of each instrumental variable (lag_var, lag_indvar) are 0.975 and −0.035 at a 1% significance level, indicating that the selection of instrumental variables is reasonable. The coefficient of X in the second-stage test is significantly negative at the 1% significance level, with a coefficient of −0.004, and the above empirical results are consistent with the obtained conclusions, proving that the findings are reliable.

In addition to the use of provinces’ and cities’ network penetration rate (internetrate) as an instrumental variable, network penetration rate, as one of the essential components of the development of digital finance itself, is highly correlated with the degree of development of digital finance. The first stage of the test network penetration rate (internetrate) and digital finance (Cindex) at the 1% level of significance shows a significant positive correlation, and the coefficient value is 0.016. The Internet penetration rate itself and corporate finance constraints do not affect each other, so the Internet penetration rate is a more reasonable and adequate instrumental variable. In the second stage of the test, the coefficient of X is significantly negative at a significant level of 1%, with a coefficient of −0.005, which indicates that the instrumental variable Internet penetration rate (internet rate) significantly alleviates corporate finance constraints (WW). The test results confirm the consistency with the already obtained conclusions, proving the findings are reliable.

5. Further Testing

5.1. Base Regression Robustness Test

Replacement of the main variables measured by the robustness test is as follows: The explanatory variable digital finance total municipal index (Cindex) and its three dimensions of the municipal secondary indicators are replaced by the provincial index, empirical testing of the digital finance (Index), digital finance coverage breadth (Coverage), digital finance usage depth (Usage_depth), digital finance (Digital) on corporate finance constraints (WW), testing the use of year and industry fixed effects, and replacing the original control variable municipal GDP for provincial GDP. On the role of corporate finance constraints (WW), test the use of year and industry fixed effects, provincial clustering standard error, and replace the original control variable, municipal GDP, with provincial GDP.

Column (1) of

Table 7. Base regression robustness tests.

Variables	(1)	(2)	(3)	(4)
Intercept	0.056 ***	0.061 ***	0.031 ***	0.061 ***
	(2.79)	(3.63)	(3.74)	(3.20)
Index	−0.006 *
	(−2.03)
Coverage		−0.006 **
		(−2.70)
Usage_depth			−0.001
			(−1.48)
Digital				−0.007 **
				(−2.16)
Size	−0.047 ***	−0.047 ***	−0.047 ***	−0.047 ***
	(−124.72)	(−124.62)	(−125.01)	(−126.72)
Roa	−0.156 ***	−0.155 ***	−0.156 ***	−0.156 ***
	(−23.56)	(−23.34)	(−24.19)	(−23.79)
Lev	0.025 ***	0.025 ***	0.025 ***	0.025***
	(12.50)	(12.53)	(12.53)	(12.74)
Growth	−0.043 ***	−0.043 ***	−0.043 ***	−0.043 ***
	(−48.62)	(−48.48)	(−48.76)	(−48.71)
TobinQ	0.002 ***	0.002 ***	0.002 ***	0.002 ***
	(9.85)	(9.99)	(9.96)	(9.85)
Board	−0.002	−0.002	−0.002	−0.002
	(−1.29)	(−1.31)	(−1.28)	(−1.31)
Big4	−0.002 *	−0.002 *	−0.002 *	−0.002 *
	(−1.78)	(−1.79)	(−1.76)	(−1.74)
Dual	−0.001	−0.001	−0.001	−0.001
	(−1.33)	(−1.30)	(−1.32)	(−1.28)
SOE	−0.001	−0.001	−0.001	−0.001
	(−1.41)	(−1.51)	(−1.26)	(−1.28)
FirmAge	0.004 ***	0.004 ***	0.004 ***	0.004 ***
	(3.27)	(3.29)	(3.23)	(3.16)
ROE	−0.004 *	−0.004 *	−0.004 *	−0.004 *
	(−1.78)	(−1.79)	(−1.74)	(−1.71)
Cashflow	−0.097 ***	−0.097 ***	−0.097 ***	−0.097 ***
	(−32.75)	(−32.81)	(−32.46)	(−32.56)
wdigitaltrans	<0.001	<0.001	<0.001	<0.001
	(−0.98)	(−0.99)	(−1.02)	(−0.91)
Gdp	−0.003 *	−0.002	−0.006 ***	−0.004 ***
	(−1.98)	(−1.66)	(−4.04)	(−2.98)
year	YES	YES	YES	YES
ind	YES	YES	YES	YES
R2	87.11%	87.11%	87.10%	87.11%
Adj R2	87.06%	87.06%	87.05%	87.06%
N	26,388	26,388	26,388	26,388
F-value	2512.242 ***	13,165.411 ***	720.693 ***	1373.644 ***

Note: *, **, *** indicate significance at the 10% percent, 5% percent, and 1% percent levels, respectively.

reports that the Digital Finance Index (Index) is significantly negative at the 10% significance level with a coefficient value of −0.006, indicating that digital finance significantly alleviates corporate finance constraints. Column (2) reports that provincial digital finance coverage breadth (Coverage) is significantly negative at the 5% significance level, with a coefficient value of −0.06, and column (4) reports that the digital finance digitisation process (Digital) is significantly negative at the 5% significance level, with a coefficient value of −0.07. The results basically confirm that, after replacing the explanatory variables, digital finance still has significant mitigating effect on corporate finance constraints, and the empirical conclusion is basically consistent with the previous results.

5.2. Analysis of the Impact of Digital Finance on Corporate Finance Constraints in Different Contexts

Property rights discrimination is one of the key factors contributing to financing constraints for Chinese enterprises [10,41]. Against the backdrop of China’s dual carbon goals and pursuit of sustainable economic development, achieving high-quality green sustainable development has become a crucial objective for enterprises. Pursuing sustainable development and enhancing green innovation capabilities require adequate financial support, which is crucial. Against this backdrop, further heterogeneous research on how digital finance can alleviate financing constraints for enterprises can be conducted. This will help us to understand the internal composition, structure, and changes of Chinese enterprises, enabling a more targeted understanding of how digital finance empowers enterprises to alleviate financing constraints across different subgroups. Consequently, more tailored solutions can be provided based on these differences.

5.2.1. Impact Analysis of Heterogeneity Based on the Nature of Corporate Property Rights

Figure 3 shows the trends for the entire sample of state-owned (national) and non-state-owned enterprises (nonational). The chart clearly shows that non-state-owned enterprises account for a larger proportion and are growing steadily. This indicates that the development and financing of non-state-owned enterprises have a significant impact on the overall financing constraints of Chinese enterprises.

Table 8. Heterogeneity analysis based on the nature of business ownership.

Group 1		(1)	(2)	(3)	(4)
Panel A national
	Intercept	0.035 *	0.035 *	0.036 *	0.035 *
		(1.96)	(1.95)	(2.01)	(1.92)
	Cindex	−0.011 **
		(−2.26)
	Ccoverage		−0.008
			(−1.24)
	Cusage_depth			−0.009 **
				(−2.26)
	Cdigital				−0.001
					(−0.53)
	Size	−0.047 ***	−0.047 ***	−0.047 ***	−0.047 ***
		(−82.47)	(−82.82)	(−84.02)	(−83.84)
	ROA	−0.174 ***	−0.175 ***	−0.174 ***	−0.175 ***
		(−12.98)	(−13.03)	(−12.88)	(−12.85)
	ROE	−0.004	−0.004	−0.004	−0.004
		(−0.93)	(−0.89)	(−0.95)	(−0.92)
	Lev	0.030 ***	0.030 ***	0.030 ***	0.030 ***
		(8.21)	(8.18)	(8.38)	(8.34)
	Growth	−0.043 ***	−0.043 ***	−0.043 ***	−0.043 ***
		(−25.30)	(−25.35)	(−25.22)	(−25.31)
	TobinQ	0.002 ***	0.002 ***	0.002 ***	0.002 ***
		(5.56)	(5.57)	(5.75)	(5.66)
	Board	−0.004	−0.004	−0.005	−0.004
		(−1.55)	(−1.54)	(−1.57)	(−1.51)
	Big4	−0.004 *	−0.004 **	−0.004 *	−0.004 **
		(−2.01)	(−2.06)	(−1.98)	(−2.05)
	SOE
	Indep	<0.001	<0.001	<0.001	<0.001
		(−0.38)	(−0.37)	(−0.34)	(−0.29)
	FirmAge	0.004	0.004	0.004	0.004
		(1.50)	(1.43)	(1.48)	(1.31)
	Cashflow	−0.094 ***	−0.094 ***	−0.093 ***	−0.094 ***
		(−19.07)	(−19.05)	(−18.99)	(−18.78)
	wdigitaltrans	−0.001 **	−0.001 **	−0.001 **	−0.001 **
		(−2.12)	(−2.17)	(−2.19)	(−2.27)
	Cgdp	0.003	0.001	<0.001	−0.006***
		(0.62)	(0.14)	(−0.13)	(−3.11)
	year	YES	YES	YES	YES
	ind	YES	YES	YES	YES
	R2	88.49%	88.48%	88.49%	88.47%
	Adj R2	88.39%	88.38%	88.39%	88.37%
	N	10,822	10,822	10,822	10,822
	F-value	1078.871 ***	216.362 ***	206.040 ***	2833.742 ***
Panel B nonational
	Intercept	0.009	0.009	0.009	0.008
		(0.98)	(0.98)	(1.02)	(0.84)
	Cindex	−0.008 ***
		(−2.88)
	Ccoverage		−0.009 **
			(−2.41)
	Cusage_depth			−0.006 ***
				(−2.95)
	Cdigital				−0.001
					(−1.04)
	Size	−0.047 ***	−0.047 ***	−0.047 ***	−0.047 ***
		(−121.99)	(−122.73)	(−120.77)	(−120.82)
	ROA	−0.156 ***	−0.156 ***	−0.156 ***	−0.157 ***
		(−19.94)	(−19.90)	(−20.19)	(−21.07)
	ROE	<0.001	<0.001	<0.001	<0.001
		(−0.03)	(−0.02)	(−0.05)	(−0.01)
	Lev	0.021 ***	0.021 ***	0.021 ***	0.021 ***
		(10.93)	(10.89)	(10.93)	(10.70)
	Growth	−0.043 ***	−0.043 ***	−0.043 ***	−0.043 ***
		(−36.19)	(−36.25)	(−36.08)	(−36.25)
	TobinQ	0.002 ***	0.002 ***	0.002 ***	0.002 ***
		(11.01)	(10.88)	(11.13)	(11.04)
	Board	0.002	0.002	0.002	0.002
		(0.80)	(0.84)	(0.78)	(0.82)
	Big4	0.001	0.001	0.001	0.001
		(0.77)	(0.78)	(0.76)	(0.75)
	SOE
	Indep	<0.001 *	<0.001 *	<0.001 *	<0.001 *
		(1.82)	(1.87)	(1.76)	(1.80)
	FirmAge	0.004 ***	0.004 ***	0.004 ***	0.004 ***
		(3.25)	(3.26)	(3.29)	(3.35)
	Cashflow	−0.097 ***	−0.097 ***	−0.097 ***	−0.098 ***
		(−27.46)	(−27.36)	(−27.59)	(−27.53)
	wdigitaltrans	<0.001	<0.001	<0.001	<0.001
		(0.24)	(0.33)	(0.13)	(0.13)
	Cgdp	0.004	0.005	0.001	−0.003 **
		(1.21)	(1.22)	(0.48)	(−2.14)
	year	YES	YES	YES	YES
	ind	YES	YES	YES	YES
	R2	85.15%	85.15%	85.15%	85.13%
	Adj R2	85.06%	85.06%	85.05%	85.04%
	N	15,566	15,566	15,566	15,566
	F-value	3046.106 ***	7437.613 ***	335.405 ***	911.899 ***

Note: *, **, *** indicate significance at the 10% percent, 5% percent, and 1% percent levels, respectively.

reports the differences in the impact of digital finance on corporate financing constraints after grouping companies according to their ownership structure. From Figure 3, we can clearly see that the proportion of non-state-owned listed companies is quite high, so the financing situation of non-state-owned enterprises deserves our close attention. As shown in Panel A and Panel B of Table 8, the test results indicate that digital finance significantly mitigates the corporate finance covenant in both SOEs and non-SOEs. Digital finance has significantly alleviated financing constraints on non-state-owned enterprises, and there is no property rights discrimination in the use of digital finance to alleviate corporate financing constraints.

5.2.2. Impact Analysis of Heterogeneity Based on Industry

Figure 4 shows the full sample trend chart for enterprises in different industries, including industrial enterprises (Indus), service enterprises (Servi), and agricultural enterprises (Agric). As can be seen from the figure, industrial enterprises and service enterprises account for a large proportion of listed companies in China and are growing in number. The financing situation of industrial enterprises and service enterprises is also very important.

Table 9. Heterogeneity analysis based on different industries.

Group 2		(1)	(2)	(3)	(4)
Panel A Industry
	Intercept	0.010	0.009	0.011	0.007
		(0.95)	(0.92)	(1.05)	(0.74)
	Cindex	−0.009 **
		(−2.61)
	Ccoverage		−0.008 *
			(−1.92)
	Cusage_depth			−0.007 **
				(−2.67)
	Cdigital				<0.001
					(−0.19)
	Size	−0.047 ***	−0.047 ***	−0.047 ***	−0.047 ***
		(−81.39)	(−82.63)	(−80.91)	(−81.84)
	ROA	−0.175 ***	−0.175 ***	−0.174 ***	−0.175 ***
		(−20.46)	(−20.47)	(−20.51)	(−20.81)
	ROE	<0.001	<0.001	<0.001	<0.001
		(−0.09)	(−0.04)	(−0.12)	(−0.06)
	Lev	0.025 ***	0.026 ***	0.025 ***	0.025 ***
		(9.88)	(9.94)	(9.89)	(9.97)
	Growth	−0.043 ***	−0.043 ***	−0.043 ***	−0.043 ***
		(−41.79)	(−41.90)	(−41.83)	(−42.00)
	TobinQ	0.002 ***	0.002 ***	0.002 ***	0.002 ***
		(9.45)	(9.55)	(9.53)	(9.75)
	Board	−0.001	−0.001	−0.001	−0.001
		(−0.58)	(−0.56)	(−0.61)	(−0.56)
	Big4	−0.002	−0.002	−0.002	−0.002
		(−1.46)	(−1.46)	(−1.48)	(−1.50)
	SOE	<0.001	<0.001	<0.001	<0.001
		(−0.24)	(−0.15)	(−0.21)	(0.13)
	Indep	<0.001	<0.001	<0.001	<0.001
		(−0.05)	(−0.01)	(−0.07)	(0.04)
	FirmAge	0.004 **	0.004 **	0.004 **	0.004 **
		(2.56)	(2.57)	(2.55)	(2.53)
	Cashflow	−0.096 ***	−0.097 ***	−0.096 ***	−0.097 ***
		(−22.70)	(−22.55)	(−22.47)	(−22.06)
	wdigitaltrans	<0.001	<0.001	<0.001	<0.001
		(−0.95)	(−0.89)	(−1.08)	(−1.19)
	Cgdp	0.003	0.003	<0.001	−0.005 **
		(0.82)	(0.59)	(0.18)	(−2.11)
	year	YES	YES	YES	YES
	ind	YES	YES	YES	YES
	R2	87.31%	87.30%	87.30%	87.29%
	Adj R2	87.26%	87.26%	87.26%	87.24%
	N	18,920	18,920	18,920	18,920
	F-value	802.167 ***	597.111 ***	683.271 ***	755.259 ***
Panel B Services Industry
	Intercept	0.023	0.023	0.023	0.022
		(1.34)	(1.37)	(1.37)	(1.35)
	Cindex	−0.013 ***
		(−3.10)
	Ccoverage		−0.012 **
			(−2.44)
	Cusage_depth			−0.009 **
				(−2.52)
	Cdigital				−0.002
					(−0.98)
	Size	−0.047 ***	−0.047 ***	−0.047 ***	−0.047 ***
		(−90.16)	(−88.92)	(−89.57)	(−88.34)
	ROA	−0.106 ***	−0.106 ***	−0.106 ***	−0.107 ***
		(−8.74)	(−8.83)	(−8.74)	(−9.19)
	ROE	−0.016 ***	−0.016 ***	−0.016 ***	−0.016 ***
		(−3.11)	(−3.10)	(−3.12)	(−3.11)
	Lev	0.023 ***	0.023 ***	0.023 ***	0.023 ***
		(6.02)	(6.01)	(6.09)	(6.01)
	Growth	−0.044 ***	−0.044 ***	−0.044 ***	−0.044 ***
		(−35.36)	(−35.34)	(−35.40)	(−35.36)
	TobinQ	0.002 ***	0.002 ***	0.002 ***	0.002 ***
		(4.21)	(4.15)	(4.32)	(4.30)
	Board	<0.001	<0.001	<0.001	<0.001
		(0.02)	(0.06)	(0.03)	(0.13)
	Big4	−0.002	−0.002	−0.002	−0.002
		(−1.16)	(−1.14)	(−1.15)	(−1.09)
	SOE	−0.003 **	−0.003 **	−0.003 **	−0.003 **
		(−2.48)	(−2.50)	(−2.42)	(−2.31)
	Indep	<0.001	<0.001	<0.001	<0.001
		(0.93)	(0.95)	(0.93)	(0.95)
	FirmAge	0.006 ***	0.006 ***	0.006 ***	0.006 ***
		(3.75)	(3.59)	(3.82)	(3.56)
	Cashflow	−0.097 ***	−0.097 ***	−0.097 ***	−0.097 ***
		(−24.32)	(−24.37)	(−24.38)	(−24.37)
	wdigitaltrans	−0.001	−0.001	−0.001	−0.001
		(−1.44)	(−1.42)	(−1.51)	(−1.47)
	Cgdp	0.003	0.004	−0.002	−0.007 ***
		(0.78)	(0.70)	(−0.51)	(−4.11)
	year	YES	YES	YES	YES
	ind	YES	YES	YES	YES
	R2	87.81%	87.80%	87.80%	87.78%
	Adj R2	87.71%	87.71%	87.70%	87.68%
	N	7120	7120	7120	7120
	F-value	1546.361 ***	2996.937 ***	1627.783 ***	3821.828 ***
Panel C Agriculture
	Intercept	−0.018	−0.031	0.011	0.011
		(−0.27)	(−0.43)	(0.16)	(0.18)
	Cindex	−0.048 **
		(−2.14)
	Ccoverage		−0.047 ***
			(−3.17)
	Cusage_depth			−0.005
				(−0.35)
	Cdigital				0.014
					(1.29)
	Size	−0.045 ***	−0.044 ***	−0.046 ***	−0.046 ***
		(−11.92)	(−11.83)	(−12.62)	(−13.01)
	ROA	−0.252 ***	−0.261 ***	−0.245 ***	−0.248 ***
		(−6.10)	(−5.93)	(−6.52)	(−6.65)
	ROE	0.018 **	0.020 **	0.017 *	0.016 **
		(2.38)	(2.55)	(2.12)	(2.13)
	Lev	0.034 ***	0.035 ***	0.035 ***	0.034 ***
		(3.18)	(3.62)	(3.07)	(3.01)
	Growth	−0.035 ***	−0.035 ***	−0.035 ***	−0.034 ***
		(−9.09)	(−8.93)	(−8.70)	(−8.89)
	TobinQ	−0.001	−0.002	−0.002	−0.002
		(−0.79)	(−0.92)	(−0.95)	(−1.07)
	Board	−0.021	−0.022	−0.020	−0.021
		(−1.28)	(−1.37)	(−1.17)	(−1.22)
	Big4	−0.047 ***	−0.054 ***	−0.049 ***	−0.053 ***
		(−4.83)	(−6.12)	(−4.09)	(−5.02)
	SOE	−0.012 *	−0.013 **	−0.010	−0.009
		(−2.09)	(−2.37)	(−1.72)	(−1.48)
	Indep	<0.001	<0.001	<0.001	<0.001
		(0.18)	(0.15)	(0.14)	(−0.27)
	FirmAge	0.004	0.005	0.005	0.007
		(0.41)	(0.48)	(0.47)	(0.57)
	Cashflow	−0.042	−0.043	−0.045	−0.044
		(−1.29)	(−1.35)	(−1.26)	(−1.25)
	wdigitaltrans	0.006 **	0.006 **	0.005 **	0.005 **
		(2.73)	(2.39)	(2.35)	(2.42)
	Cgdp	0.055 **	0.058 ***	0.014	0.007
		(2.44)	(3.55)	(1.28)	(1.06)
	year	YES	YES	YES	YES
	ind	YES	YES	YES	YES
	R2	88.56%	88.66%	88.34%	88.54%
	Adj R2	87.56%	87.67%	87.32%	87.54%
	N	348	348	348	348
	F-value	334.513 ***	41.094 ***	193.218 ***	1067.947 ***

Note: *, **, *** indicate significance at the 10% percent, 5% percent, and 1% percent levels, respectively.

reports the results of testing the differences in the impact of digital finance on corporate financing constraints across different industries based on the classification of the industries in which the sample companies operate. From the sample trends in Figure 4, we can see that industrial enterprises account for a relatively larger proportion of Chinese listed companies. Therefore, we need to pay further attention to the extent to which digital finance alleviates financing constraints for industrial enterprises and service enterprises. As shown Panel A, Panel B, and Panel C in Table 9, the results show that digital finance can significantly alleviate corporate financing constraints in industrial enterprises, service industries, and agribusinesses. This role is most empowering in service sector firms and relatively weak in agriculture

5.2.3. Heterogeneous Impact Analysis Based on High-Carbon- Versus Low-Carbon-Emitting Industries

Figure 5 shows the overall trends for high-carbon-emission enterprises (High) and low-carbon-emission enterprises (Lowc). As can be seen from the figure, high-carbon-emission enterprises still account for a relatively high proportion of Chinese enterprises and are showing a continuing upward trend. Therefore, the transformation, upgrading, and financing of high-carbon enterprises are crucial to the sustainable development of Chinese enterprises.

Table 10. Heterogeneity analysis based on different levels of carbon emissions.

Group 3		(1)	(2)	(3)	(4)
Panel A High-Carbon-Emission Industries
	Intercept	0.017	0.017	0.014	0.018
		(1.50)	(1.47)	(1.28)	(1.57)
	Ccindex	−0.003 *
		(−1.95)
	Ccoverage		−0.005 **
			(−2.14)
	Cusage_depth			<0.001
				(−0.22)
	Cdigital				−0.004 ***
					(−3.22)
	Size	−0.047 ***	−0.047 ***	−0.047 ***	−0.047 ***
		(−83.97)	(−82.65)	(−82.96)	(−85.25)
	ROA	−0.175 ***	−0.174 ***	−0.175 ***	−0.174 ***
		(−21.00)	(−20.74)	(−21.03)	(−22.22)
	ROE	<0.001	<0.001	<0.001	<0.001
		(0.11)	(0.04)	(0.04)	(0.03)
	Lev	0.026 ***	0.025 ***	0.025 ***	0.025 ***
		(10.33)	(10.14)	(9.99)	(9.95)
	Growth	−0.043 ***	−0.043 ***	−0.043 ***	−0.043 ***
		(−43.58)	(−43.66)	(−43.61)	(−43.72)
	TobinQ	0.002 ***	0.002 ***	0.002 ***	0.002 ***
		(10.29)	(10.31)	(10.41)	(10.08)
	Board	−0.002	−0.002	−0.001	−0.002
		(−0.72)	(−0.70)	(−0.65)	(−0.71)
	Big4	−0.002	−0.002	−0.002	−0.002
		(−1.41)	(−1.40)	(−1.48)	(−1.48)
	SOE	<0.001	<0.001	<0.001	<0.001
		(−0.04)	(−0.18)	(−0.06)	(−0.42)
	Indep	<0.001	<0.001	<0.001	<0.001
		(0.11)	(0.07)	(0.12)	(0.12)
	FirmAge	0.004 **	0.004 **	0.004 **	0.004 ***
		(2.53)	(2.58)	(2.62)	(2.77)
	Cashflow	−0.097 ***	−0.097 ***	−0.097 ***	−0.096 ***
		(−22.38)	(−22.40)	(−22.20)	(−22.54)
	wdigitaltrans	<0.001	<0.001	−0.001	<0.001
		(−1.38)	(−1.42)	(−1.57)	(−1.41)
	Cgdp	−0.003	−0.002	−0.005 **	−0.004 **
		(−1.47)	(−0.82)	(−2.08)	(−2.45)
	year	YES	YES	YES	YES
	ind	YES	YES	YES	YES
	R2	87.56%	87.57%	87.56%	87.58%
	Adj R2	87.52%	87.52%	87.51%	87.54%
	N	19,757	19,757	19,757	19,757
	F-value	612.534 ***	1344.707 ***	1090.377 ***	2216.489 ***
Panel B Low-carbon-emission industries
	Intercept	0.041 **	0.042 **	0.040 **	0.041 **
		(2.48)	(2.49)	(2.37)	(2.36)
	Cindex	−0.006 **
		(−2.41)
	Ccoverage		−0.012 ***
			(−3.44)
	Cusage_depth			−0.005 **
				(−2.18)
	Cdigital				−0.004
					(−1.39)
	Size	−0.048 ***	−0.048 ***	−0.048 ***	−0.048 ***
		(−93.61)	(−93.83)	(−93.31)	(−90.80)
	ROA	−0.114 ***	−0.114 ***	−0.115 ***	−0.115 ***
		(−9.83)	(−9.58)	(−9.97)	(−9.98)
	ROE	−0.013 **	−0.012 **	−0.012 **	−0.012 **
		(−2.40)	(−2.36)	(−2.38)	(−2.27)
	Lev	0.022 ***	0.022 ***	0.022 ***	0.022 ***
		(5.65)	(5.73)	(5.58)	(5.59)
	Growth	−0.044 ***	−0.044 ***	−0.044 ***	−0.044 ***
		(−35.05)	(−35.18)	(−34.95)	(−34.97)
	TobinQ	0.001 ***	0.001 ***	0.001 ***	0.001 ***
		(3.15)	(3.27)	(3.26)	(3.15)
	Board	−0.001	−0.001	−0.001	−0.001
		(−0.17)	(−0.24)	(−0.19)	(−0.25)
	Big4	−0.003	−0.003	−0.003	−0.003
		(−1.37)	(−1.49)	(−1.33)	(−1.35)
	SOE	−0.003 ***	−0.003 ***	−0.004 ***	−0.004 ***
		(−2.83)	(−3.01)	(−2.85)	(−2.92)
	Indep	<0.001	<0.001	<0.001	<0.001
		(1.03)	(1.01)	(0.96)	(0.90)
	FirmAge	0.006 ***	0.006 ***	0.006 ***	0.006 ***
		(2.88)	(3.15)	(2.96)	(2.93)
	Cashflow	−0.096 ***	−0.095 ***	−0.095 ***	−0.095 ***
		(−22.21)	(−22.14)	(−22.97)	(−22.14)
	wdigitaltrans	−0.001	−0.001	−0.001	−0.001
		(−0.59)	(−0.65)	(−0.68)	(−0.64)
	Cgdp	−0.006 **	−0.003	−0.007 ***	−0.008 ***
		(−2.38)	(−1.02)	(−2.78)	(−3.13)
	year	YES	YES	YES	YES
	ind	YES	YES	YES	YES
	R2	87.07%	87.11%	87.07%	87.07%
	Adj R2	86.97%	87.01%	86.97%	86.97%
	N	6631	6631	6631	6631
	F-value	1044.360 ***	2848.043 ***	23,179.985 ***	13,132.190 ***

Note: *, **, *** indicate significance at the 10% percent, 5% percent, and 1% percent levels, respectively.

reports on the differences in the impact of digital finance on corporate financing constraints between high-carbon- and low-carbon-emitting companies, based on the level of carbon emissions in the sample. As shown in the sample trends in Figure 5, high-carbon-emitting enterprises account for a relatively larger proportion of listed companies in China. Therefore, the extent to which digital finance alleviates financing constraints for high-carbon-emitting enterprises warrants further attention. This is critical for the transformation and upgrading of high-carbon-emitting enterprises as well as their future ability to enhance their green transformation capabilities. As shown in Panel A and Panel B in Table 10, the test results show that whether in high- or low-carbon-emitting sectors, digital finance works well to alleviate financing constraints, and this role is better played in low-carbon-emitting sectors.

5.3. An Analysis of the Impact of Different Contexts on Digital Finance, Management Power, and Financing Constraints

5.3.1. Impact Analysis of Heterogeneity Based on the Nature of Firms’ Property Rights

As shown in columns (1) and (2) of

Table 11. A group research on the regulatory role of management power in digital finance and corporate financing constraints.

Variables	Nature of Company Ownership		Industry Developments		Degree of Regional Development
	State-Owned Enterprises	Non-State-Owned Enterprises	High-Tech Industries	Traditional Industries	Highly Developed Regions	Low-Development Regions
	(1)	(2)	(3)	(4)	(5)	(6)
Intercept	4.174 ***	8.662 ***	7.061 ***	5.713 ***	−0.833	7.003 ***
	(7.24)	(8.52)	(4.94)	(8.68)	(−0.39)	(8.45)
Cindex	−0.333 ***	−0.699 ***	−0.598 ***	−0.463 ***	0.408	−0.456 ***
	(−3.06)	(−3.80)	(−3.28)	(−3.08)	(1.09)	(−4.50)
Cindex*Power	0.198 ***	0.354 ***	0.075	0.371 ***	0.390 **	0.228 ***
	(3.50)	(3.64)	(0.72)	(4.81)	(2.50)	(3.27)
	(−4.15)	(−3.79)	(−1.03)	(−5.03)	(−2.78)	(−3.51)
size	−0.066	−0.179 ***	−0.159 **	−0.122 *	−0.026	−0.190 ***
	(−1.37)	(−2.60)	(−2.16)	(−1.87)	(−0.45)	(−3.44)
roa	−0.067 ***	−0.005	−0.056 ***	−0.006	−0.002	−0.048 ***
	(−5.00)	(−1.03)	(−6.72)	(−1.05)	(−0.66)	(−10.50)
lev	0.032 ***	0.034 ***	0.027 ***	0.038 ***	0.025 ***	0.040 ***
	(4.10)	(3.99)	(3.28)	(4.52)	(3.21)	(5.61)
growth	<0.001 ***	<0.001 ***	−0.004	<0.001 *	0.003 **	<0.001 *
	(5.05)	(6.70)	(−0.28)	(1.85)	(2.10)	(1.81)
SOE			0.257 ***	0.232 ***	0.039	0.243 ***
			(4.19)	(6.24)	(0.67)	(6.07)
TobinQ	0.076 **	0.008 **	0.005	0.046 *	0.045 *	0.008
	(2.42)	(2.48)	(1.58)	(1.81)	(1.91)	(1.60)
board	−0.016 **	−0.072 ***	−0.014	−0.038 ***	−0.030 **	−0.024 ***
	(−2.03)	(−4.27)	(−0.76)	(−3.94)	(−1.98)	(−2.76)
big4	−0.132 **	−0.311 ***	−0.069	−0.176 ***	−0.024	−0.296 ***
	(−2.06)	(−3.16)	(−0.54)	(−2.68)	(−0.34)	(−3.68)
Indep	0.552 *	0.427	1.286 ***	0.752 ***	0.817 **	0.685 ***
	(1.96)	(1.03)	(2.63)	(2.80)	(2.01)	(2.64)
FirmAge	0.004 **	0.004 **	0.004 **	0.004 ***	0.006 ***	0.006 ***
	(2.53)	(2.58)	(2.62)	(2.77)	(3.82)	(3.56)
Cashflow	−0.097 ***	−0.097 ***	−0.097 ***	−0.096 ***	−0.097 ***	−0.097 ***
	(−22.38)	(−22.40)	(−22.20)	(−22.54)	(−24.38)	(−24.37)
wdigitaltrans	<0.001	<0.001	−0.001	<0.001	−0.001	−0.001
	(−1.38)	(−1.42)	(−1.57)	(−1.41)	(−1.51)	(−1.47)
Cgdp	−0.006 **	−0.003	−0.007 ***	−0.008 ***	−0.006 **	−0.003
	(−2.38)	(−1.02)	(−2.78)	(−3.13)	(−2.38)	(−1.02)
year	−0.003	−0.002	−0.005 **	−0.004 **	−0.002	−0.007 ***
ind	(−1.47)	(−0.82)	(−2.08)	(−2.45)	(−0.51)	(−4.11)
R2	55.09%	40.48%	45.33%	47.00%	36.42%	55.46%
AdjR2	54.91%	40.27%	45.01%	46.87%	36.05%	55.34%
N	7744	8881	4406	12,219	5390	11,235
F-value	582.628 ***	150.830 ***	79.539 ***	216.664 ***	61.860 ***	240.626 ***

Note: *, **, *** indicate significance at the 10% percent, 5% percent, and 1% percent levels, respectively.

, the test results indicate that management power will inversely regulate the effects of digital finance on corporate finance constraints in both state-owned and non-state-owned enterprises. This inverse regulation is stronger in non-state-owned enterprises.

5.3.2. Impact Analysis of Heterogeneity Based on Industry Development

As shown in columns (3) and (4) of Table 11, the test results show that in conventional firms, management power has a significant reverse moderating effect on digital finance for corporate finance constraints.

5.3.3. Heterogeneity Impact Analysis Based on Regional Development

As shown in columns (5) and (6) of Table 11, the results show that the management power has a considerable inverse moderating effect whether in the highly developed regions or the local development regions. This moderating effect will be more substantial in the low-development regions.

5.4. An Analysis of the Impact of Different Contexts on Digital Finance, Corporate Social Responsibility, and Financing Constraints

Table 12. A group research on the regulatory role of CSR performance in digital finance and corporate financing constraints.

Variables	Nature of Company Ownership		Industry Developments		Level of Regional Development
	Nationalised Business	Non-State Enterprise	High-Tech Industries	Traditional Industries	Highly Developed Regions	Low-Development Regions
	(1)	(2)	(3)	(4)	(5)	(6)
Intercept	1.678 **	0.825	3.493 **	−0.008	−7.692 ***	2.932 ***
	(2.33)	(0.67)	(2.31)	(−0.01)	(−3.68)	(2.74)
Cindex	−0.001	0.244	−0.136	0.246 **	1.339 ***	0.097
	(−0.01)	(1.60)	(−0.77)	(2.11)	(3.97)	(1.04)
Cindex*csr	−0.007 ***	−0.026 ***	−0.015 ***	−0.017 ***	−0.024 ***	−0.014 ***
	(−5.52)	(−9.40)	(−5.17)	(−10.63)	(−8.04)	(−8.74)
csr	0.023 ***	0.100 ***	0.051 ***	0.060 ***	0.098 ***	0.049 ***
	(3.83)	(7.39)	(3.50)	(8.33)	(6.52)	(6.90)
size	−0.034	−0.032	−0.111	−0.022	0.063	−0.137 **
	(−0.65)	(−0.45)	(−1.48)	(−0.31)	(1.05)	(−2.32)
roa	−0.059 ***	−0.003	−0.036 ***	−0.004	−0.001	−0.035 ***
	(−4.28)	(−0.78)	(−4.28)	(−0.97)	(−0.44)	(−7.65)
lev	0.033 ***	0.032 ***	0.028 ***	0.036 ***	0.024 ***	0.040 ***
	(4.27)	(3.96)	(3.50)	(4.49)	(3.22)	(5.99)
growth	<0.001 ***	<0.001 ***	−0.008	<0.001	0.002 *	<0.001
	(4.47)	(4.09)	(−0.62)	(1.45)	(1.87)	(1.23)
SOE			0.375 ***	0.259 ***	0.183 ***	0.264 ***
			(5.94)	(7.05)	(3.15)	(6.89)
TobinQ	0.087 ***	0.011 **	0.007 **	0.053 **	0.050 **	0.010 *
	(2.72)	(2.57)	(2.03)	(2.17)	(2.21)	(1.79)
board	−0.019 **	−0.056 ***	−0.005	−0.038 ***	−0.036 **	−0.020 **
	(−2.39)	(−3.58)	(−0.31)	(−4.10)	(−2.52)	(−2.37)
big4	−0.110 *	−0.242 ***	0.087	−0.145 **	0.060	−0.279 ***
	(−1.81)	(−2.61)	(0.68)	(−2.31)	(0.89)	(−3.63)
Indep	0.511 *	0.621	1.298 ***	0.626 **	0.540	0.802 ***
	(1.89)	(1.59)	(2.74)	(2.42)	(1.39)	(3.13)
FirmAge	0.004 **	0.004 **	0.004 **	0.004 ***	0.004 **	0.004 **
	(2.53)	(2.58)	(2.62)	(2.77)	(2.53)	(2.58)
Cashflow	−0.097 ***	−0.097 ***	−0.097 ***	−0.096 ***	−0.097 ***	−0.097 ***
	(−22.38)	(−22.40)	(−22.20)	(−22.54)	(−22.38)	(−22.40)
wdigitaltrans	<0.001	<0.001	−0.001	<0.001	<0.001	<0.001
	(−1.38)	(−1.42)	(−1.57)	(−1.41)	(−1.38)	(−1.42)
Cgdp	−0.003	−0.002	−0.005 **	−0.004 **	−0.003	−0.002
	(−1.47)	(−0.82)	(−2.08)	(−2.45)	(−1.47)	(−0.82)
year	YES	YES	YES	YES	YES	YES
ind	YES	YES	YES	YES	YES	YES
R2	56.56%	46.38%	47.57%	50.93%	41.42%	57.35%
AdjR2	56.40%	46.20%	47.28%	50.82%	41.09%	57.24%
N	8197	9041	4494	12,744	5559	11,679
F-value	735.626 ***	191.907 ***	103.356 ***	274.762 ***	83.831 ***	306.446 ***

Note: *, **, *** indicate significance at the 10% percent, 5% percent, and 1% percent levels, respectively.

columns (1) to (6) show that at the 1% significance level, the cross-multipliers of digital finance and CSR are all significantly negative in each subgroup. CSR plays a significant homoscedastic moderating effect in different subgroups.

5.5. A Test of the Economic Consequences of Green Innovation for Firms

In the context of the global pursuit of sustainable economic development, Chinese enterprises are also actively practicing sustainable and high-quality development. The core of high-quality corporate development is corporate green innovation. Can the synergy between digital finance and corporate finance constraints promote green innovation? We conducted a further test, as the measurement method of green innovation has not yet formed a unified international standard, we borrowed the research method of existing literature [43] and determined the number of green invention patent applications filed by enterprises based on the International Patent Classification (IPC) code in the International Patent Classification (IPC) Green List. Based on the above discussion, finding suitable metrics to directly measure landmark green innovation and materiality is challenging. Therefore, we describe two types of innovation behaviours from the perspective of innovation effects based on the definition of Chinese patent law and existing research literature [44]. Enterprises applying for “high-quality” green invention patents are recognised as having substantive green innovation (Gpi1), while those applying for green utility model patents are recognised as having iconic green innovation (Gpu1). The number of green design patents filed is not considered, as design patents are the most basic innovation with relatively low technological content; the filing process does not require the submission of reports and substantive examination, which is a more independent behaviour.

After selecting the indicators, when testing the synergy between digital finance and corporate finance constraints, as firms in developed regions are likely to operate more efficiently, we added regional urbanisation level (urban) to the original control variables and urban industrialisation level (industriallevel) using year and industry fixed effects and provincial clustering standard errors to ensure the rigour of the regression.

The results of the empirical tests show that digital finance has a significant role in promoting corporate signature green innovation, as can be understood from the report in

Table 13. Impact of synergies between digital finance and corporate finance constraints on signature green innovation.

Variables	(1)	(2)	(3)	(4)
Intercept	−2.154 ***	−2.178 ***	−1.809 ***	−1.310 ***
	(−7.21)	(−5.59)	(−4.07)	(−4.90)
Cindex	0.953 **
	(2.04)
Ccoverage		1.184 ***
		(3.25)
Cusage_depth			0.555 *
			(1.86)
Cdigital				−0.255
				(−0.77)
Cindex*WW	0.847 *
	(1.74)
Ccoverage*WW		1.144 ***
		(3.03)
Cusage_depth*WW			0.527 *
			(1.75)
Cdigital*WW				−0.360
				(−1.08)
WW	−0.679 *	−0.730 **	−0.369	0.063
	(−1.97)	(−2.66)	(−1.59)	(0.21)
Size	0.076 ***	0.076 ***	0.075 ***	0.075 ***
	(5.02)	(5.00)	(4.77)	(4.75)
ROA	0.169	0.176	0.189	0.181
	(1.16)	(1.20)	(1.27)	(1.20)
Lev	0.085 *	0.085 *	0.084 *	0.093 *
	(1.87)	(1.86)	(1.83)	(2.03)
Growth	−0.035 **	−0.035 **	−0.035 **	−0.033 **
	(−2.72)	(−2.75)	(−2.67)	(−2.56)
TobinQ	0.006	0.006	0.006	0.005
	(1.69)	(1.67)	(1.58)	(1.56)
Board	0.064	0.061	0.061	0.062
	(1.42)	(1.35)	(1.35)	(1.39)
Big4	0.099 *	0.101 *	0.096 *	0.094
	(1.78)	(1.79)	(1.73)	(1.68)
Dual	0.033	0.032	0.033	0.033
	(1.08)	(1.07)	(1.11)	(1.11)
SOE	0.020	0.022	0.020	0.023
	(0.92)	(1.05)	(0.96)	(1.12)
FirmAge	−0.099 **	−0.101 **	−0.100 **	−0.101 **
	(−2.46)	(−2.54)	(−2.49)	(−2.53)
ROE	0.059	0.061 *	0.061	0.060 *
	(1.68)	(1.73)	(1.70)	(1.72)
Cashflow	0.010	0.006	0.010	0.008
	(0.09)	(0.05)	(0.09)	(0.08)
wdigitaltrans	−0.007	−0.006	−0.005	−0.007
	(−0.71)	(−0.62)	(−0.55)	(−0.66)
Gdp	0.001	0.002	0.004	−0.029
	(0.01)	(0.02)	(0.04)	(−0.40)
Urban	−0.172	−0.169	−0.155	−0.136
	(−0.98)	(−0.99)	(−0.91)	(−1.07)
Industrylevel	0.187 *	0.160	0.181 *	−0.084
	(1.71)	(1.40)	(1.77)	(−0.70)
Cgdp	0.001	0.075	0.073	0.020
	(0.02)	(1.43)	(1.33)	(0.35)
year	YES	YES	YES	YES
ind	YES	YES	YES	YES
R2	17.26%	17.24%	17.16%	17.30%
Adj R2	16.92%	16.89%	16.82%	16.96%
N	26,381	26,381	26,381	26,381
F-value	130.795 ***	129.372 ***	93.856 ***	148.687 ***

Note: *, **, *** indicate significance at the 10% percent, 5% percent, and 1% percent levels, respectively.

. In contrast to the empirical results in

Table 14. Impact of synergies between digital finance and corporate finance constraints on firms’ substantive green innovation.

Variables	(1)	(2)	(3)	(4)
Intercept	−2.327 ***	−2.267 ***	−1.919 ***	−1.343 ***
	(−4.91)	(−4.51)	(−3.33)	(−4.53)
Cindex	0.679
	(1.67)
Ccoverage		0.720
		(1.54)
Cusage_depth			0.107
			(0.17)
Cdigital				−0.656 **
				(−2.10)
Cindex*WW	0.666
	(1.59)
Ccoverage*WW		0.729
		(1.48)
Cusage_Depth*WW			0.073
			(0.12)
Cdigital*WW				−0.824 **
				(−2.66)
WW	−1.245 ***	−1.201 ***	−0.847 ***	−0.360
	(−3.88)	(−4.06)	(−3.29)	(−1.20)
Size	0.062 **	0.061 **	0.061 **	0.063 **
	(2.62)	(2.65)	(2.53)	(2.66)
ROA	0.361 *	0.365 *	0.367 *	0.370 *
	(1.80)	(1.83)	(1.85)	(1.80)
Lev	0.026	0.025	0.024	0.035
	(0.39)	(0.39)	(0.36)	(0.52)
Growth	−0.065 ***	−0.065 ***	−0.064 ***	−0.062 ***
	(−4.80)	(−4.82)	(−4.78)	(−4.80)
Tobinq	0.026 ***	0.026 ***	0.026 ***	0.024 ***
	(4.86)	(4.64)	(4.44)	(4.50)
Board	0.069	0.068	0.069	0.074
	(0.72)	(0.71)	(0.72)	(0.77)
Big4	0.198 **	0.199 ***	0.194 **	0.197 **
	(2.75)	(2.77)	(2.73)	(2.67)
Indep	0.002	0.002	0.002	0.002
	(1.02)	(1.03)	(1.02)	(0.97)
SOE	0.021	0.021	0.021	0.029
	(0.57)	(0.60)	(0.57)	(0.80)
Firmage	−0.161 ***	−0.160 ***	−0.162 ***	−0.167 ***
	(−4.12)	(−4.16)	(−4.21)	(−4.29)
ROE	−0.032	−0.032	−0.033	−0.035
	(−0.52)	(−0.51)	(−0.52)	(−0.57)
Cashflow	−0.203	−0.203	−0.198	−0.204
	(−1.48)	(−1.51)	(−1.48)	(−1.48)
Wdigitaltrans	0.047 ***	0.047 ***	0.046 ***	0.045 ***
	(9.57)	(9.03)	(9.24)	(9.76)
Gdp	0.132	0.147	0.123	0.089
	(1.37)	(1.53)	(1.44)	(1.43)
Urban	−0.381 *	−0.387 **	−0.356 *	−0.362 **
	(−1.94)	(−2.18)	(−1.92)	(−2.43)
Industrylevel	0.199	0.210	0.213	−0.210
	(1.23)	(1.17)	(1.33)	(−1.45)
Cgdp	0.140 **	0.149 ***	0.123 **	0.038
	(2.65)	(2.85)	(2.31)	(0.64)
R2	6.87%	6.87%	6.84%	7.22%
Adj R2	6.79%	6.80%	6.77%	7.15%
N	26,381	26,381	26,381	26,381
F-Value	619.892 ***	440.918 ***	318.435 ***	351.201 ***

Note: *, **, *** indicate significance at the 10% percent, 5% percent, and 1% percent levels, respectively.

, digital finance does not have any facilitating effect on firms’ substantive green innovation

6. Discussion

Taking Chinese A-share-listed companies from 2011 to 2020 as research samples, this paper empirically examines the role of digital finance on corporate financing constraints and its impact on corporate green innovation and explores the issue from the perspectives of management power and corporate social responsibility at the micro level of enterprises. The empirical results show that digital finance can indeed alleviate corporate financing constraints, but the synergistic effect of the two has a “quantitative and qualitative separation” effect on corporate green innovation, which only promotes the enhancement of iconic green innovation and has no obvious effect on substantive green innovation. The power of the management at the corporate level and the CSR have different moderating effects on the empowerment garnered from digital financing to alleviate financing constraints.

First, the hypothesis proposed in this study is confirmed by analyzing the results of empirical tests: Digital finance can indeed significantly alleviate corporate financing constraints. A 1% increase in the composite digital finance index at the city level is associated with a 0.008% decrease in the level of financing constraints of firms, with a significance level of 1%. From the heterogeneity analysis, it can be seen that digital financial empowerment alleviates enterprise financing constraints without property rights discrimination. Specifically, for non-state-owned firms, a 1% increase in the total digital finance index is associated with a 0.008% reduction in the firm’s financing constraints, with a significance level of 1%. This is also consistent with the findings of existing studies on the financing difficulties of SMEs [10]. The results of this study’s test of the financing constraints of state-owned enterprises show that the mitigating effect on the financing constraints of state-owned enterprises is equally significant. For state-owned enterprises, a 1% increase in the total digital finance index is associated with a 0.011% decrease in the level of corporate financing constraints, with a significance level of 1%. In the heterogeneity analysis for industries, we find that digital finance alleviates financing constraints most significantly in the service sector and relatively weakly in agriculture and that this relief effect is stronger in low-carbon-emission industries compared to high-carbon-emission industries. The conclusions provide additional grounds for further promoting the policy of digital financial inclusion and empowerment. For further transformation, upgrading, and high-quality sustainable development of enterprises, further promotion of digital finance is necessary.

Second, this study attempts to explore the specific moderating effects of both management power and CSR on digitally enabled enterprises, based on the micro-level of enterprises.This is different from most studies on the impact of digitally empowering enterprises at the micro-level, most of which focus on enterprise financing issues, the low-carbon energy transition of enterprises, and ESG [10,12,16,17] with an overall path mainly exploring the external financing environment or financing mitigation without focusing on the internal aspects of enterprises. In this paper, we investigate from the perspectives of management power and corporate social responsibility (CSR) to explore how different degrees of management power and different manifestations of CSR in enterprises at the enterprise level respond to such a role when macro-factors of digital finance are empowered to alleviate the financing constraints of enterprises. These two aspects are explored based on the fact that China’s current financial system regulatory mechanism is yet to be perfected, digital finance has the characteristic attributes of financial risk, and regulation always lags behind behaviour; therefore, to solve the problem of corporate financing constraints while relying on digital finance, at the enterprise level, reasonable control of management power and improvement of the performance of corporate social responsibility are effective paths that can be followed in order to reduce the existing regulatory dependence, achieve the combination of ‘internal and external’ dual paths, alleviate the difficulty of corporate financing, and help enterprises transform, upgrade, and develop.

Finally, based on the concept of sustainable development, green innovation is a key factor in the sustainable development of enterprises. Within the framework of China’s pursuit of a green and low-carbon economic trajectory, the significance of companies’ pursuit of green development has become critical. Green innovation is a key means to revitalise the ecosystem and provide an indispensable technological base for sustainable economic development [1,51,52,53]. The definition of green innovation remains the subject of an ongoing academic debate, and there is a lack of unified understanding. Green innovation is also known as “environmental innovation” or “eco-innovation” [54]. Green innovation has obvious double externalities, promoting green transformation and upgrading of enterprises through technological spillovers while positively affecting environmental protection and citizens’ quality of life [55]. Dian et al. (2024) emphasise that green innovation refers to the enhancement of product design through the use of environmentally friendly products or processes, minimising product life cycle stages, reducing negative impacts on the environment, or achieving sustainable development [56]. It can be seen that the ultimate goal of green innovation in enterprises is to reduce or even eliminate the negative impacts on the ecological environment during the production and operation processes. Overall, green innovation is ultimately an important part of corporate sustainable development. Green innovation itself requires a large amount of technology, financial support, and also may further increase the demand for human capital, which has been proven to play a key role in modern economic growth [57,58], hence the synergies between the digital finance and corporate finance constraints The role of digital finance and corporate finance constraints on the economic outcomes of corporate green innovation is further tested. While most tests of the economic consequences of green innovation for firms have focused on tests of green innovation capabilities or the overall performance outcomes of green innovation [1,59,60], with uniformity in the test metrics, this study focuses on dividing the test of the outcomes of green innovation into two dimensions, iconic green innovation and substantive green innovation in order to clarify whether there is divergence, or harmonisation, in the economic consequences. In contrast to previous studies, this paper attempts to go further by testing whether digital finance’s alleviation of financing constraints promotes firms’ green innovation in the true sense of the word.

The results of the substantive tests show a “quantitative and qualitative separation” effect: The overall contribution to iconic green innovation, but not to substantive green innovation, is significant. This difference is very significant, which indicates that digital finance has indeed eased the difficulties of enterprise financing, but the promotion of real green innovation is not obvious, which is a new problem that enterprises need to pay attention to in the pursuit of high-quality development nowadays. Enterprises should be alert to whether they have neglected the improvement of their own green innovation capability and blindly pursued the improvement of “superficial” green innovation. After all, in the current context of corporate social responsibility and sustainable development, it is important to show that a better green innovation answer sheet will win more support from enterprises and managers in various aspects.

Research has confirmed that digital finance alleviates financing constraints and has a “quantitative and qualitative separation” effect on green innovation, but what exactly is the reason for this result? We will explore this from a theoretical perspective. Based on management power theory and agency theory [21,29], one of the reasons for this is that managers with powerful financial discipline will send signals to stakeholders by pursuing a large number of low-quality, easily disclosed innovations in order to gain more support. Research has confirmed that managerial authority significantly undermines corporate innovation [33] and that managerial power is significantly negatively correlated with green innovation [34]. There are also other corporate-level factors that contribute to this outcome, such as the characteristics of the company itself. Digital finance is more favourable to younger, smaller, or riskier firms [10], which lack substantial long-term, high-cost research and development resources and have opted for the lower-cost and more convenient minor utility patents. This has led to a phenomenon of “quantity–quality separation” in corporate green innovation.

The limitations of our study lie in the fact that, as mentioned above, the factors contributing to the “disconnect between quantity and quality” effect on corporate green innovation resulting from the synergistic interaction between digital finance and corporate financing constraints are the result of multiple factors acting in concert. Whether there are other factors contributing to this outcome requires further analysis and research, which this study does not address.

7. Conclusions

This paper empirically examines the effect of digital finance on corporate financing constraints and its impact on corporate green innovation using a sample of A-share-listed companies in China from 2011 to 2020 and explores the specific impacts on this effect from the perspectives of managerial power and corporate social responsibility (CSR) at the micro-level of firms. The key findings are as follows:

First, the study confirms through substantive tests that the hypotheses in the paper are valid and that digital finance can significantly empower the alleviation of financing constraints, which is weakened by an increase in management power and strengthened by an increase in CSR performance. Heterogeneity analyses reveal that digital finance empowers the alleviation of corporate finance constraints without property rights discrimination, and this alleviation effect is significant in both state-owned and non-state-owned firms. Across industries, this mitigating effect is most pronounced in services, relatively weaker in agriculture, and more pronounced in low-carbon-emitting industries than in high-carbon-emitting industries. The conclusion confirms that digital finance can empower the long tail in a balanced way, but there are some relative barriers to empowerment in high-carbon-emitting sectors and agriculture.

In the mechanistic study, it is confirmed that empowerment from digital finance can be inhibited as management power increases and promoted as CSR performance improves. Heterogeneity in the moderating effect reveals that the inverse moderating effect of managerial power is stronger in traditional firms versus firms in low-development regions. CSR significantly contributes to digital financial empowerment to alleviate corporate finance constraints in both state-owned and non-state-owned firms, traditional and emerging industries, and high- and low-development regions. The difference between these two moderating effects also illustrates the importance of rationally controlling management’s power and promoting corporate social responsibility for digital financial empowerment to alleviate corporate finance constraints. This enriches the research on digital finance, management power, and CSR at the firm level based on principal–agent theory and sustainable development theory.

In the economic consequences test, it is found that digital financial empowerment to alleviate corporate financing constraints has a “quantitative and qualitative separation” effect on green innovation, which only promotes the promotion of iconic green innovations, but not the promotion of green invention patents, which are the most valuable innovations. This is a new problem that enterprises need to be alerted to. Blindly pursuing “superficial” green innovation and chasing “superficial” green answers are not conducive to the high-quality sustainable development of enterprises in the long run. The results enrich the research on the economic consequences of digital finance and corporate finance from the perspective of sustainable development.

Overall, these findings enrich the research in related fields; provide empirical evidence and theoretical foundations for the further promotion of digital finance policies, the rational control of management power, the enhancement of corporate social responsibility and the improvement of the green innovation capacity of enterprises; and provide more actionable solutions for the high-quality and sustainable development of enterprises.

For future research directions, first, scholars studying this field in the future may focus on whether other corporate factors, managerial characteristics, or external macroeconomic factors contribute to this phenomenon. Second, scholars could investigate whether there are corresponding solutions, or whether other corporate-level factors may enable corporate green innovation to achieve “unity of knowledge and action” and “consistency between quantity and quality”, thereby providing a solution for genuinely enhancing corporate green innovation and high-quality sustainable development.