Navigating Green Innovation in High-Tech Manufacturing: The Roles of Customer Concentration and Digital Transformation

Abstract

The increasingly environmental issues pose challenges to the economic development of countries, particularly hindering industrial transformation in developing nations. This study, grounded in the Resource-Based View, examines factors influencing green innovation in high-tech manufacturing firms. Market interactions and digital technologies significantly impact resource investments in green innovation. Using data from Chinese high-tech manufacturing firms from 2007 to 2021, the study reveals that customer concentration negatively affects green innovation, while digital transformation promotes it and mitigates the inhibitory effect of customer concentration. To explain this mechanism, green innovation is divided into green process innovation and green product innovation, and the effect of customer concentration is more pronounced in green product innovation. Further testing discusses the roles of the external environment, internal governance, and manager characteristics. Specifically, product market competition and political resources influence firms’ reliance on major customers, allowing digital technologies to optimize resource allocation for green innovation. In terms of internal governance, flexibility and regulatory strength alter the emphasis firms place on green innovation, with higher governance efficiency reducing dependency on major customers. Managerial characteristics, particularly managers’ rationality, determine the importance placed on digital technologies versus customer demands, leading to varied investment decisions in green innovation. Our findings provide valuable insights for optimizing resource allocation and enhancing green innovation investment, thereby effectively promoting sustainable regional economic development.

1. Introduction

As global environmental issues, such as resource depletion, climate change, and ecological destruction, become increasingly severe, they have garnered significant attention worldwide. Some countries and regional organizations have implemented various environmental protection measures and held numerous important conferences. For example, the Paris Agreement, signed by 178 countries in 2016, aimed to address the potential impacts of global climate change on sustainable economic development. It required governments to reduce greenhouse gas emissions, improve energy efficiency, and develop renewable energy sources. However, some developing countries still heavily rely on the economic activities of manufacturing firms during their development process, often neglecting the sustainability benefits of environmental protection [1]. This situation presents a significant dilemma for many labor-intensive developing countries, such as China and India: should they protect the environment and potentially hinder economic development, or should they compromise the environment for rapid economic growth? In this context, the concept of green innovation has emerged as a crucial approach to balancing environmental and economic benefits. Green innovation refers to technological and managerial innovations that achieve efficient resource utilization, reduce environmental pollution, and protect ecosystems, thereby promoting sustainable economic development [1,2]. Green innovation includes not only improvements in products and processes but also requires firms to consider environmental protection and social responsibility while pursuing economic benefits [3]. The introduction of this concept marks a shift in the global economy from traditional extensive growth models to green, low-carbon, and circular development [4,5]. Therefore, exploring the factors influencing green innovation in manufacturing enterprises becomes a critical path for achieving sustainable economic development.

In the manufacturing industry, high-tech manufacturing industries, as a technology-intensive and capital-intensive sector, significantly influence the success of economic transformation, particularly in developing countries. High-tech manufacturing firms possess strong research and development (R&D) capabilities and technological reserves, enabling them to achieve innovations that improve resource utilization efficiency and reduce environmental pollution. Green innovation can help these firms reduce costs and improve efficiency [6]. By improving production processes and increasing resource utilization efficiency, high-tech manufacturing firms can reduce raw material and energy consumption, thus lowering production costs. Additionally, green innovation in high-tech manufacturing enhances the competitiveness of these firms and provides green technologies and solutions to other industries, driving the entire economic system towards sustainable development [7,8]. Through green innovation, high-tech manufacturing firms can produce more energy-efficient products, meeting market demands and gaining larger market shares. Notably, green innovation can enhance the social responsibility of high-tech manufacturing firms and build a positive corporate image [2,9]. In the era of informatization and globalization, investments in green innovation by high-tech manufacturing firms will be reflected in environmental protection and social responsibility, thereby earning the trust of the consumers and investors.

As cooperative relationships between firms improve, the relationship between manufacturing firms and their customers has gradually formed close economic links, particularly due to the potential benefits brought by major customers [10,11]. This is a key factor in strategic planning for manufacturing firms [12]. According to the Resource-Based View, high-tech manufacturing firms with higher customer concentration may face corresponding impacts on their investments in green innovation [13]. Firstly, the dependency resulting from high customer concentration may lead to over-concentration and uneven utilization of innovation resources in high-tech manufacturing firms. When these firms overly rely on a few major customers, their resource allocation and innovation strategies often skew towards the needs of these important customers, leading to an imbalance in resource distribution [14]. To meet the specific demands of major customers, firms may allocate more resources to these projects, neglecting the development opportunities presented by green innovation. Secondly, high customer concentration might lead to path dependency and risk-averse behavior in innovation. When high-tech manufacturing firms depend on a few major customers, their innovation decisions are strongly influenced by these customers’ demands and preferences [15]. This reliance causes firms to focus on incremental improvements of existing technologies and products rather than boldly attempting entirely new green technologies. Additionally, to maintain relationships with major customers, firms may become more conservative and risk-averse, unwilling to invest heavily in green innovation projects with higher uncertainties, further inhibiting their green innovation capabilities [16]. Meanwhile, digital transformation may weaken the impact of customer concentration on green innovation in high-tech manufacturing firms. In the fast-developing digital economy, firms no longer solely rely on the demands and guidance of a few major customers but can use digital technologies to obtain more diversified market information and innovation resources [17]. Digital transformation plays a critical role in green innovation in high-tech manufacturing firms, and their relationship can be explained by the theory of Innovation Value Chain [18,19]. During the knowledge acquisition phase, digital transformation expands the knowledge sources and acquisition channels for high-tech manufacturing firms. Technologies such as big data analytics, artificial intelligence, and the Internet of Things enable enterprises to extract valuable information from vast datasets, identifying potential development opportunities from green innovation [20]. During the knowledge conversion phase, digital technologies significantly enhance the innovation capability and efficiency of high-tech manufacturing firms. Virtual simulation and digital twin technologies enable firms to test and optimize green technologies in virtual environments, reducing R&D costs and risks [21]. During the knowledge diffusion phase, digital transformation provides strong support for the promotion and application of green innovation results. Through digital platforms and social media, high-tech manufacturing firms can efficiently disseminate green innovation results, increasing market awareness and consumer acceptance [22]. Precise digital marketing techniques help firms target environmentally conscious consumer groups, promote green products and technologies, and expand their market share. In this context, exploring the impact of customer concentration and digital transformation on green innovation has become a key factor in driving industrial transformation in high-tech manufacturing industries.

To investigate the relationship between customer concentration, digital transformation, and green innovation, this study uses data from high-tech manufacturing firms in China from 2007 to 2021 and constructs empirical models to explore the factors influencing green innovation. As the largest developing country by economic scale, China’s economic development depends on the technological transformation and industrial upgrading of the manufacturing sector, gradually shifting from labor-intensive to technology-intensive industries. This industrial transformation can provide relevant recommendations for the sustainable economic development of other developing countries, ensuring stable global economic growth. According to empirical results, we find that customer concentration can inhibit green innovation in high-tech manufacturing firms, while digital transformation promotes green innovation outcomes. Additionally, digital transformation can moderate the negative impact of customer concentration on green innovation, indicating that the adoption of digital technologies reduces the heavy reliance of high-tech manufacturing firms on major customers. The robustness test using propensity score matching confirms that the relationship between customer concentration, digital transformation, and green innovation remains significant. In the heterogeneity test, we find that customer concentration significantly inhibits green product innovation, and digital transformation can also moderate the relationship between customer concentration and green product innovation. However, green process innovation is not significantly affected by customer concentration. Further testing explores the different impact mechanisms of customer concentration and digital transformation on green innovation from the perspectives of external environment, internal governance, and manager characteristics. Regarding the external environment, increased product market competition intensity not only prompts high-tech manufacturing firms to rely more on customer demands but also weakens the moderating effect of digital transformation on the relationship between customer concentration and green innovation. Compared to firms receiving environmental subsidies, high-tech manufacturing firms not receiving subsidies are more likely to be influenced by customer concentration in green innovation, with digital transformation being a crucial means to reduce dependence on major customers. In terms of internal governance, although a higher state-owned shareholding prompts high-tech manufacturing firms to rely more on major customer demands, it also enhances the role of digital transformation in promoting green innovation. Notably, high-tech manufacturing firms not audited by the Big Four accounting firms are influenced by customer concentration and digital transformation in green innovation, indicating that reduced internal regulatory intensity alters decision-making preferences regarding green innovation. Regarding manager characteristics, managers with financial backgrounds are more influenced by major customers and utilize digital transformation to mitigate the negative impact of customer concentration on green innovation. Conversely, managers with weak opportunism use digital technologies to promote green innovation outcomes, indicating that managers’ risk preferences can change the impact mechanisms of customer concentration and digital transformation on green innovation.

Our work contributes to the relevant stream of literature from the following perspectives: Firstly, it enriches the understanding of factors that influence green innovation. While existing studies primarily focus on the external environment or internal governance, exploring various factors that promote or inhibit investments in green innovation, our study employs the Resource-Based View as the fundamental theoretical framework. This approach allows us to rigorously examine the impact of customer dependency and digital technology on green innovation, offering a novel perspective for ongoing research in this area. Secondly, from the supply chain management perspective, this study explores the interaction mechanism between suppliers and customers in green innovation. Unlike other forms of innovation, green innovation places a strong emphasis on the development of green technologies or products, necessitating that firms align closely with the actual innovation demands of their customers. In this supply chain context, high-tech manufacturing firms must tailor their investments in green innovation to correspond with shifts in customer demands, thereby fostering a value co-creation process that is propelled by green innovation. Lastly, our findings demonstrate the important role of digital transformation in facilitating resource coordination within high-tech manufacturing firms. The application of digital technologies significantly enhances technological upgrading and efficiency improvement in high-tech manufacturing firms and is a crucial condition for achieving green transformation. Owing to the competitive edge provided by digital technologies, high-tech manufacturing firms can achieve efficient internal resource allocation, balancing economic and environmental benefits.

The remainder of this study is organized as follows: Section 2 introduces the literature review and research hypotheses; Section 3 describes the research design, including data sources, variable construction, and model construction; Section 4 presents the research results, including descriptive statistics, baseline test, robustness check, heterogeneity test, and further test; Section 5 discusses the empirical results; the conclusion is drawn in Section 6.

2. Literature Review and Research Hypotheses

2.1. Green Innovation

Green innovation represents a unique form of innovation that allows enterprises to simultaneously secure environmental protection and obtain economic benefits through advancements in technology and product development [2]. As the concept of sustainable development continues to evolve, green innovation has become an increasingly critical tool for enterprises to gain a competitive edge, particularly in high-tech manufacturing sectors [1]. On one hand, green innovation can reduce operational costs and enhance market competitiveness [6]. By improving production processes and increasing resource utilization efficiency, enterprises can reduce energy and raw material consumption, thereby lowering pollutant emissions. Leveraging the outcomes of green innovation, enterprises can meet customer demands for environmental protection, securing a favorable position in market competition [23]. On the other hand, green innovation can help enterprises obtain policy support and enhance their brand image and social responsibility [24]. With governments placing increasing emphasis on environmental protection, enterprises’ efforts in green innovation gain recognition from local authorities, allowing them access to more policy resources [9]. In the era of rapid internet development, environmental protection measures implemented by enterprises attract public and media attention, helping to build a positive brand image and enhance corporate social responsibility performance [25]. Therefore, exploring the factors influencing green innovation is essential for high-tech manufacturing enterprises to achieve sustainable development goals.

Existing research on the factors influencing green innovation primarily delves into three aspects: external environment, internal governance, and manager characteristics. In terms of the external environment, market demand, industry competition, and government policies are crucial in determining whether enterprises invest heavily in green innovation [23,26]. Within an economic theoretical framework, market demand for green products increases enterprises’ market opportunities, encouraging them to improve production processes or product compositions, making green innovation a key tool to meet market demand. Additionally, market competition pressure is a significant driving force for green innovation, enabling enterprises to achieve differentiation advantages and stand out in the competition [27]. However, intensified market competition may also alter enterprises’ development strategies, forcing them to abandon green innovation in favor of other profit-maximizing approaches [3]. Meanwhile, government factors are pivotal in driving enterprises to undertake green innovation activities. Compared to developed countries, developing countries rely on environmental protection achievements to gain international recognition, making these nations highly focused on the transformation of the manufacturing sector [28]. By enacting stringent environmental regulations and emission standards, governments can compel enterprises to upgrade their production technologies to meet environmental protection goals [1,29]. This process is also supported by government policies such as tax incentives, subsidies, and green credits, encouraging enterprises to invest more resources in green innovation.

In terms of internal governance, ownership structure and internal regulatory intensity directly influence enterprises’ decisions regarding green innovation. According to agency theory, conflicts between managers and owners are often driven by interests, indirectly leading to challenges in balancing environmental and economic benefits [25]. In this context, a high-concentration ownership structure can enhance decision-making efficiency and resource allocation capabilities, accelerating the decision-making and implementation of green innovation projects [30]. However, overly concentrated ownership structures may lead to conservative approaches to green innovation, limiting innovation diversity [7]. Conversely, a decentralized ownership structure can introduce diverse perspectives and interests, promoting broad participation and support for green innovation, albeit potentially at the expense of decision-making efficiency. Additionally, an increase in state ownership can help enterprises access more resources and policy support, with higher expectations and demands from the government for green innovation [28]. Governments may encourage these enterprises to invest in green technology R&D and applications through special funds, tax incentives, and technical support. It is worth noting that enterprises with high state ownership may also face challenges such as insufficient innovation motivation and low management efficiency in green innovation [27]. Furthermore, changes in internal regulatory intensity can alter enterprises’ investments in green innovation. High-intensity internal regulation ensures that enterprises adhere to established environmental goals and standards in green innovation, preventing resource waste and violations [7]. In contrast, low-intensity internal regulation can lead to a lack of systematic and rigorous approaches to green innovation, resulting in resource waste and environmental violations, thereby affecting the quality and effectiveness of green innovation [2]. Strengthening internal regulation can not only improve the quality and effectiveness of green innovation but also enhance transparency and credibility in environmental protection.

Regarding manager characteristics, managers’ financial backgrounds and speculative tendencies play significant roles in decision-making related to green innovation. Firstly, managers with financial backgrounds typically possess strong financial management and capital operation capabilities, enabling them to effectively integrate and utilize resources to support green innovation projects [24]. These managers can develop scientific financial plans to ensure the sustainability and long-term returns of green innovation projects [29]. However, such managers may focus more on short-term financial returns, neglecting the long-term value and social benefits of green innovation, particularly when enterprises face financial pressures or market uncertainties [1]. In such situations, they may prioritize projects with shorter payback periods, affecting investments in green innovation. Secondly, managers’ opportunism significantly influences green innovation. Managers with strong opportunism often pursue high-risk, high-reward investment opportunities, focusing on short-term market fluctuations and quick gains rather than long-term stable development strategies [31]. Driven by this mentality, they may allocate resources to high-risk, fast-return projects rather than green innovation projects that require long-term investment and steady R&D [8]. Green innovation typically involves long R&D cycles and substantial upfront investment, with relatively slow market returns, conflicting with the risk preferences of speculative managers.

2.2. Customer Concentration and Green Innovation

Customer concentration refers to the proportion of a company’s revenue derived from a few major customers. This concentration has a dual impact on internal resource allocation [11]. On one hand, higher customer concentration can enhance resource allocation efficiency. Firms that are highly dependent on a few major customers can establish long-term stable relationships with them, gaining more market information and resource support, optimizing resource utilization, and reducing market development and customer maintenance costs [32]. Additionally, these firms can more efficiently allocate resources and respond quickly to the needs of their major customers, thereby enhancing market competitiveness [33]. On the other hand, increased customer concentration can exacerbate an enterprise’s dependence on major customers. Such firms must prioritize the needs and preferences of these customers in their strategic and operational decisions, potentially leading to passive positions in response to market changes [34]. Over-reliance on revenue from major customers exposes enterprises to significant operational risks when these customers’ demands change or if they are lost [11]. High-tech manufacturing firms with higher customer concentration tend to be conservative in resource allocation, lacking the motivation and resources to invest in high-risk and high-investment green innovation projects [12]. The impact of customer concentration on green innovation primarily manifests in several ways.

High-tech manufacturing firms heavily reliant on a few major customers must prioritize these customers’ demands and preferences, which may not align with green innovation directions. Research indicates that green innovation typically requires substantial upfront investment and long payback periods [10]. Firms with high customer concentration, facing significant demand pressure from major customers, are less likely to see immediate economic benefits from green innovation, thus tending to reduce resource allocation for it [12]. And then, the strong bargaining power of major customers reduces enterprises’ motivation for green innovation when facing environmental regulations and market pressures [35]. Major customers can negotiate and use other business means to help enterprises partially avoid environmental pressures, resulting in low enthusiasm for green innovation [12]. Additionally, firms highly dependent on major customers must consider these customers’ short-term interests in their strategic and operational decisions, limiting long-term planning and strategic deployment for green innovation [15]. Given the uncertainty and variability of major customer demands, firms face greater risks and uncertainties in green innovation projects, prompting them to avoid high-risk investments.

The inhibitory effect of customer concentration on green innovation in high-tech manufacturing firms is mainly reflected in reduced resource allocation, insufficient innovation motivation, and mismatched resource capabilities. Firms highly dependent on a few major customers tend to prioritize these customers’ short-term needs in green innovation decisions, thereby reducing resource allocation for green innovation and inhibiting innovation outcomes. Therefore, the following hypothesis is proposed:

Hypothesis 1.

There is a negative association between customer concentration and green innovation in high-tech manufacturing industries.

2.3. Digital Transformation and Green Innovation

Digital transformation refers to the comprehensive overhaul of business processes and management models through the application of information technology, significantly enhancing enterprise efficiency and fostering technological innovation [20]. For high-tech manufacturing firms, digital transformation is of paramount importance. Firstly, digital transformation substantially improves production efficiency and product quality. By integrating advanced technologies such as the Internet of Things (IoT), big data, and artificial intelligence (AI), enterprises can achieve real-time monitoring and intelligent maintenance of production equipment, thereby enhancing production line efficiency [19]. Additionally, these digital technologies facilitate the optimization of production processes, precise control of production parameters, and improvement of product consistency and quality [36]. Secondly, digital transformation enhances market responsiveness in high-tech manufacturing firms. In a digitally enabled environment, enterprises can interact directly with customers through e-commerce platforms, social media, and mobile applications, allowing them to better understand customer needs and market trends [37,38]. This capability enables enterprises to quickly respond to customer feedback and market changes, adjust product designs and production plans accordingly, launch products that meet market demands, and enhance customer satisfaction and market competitiveness [21]. Thirdly, digital transformation drives technological innovation in high-tech manufacturing firms. Digital technologies provide a wealth of innovation-based tools and platforms, such as cloud computing, virtual reality, and blockchain. These tools facilitate product design, R&D, and testing, while also reducing innovation costs and risks [39]. Lastly, digital transformation supports high-tech manufacturing firms in achieving sustainable development. By applying digital technologies, firms can optimize resource utilization, improve energy efficiency, and reduce environmental pollution [40]. Smart manufacturing technologies enable real-time monitoring of energy consumption and waste emissions, allowing enterprises to take timely measures to reduce resource consumption and environmental impact [41]. Digital transformation is strategically significant for high-tech manufacturing firms as it enhances production efficiency and market competitiveness, drives innovation and sustainable development, and provides robust technical support and management assurance for high-quality growth.

Digital transformation also plays a crucial role in promoting green innovation in high-tech manufacturing firms by enhancing technological capabilities, optimizing resource allocation, and fostering collaborative innovation [19]. Moreover, it significantly enhances technological capabilities in high-tech manufacturing enterprises, providing robust technical support for green innovation [42]. By applying digital technologies such as IoT, AI, and big data analysis, firms can achieve precise monitoring and management of production processes and environmental impacts [41]. Digital technologies support enterprises in green product design and R&D, enabling the development of low-energy, high-efficiency environmental products that meet market demands [41]. In addition, digital transformation optimizes resource allocation, providing efficient resource support for green innovation. Through digital technologies, enterprises can achieve precise resource management and efficient utilization [21]. Additionally, digital technologies support energy management and circular economy practices, achieving efficient energy use and management. Blockchain technology facilitates traceability of waste and resource recycling, promoting enterprises’ transition to green and sustainable development [43]. Furthermore, digital transformation fosters collaborative innovation in high-tech manufacturing firms, providing a broad cooperation platform for green innovation. Digital technologies offer open innovation platforms and tools, enabling firms to collaborate conveniently with external partners. Therefore, the following hypothesis is proposed:

Hypothesis 2.

There is a positive association between digital transformation and green innovation in high-tech manufacturing industries.

Digital transformation significantly enhances the efficiency of internal resource allocation in manufacturing firms [44]. By applying digital technologies, firms can achieve precise resource management and efficient utilization, reducing dependence on major customers [37]. Using digital technologies, firms can manage and optimize resources throughout the entire process. Leveraging the advantages of big data analytics, high-tech manufacturing firms can accurately forecast market demand and resource supply, reducing resource waste and inventory backlogs [27]. Additionally, digital technologies support supply chain management and collaborative optimization, achieving seamless integration and efficient collaboration in the supply chain, enhancing operational efficiency and competitiveness [44]. In this context, digital transformation reduces high-tech manufacturing firms’ dependence on major customers, allowing for more flexible and efficient responses to market and customer demands [45]. In channel management, digital technologies support multichannel sales and customer management, expanding new markets and customer resources, and reducing dependence on major customers [21]. By optimizing resource allocation and enhancing market responsiveness, firms can achieve business diversification and market diversification, enhancing competitiveness and risk resilience [37].

Based on these viewpoints, digital transformation significantly improves resource allocation efficiency in high-tech manufacturing firms by achieving precise resource management and efficient utilization, optimizing resource allocation processes, and enhancing operational efficiency. By reducing firms’ dependence on major customers, digital transformation provides robust support for business diversification and market diversification, enabling flexible and efficient green innovation and sustainable development. Therefore, the following hypothesis is proposed:

Hypothesis 3.

Digital transformation could moderate the relationship between customer concentration and green innovation in high-tech manufacturing industries.

The research framework of this study is shown in Figure 1.

3. Research Design

3.1. Data Sources

This study utilizes data from non-financial listed companies on the A-share market in China, spanning from 2007 to 2021, as the research sample. Recognizing that high-tech enterprises often have substantial research capabilities and technological reserves, we concentrate on high-tech manufacturing firms for the final sample. The “High-tech Industry (Manufacturing) Classification” issued by the National Bureau of Statistics of China is used to identify relevant manufacturing sectors, including pharmaceutical manufacturing, aerospace and equipment manufacturing, electronic and communication equipment manufacturing, computer and office equipment manufacturing, medical instrument and device manufacturing, and information chemical manufacturing. For industry classification, we adopt the “Industry Classification Guidelines for Listed Companies” issued by the China Securities Regulatory Commission in 2012. The sample exclusion criteria are as follows: (1) companies listed for less than five years; (2) companies with significant missing variable data; and (3) companies that were delisted during the sample period. After winsorizing all continuous variables at the 1% level, the final research sample consists of 7715 observations.

In the empirical analysis, data on green innovation and digital transformation of high-tech manufacturing firms are sourced from the Chinese Research Data Services Platform (CNRDS). Sales data pertinent to customer concentration are obtained from the China Stock Market and Accounting Research (CSMAR) database. Data from annual reports of companies are retrieved from the official websites of the Shanghai Stock Exchange and Shenzhen Stock Exchange. Additional information on listed companies, including capital structure, stock market performance, and operational performance data, is also sourced from CSMAR.

3.2. The Construction of Variables

3.2.1. Dependent Variable

Green innovation reflects the outcomes of enterprises’ efforts to advance technologies or products in an environmentally friendly manner, balancing environmental benefits with economic gains. Existing research typically measures innovation through R&D investment and innovation outcomes. Given that the R&D investment in green innovation closely resembles that of general innovation activities, it is challenging to highlight the unique environmental advantages of green innovation [9]. To better capture these benefits, this study focuses on the potential environmental gains from green innovation outcomes in high-tech manufacturing firms, using the number of green patent applications as a measure. Therefore, we measure the green innovation variable using the natural logarithm of the number of green patent applications plus one.

3.2.2. Independent Variables

Customer concentration denotes the close relationship between enterprises and their major customers in the product sales process. Unlike other industries, high-tech manufacturing firms heavily rely on customer demand and market trends for their operations and production outputs, making them more sensitive to changes in major customers’ demands. From a resource perspective, the actual demands of major customers significantly influence resource allocation in high-tech manufacturing firms, prompting these firms to invest more resources in specific products [11]. To effectively measure customer concentration, we use the idea of the Herfindahl–Hirschman Index (HHI) to show the degree of dependence of firms on major customers. Consistent with existing research, we calculate customer concentration using the sum of the squared ratios of the sales revenue of the top five customers to the total sales revenue.

Digital transformation represents the extent of attention and resource investment in digital technologies by enterprises. Leveraging digital technologies such as big data analytics, artificial intelligence, and blockchain, high-tech manufacturing firms can optimize production processes and predict market demand, gaining significant market advantages. Similar to technological innovation, the economic benefits brought by digital technologies to high-tech manufacturing firms exhibit a time lag, indicating that digital transformation is a process variable [46]. To measure this process variable, we focus on the attention given to digital technologies by high-tech manufacturing firms. We use text analysis methods to extract keywords related to digital transformation from the management discussion and analysis (MD&A) sections of annual reports. Following existing research, we measure the digital transformation variable using the natural logarithm of the frequency of digital-related keywords plus one.

3.2.3. Control Variables

Based on existing research, green innovation in high-tech manufacturing firms may be influenced by factors such as profitability, stock market performance, and capital structure [3,7,9]. Profitability provides enterprises with the necessary resources to invest in green technologies, suggesting that higher profits may correlate with enhanced capacity and willingness to engage in environmental practices. Meanwhile, stock market performance influences corporate strategies and investor confidence, potentially driving firms to adopt green innovations to enhance their market image and attract eco-conscious investors. Lastly, the capital structure of a company affects its financial stability and ability to fund green projects, including those aimed at environmental sustainability. Considering the characteristics of the high-tech industry, investments in green innovation by high-tech manufacturing firms can be also affected by government factors such as political resources and policy support. In this context, we select the following variables as control variables to account for the influence of other factors on green innovation in high-tech manufacturing firms: listing age (Age), leverage (Leverage), return on assets (ROA), stock return (Return), turnover (Turnover), the shareholding ratio of the largest shareholder (First), institutional investor shareholding ratio (Institution), and state ownership ratio (State).

The selected control variables comprehensively consider aspects such as operational performance, stock market performance, capital structure, and government support. Listing age measures the duration of the company’s presence in the stock market; leverage measures the debt pressure faced by the company during operations; return on assets measures the profitability of the company’s operational activities; stock return measures the company’s profitability in the stock market; turnover measures the liquidity of the company’s stocks in the stock market; the shareholding ratio of the largest shareholder and the institutional investor shareholding ratio measure the company’s capital structure; and the state ownership ratio measures the extent of government involvement in the company’s operations.

The definitions and measurements of all variables used in this study are shown in

Table 1. The definition of variables.

Variables	Type	Measurement
Green Innovation	Dependent Variable	Green Innovation is measured by the natural logarithm of the number of green innovation applications plus one.
Customer	Independent Variable	Customer is measured by the sum of the squared ratios of the top five customer sales to total sales.
Digital		Digital is measured by the natural logarithm of the frequency of digital words plus one.
Age	Control Variable	Age is measured by the natural logarithm of the number of years that firm i has been listed.
Leverage		Leverage is measured by the ratio of total debts to total assets.
ROA		ROA is measured by the ratio of net profit to total assets.
Return		Return is measured by the annual cumulative stock returns.
Turnover		Turnover is measured by the annual trading volume.
First		First is measured by the proportion of shares held by firm i’s top one shareholder.
Institution		Institution is measured by the ownership proportion of shares held by institution.
State		State is measured by the ownership proportion of shares held by state.

.

3.3. The Construction of Models

To test the hypotheses proposed in Section 2, this study constructs the following empirical models to explore the potential relationships among customer concentration, digital transformation, and green innovation in high-tech manufacturing firms.

$$\begin{gathered} Green Innovatio{n}_{i,t}=\alpha +{\beta }_{1}Custome{r}_{i,t-1}+{\beta }_{2}Digita{l}_{i,t-1}+Control{s}_{i,t-1} \\ +Fir{m}_{fixed-effects}+Yea{r}_{fixed-effects}+{\epsilon }_{i,t} \end{gathered}$$

(1)

In Equation (1), $Green Innovatio{n}_{i,t}$ represents the green innovation of firm i in year t; $Custome{r}_{i,t-1}$ represents the customer concentration of firm i in year t − 1; $Digita{l}_{i,t-1}$ represents the digital transformation of firm i in year t − 1; $Control{s}_{i,t-1}$ includes the control variables such as listing age (Age), leverage (Leverage), return on assets (ROA), stock return (Return), turnover (Turnover), largest shareholder’s shareholding ratio (First), institutional investors’ shareholding ratio (Institution), and state ownership ratio (State). Considering the research hypotheses proposed in Section 2, we predict that the coefficient (${\beta }_1$) for $Custome{r}_{i,t-1}$ will be negative, whereas the coefficient (${\beta }_2$) for $Digita{l}_{i,t-1}$ will be positive. To mitigate the impact of endogeneity, all explanatory and control variables are lagged by one period, thereby better capturing their influence on green innovation in high-tech manufacturing firms. $Fir{m}_{fixed-effects}$ and $Yea{r}_{fixed-effects}$ control for individual and year effects, respectively. ${\epsilon }_{i,t}$ represents the error term.

According to the Resource-Based View, digital transformation, while influencing the allocation of innovation resources within enterprises, may also impact customer concentration. This interaction could, in turn, affect green innovation in high-tech manufacturing firms. Therefore, we further construct the following empirical model to examine the moderating effect of digital transformation on the relationship between customer concentration and green innovation.

$$\begin{gathered} Green Innovatio{n}_{i,t}=\alpha +{\beta }_{1}Custome{r}_{i,t-1}+{\beta }_{2}Digita{l}_{i,t-1}+ \\ {\beta }_{3}Custome{r}_{i,t-1}\times Digita{l}_{i,t-1}+Control{s}_{i,t-1}+ \\ Fir{m}_{fixed-effects}+Yea{r}_{fixed-effects}+{\epsilon }_{i,t} \end{gathered}$$

(2)

In Equation (2), $Custome{r}_{i,t-1}\times Digita{l}_{i,t-1}$ represents the interaction term between customer concentration and digital transformation. In line with research hypothesis H3, we predict that the coefficient (${\beta }_3$) for the interaction term ($Custome{r}_{i,t-1}\times Digita{l}_{i,t-1}$) will be positive. To address potential endogeneity issues, the interaction term is also lagged by one period, ensuring the reliability and accuracy of the empirical results.

4. Empirical Results

4.1. Descriptive Statistics

To better illustrate the distribution characteristics of the variables, descriptive statistics are first performed on all variables before conducting empirical analysis. The results are shown in

Table 2. The results of descriptive statistics.

Variables	Observations	Mean	S.D.	Min	Median	Max
Green Innovation	7715	1.027	1.222	0.000	0.693	4.736
Customer	7715	0.284	0.197	0.000	0.234	0.895
Digital	7715	1.194	1.297	0.000	0.693	4.691
Age	7715	2.488	0.439	1.792	2.485	3.332
Leverage	7715	0.436	0.204	0.069	0.433	1.032
ROA	7715	0.029	0.084	−0.414	0.033	0.215
Return	7715	0.209	0.631	−0.663	0.053	2.714
Turnover	7715	5.667	3.867	0.760	4.603	18.546
First	7715	0.305	0.131	0.081	0.283	0.666
Institution	7715	0.435	0.231	0.004	0.446	0.930
State	7715	0.035	0.106	0.000	0.000	0.543

. The mean value of green innovation (Green Innovation) is 1.027, with a median of 0.693 and a standard deviation of 1.222, indicating significant variability in green innovation outcomes among high-tech manufacturing firms. This suggests substantial differences in the extent and effectiveness of green innovation efforts across these firms. The mean value of customer concentration (Customer) is 0.284, with a median of 0.234 and a standard deviation of 0.197. This suggests a relatively balanced distribution in the dependency on major customers among high-tech manufacturing firms, with minor differences between companies. The mean value of digital transformation (Digital) is 1.194, with a median of 0.693 and a standard deviation of 1.297. This reflects considerable variability in the attention paid to digital technologies by high-tech manufacturing firms, indicating differences in how these firms are integrating digital technologies into their operations. The distribution pattern of the digital transformation variable is similar to that of the green innovation variable, suggesting a potential link between these two aspects. In the analysis of control variables, the standard deviation of turnover (Turnover) is significantly higher than that of other variables, highlighting substantial differences in stock market activity among high-tech manufacturing firms. This indicates that some firms are much more active in the stock market than others, which could impact their investment capabilities and strategies. Additionally, the minimum values of stock return (Return) and return on assets (ROA) are both negative, indicating that some high-tech manufacturing firms are facing substantial financial pressures. This suggests that these firms may struggle to allocate sufficient resources towards green innovation, which could hinder their ability to develop and implement sustainable practices. These descriptive statistics provide a foundational understanding of the data distribution, offering insights into the variability and characteristics of the key variables under study. This foundational analysis is crucial for interpreting the subsequent empirical results and understanding the broader implications of customer concentration, digital transformation, and other control variables on green innovation in high-tech manufacturing firms.

Given that the linear mean model is a panel data model, there may be concerns regarding multicollinearity affecting the analysis. To address this, a correlation analysis of all variables was conducted before proceeding with the empirical analysis. The results of the correlation matrix are presented in

Table 3. The results of correlation matrix.

	Variables	(1)	(2)	(3)	(4)	(5)	(6)
(1)	Green Innovation	1
(2)	Customer	−0.002	1
(3)	Digital	0.350 ***	0.082 ***	1
(4)	Age	0.087 ***	−0.027 **	−0.030 ***	1
(5)	Leverage	0.187 ***	0.033 ***	−0.013	0.219 ***	1
(6)	ROA	0.007	−0.137 ***	−0.043 ***	−0.050 ***	−0.379 ***	1
(7)	Return	−0.027 **	−0.028 **	−0.063 ***	−0.052 ***	0.015	0.152 ***
(8)	Turnover	−0.038 ***	0.052 ***	0.055 ***	−0.154 ***	0.002	−0.093 ***
(9)	First	−0.058 ***	−0.058 ***	−0.079 ***	−0.013	−0.026 **	0.140 ***
(10)	Institution	0.025 **	−0.056 ***	−0.158 ***	0.229 ***	0.078 ***	0.187 ***
(11)	State	−0.070 ***	−0.019 *	−0.160 ***	0.060 ***	0.088 ***	0.015
		(7)	(8)	(9)	(10)	(11)
(7)	Return	1
(8)	Turnover	0.381 ***	1
(9)	First	0.027 **	−0.167 ***	1
(10)	Institution	0.112 ***	−0.301 ***	0.496 ***	1
(11)	State	0.097 ***	0.099 ***	0.251 ***	0.283 ***	1

Note: The lower triangular matrix reports the Pearson correlation coefficients. ***, **, and * represent the 1%, 5%, and 10% significance levels, respectively.

. The correlation analysis reveals a negative correlation between green innovation (Green Innovation) and customer concentration (Customer), suggesting that higher customer concentration might inhibit green innovation in high-tech manufacturing enterprises. Additionally, there is a significant positive correlation between green innovation (Green Innovation) and digital transformation (Digital), indicating that digital transformation can facilitate green innovation within these firms. By comparing the correlations between the control variables and the dependent and independent variables, it is observed that the absolute values of all correlation coefficients are below 0.5. This indicates that multicollinearity is not a concern in this empirical analysis, ensuring the reliability of the regression results. These initial findings lay a robust foundation for further empirical analysis, supporting the exploration of how customer concentration and digital transformation influence green innovation in high-tech manufacturing firms while considering other control variables.

4.2. Baseline Test

To provide a more comprehensive analysis of the impact of customer concentration and digital transformation on green innovation, we initially conduct univariate difference tests to explore the effects of varying levels of customer concentration and digital transformation on green innovation in high-tech manufacturing firms. During the t-test process, the median values of customer concentration and digital transformation are used as criteria for grouping. Consequently, subsamples with low Customer, high Customer, low Digital, and high Digital are constructed, respectively. The t-test results are presented in

Table 4. The results of the t-test.

Variable	Customer (Low)	Mean 1	Customer (High)	Mean 2	Differences
Green Innovation	3859	1.327	3856	1.028	0.299 **
Variable	Digital (Low)	Mean 1	Digital (High)	Mean 2	Differences
Green Innovation	4089	0.690	3626	1.408	−0.719 ***

Note: *** and ** represent the 1% and 5% significance levels, respectively.

.

In Table 4, the mean value of green innovation in high-tech manufacturing firms is significantly higher in the sub-sample with lower customer concentration compared to that in the sub-sample with higher customer concentration, and the positive intergroup difference indicates that high-tech manufacturing firms with lower customer concentration perform better in terms of green innovation. Furthermore, the mean value of green innovation in high-tech manufacturing firms is significantly lower in the sub-sample with lower levels of digital transformation than in the sub-sample with higher levels of digital transformation, and the negative intergroup difference suggests that firms with higher digital transformation exhibit better performance in green innovation. The t-test results in Table 4 can support the hypotheses proposed in Section 2.

To test the hypotheses proposed in Section 2, we employ the empirical models constructed in Equations (1) and (2) to explore the relationships between customer concentration, digital transformation, and green innovation in high-tech manufacturing firms. In our regression analysis, we first examine the impact mechanisms of customer concentration and digital transformation on green innovation without considering control variables. Subsequently, we incorporate control variables to further investigate these relationships. Finally, we include the interaction term between customer concentration and digital transformation to explore the moderating role of digital transformation in resource allocation. The baseline test results are presented in

Table 5. The results of baseline test.

	Green Innovation
	(1)	(2)	(3)	(4)	(5)
Customer	−0.1652 **		−0.1721 **	−0.1873 **	−0.3101 ***
	(−2.08)		(−2.17)	(−2.37)	(−3.39)
Digital		0.1085 ***	0.1087 ***	0.1049 ***	0.0672 ***
		(8.28)	(8.31)	(8.04)	(3.50)
Customer × Digital					0.1264 ***
					(2.68)
Age				−0.2147 *	−0.2136 *
				(−1.88)	(−1.87)
Leverage				0.4390 ***	0.4461 ***
				(5.31)	(5.40)
ROA				0.5328 ***	0.5376 ***
				(3.90)	(3.94)
Return				0.0302	0.0299
				(1.42)	(1.41)
Turnover				0.0067 *	0.0064 *
				(1.94)	(1.86)
First				−0.4107 **	−0.4179 **
				(−2.35)	(−2.40)
Institution				0.4220 ***	0.4245 ***
				(4.40)	(4.43)
State				0.0667	0.0722
				(0.64)	(0.69)
Constants	1.1251 ***	0.9570 ***	1.0047 ***	1.2331 ***	1.2640 ***
	(47.58)	(57.01)	(36.38)	(4.13)	(4.23)
Firm FE	Yes	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes	Yes
N	6545	6545	6545	6545	6545
R2	0.7233	0.7263	0.7265	0.7293	0.7296

Note: The t-statistics are in parentheses. ***, **, and * represent the 1%, 5%, and 10% significance levels, respectively.

.

Table 5 outlines the baseline test results. Columns (1) to (3) focus on the impacts of customer concentration and digital transformation on green innovation in high-tech manufacturing firms. In Column (1), there is a significant negative correlation between customer concentration and green innovation (coefficient = −0.1652, significant at the 5% level). In Column (2), the coefficient for digital transformation and green innovation is 0.1085, significant at the 1% level. Column (3) shows that both customer concentration and digital transformation significantly impact green innovation, consistent with the results in Columns (1) and (2). After including the control variables, Column (4) shows a significant negative correlation between customer concentration and green innovation (coefficient = −0.1873, significant at the 5% level) and a significant positive correlation between digital transformation and green innovation (coefficient = 0.1049, significant at the 1% level). The results from Columns (1) to (4) indicate that customer concentration inhibits investment in green innovation by high-tech manufacturing firms, while digital transformation promotes this unique form of innovation. In Column (5), the interaction term between customer concentration and digital transformation has a coefficient of 0.1264, significant at the 1% level. This indicates that digital transformation can mitigate the negative impact of customer concentration on green innovation. According to the results in Table 5, reliance on major customers can reduce the preference for green innovation outcomes in high-tech manufacturing firms. Conversely, the attention to and application of digital technologies can promote the production of green innovation outcomes, thus supporting the hypotheses proposed in Section 2.

4.3. Robustness Test

Given that the empirical models in this study are panel data models, there is a potential risk of sample selection bias, which could affect the accuracy and applicability of the results. To address this issue, we employ the propensity score matching (PSM) method for robustness testing. In the PSM analysis, we designate samples above the 75th percentile of customer concentration as the treatment group, while the remaining samples constitute the control group. For matching, we use the nearest neighbor method with replacement, employing all control variables as matching criteria. The propensity scores are calculated using the probit model. The robustness test results are shown in

Table 6. The results of the robustness test.

	PSM
	Treat	Green Innovation
	(1)	(2)	(3)
Customer		−0.2455 **	−0.4044 ***
		(−2.50)	(−3.40)
Digital		0.1327 ***	0.0678 *
		(6.22)	(1.95)
Customer × Digital			0.1745 **
			(2.37)
Age	−1.2176 ***	−0.5298 ***	−0.5317 ***
	(−3.78)	(−2.73)	(−2.74)
Leverage	−0.5042 **	0.6075 ***	0.6211 ***
	(−2.29)	(4.76)	(4.87)
ROA	−0.6363 *	0.3723 *	0.3795 *
	(−1.77)	(1.81)	(1.85)
Return	0.1792 ***	−0.0125	−0.0113
	(2.90)	(−0.35)	(−0.32)
Turnover	−0.0077	0.0086	0.0079
	(−0.80)	(1.52)	(1.39)
First	−0.4909	−0.3078	−0.3094
	(−1.01)	(−1.11)	(−1.12)
Institution	0.8290 ***	0.7347 ***	0.7467 ***
	(3.02)	(4.74)	(4.82)
State	−0.2360	−0.1917	−0.1923
	(−0.80)	(−1.18)	(−1.18)
Constants	2.3758 **	1.7632 ***	1.8173 ***
	(2.11)	(3.45)	(3.55)
Firm FE	Yes	Yes	Yes
Year FE	Yes	Yes	Yes
N	3337	2499	2499
Pseudo R2	0.2366
R2		0.6944	0.6951

Note: The t-statistics are in parentheses. ***, **, and * represent the 1%, 5%, and 10% significance levels, respectively.

.

Table 6 presents the robustness test results. Column (1) displays the regression results for the control variables within the treatment group. Column (2) presents the matched empirical results, revealing a significant negative correlation between customer concentration and green innovation (coefficient = −0.2455, significant at the 5% level) and a significant positive correlation between digital transformation and green innovation (coefficient = 0.1327, significant at the 1% level). In Column (3), the interaction term between customer concentration and digital transformation shows a coefficient of 0.1745, significant at the 5% level, indicating that the moderating effect of digital transformation remains significant after matching. According to the results in Table 6, the baseline test results are confirmed to be robust. This robustness analysis strengthens the credibility of the initial findings, demonstrating that customer concentration negatively impacts green innovation in high-tech manufacturing firms, while digital transformation not only promotes green innovation but also mitigates the negative effects of customer concentration.

4.4. Heterogeneity Test

Green innovation is a distinct form of innovation that encompasses two types of outcomes: green process innovation and green product innovation. Green process innovation focuses on technological advancements enabled by green technologies, such as improved energy efficiency and reduced resource consumption. In contrast, green product innovation emphasizes the environmentally friendly attributes of products that meet customers’ environmental needs. For the heterogeneity analysis, we categorize green innovation into these two outcomes. Green process innovation is measured by the number of green invention patents, while green product innovation is measured by the number of green utility patents. To ensure consistency in our research, we continue to use the empirical models specified in Equations (1) and (2) for the heterogeneity test. The empirical results are shown in

Table 7. The results of heterogeneity test.

	Green Process Innovation		Green Product Innovation
	(1)	(2)	(3)	(4)
Customer	−0.0566	−0.1176	−0.1540 **	−0.2498 ***
	(−0.81)	(−1.45)	(−2.25)	(−3.17)
Digital	0.1072 ***	0.0885 ***	0.0503 ***	0.0209
	(9.25)	(5.19)	(4.47)	(1.26)
Customer × Digital		0.0628		0.0986 **
		(1.50)		(2.42)
Age	−0.2487 **	−0.2482 **	0.0486	0.0495
	(−2.45)	(−2.45)	(0.49)	(0.50)
Leverage	0.4087 ***	0.4122 ***	0.2661 ***	0.2716 ***
	(5.57)	(5.61)	(3.73)	(3.81)
ROA	0.4694 ***	0.4718 ***	0.3134 ***	0.3171 ***
	(3.87)	(3.89)	(2.66)	(2.69)
Return	0.0250	0.0248	0.0376 **	0.0373 **
	(1.32)	(1.31)	(2.05)	(2.03)
Turnover	0.0105 ***	0.0104 ***	−0.0003	−0.0005
	(3.42)	(3.37)	(−0.09)	(−0.16)
First	−0.4345 ***	−0.4380 ***	−0.3258 **	−0.3314 **
	(−2.80)	(−2.83)	(−2.16)	(−2.20)
Institution	0.4121 ***	0.4133 ***	0.2093 **	0.2113 **
	(4.84)	(4.85)	(2.53)	(2.55)
State	0.1467	0.1494	−0.0083	−0.0040
	(1.57)	(1.60)	(−0.09)	(−0.04)
Constants	0.9234 ***	0.9388 ***	0.4799 *	0.5040 *
	(3.48)	(3.53)	(1.86)	(1.95)
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
N	6545	6545	6545	6545
R2	0.6963	0.6963	0.6719	0.6722

Note: The t-statistics are in parentheses. ***, **, and * represent the 1%, 5%, and 10% significance levels, respectively.

.

Table 7 presents the heterogeneity test results. Columns (1) and (2) explore the impact of customer concentration and digital transformation on green process innovation. In Column (1), the negative correlation between customer concentration and green process innovation is not significant, whereas digital transformation significantly promotes green process innovation (coefficient = 0.1072, significant at the 1% level). In Column (2), the interaction term between customer concentration and digital transformation is not significant, indicating that digital transformation does not significantly moderate the relationship between customer concentration and green process innovation. Columns (3) and (4) examine the impact of customer concentration and digital transformation on green product innovation. In Column (3), there is a significant negative correlation between customer concentration and green product innovation (coefficient = −0.1540, significant at the 5% level), and a significant positive correlation between digital transformation and green product innovation (coefficient = 0.0503, significant at the 1% level). In Column (4), the interaction term between customer concentration and digital transformation has a coefficient of 0.0986, significant at the 5% level, indicating that digital transformation significantly moderates the negative correlation between customer concentration and green product innovation. According to the results in Table 7, the impact mechanisms of customer concentration and digital transformation differ across various types of green innovation. Digital transformation significantly promotes both types of green innovation, while customer concentration significantly inhibits green product innovation.

4.5. Further Test

4.5.1. The Test of External Environment

Green innovation is a long-term and unpredictable outcome, and high-tech manufacturing firms may pursue green innovation passively due to external environmental factors that drive innovation. High-tech manufacturing firms not only belong to the manufacturing sector but also have distinct technological attributes. On one hand, the operations of manufacturing firms heavily depend on market demand, meaning that competitive relationships among firms affect their decision-making processes. On the other hand, high-tech manufacturing is a core force driving industrial transformation, and the innovation outcomes of these firms are crucial for governments to construct knowledge-intensive industries. In this context, we explore the external environment from two aspects: competitive relationships and government support, to investigate the differentiated factors influencing green innovation in high-tech manufacturing firms under different environmental conditions.

Regarding competitive relationships, high-tech manufacturing firms often face intense market competition, leading them to make green innovation decisions based on the competitive environment. As market competition intensifies, high-tech manufacturing firms pay more attention to changes in customer demand, increasing the influence of major customer demands on innovation decisions. Meanwhile, high-tech manufacturing firms rely on technological innovation to alleviate competitive pressure and may also obtain unique political resources. In this context, we use the Herfindahl–Hirschman Index (HHI) to measure market competition intensity and divide the research sample into two subsamples based on the median of this variable: high competition intensity and low competition intensity subsamples. The empirical results are shown in

Table 8. The results of product competition.

	Green Innovation
	Competition Intensity High		Competition Intensity Low
	(1)	(2)	(3)	(4)
Customer	−0.3178 ***	−0.3191 **	−0.0783	−0.2852 **
	(−2.63)	(−2.49)	(−0.69)	(−2.04)
Digital	0.0905 ***	0.0898 ***	0.0860 ***	0.0311
	(4.07)	(2.80)	(4.74)	(1.11)
Customer × Digital		0.0035		0.1613 **
		(0.03)		(2.56)
Age	−0.2606	−0.2606	−0.2545	−0.2392
	(−1.54)	(−1.54)	(−1.52)	(−1.43)
Leverage	0.2268 **	0.2269 **	0.5514 ***	0.5764 ***
	(2.00)	(2.00)	(4.27)	(4.46)
ROA	0.1415	0.1416	0.7413 ***	0.7459 ***
	(0.69)	(0.69)	(3.86)	(3.88)
Return	−0.0100	−0.0100	0.0650 **	0.0647 **
	(−0.30)	(−0.30)	(2.07)	(2.07)
Turnover	0.0012	0.0012	0.0051	0.0048
	(0.22)	(0.22)	(1.02)	(0.97)
First	0.0410	0.0410	−0.6653 **	−0.6748 **
	(0.17)	(0.17)	(−2.46)	(−2.50)
Institution	0.2362 *	0.2362 *	0.4588 ***	0.4727 ***
	(1.76)	(1.75)	(3.15)	(3.25)
State	−0.1298	−0.1298	0.2477	0.2631
	(−0.94)	(−0.94)	(1.49)	(1.58)
Constants	1.0869 **	1.0871 **	1.6481 ***	1.6691 ***
	(2.38)	(2.38)	(3.82)	(3.87)
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
N	2718	2718	3521	3521
R2	0.6013	0.6012	0.7318	0.7323

Note: The t-statistics are in parentheses. ***, **, and * represent the 1%, 5%, and 10% significance levels, respectively.

.

Table 8 presents the test results for product competition. Columns (1) and (2) explore the impact of customer concentration and digital transformation on green process innovation under high competition intensity. According to the results in Columns (1) and (2), customer concentration significantly inhibits green innovation in high-tech manufacturing firms under high competition intensity, while digital transformation significantly promotes green innovation outcomes. Notably, digital transformation cannot effectively moderate the negative impact of customer concentration on green innovation in a highly competitive environment. Columns (3) and (4) explore the impact of customer concentration and digital transformation on green process innovation under low competition intensity. According to the results in Columns (3) and (4), digital transformation still significantly promotes green innovation in high-tech manufacturing firms, but the negative impact of customer concentration is not significant. However, digital transformation significantly moderates the negative correlation between customer concentration and green innovation, indicating that digital technologies play a stronger role in promoting green innovation in a less competitive environment. According to the results in Table 8, increased competition intensity substantially enhances the inhibitory effect of customer concentration on green innovation decisions, while digital transformation effectively reduces the dependency of high-tech manufacturing firms on major customer demands in less competitive environments.

Considering the role of green innovation outcomes in environmental protection, high-tech manufacturing firms need to align their innovation directions with policy orientations to obtain more government support while exploring technological and product innovations. More government resources not only help firms gain market advantages but also enhance their market power. Compared to firms receiving government support, those without government support may place greater emphasis on the environmental benefits brought by green innovation, helping them better respond to competitors’ innovation strategies. In this context, we divide the research sample into subsamples based on whether they received government environmental subsidies. The empirical results are shown in

Table 9. The results of environmental subsidy.

	Green Innovation
	Environmental Subsidy		No Environmental Subsidy
	(1)	(2)	(3)	(4)
Customer	−0.0202	−0.0684	−0.1249	−0.2570 **
	(−0.09)	(−0.30)	(−1.41)	(−2.45)
Digital	0.0668 *	0.0378	0.1150 ***	0.0779 ***
	(1.73)	(0.69)	(7.94)	(3.63)
Customer × Digital		0.1266		0.1203 **
		(0.74)		(2.34)
Age	0.1604	0.1588	−0.2477 *	−0.2476 *
	(0.43)	(0.43)	(−1.88)	(−1.88)
Leverage	−0.0720	−0.0690	0.4818 ***	0.4899 ***
	(−0.31)	(−0.30)	(5.19)	(5.27)
ROA	0.3685	0.3702	0.5726 ***	0.5732 ***
	(0.82)	(0.83)	(3.90)	(3.90)
Return	0.0288	0.0283	0.0108	0.0108
	(0.56)	(0.54)	(0.45)	(0.45)
Turnover	0.0172 *	0.0169 *	0.0060	0.0057
	(1.78)	(1.75)	(1.54)	(1.47)
First	0.4260	0.4307	−0.4260 **	−0.4448 **
	(0.83)	(0.83)	(−2.14)	(−2.24)
Institution	−0.2506	−0.2437	0.6209 ***	0.6210 ***
	(−1.07)	(−1.04)	(5.59)	(5.59)
State	−0.3075	−0.3016	0.0948	0.1038
	(−1.45)	(−1.42)	(0.75)	(0.83)
Constants	0.3027	0.3128	1.2411 ***	1.2830 ***
	(0.31)	(0.32)	(3.60)	(3.72)
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
N	1041	1041	5345	5345
R2	0.7258	0.7257	0.7399	0.7402

Note: The t-statistics are in parentheses. ***, **, and * represent the 1%, 5%, and 10% significance levels, respectively.

.

Table 9 presents the test results for environmental subsidies. Columns (1) and (2) explore the impact of customer concentration and digital transformation on green process innovation under the condition of receiving environmental subsidies. According to the results in Columns (1) and (2), environmental subsidies substantially weaken the effects of customer concentration and digital transformation on green innovation, and high-tech manufacturing firms receiving environmental subsidies exhibit a lower investment preference for technological innovation. Columns (3) and (4) explore the impact of customer concentration and digital transformation on green process innovation under the condition of not receiving environmental subsidies. According to the results in Columns (3) and (4), customer concentration significantly inhibits green innovation in high-tech manufacturing firms, while digital transformation significantly promotes green innovation outcomes. Additionally, digital transformation significantly moderates the negative correlation between customer concentration and green innovation, indicating that environmental subsidies partially stimulate the enthusiasm of high-tech manufacturing firms without government support for technological innovation, helping to promote green innovation outcomes. According to the results in Table 9, environmental subsidies can alter the focus of high-tech manufacturing firms on technological innovation and reduce their dependency on major customers. However, this special form of government support may also decrease the resource investment in green innovation, making it difficult for high-tech manufacturing firms to achieve sustainable development goals.

4.5.2. The Test of Internal Governance

Internal governance is critical in corporate operations and development, significantly influencing decisions related to green innovation, especially in the presence of agency problems. With limited resources, high-tech manufacturing firms must balance the economic benefits of existing technologies with the potential benefits of green technologies, which greatly constrains managers’ resource allocation to green innovation. The core purpose of internal governance is to enhance the efficiency of internal resource utilization and ensure that decision-making maximizes benefits. In this context, we explore internal governance from two aspects: state ownership and the choice of auditing firms, to investigate the differentiated factors influencing green innovation in high-tech manufacturing firms under different governance conditions.

In the manufacturing sector, ownership structure largely determines the efficiency of corporate decision-making processes. Compared to other types of investor ownership, state-owned shareholding more significantly influences strategic adjustments and development directions, representing a unique form of conflict between shareholders and managers. To illustrate the impact of state-owned shareholding on green innovation, we divide the research sample based on the median of the state-owned shareholding ratio into high state-owned shareholding and low state-owned shareholding subsamples. The empirical results are shown in

Table 10. The results of state-owned shareholding.

	Green Innovation
	State-Owned Shareholding High		State-Owned Shareholding Low
	(1)	(2)	(3)	(4)
Customer	−0.3592 *	−0.8617 ***	−0.0710	−0.1025
	(−1.77)	(−3.68)	(−0.79)	(−0.99)
Digital	0.1173 ***	0.0522	0.0979 ***	0.0886 ***
	(3.14)	(0.95)	(6.83)	(4.24)
Customer × Digital		0.5784 ***		0.0320
		(4.16)		(0.61)
Age	0.0611	0.0411	−0.1157	−0.1164
	(0.17)	(0.11)	(−0.90)	(−0.90)
Leverage	0.4281 *	0.4702 **	0.4185 ***	0.4203 ***
	(1.84)	(2.04)	(4.46)	(4.48)
ROA	0.1691	0.1040	0.6016 ***	0.6041 ***
	(0.47)	(0.29)	(4.05)	(4.06)
Return	0.0130	0.0120	0.0476 **	0.0476 **
	(0.25)	(0.23)	(1.98)	(1.98)
Turnover	−0.0008	0.0001	0.0054	0.0053
	(−0.08)	(0.01)	(1.40)	(1.37)
First	−0.3722	−0.3612	−0.2732	−0.2759
	(−0.97)	(−0.95)	(−1.27)	(−1.28)
Institution	0.5511 **	0.5722 **	0.2870 ***	0.2876 ***
	(2.14)	(2.24)	(2.64)	(2.64)
State	0.3915 *	0.3367	−0.0283	−0.0258
	(1.84)	(1.60)	(−0.17)	(−0.16)
Constants	0.5974	0.7525	0.9662 ***	0.9770 ***
	(0.60)	(0.76)	(2.88)	(2.91)
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
N	1079	1079	5325	5325
R2	0.7460	0.7509	0.7357	0.7356

Note: The t-statistics are in parentheses. ***, **, and * represent the 1%, 5%, and 10% significance levels, respectively.

.

Table 10 presents the test results for the state-owned shareholding ratio. Columns (1) and (2) explore the impact of customer concentration and digital transformation on green process innovation under high state-owned shareholding. According to the results in Columns (1) and (2), customer concentration significantly inhibits green innovation in high-tech manufacturing firms, while digital transformation significantly promotes green innovation outcomes. Notably, under high state-owned shareholding, digital transformation significantly moderates the negative impact of customer concentration on green innovation. Columns (3) and (4) explore the impact of customer concentration and digital transformation on green process innovation under low state ownership. According to the results in Columns (3) and (4), digital transformation still significantly promotes green innovation in high-tech manufacturing firms, but the negative impact of customer concentration is not significant. Additionally, digital transformation does not effectively moderate the relationship between customer concentration and green innovation. According to the results in Table 10, increased state-owned shareholding substantially enhances the inhibitory effect of customer concentration on green innovation decisions, while decreased state-owned shareholding reduces high-tech manufacturing firms’ dependency on major customer demands, allowing these firms to focus more on the environmental benefits brought by green innovation.

During corporate operations, the choice of auditing firms directly influences the compliance and rationality of managerial decisions. For high-tech manufacturing firms, decisions related to green innovation reflect internal innovation resource allocation and determine the future development path. In this context, internal regulatory intensity can directly constrain high-tech manufacturing firms’ green innovation, as firms must consider various factors in green innovation decisions, including success rates and returns. Therefore, we divide the research sample based on the choice of auditing firms into subsamples that chose the Big Four auditing firms and those that did not. The empirical results are shown in

Table 11. The results of audit company.

	Green Innovation
	Big 4 Audit		Non-Big 4 Audit
	(1)	(2)	(3)	(4)
Customer	−0.5834	−0.0346	−0.1734 **	−0.3009 ***
	(−1.22)	(−0.05)	(−2.15)	(−3.25)
Digital	0.0887	0.1624	0.1007 ***	0.0603 ***
	(1.42)	(1.64)	(7.53)	(3.06)
Customer × Digital		−0.2363		0.1356 ***
		(−0.96)		(2.80)
Age	−1.0323	−0.9956	−0.1431	−0.1416
	(−1.37)	(−1.32)	(−1.23)	(−1.22)
Leverage	1.4338 ***	1.5260 ***	0.4218 ***	0.4315 ***
	(3.00)	(3.13)	(5.00)	(5.12)
ROA	2.7168 **	2.8765 **	0.5232 ***	0.5301 ***
	(2.23)	(2.34)	(3.79)	(3.84)
Return	−0.0230	−0.0358	0.0286	0.0279
	(−0.23)	(−0.35)	(1.31)	(1.28)
Turnover	0.0020	0.0042	0.0077 **	0.0075 **
	(0.09)	(0.19)	(2.20)	(2.14)
First	−0.4247	−0.2621	−0.2764	−0.2817
	(−0.55)	(−0.33)	(−1.51)	(−1.54)
Institution	0.1470	0.1568	0.4109 ***	0.4155 ***
	(0.29)	(0.31)	(4.17)	(4.22)
State	0.2166	0.2220	0.0085	0.0155
	(0.54)	(0.56)	(0.08)	(0.14)
Constants	3.4773 *	3.1146	1.0087 ***	1.0369 ***
	(1.72)	(1.51)	(3.31)	(3.40)
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
N	255	255	6283	6283
R2	0.8633	0.8632	0.7217	0.7220

Note: The t-statistics are in parentheses. ***, **, and * represent the 1%, 5%, and 10% significance levels, respectively.

.

Table 11 presents the test results for the choice of auditing firms. Columns (1) and (2) explore the impact of customer concentration and digital transformation on green process innovation under the condition of choosing the Big Four auditing firms. According to the results in Columns (1) and (2), customer concentration and digital transformation do not significantly impact green innovation in high-tech manufacturing firms. Columns (3) and (4) explore the impact of customer concentration and digital transformation on green process innovation under the condition of not choosing the Big Four auditing firms. According to the results in Columns (3) and (4), customer concentration significantly inhibits green innovation in high-tech manufacturing firms, while digital transformation significantly promotes green innovation outcomes. Additionally, digital transformation significantly moderates the negative correlation between customer concentration and green innovation, indicating that reduced internal regulatory intensity enhances high-tech manufacturing firms’ resource investment in green innovation. According to the results in Table 10, the choice of auditing firms can alter high-tech manufacturing firms’ efforts in green innovation, reflected in both core customer dependency and attention to digital technologies.

4.5.3. The Test of Manager Characteristics

Manager characteristics play a crucial role in the decision-making processes related to green innovation in high-tech manufacturing firms. Managers with different backgrounds, mindsets, and values can significantly influence the implementation and success of green innovation projects. This study focuses on two key managerial characteristics: financial background and opportunism, and how these factors impact green innovation in high-tech manufacturing firms under different governance conditions. Managers with financial backgrounds typically exhibit stronger resource allocation and risk management capabilities, enabling them to effectively identify and invest in green innovation projects. Such managers are inclined toward long-term planning, ensuring sustainable development while also achieving economic returns. Their financial expertise allows them to precisely evaluate the economic potential of green innovation projects, optimize resource allocation, and improve investment efficiency. Additionally, they are adept at using various financial tools and market opportunities to secure funding and reduce financing costs, promoting continuous investment in green technology R&D. In contrast, managers with opportunism often pursue short-term returns, leading to insufficient investment in green innovation projects, which typically require long-term R&D and market cultivation. This short-sighted approach results in inadequate funding and resources for green technology R&D, hindering the progress and overall competitiveness of green innovation projects.

Managers with financial backgrounds focus more on long-term returns and the value of the company’s image when making decisions about technological innovation. This characteristic can impact green innovation in high-tech manufacturing enterprises in three ways. Firstly, their financial knowledge and experience enable them to precisely identify and evaluate the economic potential of green innovation projects, optimizing resource allocation and improving investment efficiency. Secondly, these managers are skilled at using various financial tools and market opportunities to raise funds for green innovation projects and effectively reduce financing costs, ensuring smooth project progress. Thirdly, managers with financial backgrounds usually have strategic vision and long-term planning abilities, integrating green innovation into the overall development strategy of the enterprise, promoting continuous investment in green technology R&D and application, and enhancing the market competitiveness and sustainable development capabilities of the enterprise. Therefore, we divide the research sample based on whether the managers have financial backgrounds into subsamples. The empirical results are shown in

Table 12. The results of managers’ financial background.

	Green Innovation
	Financial Background		Non-Financial Background
	(1)	(2)	(3)	(4)
Customer	−0.2223 **	−0.3226 ***	−0.1689	−0.1454
	(−2.29)	(−2.88)	(−1.09)	(−0.81)
Digital	0.1226 ***	0.0927 ***	0.0460	0.0538
	(7.87)	(4.04)	(1.64)	(1.30)
Customer × Digital		0.0983 *		−0.0276
		(1.78)		(−0.25)
Age	−0.3241 **	−0.3198 **	−0.2950	−0.2976
	(−2.30)	(−2.27)	(−1.17)	(−1.18)
Leverage	0.4732 ***	0.4833 ***	0.2034	0.2038
	(4.81)	(4.90)	(1.04)	(1.04)
ROA	0.5548 ***	0.5606 ***	0.4866	0.4868
	(3.54)	(3.58)	(1.58)	(1.58)
Return	0.0391	0.0392	−0.0031	−0.0029
	(1.54)	(1.54)	(−0.07)	(−0.07)
Turnover	0.0074 *	0.0071 *	0.0021	0.0020
	(1.77)	(1.71)	(0.29)	(0.29)
First	−0.5793 ***	−0.5773 ***	−0.1962	−0.1914
	(−2.71)	(−2.70)	(−0.52)	(−0.51)
Institution	0.4215 ***	0.4236 ***	0.5879 ***	0.5883 ***
	(3.68)	(3.70)	(2.87)	(2.87)
State	0.1474	0.1532	−0.0929	−0.0923
	(1.18)	(1.23)	(−0.41)	(−0.40)
Constants	1.5229 ***	1.5354 ***	1.4973 **	1.4957 **
	(4.09)	(4.12)	(2.29)	(2.29)
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
N	4407	4407	1982	1982
R2	0.7558	0.7559	0.7023	0.7021

Note: The t-statistics are in parentheses. ***, **, and * represent the 1%, 5%, and 10% significance levels, respectively.

.

Table 12 presents the test results for managers with financial backgrounds. Columns (1) and (2) explore the impact of customer concentration and digital transformation on green process innovation under the condition of managers having financial backgrounds. According to the results in Columns (1) and (2), customer concentration significantly inhibits green innovation in high-tech manufacturing firms, while digital transformation significantly promotes green innovation outcomes. Furthermore, digital transformation significantly moderates the inhibitory effect of customer concentration on green innovation. Columns (3) and (4) examine the impact of customer concentration and digital transformation on green process innovation under the condition of managers lacking financial backgrounds. In these cases, customer concentration and digital transformation do not significantly impact green innovation. The results in Table 12 indicate that managers with financial backgrounds can better balance technological development and customer demands, leading to more effective resource investment in green innovation.

Opportunism in managers often inhibits green innovation in high-tech manufacturing firms. Managers with high opportunism prioritize short-term profits over long-term sustainability, resulting in inadequate investment in green innovation projects that require significant R&D and market development. This short-sighted behavior results in inadequate funding and resources for green technology R&D and application, hindering the smooth progress of green innovation projects. Additionally, opportunism may lead firms to choose projects that yield quick results, ignoring the importance of long-term sustainable development, further inhibiting the progress of green innovation and the overall competitiveness of enterprises. Therefore, we measure managers’ opportunism using the extent of earnings management and divide the research sample based on the median of this variable into high earnings management and low earnings management subsamples. The empirical results are shown in

Table 13. The results of earnings management.

	Green Innovation
	Earnings Management High		Earnings Management Low
	(1)	(2)	(3)	(4)
Customer	−0.2225 *	−0.2820 **	−0.1282	−0.2936 **
	(−1.89)	(−2.12)	(−1.08)	(−2.10)
Digital	0.0987 ***	0.0766 **	0.1100 ***	0.0667 **
	(4.88)	(2.50)	(5.72)	(2.42)
Customer × Digital		0.0676		0.1580 **
		(0.96)		(2.20)
Age	−0.2065	−0.2068	−0.3030 *	−0.3024 *
	(−1.17)	(−1.17)	(−1.78)	(−1.78)
Leverage	0.4316 ***	0.4354 ***	0.5811 ***	0.5918 ***
	(3.74)	(3.77)	(4.30)	(4.38)
ROA	0.5357 ***	0.5414 ***	0.7350 ***	0.7371 ***
	(2.94)	(2.97)	(2.91)	(2.92)
Return	0.0353	0.0350	0.0139	0.0138
	(1.11)	(1.10)	(0.42)	(0.41)
Turnover	0.0116 **	0.0115 **	0.0007	0.0001
	(2.18)	(2.17)	(0.13)	(0.02)
First	−0.0311	−0.0344	−0.7573 ***	−0.7751 ***
	(−0.12)	(−0.13)	(−2.87)	(−2.94)
Institution	0.4132 ***	0.4185 ***	0.2875 **	0.2797 *
	(2.87)	(2.90)	(1.97)	(1.92)
State	−0.0083	0.0007	0.2313	0.2353
	(−0.05)	(0.00)	(1.54)	(1.57)
Constants	1.0248 **	1.0410 **	1.6293 ***	1.6769 ***
	(2.20)	(2.24)	(3.64)	(3.74)
Firm FE	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes
N	3150	3150	3208	3208
R2	0.7103	0.7103	0.7503	0.7507

Note: The t-statistics are in parentheses. ***, **, and * represent the 1%, 5%, and 10% significance levels, respectively.

.

Table 13 presents the test results for earnings management. Columns (1) and (2) explore the impact of customer concentration and digital transformation on green process innovation under high earnings management. The results indicate a significant negative correlation between customer concentration and green innovation, while digital transformation significantly promotes green innovation outcomes. However, increased earnings management reduces the moderating effect of digital transformation, making digital technologies less effective in reducing dependency on major customers. Columns (3) and (4) examine the impact of customer concentration and digital transformation on green process innovation under low earnings management. In these cases, digital transformation still significantly promotes green innovation, but the relationship between customer concentration and green innovation is not significant. Digital transformation significantly moderates the inhibitory effect of customer concentration on green innovation, indicating that digital technologies can reduce dependency on major customers. The results in Table 13 suggest that opportunism increases high-tech manufacturing firms’ dependency on major customers and reduces managers’ attention to digital technologies, thereby inhibiting green innovation.

5. Discussion

Green innovation serves as both a pivotal mechanism for enterprises to achieve sustainable development and a critical catalyst for industrial transformation and upgrading. Existing research primarily explores the influencing factors of green innovation from the perspectives of external environmental pressures and internal resource constraints, systematically explaining the core motivations for enterprises to invest in green innovation [2,7]. In the realm of high-tech manufacturing, green innovation not only propels regional economic sustainability but also fosters industrial upgrading, marking a transition from labor-intensive to knowledge-intensive operations. In this context, exploring the factors influencing green innovation in high-tech manufacturing firms is particularly important for balancing environmental protection and economic development.

This study combines theoretical and empirical analysis to deeply investigate the relationships among customer concentration, digital transformation, and green innovation in high-tech manufacturing firms. According to the empirical results obtained in Section 4, there is a significant negative correlation between customer concentration and green innovation. This suggests that the dependency of high-tech manufacturing firms on a few major customers acts as a barrier to the advent of green innovation. Given the operational characteristics of high-tech manufacturing firms, these enterprises need to maintain technological advancement while ensuring the profitability of product sales, further leading to a supply–demand game scenario with major customers. Referencing the findings of Ni et al. (2023), customer concentration may exacerbate financial constraints faced by enterprises, intensifying the negative impact of limited internal resources on production and operations in high-tech manufacturing firms [11]. These financial constraints mean that investments in green innovation are likely tailored to meet the immediate needs of core customers, elucidating why an increase in customer concentration reduces green innovation output [34]. The inhibitory effect of customer concentration can be further explained through its impact on green product innovation. High-tech manufacturing firms must thoroughly consider the actual needs of major customers, making product-related green innovation more susceptible to the influence of existing product demands from these customers [35].

Meanwhile, there is a significant positive correlation between digital transformation and green innovation, indicating that the more high-tech manufacturing firms focus on digital technologies, the more resources they will invest in green innovation. Considering the characteristics of the production process, high-tech manufacturing firms need to use digital technologies to upgrade production lines and accurately predict market demand changes. These needs enhance the attention to digital technologies. In terms of internal resource allocation, digital transformation improves resource utilization efficiency, allowing more resources to be invested in green technology R&D, supporting the findings of Upadhayay et al. (2024) [19]. According to the Resource-Based View, the role of digital transformation in improving resource utilization efficiency offsets the resource bias generated to meet major customer demands, ensuring adequate resource allocation for green technology R&D [18]. Further examinations explore the channels through which customer concentration and digital transformation impact green innovation from three perspectives: external environment, internal governance, and managerial characteristics. In scenarios of weakened market competition intensity or lack of government support, high-tech manufacturing firms rely more on the optimization of resource allocation through digital transformation, reducing the impact of major customer demands on green innovation. However, enterprises receiving government support for green innovation are not influenced by major customer demands or digital technologies, highlighting the significant role of political resources in promoting green innovation outcomes. Furthermore, a reduction in state ownership mitigates the inhibitory effect of customer concentration on green innovation, indicating that increased corporate governance freedom helps enterprises reduce heavy reliance on major customer demands. Notably, the baseline test results are only significant for enterprises not choosing the Big Four auditing firms, suggesting that green innovation decisions are influenced by the intensity of internal supervision. For enterprises choosing the Big Four auditing firms, decision-making processes are more affected by auditing work, severely restricting green innovation-related decisions. Additionally, managers with financial backgrounds are more likely to understand the potential value of green innovation outcomes, making them more willing to use digital technologies to promote green innovation R&D. Compared to managers with high opportunism tendencies, those with lower opportunism tendencies are more inclined to use digital transformation to alleviate the inhibitory effect of major customer reliance on green innovation outcomes.

6. Conclusions

As a key driver of economic transformation, green innovation in high-tech manufacturing firms can significantly impact industrial upgrading and environmental protection. Encouraging these firms to continuously pursue green innovation is essential for sustainable economic development. This study explores the influence of customer concentration and digital transformation on green innovation in high-tech manufacturing firms, revealing that green innovation is a resource-driven form of innovation. Unlike existing studies that examine the influencing factors of green innovation from the perspectives of external environment and internal governance, our research emphasizes the significant role of resource allocation in the decision-making process related to green innovation.

Given the slow returns and long R&D cycles characteristic of green innovation, substantial resources are needed to support these R&D activities. According to the Resource-Based View, investments in green innovation by high-tech manufacturing firms need to consider resource allocation issues, indicating that factors related to internal resources may more easily influence green innovation output. In this context, this study investigates the market and technological factors influencing green innovation in high-tech manufacturing firms, finding that customer concentration hinders the emergence of green innovation outcomes, while digital transformation consistently drives resource investment in green innovation. Customer concentration reflects the degree of reliance on the changing demands of major customers, pertaining to resource allocation in product production and sales. Digital transformation reflects the focus on digital technologies, relating to resource allocation in technological upgrades. Notably, the application of digital technologies can mitigate the inhibitory effect of customer concentration on green innovation, suggesting that high-tech manufacturing firms prioritize the potential benefits of technological upgrades, consistent with their technological attributes. Further examination reveals three perspectives on how customer concentration and digital transformation influence green innovation. Externally, market competition and government support alter the dependence of high-tech manufacturing firms on major customers, enabling digital transformation to better facilitate green innovation through resource allocation. Internally, ownership structure and the intensity of internal regulation change the importance high-tech manufacturing firms place on green innovation, with improved corporate governance potentially reducing the impact of major customer demands. Managerially, rational managers prioritize the influence of digital transformation on green innovation, whereas opportunistic managers rely more on the actual demands of major customers, hindering their understanding of the environmental and economic benefits of green innovation.

This study offers practical and policy implications. First, developing countries should place greater emphasis on green innovation in high-tech manufacturing firms. Unlike traditional manufacturing firms, high-tech manufacturing firms possess strong R&D capabilities and technological reserves, providing robust support for green innovation. Developing countries need to leverage the green transformation of high-tech manufacturing to achieve long-term regional sustainable economic development. Second, green innovation in high-tech manufacturing firms is a collective outcome from an industrial chain perspective. Compared to other industries, the demands of major customers more easily influence the production and sales processes of high-tech manufacturing firms, making green innovation decisions highly correlated with customers’ green beliefs. Lastly, resource investment in green innovation by high-tech manufacturing firms is driven by digital technologies. With the advantages of digital technologies, high-tech manufacturing firms can not only coordinate supply and demand relationships with customers but also optimize resource allocation efficiency, providing essential technical support for the continuous emergence of green innovation.

There are two limitations in this study. First, the empirical analysis primarily relies on data from high-tech manufacturing firms in China, spanning from 2007 to 2021. While this provides comprehensive insights into the Chinese context, the findings may not be entirely generalizable to other geographical regions or different economic contexts. High-tech industries vary significantly across countries due to differing economic policies, industrial support, and technological advancements. Second, the research design may not adequately account for all external factors influencing green innovation in high-tech manufacturing. Factors such as global economic shifts, international trade policies, and cross-border technological transfers can significantly impact innovation trends. While the study considers some external elements like government policies and market competition, other potentially influential factors, such as international environmental agreements or global market crises, are not extensively explored. Future research could undertake comparative studies across different countries to enhance the generalizability of the findings. By comparing how high-tech manufacturing firms in various regions—such as North America, Europe, and other parts of Asia—respond to customer concentration and digital transformation, researchers can explore the influence of distinct economic policies, cultural factors, and market conditions on green innovation. Furthermore, there is a need to integrate broader global factors that influence green innovation, such as international trade policies, environmental regulations, and economic downturns. Some future studies could examine how these external pressures interact with internal strategies to influence innovation patterns.