Does Proactive Green Technology Innovation Improve Financial Performance? Evidence from Listed Companies with Semiconductor Concepts Stock in China

Abstract

Academia and business alike are paying increasing attention to innovation in green technology due to the potential environmental and financial performance benefits. However, a limited amount of research has been carried out on the effect of proactive green technology innovation on corporate financial performance. This study examines the effects of two dimensions of proactive green technology innovation, namely, proactive green process innovation and proactive green product innovation, on corporate financial performance. Moreover, the moderating role of absorptive capacity on these relationships is introduced. The proposed hypotheses were tested empirically using a dynamic panel dataset of 126 Chinese listed semiconductor concept stocks from 2010 to 2020 and a difference-GMM approach. It was found that proactive green process innovation has a significant positive effect on both short-term and long-term corporate financial performance. Moreover, proactive green product innovation has a significant positive effect on long-term corporate financial performance. However, it does not improve short-term corporate financial performance. In addition, absorptive capacity has a positive moderating effect on the relationship between proactive green process innovation and both short-term and long-term corporate financial performance, and shows a positive moderating effect on the relationship between proactive green product innovation and long-term financial performance. However, it has a significant negative moderating effect on short-term corporate financial performance. Thus, we suggest that firms adopt more supportive proactive green technology innovation practices in order to improve their financial performance.

1. Introduction

With the evolution and development of industry and the increasing public awareness of environmental issues, green technology innovation has received a great deal of attention. There is growing scientific evidence that a firm’s external pressure has increased from climate change, social development, and environmental degradation [1,2]. At the same time, firms are facing increasing competition due to globalization and the emergence of new technologies [3]. This dual environmental and competitive pressure has increased corporate focus on ‘greener’ and ‘more sustainable’ production, and has led to a focus on whether green technology innovation can address both issues simultaneously to reduce environmental consequences and improve financial performance [4]. Thus, the relationship between green technology innovation and corporate financial performance has attracted significant interest among scholars in recent years [5,6], and the literature on this relationship is rapidly growing. For example, scholars have discussed the relationship between firms’ sustainability [7], environmental strategies [8], eco-innovation [9], green innovation [10], corporate social responsibility [11], and corporate financial performance. Other scholars have examined the antecedents and consequences of the relationship between green technology innovation and financial performance, such as risk-taking level [12], the pressure of environmental regulations [13,14], and shareholder structure heterogeneity [15,16]. These existing studies are critical for business managers seeking to leverage sustainability innovations [17]. However, previous studies indicate that the link between green technology innovation and corporate financial performance is a controversial issue in the scientific community. Several studies have suggested that this relationship is a positive one. Green technology innovation is defined as innovation that includes pollution prevention, waste recycling and utilization, development of green products, and the treatment of the environment [18]. It generally involves producing a ‘double dividend’ in the form of benefits to consumers or businesses and a significant reduction in environmental damage [19]; in other words, benefiting the environment while helping to improve corporate financial performance [20]. Other studies, however, have suggested that this relationship is neutral or even negative [21] and that it may be U-shaped [22] or inverted U-shaped [23]. The inconsistency of these findings leaves room for further research.

In addition, the existing literature has begun to focus on the classification of green technology innovation, such as proactive green technology innovation and reactive green technology innovation [24]. However, the main focus is on the relationship between reactive and proactive environmental strategies, environmental performance, and competitive advantage [25,26,27]. Thus, to the best of our knowledge, there is a research gap between proactive green technology innovation and corporate financial performance, especially in the context of the semiconductor industry. It is important to focus on exploring the effect of proactive green technology innovation on corporate financial performance. A firm’s attitude towards implementing green technology innovation will influence its organizational model, ultimately affecting its financial performance [28]. Therefore, considering the limited literature, our research specifically focuses on mapping the relationship between proactive green technology innovation and corporate financial performance. It contributes to the literature from the following four perspectives.

First, based on the existing conceptual division of green technology innovation into the dimensions of green process innovation and green product innovation [29,30,31], we consider the segmentation effects of these two dimensions of proactive green technology innovation on corporate financial performance, as this division has arguably been underutilized in previous studies [32]. This adds to the existing theoretical research and provides more exemplary granular insights on the relationship between green technology innovation and corporate financial performance. Likewise, relatively little is known about the extent to which a firm’s internal capabilities influence proactive green process innovation and proactive green product innovation [33]. Thus, we introduce the moderating role of absorptive capacity to examine these relationships, expanding the understanding of how firms should improve their financial performance through proactive green technology innovation. Second, previous studies which have examined the relationship between green technology innovation and corporate financial performance only use corporate financial performance as a single dimension [34,35]. We further divide corporate financial performance into short-term and long-term performance in order to measure their relationship with proactive green technology innovation. Third, unlike previous studies that have primarily used subjective data obtained through questionnaires and focused on heavily polluting industries [36], we focus on the semiconductor industry and use a secondary dataset in order to test and mitigate existing research gaps by targeting dynamic panel data on Chinese listed semiconductor concept stocks between 2010 and 2020. As observed, the relationship between proactive green technology innovation and corporate financial performance is a process that occurs over time, resulting in a dynamic rather than a static context. Thus, dynamic panel data enables use of the available data to explore the complex relationship between proactive green technology innovation and corporate financial performance at the firm level. Finally, we use a difference-generalized method-of-moments estimation (GMM) approach. Previous studies have often used methods such as ordinary least squares [37], structural equation modeling [38], and hierarchical regression analysis [39]. However, the GMM method is seldom used to measure and examine the relationships between proactive green technology and corporate financial performance. The method is remarkable because it constructs instruments for the endogenous variables to address endogeneity concerns among them. In brief, our study uses dynamic panel data on Chinese listed semiconductor concept stocks and applying the Difference-GMM approach. We contribute to the literature by examining how a proactive green technology innovation tool that forces firms to be more innovative to continue working in a very competitive international market affects the firms’ financial performance. Our findings reveal that proactive green process innovation and proactive green product innovation indeed have different impacts on a firm’s financial performance in both the long and short term, and that absorptive capacity has a positive moderating effect on the relationships.

The rest of the paper is structured as follows: Section 2 conducts a literature review and develops hypotheses; Section 3 presents the research methodology and data sources; Section 4 reports the empirical results and analysis; and Section 5 summarizes and discusses the main findings and provides managerial implications, limitations, and future research directions.

2. Literature Review and Hypothesis Development

2.1. The Dimension of Proactive Green Technology Innovation

Chen et al. (2012) [32] define proactiveness as taking proactive action to create advantage. “Proactive green technology innovation” means actively pursuing environmentally relevant innovations. It aims to develop new technologies or products before competitors in order to achieve market leadership and gain competitive advantage [40]. Scholars [9] have suggested that proactive eco-innovation comprises three dimensions: product, process, and technology. Therefore, we believe that proactive green technology innovation can be divided into at least two main dimensions: proactive green process innovation and proactive green product innovation. Proactive green process innovation mainly emphasizes innovative approaches in the production process, such as reducing process-related environmental pollution and controlling the energy consumption [41]. On the other hand, proactive green product innovation involves pro-environmental products [42]; it is widely accepted that green product innovations result in environmentally friendly products which extend protection of the environment throughout the product life cycle, from development to distribution [43].

2.2. The Dimension of Corporate Financial Performance

A firm’s performance can take various forms and manifestations. Among them, financial performance is widely used in the strategic research on firms. The most critical dimension in this approach to financial performance is the division into short-term financial performance and long-term financial performance. The short-term financial performance dimension mainly measures a firm’s past and present financial situation [44]. This dimension is very appropriate when a firm wants to evaluate the results of its past and present actions [45]. In contrast, the long-term financial performance dimension focuses on a firm’s future value, and is often linked to shareholder interests [46]. In the literature, accounting-based measures of corporate financial performance consider past or short-term financial performance. On the other hand, market-based measures capture a firm’s future or long-term financial performance [47,48].

Numerous indicators measure a firm’s short-term financial performance from an accounting measures-based perspective. In the existing literature, common indicators used to measure short-term corporate financial performance are ROA and ROE [49]. To a certain extent, these reflect the current and short-term profitability of a firm [24]; ROIC [50], operating margin [51], sales growth rate [52], and net profit [26] are other accounting-based measures of short-term corporate financial performance.

In previous research, scholars have argued that market-based measures can capture future or long-term financial performance, which is widely accepted [45,47,48]. For example, many scholars use Tobin’s Q as an indicator of long-term corporate financial performance [9,53]. Tobin’s Q is the ratio of the market value of an asset to its replacement value. It focuses on whether the market value of a firm’s asset is overvalued or undervalued from a risk perspective. In addition, several scholars have argued that market value (MV) can be measured by the market value of a firm’s stock. MV is a virtual value that reflects investors’ expectations about future earnings. These expectations come from the evaluation and perception of the tangible and intangible assets of a firm [54,55]. In addition, long-term corporate financial performance can be expressed by the price-to-earnings ratio (P/E) and the ratio of market value to book value (MV/BV) [37].

2.3. Proactive Green Process Innovation and Corporate Financial Performance

The “Resource-Based View” (RBV) proposes that green process innovation has a first-mover advantage and can improve a firm’s green image [56,57]. For example, green process innovation can achieve environmental compliance by improving a firm’s existing production processes or adding new processes, as well as by gaining competitive differentiation, although proactive green process innovation often requires firms to systematically improve their production processes, requiring them to invest more capital or incur higher implementation costs [58]. However, from a financial perspective market investors may be willing to pay a premium for firms with a green image. At the same time, the public may be willing to pay a higher price for environmentally friendly products. This green approach has proven to be effective. In addition, Xie et al. (2016) [59] found that green process innovation is positively associated with corporate financial performance. Overall, it is meaningful for firms to actively invest in green process innovation [58]. Therefore, we argue that a firm’s proactive green process innovation improves its financial performance. Based on this, we propose the following hypotheses:

H1: 

A firm’s proactive green process innovation is conducive to enhancing its short-term corporate financial performance.

H2: 

A firm’s proactive green process innovation is conducive to enhancing its long-term corporate financial performance.

2.4. Proactive Green Product Innovation and Corporate Financial Performance

Scholars believe that green product innovation is one of the necessary conditions for achieving environmental sustainability and enables the growth of business performance [42]. Green product innovation improves products through technological components or environmentally friendly materials [60]. In order to reduce the environmental effects during the product life cycle, firms should invest in green product innovation to meet the environmental needs of their markets or customers [61]. Therefore, proactive green product innovation plays a vital role in developing a firm’s green capabilities, strengthening its green image, and improving its financial performance [62] Furthermore, the literature shows that green product innovation has a positive impact on firm profitability and market values [19], and that the more innovative a new green product is, the greater its financial performance [43]. These include cost savings [63], increased sales [62], increased Operating Margin (OM) [51]), more profits [26], higher return on investment (ROI) [50], higher return on total assets (ROA) [6] and return on equity (ROE) [64], a better price/earnings ratio (P/E) [37], higher market value (MV), [65] and better Tobin’s Q [53]. Accordingly, we argue that firms proactively developing innovative green products will achieve better financial and market performance. We propose the following hypotheses:

H3: 

A firm’s proactive green product innovation is conducive to enhancing its short-term corporate financial performance.

H4: 

A firm’s proactive green product innovation is conducive to enhancing its long-term corporate financial performance.

2.5. The Moderating Role of Absorptive Capacity

Absorptive capacity refers to a firm’s ability to harness the value of new information and apply it to business activities. Absorptive capacity thus helps transform knowledge into new products, services, or processes to support innovation. It can be argued that maintaining and improving absorptive capacity is critical to a firm’s sustainability. Therefore, absorptive capacity is often associated with innovation [66]. In their proactive green technology innovation, firms need to learn from partners, acquire external information, and integrate it into their existing knowledge base [67]. Similarly, absorptive capacity is associated with proactive green technology innovation. From a resource-based review, if a firm has valuable yet scarce resources and fully utilizes its absorptive capacity, this usually affects its competitive advantage and performance in a positive way. Absorptive capacity can help increase the speed, frequency, and scale of innovation and new knowledge to fulfil the additional knowledge requirements of firms in the process of proactive green technology innovation [68], providing a unique competitive advantage and obtaining better financial performance. It is necessary to determine the role of absorptive capacity in the relationship between proactive green technology innovation and corporate financial performance [69]; thus, we introduced the role of absorptive capacity as a moderator of the relationship between proactive green technology innovation and corporate financial performance. We hypothesize that absorptive capacity will positively moderate this relationship. The following hypotheses are proposed:

H5a: 

Absorptive capacity positively moderates the relationship between proactive green process innovation and short-term corporate financial performance.

H5b: 

Absorptive capacity positively moderates the relationship between proactive green process innovation and long-term corporate financial performance.

H5c: 

Absorptive capacity positively moderates the relationship between proactive green product innovation and short-term corporate financial performance.

H5d: 

Absorptive capacity positively moderates the relationship between proactive green product innovation and long-term corporate financial performance.

We constructed our hypotheses as a conceptual model, as shown in Figure 1.

3. Research Methods

3.1. Sample Selection and Data Resources

3.1.1. Sample Selection

Our study focuses on the impact of proactive green technology innovation on corporate financial performance in Chinese listed semiconductor concept stocks. In the securities market, semiconductor concept stocks are semiconductor listed with an energy-saving and environmental protection theme. First, according to the business descriptions of different classifications in the Classification of National Economic Industries (GB/T4754-2017), C class manufacturing industry, containing all semiconductor concept stock firms, was locked. Second, by manually searching the concept stock sector ‘semiconductor concept stocks’ in China’s Shanghai Stock Exchange and Shenzhen Stock Exchange and then combining the results with the existing CSIA member directory, we obtained 210 listed companies with semiconductor concept stocks; firms with missing data during 2010–2020 were then excluded. Ultimately, for this paper we obtained a sample of 126 Chinese listed semiconductor concept stocks, and examined dynamic panel data consisting of information related to these.

3.1.2. Data Resources

Data on proactive green technology innovation, including proactive green process innovation and proactive green product innovation, were obtained from the China Stock Market Accounting Research (CSMAR) and WIND databases. Moreover, corporate financial performance and absorptive capacity data were obtained from the CSMAR database.

3.2. Variables Measurement

3.2.1. Dependent Variables

Drawing on the division of the research literature [58], our study divides proactive green technology innovation into proactive green process innovation and proactive green product innovation. For the measurement variable of proactive green process innovation, we selected the environmental protection investment scale (EPIS) [9], that is, EPIS = Environmental protection investment amount/Capital stock. For the measurement variable of proactive green product innovation, we selected the total number of green patents (GPT) [39], that is, GPT = Green patent applications + Green patent grants.

3.2.2. Independent Variables

As per the existing literature [50], we used two-dimensional short-term and long-term indicators to represent corporate financial performance as independent variables for measurement of short-term financial performance. Consistent with the previous literature [70], we used the basic accounting measurements Return on Assets (ROA) and Return on Equity (ROE) for short-term corporate financial performance. These indicators help to examine the relationship between green innovation and corporate financial performance, as they denote the ratio of net income to total assets and net income to shareholders’ equity, respectively. For measurement of long-term financial performance, consistent with the previous literature [37,55] we selected market-based indicators to measure long-term corporate financial performance, namely, Tobin’s Q, Market Value (MV), and price-to-earnings ratio (P/E).

3.2.3. Moderating Variables

We introduced absorptive capacity as a moderating variable. In the existing literature, data on research and development expenditures and patents are the most widely used measures of absorptive capacity [71,72]. Similarly, Xie et al. (2016) [59] used total research and development expenditures as a proxy for absorptive. Thus, in this study, we used research and development intensity as proxy variables for absorptive capacity, measured by the ratio of research and development expenses divided by net sales.

3.2.4. Control Variables

In addition to the above variables, we controlled for five variables that can affect corporate financial performance. Firm Size (Size): a firm’s size is a potential factor in its financial performance [21]; in order to control for any potential side effect, we included a measure of firm size, which was measured by Log of total assets in the firm (ln TA). Firm age (Age): we measured firm age by the number of years from its founding [60]. Financial leverage (LEV): previous studies have widely used this measure [73], which represents debt or non-compliance risk; we measured it using the total liabilities/total assets ratio. Capital intensity (CAP INT): capital intensity is included in the model because it affects corporate financial performance [71]; we measured it as Capital intensity = Total Assets/Operating Income. Equity Concentration (Top1): consistent with Chen and Ma (2021) [64], we measured equity concentration by calculating the number of shareholding according to the first largest shareholder/total shares.

The summary of measurement and definition of all variables are shown in

Table 1. Summary of the measurements of the variables used in this study.

Variables	Names	Measurement	Symbols	Definition
Independent Variables	Proactive green process innovation	Environmental Protection Investment Scale	EPIS	Environmental protection investment amount/Capital stock
	Proactive green product innovation	Green Patents Total	GPT	GPT = Green patent applications + Green patent grants
Dependent Variables	Short-term corporate financial performance	Return of Assets	ROA	ROA = Net income/Average balance of total assets
		Return of Equity	ROE	ROE = Net income/Average balance of shareholders’ equity
	Long-term corporate financial performance	Tobin’s Q	Tobin’s Q	Tobin’s Q = MV/Total assets
		Market Value	MV	MV = Total equity * shares at today’s closing price + current ending value of total liabilities
		Price/earning	PE	PE = Market price per share/Earnings per share
Moderator Variables	Absorptive Capacity	R&D Intensity	R&D INT	R&D INT = R&D expenses/Net sales
Control Variables	Firm size	ln TA	Size	Size = Log of Total Assets
	Firm age	Firm age	Age	Age = the number of years since the firm was founded
	Financial leverage	Financial leverage	LEV	LEV = Total liabilities/Total assets
	Capital intensity	Capital intensity	CAP INT	CAP − INT = Total Assets/Operating Income
	Equity Concentration	Top1 Shareholder’s shareholding	Top1	Top1 = the number of shareholding by the first largest shareholder/total shares

Note: * represents the multiplication of items in the table.

.

3.3. Model Construction

To test the impact of proactive green technology innovation on corporate financial performance at the firm level, we conducted an econometric analysis using a panel sample of 126 semiconductor companies from 2010–2020. Due to the cross-sectional and short time period characteristics of the sample data, and considering that the current period’s corporate financial performance is likely to be influenced by the previous period’s corporate financial performance, a lag term for corporate financial performance was added to the dependent variables; thus, we introduce the following dynamic panel model:

$$\begin{array}{c}{Y}_{i,t}=\alpha Y_{i,t-1}+\beta X_{i,t}+u_i+{ϵ}_{i.t} \left(i=1,\dots ,N,t=1,\dots ,T\right)\end{array}$$

(1)

where $Y_{i,t}$ denotes the corporate financial performance of an individual firm i in period t, $Y_{i,t-1}$ is a lagged corporate financial performance term, $X_{i,t}$ denotes the vector of all dependent variables, $u_i$ is an unobserved firm-specific fixed effect, and ${ϵ}_{i.t}$ is a disturbance term independent and identically distributed across firms and over time.

A lagged dependent variable allows for a dynamic process, which may be crucial for recovering the validity of other parameters as well as for consistent estimates. Thus, a first-order difference transformation to (1) can be made to eliminate unobserved individual fixed effects and the dynamic model reformulated as follows:

$$\begin{array}{c}\Delta Y_{i,t}=\alpha \Delta Y_{i,t-1}+\beta \Delta X_{i,t}+\Delta {ϵ}_{i.t} \left(i=1,\dots ,N,t=1,\dots ,T\right)\end{array}$$

(2)

where $\Delta Y_{i, t}$ and $\Delta Y_{i,t-1}$ denote, respectively, the first-order difference of corporate financial performance and the lagged corporate financial performance of an individual firm i in period t, $\Delta X_{i,t}$ denotes the vector of the first-order difference of all dependent variables, and $\Delta {ϵ}_{i.t}$ is the first-order difference of time-specific intercept and disturbance term independent and identically distributed across industries and over time.

Because $\mathsf{\Delta }{Y}_{i, t-1}$ and $\mathsf{\Delta }{ϵ}_{i.t}$ are the correlation, $Y_{i, t-2}$ is used for the instrumental variable as $\mathsf{\Delta }{Y}_{i, t-1}$, that is, the “Anderson–Hsiao estimator”. $Y_{i, t-2}$ is a valid instrumental variable in the absence of autocorrelation. By analogy, using all subsequent variables as instrumental variables, these variables are called “Arellano–Bond” estimators as well.

3.3.1. The First-Order Difference Equation of Direct Effects

To further test the direct effect of proactive green technology innovation on short-term and long-term corporate financial performance, we added control variables to model (2). In addition, year dummy variables were added for fixation in order to avoid endogeneity problems caused by time effects. Thus, we further transformed the first-order difference dynamic panel model.

The following Equations (3) and (4) represent the direct effect of proactive green technology innovation on short-term corporate financial performance.

$$\begin{array}{c}\begin{array}{c}\Delta RO{A}_{i,t}=\alpha \Delta RO{A}_{i,t-1}+{\beta }_1\Delta EPI{S}_{i,t}+{\beta }_2\Delta GP{T}_{i,t}+\\ {\beta }_3\Delta LE{V}_{i,t}+{\beta }_4\Delta {\mathrm{Size}}_{i,t}+{\beta }_5\Delta Ag{e}_{i,t}+{\beta }_6\Delta CAPIN{T}_{i,t}+{\beta }_7\Delta Top{1}_{i,t}+\Delta Yea{r}_i+\Delta {ϵ}_{i.t} \left(i=1,\dots ,N,t=1,\dots ,T\right)\end{array}\end{array}$$

(3)

$$\begin{array}{c}\begin{array}{c}\Delta RO{E}_{i,t}=\alpha \Delta RO{E}_{i,t-1}+{\beta }_1\Delta EPI{S}_{i,t}+{\beta }_2\Delta GP{T}_{i,t}+\\ {\beta }_3\Delta LE{V}_{i,t}+{\beta }_4\Delta {\mathrm{Size}}_{i,t}+{\beta }_5\Delta Ag{e}_{i,t}+{\beta }_6\Delta CAPIN{T}_{i,t}+{\beta }_7\Delta \mathrm{Top}{1}_{i,t}+\Delta Yea{r}_i+\Delta {ϵ}_{i.t} \left(i=1,\dots ,N,t=1,\dots ,T\right)\end{array}\end{array}$$

(4)

The following Equations (5)–(7) represent the direct effect of proactive green technology innovation on long-term corporate financial performance.

$$\begin{array}{c}\begin{array}{c}\Delta \mathrm{Tobinsqi},\mathrm{t}=\alpha \mathsf{\Delta }Tobins{q}_{i, t-1}+{\beta }_1\Delta EPI{S}_{i,t}+{\beta }_2\Delta GP{T}_{i,t}+\\ {\beta }_3\Delta LE{V}_{i,t}+{\beta }_4\Delta {\mathrm{Size}}_{i,t}+{\beta }_5\Delta {\mathrm{Age}}_{i,t}+{\beta }_6\Delta \mathrm{CAP}IN{T}_{i,t}+{\beta }_7\Delta \mathrm{Top}{1}_{i,t}+\Delta Y{\mathrm{ear}}_i+\Delta {ϵ}_{i.t} \left(i=1,\dots ,N,t=1,\dots ,T\right)\end{array}\end{array}$$

(5)

$$\begin{array}{c}\begin{array}{c}\Delta M{V}_{i,t}=\alpha \Delta M{V}_{i,t-1}+{\beta }_1\Delta EPI{S}_{i,t}+{\beta }_2\Delta GP{T}_{i,t}+\\ {\beta }_3\Delta LE{V}_{i,t}+{\beta }_4\Delta {\mathrm{Size}}_{i,t}+{\beta }_5\Delta Ag{e}_{i,t}+{\beta }_6\Delta CAPIN{T}_{i,t}+{\beta }_7\Delta \mathrm{Top}{1}_{i,t}+\Delta Yea{r}_i+\Delta {ϵ}_{i.t} \left(i=1,\dots ,N,t=1,\dots ,T\right)\end{array}\end{array}$$

(6)

$$\begin{array}{c}\begin{array}{c}\Delta P{E}_{i,t}=\alpha \Delta P{E}_{i,t-1}+{\beta }_1\Delta EPI{S}_{i,t}+{\beta }_2\Delta GP{T}_{i,t}+\\ {\beta }_3\Delta LE{V}_{i,t}+{\beta }_4\Delta {\mathrm{Size}}_{i,t}+{\beta }_5\Delta Ag{e}_{i,t}+{\beta }_6\Delta CAPIN{T}_{i,t}+{\beta }_7\Delta \mathrm{Top}{1}_{i,t}+\Delta Yea{r}_i+\Delta {ϵ}_{i.t} \left(i=1,\dots ,N,t=1,\dots ,T\right)\end{array}\end{array}$$

(7)

where $\mathsf{\Delta }RO{A}_{i,t}$ and $\mathsf{\Delta }RO{E}_{i,t}$ all reflect the first-order difference of the short-term corporate financial performance of an individual firm i in period t; $\mathsf{\Delta }RO{A}_{i, t-1}$ and $\mathsf{\Delta }RO{E}_{i, t-1}$ all reflect the first-order difference of the lagged short-term corporate financial performance; $\mathsf{\Delta }Tobins{q}_{i,t}$, $\mathsf{\Delta }M{V}_{i,t}$, and $\mathsf{\Delta }P{E}_{i,t}$ all reflect the vector of the first-order difference of the long-term corporate financial performance; $\mathsf{\Delta }Tobins{q}_{i, t-1}$, $\mathsf{\Delta }M{V}_{i, t-1}$, and $\mathsf{\Delta }P{E}_{i, t-1}$ all reflect the first-order difference of the lagged long-term corporate financial performance; $\mathsf{\Delta }EPI{S}_{i, t}$ and $\mathsf{\Delta }GP{T}_{i, t}$ denote the first-order difference of the green process innovation and green product innovation of an individual firm i in period t, respectively, which implements proactive green technology innovation; $\mathsf{\Delta }LE{V}_{i, t}$, $\mathsf{\Delta }Siz{e}_{i, t}$, ${\mathsf{\Delta }\mathrm{Age}}_{i, t}$, ${\mathsf{\Delta }\mathrm{CAP}\mathrm{INT}}_{i, t}$, and $\mathsf{\Delta }\mathrm{Top}{1}_{i, t}$ are control variables which denote, respectively, the first-order difference of financial leverage, firm size, firm age, capital intensity, and Top1 Shareholder’s shareholding; $\mathsf{\Delta }Yea{r}_i$ denotes the first-order difference of dummy year; and $\Delta {ϵ}_{i.t}$ is the first-order difference of time-specific intercept and disturbance term independent and identically distributed across industry and over time.

3.3.2. The First-Order Difference Equation of the Moderating Effects

We established the following model to examine the moderating role of proactive green technology innovation and corporate financial performance.

The following Equations (8) and (9) represent the moderating role of absorptive capacity on proactive green technology innovation and short-term corporate financial performance.

$$\begin{array}{c}\begin{array}{c}\Delta RO{A}_{i,t}=\alpha \Delta RO{A}_{i,t-1}+{\beta }_1\Delta EPI{S}_{i,t}+{\beta }_2\Delta GP{T}_{i,t}+{\beta }_3\Delta RDIN{T}_{i,t}\ast \Delta EPI{S}_{i,t}+{\beta }_4\Delta RnD IN{T}_{i,t}\ast \Delta GP{T}_{i,t}+\\ {\beta }_5\Delta LE{V}_{i,t}+{\beta }_6\Delta {\mathrm{Size}}_{i,t}+{\beta }_7\Delta Ag{e}_{i,t}+{\beta }_8\Delta CAPIN{T}_{i,t}+{\beta }_9\Delta Top{1}_{i,t}+\Delta Yea{r}_i+\Delta {ϵ}_{i.t} \left(i=1,\dots ,N,t=1,\dots ,T\right)\end{array}\end{array}$$

(8)

$$\begin{array}{c}\begin{array}{c}\Delta RO{E}_{i,t}=\alpha \Delta RO{E}_{i,t-1}+{\beta }_1\Delta EPI{S}_{i,t}+{\beta }_2\Delta GP{T}_{i,t}+{\beta }_3\Delta RDIN{T}_{i,t}\ast \Delta EPI{S}_{i,t}+{\beta }_4\Delta RnD IN{T}_{i,t}\ast \Delta GP{T}_{i,t}+\\ {\beta }_5\Delta LE{V}_{i,t}+{\beta }_6\Delta {\mathrm{Size}}_{i,t}+{\beta }_7\Delta Ag{e}_{i,t}+{\beta }_8\Delta CAPIN{T}_{i,t}+{\beta }_9\Delta Top{1}_{i,t}+\Delta Yea{r}_i+\Delta {ϵ}_{i.t} \left(i=1,\dots ,N,t=1,\dots ,T\right)\end{array}\end{array}$$

(9)

The following Equations (10)–(12) represent the moderating role of absorptive capacity on proactive green technology innovation and long-term corporate financial performance.

$$\begin{array}{c}\begin{array}{c}\Delta {\mathrm{Tobinsq}}_{i,t}=\alpha \Delta \mathrm{Tobins}{q}_{i,t-1}+{\beta }_1\Delta EPI{S}_{i,t}+{\beta }_2\Delta GP{T}_{i,t}+{\beta }_3\Delta RD{\mathrm{INT}}_{i,t}\ast \Delta EPI{S}_{i,t}+{\beta }_4\Delta RnD{\mathrm{INT}}_{i,t}\ast \Delta GP{T}_{i,t}+\\ {\beta }_5\Delta LE{V}_{i,t}+{\beta }_6\Delta {\mathrm{Size}}_{i,t}+{\beta }_7\Delta {\mathrm{Age}}_{i,t}+{\beta }_8\Delta {\mathrm{CAP}\mathrm{INT}}_{i,t}+{\beta }_9\Delta \mathrm{Top}{1}_{i,t}+\Delta Y{\mathrm{ear}}_i+\Delta {ϵ}_{i.t} \left(i=1,\dots ,N,t=1,\dots ,T\right)\end{array}\end{array}$$

(10)

$$\begin{array}{c}\begin{array}{c}\Delta M{V}_{i,t}=\alpha \Delta M{V}_{i,t-1}+{\beta }_1\Delta EPI{S}_{i,t}+{\beta }_2\Delta GP{T}_{i,t}+{\beta }_3\Delta RnD IN{T}_{i,t}\ast \Delta EPI{S}_{i,t}+{\beta }_4\Delta RnD IN{T}_{i,t}\ast \Delta GP{T}_{i,t}+\\ {\beta }_5\Delta LE{V}_{i,t}+{\beta }_6\Delta Siz{e}_{i,t}+{\beta }_7\Delta Ag{e}_{i,t}+{\beta }_8\Delta CAP IN{T}_{i,t}+{\beta }_9\Delta \mathrm{Top}{1}_{i,t}+\Delta Yea{r}_i+\Delta {ϵ}_{i.t} \left(i=1,\dots ,N,t=1,\dots ,T\right)\end{array}\end{array}$$

(11)

$$\begin{array}{c}\begin{array}{c}\Delta P{E}_{i,t}=\alpha \Delta P{E}_{i,t-1}+{\beta }_1\Delta EPI{S}_{i,t}+{\beta }_2\Delta GP{T}_{i,t}+{\beta }_3\Delta RnD IN{T}_{i,t}\ast \Delta EPI{S}_{i,t}+{\beta }_4\Delta RnD IN{T}_{i,t}\ast \Delta GP{T}_{i,t}+\\ {\beta }_5\Delta LE{V}_{i,t}+{\beta }_6\Delta Siz{e}_{i,t}+{\beta }_7\Delta Ag{e}_{i,t}+{\beta }_8\Delta CAP IN{T}_{i,t}+{\beta }_9\Delta Top{1}_{i,t}+\Delta Yea{r}_i+\Delta {ϵ}_{i.t} \left(i=1,\dots ,N,t=1,\dots ,T\right)\end{array}\end{array}$$

(12)

where $\mathsf{\Delta }R\&D IN{T}_{i,t}\ast \mathsf{\Delta }EPI{S}_{i,t}$ and $\mathsf{\Delta }R\&D IN{T}_{i,t}\ast \mathsf{\Delta }GP{T}_{i,t}$ represent the first-order difference vectors of the product terms of absorptive capacity and proactive green process innovation and the product terms of absorptive capacity and proactive green product innovation, respectively.

In order to examine our hypotheses and enhance the reliability of the results, we needed to choose an appropriate estimation procedure for our theoretical framework and dynamic panel data analysis. Based on previous works [74], we used the difference-generalized method-of-moments (GMM) estimator for dynamic panel data to examine our predictions. Furthermore, we used Stata 15.0 statistical software. In addition, we used the Sargan test and Arellano–Bond autocorrelation (AR) (2) to determine whether the model was over-identified or mis-specified. Based on the AR serial autocorrelation test results, none of the disturbance terms had order two or higher autocorrelation, consistent with the premise of a GMM regression model. In addition, the Sargan over-identification test results accepted the original hypothesis of ‘all instrumental variables are valid,’ indicating that the model applicability test was passed. The regression results showed that the first-order lags of the dependent variables had a significant positive correlation with the current period, fully indicating that the dependent variables had ‘time stickiness,’ which can be solved by difference GMM.

4. Results

4.1. Descriptive Statistics

The descriptive statistics results for all of the variables are presented in

Table 2. Descriptive Statistics.

Variables	Obs	Mean	Std. Dev.	Min	Max
ROA	1491	0.51	0.061	−0.169	0.261
ROE	1491	0.077	0.108	−0.47	0.428
Tobinsq	1476	3.114	2.296	0.925	14.321
MV	1476	22.872	1.085	20.711	26.652
PE	1474	101.591	168.503	−97.377	1208.302
EPIS	1156	0.042	0.054	0	0.274
GPT	1449	1.887	1.591	0	6.675
R&D INT	1682	0.069	0.066	0.001	0.376
LEV	1491	0.367	0.201	0.033	0.861
Size	1491	21.905	1.241	19.727	26.483
Age	1474	2.73	0.397	1.386	3.466
CAP INT	1491	2.674	2.262	0.339	15.15
Top1	1474	0.325	0.151	0.088	0.73

.

In Table 2, the mean value of ROA and ROE is 0.051 and 0.077, respectively, indicating that the short-term corporate financial performance is low, as a higher ratio indicates better profitability.

The mean values of Tobin’s Q, MV, and P/E are, respectively, 3.114, 22.873, 101.591, indicating that the long-term financial performance of the sample firms is good and investors are bullish about their future market investment value. When these values are higher, it indicates better profitability. When the value of q is higher than 1 (q > 1), this means that market investors overestimate the firm’s value and firms pursuing profit maximization will choose to make additional investments. When the value of q is less than 1 (q < 1), it means that market investors underestimate the firm’s value, and it is rational not to invest or to reduce investment [75].

The mean value of EPIS is 0.042, showing that the average ratio of environmental protection investment to equity of the sample firms is only 4.2%, indicating low motivation for proactive green process innovation. The mean value of PGT is 1.887, showing that the average total number of green patents of the sample firms is 1.887, indicating a standard output in proactive green product innovation. The mean value of research and development intensity is 0.069, representing an average ratio of research and development expenses to net sales in the sample firms of 6.9%, indicating a low absorption capacity.

4.2. Correlation Coefficient Matrix

The correlation coefficient matrix of the study results is shown in

Table 3. Correlations.

	1	2	3	4	5	6	7	8	9	10	11	12	13
1. ROA	1
2. ROE	0.908 ***	1
3. PE	−0.315 ***	−0.307 ***	1
4. MV	0.034	0.117 ***	−0.055 **	1
5. Tobinsq	0.339 ***	0.223 ***	0.173 ***	0.086 ***	1
6. R&D INT	−0.035	−0.094 ***	0.181 ***	0.016	0.311 ***	1
7. EPIS	0.219 ***	0.122 ***	0.097 ***	−0.353 ***	0.381 ***	0.097 ***	1
8. GPT	−0.071 ***	0.015	−0.069 ***	0.615 ***	−0.206 ***	0.038	−0.341 ***	1
9. LEV	−0.369 ***	−0.164 ***	−0.024	0.347 ***	−0.347 ***	−0.294 ***	−0.285 ***	0.330 ***	1
10. Size	−0.161 ***	−0.026	−0.122 ***	0.871 ***	-0.371 ***	−0.144 ***	−0.508 ***	0.662 ***	0.503 ***	1
11. Age	−0.167 ***	−0.126 ***	0.037	0.214 ***	−0.076 ***	−0.102 ***	−0.030	0.067 **	0.231 ***	0.252 ***	1
12. CAP INT	−0.174 ***	−0.181 ***	0.215 ***	−0.014	0.050 *	0.404 ***	0.024	−0.113 ***	−0.179 ***	−0.024	0.014	1
13. Top1	0.120 ***	0.189 ***	−0.102 ***	0.078 ***	−0.076 ***	−0.199 ***	−0.037	−0.069 ***	0.117 ***	0.111 ***	−0.046 *	−0.038	1

Note: *** Significance at the 1% level; ** Significance at the 5% level; * Significance at the 10% level.

.

The correlations show that proactive green process innovation and proactive green product innovation are significantly correlated with short-term and long-term corporate financial performance. Moreover, these two forms of proactive green technology innovation correlate with absorptive capacity.

4.3. Variable Multicollinearity Test

In order to avoid co-linearity, a multicollinearity test is required. Multicollinearity is generally detected by calculating the variance inflation factor (VIF), which is the ratio of variance in the presence of multicollinearity between independent variables to variance in the absence. The empirical judgment method shows that the larger the VIF, the more severe the covariance. When 0 < VIF < 10, there is no multicollinearity. When 10 ≤ VIF < 100, there is strong multicollinearity. When VIF ≥ 100, there is severe multicollinearity. The results of the multicollinearity test of the model in this study are shown in

Table 4. The variance inflation factor.

Variables	VIF	1/VIF
GPT	2.105	0.475
EPIS	1.474	0.678
LEV	1.645	0.608
Size	3.511	0.285
Age	1.208	0.828
CAP INT	1.132	0.883
Top1	1.103	0.906
Mean VIF	1.684	0.594

.

The results in Table 4 show that the VIF of all variables is less than 10. Thus, the variables selected in this study do not have collinearity.

4.4. Results Analysis

4.4.1. The Results of Direct Effects

We report the direct effect of the difference GMM estimation in

Table 5. GMM Estimation Results.

Variables	Model 1	Model 2	Model 3	Model 4	Model 5
	ROA	ROE	MV	PE	Tobinsq
L.ROA	0.2533 ***
	(0.01)
L.ROE		0.1425 ***
		(0.01)
L.MV			0.4323 ***
			(0.01)
L.PE				0.1166 ***
				(0.01)
L.Tobinsq					0.2894 ***
					(0.01)
EPIS	0.0375 **	0.0289 **	95.9727 ***	0.6020 ***	5.1475 ***
	(0.02)	(0.01)	(25.83)	(0.08)	(0.72)
GPT	−0.0071 ***	−0.0140 ***	12.0647 ***	0.0204 ***	0.0095 **
	(0.00)	(0.00)	(2.30)	(0.00)	(0.01)
LEV	−0.0505 ***	−0.1058 ***	63.6872 ***	0.0080	−0.8940 ***
	(0.01)	(0.01)	(11.23)	(0.03)	(0.10)
Size	0.0139 ***	0.0354 ***	−79.4254 ***	0.5951 ***	−0.6154 ***
	(0.00)	(0.00)	(5.17)	(0.01)	(0.04)
Age	−0.1122 ***	−0.1900 ***	85.8699 **	−0.0190	−0.4957 **
	(0.02)	(0.01)	(37.32)	(0.10)	(0.25)
CAP INT	−0.0074 ***	−0.0144 ***	10.1609 ***	−0.0078 ***	−0.1356 ***
	(0.00)	(0.00)	(2.61)	(0.00)	(0.02)
Top1	−0.0593 ***	0.0007	−339.6041 ***	−0.0676	−0.8015 ***
	(0.02)	(0.01)	(42.36)	(0.09)	(0.27)
Constant	0.1150	−0.0857	1601.4076 ***	7.4040 ***	18.2475 ***
	(0.08)	(0.06)	(136.34)	(0.30)	(0.75)
AR (2)	0.227	0.115	0.403	0.849	0.702
Sargan	0.676	0.481	0.163	0.476	0.421
Observations	636	636	636	636	636
Number of stkcd	126	126	126	126	126
Wald test	2894	6.103 × 106	24,209	2.340 × 107	1.001 × 106

Notes: Standard errors in parentheses, *** p < 0.01, ** p < 0.05.

. Model 1 and Model 2 show the results of Equations (3) and (4), respectively, and verify hypotheses H1 and H3. Model 3, Model 4, and Model 5 show the results of Equations (5)–(7), respectively, and verify hypotheses H2 and H4.

The results in Table 5 show that firms’ proactive green process innovation positively relates to their short-term and long-term financial performance supporting H1 and H2, respectively. The regression coefficient of EPIS is significantly positive in model 1 (β₁ = 0.0375, p < 0.05) and in model 2 (β₁ = 0.0289, p < 0.05), implying that when a firm’s the scale of environmental investment increases by 1%, its ROA and ROE increase by 3.75% and 2.89%, respectively. This illustrates that a semiconductor firm’s proactive green process innovation is conducive to enhancing its short-term corporate financial performance. Meanwhile, the regression coefficient of EPIS is significantly positive in model 3 (β₁ = 95.9727, p < 0.01), model 4 (β₁ = 0.6020, p < 0.01), and model 5 (β₁ = 5.1475, p < 0.01), implying that when a firm’s the scale of environmental investment increases by 1%, its MV, P/E, and Tobin’s Q increase by 95.9727, 0.602, and 5.1475, respectively. This illustrates that a semiconductor firm’s proactive green process innovation is conducive to enhancing its long-term corporate financial performance.

In addition, the results show that proactive green product innovation is significantly and negatively related to short-term corporate financial performance; the regression coefficient of GPT is significantly negative in model 1 (β₂ = −0.0071, p < 0.01) and model 2 (β₂ = −0.0140, p < 0.01). Thus, H3 is not supported. When semiconductor firms engage in green product innovation, ROA and ROE, which measure their short-term financial performance, decrease by 0.71% and 1.4%, respectively. That indicates that firms do not profit in the short term when they adopt proactive green product innovation. However, the regression coefficient of GPT is significantly positive in model 3 (β₂ = 12.0647, p < 0.01), Model 4 (β₂ = 0.0204, p < 0.01), and model 5 (β₂ = 0.0095, p < 0.05), providing support for H4. This illustrates that a semiconductor firm’s proactive green product innovation is conducive to enhancing its long-term corporate financial performance.

4.4.2. The Results of the Moderating Effect

We report the moderating results of absorptive capacity on the relationship between proactive green technology innovation and corporate financial performance in

Table 6. Regression results for the moderating role of absorptive capacity.

Variables	Model 6	Model 7	Model 8	Model 9	Model 10
	ROA	ROE	MV	PE	Tobinsq
L.ROA	0.2412 ***
	(0.02)
L.ROE		0.1225 ***
		(0.01)
L.MV			0.4023 ***
			(0.01)
L.PE				0.1130 ***
				(0.01)
L.Tobinsq					0.2935 ***
					(0.01)
EPIS	−0.0128	−0.0090	−320.4243 ***	0.4053 ***	3.0506 ***
	(0.03)	(0.02)	(52.92)	(0.15)	(0.58)
GPT	−0.0082 ***	−0.0140 ***	13.5756 ***	−0.0218 ***	−0.0059
	(0.00)	(0.00)	(2.27)	(0.00)	(0.01)
R&D INT	−0.3177 ***	−0.5105 ***	277.5927 ***	−1.0538 ***	−2.9109 ***
	(0.04)	(0.04)	(38.39)	(0.08)	(0.34)
R&D INT × EPIS	0.7866 ***	0.6571 ***	329.4047 ***	2.5569 ***	29.4641 ***
	(0.25)	(0.15)	(452.40)	(0.86)	(2.22)
R&D INT × GPT	−0.0008 ***	−0.0020 ***	3.5161 ***	0.0060 ***	0.0255 ***
	(0.00)	(0.00)	(0.33)	(0.00)	(0.00)
LEV	−0.0624 ***	−0.1244 ***	58.9443 ***	0.0099	−0.8344 ***
	(0.01)	(0.01)	(11.83)	(0.02)	(0.16)
Size	0.0151 ***	0.0306 ***	−84.1169 ***	0.6004 ***	−0.5443 ***
	(0.00)	(0.00)	(6.03)	(0.01)	(0.03)
Age	−0.1487 ***	−0.2180 ***	137.3074 ***	−0.1784 *	−1.1106 ***
	(0.02)	(0.02)	(38.67)	(0.10)	(0.28)
CAP INT	−0.0058 ***	−0.0131 ***	15.8373 ***	−0.0216 ***	−0.2325 ***
	(0.00)	(0.00)	(2.19)	(0.00)	(0.01)
Top1	−0.0571 **	−0.0215	−278.7061 ***	−0.2380 ***	−1.3120 ***
	(0.03)	(0.01)	(34.68)	(0.08)	(0.32)
Constant	0.2252 **	0.1579 **	1502.7126 ***	8.0004 ***	19.0619 ***
	(0.11)	(0.07)	(147.67)	(0.27)	(0.91)
AR(2)	0.202	0.113	0.357	0.620	0.648
Sargan	0.360	0.459	0.071	0.483	0.452
Observations	632	632	632	632	632
Number of stkcd	126	126	126	126	126
Wald test	1661	1.180 × 107	20,607	1.750 × 107	1.550 × 107

Note: Standard errors in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1.

. Models 6 and 7 show the results of Equations (8) and (9), respectively, and verify hypotheses H5a and H5c. Models 8, 9, and 10 show the results of Equations (10)–(12), respectively, and verify hypotheses H5b and H5d.

The results show that absorptive capacity enhances the relationship between proactive green process innovation and its short-term and long-term financial performance effects, supporting H5a and H5b. The regression coefficients of the product terms of absorptive capacity (R&D intensity) and proactive green process innovation (Environmental Protection Investment Scale) show a significantly positive moderating effect in models 6 (β₁ =0.7866, p < 0.01) and 7 (β₁ = 0.6571, p < 0.01). When a firm’s absorptive capacity increases by 1%, moderating its ROA and ROE, they increase by 78.66% and 65.71%, respectively. This indicates that the absorptive capacity of semiconductor firms has a positive moderating effect on the relationship between proactive green process innovation and its short-term financial performance effect. In addition, the regression coefficients of the product terms of absorptive capacity and proactive green process innovation have a significant positive moderating effect in models 8 (β₁ = 329.4047, p < 0.01), 9 (β₁ = 2.5569, p < 0.01), and 10 (β₁ = 29.4641, p < 0.01), implying that when a firm’s absorptive capacity increases by 1%, moderating its MV, P/E and Tobin’s Q, they increase by 329.4047, 2.5569, and 29.4641, respectively. This proves that the absorptive capacity of semiconductor firms positively moderates the relationship between their proactive green process innovation and long-term financial performance [67].

Furthermore, the results show that absorptive capacity has a significant effect in the relationship between proactive green product innovation and corporate financial performance. However, its moderating effect on short-term corporate financial performance is negative effect, while it has a significantly positive role with respect to long-term corporate financial performance. These results provide support for H5d and do not support H5c. The regression coefficients of the product terms of absorptive capacity (R&D intensity) and proactive green product innovation (GPT) have significantly negative moderating effects in models 6 (β₂ = −0.0008, p < 0.01) and 7 (β₂ = −0.0020, p < 0.01). This implies that when a firm’s absorptive capacity improves by 1%, moderating its ROA and ROE, they decrease by 0.08% and 0.2%, respectively. This shows that the moderating effect of a semiconductor firm’s absorptive capacity in the process of proactive green product innovation does not lead to good financial performance in the short term. Moreover, we observe that the negative value of the correlation coefficient of the product term of absorptive capacity (R&D intensity) and proactive green product innovation (GPT) is small. This negative moderating effect does not massively affect semiconductor firms’ short-term financial performance. It can even be considered negligible. However, the regression coefficient of the product term of absorptive capacity and proactive green product innovation has a significantly positive moderating effect in models 8 (β₂ = 3.5161, p < 0.01), 9 (β₂ = 0.0060, p < 0.01), 10 (β₂ = 0.0255, p < 0.01), implying that when a firm’s absorptive capacity improves by 1%, moderating its MV, P/E and Tobin’s Q, they are increased by 3.5161, 0.006, and 0.0255, respectively. This proves that in the long run, this positive moderating effect can bring about growth in a firm’s future financial performance [69].

4.5. Robustness Tests

To avoid the chance of regression results, we applied fixed effects estimation to the static panel model for robustness testing and constructed the following static panel model:

$$\begin{array}{c}{\mathrm{S}}_{\mathrm{term}}{\mathrm{CFP}}_{i,t}=\lambda +{\gamma }_1{\mathrm{EPIS}}_{i,t}+{\gamma }_2{\mathrm{GPT}}_{i,t}+{\gamma }_3{\mathrm{LEV}}_{i,t}+{\gamma }_4{\mathrm{Size}}_{i,t}+{\gamma }_5{\mathrm{Age}}_{i,t}+{\gamma }_6{\mathrm{CAPINT}}_{i,t}+{\gamma }_7\mathrm{Top}{1}_{i,t}+{\gamma }_{8{{\displaystyle \sum }}^{}\mathrm{Year}}+{\epsilon }_{i,t}\end{array}$$

(13)

$$\begin{array}{c}{\mathrm{L}}_{\mathrm{term}}{\mathrm{CFP}}_{i,t}=\lambda +{\gamma }_1{\mathrm{EPIS}}_{i,t}+{\gamma }_2{\mathrm{GPT}}_{i,t}+{\gamma }_3{\mathrm{LEV}}_{i,t}+{\gamma }_4{\mathrm{Size}}_{i,t}+{\gamma }_5{\mathrm{Age}}_{i,t}+{\gamma }_6{\mathrm{CAP}\mathrm{INT}}_{i,t}+{\gamma }_7\mathrm{Top}{1}_{i,t}+{\gamma }_{8{{\displaystyle \sum }}^{}\mathrm{Year}}+{\epsilon }_{i,t}\end{array}$$

(14)

where t denotes year, i denotes fixed-effect individual, the dependent variable S_term CFP includes two measures (ROA, ROE) and the L_term CFP includes three measures (Tobin’s Q, MV, PE), ${\mathrm{EPIS}}_{i,t}$ and ${\mathrm{GPT}}_{i,t}$ are the proxy variables for proactive green process innovation and proactive green product innovation by individual i in year t, respectively, LEV (financial leverage), Size (firm size), Firm age, CAP INT (Capital intensity), and Top1 (Top1 Shareholder’s shareholding) are control variables, Year is a dummy variable for year, and ${ϵ}_{i.t}$ is a disturbance term. The estimation results of the static panel model are shown in

Table 7. Robustness test results.

Variables	Model 11	Model 12	Model 13	Model 14	Model 15
	ROA	ROE	MV	PE	Tobinsq
EPIS	0.0561 **	0.0896 **	214.6661 **	1.1564 ***	6.7981 ***
	(0.04)	(0.05)	(132.39)	(0.28)	(1.69)
GPT	−0.0041 ***	−0.0066 ***	10.8707 ***	0.0112 ***	0.0049 **
	(0.00)	(0.00)	(4.08)	(0.00)	(0.02)
LEV	−0.0997 ***	−0.1414 ***	−78.5460	−0.1721	−1.3929 **
	(0.01)	(0.03)	(53.63)	(0.16)	(0.64)
Size	0.0160 **	0.0347 **	−38.1769 *	0.7412 ***	−0.4015 ***
	(0.01)	(0.01)	(21.98)	(0.06)	(0.15)
Age	−0.0916 ***	−0.1245 *	89.9165 **	−0.1716	−0.9264
	(0.03)	(0.06)	(43.69)	(0.20)	(0.77)
CAP INT	−0.0099 ***	−0.0176 ***	10.2071 *	−0.0528 ***	−0.2309 ***
	(0.00)	(0.00)	(5.61)	(0.02)	(0.07)
Top1	0.0703 **	0.1600 ***	−114.1910 *	0.0869	0.1057
	(0.03)	(0.05)	(67.62)	(0.17)	(0.58)
Constant	0.0153	−0.2550	681.2390	7.5065 ***	15.9924 ***
	(0.19)	(0.32)	(433.21)	(1.57)	(5.06)
Observations	829	829	829	829	829
R-squared	0.204	0.132	0.0761	0.760	0.427
Number of groups	128	128	128	128	128
Industry	YES	YES	YES	YES	YES
year	YES	YES	YES	YES	YES
F	11,993	55,482	3917	12,642	2599

Note: Standard errors in parentheses, *** p < 0.01, ** p < 0.05, * p < 0.1.

.

As seen in Table 7, the static estimation results are consistent with the previous dynamic estimation results. This finding further confirms the significant effect of proactive green technology innovation on short- and long-term corporate financial performance, and indicates that the results of our tests are robust.

5. Discussion and Conclusions

In recent years, whether it in fact pays to be “green” has been a core debate in both business and academic circles. Our findings shed light on this debate: proactive green technology innovation improves a firm’s financial performance. We analyzed dynamic panel data from 126 Chinese listed semiconductor concept stocks in this study using the difference-GMM approach. We tried to shed light on the dynamic development process and explore the relationships between proactive green process and product innovation and both short- and long-term corporate financial performance. Furthermore, we examined the role of absorptive capacity in moderating these relationships. Our main findings and conclusions are discussed below.

First, our empirical results show that a firm’s proactive green process innovation is conducive to enhancing its short- and long-term corporate financial performance. These findings are consistent with the previous literature examining the relationship between green innovation and firm performance [29,39,76]. This may be because, although green process innovation requires a higher Environmental Protection Investment Scale (EPIS), it can improve existing production processes or add new ones to reduce harmful emissions. A firm which achieves competitive differentiation and has an image of being a good corporate citizen in society is more likely to be favored by consumers. This in turn helps firms obtain a better return on their investment, thus proving that the pursuit of proactive green process innovation can provide both environmental benefits and superior financial performance. Therefore, firms should adopt proactive green process innovation in order to improve their value.

Second, our empirical results further show that while a firm’s proactive green product innovation is not necessarily conducive to enhancing its short-term corporate financial performance, it is conducive to enhancing its long-term performance. In other words, while implementing proactive green product innovation is slightly detrimental to corporate financial performance in the short term, it will benefit performance in the long term. This offers a fresh perspective on the long-standing debate on the link between green technology innovation and corporate financial performance, and is consistent with the findings of Lu et al. (2018) [34] and contrary to those of Amores-Salvadó et al. (2014) [77], who hold that environmental product innovation does not have a statistically significant effect on firm performance. We find that it has a statistically significant effect, and a negative effect on short-term corporate financial performance. This is probably because proactive green product innovation requires a long process to generate the corresponding financial benefits. Therefore, we believe that proactive green product innovation is worthwhile. Investors should focus on long-term profitability and improve their financial performance by implementing proactive green product innovation.

Third, we find that absorptive capacity significantly moderates the relationship between proactive green technology innovation and corporate financial performance. On the one hand, absorptive capacity positively moderates the relationship between a firm’s proactive green process innovation and its short- and long-term corporate financial performance. This is consistent with the findings of Xie et al. (2015) [59]. They argue that absorptive capacity has a positive moderating effect on the relationship between clean technology in green process innovation and financial performance. Thus, for firms that proactively implement green process innovation, their more robust absorptive capacity can lead to good profits in the short and long term. On the other hand, absorptive capacity negatively moderates the relationship between proactive green product innovation and short-term corporate financial performance. However, surprisingly, it positively moderates the relationship between a firm’s proactive green product innovation and its long-term corporate financial performance. This may be due to absorptive capacity making it difficult to advance green product innovation in the short term [59]. Green product innovation requires firms to have high knowledge and technology transfer capabilities. Usually, however, this usually involves a continuous and long-term process before firms are able to effectively and successfully transform technology and research and development activities into green or new products which meet the growing consumer demand for environmentally friendly products. In doing this, firms become better positioned to profit from proactive green product innovation [69]. This suggests that even though absorptive capacity fails to positively moderate the relationship between proactive green product innovation and corporate financial performance in the short term, it can be a positive moderator in the long term and help firms to develop green products better. Therefore, we encourage firms to pay attention to improving their absorptive capacity in proactive green product innovation in order to bring about significant benefits in their future.

5.1. Managerial Implications

Several managerial implications can be drawn from our research. First, firms should make more excellent investments in proactive green process innovation, as it is often more advantageous than product innovation for short-term corporate financial performance. As a matter of fact, many stakeholders are more interested in timely profit returns [78]. Second, firms should take a long view about investing in these technologies, and proactive green product innovation should be considered as well, as it can deliver better future market value to firms in the long term. Seeking to develop environmentally friendly products and services that meet the growing consumer awareness of environmental protection is critical to a firm’s success in achieving sustainability. Finally, firms should use their internal absorptive capacity to improve their financial performance in proactive green technology innovation. Because absorptive capacity plays a positive moderating role in the relationship between proactive green technology innovation and corporate financial performance, it is an essential resource in enhancing the competitiveness of firms and helps them to better realize the sustainable value in proactive green technology innovation. Therefore, firms should strive to communicate the value of green technology to their employees and raise their awareness of green technology. In particular, for employees in research and development departments it is necessary to continuously and effectively learn about green technology in order to promote better implementation of their firm’s proactive green process and product innovation goals, as they will eventually find that it is worth paying to be green.

5.2. Limitations and Future Research

Our research contains several limitations that need to be acknowledged. Nonetheless, these limitations may provide additional opportunities for future studies. First, certain potential factors may change the relationship between proactive green technology innovation and corporate financial performance, including shareholder structure heterogeneity and investor risk dimension. We could not fully consider these factors in our current work, and future studies should consider introducing these variables. Second, our research data was obtained in the context of listed semiconductor companies in China. Different countries may conduct different proactive green technology innovation practices, and these exercises may lead to various outcomes. Future studies should be conducted in other emerging economies to determine the relationship between proactive green technology innovation and corporate financial performance. Third, it is possible to reverse the causality direction in our research model. For example, corporate financial performance was used as the independent variable in this study. Therefore, future studies could explore this relationship in the reverse direction. Finally, depending on its origin, green technology innovation can be divided into proactive green technology innovation and reactive green technology innovation, and the influence of proactive green technology innovation may differ from that of reactive green technology innovation. However, our current study focuses only on proactive innovation. Future studies could introduce reactive green technology innovation for further comparative exploration.