Are Higher-Educated Employees More Responsible? A Study about Employee Quality and Corporate Environmental Sustainability

Abstract

This paper explores how employee quality affects companies’ actions in terms of environmental protection and sustainable development. We use the proportions of the workforce with three education levels as proxies for employee quality and examine their respective impacts on three dimensions of corporate environmental sustainability: (1) reporting, (2) quality and standardization, and (3) attention and initiative. We investigate over 1300 firms listed in Chinese stock markets from 2006, the year when Chinese firms first released environmental issues and the Chinese regulatory divisions first announced a guideline, to 2021. The results indicate the following: (a) employee quality is positively associated with the reporting of corporate environmental sustainability, (b) companies with more educated employees are more likely to provide environmental and sustainable information with high standards and quality, and (c) companies with more educated employees are more likely to release environmental and sustainable information in solo reports as a reflection of the corporate attention on and initiative in environmental sustainability. Moreover, we find that employees at higher education levels have a stronger impact on corporate environmental sustainability. This paper contributes to the literature on human capital and environmental management.

1. Introduction

Concerns about environmental and sustainable development have become mainstream over many years (Haddock-Fraser and Tourelle, 2010) [1]. Early in 1987, the World Commission on Environment and Development (WCED) pointed out the concept of “environmentally sound and sustainable development” (ESSD) and defined it as “development that meets the needs of the present without compromising the ability of future generations to meet their own needs” (WCED, 1987; Harger and Meyer, 1996; Moldan et al., 2012) [2,3,4]. The definition was extended by the United Nations (UN, 1992) [5] at the Earth Summit in 1992, and the formalization was eventually completed by the World Summit on Sustainable Development in 2002 (UN, 2002) [6] with one notion, “environmental sustainability”, to highlight the human impact on natural systems (Goodland, 1995; Morelli, 2011) [7,8]. Today, the term itself and its tenor have become so widespread and well known that they have become common sense to every party in society including individuals, governments, and corporations. From the corporate perspective, environmental sustainability is one crucial requirement to meet, and large corporations bear much responsibility for key environmental issues. Today, companies spend a great deal of effort on corporate social responsibility (CSR) to fulfill requirements and conditions related to social sustainability, economic sustainability, and environmental sustainability. Thus, aspects of CSR become channels and measures to observe and evaluate corporate performance in terms of environmental sustainability.

During recent years, employee quality has gained great attention as an important factor affecting aspects of corporate actions and performance. Recent studies on corporate finance have usually used the compositions of employees’ education levels as measures and discussed their impacts. For example, Chen and Yoon (2021) [9] found that employee quality positively affected firm values during 2010–19. Sun et al. (2020) [10] found that during 2012–2016, Chinese listed firms with a high-quality workforce exhibited higher CSR scores as per a credit rating organization. Similar results also appeared in studies on Chinese listed firms through 2011–18 (Wang and Liao, 2021) [11] and through 2011–2017 (Wang and Yan, 2022) [12]. Si and Xia (2022) [13] studied a 2012–2017 sample of Chinese firms and found that employee quality improves the corporate internal information environment and reduces the stock price crash risk. Jin et al. (2023) [14] studied corporate risk taking among China’s A-share-listed firms during 2011–2017 and showed that higher levels of employee education may result in a higher level of corporate risk taking. The results of these studies suggest that employee education has a strong impact on aspects of corporate actions.

The literature above has shown that employees can influence their companies in many aspects and that employees with different levels of education may have different impacts. Inspired by those existing studies, this paper explores whether employee education has a significant impact on corporate environmental sustainability.

We evaluate the workforce, using different education levels as proxies for employee quality, and construct research hypotheses to examine the impact of employee quality at different levels on corporate actions related to environmental and sustainable development. We investigate environmental issues involving 1300 firms listed in Chinese stock markets during the period between 2006 and 2021. Our findings show that employees’ educations positively affect the reporting of corporate environmental and sustainable development. Moreover, a company with more educated employees has a higher likelihood of complying with a sustainable development framework and thus of providing information of high standards and quality. Furthermore, improvements in employee education quality are positively associated with corporate attention on and initiative in environmental and sustainable development and may influence the company to provide information in a solo report. In addition, we observe that employees with higher education levels have a stronger impact in contrast with employees with relatively lower education levels.

This paper makes highlights and contributions. First, our paper conducts a thorough examination to see how employee quality affects different dimensions of corporate environmental sustainability in China. Most previous studies usually used a third-party CSR rating score as the only measure. By contrast, in our paper, we have a greater concentration on issues of environmental sustainability instead of a general CSR rating performance. We conduct three hypothesis tests on three dimensions of corporate environmental sustainability, including (1) reporting, (2) quality and standardization, and (3) attention and initiative. Through these test results, we can have an in-depth perception on employee quality and its influences on specific aspects.

Second, our paper provides a more accurate classification of employee education quality. Previous studies used diverse proxies and methods to measure employee education quality. Some studies set up a solo measure to represent the overall employee quality (Call et al., 2017; Sun et al., 2020; Dong et al., 2022; Li, Liu and Wu, 2022) [10,15,16,17]. Some studies focused on the proportions of employees with bachelor’s degrees or above to measure human capital quality (Wang and Liao, 2021; Si and Xia, 2022; Jin et al., 2023) [11,13,14]. Some studies created two measures to describe employees with undergraduate and graduate education degrees (Li and Lu, 2020; Xue et al., 2020; Chen and Yoon, 2021; Wang and Yan, 2022) [9,12,18,19]. In our paper, we create three measures to categorize employees into three education levels: the graduate level, undergraduate level, and technical-school level. Using these measures, we can conduct a more thorough study on employee quality levels from different education groups and compare their respective impacts on corporate actions.

Third, our paper provides a complete tracking of corporate actions related to environmental sustainability in China. According to the historical records (Shenzhen Stock Exchange, 2006; Mu, 2020) [20,21], the earliest report about environmental sustainability from Chinese firms was released in 2006 and the first governmental guideline about CSR was issued in 2006. This paper tracks corporate actions related to environmental sustainability from 2006 to 2021, enabling a thorough observation of corporate actions in China.

The remainder of this paper is organized as follows. Section 2 presents the literature review and hypothesis development. Section 3 presents the methodology, including the settings of variables and the statistical description of data. Section 4 presents the test results. Section 5 concludes the paper.

2. Literature Review and Hypothesis Development

Many existing studies have pointed out that employees and employee quality are highly influential in various firm behaviors. A series of recent studies have taken employee education as the measure of employee quality and empirically shown that employee education has large impacts on firms’ productivity, value growth, and innovation. Chen and Huang (2009) [22] found that the knowledge management capacity along with education relates positively to firms’ innovation performance. Chowdhury et al. (2014) [23] found that the experience and education of firm employees represent key resources to improve firm productivity. Chen and Yoon (2021) [9] studied employee education quality in the classes of undergraduate and graduate degrees and found that employees with higher education experiences enhance firm value at a larger magnitude than those with lower education experiences. Kong et al. (2022) [24] confirmed the contribution of education on innovation and productivity growth from corporate standpoints.

As part of sustainable development practices, the innovation process and value performance of one company depend on its knowledge base. Employees with higher education represent the valuable intelligent input to innovation. Therefore, we posit that employee education could have effects on corporate dedication when we observe a number of companies from a perspective of sustainable development.

Some studies have investigated employee education and its influence on CSR at the firm level. Sobczak et al. (2006) [25] conducted a questionnaire among student interns and young managers to test whether higher education may affect the perception of CSR among employees. Their results showed that both the content of education and the culture of academic institutions have impacts on the perception of CSR. Rosati et al. (2018) [26] observed attitudes towards CSR among 153 employees with different levels of education in the Italian banking system. They found that employees with graduate education are more concerned with and trustful of a company’s CSR commitment than non-graduate employees. Sun et al. (2020) [10] found that the educational levels of employees are positively related to employers’ CSR rating grades from a rating institution. Li and Li (2024) [27] found that employee educations promote the quality of CSR reporting.

However, studies about the impact of education on CSR are not equivalent to those on the impact on environmental protection and sustainable development. One important reason is that CSR includes many aspects and components (Sun et al., 2020; Cai and He, 2022) [10,28]. In practice, CSR includes social, economic, and environmental aspects of business activities in collective actions (O’Riordan and Fairbrass, 2008; Babiak and Trendafilova, 2011; Tang et al., 2018) [29,30,31]. Many studies have usually used CSR rating scores to assess the influence of employee education quality. CSR rating scores are mixtures of diverse aspects of business activities and thus cannot accurately reflect the corporation’s output in terms of environmental sustainability. Moreover, when a company releases a CSR report (which is usually used as the primary reference for making a CSR rating score), it does not guarantee that environmental sustainability content must be included. In fact, when we check the CSR reports from Chinese firms, we find that about 59% of the CSR reports contain environmental sustainability content while 41% of them have no relevant information to release. Furthermore, even the bough of environmental sustainability has multiple branches. For instance, environmental sustainability projects under CSR consist of six main components: waste reduction; non-polluting production; production energy efficiency; safe and healthy work environments; environmentally sound products; and environmentally sound packaging (Li, Khalili and Cheng, 2019) [32]. A change in employee education quality ought to influence those specific CSR components to different extents. Therefore, it is critical to examine the specific impact of employee education on corporate activities related to environmental sustainability.

In practice, we still take the released CSR reports as the main channel to track corporate activities related to environmental sustainability. Firms are inclined to participate in CSR activities and disclose them in various forms, which has gradually become a trend during the last 15 years (Li, Khalili and Cheng, 2019; Li, He, Dbouk and Zhao, 2021; Li, 2023) [32,33,34].

In 2006, the Shenzhen Stock Exchange (SZSE) released a guideline about CSR report preparation and more listed companies initiated CSR disclosure. In 2008, the Shanghai Stock Exchange (SSE) released an official announcement about mandatory reporting to some listed companies in specific sectors. In 2011, the China Securities Regulatory Commission (CSRC) had an announcement to encourage listed companies in China to release CSR reports as independent issues instead of CSR chapters in companies’ annual reports. This was the first time that the CSRC gave official opinions on the CSR disclosure and report design (CSRC, 2011) [35]. In June 2021, the CSRC updated the guidelines on annual report formation and encouraged companies to disclose their CSR activities in more aspects, including environmental protections (CSRC, 2021) [36]. Thus, CSR reporting has become a main channel for companies to release information in environmental sustainability.

Hence, we propose the following:

Hypothesis 1.

Employee education quality has a positive impact on the reporting of corporate environmental and sustainable development.

However, the quality of CSR reporting differs across companies. Take Chinese firms as examples. According to the annual rating report of Environment, Social and Governance (ESG) by Rankins (RKS, 2020) [37], a major flaw of the CSR reports from Chinese firms is that most of these CSR reports do not release information in normative standards, and consequently, lower the quality of information. Quan et al. (2015) [38] also pointed out that the releasing of CSR reports from Chinese firms is still at the primitive stage. Some firms simply “clone” their CSR reports from previous years and intentionally skip negative information disclosure in reports. Therefore, information disclosure is not a sufficient condition to confirm that one company does well in environmental sustainability. It is necessary to consider and check the quality and standardization of the disclosure.

During the last two decades, new global governance institutions have emerged and guided companies to take into account their social and environmental impacts (Bartley 2007; Gilbert et al., 2011) [39,40]. A set of global standards have been established in order to help companies implement, manage, and report their activities related to social and environmental sustainability (Waddock, 2008) [41].

One example, the Global Reporting Initiative (GRI), is a multi-stakeholder initiative to encourage companies’ information disclosure on economic, social, governance-related, and environmental performance (Waddock 2007; GRI 2011a) [42,43]. The GRI provides information on the scope and quality of reporting and develops reporting norms on the content on and forms in which to report (Hussey et al., 2001; del Mar Alonso-Almeida et al., 2014; Vigneau et al., 2015) [44,45,46]. The GRI focuses on standardizing the reporting practices of environmental protection and sustainability and provides a guideline to companies on what to report (indicators on social, environmental, and economic performance) and how to report (protocol of reporting) (Brown et al., 2009a, 2009b; GRI, 2011b) [47,48,49]. The GRI also provides a grading system on companies’ disclosure levels to give companies scores of their reporting quality (GRI, 2011c) [50].

Although thousands of multi-national corporations (MNCs) recognize the contribution and dominant power of the GRI, its CSR reporting guideline is still a voluntary standard (Willis, 2003) [51]. So, it is each company’s own choice whether to comply with this international sustainability standard. In reality, many companies volunteer to join the GRI and release their sustainability reports as communication tools for their socially responsible behavior and thus gain credibility and reputation (Hedberg and Von Malmborg, 2003; Brown et al., 2009a; Fortanier et al., 2011; Fernandez-Feijoo et al., 2014) [47,52,53,54]. Today, the GRI has become a worldwide reporting standard among MNCs across the world (Waddock, 2007; Marimon et al., 2012; Vigneau et al., 2015; Halkos and Nomikos, 2021) [43,46,55,56]. It is a brand or icon to assess and justify the quality and standardization of a company’s practices related to environmental and sustainable development. Therefore, in this paper, we use the GRI as an indicator to examine the reporting quality and standardization of environmental and sustainable development.

Since 1998, China has experienced several rounds of expansion in higher education, and more people have had admission to undergraduate and postgraduate education. Meanwhile, Chinese companies have started to introduce the GRI and paid attention to the quality of environmental and sustainable development. Accordingly, we propose the following:

Hypothesis 2.

Employee education quality has a positive impact on the reporting quality and standardization of corporate environmental and sustainable development.

Currently in China, most listed firms are encouraged by the CSRC to voluntarily release CSR reports or relevant information. Similar policies are also applied in other countries, and CSR disclosure could have diverse determinants (Gamerschlag et al., 2011; Kansal, et al., 2014; Ali et al., 2017; Fahad and Nidheesh, 2020; Malik et al., 2020) [57,58,59,60,61]. Among these determinants, employee education quality is considered an important driver to affect the corporate attention on and initiative in corporate environmental and sustainable development. Some studies have confirmed that employee education can affect employees’ attitudes and cognitions related to CSR issues (Luthar et al., 1997; Elias, 2004) [62,63]. According to Quazi (2003); Li, Tang, and Chau (2019); and Sun et al. (2020) [10,64,65], education equality may affect the awareness of and attention paid to CSR and consequently affect a company to be more willing to fulfill CSR requirements. So, in this paper, we plan to examine whether a similar impact of employee quality exists in the context of environmental sustainability.

Hypothesis 3.

Employee education quality has a positive impact on the corporate attention on and initiative in environmental and sustainable development.

Previous studies have used diverse proxies and methods to measure employee education quality. Call et al. (2017), Sun et al. (2020), and Dong et al. (2022) [10,15,16] created a measure for evaluating the educational backgrounds of employees using a weighted average method. Rosati et al. (2018) [26] examined educational differences by classifying employees into two groups: graduate employees and non-graduate employees. Garrido-Ruso and Aibar-Guzmán (2022) [66] also had similar methods in their meta-analysis. Li et al. (2020) [67] constructed two firm-level measures: the average number of years of education of a firm’s employees and the total proportion of employees with a bachelor’s degree (or higher). By contrast, it has been more popular to create two measures to describe the proportions of employees with undergraduate and graduate education degrees, as seen in the recent studies by Li and Lu (2020), Xue et al. (2020), Chen and Yoon (2021), and Wang and Yan (2022) [9,12,18,19]. In our paper, we follow the widely used classification to distinguish employees with different education levels and compare their respective impacts. We have one additional hypothesis:

Hypothesis 4.

Employees with higher education levels have a larger impact on aspects of corporate environmental and sustainable development than employees with relatively lower education levels.

The above-mentioned aspects of corporate environmental and sustainable development refer to the three research targets in Hypotheses 1, 2, and 3: (1) disclosure, (2) quality and standardization, and (3) attention on and initiative in environmental and sustainable development.

3. Methodology

3.1. Dependent Variables

We have the following three dependent variables (${DV}_{i,t}$).

First, we use a dummy variable, ${Environment}_{i,t}$, to indicate whether content about environmental and sustainable development is reported by company i in Year t. The variable equals 1 when such content is reported and 0 otherwise. This variable corresponds to the research target of Hypothesis 1.

Second, we use another dummy variable, ${GRI}_{i,t}$, to indicate whether the reporting of environmental sustainability is in line with the framework of the GRI reporting guidelines for company i in Year t. The GRI provides a framework of issues about environmental sustainability that should be disclosed in corporate reports. We use ${GRI}_{i,t}$ to indicate the quality and standardization of the corporate report disclosure. The variable equals 1 when the content is disclosed following the GRI guidelines and 0 otherwise. This variable corresponds to the research target of Hypothesis 2.

Third, we use another dummy variable, ${Source}_{i,t}$, to indicate whether content on environmental sustainability is published as an independent report by company i in Year t. The variable equals 1 when the content about environmental and sustainable development is disclosed in an independent corporate report and 0 when the content is merged as a chapter in the annual report. The variable ${Source}_{i,t}$ corresponds to the research target of Hypothesis 3. In China, disclosing the information is voluntary for most firms, and they are free to choose the form of disclosure. Thus, when a firm releases an independent environmental sustainability report instead of a combined one, it suggests that the firm has paid more attention to its environmental and sustainable development and is more willing to fulfill it. So, we use ${Source}_{i,t}$ to assess the company’s attention to environmental sustainability.

3.2. Independent Variables

A number of previous studies constructed two education-level measures to capture employee quality. They categorized the employees by two education levels: the undergraduate level and the graduate level (Li and Lu, 2020; Xue et al., 2020; Chen and Yoon, 2021; Wang and Yan, 2022) [9,12,18,19]. In China, a large number of employees do not have undergraduate education degrees. According to the Ministry of Education report (2021) [68], the university student enrollment (including both undergraduates and graduates) comprised 44.3 million students while the overall student enrollment was 291 million. Besides employees with higher education, we also need to pay attention to the employees whose education levels are below undergraduate level.

Following Li and Li (2024) [27], we create three education-based independent variables to precisely describe the human capital intensity in each company. For company i in Year t, we calculate the percentage of employees whose highest degree is a postgraduate degree (master and Ph.D.) as $postgraduat{e}_{i,t}$, the percentage of employees whose highest degree is a bachelor’s degree as $bachelo{r}_{i,t}$, and the percentage of employees whose highest degree is from a two- or three-year college or technical school as $associat{e}_{i,t}$. These three measures present the education composition of employees. Higher values on these measures mean a high quality of employee composition. These three measures will enable us to examine whether employees with different education levels may influence the company to different extents, which corresponds to Hypothesis 4.

3.3. Control Variables

To control for firm characteristics, we use control variables to proxy for corporate financial performance and market trade ($Control{s}_{i,t}$). We use return on equity (${ROE}_{i,t}$), leverage ratio (${LEV}_{i,t}$), market-to-book ratio (${MB}_{i,t}$), market value (${TRADE}_{i,t}$), average return (${RETURN}_{i,t}$), and total employee number (${Employee}_{i,t}$), following a series of previous studies (Lins, Servaes and Tamayo, 2017; Chen, Hung and Wang, 2018; He et al., 2019; Li et al., 2020; Wang and Yan, 2022) [12,67,69,70,71]. Definitions of variables are provided in

Table 1. Variable definition.

Variables	Definitions
Dependent variables
${Environment}_{i,t}$	A dummy to indicate whether content about environmental and sustainable development is reported.
${GRI}_{i,t}$	A dummy to indicate whether the reporting of environmental sustainability is in line with the GRI guidelines.
${Source}_{i,t}$	A dummy to indicate whether content on environmental sustainability is published as an independent report.
Independent variables
$postgraduat{e}_{i,t}$	proportion of employees with postgraduate degrees.
$bachelo{r}_{i,t}$	proportion of bachelor’s degree employees.
$associat{e}_{i,t}$	proportion of employees who have graduated from a two- or three-year college or technical school.
Control variables
${ROE}_{i,t}$	income before extraordinary items divided by lagged total equity.
${LEV}_{i,t}$	total long-term debts divided by total assets.
${MB}_{i,t}$	the market value of equity divided by the book value of equity.
${TRADE}_{i,t}$	the market value of traded equity.
${RETURN}_{i,t}$	the mean of firm-specific weekly returns over the fiscal year period.
${Employee}_{i,t}$	the total number of employees over the fiscal year period.

.

3.4. Data and Sample

We use firm-yearly data about Chinese listed firms from the China Stock Market & Accounting Research (CSMAR) Database. The database is well known as the most comprehensive research-oriented database focusing on China’s finance and economy. The data were retrieved from the corporate governance division.

The observation period comprises the years between 2006 (the first year when Chinese companies released CSR) and 2021. We keep firms with more than 100 employees and exclude observations due to missing information for dependent, independent, and control variables. To mitigate the effect of outliners, we winsorize all continuous variables at the 1% and 99% levels. The final sample contains 18,318 firm-yearly observations from 1357 Chinese listed firms. The firms included cover all the industries according to the Industry Classification Guidelines (CSRC, 2012) [72].

Table 2. Summary statistics.

Variable	N	MEAN	Standard Deviation	MIN	MAX
Dependent
${Environment}_{i,t}$	18,318	0.60	0.49	0	1
${GRI}_{i,t}$	18,318	0.06	0.24	0	1
${Source}_{i,t}$	18,318	0.30	0.46	0	1
Independent	18,318
$postgraduat{e}_{i,t}$	18,318	0.06	0.09	0	0.53
$bachelo{r}_{i,t}$	18,318	0.20	0.18	0	0.77
$associat{e}_{i,t}$	18,318	0.63	0.27	0	0.98
Control	18,318
${ROE}_{i,t}$	18,318	0.06	0.16	−1.17	0.45
${LEV}_{i,t}$	18,318	2.06	2.65	0	19.25
${MB}_{i,t}$	18,318	0.61	0.26	0.10	1.20
${TRADE}_{i,t}$	18,318	3.32 × 1010	4.64 × 1010	9.28 × 108	3.10 × 1011
${RETURN}_{i,t}$	18,318	0.31	0.81	−0.73	3.63
${Employee}_{i,t}$	18,318	6539	11,463	138	76,126

presents the descriptive statistics. The average of variable ${Environment}_{i,t}$ is 0.60. Among the overall samples, the values of about 60% of them (11,076 observations) are equal to one and indicate that these firms participate in the social activities related to environmental protection and sustainable development. This result also indicates that about 60% of the CSR reports from Chinese firms contain environmental sustainability content while 40% of them have no relevant information to release.

By contrast, the average values of ${GRI}_{i,t}$ and ${Source}_{i,t}$ are smaller. ${GRI}_{i,t}$ has an average of 0.06, indicating that a small group of cases comply with the framework of the GRI guidelines. ${Source}_{i,t}$ has an average of 0.30, indicating that only a small group of companies prefer to publish their environmental protection practice data in an independent report. These results are in line with the current status of CSR disclosure in Chinese companies as previously discussed in the hypothesis development section. It is necessary and valuable to study how to encourage companies to join a standard like GRI to improve the quality of corporate environmental and sustainable development and enhance the corporate attention and initiative in this area.

The average values of the independent variables ($postgraduat{e}_{i,t}$, $bachelo{r}_{i,t}$, and $associat{e}_{i,t}$) are 0.06, 0.20 and 0.63, respectively. These results imply that in Chinese companies, the majority of employees have education experiences below the undergraduate level. This is reasonable when we see that the average of the total employee number (${Employee}_{i,t}$) is 6539. Most of firms in the sample have a large number of employees, and some firms may have more than 70,000 employees. The employees with undergraduate and graduate degrees count for a relatively smaller proportion of the overall employees. These results are in line with the statistics in the Ministry of Education report.

Table 3. Correlations.

	${Environment}_{i,t}$	${GRI}_{i,t}$	${Source}_{i,t}$	${ROE}_{i,t}$	${LEV}_{i,t}$	${MB}_{i,t}$	${TRADE}_{i,t}$	${RETURN}_{i,t}$	${Employee}_{i,t}$	$postgraduat{e}_{i,t}$	$bachelo{r}_{i,t}$
${GRI}_{i,t}$	0.20
${Source}_{i,t}$	0.49	0.37
${ROE}_{i,t}$	0.04	0.05	0.09
${LEV}_{i,t}$	−0.02	−0.02	−0.04	−0.23
${MB}_{i,t}$	0.19	0.16	0.20	−0.05	0.15
${TRADE}_{i,t}$	0.22	0.25	0.28	0.13	−0.06	−0.09
${RETURN}_{i,t}$	−0.25	−0.06	−0.16	0.03	−0.01	−0.23	0.09
${Employee}_{i,t}$	0.20	0.41	0.31	0.10	−0.01	0.27	0.39	−0.04
$postgraduat{e}_{i,t}$	0.06	0.06	0.13	0.02	−0.03	0.01	0.08	−0.03	−0.05
$bachelo{r}_{i,t}$	0.05	0.08	0.10	0.04	−0.06	0.04	0.12	−0.05	0.00	−0.10
$associat{e}_{i,t}$	0.05	−0.06	−0.06	−0.03	0.04	0.01	−0.07	−0.01	0.01	−0.22	−0.52

presents the correlation matrix of the variables. The employee education quality variables ($postgraduat{e}_{i,t}$, $bachelo{r}_{i,t}$, and $associat{e}_{i,t}$) are positively associated with ${Environment}_{i,t}$, providing preliminary support for our fundamental assumption. Among the three variables, $postgraduat{e}_{i,t}$ and $bachelo{r}_{i,t}$ show positive relationships to ${GRI}_{i,t}$ and ${Source}_{i,t}$, indicating that the employees with higher education may have a higher inclination to be positively associated with the environmental and sustainable development of the company. These correlation coefficients are statistically significant.

4. Empirical Results

We create a logit model in (1) to test how employee education influences aspects of environmental and sustainable development. On the left side, ${\mathrm{D}\mathrm{V}}_{i,t}$ represents ${Environment}_{i,t}$, ${GRI}_{i,t}$, and ${Source}_{i,t}$, one for each regression. The main target independent variables are $postgraduat{e}_{i,t}$, $bachelo{r}_{i,t}$, and $associat{e}_{i,t}$, and the control variable group ($Control{s}_{i,t}$) contains return on equity (${ROE}_{i,t}$), leverage ratio (${LEV}_{i,t}$), market-to-book ratio (${MB}_{i,t}$), market value (${TRADE}_{i,t}$), average return (${RETURN}_{i,t}$), and total employee number (${Employee}_{i,t}$). In order to examine multicollinearity, we use a variance inflation factor (VIF) test (see Appendix A).

$${\mathrm{D}\mathrm{V}}_{i,t}={\beta }_1\mathit{p}\mathit{o}\mathit{s}\mathit{t}\mathit{g}\mathit{r}\mathit{a}\mathit{d}\mathit{u}\mathit{a}\mathit{t}{\mathit{e}}_{\mathit{i},\mathit{t}}+{\beta }_2\mathit{b}\mathit{a}\mathit{c}\mathit{h}\mathit{e}\mathit{l}\mathit{o}{\mathit{r}}_{\mathit{i},\mathit{t}}+{\beta }_3\mathit{a}\mathit{s}\mathit{s}\mathit{o}\mathit{c}\mathit{i}\mathit{a}\mathit{t}{\mathit{e}}_{\mathit{i},\mathit{t}}+\sum \gamma \times Control{s}_{i,t}+\epsilon$$

(1)

The three dependent variables are dichotomous. They can measure the likelihood of occurrence of the corresponding events. For example, the dependent variable ${Environment}_{i,t}$ can measure the probability of corporate reporting on environmental sustainability relative to the probability of non-reporting. Following Agresti (2007) and Sharma (1996) [73,74], we use the logit regression model to examine how the likelihoods of corporate reporting on environmental sustainability are driven by those independent and control variables.

We first test Hypothesis 1. Using the logit regression model (1) and the dichotomous dependent variable ${Environment}_{i,t}$, we set up the probability of corporate reporting on environmental sustainability relative to the probability of non-reporting, which is referred to as the odds.

Table 4. Parameter estimates of logit regression with ${\mathit{E}\mathit{n}\mathit{v}\mathit{i}\mathit{r}\mathit{o}\mathit{n}\mathit{m}\mathit{e}\mathit{n}\mathit{t}}_{\mathit{i},\mathit{t}}$.

	Odds Ratio
$\mathrm{p}\mathrm{o}\mathrm{s}\mathrm{t}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{u}\mathrm{a}\mathrm{t}\mathrm{e}$	6.35 *** (0.00)
$\mathrm{b}\mathrm{a}\mathrm{c}\mathrm{h}\mathrm{e}\mathrm{l}\mathrm{o}\mathrm{r}$	2.57 *** (0.00)
$\mathrm{a}\mathrm{s}\mathrm{s}\mathrm{o}\mathrm{c}\mathrm{i}\mathrm{a}\mathrm{t}\mathrm{e}$	2.55 *** (0.00)
$\mathrm{R}\mathrm{O}\mathrm{E}$	1.06 (0.564)
$\mathrm{L}\mathrm{E}\mathrm{V}$	0.98 *** (0.00)
$\mathrm{M}\mathrm{B}$	3.86 *** (0.00)
$\mathrm{T}\mathrm{R}\mathrm{A}\mathrm{D}\mathrm{E}$	1.00 *** (0.00)
$\mathrm{R}\mathrm{E}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{N}$	0.49 *** (0.00)
$\mathrm{E}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{e}$	1.00 *** (0.00)
Observations	18,318
R-squared value	0.1504

The p-values are reported in parenthesis. ***, **, and * indicate p < 1%, 5%, and 10%, respectively.

presents the results of the logit regression model with ${Environment}_{i,t}$. The three independent variables $postgraduat{e}_{i,t}$, $bachelo{r}_{i,t}$, and $associat{e}_{i,t}$ are statistically significant at 1%. The odds-ratio coefficient of $postgraduat{e}_{i,t}$ is 6.35, indicating that a one-unit increase in the proportion of graduate-level employees will enhance the reporting/non-reporting odds ratio by over six times. Consequently, the probability of reporting may rise from 1/2 to 86%, that is, showing a change of 36%, as a reaction to the increase in the number of graduate-level employees in one company. Similar influences also appear from the other two groups of employees. $bachelo{r}_{i,t}$ and $associat{e}_{i,t}$ have odds-ratio coefficients of 2.57 and 2.55, respectively, indicating that the increase in the number of employees with bachelor’s degrees or technical school educations will increase the reporting/non-reporting odds ratio by 2.5 times and thus increase the probability of reporting by 21%. After controlling the firm characteristics, all three independent variables are positively associated to the odds ratio of ${Environment}_{i,t}$, so the results support the argument of Hypothesis 1. The improvement in employee education quality at any education level has a positive impact on the reporting of corporate environmental and sustainable development.

Table 4 also presents another finding when we look at the three independent variables. The odds-ratio coefficient of $postgraduat{e}_{i,t}$ (6.35) is higher than the coefficients of $bachelo{r}_{i,t}$, and $associat{e}_{i,t}$ (2.57 and 2.55, respectively). These results suggest that employees with graduate education experience have a stronger impact on corporate reporting on environmental sustainability compared with those employees with undergraduate- or lower-level educations. This finding provides supportive evidence for Hypothesis 4.

Second, we test Hypothesis 2. Using the dichotomous dependent variable ${\mathit{G}\mathit{R}\mathit{I}}_{\mathit{i},\mathit{t}}$, we set up the odds between the compliance and non-compliance of the GRI framework so as to examine the reporting quality and standardization of corporate environmental and sustainable development.

Table 5. Parameter estimates of logit regression for ${\mathit{G}\mathit{R}\mathit{I}}_{\mathit{i},\mathit{t}}$.

	Odds Ratio
$postgraduate$	18.99 *** (0.00)
$bachelor$	5.90 *** (0.00)
$associate$	1.31 * (0.10)
$ROE$	2.04 ** (0.032)
$LEV$	0.99 (0.313)
$MB$	6.04 *** (0.00)
$TRADE$	1.00 *** (0.00)
$RETURN$	0.68 *** (0.00)
$Employee$	1.00 *** (0.00)
Observations	18,318
R-squared value	0.2452

The p-values are reported in parenthesis. ***, **, and * indicate p < 1%, 5%, and 10%, respectively.

presents the results of the logit regression model on ${\mathit{G}\mathit{R}\mathit{I}}_{\mathit{i},\mathit{t}}$. Similar to the test of Hypothesis 1, the three independent variables have odds-ratio coefficients above 1. The coefficients of $\mathit{p}\mathit{o}\mathit{s}\mathit{t}\mathit{g}\mathit{r}\mathit{a}\mathit{d}\mathit{u}\mathit{a}\mathit{t}{\mathit{e}}_{\mathit{i},\mathit{t}}$ and $\mathit{b}\mathit{a}\mathit{c}\mathit{h}\mathit{e}\mathit{l}\mathit{o}{\mathit{r}}_{\mathit{i},\mathit{t}}$ (18.99 and 5.90, respectively) are statistically significant at 1%, and the coefficient of $\mathit{a}\mathit{s}\mathit{s}\mathit{o}\mathit{c}\mathit{i}\mathit{a}\mathit{t}{\mathit{e}}_{\mathit{i},\mathit{t}}$ (1.31) is statistically significant at 10%. These results suggest that an increase in the number of highly educated employees can increase the corporate inclination to adopt the GRI standards and thus support the argument of Hypothesis 2. An improvement in employee education quality at any education level has a positive impact on the quality and standardization of corporate environmental and sustainable development.

Table 5 also provides evidence to support Hypothesis 4. The odds-ratio coefficient of $postgraduat{e}_{i,t}$ (18.99) is higher than the coefficients of $bachelo{r}_{i,t}$, and $associat{e}_{i,t}$ (5.90 and 1.31, respectively). Similar to those in Table 4, these results suggest that employees with graduate education experience have a stronger power to affect the company to comply with an environmental and sustainable standard.

Third, we test Hypothesis 3. Using the dichotomous dependent variable ${Source}_{i,t}$, we set up the odds between the independent and combined report release on environmental sustainability and examine the corporate attention on and initiative in environmental and sustainable development.

Table 6. Parameter estimates of logit regression for ${\mathit{S}\mathit{o}\mathit{u}\mathit{r}\mathit{c}\mathit{e}}_{\mathit{i},\mathit{t}}$.

	Odds Ratio
$\mathrm{p}\mathrm{o}\mathrm{s}\mathrm{t}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{d}\mathrm{u}\mathrm{a}\mathrm{t}\mathrm{e}$	30.59 *** (0.00)
$\mathrm{b}\mathrm{a}\mathrm{c}\mathrm{h}\mathrm{e}\mathrm{l}\mathrm{o}\mathrm{r}$	3.77 *** (0.00)
$\mathrm{a}\mathrm{s}\mathrm{s}\mathrm{o}\mathrm{c}\mathrm{i}\mathrm{a}\mathrm{t}\mathrm{e}$	1.38 *** (0.00)
$\mathrm{R}\mathrm{O}\mathrm{E}$	3.21 *** (0.00)
$\mathrm{L}\mathrm{E}\mathrm{V}$	0.97 *** (0.00)
$\mathrm{M}\mathrm{B}$	4.36 *** (0.00)
$\mathrm{T}\mathrm{R}\mathrm{A}\mathrm{D}\mathrm{E}$	1.00 *** (0.00)
$\mathrm{R}\mathrm{E}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{N}$	0.57 *** (0.00)
$\mathrm{E}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{o}\mathrm{y}\mathrm{e}\mathrm{e}$	1.00 *** (0.00)
Observations	18,318
R-squared value	0.1705

The p-values are reported in parenthesis. ***, **, and * indicate p < 1%, 5%, and 10%, respectively.

presents the results of the logit regression model on ${Source}_{i,t}$. All three independent variables ($postgraduat{e}_{i,t}$, $bachelo{r}_{i,t}$, and $associat{e}_{i,t}$) have odds-ratio coefficients above 1 (30.59, 3.77 and 1.38, respectively) with statistical significance at 1%. These results support Hypothesis 3. An increase in the number of highly educated employees can increase the corporate inclination to release information on environmental sustainability in an independent report, which reflects the corporate attention on and initiative in environmental and sustainable development. Moreover, the results in Table 6 provide evidence to support Hypothesis 4 as well. The odds-ratio coefficient of $postgraduat{e}_{i,t}$ (30.59) is higher than the coefficients of $bachelo{r}_{i,t}$ and $associat{e}_{i,t}$ (3.77 and 1.38, respectively). The differences between these coefficients across education levels further suggest that employees with higher educations have a stronger power to affect the corporate attention on and initiative in environmental and sustainable development.

In summary, the empirical analyses through the logit regression model provide supportive evidence for the hypotheses. An improvement in employee quality, which is measured through the increase in the number of highly educated employees, has a strong and multi-dimensional positive impact on the construction of corporate environmental and sustainable development. The composition of employees with higher-level educations may impel the company to adopt and disclose policies and activities on environmental and sustainable development, comply with an environmental and sustainable standard, and pay more attention and take more initiative.

5. Discussions and Conclusions

The need for environmental and sustainable development has become a worldwide consensus among individuals, governments, and, especially, corporations. From the corporate perspective, companies have responsibilities to fulfill requirements and conditions about environmental sustainability in their business practices. Many companies have a periodical disclosure of their dedications towards environmental sustainability in the form of CSR reports. But these outputs are diverse across companies and industries.

According to previous studies, many factors may affect the outputs and performance of a company in terms of environmental sustainability. During recent years, employee quality has gained great attention and has been tested as an important factor on different aspects of corporate actions and performance. In this paper, we have explored how employee quality affects companies’ actions in areas of environmental protection and sustainable development. Compared with those studies that have examined the impact of employee quality on a general CSR rating score, our analyses have brought a greater concentration on corporate environmental sustainability.

In the analyses, we have examined three dimensions of corporate environmental sustainability: (1) reporting, (2) quality and standardization, and (3) attention and initiative. We have evaluated the workforce, using different education levels as proxies for employee quality, and categorized employees into three education levels: the graduate level, undergraduate level, and technical-school level. We have investigated 1357 firms listed in Chinese stock markets from 2006, the year when Chinese firms first released environmental issues and the Chinese regulatory divisions first announced a guideline, to 2021.

Our results from the logit regression models confirm the positive impact of employee quality on corporate environmental and sustainable development. First, an improvement in employee education is positively associated with the reporting of corporate environmental sustainability. Second, a company with more educated employees may have a larger probability to accept the GRI standard framework and consequently provide environmental and sustainable information with high standards and quality. Third, a company with more educated employees is more likely to release environmental and sustainable information in a solo report rather than a chapter in financial reports, implying that the high employee education quality is positively associated to the corporate attention on and initiative in environmental and sustainable development. Fourth, through the three analyses above, we have observed that employees with higher educations (graduate-level) have a stronger impact on corporate environmental sustainability in contrast with employees with relatively lower educations (undergraduate- and technical-level).

This paper contributes to the existing literature. First, it provides new evidence to support those studies regarding employee equality and CSR and further presents an in-depth perception on employee quality and its influences on specific aspects of environmental sustainability. Second, this paper provides a more detailed analysis of employee quality through the classification of three education levels and has created three measures correspondingly. Using these measures, we can observe the distinctions of employee quality across education levels and compare their respective impacts on corporate actions. Third, on the basis of relevant studies on CSR in China, this paper provides new findings on corporate actions related to environmental sustainability in China. We have tracked the corporate actions of environmental sustainability in China back to their origin time and presented a thorough developing picture on corporate environmental sustainability in China.

This paper also suggests practical implications. First, from the corporate standpoint, a firm may benefit from investing in human capital in diverse aspects. It is widely known that the enhancement of employee education can improve the capacity, efficiency, and innovation of firms. Now, corporate environmental sustainability is also related to employee education. Second, the composition of human capital is highly associated with aspects of corporate environmental sustainability. It is valuable for firms to hire employees with higher educations. Highly educated employees may have a larger probability to engage in the environmental and sustainability-related issues of their employers. Third, the contribution of education has been confirmed. During the last two decades, China has launched several rounds of expansion in higher education. The number of enrollments in college and graduate studies has risen significantly. The boom in education has facilitated firms in any industry to develop in a sustainable way.