1 Introduction
Corporate social responsibility (CSR), a self-regulating business model, is widely acknowledged in the existing literature as a company’s commitment to mitigating the potential adverse effects of its activities on its stakeholders and maximizing its long-term beneficial influence on the general public and society (Carroll 1979; McWilliams and Siegel 2001; Cheng et al. 2014; Ferrell et al. 2016; Liang and Renneboog 2017b). Specifically, CSR, also known as corporate citizenship, is a corporate action taken on behalf of stakeholders (McWilliams and Siegel 2000). The interest in and scope of practicing CSR have expanded from a minority of academic researchers to a vast majority of business communities, investors, senior executives, and policymakers (Shahab et al. 2018a; Ma et al. 2020).1 The effective use of CSR can enhance firm value, referred to as ‘doing well by doing good’ (Brown and Dacin 1997; Deng et al. 2013; Boubakri et al. 2016; Nguyen et al. 2020). CSR activities, such as contributions to employee benefits, compliance with international environmental protection guidelines and social welfare practices, contributions to workplace and product safety, and human rights protection, motivate a company’s customers, employees, investors, suppliers, and communities to continuously support its business dealings and operations, thus enhancing firm performance (Liang and Renneboog 2017a; Li et al. 2021b). Nevertheless, we still have limited knowledge on the determinants of a company’s commitment to socially responsible activities, particularly from the top management team (TMT) perspective (Li and Zhang 2010; Siegel 2014; Orlitzky et al. 2015).
Prior research has, to a large extent, focused on the influence of CEOs on CSR strategies (Slater and Dixon-Fowler 2009; Tang et al. 2015; McCarthy et al. 2017; Al-Shammari et al. 2019; Hegde and Mishra 2019; Shahab et al. 2020). Until quite recently, the influence of the TMT on CSR has been little understood (Siegel 2014), especially in emerging markets. This is crucial negligence, as all companies’ strategic decisions are made by the TMT, not CEOs alone. Along with the increased importance of and pressure from various stakeholders, CSR is increasingly matrixed into companies’ business strategies, and it even constitutes a source of competitive advantage (Siegel and Vitaliano 2007). However, the dominated neoclassical economics and agency theories are not able to explain the CSR strategies as they are not obviously profit-maximizing. Indeed, the upper echelon theory (UET) postulates that the demographic and functional backgrounds of senior executives play a critical role in strategic decision making (Hambrick and Mason 1984). In regard to the implementation of CSR strategies, prior studies indicate that the demographic and functional backgrounds of senior executives can encourage and cultivate a stronger executive commitment to compliance with institutional regulations and guidelines, which exert a positive influence on social and environmental conduct (Ntim and Soobaroyen 2013; Shahab et al. 2018b, 2020; Al-Shammari et al. 2019; Hegde and Mishra 2019; Chen et al. 2020a). In other words, senior executives’ specific characteristics or experiences may serve as crucial antecedents of the extent of CSR performance and reporting activities.
Anzeige
In this study, we draw on the UET and respond to the recent call for investigating specific TMT demographic backgrounds as drivers of CSR (Siegel 2014; Orlitzky et al. 2015) by exploring whether and how senior executives’ financial working experience (‘financial expertise’) affects CSR. The financial expertise acquired by the TMT becomes increasingly important and is cast as best equipped to make strategies (Zorn 2004; Uhde et al. 2017). In particular, the increased financial constraint inflicted by the recession of 2008–2009 makes senior executives with financial acumen more desirable because of the importance of financial skills when appraising business strategies and corporate policies and managing limited capital resources and business risk (Custódio and Metzger 2014). Prior studies have demonstrated that boards with financial experts affect corporate policies and strategies (Defond et al. 2005; Burak Güner et al. 2008; Abernathy et al. 2013; Badolato et al. 2014; Farber et al. 2018). We extend this strand of research by investigating whether financial experts with financial institutions working experience affect CSR strategies. Traditionally, financial experts (i.e., accountants) are viewed as profit-driven, whereas evidence finds that CEOs with financial expertise engage more in CSR activities (Shahab et al. 2020). This suggests that some backgrounds of financial experts are not solely profit-driven. Therefore, we focus on senior executives’ working experience in financial institutions, not only because of the financial skills obtained from such experience, but also the special cognitions of responsible investments cultivated while working in financial institutions matters in CSR practices.
The Chinese market provides a unique and ideal environment to study companies’ CSR strategies and decisions regarding the existence of senior executives with financial experience for the following reasons. China is the largest emerging market and has weak corporate sustainable performance and minority shareholder protections, severe environmental problems, and inferior law enforcement (Li et al. 2021b). The average CSR ratings of Chinese listed companies are, in general, lower than those of companies located in countries with common or civil law regimes (Demirbag et al. 2017; Liang and Renneboog 2017b). China has significantly improved its economic development and successfully transformed its industrial structure within a relatively short period since its capital markets were open to international investors in the 1990s; however, this remarkable ‘industrial upgrade’ has, to a certain extent, driven up societal and environmental costs such as excessive sulphur dioxide emissions, industrial wastewater treatment, and poor food quality (Elmagrhi et al. 2019). A series of actions have been conducted by the Chinese government and regulatory authorities to address environmental problems, and sustainability regulations have been enacted at all levels in response to the pressure from various stakeholders to encourage companies to promote CSR.2 Moreover, in response to drastic climate change and deteriorating environmental conditions, Chinese environmental organizations highlighted the slogan ‘greener GDP’ in the mid-2000s (Shahab et al. 2018b, 2020). In September 2016, the Chinese authority signed the Paris Climate Agreement to demonstrate its commitment to changing the worsening environment situation in the country in particular and the globe in general. However, despite all these efforts, social and environmentally sustainable problems are still prevalent because of the poor law enforcement in China (Li et al. 2021b). Compared with those in developed countries, where CSR standards have been successfully implemented, Chinese executives are still a long way from fully practicing socially sustainable activities in alignment with international practices and standards; this alarming situation and China’s institutional background motivate us to timely examine the influence of senior executives’ attributes on CSR activities and sustainable reporting.
Using a sample consisting of 847 unique Chinese listed companies and 5,158 firm-year observations over the period 2009–2018, we find that the TMT’s financial experience has a positive influence on CSR and that the influence has its economic significance. On average, the CSR rating is approximately 4.05% higher for companies with financial experts as senior executives than for companies without financial experts as senior executives, ceteris paribus. In addition, companies run by senior executives with financial expertise tend to issue longer sustainability reports. We further find that the positive influence of the TMT’s financial experience on CSR is particularly driven by senior executives who have work experience in regulatory-oriented financial institutions. By examining the role of the TMT’s latitude of action in the link between senior executives’ financial experience and CSR performance, we reveal that this positive influence is more salient in non-state-owned enterprises (non-SOEs) or in cash-rich firms. Our results are also robust to a battery of additional tests, including a firm fixed-effect model, alternative measures of the TMT’s financial experience, an industry-mean-adjusted social responsibility rating, the inclusion of additional control variables, the weighted least squares (WLS), the propensity score matching (PSM) technique, the Heckman correction approach, and the dynamic generalized method of moments (GMM) model. Last, our further analysis illustrates that the enhancement in companies’ CSR driven by financial experts on the TMT leads to higher firm value.
We contribute to the extant literature in the following ways. First, we update the growing understanding of the effect of financial experience on companies’ strategies by providing novel evidence of its effect on CSR, a crucial aspect that has been omitted in related literature. Extant studies on the effects of TMT members with financial experience mainly focus on conservative financial disclosure (Bamber et al. 2010), internal control weaknesses (Oradi et al. 2020), general financial policies (Custódio and Metzger 2014),3 earnings management (Jiang et al. 2013; Dimitrios and Hang 2018), investment efficiency and firm performance (Li et al. 2021a), tax avoidance (Chen et al. 2020b), and the interplay between the Sarbanes–Oxley Act and the likelihood of appointing a financial expert CEO (Cullinan and Roush 2011). We demonstrate that financial experts in the TMT not only focus on short-term profit via financial performance, innovation, and M&As but also take the long-term stakeholders’ interests into consideration. Our finding contributes to upper echelons theory (Hambrick and Mason 1984; Hambrick 2007) and early-career formative experiences (Gibbons and Waldman 2004, 2006).
Anzeige
Second, our research contributes to the literature on the driving factors of CSR. To the best of our knowledge, this is the first attempt to investigate whether TMTs’ financial experience affects corporate outcomes with a specific focus on socially responsible activities. The extant literature indicates that a company’s CSR is influenced by female board directors (McGuinness et al. 2017; Elmagrhi et al. 2019), analyst coverage (Adhikari 2016), institutional ownership (Dyck et al. 2019; Li et al. 2021b), family ownership (Block and Wagner 2014; Oh et al. 2019), cross-listing (Del Bosco and Misani 2016; Lu and Wang 2021), legal origins and national institutional characteristics (Demirbag et al. 2017), and CEO management skills (Chen et al. 2020a). Our study highlights that the financial backgrounds of senior executives play a part in improving CSR performance and facilitating nonfinancial disclosures, such as the readability and reporting contents of sustainability reports.
The rest of the study is organized as follows. Section 2 discusses the institutional background of China. Section 3 reviews the related literature and develops our hypotheses. Section 4 describes the research design. The discussion of key findings and robustness checks is covered in Sect. 5. Endogeneity tests are presented in Sect. 6. Section 7 presents further analysis, and Sect 8 concludes the study.
2 CSR in China
The concept of CSR was formally introduced into Chinese capital markets between the 1990s and early 2000s following extensive pressure from domestic and foreign stakeholders for more information on employee contributions, product standards and quality, environmental conditions and social problems.4 However, Chinese firms saw CSR not as a core value but as a task to complete, and as a consequence, a wide range of labour, health, and sustainability problems still continued to break out. Since then, Chinese firms have faced greater pressure to engage in stakeholder-oriented activities, and a series of localized sustainability guidelines and standards have been enacted to reflect the view that companies’ socially responsible activities can help build a harmonious community and society, which is a key goal that was highlighted by the Chinese regulatory authorities at the 2006 National People’s Congress. Later, both the Shanghai Stock Exchange (SSE) and the Shenzhen Stock Exchange (SZSE) enacted guidelines for the social responsibility and environmental information disclosure of listed companies in September 2006 and December 2008, respectively, to further strengthen CSR consciousness and construct a harmonious society.5 Since the turning point of 2006, when the concept of CSR was included in Chinese Corporate Law, many firms have begun to publish corporate sustainability reports. This was the time when non-governmental organizations and local academic institutions started researching the topic and educating those around them on best practices. Driven by the intensive intervention of the central government and regulatory authorities, the total annual amount of Chinese listed companies’ donations in kind experienced an increase of nearly 42%, from $3.24 billion in 2015 to $4.6 billion in 2018, as noted by the RKS Runling CSR rating system, which is a leading and independent sustainability rating agency in the Chinese market.
The execution of these CSR policies, however, still remained substandard, mainly because of the weak commitment of companies’ management teams and inferior corporate governance (Du et al. 2014; Shahab et al. 2020). Moreover, our data show that between 2009 and 2018, only approximately 15.12% of Chinese listed firms issued their CSR reports in accordance with the Global Reporting Initiative (GRI) sustainability and ‘green’ standards. Prevailing evidence suggests that in the absence of the total commitment of the TMT, the drive to bring positive changes to China’s sustainable and social conduct is unlikely (Shahab et al. 2018b, 2020). In sum, the present environmental and social concerns in China and its unique institutional background motivate us to investigate the effects of the TMT’s attributes on CSR.
3 Theoretical framework and hypotheses
3.1 TMT’s influence on CSR
Although CSR is not a traditional profit-driven investment, it can still benefit companies financially. For instance, firms that invest in CSR strategies to satisfy both internal and external stakeholders can gain enhanced reputation, greater support from customers, suppliers, communities, and governments (Wang and Qian 2011), and deepened employee loyalty and commitment (Greening and Turban 2000). All of these developments ultimately reflect on the increase in financial performance. Furthermore, a remarkable CSR record can help companies secure critical resources from stakeholders and mitigate the possible risks of reputational loss, thereby providing insurance-like protection in the long run (Fombrun and Gardberg 2000; Williams and Barrett 2000; Godfrey 2005). Despite these benefits, companies still vary in terms of CSR activities. In fact, TMTs, as the main driver of company strategies, face a significant increase in pressure from various stakeholders, including shareholders and debtholders (Wiengarten et al. 2017), which has a great influence on companies’ CSR strategies (Defond et al. 2005; Pondeville et al. 2013).
Nevertheless, the dominated neoclassical economics and agency theories are not able to explain the CSR strategies as they are not obviously profit-maximizing (Huang and Watson 2015; Plöckinger et al. 2016). Whether and how a company balances the conflicting needs of multiple stakeholders and achieves the abovementioned financial benefits is attributed to how the TMT understands and identifies the needs of stakeholders and incorporates them into firm strategies (Brower and Mahajan 2013; Parker 2014). We, therefore, introduce the UET, which postulates that individual characteristics play a significant role in corporate-level decision making (Hambrick and Mason 1984).
UET posits that a company’s strategic choices can reflect the values and cognitions of its TMT (Hambrick and Mason 1984; Hambrick 2007). Based on bounded rationality, this theory posits that senior executives will percept and interpret facets, details, and external pressures through a set of cognitive bases and values which affect the way they perceive the information and make decisions (Hambrick and Mason 1984; Parker 2014; Heyden et al. 2017). These cognitions and values reflect individual characteristics and idiosyncrasies of decision-making and create an individual managerial perception that evaluatesall situations and makes corporate strategic decisions. Therefore, UET provides theoretical roots to explain how TMT may react to the requests of stakeholders and make CSR decisions (Plöckinger et al. 2016; Reimer et al. 2018). Based on this theory, various observable experiences (i.e., age, gender, working experience, functional and educational backgrounds) are more often used as proxies for actual managerial cognitions in empirical investigations of selective perception (Carpenter et al. 2004). Working experience in financial institutions (or financial working experience) is one of such proxies.
Prior studies offer some evidence on how the demographic backgrounds of TMT members affect CSR activities. For example, Brower and Mahajan (2013) find that TMTs with a chief marketing officer will be more sensitive to external stakeholders and thus achieve better CSR performance. Wiengarten et al. (2017) contend that the gender and functional background of newly appointed CSR chief officers positively affect the financial performance of CSR strategies. Lau et al. (2016) also find that TMTs with more foreigners or managers having international experience achieve increased CSR performance. As noted by Reimer et al. (2018), a complementary relationship between the TMT and CEO can explain the heterogeneity of CSR strategies. Dezső and Ross (2012) and Christensen et al. (2014) find that having female executives in the TMT improves CSR performance. However, the influence of the financial working experience of TMTs, an increasingly important demographic background, on companies’ CSR has received little attention in the literature.
Different from others, corporate managers who build up their knowledge, skills, and social responsibility values through financial working experience may have conflicting effects in terms of CSR strategies. One would expect that financial experts associated with cost control do not have motivations to support nonfinancial activities, such as CSR activities. By contrast, we conjecture that senior executives with financial career experience may better understand the dependence on stakeholders, which may motivate them to advise more CSR engagement. Shahab et al. (2020) provide initial support to our conjunction by finding that financial expert CEOs positively improve a company’s sustainable performance and environmental management.
3.2 Financial experience of TMT members and CSR
Unlike other financial experts who have only worked in non-financial institutions, people who have ever worked in financial institutions are more likely to establish CSR cognition and values. This is because financial institutions, particularly the regulatory-oriented financial institutions, may be more sensitive to the trend of CSR/sustainability and subject to strict CSR concerns (Guler et al. 2002; Yin and Zhang 2012). The CSR cognition that has been built in organizations’ culture/policies enables their employees to be aware of socially responsible and environmentally sustainable engagement and commitment. Their employees are more likely to receive CSR-related budgeting and reporting training (Baumann-Pauly et al. 2013; Edinger-Schons et al. 2019).
Financial experience is related to a general responsibility of creating budgets and analyzing investments. Managers who lack good budgeting and investment appraisal skills may be less likely to invest in CSR projects because CSR activities are usually costly and uncertain. The relevant future cash inflows are difficult to predict. However, there is a natural link between CSR and financial professionals because of the high demand for measuring, analyzing, and reporting CSR-related information (Huang and Watson 2015). First, along with the popularity of integrated reporting and CSR reporting in the last few decades, financial professionals have a great opportunity to participate in the creation and analysis of CSR-related information. In this way, senior executives with financial experience should have a better understanding of the expenses and incomes of CSR projects and, hence, may provide useful advice on their forward-looking valuations. This may increase the likelihood of investing in CSR projects.
Second, senior executives with financial experience tend to be more proficient in risk management (Bamber et al. 2010). Research indicates that financial experts play predominant roles in risk assessment and management (Hall et al. 2015). Regarding CSR, the costs related to environmental and social risks can be substantial. Senior executives with financial expertise are qualified to advise risk management and avoid such costs by improving CSR (Helfaya and Moussa 2017). Furthermore, the abovementioned benefits of CSR activities come from strengthened relationships with stakeholders and enhanced reputation. These benefits can moderate the impact of adverse events on shareholder value and offset firm risk (Godfrey 2005; El Ghoul et al. 2018). As such, senior executives with financial experience may be more likely to employ CSR activities as an insurance-like tool to offset firm risk in both the short and long terms.
Third, senior executives with financial experience may improve CSR performance by improving the disclosure quality of CSR information. Information disclosure is one of the general responsibilities of financial professionals. They are in a good position to understand the financial implications of good reporting practices. Regarding CSR reporting, previous literature has documented that auditing committee members with financial expertise have a good understanding of the financial implications of good CSR reporting practices and compliance with recommended best reporting practices in particular (Li et al. 2012; Appuhami and Tashakor 2017). High-quality CSR reports can, for example, reduce the cost of capital (Dhaliwal et al. 2011) and improve reputation (Brammer and Millington 2005). Dhaliwal et al. (2014) contend that CSR improves with scrutiny from financial analysts since extensive coverage exerts pressure on firms to disclose information, including information related to social issues. One would expect senior executives with financial experience to act in a similar way to improve CSR disclosure quality and, hence, CSR performance.
Finally, owing to their experience in financial institutions, senior executives may be more sensitive to stakeholder needs and expectations. This is because financial institutions’ business nature plays an important role among sectors in contributing to sustainable development and is required for developing socially responsible investment (Crifo and Forget 2013).6 In particular, the Chinese market provides a better context to investigate this research issue, as Chinese financial institutions were under stronger government supervision/control than those in other countries in the 1990s and early 2000s. China has enacted policies to encourage the financial sector to develop green finance (i.e., China’s 11th and 12th five-year plans in 2006 and 2011, respectively) (Dong et al. 2021). Senior executives who used to be among such responsible investors were more likely to know the expectations of these financial stakeholders (i.e., as shareholders or debtholders). Furthermore, employees of financial institutions usually have broad networks, such as clients from various industries, peer colleagues and financial professional organizations. The different external networks enable them to be more sensitive to the demands of stakeholders (Bear et al. 2010). For these reasons, they may have high sensitivity to society and stakeholder interests and motivation to vote for CSR activities (Ramón-Llorens et al. 2019).
However, given that China is characterized by inferior governance mechanisms and weak law enforcement (Guariglia and Yang 2016), conflicts of interest between corporate executives and shareholders may be likely to result in the opportunistic use of philanthropic activities or overinvestment. Moreover, Krüger (2015) draws on the agency/negative view of CSR and argues that executives may use CSR engagement for reputation building, wasteful pet projects, and social standing at the expense of shareholder wealth. This may indicate that CSR improvement may be driven by the opportunistic managerial use of philanthropy. In contrast, we posit that senior executives with financial working experience may better perceive the long-run benefits of CSR and the importance of stakeholder-oriented engagement. Based on the above discussions, senior executives with financial working experience have the potential to actively propel CSR practices. Therefore,
Hypothesis 1
Companies with senior executives who have financial experience have significantly better CSR performance than those without such executives, ceteris paribus.
Experiences of exposure to CSR activities also help firms take on more CSR activities (Lau et al. 2016). If a senior executive has working experience in a regulatory-oriented financial institution (i.e., regulatory commission, policy bank, stock exchange), s/he should have a CSR mindset and be more likely to support CSR strategies. This is because regulatory-oriented financial institutions owned by the government usually take CSR into consideration. For instance, private equity funds invested by government funds have better CSR performance than others (Crifo and Forget 2013). Thus,
Hypothesis 2
Companies with senior executives who have working experience in regulatory-oriented financial institutions have significantly better CSR performance than those without such executives, ceteris paribus.
As discussed earlier, firms with senior executives with financial experience have incentives and the capability to propel CSR investments and strategies, thereby pushing firms towards better sustainability performance. Finkelstein and Hambrick (1990) argue that senior executives, however, do not always have a full latitude of action (also known as managerial discretion or flexibility) in terms of decision making. According to the UET, whether senior executives’ demographic characteristics have a stronger or weaker influence on firms’ business plans or organizational outcomes largely depends on how much discretion these executives have (Xie 2014; Ma et al. 2019). Prior research has documented some important political and organizational factors that could affect the scope of decision-making flexibility afforded to corporate executives, for example, the control of the state (Li and Tang 2010) and the availability of slack resources (Tang et al. 2015). Motivated by these studies, we further investigate whether these factors moderate the influence of senior executives’ financial experience on CSR.
The presence of government control in Chinese companies has a great influence on managerial decisions (Lau et al. 2016). To a large extent, senior executives’ latitude of action is restricted by state ownership (Li and Tang 2010; Liu and Liu 2013; Shen et al. 2020). Higher state ownership and ultimate government control mean higher pressure to comply with the government’s macro-level strategies (Delios et al. 2008), including, for example, maintaining and increasing the value of state-owned assets. Also, SOE firms, which are more sensitive to government policy changes and exhibit a strong state intervention (Liu et al. 2017), may not allow a high level of managerial discretion and flexibility. In other words, TMT per se has a weak influence on some of the companies’ strategies. Hence, the explanatory power of senior executives’ demographic backgrounds is weak. In the same way, the TMT of state-owned companies can lose control of CSR-related decisions because institutional pressure drives these companies to engage in more CSR activities to show their social responsibility and echo the call of the government regarding CSR strategies (Marquis and Qian 2014; Lau et al. 2016). Therefore, we expect that in state-owned companies, senior executives’ financial experience will have a weaker influence on CSR decisions.
Hypothesis 3
State control weakens the positive relationship between the financial experience of the TMT and CSR performance.
A firm’s financial slack is a crucial factor of managerial decisions on CSR. Firms with financial slack are more likely to invest in CSR and exhibit higher CSR ratings because free resources provide corporate executives more flexibility and make them less resistant to meeting stakeholders’ demands (Surroca et al. 2010; Arora and Dharwadkar 2011; Marquis and Qian 2014; Li et al. 2021b). The CSR investment advice of senior executives with financial experience may be more likely to be accepted by other senior managers when there are more free sources. Hence, we expect financial slack to strengthen the positive link between CSR and managerial financial experience.
Hypothesis 4
If a company has more financial slack, the positive link between the financial experience of the TMT and CSR performance is stronger.
4 Research design
4.1 Data and sample
We start with all Chinese A-share companies listed on either the Shenzhen Stock Exchange (SZSE) or Shanghai Stock Exchange (SSE) with CSR performance ratings during the period from 2009 to 2018, where A-shares, also known as domestic shares, are the stock shares of mainland Chinese firms that trade on the above two stock exchanges and are stock shares that are denominated in Renminbi.7 Data on CSR performance and its dimensions are extracted from the RKS Runling rating system, which evaluates the quality of socially sustainable and environmentally friendly corporate conduct and covers Chinese listed companies issuing sustainability reports (Marquis and Qian 2014; Lau et al. 2016; McGuinness et al. 2017; Guo et al. 2018; Long et al. 2020; Li et al. 2021b).8
We then extract data on the TMT’s financial experience, financial and accounting data, and governance quality from the China Stock Market and Accounting Research (CSMAR) Platform, and combine the information on TMTs’ financial experience with these control variables. This generates an initial sample of 20,914 firm-year observations. We next exclude all financial industry companies because some financial fundamentals (i.e., the leverage ratio) that are normal for these companies may not always have the same meaning as for non-financial companies. This results in a sample of 20,688 firm-year observations. After requiring non-missing data on the TMT’s financial experience, CSR performance ratings, and control variables, we are left with a panel data sample of 5,158 firm-year observations and 847 listed companies between 2009 and 2018. To eliminate the influence of outliers on our results, we winsorize all continuous variables at the 1st and 99th percentiles of their respective distributions.
4.2 Measurement of key variables
4.2.1 Proxies for CSR
The CSR performance score (CSRSCORE) is a composite proxy of the sustainable performance of a company. This rating ranges from 0 to 100, with a higher value indicating better CSR engagement. The CSR ratings encapsulate a firm’s strategy, ability, and orientation to meet stakeholders’ concerns and focus on charitable and philanthropic works. This composite performance measure includes three dimensions under the headings of MSCORE, CSCORE and TSCORE. The level of MSCORE evaluates the quality of sustainability strategies and CSR information disclosure, the quality of the engagement of stakeholder-oriented activities, the assessment of risk control mechanisms, sustainability values and codes of conduct. CSCORE evaluates a company’s fair business operations, economic outcome, effectiveness of environmental protection, human rights and labour rights protections, customer concerns and feedback, and contributions to community development. TSCORE evaluates the coverage, scope, accuracy, normalization, consistency of sustainable reporting, and reporting innovation. Our study also uses the length of CSR reports (Page) as an alternative measurement of CSR engagement, which is measured as the natural logarithm of the number of pages in the CSR report. Regarding CSR disclosure and reporting, some companies choose to publish this information as part of their annual reports or sustainability reports. Others use stand-alone CSR reports with improved readability and more reporting material to show their dedication and attempts to increase accountability and efficiency (Li et al. 2021b). As noted by Dhaliwal et al. (2011) and Dhaliwal et al. (2014), the required rate of return by investors is generally lower for firms that issue more comprehensive and detailed standalone CSR reports.
4.2.2 Measuring TMT members’ financial experience
Our main independent variable, Financial_D, is a categorical variable that equals one if a company has at least one senior executive with financial experience and zero otherwise. Specifically, the financial experience by CSMAR refers to the experience gained by senior executives who have working experience in regulatory commissions, policy and commercial banks, insurance firms and investment banks, fund management firms, securities depository and clearing organizations, futures and trust firms, investment management firms, and stock exchanges (Shahab et al. 2020).9 Following Wiengarten et al. (2017), Reimer et al. (2018), Ma et al. (2019), Shen et al. (2020), and Ma et al. (2020), we define the TMT as comprising senior executives who directly engage in a company’s investment and financing choices and corporate policies, including a company’s CEO, executive chairperson, vice president, chief financial officer (CFO), financial controller, general manager, executive/deputy general manager, and vice manager.10 To further examine the extent to which senior executives with financial experience influence CSR, we introduce Financial_Ratio, which is calculated as the number of financial expert executives divided by the total number of TMT members as a robustness test.
4.3 Empirical model
To examine our hypotheses, we use the following ordinary least squares (OLS) regression (Ferrell et al. 2016; Liang and Renneboog 2017b):where Financial_D denotes the presence of senior executives with financial experience in a firm; CSRSCORE is a measure of the overall CSR rating of firm i in year t. Based on Hypothesis 1, we anticipate the estimate on Financial_D to be significantly positive.
$$CSRSCORE_{i,t} = {\upalpha } + \beta_{1} Financial\_D_{i,t - 1} + \beta_{2} Control_{i,t - 1} + Dummy_{year} + Dummy_{industry} + \varepsilon_{i,t}$$
(1)
In accordance with prior CSR studies (Ferrell et al. 2016; Liang and Renneboog 2017a; McGuinness et al. 2017; Li et al. 2021b), we control for a set of factors (Control) that may influence CSR. More specifically, we control for the influence of the state-owned enterprises (SOEs), a categorical variable equal to one if a corporation’s ultimate controlling shareholder is the state or government, and zero otherwise. We also control for company size (Size), which is calculated as the book value of total assets in the form of a natural logarithm, since larger companies are likely to have more resources to propel CSR engagements (Li et al. 2021b). Firm age (Age), measured as the natural logarithm of the number of years since listing, can influence CSR in both directions. Mature firms are more able to engage in CSR (Boubakri et al. 2016), while new or start-up firms have incentives to invest in CSR to establish their reputation and enhance their competitive advantages (Farag et al. 2015).
A company’s leverage (Lev) is the ratio of total liabilities to total assets and indicates resource constraints on socially responsible activities. Return on assets (ROA) is calculated as the earnings before interest and taxes scaled by the book value of total assets. Profitable companies are more likely to have financial resources for socially sustainable activities and thus higher CSR performance (Boubakri et al. 2016). By contrast, firms with weak past operating performance may use CSR strategies as a reputation-building tool (Farag et al. 2015), thus resulting in a negative sign for ROA. A firm’s growth opportunities are captured by Tobin’s Q (Q). Tobin’s Q can affect CSR both positively and negatively. High-growth companies have the capability to take on CSR initiatives (Ferrell et al. 2016). By contrast, it may have a negative influence because of the financial constraints imposed by other investment projects (Li et al. 2021b).
Further, we allow for the effect of ownership concentration, HERF10, defined as the sum of squares of the shareholding percentage of the top ten negotiable shareholders (also known as the Herfindahl index for a company’s ownership by top-ten shareholders). Large shareholders may demand stronger social responsibility commitment (Li and Zhang 2010). A company’s free cash flow (FCF), measured as the ratio of the net operating cash flow to the book value of total assets, is included as a control because cash-rich companies are better able to engage in CSR. We control for RD_Intensity, measured as R&D expenditures divided by total sales. If there exists a strong inclination towards R&D expenditures, companies are likely to choose to expend fewer resources on CSR (Pavelin and Porter 2008). Alternatively, drawing on the resource-based theoretical perspective, CSR and R&D activities may bring together valuable resources and establish competitive edges, thus driving a positive link between CSR and R&D intensity (Hegde and Mishra 2019; Li et al. 2021b).
Following McGuinness et al. (2017) and Li et al. (2021b), we also control for several proxies for a company’s governance quality, for example, the natural logarithm of the total number of directors (BoardSize), CEO-chair duality (Duality), the proportion of independent directors (Indep) on the board, the size of the executive team (Managerial), and board gender diversity (Female). Having more board directors and independent directors is associated with better CSR performance, while CEO-chair duality is negatively related to CSR ratings because companies with better internal governance quality are more likely to behave in socially responsible ways. The size of the executive team, measured as the natural logarithm of the total number of executive managers, matters in promoting effective CSR. Female is calculated as the proportion of female directors on board. A higher proportion of female directors is associated with higher CSR ratings. Finally, we control for a categorical variable, SSE, to account for the effects of stock exchange guidelines on corporate social conduct.11 This variable is set to one if the company is listed on the SSE and zero otherwise.
This study includes year and industry dummies to capture time and industry effects. A firm fixed-effect model is used for robustness tests. We lag all explanatory variables except SSE by one year. The variable construction and data sources are displayed in Appendix 1.
5 Empirical results
5.1 Univariate results
Table 1 displays the annual distributions of CSR ratings and the characteristics of senior executives with financial working experience during the sample period. The percentage of firms having senior executives with financial experience has increased over the years, from nearly 24.45% in 2009 to 28.92% in 2018. In addition, the percentage of senior executives with financial experience has increased over the years. The average CSR rating in 2009 is approximately 28.8922, while it significantly increases to 41.7502 in 2018. Taken together, CSR performance increases as the percentage of firms with senior executives who have financial experience increases.
Table 1
Full sample distribution
Year | Percentage of firms with senior executives who have financial experience | Proportion of executives with financial experience on the TMT | CSR performance | No. of obs |
|---|---|---|---|---|
2009 | 0.2445 | 0.0478 | 28.8922 | 274 |
2010 | 0.2555 | 0.0546 | 31.6110 | 411 |
2011 | 0.2598 | 0.0579 | 34.1646 | 281 |
2012 | 0.2649 | 0.0508 | 36.0147 | 521 |
2013 | 0.2543 | 0.0482 | 37.9659 | 586 |
2014 | 0.2516 | 0.0466 | 39.4625 | 624 |
2015 | 0.2504 | 0.0465 | 41.6688 | 643 |
2016 | 0.1979 | 0.0338 | 41.6022 | 672 |
2017 | 0.2314 | 0.0389 | 42.3867 | 700 |
2018 | 0.2892 | 0.0512 | 41.7502 | 446 |
Average/total | 0.2470 | 0.0462 | 38.6169 | 5158 |
The descriptive statistics are shown in Table 2. CSRSCORE has a mean (median) value of 38.62 (36.03), which is comparable to the findings reported by McGuinness et al. (2017), Elmagrhi et al. (2019), and Li et al. (2021b). CSR_Industry, an industry-mean-adjusted CSR rating, ranges between − 35.67 and 47.06 and has a mean (median) value of approximately 0.03 (− 1.61). The low average CSR ratings show that Chinese companies are still significantly behind in regard to implementing CSR guidelines. A total of 24.7% of the sample firms have financial experts on their management teams (Financial_D). The mean value (0.0462) of Financial_Ratio indicates that on average, 4.62% of members of the TMT have financial experience. The mean value of Financial_regulatory (Financial_nonregulatory) is 0.0111 (0.2359), suggesting that 1.11% (23.59%) of sample firms have senior executives with financial experience in regulatory authorities (nonregulatory authorities). A total of 5.33% of the sample firms have a CEO with a functional background in finance (Financial_CEO).
Table 2
Descriptive statistics
Variable | No. of obs | Mean | SD | Min | P25 | P50 | P75 | Max |
|---|---|---|---|---|---|---|---|---|
CSRSCORE | 5158 | 38.6169 | 12.1592 | 14.1400 | 30.2800 | 36.0304 | 44.3377 | 87.9478 |
MSCORE | 4884 | 13.2619 | 4.4435 | 2.8100 | 10.0800 | 12.8906 | 15.9375 | 27.5625 |
CSCORE | 4884 | 17.2572 | 5.7795 | 2.6367 | 13.3447 | 16.3100 | 20.2148 | 39.5903 |
TSCORE | 4884 | 6.9334 | 1.8926 | 1.7763 | 5.6250 | 6.5132 | 7.7961 | 17.6500 |
CSR_Industry | 5158 | 0.0266 | 10.3259 | − 35.6672 | − 6.7729 | − 1.6149 | 4.5585 | 47.0567 |
Page | 3669 | 2.8571 | 0.8127 | 0.6931 | 2.1972 | 2.7726 | 3.4340 | 5.3083 |
MandatoryPolicy | 20,688 | 0.1083 | 0.3107 | 0.0000 | 0.0000 | 0.0000 | 0.0000 | 1.0000 |
CSR_Presence | 20,688 | 0.2493 | 0.4326 | 0.0000 | 0.0000 | 0.0000 | 0.0000 | 1.0000 |
Financial_D | 5158 | 0.2470 | 0.4313 | 0.0000 | 0.0000 | 0.0000 | 0.0000 | 1.0000 |
Financial_D_ExCEOCFO | 5158 | 0.1644 | 0.3707 | 0.0000 | 0.0000 | 0.0000 | 0.0000 | 1.0000 |
Financial_regulatory | 5158 | 0.0111 | 0.1046 | 0.0000 | 0.0000 | 0.0000 | 0.0000 | 1.0000 |
Financial_nonregulatory | 5158 | 0.2359 | 0.4246 | 0.0000 | 0.0000 | 0.0000 | 0.0000 | 1.0000 |
Financial_CEO | 5158 | 0.0533 | 0.2247 | 0.0000 | 0.0000 | 0.0000 | 0.0000 | 1.0000 |
Financial_Ratio | 5158 | 0.0462 | 0.1018 | 0.0000 | 0.0000 | 0.0000 | 0.0000 | 1.0000 |
SOE | 5158 | 0.6262 | 0.4839 | 0.0000 | 0.0000 | 1.0000 | 1.0000 | 1.0000 |
Size | 5158 | 22.9928 | 1.4450 | 18.2659 | 21.9548 | 22.8300 | 23.8405 | 28.5087 |
Age | 5158 | 2.3638 | 0.6168 | 0.0000 | 2.0794 | 2.5649 | 2.8332 | 3.3322 |
Lev | 5158 | 0.4955 | 0.2002 | 0.0478 | 0.3485 | 0.5090 | 0.6484 | 1.0548 |
ROA | 5158 | 0.0468 | 0.0614 | − 0.2878 | 0.0149 | 0.0401 | 0.0760 | 0.2300 |
Q | 5158 | 1.7766 | 1.1987 | 0.5520 | 1.0512 | 1.3933 | 2.0524 | 10.4585 |
HERF10 | 5158 | 0.1919 | 0.1296 | 0.0141 | 0.0878 | 0.1675 | 0.2684 | 0.5636 |
FCF | 5158 | 0.0362 | 0.1705 | − 0.7972 | − 0.0263 | 0.0631 | 0.1345 | 0.4203 |
RD_Intensity | 5158 | 0.0223 | 0.0330 | 0.0000 | 0.0000 | 0.0090 | 0.0340 | 0.2170 |
BoardSize | 5158 | 2.3118 | 0.1883 | 1.6094 | 2.1972 | 2.3026 | 2.3979 | 2.9444 |
Duality | 5158 | 0.1634 | 0.3698 | 0.0000 | 0.0000 | 0.0000 | 0.0000 | 1.0000 |
Indep | 5158 | 0.3740 | 0.0562 | 0.3077 | 0.3333 | 0.3636 | 0.4000 | 0.5714 |
Managerial | 5158 | 2.0664 | 0.3279 | 0.0000 | 1.7918 | 2.0794 | 2.3026 | 3.3673 |
Female | 5158 | 0.1454 | 0.0997 | 0.0000 | 0.0690 | 0.1304 | 0.2000 | 0.5556 |
SSE | 5158 | 0.5931 | 0.4913 | 0.0000 | 0.0000 | 1.0000 | 1.0000 | 1.0000 |
AnalystCoverage | 5158 | 2.0346 | 1.1007 | 0.0000 | 1.0986 | 2.1972 | 2.9444 | 4.1897 |
BigFour | 5158 | 0.1503 | 0.3574 | 0.0000 | 0.0000 | 0.0000 | 0.0000 | 1.0000 |
TMT_AcademicRatio | 5158 | 0.0996 | 0.1492 | 0.0000 | 0.0000 | 0.0000 | 0.1667 | 1.0000 |
TMT_ForeignRatio | 5158 | 0.0456 | 0.1045 | 0.0000 | 0.0000 | 0.0000 | 0.0000 | 1.0000 |
TMT_Age | 5158 | 3.8571 | 0.0773 | 3.5205 | 3.8089 | 3.8636 | 3.9120 | 4.1026 |
MTB | 4526 | 2.4601 | 2.3829 | 0.0384 | 1.1450 | 1.8301 | 2.9486 | 25.1837 |
Table 3 displays the correlation matrix. The correlation coefficient between Financial_D and CSRSCORE (MSCORE, CSCORE, TSCORE, and Page) is significantly positive. This lends initial support to our Hypothesis 1 (H1). Regarding the operating performance, the correlation coefficient between ROA and FCF is 0.2019, which is at an acceptable level, indicating that our result might not be driven by the collinearity problem. None of the other variables are highly correlated, thus eliminating potential concerns with respect to multicollinearity.
Table 3
Correlation matrix
Variable | (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | (11) | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
(1) | CSRSCORE | 1.0000 | ||||||||||
(2) | MSCORE | 0.9144* | 1.0000 | |||||||||
(3) | CSCORE | 0.9308* | 0.7379* | 1.0000 | ||||||||
(4) | TSCORE | 0.8172* | 0.7377* | 0.6934* | 1.0000 | |||||||
(5) | Page | 0.8799* | 0.8361* | 0.8279* | 0.7266* | 1.0000 | ||||||
(6) | Financial_D | 0.0616* | 0.0578* | 0.0529* | 0.0775* | 0.1038* | 1.0000 | |||||
(7) | SOE | 0.1180* | 0.0676* | 0.1516* | 0.0717* | 0.1194* | − 0.1206* | 1.0000 | ||||
(8) | Size | 0.4748* | 0.3984* | 0.4332* | 0.4114* | 0.4277* | 0.0874* | 0.2895* | 1.0000 | |||
(9) | Age | 0.0343* | 0.0504* | − 0.0222 | 0.0466* | 0.0101 | − 0.0509* | 0.2845* | 0.1924* | 1.0000 | ||
(10) | Lev | 0.1095* | 0.0864* | 0.1196* | 0.0913* | 0.1554* | 0.0442* | 0.2330* | 0.4895* | 0.2475* | 1.0000 | |
(11) | ROA | 0.0153 | − 0.0292* | 0.0633* | 0.0038 | 0.0556* | 0.0289* | − 0.1470* | − 0.0341* | − 0.1533* | − 0.4310* | 1.0000 |
(12) | Q | − 0.1091* | − 0.0895* | − 0.1654* | − 0.0688* | − 0.1270* | − 0.0237* | − 0.1616* | − 0.4099* | 0.0300* | − 0.3373* | 0.2078* |
(13) | HERF10 | 0.2029* | 0.1485* | 0.2159* | 0.1376* | 0.1633* | − 0.0092 | 0.2831* | 0.3442* | − 0.1297* | 0.0755* | 0.0722* |
(14) | FCF | 0.0509* | 0.0204 | 0.0786* | 0.0293* | 0.0208 | − 0.0411* | 0.0356* | − 0.0063 | 0.0019 | − 0.0774* | 0.2019* |
(15) | RD_Intensity | 0.0184 | 0.0479* | − 0.0487* | − 0.0178 | − 0.0661* | − 0.0535* | − 0.2655* | − 0.2240* | − 0.2069* | − 0.3012* | − 0.0063 |
(16) | BoardSize | 0.1174* | 0.0721* | 0.1559* | 0.0847* | 0.1273* | − 0.0146 | 0.2442* | 0.2117* | 0.0542* | 0.1024* | − 0.0019 |
(17) | Duality | − 0.0362* | − 0.0142 | − 0.0581* | − 0.0320* | − 0.0450* | 0.0544* | − 0.2806* | − 0.1101* | − 0.1562* | − 0.1204* | 0.0703* |
(18) | Indep | 0.0503* | 0.0515* | 0.0244* | 0.0379* | 0.0275* | 0.0549* | − 0.0228 | 0.1222* | − 0.0301* | 0.0343* | − 0.0126 |
(19) | Managerial | 0.2280* | 0.1877* | 0.2432* | 0.1524* | 0.2279* | 0.0104 | 0.1357* | 0.2739* | − 0.0255* | 0.1528* | 0.0225 |
(20) | Female | − 0.0479* | − 0.0016 | − 0.1053* | − 0.0262* | − 0.0508* | 0.0483* | − 0.2866* | − 0.2316* | 0.0188 | − 0.1757* | 0.1005* |
(21) | SSE | − 0.0088 | − 0.0711* | 0.0158 | 0.0263* | − 0.0150 | 0.0324* | 0.3266* | 0.1818* | 0.1671* | 0.2099* | − 0.0987* |
Variable | (12) | (13) | (14) | (15) | (16) | (17) | (18) | (19) | (20) | (21) | |
|---|---|---|---|---|---|---|---|---|---|---|---|
(12) | Q | 1.0000 | |||||||||
(13) | HERF10 | − 0.1657* | 1.0000 | ||||||||
(14) | FCF | 0.0267* | 0.1125* | 1.0000 | |||||||
(15) | RD_Intensity | 0.2714* | − 0.1880* | − 0.0423* | 1.0000 | ||||||
(16) | BoardSize | − 0.1640* | 0.0578* | 0.0481* | − 0.1140* | 1.0000 | |||||
(17) | Duality | 0.0917* | − 0.1014* | − 0.0255* | 0.1964* | − 0.1579* | 1.0000 | ||||
(18) | Indep | 0.0318* | 0.0891* | − 0.0351* | 0.0329* | − 0.3835* | 0.0933* | 1.0000 | |||
(19) | Managerial | − 0.1141* | 0.0352* | 0.0099 | 0.0323* | 0.2188* | 0.0237* | 0.0051 | 1.0000 | ||
(20) | Female | 0.1753* | − 0.1912* | − 0.0487* | 0.0982* | − 0.1818* | 0.1221* | 0.0180 | − 0.1984* | 1.0000 | |
(21) | SSE | − 0.1346* | 0.2021* | 0.0294* | − 0.2463* | 0.1280* | − 0.1760* | − 0.0170 | 0.0357* | − 0.1195* | 1.0000 |
5.2 Main regression results
5.2.1 Effects of the TMT’s financial experience on a firm’s CSR and its dimensions
To investigate the effects of senior executives’ functional backgrounds in finance on a firm’s social responsibility, we estimate Eq. (1) and display the results in Table 4. Model 1 presents the results of the baseline OLS regression; the coefficient on Financial_D is positive and significant at the 1% level, indicating that firms with senior executives who have financial experience exhibit an increase in subsequent CSR performance. Economically, the CSR rating is approximately 4.05% higher in companies with financial expert executives than in companies without.12 In addition, we introduce a firm fixed-effects model to address concerns related to unobservable time-invariant firm-specific issues and report the results in Model 2. Notably, the estimate on Financial_D remains positive and significant (coefficient = 1.0081, t = 3.6999), and the magnitude of the estimate is similar to that reported in Model 1.
Table 4
Effects of TMT’s financial experience on a firm’s CSR and its dimensions
Dependent variable = | CSRSCORE | CSRSCORE | MSCORE | CSCORE | TSCORE | Page |
|---|---|---|---|---|---|---|
Baseline OLS | Firm fixed effects | Overall quality of CSR strategy, disclosure, and risk control | The effectiveness of the implemented environmental and social strategy | The extent and accuracy of CSR reporting | Sustainability report pages | |
(1) | (2) | (3) | (4) | (5) | (6) | |
Financial_D | 1.5652*** | 1.0081*** | 0.5566*** | 0.6815*** | 0.2929*** | 0.1374*** |
(4.6222) | (3.6999) | (4.5618) | (3.8824) | (4.9369) | (4.7429) | |
SOE | 0.9243** | 0.3330 | 0.3152** | 0.5866*** | − 0.0121 | 0.0819** |
(2.4873) | (0.4467) | (2.3870) | (2.9605) | (− 0.2070) | (2.4994) | |
Size | 3.9677*** | 1.2908*** | 1.2456*** | 1.8310*** | 0.5619*** | 0.2679*** |
(23.6325) | (4.1919) | (21.2281) | (21.5962) | (19.9030) | (20.5388) | |
Age | − 1.4797*** | − 0.9651* | − 0.5494*** | − 0.6330*** | − 0.2774*** | − 0.1200*** |
(− 5.2333) | (− 1.8390) | (− 5.4272) | (− 4.3638) | (− 5.7486) | (− 5.0487) | |
Lev | − 5.0179*** | 0.9502 | − 1.6139*** | − 2.3036*** | − 0.8435*** | − 0.2248** |
(− 5.1065) | (0.9023) | (− 4.4280) | (− 4.4697) | (− 5.1493) | (− 2.5089) | |
ROA | − 7.1133*** | 5.2476** | − 4.7652*** | − 0.9341 | − 1.7307*** | − 0.4526* |
(− 2.7334) | (2.4354) | (− 4.7653) | (− 0.6681) | (− 4.1635) | (− 1.8974) | |
Q | 0.4130*** | − 0.1148 | 0.2030*** | 0.0864 | 0.0869*** | 0.0240** |
(2.9870) | (− 0.9921) | (4.0850) | (1.1960) | (3.8800) | (1.9757) | |
HERF10 | 5.5505*** | − 0.3029 | 2.0320*** | 2.6378*** | 0.5229** | 0.2998** |
(4.0059) | (− 0.1573) | (4.2422) | (3.5742) | (2.4386) | (2.4833) | |
FCF | 2.5112*** | 0.0367 | 0.4643* | 1.6090*** | 0.4858*** | 0.1416** |
(3.3549) | (0.0741) | (1.6769) | (3.9517) | (4.2194) | (2.0353) | |
RD_Intensity | 7.4721 | − 0.1284 | 1.7468 | 4.4191 | 0.3299 | 0.0471 |
(1.3318) | (− 0.0211) | (0.8948) | (1.4963) | (0.4048) | (0.1011) | |
BoardSize | 2.2529** | 0.6121 | 0.8834*** | 1.2177** | 0.3466** | 0.1950** |
(2.3953) | (0.6291) | (2.6120) | (2.5329) | (2.1782) | (2.5488) | |
Duality | − 1.0818*** | − 0.3287 | − 0.3501*** | − 0.4357** | − 0.1698*** | − 0.0836*** |
(− 2.9462) | (− 0.9560) | (− 2.6129) | (− 2.2678) | (− 2.8595) | (− 2.6908) | |
Indep | − 0.2382 | − 3.7755 | 0.4720 | − 0.1821 | − 0.0324 | − 0.0050 |
(− 0.0902) | (− 1.4589) | (0.4946) | (− 0.1310) | (− 0.0732) | (− 0.0219) | |
Managerial | 3.1500*** | 0.4867 | 0.9262*** | 1.7159*** | 0.2259*** | 0.2254*** |
(6.6541) | (1.0790) | (5.4379) | (7.0962) | (2.9994) | (5.7905) | |
Female | 4.3538*** | − 2.2359 | 1.7960*** | 1.7211** | 0.7017*** | 0.4282*** |
(2.8079) | (− 1.3125) | (3.2709) | (2.1130) | (2.8559) | (3.2202) | |
SSE | − 1.6981*** | – | − 0.9634*** | − 0.7816*** | − 0.0584 | − 0.1867*** |
(− 5.6754) | – | (− 8.7198) | (− 4.8951) | (− 1.2276) | (− 7.2683) | |
_intercept | − 68.8300*** | 0.2462 | − 22.9928*** | − 29.0217*** | − 7.3801*** | − 4.4075*** |
(− 17.0539) | (0.0350) | (− 15.4883) | (− 13.8197) | (− 11.1943) | (− 13.2877) | |
Year effects | Yes | Yes | Yes | Yes | Yes | Yes |
Industry effects | Yes | No | Yes | Yes | Yes | Yes |
Firm fixed effects | No | Yes | No | No | No | No |
No. of obs | 5158 | 5158 | 4884 | 4884 | 4884 | 3669 |
Adj. R-square | 0.405 | 0.273 | 0.446 | 0.322 | 0.379 | 0.300 |
Table 4 also displays the regression results using CSR subcategories (MSCORE in Model 3, CSCORE in Model 4, and TSCORE in Model 5) as the dependent variables. The coefficients on Financial_D in these three models are significantly positive at the 1% level, which further conforms to H1. Finally, we employ Page as a dependent variable in Model 6 to investigate whether firms with senior executives who have financial experience issue lengthier sustainability reports to enhance the readability of CSR disclosure. The positive and highly significant coefficient on Financial_D indicates that senior executives with financial career experience have a positive influence on the contents and readability of corporate sustainable reporting. Overall, the evidence affirms that senior executives with financial experience drive firms to adopt better CSR strategies, thus supporting H1.
5.2.2 TMT’s financial experience in regulatory authorities and nonregulatory authorities
To test the validity of Hypothesis 2 (H2), we introduce two key independent variables to our model specification to investigate the influence of TMT members’ financial experience in regulatory authorities and in nonregulatory authorities on a firm’s CSR. Specifically, Financial_regulatory is a categorical variable assigned a value of one if any executives on the TMT have financial working experience in regulatory authorities (i.e., policy banks, regulatory commissions, or stock exchanges) and zero otherwise. Financial_nonregulatory is a categorical variable assigned a value of one if any executives on the TMT have financial working experience in nonregulatory authorities, and zero otherwise. We replace Financial_D with Financial_regulatory and Financial_nonregulatory in Eq. (1) and re-estimate the specification using the overall CSR and its subcategories and report the results in Table 5. In all models, the coefficients on Financial_regulatory and Financial_nonregulatory are positive and significant; notably, the magnitudes of the coefficients on Financial_regulatory are generally larger than those of the coefficients on Financial_nonregulatory. For example, in Model 1 of Table 5, the estimate on the former is 3.5712 and that on the latter is 1.4740. We reasonably conclude that the improvement in CSR is, to a large extent, driven by senior executives who have work experience in regulatory-oriented financial institutions. Taken together, these results suggest that companies with senior executives who have ever worked in financial regulatory authorities exhibit higher CSR scores than firms with senior executives who have financial experience in nonregulatory authorities, thereby strongly supporting H2.
Table 5
Effects of TMT’s financial experience in regulatory authorities and nonregulatory authorities on a firm’s CSR and its dimensions
Dependent variable = | CSRSCORE | CSRSCORE | MSCORE | CSCORE | TSCORE | Page |
|---|---|---|---|---|---|---|
OLS | Firm fixed effects | Overall quality of CSR strategy, disclosure, and risk control | The effectiveness of the implemented environmental and social strategy | The extent and accuracy of CSR reporting | Sustainability report pages | |
(1) | (2) | (3) | (4) | (5) | (6) | |
Financial_regulatory | 3.5712*** | 2.8805** | 1.6736*** | 1.1123* | 0.6330*** | 0.2087* |
(2.7962) | (2.5103) | (3.7600) | (1.8301) | (2.8711) | (1.8283) | |
Financial_nonregulatory | 1.4740*** | 0.9321*** | 0.5033*** | 0.6610*** | 0.2767*** | 0.1337*** |
(4.2778) | (3.3751) | (4.0585) | (3.6889) | (4.5687) | (4.5350) | |
SOE | 0.9356** | 0.3590 | 0.3223** | 0.5894*** | − 0.0099 | 0.0824** |
(2.5154) | (0.4816) | (2.4393) | (2.9719) | (− 0.1694) | (2.5115) | |
Size | 3.9716*** | 1.2830*** | 1.2480*** | 1.8320*** | 0.5626*** | 0.2681*** |
(23.6443) | (4.1671) | (21.2576) | (21.5987) | (19.9246) | (20.5233) | |
Age | − 1.4952*** | − 0.9435* | − 0.5585*** | − 0.6366*** | − 0.2802*** | − 0.1209*** |
(− 5.2885) | (− 1.7977) | (− 5.5302) | (− 4.3827) | (− 5.7991) | (− 5.0678) | |
Lev | − 5.0283*** | 0.9724 | − 1.6212*** | − 2.3064*** | − 0.8457*** | − 0.2254** |
(− 5.1174) | (0.9234) | (− 4.4475) | (− 4.4754) | (− 5.1656) | (− 2.5149) | |
ROA | − 7.2735*** | 5.1783** | − 4.8577*** | − 0.9698 | − 1.7589*** | − 0.4587* |
(− 2.7967) | (2.4033) | (− 4.8657) | (− 0.6933) | (− 4.2364) | (− 1.9231) | |
Q | 0.4203*** | − 0.1112 | 0.2072*** | 0.0881 | 0.0882*** | 0.0244** |
(3.0367) | (− 0.9603) | (4.1644) | (1.2170) | (3.9324) | (1.9995) | |
HERF10 | 5.5455*** | − 0.3644 | 2.0258*** | 2.6354*** | 0.5210** | 0.2992** |
(4.0023) | (− 0.1892) | (4.2308) | (3.5697) | (2.4302) | (2.4767) | |
FCF | 2.5481*** | 0.0574 | 0.4861* | 1.6175*** | 0.4925*** | 0.1427** |
(3.4048) | (0.1159) | (1.7577) | (3.9685) | (4.2849) | (2.0493) | |
RD_Intensity | 7.8198 | − 0.3140 | 1.9590 | 4.5010 | 0.3946 | 0.0663 |
(1.3929) | (− 0.0516) | (1.0027) | (1.5219) | (0.4841) | (0.1419) | |
BoardSize | 2.2036** | 0.5398 | 0.8540** | 1.2063** | 0.3376** | 0.1923** |
(2.3414) | (0.5544) | (2.5252) | (2.5054) | (2.1199) | (2.5060) | |
Duality | − 1.0658*** | − 0.3180 | − 0.3410** | − 0.4322** | − 0.1670*** | − 0.0834*** |
(− 2.9004) | (− 0.9250) | (− 2.5463) | (− 2.2467) | (− 2.8103) | (− 2.6821) | |
Indep | − 0.2898 | − 3.9537 | 0.4387 | − 0.1950 | − 0.0426 | − 0.0093 |
(− 0.1095) | (− 1.5268) | (0.4581) | (− 0.1401) | (− 0.0959) | (− 0.0402) | |
Managerial | 3.1848*** | 0.4718 | 0.9457*** | 1.7234*** | 0.2318*** | 0.2264*** |
(6.7102) | (1.0459) | (5.5378) | (7.1125) | (3.0693) | (5.8121) | |
Female | 4.4276*** | − 2.2454 | 1.8372*** | 1.7370** | 0.7143*** | 0.4306*** |
(2.8579) | (− 1.3183) | (3.3541) | (2.1326) | (2.9084) | (3.2399) | |
SSE | − 1.6856*** | – | − 0.9565*** | − 0.7789*** | − 0.0563 | − 0.1860*** |
(− 5.6369) | – | (− 8.6667) | (− 4.8796) | (− 1.1839) | (− 7.2369) | |
_intercept | − 68.8070*** | 0.6395 | − 22.9867*** | − 29.0193*** | − 7.3783*** | − 4.4058*** |
(− 17.0459) | (0.0910) | (− 15.4915) | (− 13.8175) | (− 11.1945) | (− 13.2799) | |
Year effects | Yes | Yes | Yes | Yes | Yes | Yes |
Industry effects | Yes | No | Yes | Yes | Yes | Yes |
Firm fixed effects | No | Yes | No | No | No | No |
No. of obs | 5158 | 5158 | 4884 | 4884 | 4884 | 3669 |
Adj. R-square | 0.405 | 0.274 | 0.447 | 0.322 | 0.379 | 0.300 |
5.2.3 Moderating effects of SOEs and slack resources
To test the validity of Hypothesis 3 (H3), we re-estimate Eq. (1) by introducing an interaction term Financial_D × SOE. This interaction term captures the moderating effect of SOEs. The results are presented in Panel A of Table 6. The coefficient on Financial_D in all models is significantly positive, while that on Financial_D × SOE is negative and significant at the 5% level, broadly meaning that TMT members’ financial experience indeed has a positive influence on firms’ CSR and that such a positive influence is more pronounced in non-SOEs than in SOEs. In line with H3, senior executives tend to have a lower latitude of action in SOEs (Li and Tang 2010; Liu and Liu 2013; Shen et al. 2020), mainly due to the higher pressure that managers face regarding compliance with the government’s macro-level strategies (Delios et al. 2008); hence, the influence of senior executives’ financial experience on CSR is greatly weakened.
Table 6
Role of TMT’s latitude of action in the link between financial experience and a firm’s CSR and its dimensions
Dependent variable = | CSRSCORE | CSRSCORE | MSCORE | CSCORE | TSCORE | Page |
|---|---|---|---|---|---|---|
OLS | Firm fixed effects | Overall quality of CSR strategy, disclosure, and risk control | The effectiveness of the implemented environmental and social strategy | The extent and accuracy of CSR reporting | Sustainability report pages | |
(1) | (2) | (3) | (4) | (5) | (6) | |
Panel A Moderating effect of state-own-enterprises (SOEs) in the link between TMT’s financial experience and CSR | ||||||
Financial_D | 2.4892*** | 1.7489*** | 0.8239*** | 1.0881*** | 0.3487*** | 0.1496*** |
(5.1621) | (4.2308) | (4.9237) | (4.3408) | (4.2645) | (3.6449) | |
Financial_D × SOE | − 1.6509** | − 1.2820** | − 0.4832** | − 0.7348** | − 0.1009 | − 0.0234 |
(− 2.4675) | (− 2.3819) | (− 2.0371) | (− 2.1175) | (− 0.8770) | (− 0.4094) | |
SOE | 1.3329*** | 0.6206 | 0.4347*** | 0.7683*** | 0.0129 | 0.0875** |
(3.4650) | (0.8222) | (3.1213) | (3.7167) | (0.2166) | (2.5151) | |
Size | 3.9868*** | 1.2887*** | 1.2509*** | 1.8390*** | 0.5630*** | 0.2681*** |
(23.6913) | (4.1873) | (21.2511) | (21.6382) | (19.9488) | (20.4982) | |
Age | − 1.4814*** | − 0.9156* | − 0.5508*** | − 0.6352*** | − 0.2777*** | − 0.1204*** |
(− 5.2568) | (− 1.7444) | (− 5.4558) | (− 4.3888) | (− 5.7636) | (− 5.0513) | |
Lev | − 5.2466*** | 0.8769 | − 1.6788*** | − 2.4023*** | − 0.8570*** | − 0.2275** |
(− 5.3185) | (0.8327) | (− 4.5783) | (− 4.6482) | (− 5.2420) | (− 2.5291) | |
ROA | − 7.1734*** | 5.1266** | − 4.7864*** | − 0.9664 | − 1.7351*** | − 0.4532* |
(− 2.7500) | (2.3799) | (− 4.7808) | (− 0.6898) | (− 4.1732) | (− 1.8984) | |
Q | 0.4084*** | − 0.1095 | 0.2015*** | 0.0841 | 0.0865*** | 0.0240** |
(2.9436) | (− 0.9459) | (4.0426) | (1.1603) | (3.8601) | (1.9715) | |
HERF10 | 5.5669*** | − 0.3656 | 2.0349*** | 2.6422*** | 0.5235** | 0.2994** |
(4.0161) | (− 0.1899) | (4.2462) | (3.5791) | (2.4416) | (2.4802) | |
FCF | 2.5646*** | 0.0467 | 0.4767* | 1.6278*** | 0.4884*** | 0.1421** |
(3.4153) | (0.0944) | (1.7177) | (3.9894) | (4.2350) | (2.0410) | |
RD_Intensity | 7.1507 | − 0.2608 | 1.6537 | 4.2776 | 0.3105 | 0.0413 |
(1.2677) | (− 0.0428) | (0.8427) | (1.4428) | (0.3803) | (0.0885) | |
BoardSize | 2.2493** | 0.5703 | 0.8842*** | 1.2189** | 0.3467** | 0.1955** |
(2.3943) | (0.5864) | (2.6165) | (2.5385) | (2.1788) | (2.5554) | |
Duality | − 1.0997*** | − 0.3174 | − 0.3563*** | − 0.4452** | − 0.1711*** | − 0.0841*** |
(− 2.9884) | (− 0.9236) | (− 2.6565) | (− 2.3108) | (− 2.8761) | (− 2.6985) | |
Indep | − 0.1416 | − 3.7588 | 0.5000 | − 0.1396 | − 0.0266 | − 0.0027 |
(− 0.0537) | (− 1.4532) | (0.5245) | (− 0.1004) | (− 0.0600) | (− 0.0118) | |
Managerial | 3.1827*** | 0.4807 | 0.9351*** | 1.7293*** | 0.2277*** | 0.2259*** |
(6.7260) | (1.0663) | (5.4937) | (7.1536) | (3.0129) | (5.7970) | |
Female | 4.3919*** | − 2.1774 | 1.8104*** | 1.7430** | 0.7047*** | 0.4296*** |
(2.8305) | (− 1.2787) | (3.2974) | (2.1368) | (2.8706) | (3.2280) | |
SSE | − 1.6805*** | – | − 0.9579*** | − 0.7732*** | − 0.0572 | − 0.1863*** |
(− 5.6146) | – | (− 8.6696) | (− 4.8404) | (− 1.2023) | (− 7.2297) | |
_intercept | − 69.6344*** | 0.1453 | − 23.2086*** | − 29.3499*** | − 7.4252*** | − 4.4164*** |
(− 17.1902) | (0.0207) | (− 15.5539) | (− 13.9134) | (− 11.2151) | (− 13.2331) | |
Year effects | Yes | Yes | Yes | Yes | Yes | Yes |
Industry effects | Yes | No | Yes | Yes | Yes | Yes |
Firm fixed effects | No | Yes | No | No | No | No |
No. of obs | 5158 | 5158 | 4884 | 4884 | 4884 | 3669 |
Adj. R-square | 0.406 | 0.274 | 0.447 | 0.323 | 0.379 | 0.300 |
Panel B Moderating effect of corporate free cash flow in the link between TMT’s financial experience and CSR | ||||||
Financial_D | 1.4523*** | 0.9389*** | 0.5315*** | 0.6344*** | 0.2733*** | 0.1345*** |
(4.2596) | (3.4114) | (4.3212) | (3.5842) | (4.6326) | (4.5801) | |
Financial_D × FCF | 3.8910** | 1.9736* | 0.8951 | 1.6838* | 0.6993** | 0.0980 |
(2.2181) | (1.7673) | (1.4119) | (1.8120) | (2.4682) | (0.6312) | |
SOE | 0.8947** | 0.2995 | 0.3086** | 0.5743*** | − 0.0172 | 0.0811** |
(2.4063) | (0.4017) | (2.3345) | (2.8990) | (− 0.2945) | (2.4701) | |
Size | 3.9643*** | 1.3059*** | 1.2446*** | 1.8292*** | 0.5611*** | 0.2677*** |
(23.6486) | (4.2403) | (21.2376) | (21.5920) | (19.9194) | (20.5279) | |
Age | − 1.4854*** | − 0.9891* | − 0.5499*** | − 0.6340*** | − 0.2778*** | − 0.1202*** |
(− 5.2601) | (− 1.8846) | (− 5.4322) | (− 4.3745) | (− 5.7584) | (− 5.0563) | |
Lev | − 4.9571*** | 0.9355 | − 1.6021*** | − 2.2813*** | − 0.8342*** | − 0.2226** |
(− 5.0612) | (0.8886) | (− 4.4059) | (− 4.4373) | (− 5.1087) | (− 2.4878) | |
ROA | − 7.0424*** | 5.2235** | − 4.7474*** | − 0.9006 | − 1.7168*** | − 0.4519* |
(− 2.7124) | (2.4248) | (− 4.7518) | (− 0.6458) | (− 4.1285) | (− 1.8980) | |
Q | 0.4022*** | − 0.1163 | 0.2001*** | 0.0810 | 0.0846*** | 0.0238* |
(2.9151) | (− 1.0049) | (4.0358) | (1.1215) | (3.7864) | (1.9531) | |
HERF10 | 5.6227*** | − 0.2885 | 2.0521*** | 2.6755*** | 0.5386** | 0.3017** |
(4.0618) | (− 0.1498) | (4.2842) | (3.6279) | (2.5110) | (2.4985) | |
FCF | 1.5758* | − 0.3997 | 0.2465 | 1.1993*** | 0.3157** | 0.1189 |
(1.9008) | (− 0.7220) | (0.7889) | (2.6347) | (2.5096) | (1.5052) | |
RD_Intensity | 7.4984 | − 0.5729 | 1.7556 | 4.4357 | 0.3368 | 0.0487 |
(1.3386) | (− 0.0940) | (0.9002) | (1.5033) | (0.4141) | (0.1046) | |
BoardSize | 2.2792** | 0.6054 | 0.8913*** | 1.2327** | 0.3528** | 0.1958** |
(2.4206) | (0.6224) | (2.6330) | (2.5603) | (2.2142) | (2.5575) | |
Duality | − 1.0964*** | − 0.3478 | − 0.3539*** | − 0.4428** | − 0.1727*** | − 0.0843*** |
(− 2.9839) | (− 1.0114) | (− 2.6400) | (− 2.3025) | (− 2.9072) | (− 2.7108) | |
Indep | − 0.2455 | − 3.7610 | 0.4733 | − 0.1798 | − 0.0314 | − 0.0056 |
(− 0.0930) | (− 1.4536) | (0.4958) | (− 0.1293) | (− 0.0710) | (− 0.0241) | |
Managerial | 3.1588*** | 0.4829 | 0.9270*** | 1.7174*** | 0.2265*** | 0.2255*** |
(6.6721) | (1.0708) | (5.4421) | (7.1027) | (3.0091) | (5.7909) | |
Female | 4.3555*** | − 2.2489 | 1.7924*** | 1.7144** | 0.6989*** | 0.4284*** |
(2.8163) | (− 1.3204) | (3.2684) | (2.1087) | (2.8510) | (3.2246) | |
SSE | − 1.7010*** | – | − 0.9644*** | − 0.7835*** | − 0.0592 | − 0.1872*** |
(− 5.6865) | – | (− 8.7250) | (− 4.9083) | (− 1.2448) | (− 7.2734) | |
_intercept | − 68.9086*** | 0.0299 | − 23.0115*** | − 29.0570*** | − 7.3948*** | − 4.4058*** |
(− 17.0917) | (0.0043) | (− 15.5112) | (− 13.8454) | (− 11.2363) | (− 13.2783) | |
Year effects | Yes | Yes | Yes | Yes | Yes | Yes |
Industry effects | Yes | No | Yes | Yes | Yes | Yes |
Firm fixed effects | No | Yes | No | No | No | No |
No. of obs | 5158 | 5158 | 4884 | 4884 | 4884 | 3669 |
Adj. R-square | 0.406 | 0.274 | 0.446 | 0.323 | 0.379 | 0.300 |
Similarly, we re-run Eq. (1) by including an interaction term Financial_D × FCF, which captures the moderating effect of corporate slack resources (measured by free cash flow) (Li et al. 2021b). The results are displayed in Panel B of Table 6. In all models, the coefficient on Financial_D remains positive and significant at the 1% level. The coefficient on Financial_D × FCF in Models 1–2 and 4–5 is significantly positive at the conventional levels and the coefficient on FCF in Models 1, 4, and 5 is also significantly positive. These results suggest that firms whose executives have financial expertise exhibit a significant increase in future CSR and that such a positive influence is more prominent in cash-abundant firms, which strongly conforms to H4.
In sum, our findings support that when a company with senior executives with financial working experience will exhibit better CSR performance. In particular, the performance is even better when such working experience is from regulatory-oriented financial institutions and when the company is non-state-owned or has more financial slack.
5.3 Robustness checks
5.3.1 Percentage of financial experts on the TMT
As discussed, we use Financial_Ratio as an alternative measure of Financial_D and report the results of OLS and firm fixed effects in Models 1–2 of Table 7, respectively. The estimate on Financial_Ratio is significantly positive in both models; specifically, the estimate indicates that a unit increase in this ratio is associated with an increase in a firm’s CSR performance of approximately 4.1400 (3.4966) units in Model 1 (2). These results are consistent with our baseline result.
Table 7
Robustness checks
Percentage of financial experts on the TMT | Effects of CEOs with financial experience | Excluding CEOs and CFOs from the TMT | Industry-mean-adjusted CSR | Analyst coverage | Audit quality | TMT heterogeneity | Sample and weights issues | 2009–2013 | 2014–2018 | |||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
Dependent variable = | CSRSCORE | CSRSCORE | CSRSCORE | CSRSCORE | CSRSCORE | CSR_Industry | CSRSCORE | CSRSCORE | CSRSCORE | CSRSCORE | CSRSCORE | CSRSCORE |
OLS | Firm fixed effects | OLS | Firm fixed effects | OLS | OLS | OLS | OLS | OLS | WLS | OLS | OLS | |
(1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) | (9) | (10) | (11) | (12) | |
Financial_Ratio | 4.1400*** | 3.4966** | ||||||||||
(2.6544) | (2.0985) | |||||||||||
Financial_CEO | 3.3809*** | 2.1323*** | ||||||||||
(4.5981) | (3.9079) | |||||||||||
Financial_Ratio_ExCEOCFO | 1.2012*** | |||||||||||
(3.1422) | ||||||||||||
Financial_D | 1.4726*** | 1.5870*** | 1.6868*** | 1.5772*** | 1.0900*** | 1.0012* | 2.0416*** | |||||
(4.4423) | (4.6896) | (5.0430) | (4.7229) | (3.6356) | (1.8817) | (4.5813) | ||||||
AnalystCoverage | 0.2210 | 0.2893* | 0.2960* | |||||||||
(1.3434) | (1.7982) | (1.8530) | ||||||||||
BigFour | 4.9688*** | 4.8706*** | ||||||||||
(8.8371) | (8.6909) | |||||||||||
TMT_AcademicRatio | 5.5745*** | |||||||||||
(5.7251) | ||||||||||||
TMT_ForeignRatio | 4.6388*** | |||||||||||
(3.1606) | ||||||||||||
TMT_Age | 6.2294*** | |||||||||||
(3.0803) | ||||||||||||
SOE | 0.8319** | 0.2818 | 0.8564** | 0.1484 | 0.8195** | 0.8890** | 0.9457** | 0.9709*** | 0.7337** | 1.0886*** | 0.5559 | 1.0738** |
(2.2156) | (0.2728) | (2.3155) | (0.1458) | (2.1832) | (2.4237) | (2.5419) | (2.6551) | (1.9959) | (3.3061) | (0.9390) | (2.1773) | |
Size | 4.0129*** | 1.3141*** | 4.0162*** | 1.2733*** | 4.0364*** | 3.8723*** | 3.8667*** | 3.1673*** | 2.9198*** | 3.1324*** | 4.1639*** | 3.9528*** |
(23.9427) | (2.6683) | (24.2145) | (2.5955) | (24.1683) | (23.5640) | (20.8322) | (16.8548) | (15.4907) | (20.4609) | (15.6018) | (17.7132) | |
Age | − 1.5503*** | − 1.0472 | − 1.6186*** | − 1.0510 | − 1.5080*** | − 1.4934*** | − 1.4321*** | − 1.2335*** | − 1.1027*** | − 1.1116*** | − 2.1420*** | − 1.0142*** |
(− 5.4919) | (− 1.1588) | (− 5.7496) | (− 1.1654) | (− 5.3358) | (− 5.4211) | (− 5.0380) | (− 4.3198) | (− 3.9168) | (− 4.3123) | (− 4.6315) | (− 2.7904) | |
Lev | − 5.0507*** | 0.9794 | − 5.0481*** | 0.8895 | − 5.1248*** | − 4.6428*** | − 4.9588*** | − 3.8227*** | − 3.7630*** | − 3.9191*** | − 5.2434*** | − 4.6224*** |
(− 5.1224) | (0.7206) | (− 5.1325) | (0.6643) | (− 5.1894) | (− 4.8275) | (− 5.0434) | (− 3.8858) | (− 3.8637) | (− 4.3806) | (− 3.2955) | (− 3.5649) | |
ROA | − 7.3130*** | 5.3035* | − 6.8708*** | 5.3812* | − 7.4627*** | − 7.7310*** | − 8.3880*** | − 8.3076*** | − 7.7136*** | − 4.1391* | − 8.9869** | − 7.0902** |
(− 2.8069) | (1.8449) | (− 2.6551) | (1.8896) | (− 2.8650) | (− 3.0430) | (− 3.0088) | (− 3.0006) | (− 2.7991) | (− 1.6649) | (− 2.0714) | (− 2.1240) | |
Q | 0.4179*** | − 0.1207 | 0.4109*** | − 0.1356 | 0.4365*** | 0.3871*** | 0.3839*** | 0.3128** | 0.2523* | 0.2888** | 0.5557** | 0.4302** |
(3.0178) | (− 0.8649) | (2.9730) | (− 0.9783) | (3.1713) | (2.9177) | (2.7688) | (2.2681) | (1.8329) | (2.1536) | (2.0996) | (2.5453) | |
HERF10 | 5.4744*** | − 0.4194 | 5.3827*** | − 0.4845 | 5.5628*** | 5.1969*** | 5.6410*** | 4.5963*** | 4.5228*** | 5.2377*** | 3.4932 | 7.0026*** |
(3.9492) | (− 0.1625) | (3.8764) | (− 0.1892) | (4.0060) | (3.7794) | (4.0663) | (3.4051) | (3.3996) | (4.2551) | (1.6332) | (3.7479) | |
FCF | 2.4715*** | 0.0359 | 2.5277*** | 0.0467 | 2.4868*** | 2.1549*** | 2.5751*** | 1.9881*** | 1.8607** | 1.8301*** | 2.6035** | 1.9581** |
(3.3004) | (0.0773) | (3.3676) | (0.1005) | (3.3232) | (2.9439) | (3.4365) | (2.6635) | (2.4796) | (2.6422) | (2.1777) | (2.0432) | |
RD_Intensity | 7.1406 | − 0.4486 | 7.6273 | 0.1276 | 7.6619 | 10.1714* | 6.6749 | 4.3319 | 2.8265 | 4.9016 | 32.1277*** | 0.2089 |
(1.2728) | (− 0.0643) | (1.3659) | (0.0183) | (1.3612) | (1.8952) | (1.1801) | (0.7689) | (0.5018) | (0.8801) | (2.9727) | (0.0313) | |
BoardSize | 2.2579** | 0.6925 | 2.3931** | 0.7392 | 2.2076** | 2.2067** | 2.1955** | 1.9482** | 2.0659** | 2.2272** | 0.4455 | 3.3066*** |
(2.3967) | (0.4990) | (2.5425) | (0.5343) | (2.3448) | (2.4093) | (2.3367) | (2.0999) | (2.2356) | (2.5701) | (0.2760) | (2.8055) | |
Duality | − 1.0745*** | − 0.3416 | − 1.2467*** | − 0.4237 | − 1.0093*** | − 1.1958*** | − 1.0840*** | − 1.0414*** | − 1.2580*** | − 0.6831* | − 0.3467 | − 1.6252*** |
(− 2.9125) | (− 0.7504) | (− 3.3252) | (− 0.9390) | (− 2.7568) | (− 3.3110) | (− 2.9514) | (− 2.8468) | (− 3.4408) | (− 1.9580) | (− 0.5413) | (− 3.6027) | |
Indep | − 0.0539 | − 3.7097 | 0.1892 | − 3.7096 | − 0.0235 | − 0.0721 | − 0.1807 | − 1.9915 | − 3.0363 | 0.3788 | − 1.9361 | 1.1585 |
(− 0.0204) | (− 0.9391) | (0.0721) | (− 0.9366) | (− 0.0089) | (− 0.0279) | (− 0.0683) | (− 0.7576) | (− 1.1654) | (0.1523) | (− 0.4700) | (0.3302) | |
Managerial | 3.3192*** | 0.6195 | 3.1671*** | 0.5360 | 3.1281*** | 3.0474*** | 3.0862*** | 2.9132*** | 2.9474*** | 3.4223*** | 3.7591*** | 2.8468*** |
(6.8922) | (1.0510) | (6.7033) | (0.9054) | (6.6009) | (6.6040) | (6.4471) | (6.2546) | (6.3713) | (7.9761) | (4.8912) | (4.6540) | |
Female | 4.3001*** | − 2.2340 | 3.9031** | − 2.5235 | 4.3541*** | 4.5244*** | 4.2927*** | 4.5780*** | 5.1678*** | 6.4286*** | 8.4047*** | 2.4463 |
(2.7714) | (− 0.9215) | (2.5089) | (− 1.0378) | (2.8070) | (2.9575) | (2.7641) | (3.0216) | (3.3987) | (4.2987) | (3.0144) | (1.3045) | |
SSE | − 1.6502*** | – | − 1.6294*** | – | − 1.6177*** | − 1.5967*** | − 1.6603*** | − 1.8958*** | − 1.4504*** | − 1.9969*** | − 1.6387*** | − 1.7264*** |
(− 5.4957) | – | (− 5.4460) | – | (− 5.3816) | (− 5.4126) | (− 5.5275) | (− 6.4071) | (− 4.7669) | (− 7.2521) | (− 3.2947) | (− 4.5504) | |
_intercept | − 69.7859*** | − 0.4748 | − 69.5088*** | 0.7826 | − 70.0237*** | − 91.7208*** | − 66.8679*** | − 50.7366*** | − 70.3375*** | − 49.9554*** | − 63.1054*** | − 66.4104*** |
(− 17.2527) | (− 0.0427) | (− 17.2461) | − 0.0708 | (− 17.3749) | (− 23.8799) | (− 15.4930) | (− 11.4346) | (− 8.5165) | (− 13.4131) | (− 10.4616) | (− 12.3394) | |
Year effects | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
Industry effects | Yes | No | Yes | No | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
Firm fixed effects | No | Yes | No | Yes | No | No | No | No | No | No | No | No |
No. of obs | 5158 | 5158 | 5158 | 5158 | 5158 | 5158 | 5158 | 5158 | 5158 | 5158 | 2073 | 3085 |
Adj. R-square | 0.403 | 0.392 | 0.406 | 0.394 | 0.404 | 0.206 | 0.405 | 0.420 | 0.427 | 0.405 | 0.383 | 0.343 |
5.3.2 Effect of CEOs with financial expertise on CSR
We next investigate the effect of CEOs’ financial experience because CEOs may have a greater impact on companies’ strategic management process, implementation of policies, and decision making than other senior executives (Custódio and Metzger 2014; Shahab et al. 2020). As noted by Zhang and Wiersema (2009), the attributes of the CEO send important signals to the investment community regarding the credibility of the CEO certification. Hence, the quality of the company’s financial statements, in turn, affects the stock market reaction to CEO certification. Early formative experiences, such as financial experience, may shape individuals’ belief systems and behaviours and may influence their actions when they become senior executives, such as CEOs (Gibbons and Waldman 2004; Ma et al. 2019, 2020). We, therefore, employ a categorical variable, Financial_CEO, and re-estimate Eq. (1). Financial_CEO is assigned a value of one if the CEO has a functional background in finance and 0 otherwise. Given our theoretical inference, the coefficient on Financial_CEO is expected to be positive. The results of OLS and firm fixed effects reported in Models 3–4 of Table 7 confirm a positive influence of CEOs’ financial expertise on improving a company’s CSR ratings.
5.3.3 Excluding CEOs and CFOs from the TMT
There is no denying that the CEO is one of the most important characters within a firm. Moreover, CFOs are responsible for managing firms’ financial actions and investment affairs (Ma et al. 2019, 2020). Their duties generally include tracking cash flow and financial planning, such as CSR investments. Together, these findings demonstrate the essential role that CEOs/CFOs hold. However, the role of TMT cannot be underestimated, as the management literature has long pointed out the importance of the roles of all TMT members in day-to-day operations and in decision-making and business strategies (Ma et al. 2020). UET (Hambrick and Mason 1984; Hambrick 2007) argues that it is the positive interaction among all TMT members that creates a magnified impact on firm policies. Our current measure that includes all financial experts on the TMT may, to some extent, result in biased results because the positive relationship may be largely driven by CEOs or CFOs. Hence, it is not clear whether the financial experience of TMT members other than CEOs or CFOs plays a part in improving CSR.
To address this concern and validate the importance of the general TMT, we repeat our main analysis by using a redefined proxy for the TMT’s financial experience, that is, Financial_D_ExCEOCFO, which is a categorical variable assigned a value of one if at least one of the executives on the TMT (i.e., executive chairperson, vice president, general manager/managing director, executive/deputy general manager, and vice manager) has financial working experience and zero otherwise.13 Model 5 of Table 7 shows a coefficient on Financial_D_ExCEOCFO of 1.2012 (t = 3.1422), which is similar to that reported in Model 1 of Table 4 (coefficient = 1.5652, t = 4.6222), implying that other TMT members with financial experience, other than CEOs or CFOs with financial expertise, can improve a firm’s CSR.
5.3.4 Industry-adjusted CSR performance
Since CSR scores vary widely across different industries, we follow Li et al. (2021b) and use an industry-mean-adjusted CSR measure to assess a company’s CSR engagement compared with that of its peers within the same industry. This measure (CSR_Industry) is calculated as the deduction of the company’s CSR rating from the average rating for all listed companies operating in the same sector for a fiscal year.14 We replace CSRSCORE with CSR_Industry and re-estimate Eq. (1). The result in Model 6 of Table 7 is consistent with our baseline result.
5.3.5 Accounting for analyst coverage, audit quality, and TMT heterogeneity
Next, we test whether our key result is robust to controlling for additional variables. First, external monitoring can mitigate opportunistic managerial behaviour and thus improve social engagement and practices. For example, Dhaliwal et al. (2014) contend that CSR improves with scrutiny from financial analysts since extensive coverage exerts pressure on firms to disclose information, including information related to social issues. Brammer and Millington (2005) document that analyst coverage helps firms accumulate more reputational capital through CSR by improving organizational visibility, which in turn encourages firms to engage in more philanthropy. To rule out the impact of external monitoring, we account for the influence of analyst coverage on CSR performance by augmenting Eq. (1) with AnalystCoverage, which is measured as the natural logarithm of the number of financial analysts covering a focal company in a fiscal year. The result in Model 7 of Table 7 demonstrates a significantly positive estimate on Financial_D at the 1% level, supporting H1.
Next, prior studies have documented that better due diligence by large auditing firms plays an external monitoring role in facilitating a firm’s stakeholder-oriented activities (De Beelde and Tuybens 2015; Pucheta‐Martínez et al. 2019). Thus, we additionally include a categorical variable BigFour as a proxy for audit quality, which is assigned a value of one if the client company is audited by a Big Four auditor in a fiscal year, and zero otherwise. The result presented in Model 8 of Table 7 is robust.
Ma et al. (2020) and Shahab et al. (2020) find that other demographic characteristics (i.e., academic career experience, foreign experience, and age) of senior executives are positively associated with a company’s CSR performance or disclosure behaviours. Specifically, academic career experience may instil in senior executives a sense of social responsibility and shape the ways they run a firm and engage with stakeholders (Cho et al. 2017; Ma et al. 2019, 2020; Shahab et al. 2020). Lau et al. (2016) and Shahab et al. (2020) find that under the institutional background of China, top executives with foreign experience are likely to align institutional guidelines and regulations with the international standards of advanced markets, consequently driving better CSR implementation. Further, Shahab et al. (2020) argue that young executives may engage in empire building activities and focus more on short-term profit maximization, while older executives are more experienced and more likely to emphasize social capital enhancement and long-term objectives, such as socially responsible practices. To mitigate the concern that our results may be driven by the omitted executive team heterogeneity, we additionally control for TMT_AcademicRatio, measured as the proportion of executives with academic career experience on the TMT; TMT_ForeignRatio, measured as the percentage of senior executives with foreign work experience or study experience on the TMT; and TMT_Age, the natural logarithm of the average age of TMT members.15 We then re-run Eq. (1) and present the result in Model 9 of Table 7. It is worth noting that the coefficient on Financial_D remains significantly positive (coefficient = 1.5772, t = 4.7229), indicating that our key finding is robust to controlling for TMT heterogeneity.
5.3.6 Weighted least squares
As shown in Table 1, our sample is unevenly distributed across years, with 274 observations in 2009 and 446 observations in 2018. Given that each firm-year observation may not be treated equally, which drives potential concerns regarding the homogeneity of error terms, we follow Callan and Thomas (2011) to introduce the WLS regression to maximize the parameter estimation efficiency. The result is displayed in Model 10 of Table 7; the estimate on Financial_D is significantly positive at the 1% level, confirming the positive link between the TMT members’ financial experience and CSR as posited in H1.
5.3.7 Role of CSR rules by the regulatory authority in the link between financial experts and CSR
The Chinese government and regulatory authorities have widened socially responsible engagement in more recent years. For instance, the CSRC has enacted guidelines for Chinese listed firms regarding how to voluntarily disclose CSR information and environmental management in the annual report since 2013.16 Therefore, one might argue that the enhancement in firms’ CSR is not mainly driven by financial expert executives but attributable to the government’s motive for social reputation building. We then empirically examine the varying scope of the influence of financial expert executives on CSR performance by partitioning our sample around 2013 as a cut-off point and present the regression results in Models 11 and 12 of Table 7, respectively. Notably, Financial_D attracts a significantly positive coefficient in both models. Although the magnitude of the coefficient on Financial_D is larger during the period 2014–2018 (coefficient = 2.0416) than that in the period 2009–2013 (coefficient = 1.0012), the seemingly unrelated test shows no systematic difference between these estimates, which is evidenced by a p value of 0.1338 in the Chi test. These results suggest that such a change in CSR-related regulations does not bias our main finding because senior executives with financial working experience intrinsically initiate efforts in propelling CSR in both periods.
6 Endogeneity concerns
We cannot completely rule out endogeneity in our empirical context because socially responsible firms may attract senior executives with financial expertise. It is also possible that CSR and the appointment of financial expert managers are simultaneously influenced by omitted variables. Although the fixed-effect model in prior sections could help alleviate these concerns, this model cannot sufficiently rule out reverse causality concerns and capture unobserved characteristics because of sample selection bias. In this section, we carry out the PSM analysis, the Heckman correction model, and the dynamic GMM approach to further address these concerns. For brevity, we report the results based on the baseline model.
6.1 PSM approach
There may exist differences in observable firm fundamentals between firms with senior executives with a career background in finance and firms without, thereby driving biased results in our study. We follow Ma et al. (2019) to carry out a PSM analysis to address concerns that firms with executives with financial expertise are fundamentally different from firms without. Along with the caliper set at 0.01, we adopt a matching process with a ‘replacement’ procedure and use the nearest neighbour PSM technique to match each firm-year observation with senior executives who have financial experience with a firm-year observation without such executives based on a battery of factors used as control variables in Eq. (1). The industry and time dummies are included as matching criteria. We first estimate the average treatment effect on the treated (ATT). Specifically, the ATT is the average difference between the CSR performance of firms with senior executives with a career background in finance and their counterfactual peers’ CSR. Next, we re-estimate Eq. (1) based on the PSM sample to test the validity of H1. Table 8 reports the results of the PSM analysis. Panel A shows that firms run by senior executives with financial experience are associated with higher CSR ratings than firms headed by senior executives without such experience (ATT = 1.834, t = 3.05). After matching, our univariate results from the balancing test reported in Panel B show that the mean values of control variables (matching criteria) are not statistically different between firms with financial expert TMTs and those without, suggesting our sample is well matched. In Panel C, the estimate on Financial_D is positive and significant; this indicates that senior executives with financial experience have the potential to propel CSR practices.
Table 8
Propensity-score-matching (PSM) analysis
Panel A ATT in PSM | |||||
|---|---|---|---|---|---|
Estimated ATT with replacement | |||||
Outcome variable = CSRSCORE | ATT (T-Statistics) | Treatment group’s CSR level | Control group’s CSR level | No. of obs | Treated: control |
Financial_D | 1.834*** (3.05) | 39.9571 | 38.1231 | 5075 | 1243: 3832 |
Panel B Average firm characteristics in the treatment group and the control group | ||||
|---|---|---|---|---|
Variables in the matching process | Mean of the treatment group | Mean of the control group | Difference | Significance level (p value) |
SOE | 0.5326 | 0.5157 | 0.0169 | 0.399 |
Size | 23.2280 | 23.2550 | − 0.0270 | 0.662 |
Age | 2.3205 | 2.3106 | 0.0099 | 0.701 |
Lev | 0.5128 | 0.5223 | − 0.0095 | 0.222 |
ROA | 0.0499 | 0.0482 | 0.0018 | 0.432 |
Q | 1.7187 | 1.6372 | 0.0815 | 0.073 |
HERF10 | 0.1901 | 0.1837 | 0.0064 | 0.238 |
FCF | 0.0240 | 0.0218 | 0.0023 | 0.745 |
RD_Intensity | 0.0195 | 0.0203 | − 0.0008 | 0.546 |
BoardSize | 2.3065 | 2.3082 | − 0.0017 | 0.829 |
Duality | 0.1939 | 0.2076 | − 0.0137 | 0.395 |
Indep | 0.3795 | 0.3798 | − 0.0003 | 0.914 |
Managerial | 2.0738 | 2.0830 | − 0.0092 | 0.500 |
Female | 0.1511 | 0.1525 | − 0.0015 | 0.711 |
SSE | 0.6291 | 0.6155 | 0.0137 | 0.482 |
Panel C PSM sample | |
|---|---|
Outcome (Explained) variable = CSRSCORE | PSM sample with replacement |
Financial_D | 1.9366*** |
(3.7024) | |
SOE | − 0.0852 |
(− 0.1001) | |
Size | 4.5759*** |
(12.9400) | |
Age | − 1.1990** |
(− 2.1548) | |
Lev | − 3.1149 |
(− 1.3223) | |
ROA | − 13.2002** |
(− 2.4174) | |
Q | 0.2345 |
(0.9169) | |
HERF10 | 10.5217*** |
(3.6707) | |
FCF | 5.1502*** |
(3.1093) | |
RD_Intensity | 23.1471* |
(1.8496) | |
BoardSize | − 0.1797 |
(− 0.0854) | |
Duality | − 0.3639 |
(− 0.5018) | |
Indep | 1.3523 |
(0.2266) | |
Managerial | 3.0218*** |
(2.7781) | |
Female | − 1.7561 |
(− 0.5173) | |
SSE | − 0.0122 |
(− 0.0160) | |
_intercept | − 77.9790*** |
(− 9.1827) | |
Year effects | Yes |
Industry effects | Yes |
No. of obs | 5075 |
Adj. R-square | 0.520 |
6.2 Heckman two-stage procedure
Not all Chinese listed companies disclose sustainability reports every year; hence, this may introduce a sample selection bias to our study (Kong et al. 2020; Li et al. 2021b). We turn to the Heckman (1979) two-stage estimator to test whether our key finding is robust. First, we conduct a probit model, as specified in Eq. (2) to estimate the likelihood that companies issue sustainability reports and obtain social responsibility ratings from the RKS Runling rating system to calculate the inverse Mills ratio (also known as Lambda) from the residuals. Next, we include Lambda as a control variable in Eq. (3) as follows:where MandatoryPolicy is measured as a categorical variable set to one if a company is mandated to disclose a sustainability report in a fiscal year and 0 otherwise.17CSR_Presence is a categorical variable that is assigned a value of one if a company issues a sustainability report and is rated by the RKS Runling rating system in a fiscal year; otherwise, it is assigned a value of zero. MandatoryPolicy influences CSR at the company level but has no direct impact on the appointment of senior executives with financial experience (Chen et al. 2018; Kong et al. 2020).
$$CSR\_Presence_{i,t} = {\upalpha } + \beta_{1} MandatoryPolicy_{i,t - 1} + \beta_{2} Control_{i,t - 1} + Dummy_{year} + { }Dummy_{industry} + { }\varepsilon_{i,t}$$
(2)
$$CSRSCORE_{i,t} = {\upalpha } + \beta_{1} Financial\_D_{i,t - 1} + \beta_{2} Lambda_{i,t - 1} + \beta_{3} Control_{i,t - 1} + Dummy_{year} + { }Dummy_{industry} + { }\varepsilon_{i,t}$$
(3)
The regression results are presented in Table 9. In Model 1, the estimate on MandatoryPolicy is significantly positive (coefficient = 2.7400, t = 30.8129), meaning that companies that are subject to a mandatory CSR reporting policy tend to issue CSR reports and are highly likely to be rated by the RKS Runling rating agency. In Model 2, the coefficient on Financial_D is significantly positive (coefficient = 1.5175, t = 4.5699), strongly supporting H1.
Table 9
Heckman selection model
Dependent variable = | CSR_Presence | CSRSCORE |
|---|---|---|
First-stage regression | Second-stage regression | |
(1) | (2) | |
MandatoryPolicy | 2.7400*** | |
(30.8129) | ||
Financial_D | 1.5175*** | |
(4.5699) | ||
Lambda | 3.2443*** | |
(6.6069) | ||
SOE | 0.1588*** | 1.5090*** |
(5.2541) | (4.0225) | |
Size | 0.4679*** | 4.9023*** |
(28.3980) | (21.3414) | |
Age | 0.0338 | − 0.8268*** |
(1.5616) | (− 2.7678) | |
Lev | − 0.7034*** | − 6.1065*** |
(− 9.0951) | (− 6.1879) | |
ROA | 0.9700*** | − 3.4696 |
(4.0656) | (− 1.3222) | |
Q | 0.0646*** | 0.5714*** |
(6.0793) | (4.1572) | |
HERF10 | 0.0857 | 3.9504*** |
(0.7643) | (2.7993) | |
FCF | 0.1590** | 2.9256*** |
(2.4056) | (3.9028) | |
RD_Intensity | 0.6128 | 13.0847** |
(1.4112) | (2.3395) | |
BoardSize | 0.1562* | 2.2845** |
(1.8689) | (2.4670) | |
Duality | − 0.0414 | − 1.1287*** |
(− 1.3742) | (− 3.1005) | |
Indep | 0.7467*** | 0.3363 |
(2.9199) | (0.1282) | |
Managerial | 0.2299*** | 3.7122*** |
(5.3944) | (7.7168) | |
Female | − 0.2891** | 2.9386* |
(− 2.2986) | (1.8797) | |
SSE | 0.1287*** | − 0.1416 |
(4.6610) | (− 0.3587) | |
_intercept | − 12.4177*** | − 96.0718*** |
(− 29.3300) | (− 16.0516) | |
Year effects | Yes | Yes |
Industry effects | Yes | Yes |
No. of obs | 20,688 | |
Selected | 5158 | |
Censored obs | 15,530 | |
Wald test of independent equations Chi2 (p-value) | 40.10***(0.0000) | |
ρ | 0.3582*** | |
σ | 2.2451*** |
6.3 Dynamic GMM approach
The dynamic nature of our variables, according to which the current values of the independent variables are a function of past values of the explained variable, may drive endogeneity issues in our empirical setting. Following Blundell and Bond (1998), Wintoki et al. (2012), and Cui et al. (2018), we include the one-year lagged CSR ratings as an independent variable in Eq. (1) to implement the dynamic GMM estimation. We employ the Arellano–Bond system GMM method, which is a procedure involving two models. One is the dynamic regression that is transformed into a first-differenced mode. The other is the dynamic regression that is transformed into a level form (Arellano and Bover 1995; Blundell and Bond 1998).18
The result from the dynamic panel system GMM approach presented in Model 1 of Table 10 shows that the estimate on Financial_D is significantly positive. The results of the Arellano-Bond test—AR(1) and AR (2) for first-order and second-order serial correlation in the first-differenced residuals—support the null hypothesis of no serial correlation because AR(1) is serially correlated, but AR(2) is uncorrelated. To validate the use of our instruments, we adopt the Hansen test of overidentification and find a p value of 0.271, suggesting that the instruments are uncorrelated with the error terms.
Table 10
Dynamic GMM estimation
Dependent variable = CSRSCORE | Dynamic panel-data estimation |
|---|---|
System GMM | |
(1) | |
Financial_D | 0.6580*** |
(5.6242) | |
LAG_CSRSCORE | 0.7706*** |
(103.1311) | |
SOE | 0.5953*** |
(2.8431) | |
Size | 1.0060*** |
(10.4593) | |
Age | − 0.6270*** |
(− 4.9040) | |
Lev | 0.4680 |
(0.8432) | |
ROA | 0.4463 |
(0.4132) | |
Q | 0.2069*** |
(4.1717) | |
HERF10 | 2.1314** |
(2.3719) | |
FCF | 0.3483* |
(1.8110) | |
RD_Intensity | − 0.2499 |
(− 0.0947) | |
BoardSize | 2.0132*** |
(4.1978) | |
Duality | − 0.5162*** |
(− 3.7830) | |
Indep | 5.2272*** |
(4.3418) | |
Managerial | 0.5162** |
(2.3100) | |
Female | 0.1206 |
(0.1354) | |
SSE | − 0.5298*** |
(− 2.9779) | |
Year effects | Yes |
Industry effects | Yes |
AR(1) test (p value) | 0.000 |
AR(2) test (p value) | 0.321 |
Standard errors | Corrected |
Sargan test over-identification (p value) | 0.000 |
Hansen test over-identification (p value) | 0.271 |
No. of obs | 4226 |
No. of Firms | 770 |
In summary, our findings still hold when we employ the PSM technique, Heckman selection analysis, and the dynamic system GMM approach.19
7 Further analysis
As an extension, we empirically explore whether the enhancement in a firm’s philanthropic activities driven by financial expert executives may influence business outcomes. Indeed, stakeholder-oriented engagements have significant effects on business strategies (Boubakri et al. 2016; Ferrell et al. 2016; Liang and Renneboog 2017a; Lins et al. 2017; Li and Wang 2021); the positive view of CSR suggests that if a company behaves socially responsibly in its operation, actively addresses stakeholder concerns and develops relationships of mutual trust with stakeholders, then the company will be financially rewarded. In markets where investors have relatively high social consciousness and awareness, CSR activities can improve brand value and corporate reputation, which eventually leads to stronger firm performance (Li et al. 2021b). Hence, we specify Eq. (4) to examine whether companies are rewarded by the market because of the superior CSR performance driven by senior executives with financial expertise by introducing MTB as a dependent variable:where MTB, a proxy for firm value, is measured as the market value of equity divided by the book value of equity. Financial_D × CSRSCORE captures the incremental effects of CSR through the presence of senior executives with financial experience on company performance.
$$\begin{aligned} MTB_{i,t} = & \alpha + \beta_{1} Financial\_D_{i,t - 1} + \beta_{2} CSRSCORE_{i,t - 1} + \beta_{3 } Financial\_D_{i,t - 1} \times CSRSCORE_{i,t - 1} \\ & + \beta_{4} Control_{i,t - 1} + Dummy_{year} + Dummy_{industry} + \varepsilon_{i,t} , \\ \end{aligned}$$
(4)
Models 1 and 2 of Table 11 display the regression results. We find that the estimate on Financial_D × CSRSCORE in Model 2 is significantly positive at the 1% level, implying that the CSR enhancement attributed to senior executives with financial experience results in higher firm value. Moreover, consistent with Boubakri et al. (2016), firms with higher CSR performance indeed exhibit higher investor valuations, as evidenced by the significantly positive coefficient on CSRSCORE in both models.
Table 11
Incremental effect of CSR on firm value
Dependent variable = | MTB | MTB |
|---|---|---|
(1) | (2) | |
Financial_D | 0.1499** | − 0.3726 |
(2.0077) | (− 1.6366) | |
CSRSCORE | 0.0128*** | 0.0086*** |
(4.9353) | (3.0613) | |
Financial_D × CSRSCORE | 0.0135*** | |
(2.6196) | ||
SOE | 0.1627* | 0.1658* |
(1.8480) | (1.8761) | |
Size | − 0.7958*** | − 0.7986*** |
(− 17.8648) | (− 17.9244) | |
Age | 0.2725*** | 0.2834*** |
(3.6667) | (3.8009) | |
Lev | 2.2256*** | 2.2409*** |
(6.4790) | (6.5378) | |
ROA | 0.2998 | 0.3098 |
(0.2210) | (0.2285) | |
HERF10 | 1.3995*** | 1.3968*** |
(4.2678) | (4.2582) | |
BoardSize | − 0.0912 | − 0.0720 |
(− 0.4928) | (− 0.3882) | |
Duality | 0.0454 | 0.0499 |
(0.4954) | (0.5461) | |
Indep | 2.1335*** | 2.1258*** |
(3.1564) | (3.1473) | |
_intercept | 19.4528*** | 19.6150*** |
(20.7528) | (20.9106) | |
Year effects | Yes | Yes |
Industry effects | Yes | Yes |
No. of obs | 4526 | 4526 |
Adj. R-square | 0.279 | 0.280 |
8 Conclusions
This study, conducted in the context of Chinese listed firms, investigates the influence of TMT members’ financial working experience on CSR performance and reporting behaviours. Our analyses are mainly informed by the predictions of UET (Hambrick and Mason 1984; Hambrick 2007). Our findings demonstrate that senior executives with a career background in finance play a vital and integral role in the successful implementation of institutionally oriented social and environmental practices. In particular, firms with senior executives with financial expertise are more likely to be associated with CSR performance improvement and have longer sustainability reports than firms without. In addition, we confirm that the significant improvement in future CSR is mainly attributed to senior executives who have financial work experience in regulatory authorities. Next, by investigating the role of senior executives’ latitude of action in the relation between the TMT members’ financial experience and CSR, we reveal that this positive influence is more salient in non-SOE entities or more prominent in cash-abundant firms. Our result still holds when we carry out a series of robustness checks and endogeneity tests. Our further analysis illustrates that the CSR improvement attributed to financial experts on the TMT results in higher firm value.
This study is subject to common limitations. First of all, there may be other firm characteristics that are not investigated in this study but are affected by the presence of financial experts on the TMT. To provide convincing evidence on the mechanisms that link financial experience to CSR performance, it will be necessary to collect more consistent information on firm policies. We next provide some suggestions for future research. Conditional on the results of the current study, it would be interesting to see how the increase in CSR ratings driven by financial experts might affect investment efficiency. As indicated by Cook et al. (2019), socially responsible companies may invest more efficiently, and the incremental effects of CSR may lead to higher capital allocation efficiency.
This study sheds light on the influence of senior executives’ financial career experience on CSR strategies. Our findings have valuable implications for policymakers, corporate managers, and professionals in China and other countries, especially those with similar institutional backgrounds (i.e., other emerging countries in Asia with inferior social preference and governance quality). Instead of appointing CSR experts whom the executive labour markets in emerging countries may have a shortage of, firms could consider appointing top executives with financial working experience since these executives may better incorporate ethical values and social norms into their business strategies, potentially generating capital market benefits for various stakeholders. In particular, the experience in regulatory-oriented financial institutions can provide an essential avenue for firms in the emerging markets to engage more in CSR-relevant activities.
Declarations
Conflict of interest
There are no conflicts of interest to declare.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.