Analysis of the Motivation behind Corporate Social Responsibility Based on the csQCA Approach

Abstract

Contemporary society is paying increasing attention to corporate social responsibility (CSR), the motivation for enterprises to perform social responsibility, and ways to encourage enterprises to actively take action towards social responsibility have become a focus among scholars. Following previous studies, this paper systematically summarises the motivation behind CSR in four aspects: (1) senior management and environmental impact; (2) the enterprise level; (3) external influences on the enterprise; and (4) ideological and political construction. This paper adopts a qualitative comparative analysis (QCA) method to examine these four aspects, including specific variables of “Chinese Communist Party member on the board”, “women on the senior management team”, “CSR vision of the enterprise”, “overseas background among leaders”, and “enterprise profitable in the current year”. This paper finds that: (1) As a necessary condition, “enterprise profitable in the current year” plays a decisive role in the fulfilment of social responsibility. (2) The existence of women on the senior management team plays a vital role in CSR. (3) “Chinese Communist Party member on the board”, “CSR vision of the enterprise”, and “overseas background among leaders” complement each other.

1. Introduction

The socialist market economy with Chinese characteristics emphasises the leadership of the Communist Party of China and provides a certain degree of guidance and restraint on the relevant norms of conduct for enterprises. Modern enterprises in capitalist countries and some Chinese enterprises are obviously under the control and constraints of the capital market, but the shareholders are too diversified, remote, and ignorant of the situation they face. Simultaneously, the daily lives of people who do not hold corporate shares are affected by the economic activities of the enterprise, but they cannot effectively influence the relevant decisions of the company. Therefore, through a series of means and mechanisms, including government behaviour, substantive guidance and effective constraints on corporate behaviour have become very important research topics, and research on corporate social responsibility (CSR) is one of the most important areas of study today.

After an enterprise develops to a certain extent, fulfilling its social responsibility is a necessary way to ensure the sustainable and stable development of the enterprise. With the rapid growth of China’s economy, scientific and technological prowess, and improvement in people’s living standards, the government and people from all walks of life are paying increasing attention to the fulfilment of CSR. For example, the Shanghai Stock Exchange issued a series of notices and guidelines in 2008 aimed at guiding listed companies to fulfil their social responsibilities [1]. As time has passed, the meaning of CSR has become increasingly broad, and the narrow sense of “public welfare undertakings”, “charity”, “donations”, “poverty”, and “environmental protection” gradually evolved to include stakeholders and issues such as “shareholders, suppliers, creditors, and customers”, “product quality”, “staff”, and “tax”, as mentioned above, making it in many ways a “hodgepodge” concept. As stated in the guidelines on the social responsibility of listed companies issued by the Shenzhen Stock Exchange, CSR mainly refers to the responsibilities assumed for the benefit of society, the environment, employees, suppliers, customers, and other stakeholders [2].

Since 2005, the relevant state departments in China have successively issued relevant documents to guide enterprises to fulfil their social responsibilities [3], although the action by enterprises on this front is still lacking in terms of food and drug safety, ecological damage, product damage, environmental pollution, and other scandals. These problems seriously harm people’s health and are not conducive to the construction of a harmonious society today. Ultimately, the fundamental reason for this lack of successful CSR action is that the purpose of an enterprise is to maximise profits. According to the “rational economic man hypothesis” [4], enterprises fulfil their social responsibilities in order to gain substantial benefits. The performance of CSR requires enterprises to spend many human, material, and financial resources, and the benefits of CSR come at the cost of enterprises’ profits.

The qualitative comparative analysis method (QCA) is a new method that has gained widespread attention in social science research [5]. It has unique advantages in studying the effects of multiple variables on the results [6]. Therefore, the clear-set qualitative comparative analysis (csQCA) method was selected as the research method for this study. Five dimensions, “Chinese Communist Party member on the board”, “women on the senior management team”, “CSR vision of the enterprise”, “overseas background among leaders”, and “enterprise profitable in the current year”, are used to explore the motivation behind social responsibility.

Previous literature has investigated the motivations of corporations conducting CSR practices, yet research on how these factors function together is still lacking. Therefore, the adoption of the csQCA approach allows us to investigate the group effect of contributing factors to CSR in a more comprehensive way, which appears to be necessary. This study consolidates previous literature on the motivations of CSR performance to provide a more comprehensive reference for subsequent scholars studying the motivations of CSR. In addition, based on China’s unique national context, this paper considers a unique perspective in exploring CSR fulfilment in this country, namely whether corporate leaders are CPC members, enriching the findings of studying CSR from a political and ideological perspective.

The remaining sections of this article are arranged as follows: Section 2 discusses the definition of CSR through a literature review and reviews previous analyses of the motivation for implementing CSR behaviour; Section 3 describes the data selection process, the feasibility of using csQCA, and the advantages of this method over traditional regression and case analysis methods; Section 4 describes the data sources and the specific meaning and selection of each variable; Section 5 covers the specific analysis of the data collection and the presentation of the results; Section 6 tests the stability of conclusions by changing consistency thresholds and deleting certain cases; Section 7 provides the study’s conclusions, theoretical contributions, and limitations, as well as the outlook for future studies.

2. Literature Review

As early as 1979, Carroll [7] proposed that CSR should cover four areas, namely, economy, law, morality, and charity, in which economic and legal levels are fundamental and morality is valued, but charity is less important. Twenty years later, British scholars put forward the triple bottom-line theory of CSR, which includes three aspects: society, environment, and economy. The triple-bottom-line theory not only proposes requirements for the economic, social, and environmental performance of enterprises but also examines the values, problems, and processes of enterprises in terms of their social responsibility. This requires enterprises to pay attention to stakeholders and social expectations in their daily activities and to achieve a balance between the pursuit of development and environmental damage so as to create environmental, social, and economic value at the same time. McWilliams and Siegel [8] proposed that in terms of corporate interests and laws, any responsibility beyond daily business activities is CSR, which includes multiple levels, such as the environmental level, consumer level, employee level, and social donation level [9].

CSR is multidimensional, covering business performance, ethics, politics, the economy, and the environment, meaning that enterprises can fulfil their social responsibility in many ways, such as by creating social wealth, abiding by law and order, paying attention to staff safety and occupational health, ensuring product quality, and caring for underprivileged groups [10]. In recent years, most scholars’ research on CSR has focused on the motivation behind CSR, quantification of CSR, and effects of CSR, among which, the motivation of corporate social responsibility has been the subject of the most research.

Campbell et al. [11] explore the motivation of enterprises to fulfil their social responsibilities from three aspects, namely, economic incentives, legitimacy motivation, and altruism. The economic incentive perspective suggests that CSR is fulfilled out of a desire to gain high visibility and brand image and to increase customer loyalty [12,13]. The view of legitimacy motivation is that enterprises fulfil their social responsibilities due to external environmental factors, especially the pressure of institutional ethics, standards, and expectations for obtaining the resources needed to sustain and conform to society [14,15]. The altruistic view favours the explanation of considering society and nature as the most important stakeholders, which drives entrepreneurs to make decisions that are not entirely tied to business profits [16].

Other literature also identify multiple contributing factors affecting CSR [17,18], particularly the industrial effect. Since industries have their own specific rules, norms, and trends, the motivation for CSR varies depending on the industry in which the company is located [19]. The energy industry, for example, has a particular responsibility for CSR since it is crucial for sustainable development. In addition to pressure from laws, regulations, and social groups, many companies in the energy sector are proactive in incorporating CSR into their strategies and gaining stakeholder support for their business operations through social responsibility activities [20]. Moreover, Cai et al. [21] found that for companies in controversial industries, even though their products may be harmful, these companies still enhanced their corporate value by engaging in CSR, while Thorisdottir and Johannsdottir [22] discussed the driving factors for companies in the fashion industry to engage in CSR, including brand equity, culture, activism, and human rights. Besides, in the hospitality sector, CSR has been found to drive employee creativity significantly, which could better position a hotel in the face of competition [23].

Based on the above views, this paper divides the motivation of enterprises to fulfil their social responsibility into four areas: senior management and environmental impact (the human level), the enterprise level, external influences on the enterprise, and ideological and political construction. The area of senior management and environmental impact mainly includes altruistic aspects, such as personal characteristics, culture, gratitude, and the motivation of individual executive team members; the enterprise level covers business operations and the potential benefits of conducting CSR; and the external influences include public opinion, regulatory environment, economic and legal environment, and government intervention. Finally, based on China’s national conditions, the influence of the CPC and Party organisations on CSR is studied.

2.1. Senior Management and Environmental Impact

As mentioned above, a person’s values determine his or her behaviour. CSR is part of the behaviour of the enterprise, and the form it takes is determined by senior management. Therefore, the values and characteristics of senior management profoundly affect the behaviour of enterprises in their CSR activity. As early as 2010, some scholars, based on the “upper echelon theory”, explored the motivation behind CSR from the perspective of the level of corporate executives [24].

2.1.1. Personal Characteristics of Senior Executives

Some of the personal characteristics of senior executives shape their personal values, and the aforementioned literature also focuses on the long-term influence of these personal values. For example, female CEOs have a significantly enhancing effect on CSR performance [25] and CEOs with overseas experience will rate health care, which is one of the CSR-related areas, as more important on average for companies to address than those without overseas experience [26]. Compared with CEOs who do not raise daughters, CEOs who raise daughters have a higher altruistic preference, which is reflected in the better performance of corporate social responsibilities [27]. In practice, executives with “career concerns” are more inclined to perform CSR [28], and the higher their confidence is, the lower the level of CSR found among their peers [29].

2.1.2. Senior Management Team and Corporate Governance Structure

In addition to the characteristics of senior executives themselves, the corporate governance structure related to senior executives and directors also plays a decisive role in the decision of enterprises to engage in CSR. Hu et al. [30] selected education level, overseas background, government background, academic background, and financial background to analyse the impact of executive team career development potential on corporate social responsibility from multiple perspectives and concluded that CSR performance is significantly correlated with TMT education level and overseas background. On the other hand, the government background, educational background, and financial background of top managers are negatively correlated with CSR performance. From a dynamic perspective, Huang et al. [31] found that top management team stability has a positive effect on sustainable CSR because the top management team improves team behaviour integration through team cohesion, and executives can focus on long-term CSR strategies. El-Bassiouny and Letmathe [32] examined how internal corporate governance influences CSR and proposed: the effectiveness of the board of directors, including having independent directors and board committees, and whether CEO duality exists, would positively affect CSR practices.

2.1.3. Cultural Influence

Traditional Chinese culture is essentially a culture of kindness, so gratitude is naturally rooted in the depths of people’s minds. Han et al. [33] examined the childhood famine experiences of senior executives and found that when a CEO has experienced famine, his or her company makes more philanthropic donations. Relatedly, Liu et al. [34] studied the decision problems of a retailer-dominated supply chain considering CSR under government subsidy based on game analysis technology and found a promoting role of government subsidies on CSR undertakings. These two studies have proven the positive impact of gratitude factors on CSR.

The cultural environment plays a fundamental guiding role in the formation of people’s values and behavioural patterns. The daily behaviour of senior executives is influenced by the surrounding values, including Confucian culture, business gang culture, and the external economic environment of Chinese society. For example, based on the data of China’s A-share listed companies from 2010 to 2019, Kong et al. [35] found that the Confucian traditional culture behind the imperial examination system can encourage firms to improve their CSR level. With traditional Confucian culture at its core, merchant group culture all over the country values mutual help, “integrity and benevolence”, so the influence of merchant group culture on CSR cannot be ignored. Although there are various business groups all over the country, most of the cultural subjects of merchant groups emphasise the trust of business people in the process and the integrity involved in signing and carrying out contracts. It is because of the moral improvement of merchant group culture that naturally enhances the positive relationship between the strength of merchant group culture and CSR performance [36].

Since China’s reform and opening-up policy began in 1978, an increasing number of multinational companies have entered the Chinese market, and they have also introduced foreign corporate culture to China. As early as 2009, Tang and Li [37] used content analysis of the websites of these companies to explore how leading Chinese domestic and multinational companies operating in China communicate their CSR principles and practices to stakeholders in China. They found that despite the trend towards convergence, Chinese and multinational companies still demonstrate their CSR principles and practices differently. Multinational firms are more likely to discuss CSR on their corporate websites compared to domestic firms. In terms of the “content” of CSR communication, product safety is discussed more often as part of multinational firms’ CSR practices. In terms of the “way” of CSR communication, multinational firms are more likely to report having a partnership with the government as a sign of their growing legitimacy and prestige in Chinese society, while such legitimacy is not a major issue for domestic companies. In addition, global firms are more likely to institutionalise their CSR commitments through formal corporate policies than domestic companies, which suggests that global companies are more inclined to adopt CSR as an inherent part of their business practices, while domestic companies tend to regard CSR as some of the philanthropic activities they additionally engage in.

In summary, the values and characteristics of senior managers, as key decision-makers in determining corporate behaviour, are influenced by the surrounding culture and profoundly influence whether and how the company engages in CSR activities.

2.2. Enterprise Level

Just as the values and characteristics of senior executives affect how enterprises fulfil their social responsibilities, the nature and characteristics of enterprises also affect how decisions about economic activities are made, including engaging in CSR activities. These enterprise characteristics specifically include the ownership structure of enterprises, such as whether they are state-owned enterprises or family-owned enterprises; the financial situation of the enterprise, namely, whether there are sufficient resources to implement CSR activities; and whether there are substantial benefits to doing so, such as economic incentives.

2.2.1. The Ownership Structure of the Enterprise

Carroll [38] proposed that organisational stakeholders affect an enterprise’s CSR activities. Likewise, given information asymmetry, Hickman [39] found that public-traded enterprises should be discussed separately from privately-held firms when exploring CSR reporting. Previous literature shows that companies owned by the government, institutional investors, and foreign investors have a more positive impact on CSR disclosure under conditions of the high level of board independence [40], and the social responsibility overflow of state-owned enterprises can have a “leading model effect” for other enterprises and society [41].

Additionally, the influence of being family-owned on an enterprise’s CSR activity has also aroused wide interest from scholars. For example, Gallo [42] examined the impact of the family-owned structure on CSR performance. Berrone et al. [43] found that family-controlled public firms protect their socioemotional wealth through better environmental performance than non-family-controlled public firms, especially at the local level. Later, Sahasranamam et al. [44] combined this topic with a consideration of India’s national conditions and concluded that family ownership is beneficial for community-related CSR, referring to agency theory and the sociological perspectives of institutions. However, Block and Wagner [45] used Bayesian regressions on a dataset of large US firms to show that family ownership is negatively associated with community-related CSR performance. On the other hand, family ownership turned out to be positively associated with the diversity-, employee-, environment-, and product-related aspects of CSR. Further research has found that two dimensions of family involvement, namely family ownership and time spent in family control, have significant positive effects on a firm’s charitable donations [46].

Thus, enterprise ownership characteristics affect the fulfilment of CSR. In addition, there is a study from a “principal–principal perspective”, which states that the objectives of a company are not aligned between groups of shareholders, especially between controlling and minority shareholders, which explains how controlling shareholders affect corporate donation behaviour [47]. Specifically, there is a significant negative correlation between corporate donations (both the likelihood and the amount of charity giving) and the equity shares held by the largest shareholder, indicating that the controlling shareholder has opportunism in guiding corporate charitable donations.

2.2.2. Financial Status of Enterprises

Enterprises often need to invest certain resources to fulfil their social responsibilities. According to the capital supply hypothesis, the more capital a company can provide, the more social responsibilities it can fulfil [48]. However, such relationships are not absolute. Otero-González et al. [49] conducted an analysis based on the managerial opportunism hypothesis and concluded contrary to the above, finding a negative relationship between economic performance and more CSR activities implemented. The relationship between CSR and financial performance is complex [50] and needs to be considered across multiple dimensions. Margolis and Walsh [51] analysed 109 studies on predicting financial performance with CSR and 54 found positive predictions, 7 found negative predictions, 28 were unrelated, and 20 found inconclusive results. Further research revealed that there are cross-temporal effects between CSR and corporate financial performance. McGuire et al. [52] found that a firm’s prior performance is closely related to corporate social responsibility, with profitability measures overall showing positive relationships and risk measures mostly showing negative relationships. Awaysheh et al. [53] further studied the endogenous interaction between a firm’s financial performance and its CSR.

2.2.3. Substantial Benefits Such as Economic Incentives

According to the rational economic man hypothesis, rational people are motivated by the prospect of substantial gain. Therefore, when undertaking a series of CSR behaviours, such as making donations, an enterprise invests considerable human, material, and financial resources; enterprises consider the pursuit of delayed or “invisible interests”, the most significant of which is the reputation and social status of the enterprise. Enterprises engaged in CSR activities, such as announcing environmentally friendly initiatives, will be recognised for this through market forces, so their stock price rises [54], which effectively enhances investors’ investment intentions [55]. In addition, the fulfilment of corporate social responsibilities can also affect consumers’ purchase decisions, such as encouraging consumers to purchase their own products [56].

Regarding “invisible interests”, Lin et al. [57] believe that enterprises fulfil their responsibility in order to secure political connections, as such connections can bring enterprises many potential benefits, including higher levels of government subsidies or a greater likelihood of receiving future government subsidies, as well as tax benefits and government financing [58]. In the United Kingdom, CSR is considered a tool used by neoliberal-leaning political and economic elites to reduce government intervention [59].

Prior research also considered companies’ CSR activities to be a compensation mechanism. For example, Chen et al. [60] studied corporate charitable contributions made by US firms and found that those enterprises with low product safety evaluation and poor environmental evaluation are more willing to spend money on charitable donations, which means that they can cover up the mediocrity or defects of products or divert the public’s attention from the environmental problems of enterprises. Gong et al. [61] found that previous CSR activity has a “reputation insurance” effect for companies, which helps them reduce the impact of punishment on their reputation.

In summary, the characteristics of enterprises, such as their ownership structure, financial status, and “invisible economic interests”, are all related to whether an enterprise engages in CSR activities. Together, these constitute the main framework of motivation at the enterprise level, that is, why enterprises are willing to spend a great deal of human, material, and financial resources on seemingly unprofitable CSR activities.

2.3. External Influences on the Enterprise

2.3.1. Public Opinion of the Enterprise

Enterprise activities in commercial society, including CSR performance, are affected, on the one hand, by the internal elements of the enterprise, such as the enterprise’s executives and characteristics, and, on the other hand, by the external environment, including media supervision, public opinion, the institutional environment, and the market environment (competition intensity) [62]. For example, the media is an important external driver of CSR [63,64], public opinion affects CSR behaviour [65], and the use of CSR claims could provide an effective tool to counter the negative publicity of a crisis for certain companies [66].

2.3.2. The Economic and Legal Environment of the Enterprise

The local economic and legal environment of the enterprise also impact CSR performance. The aforementioned study found that the higher the degree of marketisation, the more sound the legal environment, and the more transparent the culture of information disclosure, the weaker the effect of certain factors on CSR performance will be. For example, Lu et al. [67] found that in areas with lower marketisation due to an underdeveloped institutional environment, enterprises have more motivation to replace the two indicators of female executives and CSR to gain legitimacy and fill the institutional gap, so the negative correlation between them is stronger. Gong et al. [61], in a study of the role of enterprises’ “reputation insurance”, found that the marketisation degree, legal environment, and culture of information transparency have obvious effects on “reputation insurance”; namely, the CSR has a weak “reputation insurance” effect in an environment with a low degree of marketisation, weak legal norms, and a weak culture of information transparency.

2.3.3. Government Intervention in Enterprises

Under the socialist market economy system with Chinese characteristics, the government’s degree of intervention in enterprises has also impacted CSR performance. For example, Zhang [68] found that the level of government intervention significantly affects corporate environmental performance; that is, the environmental performance of firms located in civilised cities is higher than that of firms located in uncivilised cities. Albareda et al. [69] also pointed out that government policy is key to encouraging increased awareness of CSR, summarising the changing role of government in promoting CSR.

The above discussion shows that the aforementioned literature has studied CSR performance from the perspectives of the characteristics of senior executives, corporate governance structure, enterprises’ financial situations, and the market environment of enterprises. In such research, it is difficult to ignore the impact of national conditions on CSR. Especially as a socialist country, China is fundamentally different from Western countries in terms of its systems, and researchers should not ignore the impact of national conditions on CSR. Therefore, this article specifically examines the effect of the Communist Party of China and Party organisations and a series of ideological and political constructions on enterprises’ CSR activities.

2.4. Ideological and Political Context

In 2021, the total revenue of state-owned enterprises in China was 75.55 trillion yuan, accounting for 66.3% of the country’s GDP. As a socialist country, China has a unique system that affects all aspects of people’s daily lives. Therefore, this paper considers ideological and political construction as distinct dimensions and divides them into party membership and political participation (human level) and Party organisation (system level).

The aforementioned research suggests that state-owned enterprises are more willing to engage in CSR practices than private companies [40]. In the context of the socialist market economy with Chinese characteristics, state-owned enterprises are owned by the people, and thus, the means of production are jointly owned by all people. This means that state-owned enterprises represent the interests of the masses, so they are more willing to advance the interests of the people by engaging in CSR activities. Thus, it can be seen that the ideological and moral context profoundly affects the behaviour of individuals and enterprises through internal mechanisms and then encourages enterprises to engage in CSR.

2.4.1. Party Membership and Political Participation

In the aforementioned literature, the research on the ideological and moral construction of enterprises mainly involves the Party membership of entrepreneurs and senior executives, as well as the mechanisms related to Party organisations. For example, regardless of the party status of private entrepreneurs [70] or senior executives [71], party membership can obviously promote engagement in CSR, such as employee-related CSR. Besides, Gustafsson et al. [72] found that philanthropy behaviour of Chinese private enterprises is positively correlated with owners’ political participation, such as joining all levels of people’s congresses and Chinese People’s Political Consultative Conference. It can be seen that when the senior executives and other key personnel of the enterprise actively participate in the ideological and moral construction, it will subtly affect their daily life, making them more inclined to follow the guidance of the Party and urging the enterprise to fulfil its social responsibilities.

2.4.2. Party Organisations

In recent years, researchers have begun to study the impact of Party organisations on CSR in terms of charitable donations, environmental protection, and improving labour rights. Motivated by the reciprocal logic, private entrepreneurs would respond positively to participating in environmental corporate socially responsible investments by reviewing the “spillovers” created by Communist Party of China branches and anticipating their future contributions [73]. The establishment of party organisations also helps enterprises make more charitable donations [72,74]. At the same time, having a Chinese Communist Party branch also brings a series of positive effects related to CSR, such as signing labour contracts to provide stable jobs to their employees [75], increasing the coverage of social insurance [76,77], and other employee protection behaviours.

2.5. Theory Framework and Hypothesis Development

Based on the above literature review, we can see that CSR decisions are influenced by a variety of factors. Contingency theory advocates that there is no best way to organise or lead a company. Rather, in the operation of an organisation, company leaders can only be flexible in choosing the best possible option to suit the needs and circumstances of a given period, based on specific internal and external changes [78].

In this paper, we consider CSR-related decisions as a process in which management interacts with factors at the level of the enterprises and the external environment, especially the political context. This study uses a csQCA approach to explore the complex dependencies between these interrelated conditions and equivalence pathways, and a theoretical framework based on contingency theory is constructed as shown in Figure 1. This theoretical framework incorporates multiple conceptual dimensions that enable us to examine how a firm’s external social environment and internal organisational factors interact with management perceptions to influence the choice of a firm’s CSR strategy.

In the book Comparative Methods, Ragin [79] points out that QCA consists of two main assumptions: firstly, concurrent causality, which refers to multiple concurrent causal relationships consisting of interdependent and different combinations of traction conditions; and secondly, equivalence, which means that there are multiple combinations that produce the same particular outcome. Based on the csQCA approach, we explore the grouping pathways that influence the social responsibility performance of Chinese-listed enterprises. The research question of this study is: based on a systematic review of previous literature, what are the key grouping/combination of factors contributing to CSR practice.

3. Research Methods

3.1. QCA Methods

The traditional regression analysis method is based on three considerations: independent variables, causal symmetry, and linear relationships. This analysis method involves controlling for certain variables to ensure that the influence of the independent variable on the dependent variable is constant, that is, ensuring the “net effect”. However, in cases where there are mutual influences and mutual dependence between independent variables, the traditional regression analysis method has limitations, which is where the qualitative analysis and comparison method can be useful.

In the 1980s, sociologist Ragin [79] adopted QCA to solve issues related to the interaction of independent variables and common effects on the results. From the perspective of configuration, the QCA method can be used to study the influence of multiple variables on the results. The qualitative comparative analysis method believes that each case (phenomenon) comprises multiple causal conditions (i.e., independent variables). Through the QCA method, the complex causal relationships in the configuration of the conditions can be identified.

The QCA method is based on Boolean algebra, which is binary into the Boolean algebra, wherein 0 and 1 form the basic operational logic; 0 indicates the absence of the variable, and 1 indicates the presence of the variable. In addition, different symbols also represent different meanings; for example, “*” represents “and”), meaning that two conditions are required to cause the focal result; “+” means “or”, meaning that any one of the conditions can cause the focal result; and “~” means “non”, meaning that the conditions are the opposite of those in the previous formula. Thus, configuration 1 (Party member * ~ overseas background * profitable) means that the executive board includes a Party member, but none of the executives has an overseas background, and the enterprise was profitable in the current year.

3.2. Advantages of QCA Methods

The qualitative comparative analysis method changes from investigating linear relationships, as in traditional linear regression, to considering the configuration of multiple relationships, that is, from analysing the influence of individual variables on the results to analysing the influence of the specific configuration of multiple simultaneous variables on the results.

Unlike the causal symmetry of traditional linear regression, QCA has the characteristics of causal asymmetry; that is, in the traditional perspective, the larger independent variable A is, the larger dependent variable X is, but with causal asymmetry, a larger A does not necessarily mean a larger X, and B may also influence X. The influence of a given configuration based on QCA precisely considers the effect of both A and B on X. Because of causal asymmetry, the QCA method also considers linear paths; that is, the same result can be found when there is a small A and B has a positive effect on X as when there is a large A and small B.

Compared with a single-case analysis, QCA focuses on multiple cases to avoid the occurrence of individual accidents and make the conclusions more general. This is because the method considers the influence of the configuration on the result. In its early stages of development, QCA was widely used in the social sciences, such as sociology, political science, and management, to carry out research with small samples. With the development of QCA techniques, in recent years, it has also become frequently used to deal with large-sample analysis and the analysis of complex causal problems.

Compared with the traditional regression analysis used in quantitative studies and the case analysis used in qualitative studies, QCA is advantageous in that: (1) After a cross-case comparison of samples, researchers can not only identify the mechanism of action of conditional variables but also improve the external generalisation of conclusions. (2) At the same time, researchers can also analyse the conditional configuration with equivalent results by using the QCA method, which helps researchers explore differentiated driving mechanisms leading to the final results in different situations and analyse the mutual substitution and mutual adaptation relationship between conditions according to the results. (3) Due to causal asymmetry, the conditions that led to the result may differ from the conditions that did not appear, so these conditions can be compared to provide another theoretical dimension in which to interpret the problem.

3.3. Specific QCA Method Selection

Qualitative comparative analysis methods are divided into three categories according to the different categories of treatment variables: the fuzzy set qualitative comparative analysis (fsQCA) method, the multi-value set qualitative comparative analysis (mvQCA) method, and the clear-set qualitative comparative analysis (csQCA) method. For handling dichotomous variables, it is appropriate to use csQCA, where both conditions and results are either 0 or 1. For multi-categorical variables, it is suitable to use mvQCA, while fsQCA is needed for continuous variables.

Given time and data availability constraints, most data are logged as yes or no (1 and 0, respectively). The data in this study include information on whether the executive board includes Party members or women, the CSR vision of the enterprise, and whether the business leaders have overseas backgrounds, as well as quantitative data, such as net profit, total donations, and profit derived from CSR activities. Thus, the method selected for analysis was csQCA.

4. Data Collection and Construction

4.1. Data Collection

Five representative motivating factors were selected: “Party member on the board”, “women on the senior management team”, “CSR vision of the enterprise”, “overseas background among leaders”, and “enterprise profitable in the current year”. These factors cover the values and personal characteristics of enterprise executives, the corporate nature of the enterprise, and the ideological and political context. At the same time, all the data in this paper come from 2020, which limits the impact of external environmental factors such as the generational effect and public opinion on CSR performance.

We first used the CNRDS database as the basic data source, from which we derived data on all Chinese A-share listed companies in 2008 on the aforementioned five aspects. Then, we combined CSMAR and WIND data to corroborate the CNRDS data. In order to further ensure the authenticity and accuracy of the data, we also searched information related to the political party status, gender, and overseas experience of company leaders from the online search engine Baidu (https://www.baidu.com/ (accessed on 1 March 2022)) while companies’ data about donation and profitability are further manually collected from information platforms such as RoyalFlush (https://www.10jqka.com.cn/) (accessed on 1 March 2022), CSRC (http://www.csrc.gov.cn/) (accessed on 1 March 2022), Juchao Information (http://www.cninfo.com.cn/new/index) (accessed on 1 March 2022), and Sina Finance (https://finance.sina.com.cn/ (accessed on 1 March 2022)). The data were compared with those in the aforementioned databases and supplemented them. Finally, after excluding data samples with missing values for variables, a total of 1106 sets of sample data were obtained.

4.2. Data Construction

4.2.1. Outcome Variables

As mentioned above, there are many measures of CSR engagement, including global assessment reports such as that Rankings (RKS) or the voluntary disclosure of shares, which are quantitative indexes for CSR; donations can also be used as a CSR index. This paper examines the actual situation and considers the size, revenue, and amount of donations of each enterprise to measure whether the enterprise is meeting its social responsibilities. Considering that donations have certain tax incentives, enterprises may reasonably avoid tax through donations, so the proportion of donations to EBIT is used as a more reliable measure than other methods. This paper uses the measure of donations as follows: if the ratio of donations to EBIT is more than 1/10,000 or, if the current year is negative, donation expenditure (i.e., the proportion of donations is less than 0) is identified, this is considered as the enterprise fulfilling its social responsibilities. If EBIT is positive, but the ratio of donations to EBIT is less than 1/10,000, the enterprise is identified as not fulfilling its social responsibilities.

Table 1. Implications of the outcome variables and their assignments.

Outcome Variable	The Portion of Donations in EBITDA	Variable Assignment
Enterprise has fulfilled its social responsibilities	More than one in ten thousand	1
	Less than zero	1
Enterprise has not fulfilled its social responsibilities	Simultaneously more than or equal to zero and less than one in ten thousand	0

Source: The authors’ construction.

shows the implications of the outcome variables and their assignments.

4.2.2. Conditional Variables

Table 2. Implications of conditional variables and their assignments.

Conditional Variable	Variable Definition	Variable Assignment
Party member on the board	Enterprise board includes Party members	1
	Enterprise board does not include Party members	0
Women on the senior management team (or female executives)	The corporate executive team includes women	1
	The corporate executive team does not include women	0
CSR vision of the enterprise	Company already has ideas, a vision or values regarding its economic, social and environmental responsibilities	1
	Company has no ideas, vision, or values regarding its economic, social, or environmental responsibilities	0
Overseas background among leaders	Business leaders have an overseas background	1
	There is no overseas background among business leaders	0
Enterprise profitable in current year (simple profit)	Enterprise made a profit in the current year	1
	Enterprise did not make a profit in the current year	0

Source: The authors’ construction.

shows the implications of conditional variables and their assignments (taking the end of 2020 as the data collection time node).

5. Data Analysis and Positive Results

This study conducts a sufficient condition analysis of the configuration of each conditional variable and determine the conditional configuration that best explains the target case through the Boolean algebra minimisation method.

5.1. Necessity Analysis of Individual Conditions

In the necessity analysis of a single condition, there is an adequacy and necessity relationship between the conditional variables (independent variables) and the outcome variable (dependent variable), identified by two indicators of consistency and coverage. Concordance refers to the extent to which the sample cases share a particular condition that causes the result to occur. In general, the condition is a necessary condition for the result of a consistency level higher than 0.9 [79]. Coverage refers to the extent to which conditional variables explain the emergence of the results, and the greater the coverage rate is, the greater the explanatory power of the conditional variables for the outcome variables.

In this paper, there are 1106 companies considered, of which 911 performed CSR activities. In these 911 cases, the number of companies that are profitable is 841, and the consistency level of companies is 841/911 = 0.923161 for the outcome variable (fulfilment of CSR responsibilities). In a total of 1106 cases, 1014 met the “profitable” condition, and in these 1014 cases, 841 showed evidence of CSR performance, resulting in a coverage of 841/1014 = 0.829389.

This study used fsQCA software (Ragin and Sean, 2022 [80]) to analyse the necessity of the two outcome variables of CSR fulfilment and failure to fulfil CSR.

Table 3. Results of the individual conditional essentiality analysis.

	Fulfil CSR		Failure to Fulfil CSR
Conditional Variable	Consistency	Coverage	Consistency	Coverage
There is Party member on the board	0.785950	0.809955	0.861538	0.190045
There is no Party member on the board	0.214050	0.878378	0.138462	0.121622
There is at least one woman on the senior management team	0.658617	0.835655	0.605128	0.164345
There are no women on the senior management team	0.341383	0.801546	0.394872	0.198454
The company has a CSR vision	0.863886	0.816390	0.907692	0.183610
The company has no CSR vision	0.136114	0.873239	0.092308	0.126761
At least one board member has an overseas background	0.128430	0.812500	0.138462	0.187500
No board members have an overseas background	0.871570	0.825364	0.861538	0.174636
Company was profitable in the current year	0.923161	0.829389	0.887179	0.170611
The company was not profitable in the current year	0.076839	0.760870	0.112821	0.239130

Source: Developed by the author using fsQCA software.

presents the results of the individual conditional essentiality analysis.

According to the analysis in the table above, for the outcome variable of “CSR”, the consistency of “profitable” is approximately 0.92 (92% of the results), and the remaining level of consistency is below the empirical threshold of 0.9 [81], indicating that profit may be a necessary condition for CSR. At the same time, considering that its coverage is approximately 0.83 (meaning that it can explain 83% of the cases), which is higher than the empirical threshold value of 0.5, “profitable” can be identified as a necessary condition for a company to “fulfil its social responsibilities”.

This conclusion conforms to Preston and O’Bannon’s [48] capital supply hypothesis (the more money enterprises can provide, the more they can fulfil their social responsibilities), and combined with the social phenomenon in which public opinion regarding large profitable companies requires them to engage in more CSR activities and the legal situation in which donations provide tax breaks and thus are conducive to reasonable tax avoidance, it is not difficult to understand why “profitable” is a necessary condition for “enterprise has fulfilled its social responsibilities”.

For the outcome variable of failure to engage in CSR, the consistency level of the conditional variable is below 0.9, indicating that having a CSR vision may be a necessary condition for a company’s failure to fulfil its social responsibilities. However, its coverage is below 0.5, indicating that its consistency is above 0.9 only by chance. Therefore, “having a CSR vision” is not considered a necessary condition for “enterprise has not fulfilled its social responsibilities”.

For the other conditions, although having a CSR vision has some explanatory strength, the results are all below the empirical cut-off value of 0.9, so it cannot be used as a necessary condition for the results. From the above analysis, it can be seen that whether an enterprise engages in CSR activities is determined by multiple factors, so it is necessary to conduct a conditional configuration adequacy analysis to analyse the impact of the specific configuration on the results.

5.2. Sufficiency Analysis of the Conditional Configuration

Unlike the previous analysis of individual conditional necessity, the adequacy analysis of conditional configurations attempts to study the likelihood of results under different configurations based on the composition of multiple conditions. From the perspective of set theory, this step investigates whether a set of configurations composed of multiple conditions is a subset of the resulting set. As before, consistency and coverage are also used as measures of configuration, but their representative meaning and calculation method are different from the above.

This paper establishes the truth table based on the data obtained previously, as recorded in

Table 4. True values table.

Party Member on the Board	Women on the Senior Management Team	CSR Vision of the Enterprise	Overseas Background Among Leaders	Enterprise Profitable in the Current Year	Number	Raw Consist	PRI Consist	SYM Consist
0	1	0	1	1	5	1	1	1
0	1	1	1	0	3	1	1	1
0	0	0	0	0	2	1	1	1
0	1	0	0	0	2	1	1	1
1	0	0	0	0	1	1	1	1
0	1	0	1	0	1	1	1	1
0	0	1	1	1	14	0.928571	0.928571	0.928571
0	1	1	0	1	96	0.90625	0.90625	0.90625
0	1	1	0	0	9	0.888889	0.888889	0.888889
1	0	0	0	1	32	0.875	0.875	0.875
1	1	0	0	0	8	0.875	0.875	0.875
1	1	0	1	1	7	0.857143	0.857143	0.857143
1	1	0	0	1	62	0.854839	0.854839	0.854839
0	0	1	0	1	39	0.846154	0.846154	0.846154
1	1	1	1	1	45	0.822222	0.822222	0.822222
1	1	1	1	0	11	0.818182	0.818182	0.818182
1	0	1	0	1	246	0.813008	0.813008	0.813008
1	1	1	0	1	393	0.811705	0.811705	0.811705
0	1	1	1	1	29	0.793103	0.793103	0.793103
1	1	1	0	0	32	0.78125	0.78125	0.78125
1	0	1	1	1	24	0.75	0.75	0.75
1	0	0	1	1	3	0.666667	0.666667	0.666667
1	0	1	0	0	18	0.611111	0.611111	0.611111
0	0	0	0	1	4	0.5	0.5	0.5
0	0	1	0	0	3	0.333333	0.333333	0.333333
1	0	1	1	0	2	0	0	0
0	0	0	1	0	0
1	1	0	1	0	0
0	0	1	1	0	0
0	0	0	1	1	0

Source: Developed by the author using fsQCA software.

.

According to Schneider and Wagemann [82], raw consist should not be lower than the empirical critical value of 0.75, and PRI consist should not be lower than the empirical critical value of 0.75. Additionally, because 0.75 served as the naturally truncated point in the true values table, the threshold for raw consist and PRI consist was determined to be 0.75. The case frequency threshold also needs to be determined according to the specific number of study cases. In this study, the frequency of cases was determined to be 15, so more than 75% of the observed cases were retained.

The calculations were performed based on the previously mentioned raw consist and PRI consist threshold of 0.75 and case frequency threshold of 15 to obtain the true value data in

Table 5. The truth value table of screening results.

Party Member on the Board	Women on the Senior Management Team	CSR Vision of the Enterprise	Overseas Background among Leaders	Enterprise Profitable in the Current Year	Number	Result	Raw Consist	PRI Consist	SYM Consist
0	1	1	0	1	96	1	0.90625	0.90625	0.90625
1	0	0	0	1	32	1	0.875	0.875	0.875
1	1	0	0	1	62	1	0.854839	0.854839	0.854839
0	0	1	0	1	39	1	0.846154	0.846154	0.846154
1	1	1	1	1	45	1	0.822222	0.822222	0.822222
1	0	1	0	1	246	1	0.813008	0.813008	0.813008
1	1	1	0	1	393	1	0.811705	0.811705	0.811705
0	1	1	1	1	29	1	0.793103	0.793103	0.793103
1	1	1	0	0	32	1	0.78125	0.78125	0.78125
1	0	1	1	1	24	1	0.75	0.75	0.75
1	0	1	0	0	18	0	0.611111	0.611111	0.611111

Source: Developed by the author using fsQCA software to screen the data.

.

Since “profitable” was mentioned above as a necessary condition for “enterprise has fulfilled its social responsibilities”, it was treated as present in the analysis and obtained the following results in

Table 6. Simple solution.

Condition Configuration	Raw Coverage	Unique Coverage	Consistency
Female executive	0.658617	0.0603732	0.835655
Profitable	0.923161	0.324918	0.829389
Solution coverage	0.983535
Solution consistency	0.82963

Source: Developed by the author using fsQCA software.

(note: ~ means “non”, that is, the element does not exist; * means “and”, indicating that the listed conditions need to be met at the same time).

The specific meanings of each configuration are as follows: (note: ~ means “non”, indicating that the element does not exist; * means “and”, indicating that the conditions need to be met simultaneously)

Configuration 1: Party member on the board * ~ overseas background * profitable; that is, “profitable” is the core condition, “Party member on the board” and “~ overseas background” are the marginal conditions, which are combined as configuration 1 to produce “enterprise has fulfilled its social responsibilities”.

Configuration 2: female executive * ~ overseas background * profitable; that is, “profitable” and “female executive” are the core condition and “~ overseas background” is the marginal condition, which are combined as configuration 2 to produce “enterprise has fulfilled its social responsibilities” fulfilling corporate social responsibility”.

Configuration 3: CSR vision * ~ overseas background * profitable; that is, “profit” is the core condition, and “CSR vision” and “~ overseas background” are the marginal conditions, which are combined as configuration 3 to realize “enterprise has fulfilled its social responsibilities”.

Configuration 4: Party member on the board * CSR vision * profitable; that is, “profit” is the core condition, and “Party member on the board” and “CSR vision” are the marginal conditions, which are combined as configuration 4 to realize “enterprise has fulfilled its social responsibilities” fulfilling corporate social responsibility”.

Configuration 5: female executive * CSR vision * profitable; that is, “female executive” and “profit” are the core conditions, and “CSR vision” is the marginal condition, which are combined as configuration 5 to realize “enterprise has fulfilled its social responsibilities”.

Configuration 6: Party member on board * female executive * CSR vision * ~ overseas background; that is, “female executive” is the core condition, “Party member on board”, “CSR vision” and “~ overseas background” are the marginal conditions, which are combined as configuration 6 to realize “enterprise has fulfilled its social responsibilities”.

Following Pappas and Woodside [83], all parsimony solutions are used as the core conditions. As shown in Configuration 2 (female executive * ~ overseas background * profitable), the intermediate solution corresponds to two simple solutions, both “female executive” and “profitable”, which are taken as the core conditions of this configuration.

In the above table, the consistency representative indicates the configuration, and the case indicates the proportion of all cases. In the intermediate solution condition configuration “Party member on the board * ~ overseas background * profitable”, that is, the board has Party members but no members with an overseas background, and the enterprise was profitable in all 733 cases; of those cases, 600 fit the result of “enterprise has fulfilled its social responsibilities”, so the consistency is 600/733 = 0.818554. The raw coverage represents the percentage of all cases that can be explained by any configuration, while unique coverage represents the percentage of cases that can be only explained by the specific configuration being considered. Taking the conditional configuration of “Party member on the board * ~ overseas background * profitable” in the intermediate solution of

Table 7. Intermediate solution.

Condition Configuration	Raw Coverage	Unique Coverage	Consistency
Party member on board * ~ overseas background * profitable	0.658617	0.0307355	0.818554
Female executive * ~ overseas background * profitable	0.520307	0.0164654	0.837456
CSR vision * ~ overseas background * profitable	0.701427	0.0362239	0.825581
Party member on board * CSR vision * profitable	0.630077	0.0197585	0.810734
Female executive * CSR vision * profitable	0.511526	0.025247	0.827709
Party member on board * female executive * CSR vision * ~ overseas background	0.377607	0.0274424	0.809412
Solution coverage	0.919868
Solution consistency	0.827246

Source: Developed by the author using fsQCA software. Note: The specific meanings of each configuration are as follows: (note: ~ means “non”, indicating that the element does not exist; * means “and”, indicating that the conditions need to be met simultaneously).

(i.e., the condition combination that can realize the target result) as an example, a total of 911 cases result in “enterprise has fulfilled its social responsibilities”. Among them, 600 can be explained by “Party member on the board * ~ overseas background * profitable”, and its original coverage is 600/911 = 0.658617. Then, among these cases, 28 can be explained only by “Party member on the board * ~ overseas background * profitable” and not by other conditional configurations. Thus, its unique coverage is 28/911 = 0.0307355.

As shown in Table 6, “female executives” and “profit” play a central role in each configuration as a simple solution to “enterprise has fulfilled its social responsibilities”. This conclusion is consistent with the previous propositions by McGuinness et al. [25] that “female CEOs have a significantly enhancing effect on CSR performance”. At the same time, “profit” is a necessary condition for “enterprise has fulfilled its social responsibilities”. Therefore, this simple solution is both logical and reasonable.

Some contradiction arises with the previous literature because “~ leader overseas background”, which is repeatedly mentioned in the intermediate solution, can promote the performance of CSR, which is contradictory to the aforementioned argument that “CEOs with overseas experience will rate health care, which is one of the CSR-related areas, as more important on average for companies to address than those without overseas experience” [26]. First, QCA emphasises that “causal asymmetry” and “way”, namely, for “overseas background”, because of “Party member on the board” or “female executive” or “CSR vision”, return a positive for the results and dilute the negative effect of “~ overseas background” on the results, which is the embodiment of the unique QCA method. At the same time, after in-depth research, this paper found that: as previous studies have said, state-owned enterprises are more inclined to fulfil their social responsibilities than private enterprises. As these enterprises are the lifeblood of the country, their leaders need to undertake certain political missions. Therefore, the leaders of state-owned enterprises are stricter and tend to be “born and raised without overseas background”. Private enterprises’ board executives are less strict than state-owned enterprises’, thus lowering the threshold for “no overseas background”.

As shown in

Table 8. Analysis of CSR configuration.

Condition Configuration	Configuration 1	Configuration 2	Configuration 3	Configuration 4	Configuration 5	Configuration 6
Party member on board	♦			♦		♦
Female executive		⬤			⬤	⬤
CSR vision			♦	♦	♦	♦
Overseas background	⊗	⊗	⊗			⊗
Profitable	⬤	⬤	⬤	⬤	⬤
Consistency	0.818554	0.837456	0.825581	0.810734	0.827709	0.809412
Raw coverage	0.658617	0.520307	0.701427	0.630077	0.511526	0.377607
Unique coverage	0.0307355	0.0164654	0.0362239	0.0197585	0.025247	0.0274424
Solution coverage	0.919868
Solution consistency	0.827246

The author determined the above intermediate solution and the simple solution; Note: ⬤ represents the existence of the core condition, ♦ represents the existence of the edge condition, ⊗ represents the absence of the edge condition, and a blank cell represents the existence of a conditional variable that is irrelevant to the outcome.

, the conditional configurations of CSR are divided into six categories, each above the acceptable minimum level of 0.75, and the overall solution coverage (solution coverage) is approximately 0.92, which means that the six categories of CSR can explain approximately 92% of the cases of CSR performance. The overall solution consistency is approximately 0.83, which means that in all cases of these 6 categories, approximately 83% of companies fulfil their social responsibilities. The 6 configurations in Table 8 can be considered sufficient conditions for fulfilling CSR, specifically analysed as follows.

6. Robustness Test

6.1. Methods Introduction

Generally, the robustness test of the QCA method (QCA) can be conducted by readjusting the consistency threshold during the analysis and the case frequency, increasing or decreasing the cases appropriately, and replacing the conditional variables. Based on previous studies, this paper used an adjusted consistency threshold to test the robustness of increases or decreases in the number of cases. If the intermediate solutions and simple solutions obtained after the change condition do not change much overall, or if they are the same as for previous subsets, the result can be considered robust.

6.2. A Robustness Test Was Performed to Change the Consistency Threshold

6.2.1. Operation Method

Changing the consistency threshold is a common method of robustness testing in QCA. Above, the empirical critical value of 0.75 is selected as the consistency threshold (i.e., 1 for raw consist to 0.75 and 0 for raw consist to 0.75). This section raises the value to 0.8 to determine the robustness of the conclusion.

6.2.2. Sufficiency Analysis of the Conditional Configuration

As in the previous operation, after considering the true value table (Table 4), we set the frequency threshold at 15 (as above) and the consistency threshold at 0.8 (increased by 0.75). The filtered true value table is shown below in

Table 9. True value of the filtered results (changing consistency threshold).

Party Member on Board	Female Executive	CSR Vision	Overseas Background	Profitable	Number	Result	Raw Consist	PRI Consist	SYM Consist
0	1	0	0	1	15	1	1	1	1
0	1	1	0	1	96	1	0.90625	0.90625	0.90625
1	0	0	0	1	32	1	0.875	0.875	0.875
1	1	0	0	1	62	1	0.854839	0.854839	0.854839
0	0	1	0	1	39	1	0.846154	0.846154	0.846154
1	1	1	1	1	45	1	0.822222	0.822222	0.822222
1	0	1	0	1	246	1	0.813008	0.813008	0.813008
1	1	1	0	1	393	1	0.811705	0.811705	0.811705
0	1	1	1	1	29	0	0.793103	0.793103	0.793103
1	1	1	0	0	32	0	0.78125	0.78125	0.78125
1	0	1	1	1	24	0	0.75	0.75	0.75
1	0	1	0	0	18	0	0.611111	0.611111	0.611111

Source: Developed by the author using fsQCA software to screen the data.

.

6.2.3. Data Results

As above, “profitable” is set as present, and the following results in

Table 10. Simple solution (changing consistency threshold).

Condition Configuration	Raw Coverage	Unique Coverage	Consistency
~ Overseas background * Profitable	0.809001	0.400659	0.830891
Party member on board * female executive * ~ overseas background	0.0570801	0.00987929	0.825397
Party member on board * female executive * profitable	0.455543	0	0.81854
Solution coverage	0.866081
Solution consistency	0.830526

Source: Developed by the author using fsQCA software. Note: The specific meanings of each configuration are as follows: (note: ~ means “non”, indicating that the element does not exist; * means “and”, indicating that the conditions need to be met simultaneously).

and

Table 11. Intermediate solution (changing consistency threshold).

Condition Configuration	Raw Coverage	Unique Coverage	Consistency
Party member on board * ~ overseas background * profitable	0.658617	0.0307354	0.818554
Female executive * ~ overseas background * profitable	0.520307	0.0164654	0.837456
CSR vision * ~ overseas background * profitable	0.701427	0.0362239	0.825581
Party member on board * female executive * CSR vision * profitable	0.390779	0.0406147	0.812785
Solution coverage	0.84742
Solution consistency	0.831897

Source: Developed by the author using fsQCA software. Note: The specific meanings of each configuration are as follows: (note: ~ means “non”, indicating that the element does not exist; * means “and”, indicating that the conditions need to be met simultaneously).

are obtained (note: ~ means “non”, that is, the element does not exist; * means “and”, meaning that the conditions need to be met simultaneously).

As seen from the above tables, there are three simple solutions: (1) overseas background * profitable; (2) Party member on the board * female executive * overseas background; and (3) Party member on the board * female executive * profitable. These three simple solutions are subsets of simple solutions (female executive; profitable) at a threshold of 0.75, where (1) and (3) are subsets of “profitable” at a threshold of 0.75, and (2) and (3) are subsets of “female executive” at a threshold of 0.75.

The intermediate solution, with a threshold of 0.8, is divided into (1) Party member on the board * ~ overseas background * profitable; (2) Party member on the board * ~ overseas background * profitable; (3) CSR vision * ~ overseas background * profitable; and (4) Party member on the board * female executive * CSR vision * profitable. After a comparison with the previous article, it is found that the three intermediate solutions at 1, 2, and 3 and the threshold of 0.75 correspond one to one. For the intermediate solution at 4, it is a subset of the intermediate solution at the threshold of 0.75.

It can therefore be argued that these results verify the robustness of the conclusions using the method of changing consistency thresholds.

6.3. Robustness Test with Censored Cases

6.3.1. Data Filtering

In this paper, we randomly deleted approximately half of the cases of enterprises that engage in CSR activities (assigned 1) and half of the cases of enterprises that do not engage in CSR activities (assigned 0) to obtain a total of 553 valid data points, which are imported into fsQCA software.

6.3.2. Necessity Analysis of Individual Conditions

The results of the necessity analysis are shown in

Table 12. Necessity analysis results of robustness test (deleted cases).

	Fulfilled Its CSR		Failed to Fulfil Its CSR
Conditional Variable	Consistency	Coverage	Consistency	Coverage
Party member on board	0.788079	0.802247	0.880000	0.197753
No Party member on board	0.211921	0.888889	0.120000	0.111111
Female executives	0.651214	0.828652	0.610000	0.171348
No female executives	0.348786	0.802030	0.390000	0.197970
CSR vision	0.854305	0.816456	0.870000	0.183544
No CSR vision	0.145695	0.835443	0.130000	0.164557
Overseas background	0.110375	0.781250	0.140000	0.218750
No overseas background	0.889625	0.824131	0.860000	0.175869
Profitable	0.938190	0.825243	0.900000	0.174757
Not profitable	0.061810	0.736842	0.100000	0.263158

Source: Developed by the author using fsQCA software.

.

As seen from the above table, for the new data obtained by removing half of the original cases, the consistency of “profitable” is above the empirical threshold of 0.9, and the coverage is above 0.5, while the consistency of “CSR Vision” is reduced from 0.908 to 0.870, and the coverage also does not exceed 0.5. This means that “CSR Vision” is not a necessary condition to achieve the target result (no CSR activities), just coincidental. At the same time, the table shows that for the result of no CSR activities, the consistency of “profitable” is higher than 0.9, but its coverage is less than 0.5. Additionally, the necessity analysis in Table 3 shows that the consistency is below 0.9; the variable is thus identified as coincidental and not as a necessary condition for CSR implementation.

6.3.3. Sufficiency Analysis of the Conditional Configuration

The true value in

Table 13. Robustness test (deleted cases).

Party Member on Board	Female Executive	CSR Vision	Overseas Background	Profitable	Number	Raw Consist	PRI Consist	SYM Consist
0	1	0	0	1	8	1	1	1
0	0	1	1	1	5	1	1	1
0	1	0	1	1	4	1	1	1
1	1	0	0	0	2	1	1	1
0	1	0	0	0	1	1	1	1
0	1	0	1	0	1	1	1	1
0	1	1	1	0	1	1	1	1
0	1	1	0	1	49	0.959184	0.959184	0.959184
1	1	1	0	0	16	0.875	0.875	0.875
1	0	1	0	1	127	0.834646	0.834646	0.834646
1	1	0	0	1	36	0.833333	0.833333	0.833333
1	0	1	1	1	10	0.8	0.8	0.8
1	1	1	0	1	196	0.795918	0.795918	0.795918
1	0	0	0	1	19	0.789474	0.789474	0.789474
0	0	1	0	1	19	0.789474	0.789474	0.789474
0	1	1	1	1	13	0.769231	0.769231	0.769231
1	1	1	1	1	21	0.761905	0.761905	0.761905
0	1	1	0	0	3	0.666667	0.666667	0.666667
1	0	0	1	1	3	0.666667	0.666667	0.666667
1	1	0	1	1	3	0.666667	0.666667	0.666667
1	0	1	0	0	9	0.555556	0.555556	0.555556
0	0	1	0	0	2	0.5	0.5	0.5
1	1	1	1	0	2	0.5	0.5	0.5
0	0	0	0	1	2	0.5	0.5	0.5
1	0	1	1	0	1	0	0	0
0	0	0	0	0	0
1	0	0	0	0	0
0	0	0	1	0	0
1	0	0	1	0	0
1	1	0	1	0	0
0	0	1	1	0	0
0	0	0	1	1	0

Source: Developed by the author using fsQCA software to import the data.

shows the results of the necessity analysis of a single condition.

The results were filtered using a consistency threshold of 0.75 and in case frequency of 7 (the number of cases for the robustness test should be reduced compared to all cases), and the filtered true values were recorded in

Table 14. Screening results of robustness test, true values table (deleted cases).

Party Member on Board	Female Executive	CSR Vision	Overseas Background	Profitable	Number	Result	Raw Consist	PRI Consist	SYM Consist
0	1	0	0	1	8	1	1	1	1
0	1	1	0	1	49	1	0.959184	0.959184	0.959184
1	1	1	0	0	16	1	0.875	0.875	0.875
1	0	1	0	1	127	1	0.834646	0.834646	0.834646
1	1	0	0	1	36	1	0.833333	0.833333	0.833333
1	0	1	1	1	10	1	0.8	0.8	0.8
1	1	1	0	1	196	1	0.795918	0.795918	0.795918
1	0	0	0	1	19	1	0.789474	0.789474	0.789474
0	0	1	0	1	19	1	0.789474	0.789474	0.789474
0	1	1	1	1	13	1	0.769231	0.769231	0.769231
1	1	1	1	1	21	1	0.761905	0.761905	0.761905
1	0	1	0	0	9	0	0.555556	0.555556	0.555556

Source: Developed by the author using fsQCA software to screen the data.

.

6.3.4. Data Results

By analysing the true value table after screening, since it is mentioned above that “profitable” is a necessary condition to achieve the target result, which can also be found after eliminating half of the cases and considering the simple solution and the intermediate solution obtained by indicating “profitable” as present, as shown in

Table 15. Simple solution for the robustness test (deleted cases).

Condition Configuration	Raw Coverage	Unique Coverage	Consistency
Female executive	0.651214	0.0485651	0.828652
Profitable	0.93819	0.335541	0.825243
Solution coverage	0.986755
Solution consistency	0.826248

Source: Developed by the author using fsQCA software.

and

Table 16. Robustness test, intermediate solution (deleted cases).

Condition Configuration	Raw Coverage	Unique Coverage	Consistency
Party member on board * ~ overseas background * profitable	0.677704	0.0331126	0.812169
Female executive * ~ overseas background * profitable	0.532009	0.0176601	0.83391
CSR vision * ~ overseas background * profitable	0.715232	0.0331126	0.828645
Party member on board * CSR vision * profitable	0.631347	0.0176601	0.80791
Female executive * CSR vision * profitable	0.505519	0.0220751	0.820789
Party member on board * female executive * CSR vision * ~ overseas background	0.375276	0.0309051	0.801887
Solution coverage	0.93819
Solution consistency	0.826848

Source: Developed by the author using fsQCA software. Note: The specific meanings of each configuration are as follows: (note: ~ means “non”, indicating that the element does not exist; * means “and”, indicating that the conditions need to be met simultaneously).

.

As seen from the above table, after removing half of the cases, the path combination of the intermediate solutions and the corresponding original coverage (raw coverage), the unique coverage (unique coverage), and the consistency level (consistency) do not change significantly. The path combination also does not change greatly. The coverage degree (solution coverage) of the overall solution changed from 0.919868 to 0.93819, and the consistency level (solution consistency) of the overall solution changed from 0.827246 to 0.826848.

Therefore, by eliminating half of the cases, this section proves the robustness of the configuration analysis results for the performance of CSR.

6.4. Robustness Test Conclusion

In this section, the robustness of the aforementioned conclusions is tested by changing the consistency threshold and removing half of the cases, and the results prove to be very robust. At the same time, the conclusions proposed in this paper are also consistent with the conclusions proposed by previous literature and the theoretical analysis of the slight differences. Furthermore, the conclusions proposed herein also agree with the logic of operation in real society, so the conclusions of this paper can be considered effective.

7. Conclusions and Outlook

7.1. Review of Previous Literature

In terms of executive values, characteristics, and environmental impact, the study found that CEOs who have poverty experience, have daughters, overseas experience, low confidence, or “career concerns”; companies whose executive team includes women, whose average tenure is low, and whose social capital is heterogeneous (including term heterogeneity and education heterogeneity); enterprises with a chain director system (except chairman); and enterprises with cultural practices and ideals such as gratitude, Confucian values, and multiculturalism are more likely to engage in CSR activities.

In terms of enterprise characteristics, the study found that: (1) state-owned enterprises, large enterprises, listed enterprises, and family-owned enterprises; enterprises with family members on the board of directors; enterprises with long-term family control of the company; and enterprises with lower shareholder ratios are more inclined to engage in CSR activities. (2) The relationship between financial performance and CSR performance is very complex. Generally, the more funds that enterprises have, the more likely they are to engage in CSR activities. However, there are exceptions. According to signal transmission theory, worse-performing companies perform more CSR activities to signal to society that the more money they have due to government subsidies, the more money they can designate for CSR activities. In addition, there is an endogenous interaction between CSR and financial performance in the current year and inter-period timeframes. (3) Enterprises also fulfil their social responsibilities for many hidden reasons, such as a desire to raise their stock price, guide investors’ long-term investment, influence consumers’ purchase decisions, gain political relevance, and direct attention from the outside world.

In terms of the external environment of enterprises, this research found that: (1) public opinion can encourage enterprises to fulfil their social responsibilities. (2) In more civilized external environments (e.g., with a high degree of marketisation, sound rule of law, transparent information policies), the factors affecting the performance of CSR and the “reputation insurance” effect will be significantly weakened, but the influence of participating in Party organisation will be magnified. (3) The higher the degree of government intervention is, the weaker the heterogeneous influencing factors of the senior executive team will be.

In terms of ideological and political construction, this research found that (1) the Party membership of entrepreneurs, senior executives, and chairmen plays an obvious role in promoting CSR performance. (2) The establishment of Party organisations helps enterprises fulfil their social responsibilities, including reducing the degree of surplus management, protecting the environment, improving employee welfare, narrowing the gap between rich and poor, and increasing the employment rate.

Research on the motivations for CSR implementation is fairly rich and well established, but there is little literature that examines the combination of different motivations. This paper adopts a relatively novel approach, i.e., the csQCA approach, to examine the combined role of multiple aspects of motivation more comprehensively. This not only complements the previous literature on CSR motivations but also provides a useful reference for subsequent research in this field. In addition, the paper explores whether corporate leaders’ membership in the Communist Party has an impact on corporate social responsibility in the context of China’s unique national context, which enriches research findings at the intersection of political ideology and corporate social responsibility.

7.2. QCA Empirical Conclusions

Data on listed companies in 2020 (1106 valid data) were selected through the QCA method to indicate the existence of women on the board, the presence of a CSR vision, the overseas background of leaders, and the profitability of the enterprise to explore the motivation to engage in CSR activities, and the following conclusions were made:

(1)

“Profitability” is a necessary condition for enterprises to engage in CSR activities. That is, under general circumstances, if an enterprise wants to engage in CSR activities, it must make a profit that year. When the enterprise loses money in a given year, it will most likely not engage in CSR activities.

(2)

The simple solution of “women on the executive team” and “profitability” for the outcome variable of “enterprise has fulfilled its social responsibilities” plays a core role in each configuration path (intermediate solution). This view and the previous literature review suggest that having “female CEOs have a significantly enhancing effect on CSR performance” by McGuinness et al. [25] and that the “capital supply hypothesis” proposed by Preston and O’Bannon [48] are both confirmed.

(3)

A total of six configurations provide a path to a positive result for engagement in CSR activities. These configurations explain approximately 92% of the positive results, and approximately 83% of all the companies aligning with one of the six configurations engage in CSR activities.

(4)

For the first path (configuration 1), “enterprise profitable in the current year” plays a core role, while “Party member on the board” and “overseas background of leaders” play a key auxiliary role. Approximately 81.86% of all the cases with this configuration of conditions showed a positive result for “enterprise has fulfilled its social responsibilities”. Of all such positive cases, approximately 65.86% can be explained by this path, and approximately 3.10% of cases can be explained only by this path.

(5)

For the second path (configuration 2), “profitable in the current year” and “women on the senior management team” play a core role, and “overseas background of leaders” plays an auxiliary role. Approximately 83.75% of all the cases following this configuration of conditions showed a positive result for “enterprise has fulfilled its social responsibilities”. Of all such positive cases, approximately 52.03% can be explained by this path, and approximately 1.65% can be explained only by this path.

(6)

For the third path (configuration 3), “profitable in the current year” plays a core role, while “CSR vision of the enterprise” and “overseas background of leaders” play a supporting role. Approximately 82.56% of all the cases with this configuration of conditions showed a positive result for “enterprise has fulfilled its social responsibilities”. Of all such positive cases, approximately 70.14% can be explained by this path, and approximately 3.62% can be explained only by this path.

(7)

For the fourth path (configuration 4), “profitable in the current year” plays a core role, while “CSR vision of the enterprise” and “Party member on the board” play an auxiliary role. Approximately 81.07% of all the cases with this configuration of conditions showed a positive result for “enterprise has fulfilled its social responsibilities”. Of all such positive cases, approximately 63.01% can be explained by this path, and approximately 1.98% can be explained only by this path.

(8)

For the fifth path (configuration 5), “profitable in the current year” and “women on the senior management team” play a core role, while “CSR vision of the enterprise” plays a supporting role. Approximately 82.77% of all the cases with this configuration of conditions showed a positive result for “enterprise has fulfilled its social responsibilities”. Of all such positive cases, approximately 51.15% can be explained by this path, and approximately 2.52% of them can be explained only by this path.

(9)

For the sixth path (configuration 6), “women on the senior management team” plays a core role, while “Party members on the board”, “CSR vision of the enterprise” and “overseas background of leaders” play an auxiliary role. Approximately 80.94% of all the cases with this configuration of conditions showed a positive result for “enterprise has fulfilled its social responsibilities”. Of all such positive cases, approximately 37.76% can be explained by this path, and approximately 2.74% can be explained only by this path.

7.3. Theoretical and Practical Contribution

7.3.1. Theoretical Contribution

This paper summarises the motivation behind CSR and provides insights in line with China’s national conditions. At the same time, in terms of the quantitative research analysis of CSR, QCA involves the configuration analysis of the mutual influence of factors and the degree of influence these configurations have on the results. In real life, various factors are often related to each other. For example, in this paper, when the leaders of state-owned enterprises are often party members, the QCA method pays more attention to the synergistic effect of multiple conditional variables between cases than traditional regression analysis methods do, which is also more in line with the actual situation.

In this study, we extend the contingency theory by building a theoretical framework that covers a wide range of factors influencing CSR activities, emphasising the importance of considering management perceptions and contextual variables when studying CSR decisions. On the other hand, our theoretical framework incorporates a consideration of the unique Chinese context, projecting the influence of political ideology on business leaders and, in turn, on CSR strategic decisions. As a result, our research also contributes to a better understanding of the long-standing and much-talked-about issue of the nature of the relationship between the business, market, and government.

7.3.2. Practical Contribution

Based on this paper’s findings, this paper suggests the following ways to encourage enterprises to fulfil their social responsibilities:

Senior management and environmental impact: In essence, the decision of the enterprise is made by the senior management of the enterprise, so when exerting influence on the enterprise decision, we can often start with the characteristics of corporate management. Following education about CSR, enterprises can add more women to their executive teams, the length of tenure of executive team members can be diversified, the heterogeneity of the executive team can be improved, and a chain director system can be established, which are also cultural factors that can encourage enterprise managers to enhance the enterprise’s CSR performance.

Enterprise level: The proportion of enterprises’ state-owned assets can be increased through investment and acquisition, and the performance of CSR can be promoted through various means, such as transforming enterprises into “large-scale”, “listed”, and “family-oriented” entities. Enterprises can also be encouraged to fulfil their responsibilities by promulgating a series of policies, such as providing tax incentives for engaging in CSR activities.

External influence level: By strengthening the influence of public opinion, the power of the media can be used to encourage CSR and expose enterprises that endanger the environment and the safety of people’s lives and property. Government intervention in enterprises can also be strengthened, and relevant systems can be formulated, such as those that use CSR indicators as an important assessment method, require the compulsory disclosure of CSR performance, or use other means to encourage enterprises to fulfil their social responsibilities.

Ideological and political construction: The literature search found that Party organisations are an important part of enterprise organisational structures and thus affect the enterprise’s CSR activities, such as poverty alleviation work. Indeed, this factor has played an indispensable role, embodied in the enterprise’s internal management, especially the influence of the behaviour of internal personnel and external forces.

Through these implementations in the four abovementioned areas, although the paradox of CSR cannot be fundamentally solved, the performance of CSR can be promoted to accelerate the construction of a harmonious society and the realisation of the Chinese dream.

7.4. Study Limitations ad Future Outlook

There are some limitations to this paper. For example, although this paper analyses the causes of CSR performance in five dimensions, according to previous research, the motivation behind CSR performance is often more complex, and these five dimensions are thus insufficient to cover all possible causes. At the same time, the ways CSR manifests often vary, such as ensuring a happy workplace, improving the employment rate, paying taxes, carrying out national construction projects, protecting the rights and interests of all parties, making public donations, and participating in disease prevention and mitigation, environmental protection, and poverty alleviation. This article is based on the more typical CSR performance measure of public welfare donations, which is relatively one-sided. Follow-up research can consider comprehensively measuring the level of CSR performance from multiple perspectives. In addition, due to the limited time and resources available to collect data, the csQCA adopted in this paper does not allow us to further examine related degree changes and some alternative variables, unlike the fuzzy set qualitative comparative analysis method (fsQCA). It is hoped that future research can learn from the ideas of this paper and examine the motivation behind CSR in more depth.

This paper uses a csQCA of data from A-share listed companies in 2020, but because of time, resource, and data availability limitations, the motivation dimension is not thoroughly considered, and the evaluation of the degree of CSR performance is overly simplified. Furthermore, using csQCA prevents us from conducting degree change analysis to resolve these problems. We hope that this paper can be a valuable reference for future scholars researching CSR-related issues and provide new ideas. We hope that subsequent scholars from other areas consider the motivation behind CSR or conduct a comprehensive evaluation of CSR using different analysis methods to fill the gaps remaining after this paper.

In addition, this paper is based on a study of China, a rapidly developing economy in transition that has its own particular economic and political context, so the results may be somewhat narrowly focused. In fact, there is similar literature on the impact of political ideology on CSR based on the US context. Some scholars have measured the internal political environment of companies using the partisan leanings of campaign contributions and found that companies engage in more CSR when their founders, CEOs, directors or employees are Democrats (liberal leanings) rather than Republicans (conservative leanings) [84,85,86]. Future research could further explore the relationship between political ideology and corporate social responsibility in other economies, both developed and developing. For example, in Japan, there are various political parties representing different sectors of the population, among which the conservative-leaning Liberal Democratic Party (LDP) has been in power most of the time since the Second World War. Despite the mismanagement of the LDP’s ‘political clientelism’ and the growing inequalities it has created, other parties advocating fairer welfare policies, generous allowance programs, and renewable energy have failed to gain enough support from voters [87,88,89]. It could be meaningful to explore the impact of CSR more broadly based on these more diverse institutional and cultural contexts.