When Does CSR Fail to Incentive Employees’ Affective Organizational Commitment? Exploring the Moderating Effects Based on the C-S-R Concerns Model

Abstract

Corporate social responsibility (CSR) is a strategy employed to engage stakeholders. Prior studies explored the different mechanisms by which employees, as one of an organization’s biggest stakeholder groups, positively respond to organizational CSR behavior. However, limited attention has been paid to the conditional boundaries when employees’ positive responses are stronger or weaker, overlooking the possibility that sometimes CSR fails to obtain employees’ support. The study employed the “C-S-R concerns” model to systematically investigate how caring-based (i.e., moral identity), self-based (i.e., corporate ability), and relational-based (i.e., importance of CSR) factors moderate the relationship between perceived CSR (PCSR) and affective organizational commitment (AOC). Using data collected from 326 employees in the Chinese high-tech industry, we found that the positive effects of employees’ PCSR on their AOC becomes negative when employees have higher levels of moral identity, when their organizations are perceived as having greater ability, and when CSR is perceived as more important to the success of their organization. The findings advance our understanding of the inconclusive impacts of CSR on firm performance by providing micro-level evidence from employees. This also provides practical implications for managers about how to select employees, set CSR strategies and cultivate organizational cultures.

1. Introduction

Corporate social responsibility (CSR) has drawn increasing attention from management researchers over the last three decades. Carroll [1] claimed that CSR encompasses not just ethical and discretionary duties but also financial and legal ones. Recent studies also included a stakeholder approach and adopted a broader definition [2]. For example, Aguinis and Glavas [3] defined CSR as “organizational activities and policies to integrate stakeholder expectations and meet a triple bottom line of economic, social, and environmental performance”. It is generally agreed that the purpose of CSR is to build mutual benefit [4], boost intangible resources and corporate capabilities [5,6], and influence stakeholder behavior patterns [7,8], to make an overarching contribution to firms’ competitive advantage. In today’s shifting market, a firm’s CSR strategy should emphasize close bonds with employees, customers, suppliers, shareholders, creditors, communities, and other stakeholders, while also encouraging strategic goals for long-term and sustainable success [9]. As a result, most organizations employ CSR as a business strategy [10].

Previous studies on CSR can be split into the following two streams: macro-level studies and micro-level studies. Previously, CSR studies primarily considered firm level outcomes, examining the relationship between CSR and financial performance. However, their findings are inconclusive, and the question of why CSR cannot always contribute to improving firms’ financial performance remains unanswered [11]. As a consequence, an increasing number of studies focus on micro-level studies, examining the impact of CSR on individual stakeholders. Given that employees can be considered an organizations’ most important stakeholders, a range of studies investigated employees’ perceptions of organizational CSR activities, and how these perceptions can influence employee behavior. Numerous studies found that employees’ recognition of their organization’s CSR activities can incentivize positive behavior, such as affective organizational commitment (AOC) and organizational citizenship behavior, as well as increase employee retention [12,13,14,15]. Furthermore, these studies also explored the potential mechanisms through which employees’ perceived CSR (PCSR) affects their positive behaviors, which advances our understandings of how CSR increases firm-level performance by encouraging individual stakeholders’ positive behaviors.

However, the existing literature overlooks the fact that PCSR may not always lead to positive behaviors of employees, and comprehensive knowledge about the circumstances in which PCSR may fail to result in positive employee attitude and behavior is lacking. AOC is one of the most widely explored positive attitudes in organization studies that indicates the extent to which employees want to work in their organizations, which is linked with a wide range of potential positive outcomes, such as job satisfaction, job involvement, work experiences and turnover. Therefore, this study aimed to systematically examine the boundary conditions of the relationship between PCSR and AOC. To be specific, this study sought to answer the following research questions: (1) Does CSR positively influence employees’ AOC? (2) In which circumstances can CSR not help to enhance employees’ AOC?

We performed this study by adopting the “C-S-R-concern” model to advance our understanding about when CSR activities conducted by organizations bring about perceived positive impacts for their employees, as well as for overall firm performance. Following the literature on the relationship between PCSR and AOC, we first theorized and examined the relationship based on social identity theory and social exchange theory. Then, we explored the moderating effects of moral identity (MI), corporate ability (CA), and the importance of CSR (ICSR). These three moderators can reflect the three dimensions of the “C-S-R Concerns” model proposed by David and Deborah [16], which describes the ethic-, instrument-, and relational motives of employees. We argued that MI, CA, and ICSR may all raise employees’ expectations of CSR, thus weakening the positive effects of PCSR on AOC.

This study employed a quantitative research methodology to test hypotheses by using survey data of 326 employees from the Chinese high-tech industry [17]. Ordinary Least Squares (OLS) Regression was used to estimate the relationship between PCSR and AOC and the moderating effects. Consistent with prior studies, we found that PCSR has significant positive effects on AOC [18]. Interestingly, we found that this positive effect becomes negative when employees have high moral identity (MI), when organizations have greater abilities (CA), and when employees perceive CSR as important to their organizations (ICSR). This suggests that employees cannot increase their commitment to their organizations if they believe their organizations should take responsibility for stakeholders, their organizations have the ability to do so, and they believe their organizations conduct CSR also for their own good. Such beliefs may set higher expectations of CSR, consequently failing to incentivize employees to commit more to their organizations. These findings enrich the literature on the relationship between PCSR and AOC by systematically exploring the conditional boundaries where PCSR may not effectively incentivize AOC [19,20,21]. The boundary conditions also contribute to specifying the circumstances where social identity theory and social exchange theory can be applied to explain the positive effects of PCSR on AOC. More generally, our findings also advance the knowledge of why CSR cannot always contribute to an increase in firm performance, a question yet to be answered in the CSR literature.

The following is a summary of the structure of this paper. Section 2 reviews the literature and our hypotheses are presented in Section 3, followed by a description of the research methodology, including the analysis, in Section 4. Our findings are then discussed in Section 5, and the paper concludes by presenting the study’s contributions, limitations, and possibilities for future research.

2. Literature Review

2.1. Two Streams of CSR Research

CSR is a self-regulatory business model, according to which firms are required to be socially responsible for their stakeholders and the general public in economic, social, and environmental aspects [3]. Corporations engaging in CSR make active efforts to strengthen their social and environmental performance [22]. CSR is based on the premise that businesses should seek additional pro-social aims to maximize profitability. This intrinsic value of CSR was not identified until the 1990s, with most academics previously viewing CSR as a necessary “cost” to maintain an organization’s reputation or positive image. In recent year, CSR has gradually become of strategic importance as a way to proactively engage stakeholders and avoid negative spillovers to the wider public. Moreover, CSR practices have been developed with a broader scope than before. Numerous organizations have attempted to increase the environmental sustainability of their operations, such as implementing renewable energy sources or acquiring carbon offsets [23]. In supply chain management, efforts have also been undertaken to avoid illegal employment practices, including child labor and slavery [7]. In relation to employees, CSR programs can assist in establishing more robust bonds with the firm, enhance morale, and create a perception of connectedness with the wider world in which the organization operates [24].

The present status of mainstream CSR research can be divided into the following two streams: macro-level research and micro-level research. In general, macro-CSR studies focus on firm-level outcomes, exploring the extent to which CSR practices can contribute to objective firm outcomes (“objective research”), whereas micro-CSR research focuses on individual-level behavior, investigating how stakeholders’ perceptions of CSR may influence their attitude, emotions, and behaviors (“subjective research”). According to Bouraoui et al. [25], prior to 2015, the majority of research on CSR concentrated on the macro-level (institutional and firm-levels), including the efficacy of CSR programs on corporate financial performance (CFP) [24,26]. However, research findings for the relationship between CSR and CFP are mixed.

Some scholars, on the one hand, argue that CSR has a beneficial effect on CFP because CSR practices are beneficial for stakeholders [27,28] and therefore firms are more likely to obtain support from their stakeholders, especially customers and employees [11]. On the other hand, several scholars indicate a negative relationship because CSR practices incur expenses and decrease profitability [29]. Moreover, the stakeholders who benefit from CSR practices in firms may not necessarily recognize the benefits they receive, or these benefits may not flow to firms immediately [11,30]. Other scholars who observe non-significant effects, argue that CSR does not boost profitability, but also does not harm it [31,32,33,34,35], that is, the benefits and drawbacks of CSR appear to balance one another out. Finally, a few scholars also found an inverted U-shape relationship between CSR and CFP [7,36,37]. Specifically, a moderate level of CSR is linked to high performance, but lower and higher CSR is linked to poorer financial results. In conclusion, the empirical research on CSR-CFP does not present conclusive results. It is still unclear why sometimes CSR is unable to boost firm performance. Furthermore, the question of why stakeholders cannot recognize organizations’ CSR practices has yet to be answered.

This puzzle motivates research exploring the impact of individual stakeholders’ perceptions of CSR practices. Given that employees are considered the most important stakeholders within organizations, organization research investigates how employees respond to organizational CSR strategies [38], in a shift of focus from the macro-level to the micro-level. Most micro-level CSR research employs a subjective method (self-report) for individual investigation. In addition, qualitative research emphasized the importance of understanding various employees, relationships, conflicts, human resource managements practices, and emotions associated with CSR [39,40]. Jones [41] argued that in micro-CSR studies, the success of CSR for financial performance is contingent on individual stakeholders’ perception of cooperative social value via CSR programs. In other words, firms’ CSR cannot be entirely successful when individuals fail to acknowledge it. Therefore, scholars have explored various methods for measuring CSR perception in micro-CSR research to effectively and efficiently analyze employees’ perceptions of value.

2.2. Perceived Corporate Social Responsibility and Affective Organizational Commitment

2.2.1. Affective Organizational Commitment

In micro-CSR studies, employees are widely considered to be one of the most central stakeholders for organizational value creation. According to Onkila and Sarna [38], the research is dominated by studies indicating how employees should be involved in CSR (94 studies) and studies concentrating on the advantages of employee involvement in CSR (131 studies), with far fewer studies addressing disparities in employees’ perceptions of CSR (48 studies). The research on the outcomes of perceived CSR is thus also limited.

Organizational commitment refers to the connection or association between a person and the organization in which they work. There are three forms of organizational commitment, which are as follows: affective, continuity, and normative commitment [42]. Employees’ affective organizational commitment (AOC) has been a commonly studied variable in analyzing the success of CSR practices and employees’ internal psychological mechanisms. AOC incorporates employees’ emotional attachment, identification, and participation, indicating the extent to which employees want to work in their organizations [43]. Employees’ AOC therefore has the potential to boost job performance and well-being while reducing absenteeism and the likelihood of leaving [43]. Employees who are committed are more likely to engage in “out-of-role” activities such as innovation and invention, which allows enterprises to stay competitive [44], suggesting that successful firms rely on a robust employee–employer emotional relationship (AOC).

2.2.2. The PCSR–AOC Relationship

Several studies have investigated the psychological factors associated with employee-related attitudes and behaviors involving AOC, and have demonstrated how employees perceive, understand, respond to, and engage in CSR (see

Table 1. Summary of prior research into PCSR–AOC.

Ref.	Year	Theory	Mediators	Moderators
[12]	2009	Social identity theory	NA	Importance of CSR
[52]	2014	NA	NA	Collectivism Masculinity orientations
[18]	2014	Social exchange theory Social identity theory	Organizational trust Organizational identification	NA
[50]	2016	Social exchange theory	Employee CSR values Employee engagement	NA
[53]	2017	Commitment-trust theory	Reputation	NA
[54]	2018	Social identity theory Social learning theory	Organizational Pride	Collectivism Person–Organization Fit
[51]	2018	Social identity theory Social exchange theory	Meaningfulness of work Perceived organizational support	NA
[14]	2019	Cue consistency theory Social identity theory	Perceived external prestige Organizational identification Work meaningfulness	NA
[55]	2020	Social identity theory	Organizational trust Organizational identification	Importance of CSR
[56]	2021	Social identity theory	Procedural justice Organization-based self-esteem	NA
[57]	2021	Social identity theory	Job satisfaction	NA
[15]	2021	Role theory	Intrinsic motivation	Job level
[58]	2021	Deontic justice theory Social exchange theory	Organizational Trust	Importance of CSR
[13]	2022	Attribution theory	Organizational CSR climate Employees’ CSR-induced intrinsic attribution	NA
[59]	2022	Social identity theory Theory of met expectations	Organizational Identification Organizational Commitment	NA

). Social identity theory and social exchange theory are the two most widely employed theories to illustrate the mechanisms underlying how PCSR influences AOC. Most employees build commitment to their organization through identification and exchange procedures. Social identity theory is a broad theory about the psychological basis of intergroup discrimination [45]. According to social identity theory, individuals identify themselves as members of particular communities. When people believe that their ingroup identity separates them from outgroups, this identification boosts their self-esteem. It fosters a perception of belonging to the group and, as a result, interprets the firm’s achievements and failures as personal achievements and failures [46]. Social identity theory is frequently used to explain psychological mechanisms applying to individual employees, such as organizational trust and organization identification [18,41,47].

Social exchange theory is also widely used to explore the mechanisms through which PCSR affects AOC. Social exchange theory considers social interaction as an exchange of benefits and costs, in which reciprocity is the basis for establishing and maintaining social relationships [48]. The concept explains how interacting parties build up and profit from economic, social, and psychological resources to form long-term social connections. Many positive benefits of constructive exchange contacts in the workplace have been proven through organizational research [49]. A social exchange approach is more likely to be used to explain the relational mechanisms based on interaction between stakeholders and their organizations, such as employees’ perceived organizational support and employee engagement [18,50,51].

While previous studies on the PCSR–AOC relationship illustrated the mechanisms underlying how PCSR may influence employees’ AOC, most of these studies demonstrated a positive relationship. A knowledge gap in the existing research remains regarding the boundary conditions that enhance or weaken the positive effect of PCSR [29]. For business decision makers, the issue of “when” CSR efforts improve or impair positive employee attitudes is essential. Moreover, the majority of prior studies separately explored the boundary conditions according to either the social identity mechanisms or the social exchange mechanisms. Yet, these two mechanisms are most likely to function simultaneously when influencing employees’ AOC. While the existing literature identified several moderators at the organizational level (e.g., the importance of CSR) [12], work level (e.g., Job level), and individual levels (e.g., employees’ cultural collectivism, masculine tendencies) [52], we still lack a holistic view of the conditional boundaries that influence the social identity and social exchange mechanisms simultaneously. Therefore, this study systematically explored the boundaries of PCSR–AOC from various levels.

Moreover, although the current research demonstrated that CSR has a positive feedback impact on employee AOC, few studies examined the circumstances in which the positive PCSR–AOC relationship is lessened or may even have a damaging impact. To date, there are few studies on the negative side of PCSR–AOC. As a result, this study focused on exploring the negative boundaries of PCSR–AOC, thus enriching social identity and social exchange theory from a broader perspective.

2.2.3. Moderating Framework in PCSR–AOC

Having identified the above research gap, this study explored the boundaries of PCSR–AOC based on the “C-S-R Concerns” model proposed by David and Deborah [16] to construct a more comprehensive understanding of PCSR–AOC moderating mechanisms. The model aimed to capture several concepts and aspects of the psychology of CSR (e.g., psychological needs and other motivational mechanisms, signal-based mechanisms, attribution processes, cultural values, individual differences, contextual variables, etc.). The first dimension is a personal attribute (i.e., caring-based concern), which primarily represents a person’s ethical values. Individuals view the importance of an organization’s CSR activities differently based on variations in their experience and social background. The second dimension is the employee’s self-perspective (i.e., self-based concern), which essentially represents how the employee views the transactional value provided by the employer to society. The last dimension is the relationship perspective (i.e., relational-based concern), which includes the relationship of the employee–employer and the relationship between various organizational stakeholders.

All three of these dimensions determine the value an employee places on their organization’s CSR practices, but are based on different rationales [16]. Specifically, caring-based concerns describe employees’ moral acknowledgement of the CSR practice itself. Thus, the underlying rationale is the ethic-motive, reflecting the extent to which CSR is the ‘right thing to do’. The more employees have caring-based concerns, the more they believe their organizations should conduct CSR practices to contribute to society. Self-based concerns are related to the instrumental-motive, which reflects the extent to which employees perceive their organization to fairly treat stakeholders. The perception is normally influenced by the ability of an organization to conduct CSR or the culture and actions of the organizations. The more employees feel they are being treated fairly, the more they are likely to commit to organizations. Relational-based concern reflects the relational motive, indicating the extent to which employees believe they are connected with other stakeholders and thus influenced by other stakeholders. The more relational concerns employees have, the greater they recognize that CSR practices can contribute both to them as individuals and/or to their organization. These three dimensions consequently influence how employees identify the value created by CSR practices and how they perceive the value exchange process, and then influence the effectiveness of the social identity and social exchanges mechanisms. Therefore, this study utilized the “C-S-R Concerns” model to explore the boundaries of PCSR–AOC systematically.

3. Hypotheses Development

3.1. PCSR and AOC

Consistent with prior studies, we utilized social identity and social exchange mechanisms to explain the relationship between PCSR and AOC. As CSR generates external praise for companies, enhancing corporate image and social influence, from an employee perspective, it also adds value to themselves if they perceive they are an in-group member of their organizations, which indirectly improves employee identification and increases commitment. For employees, especially young employees, the value of work goes far beyond making a living; it also has a crucial impact on personal self-fulfillment. Thus, according to social identity theory, when a firm creates value for its stakeholders, employees will gain in-group identification as a result of their membership in the firm [46]. Along with identifying the firm’s behavior, employees reevaluate the firm itself and the value of their work. Corporate involvement in CSR practices can be a means by which employees perceive they are contributing to society through their work. Ultimately, this recognition of the organization’s behavior itself and the perceived value of their work will positively impact employees, motivating them to form a more positive AOC towards the company, which in turn will further stimulate a cooperative relationship between employees and the company. Ultimately, this acknowledgment of the organization’s behavior and the perceived worth of their job will support employees in developing a more favorable AOC toward the company.

Further, organizations’ CSR actions motivate reciprocal behavior and directly increase employee commitment. Based on social exchange mechanisms, CSR practices convey to employees that their employer is caring, kind, and benevolent. A sense of justice based on fairness motivates them to reciprocate the firm’s positive actions [29]. In addition, employees are also more likely to appreciate their company’s efforts since they rationally believe that the immediate benefactors of internal CSR programs are the employees themselves. Employees that provide positive feedback are seen as fulfilling their commitments in exchange for the benefits they obtain from the organization. The influence of CSR programs on the balance of organizational interests in the employee mindset results in a feeling of affective commitment. Thus, social exchange theory also could be a facilitator of the PCSR–AOC relationship. Thus, the following hypothesis was developed:

Hypothesis 1.

PCSR positively impacts employee AOC.

3.2. C-S-R Concerns Based Moderating Effects

This study extracted three variables from the “C-S-R concern” model proposed by David and Deborah [16] to examine the moderating variables (care-based concern: moral identity; self-care: corporate ability; and relational concern: the importance of CSR). The first variable is moral identity, which is a personal attribute that indicates whether an individual employee “cares” about society and adheres to social values. The second variable is corporate ability, which measures employees’ “self” value assessment, examining if the organization’s social value to employees or other stakeholders fairly aligns with its abilities. Finally, the third variable is the importance of CSR, which evaluates how employees perceive whether CSR practices to stakeholders inside and outside organizations are important to organizational growth. Employees assess CSR for rationality, which determines their attitude toward corporate commitment. As a result, these three variables may be treated from multiple angles within the “C-S-R Concerns” framework, spanning from “internal stakeholders-organization-external stakeholders.” [16].

3.2.1. Moral Identity (MI)

At the level of individual traits, moral identity is a system that exists at the confluence of moral development and identity formation. It is regarded as a source of ethical motivation since it connects moral reasoning to actual behavior [60]. In other words, individuals who already have a higher moral identity seem to be more motivated to do what they believe is ‘right’ and generate a long-term affective commitment. Most employees do not always act rationally; people’s motivation or inducement to execute a particular behavior is determined by their assessment of the activity’s expected consequence and their estimation of the likelihood of obtaining that goal [61]. Highly ethical employees have a relatively high moral standard for themselves and, as a result, have a greater ethical expectation of anything around them. Hence, employees with a higher moral identity may set higher expectations on organizational social responsibility and are more likely to believe that their organization’s CSR practices are just ‘doing the right thing’, rather than doing an extra good for the wider public. The higher expectation may consequently reduce their identification with the organization when their expectations of the firm surpass the company’s actual worth to society. Meanwhile, higher expectations may also hinder reciprocal behavior from employees even though their organizations engage in CSR practices. This is because employees are likely to take it for granted that their organization is engaging in CSR as they believe their organizations should do. When the actual CSR is insufficient to meet the benchmarks of high socially identified workers, the positive PCSR–AOC link is diminished due to their personal characteristics. Thus:

Hypothesis 2.

MI weakens the relationship between PCSR and employee AOC.

3.2.2. Corporate Ability (CA)

Corporate ability encompasses more than just a summary of an organization’s resources. It considers both hard and soft aspects, such as human resources, knowledge base, learning skills, organizational culture, communication strategy, management-leadership, organizational techniques, values, and so on [32]. While firms with greater ability have more slack and extra resources to contribute to society it also suggests a greater responsibility. Thus, when corporate ability is stronger and industry influence is higher, employees may feel it is fair for their organizations to take greater responsibility toward society. It will eventually raise employees’ expectation of their organization’s responsibility, which in turn will diminish their identity as an ingroup member and reciprocal behavior because they may think their organization should contribute more to its stakeholders because it has the ability to do so. As a result, increased corporate ability impairs the positive PCSR–AOC relationship. The following hypothesis was therefore developed:

Hypothesis 3.

CA weakens the relationship between PCSR and employee AOC.

3.2.3. Importance of CSR (ICSR)

Employees who believe their organizations are socially influenced by a wide range of stakeholders may place high importance on engaging in CSR. The priority given to the value provided by CSR is determined by stakeholder demands and human attributes [62]. The greater the tendency of firms to attribute their success to support from stakeholders, the more likely they are to conduct CSR practices for the purpose of self-interest [63]. Previously, two types of CSR were categorized based on motivation: “other-centered” and “self-centered.” According to Vlachos et al. [64], various CSR incentives can significantly impact the attitudes of key stakeholders. Compared with the “self-centered” CSR, which is motivated by self-interest, employees are more likely to have in-group identity to “other-centered” CSR practices. They are also more likely to exchange their commitment to their organizations if their organizations engage in CSR aimed at internal stakeholders because employees believe that CSR efforts are authentic. When they perceive CSR practices from the organization as in their interests, rather than because employees are important stakeholders to the organization and therefore aimed at strategically making use of them, PCSR may not increase employees’ AOC. Therefore, the following hypothesis was formed:

Hypothesis 4.

ICSR weakens the relationship between PCSR and employee AOC.

The conceptual framework of the research, incorporating the hypotheses developed, is illustrated in Figure 1.

4. Methods

4.1. Sample and Procedure

Full-time white-collar individuals working in high-tech companies in China were surveyed for this study. We targeted the high-tech industries to serve the purpose of this study. On the one hand, there is a high turnover rate of employees in high-tech companies and therefore organizational commitment and loyalty is important. On the other hand, with the emergence of the data-driven era in China, high-tech industries have experienced a significant growth rate without requiring the provision of benefits for their employees, especially in internet companies. There are numerous international companies located in the major cities of China, such as Amazon, Alibaba, Tencent, Shopee, Bytedance, Lala-Move, Baidu, and JD, etc. Although our study was conducted in China, the sample incorporated both domestic and international companies with well-established CSR practices in different sectors, which is beneficial for enhancing the generalization of our findings.

We first approached 14 high-tech companies located in the following four major Chinese cities: Beijing, Shanghai, Hangzhou and Shenzhen (8 in the internet industry, 3 in the electronics and communication industry, and 3 in the pharmaceutical industry), and 9 allowed us to collect data. Four out of nine sample companies are international companies.

Prior to collecting data for this study, representatives from various high-tech enterprises in China were approached to obtain approval and clarify the survey’s purpose using a sample questionnaire. We initially collected 378 questionnaires. For each returned questionnaire, we checked the time respondents took to complete the survey and determined if there were a large number of missing responses. After discarding the invalid questionnaires, we included a total of 326 questionnaires in our final analysis. The demographic characteristics of the respondents were described in

Table 2. Demographic characteristics of the respondents.

Demographic Characteristics	Percentage
Gender
Female	49
Male	51
Age
18–28 years	19
29–40 years	57
More than 41 years	24
Level of education
Not completed bachelor	39
Bachelor	48
Master	13
Management level
Nonmanagement	64
Lower level managers	29
Middle and upper middle managers	7
Job tenure
0–5 years	28
6–10 years	50
More than 10 years	22

Note: Percentage value are rounded off.

.

The questionnaire design included a cover letter explaining the study’s purpose and provided directions for filling out the questionnaire. The demographic variable section asked for information about the participants and the questions in the following section were aimed at identifying workers’ CSR views, AOC, and moderating factors (moral identity, corporate ability, and importance of CSR). Participants were asked to score items on a five-point Likert scale, with 1 indicating total disagreement and 5 indicating strong agreement. When measuring AOCs, several studies in the literature used supervisor or colleague measures rather than self-reported measures. However, this study used the self-reported measure in light of previous studies [65] suggesting that individuals have a greater understanding of their AOCs than their supervisors or workmates. Participation was anonymous and voluntary, and to eliminate the risk of common method bias there were no right or wrong responses [66]. In addition, Harman’s single factor methodology was utilized to conduct a statistical analysis of common method variance in the research. We performed an exploratory factor analysis without rotation on all of the items for the research variables to see if a single general factor could explain most of the variation [67]. Finally, the overall cumulative variance of the one component that resulted from this assessment was 26.4%, indicating that standardization was not an issue in this study. All scales utilized in this study were developed and validated from prior research. The next section introduces the measure of each variable.

4.2. Measurements

4.2.1. Dependent Variable

Affective Organizational Commitment (AOC): Employees’ positive views toward the organization were measured by whether they agree with the statement “this organization has considerable personal worth for me.” Based on Dutton et al.’s [68] AOC scale, we incorporated four items in this study (see Appendix A). Cronbach’s alpha value was 0.91.

4.2.2. Independent Variable

Employees’ Perceived Corporate Social Responsibility (PCSR): We selected Arnold et al.’s [69] employee CSR perception scale based on stakeholder theory [3,70]. Respondents were asked to rate how much CSR their companies engage in for social and non-social stakeholders, employees, customers, and government. Cronbach’s α for the scale was 0.97.

4.2.3. Moderators

Moral Identity (MI): Moral identity emphasizes that employees’ cultural attributes will significantly influence employees’ responses to corporate CSR initiatives. This study integrated Aquino et al.’s [71] scales for employees’ moral identity. Cronbach’s α for the scale was 0.92.

Corporate Ability (CA): David and Deborah [16] emphasize that the moderator of self-centered CSR emphasizes the transactional relationship between employees and the company and highlights the influencing factors that employees can benefit and gain from the organization. In the corporate ability scales, previous researchers attached three impact factors to this exchange relationship [72]. Cronbach’s alpha for the scale was 0.87.

Importance of Corporate Social Responsibility (ICSR): The importance of the CSR aspect emphasizes employees’ attitudes towards the values of CSR practices toward organizations. This study used Etheredge’s scale [73] to investigate the employees’ perception of importance through four items. Cronbach’s α value was 0.86.

4.2.4. Covariates

This indicated five sociodemographic characteristics that needed to be taken into consideration in the data analysis to exclude potential confounding influences (i.e., gender, age, level of education, management level, tenure).

All the aforementioned measures are presented in Appendix A.

4.3. Common Method Bias

Common method bias (CMB) may have occurred because the study’s independent and dependent variables were self-reported and collected using the same questionnaire. Due to restrictions in the research design, it was difficult to remove the same-method deviation entirely, but it is feasible to determine if this divergence produces mistakes [74]. To test for the presence of CMB, we used Harman’s one-factor technique and common latent factors. First, we used Harman’s one-factor test, which entailed performing an unrotated factor analysis on all variables. CMV is considered a major concern if a single factor appears or if one general factor explains most of the covariance in the independent and dependent variables [74]. The test revealed a total explained variance of 48.36%, indicating that CMV is not a significant issue in this investigation. Similarly, the common method latent factor provided just 11.3% of the common factor for all independent, moderating, and dependent variables, much less than the acceptance criterion of 25%. As a result, CMB posed no significant risk to our data.

5. Results

Covariance-based structural equation modeling was performed using AMOS 22.0 (CB-SEM). To begin, a confirmatory factor analysis (CFA) was used to determine the measurement model and confirm the data’s reliability and validity. Then, to assess causation, we utilized SPSS 25 to run a hierarchical regression analysis.

5.1. Reliability and Validity of the Measures

Table 3. Confirmatory factor analysis.

Construct	Item	Mean	SD	Loadings	AVE	CR
PCSR	PCSR1	3.79	1.282	0.852	0.729	0.97
	PCSR2	3.83	1.284	0.897
	PCSR3	3.84	1.259	0.840
	PCSR4	3.83	1.212	0.867
	PCSR5	3.76	1.249	0.840
	PCSR6	3.87	1.131	0.834
	PCSR7	3.78	1.309	0.859
	PCSR8	3.79	1.253	0.874
	PCSR9	3.78	1.288	0.836
	PCSR10	3.73	1.275	0.835
	PCSR11	3.80	1.275	0.864
	PCSR12	3.92	1.238	0.846
AOC	AOC1	3.71	1.349	0.901	0.782	0.935
	AOC2	3.74	1.249	0.871
	AOC3	3.65	1.304	0.885
	AOC4	3.74	1.292	0.880
MI	MI1	3.78	1.291	0.912	0.810	0.944
	MI2	3.73	1.308	0.894
	MI3	3.70	1.359	0.892
	MI4	3.74	1.227	0.903
CA	CA1	3.68	1.309	0.910	0.797	0.922
	CA2	3.68	1.256	0.894
	CA3	3.75	1.223	0.874
ICSR	ICSR1	3.75	1.283	0.883	0.783	0.935
	ICSR2	3.80	1.261	0.901
	ICSR3	3.80	1.324	0.897
	ICSR4	3.76	1.158	0.859
Model fitness	χ2 = 214.91, df = 118, CFI = 0.976, GFI = 0.936, TLI = 0.934, SRMR = 0.051, RMSEA = 0.061

reveals composite reliability values of at least 0.922, Cronbach’s alpha values of at least 0.872, and average variance extracted (AVE) values of a minimum of 0.729 for each construct, producing promising reliability and convergent validity [75,76]. The 2/(df) value was less than 2.5 (214.91/118 = 1.82), demonstrating an excellent fit [77]. The Tucker–Lewis index (TLI) and comparative fit index (CFI) were both more than 0.9 (0.934 and 0.936, respectively), demonstrating an excellent model fit [78]. The root means of the square error of approximation (RMSEA) and standardized root mean square residual (SRMR) values were both below 0.08, meaning that the model fit was satisfactory [79]. The results correspond to the convergent and discriminant validity variables, especially when combined.

The means, standard deviations, and Pearson correlation coefficients for the analysis are presented in

Table 4. Descriptive statistics and correlations ^a.

	Mean	SD	1	2	3	4	5
1. Gender b	1.49	0.50	1
2. Age c	2.05	0.65	−0.032	1
3. Education d	1.73	0.67	0.071	−0.144 **	1
4. Man_Level e	1.43	0.62	−0.056	0.134 *	−0.130 *	1
5. Job_Tenure f	1.94	0.70	0.047	0.377 **	0.032	0.043	1
6. PCSR	3.81	1.07	−0.015	0.009	0.139 *	−0.175 **	0.031
7. AOC	3.71	1.15	−0.029	−0.008	0.139 *	−0.162 **	0.006
8. MI	3.74	1.17	0.037	−0.007	0.071	−0.117 *	0.045
9. CA	3.71	1.13	−0.083	0.082	−0.035	−0.076	0.050
10. ICSR	3.78	1.11	−0.015	0.100	0.129 *	−0.090	0.033
	Mean	SD	6	7	8	9	10
1. Gender b	1.49	0.50
2. Age c	2.05	0.65
3. Education d	1.73	0.67
4. Man_Level e	1.43	0.62
5. Job_Tenure f	1.94	0.70
6. PCSR	3.81	1.07	1
7. AOC	3.71	1.15	0.371 **	1
8. MI	3.74	1.17	0.374 **	0.404 **	1
9. CA	3.71	1.13	0.301 **	0.297 **	0.438 **	1
10. ICSR	3.78	1.11	0.407 **	0.370 **	0.413 **	0.306 **	1

Note: ^a n = 326, coefficient α reliability estimates are reported on the main diagonal; * p < 0.05, ** p < 0.01; ^b Gender (1 = female, 2 = male); ^c Age (1 = 18–28 years, 2 = 29–40 years, 3 = More than 41 years); ^d Level of Education (1 = Not completed bachelor, 2 = Bachelor, 3 = Master); ^e Management level (1 = Nonmanagement, 2 = Lower level managers, 3 = Middle and upper middle managers); ^f Job tenure (1 = 0–5 years, 2 = 6–10 years, 3 = More than 10 years).

. A Cronbach’s reliability analysis was performed for all of the variables in this study, as given in parentheses, and each value for reliability was above 0.70. (PCSR: 0.97, MI: 0.92, CA: 0.87, ICSR: 0.86, AOC: 0.91). Overall, these findings indicate a sufficient level of reliability.

5.2. Hypothesis Testing Results

The results are shown in

Table 5. Hierarchical moderated regression analysis results.

	Model 1 (H1)		Model 2 (H2)		Model 3 (H3)		Model 4 (H4)
Variable	B	SE	B	SE	B	SE	B	SE
Constant	2.393	0.414	3.266	0.413	3.186	0.400	3.231	0.412
Control variables
Gender	−0.078	0.119	−0.077	0.112	−0.096	0.110	−0.054	0.112
Age	0.026	0.100	0.002	0.094	0.044	0.093	−0.008	0.094
Education	0.144	0.091	0.031	0.088	0.031	0.086	0.049	0.087
Man_Level	−0.176	0.098	−0.1	0.093	−0.098	0.092	−0.057	0.094
Job_Tenure	−0.011	0.092	−0.051	0.087	−0.091	0.086	−0.082	0.087
Independent variables
PCSR	0.367 ***	0.057	0.259 ***	0.058	0.248 ***	0.055	0.225 ***	0.058
Moderating effects
PCSRXMI			−0.226 ***	0.040
PCSRXCA					−0.298 ***	0.042
PCSRXICSR							−0.272 ***	0.043
R2 within	0.139		0.235		0.257		0.235
F static	9.766 ***		15.295 ***		17.033 ***		15.257 ***
VIF	1.05		1.302		1.221		1.345

Note: The number of observations (n) = 326. The regression model coefficients (estimators) are listed in the table; standard deviations are provided as follows. *** p < 0.001.

. The variance inflation factors (VIF) for each model provided were less than 5, demonstrating that the findings were not skewed by multicollinearity [75]. All adjusted R² values were higher than the models’ acceptable limit. Model 1 shows that PCSR is significantly and positively associated with AOC (β = 0.367; p < 0.001), which supports H1. Module 2 shows the that the interaction items between PCSR and MI are negative, suggesting that MI negatively influence the relationship between PCSR and AOC (β = −0.226; p < 0.001). Model 3 demonstrates that the interaction item between PCSR and CA is negative, which suggests that CA also negatively moderates the relationship between PCSR and AOC (β = −0.298; p < 0.001). Model 4 is the result of the moderating effect of ICSR, where the interaction item between ICSR and PCSR is significant and negative (β = −0.272; p < 0.001). The moderating effects are also plotted in Figure 2, Figure 3 and Figure 4. Interestingly, Figure 2, Figure 3 and Figure 4 shows that the effect of PCSR on AOC even becomes negative when MI, CA, and ICSR are at a higher level. Therefore, H2–H4 are supported.

6. Discussion and Conclusions

6.1. Discussion

The objective of this article was to explore when CSR practice fails to increase employees’ AOC. Although prior research has shown that CSR can positively impact employees’ AOC, the possible negative moderating factors have not been considered. This research analyzes the boundary conditions on the relationship between PCSR and AOC based on the C-S-R concerns model to address this gap. The significant positive results are consistent with previous studies [43], emphasizing that when firms conduct CSR practices, employees’ favorable views of the company motivate their AOC. The firm’s attributes as a benevolent and caring organization are reflected in these CSR practices. On the one hand, this is in line with social identity theory, which claims that employees strive to maintain their organization’s image to align their identity and values. On the other hand, employees are eager to establish more emotionally binding ties in exchange for a high-value perceived image of the organization according to social exchange theory. As a result, employee behavior is positively influenced by PCSR, which is crucial to the firm’s performance.

However, the results also show that the positive employee PCSR–AOC link is weakened by a higher level of moral identity, corporate ability, and greater importance of CSR as perceived by employees. This is because these ethical, instrumental, and relational motives may encourage employees to set high expectations of the organization’s CSR activities, and they are more likely to believe their organizations should take responsibility to benefit society. Therefore, their organizations are less likely to increase their employees’ AOC by conducing CSR practices.

6.2. Theoretical Implications

This research contributes to the research on the consequence of CSR in several ways. First, this research enriches the literature on the PCSR–AOC relationship by systematically exploring conditional boundaries based on the C-S-R model. Although previous studies on the PCSR–AOC relationship explored several moderators, these moderators were examined separately as to the way they influence either social identity or social exchange mechanisms, and there is a lack of holistic and systematic analysis on moderating effects [12,80]. Adopting the C-S-R model, this study explored the boundary conditions based on employees’ ethic-, instrument- and relational-motives that may influence their expectations of the CSR practices conducted by their organizations. Interestingly, our findings show that CSR practices may lose their ability to incentivize employees when employees intrinsically believe that CSR is the right thing to do, when they believe their organizations have the ability to engage in more CSR practices, and when they recognize that CSR practices are also beneficial for their organizations.

Furthermore, prior studies mostly focused on the bright side of CSR, for example, how CSR can increase positive organizational behaviors and which factors may enhance the positive mechanism of CSR [18,51]. By contrast, this study pays attention to the factors that negatively influence the positive impacts of PCSR on AOC. The findings also specify the boundary conditions for the application of social identity theory and social exchange theory in the relationship between PCSR and AOC. When the levels of IM, CA, and ICSR are higher, PCSR is less likely to encourage employees’ in-group identity and reciprocal behavior in their relationship with their organizations.

Finally, our study also enriched the CSR literature in general by providing new insights into the unsolved puzzle of why organizations cannot always benefit from engaging in CSR [31,32]. The results suggest that CSR cannot play a positive role if organizations’ stakeholders fail to recognize the value of CSR activities, which in turns means stakeholders have less positive behavior in their relationship with organizations. Instead, CSR activities may bring about negative impacts, especially when organizations fail to satisfy stakeholders’ expectations of their CSR practice.

6.3. Managerial Implications

This study also has practical implications for organizations. Given the growing relevance of corporate investment in CSR programs [81], clearly understanding how employees build positive attitudes and actions toward the organization in various circumstances is crucial. When a company’s decision-makers are aware of the boundaries of the role of CSR, they may be able to make more successful strategic corporate management decisions. First, the results of this study provide insight into how firms might recruit staff with the requisite moral principles that make it easier for companies to acquire internal employee recognition. Furthermore, examining the consistency or inconsistency between employee perceptions and reality should assist practitioners in integrating their practices through forums, internal reports, training, and targeted culture development and communication [29]. Finally, this research may be used to inform CSR efforts by providing insight into how much a firm should invest in CSR, how it may balance internal and external social value, and how to reconcile social value with corporate financial performance. To put it another way, “C-S-R Concerns” serve as a diagnostic tool and a practical assessment of the impact of CSR programs in various circumstances. Consequently, businesses may more be able to avoid a one-size-fits-all management strategy by tailoring their methods based on the assessment of their employees’ opinions of their CSR activities

6.4. Limitations and Future Research

As for all studies, this study had some limitations that can be addressed by future research. First, this study was conducted in China using a sample of Chinese high-tech industry employees. Although the sample incorporated both domestic firms and international firms located in China and the high-tech industry includes a wide range of sectors, it may still potentially influence the generalization of our findings. Thus, it is recommended that future studies may involve more industries across countries. Moreover, although we conducted several tests to rule out common method bias, this study employed a self-report survey method. The study also did not control for potential endogeneity issues that could find a reverse effect for employees’ AOC and the influence on their perception of CSR [82]. Future studies could verify the findings by using data from multiple sources (e.g., from supervisors), using data collected at different times, and also using objective data.