This study aims to investigate the role of corporate social responsibility (CSR) in reducing employees’ negative emotions. Based on social information processing, a new conceptual framework was developed and tested in the Indian context.
Cross-sectional design was used in the study. A total of 240 managers of financial institutions were surveyed through a self-administered questionnaire, and the responses were collected using a convenience sampling technique.
The findings indicated that CSR induces compassion and reduces employees’ negative emotions. Contrary to expectations, compassion did not emerge as a mediator in the relationship between CSR and negative emotions.
The findings contribute to the literature by scrutinizing the persuading influence of CSR on employees’ negative emotions directly and indirectly through compassion. Furthermore, this study also tested the moderating effects of positive emotion in the relationship between CSR and negative emotions. The linkage between CSR and negative emotions was established by introducing the social information processing mechanism, adding another novelty to the literature.
1. Introduction
Corporate social responsibility (CSR) is the practice adopted by a firm beyond its economic interest to meet all its stakeholders’ psychological and physiological needs (Turker, 2009). Companies are obliged to perform CSR activities, and this obligation has attracted the attention of researchers and policymakers. Research has been mainly conducted on the positive impact of CSR at different levels. For instance, at the macro level, a study has been carried out on the effects of CSR on enhancing sustainable performance (Latif et al., 2022), financial performance (Orlitzky et al., 2003), reputation (Verschoor, 1998), competitive gain (Mai et al., 2021) and effective management practices (Waddock and Graves, 1997. At the micro level, the CSR practices of the firm may have a huge impact on the customers and employees. Previous CSR research focusing on customers indicated that CSR actions influence loyalty (Martínez and Del Bosque, 2013), positive behavioural intention (Ahn and Lee, 2019), consumer engagement (Fatma and Khan, 2023) and perceived quality (Bello et al., 2021).
Similarly, at the micro level, many studies have investigated the influence of CSR on employees’ positive behaviour. Some studies have investigated the role of CSR in influencing positive behaviour, for example, work engagement, organizational identification, organizational commitment, organizational citizenship behaviour and job satisfaction (Aguinis and Glavas, 2012; De Roeck et al., 2016; Kim et al., 2010; Gao and He, 2017; Glavas and Kelley, 2014). Some recent studies are also in line with earlier research that has examined the influence of CSR on positive employee behaviour, e.g. organizational citizenship behaviour, compassion and organizational commitment (Sarfraz et al., 2021; Nazir and Islam, 2020; Gahlawat and Kundu, 2021). These studies mainly investigated how CSR induces positive work behaviour. Unfortunately, limited studies have examined how CSR can induce negative behaviour, such as cheating (Luan et al., 2022), turnover intention (Lin, and Liu, 2017) and employee deviance (Evans et al., 2010). It is evident from the existing literature that there is insufficient data concerning CSR and negative work behaviour. The present study investigated the impact of CSR on negative emotions. Prior literature revealed that CSR plays an important role in prompting emotional responses among employees. The findings of Khan et al. (2021) argued that both external and internal CSR actions induce an emotional attachment of employees with their employer. Since the employee is the most important internal stakeholder for every employer, the employer should pay special attention to strengthening the emotional bond between the two (Budhiraja and Yadav, 2020). The various internal and external CSR initiatives strengthen the emotional bonding of employees with their employers (Budhiraja and Yadav, 2020). Subba and Kumar (2018) suggested that any initiative beyond profit maximization, i.e. CSR to meet the needs of the stakeholders, induces positive emotions among employees. However, this CSR research focusing on emotions did not investigate how CSR influences negative emotions. Based on social information processing (SIP), the present study investigated the relationship between CSR and negative emotions. Theorists suggest that “… individuals, as adaptive organisms, adapt attitudes, behaviour, and beliefs to their social context and to the reality of their own past and present behaviour and situation” (Salancik and Pfeffer, 1978, p. 226). Earlier studies used SIP to establish the linkage between CSR and career satisfaction and voice behaviour (Ilkhanizadeh and Karatepe, 2017), employee cheating (Luan et al., 2022) and workplace deviance (Mohammed Sayed Mostafa and Shen, 2020). However, no evidence exists to establish the linkage between CSR and negative emotions through SIP.
CSR actions may influence emotions through some variables. Well-being acts as a mediator between CSR and green behaviour (Ahmed et al., 2020). The linkage of CSR with green behaviour and well-being is mediated partially by identification and trust (Su and Swanson, 2019). Sarfraz et al. (2018) suggested that organizational justice partially mediates the linkage between CSR and employee output. Personal environmental norms and green employee behaviour partly mediate the relationship between CSR and employee engagement (Nasir Ansari and Irfan, 2023). Subba and Kumar (2018) conducted a study among banking employees and revealed that compassion emerged as a mediator in the relationship between CSR and positive emotions. Ali et al. (2020) found that CSR and employee performance relationships are partially mediated by employee engagement. Sarfraz et al. (2021) found organizational justice as a mediator between CSR and organizational citizenship behaviour. Surprisingly, no empirical research has identified compassion as a mediator between CSR and negative emotions.
The literature revealed that the relationship between CSR and employee output is moderated by numerous factors. Brammer et al. (2015) argued that CSR actions positively impact employees’ organizational identification in the presence of strong corporate ability rather than weak. The findings of Bouichou et al. (2022) suggested that under responsible leadership, employees display more engagement, commitment and organizational citizenship behaviours. Restoration of belief increased the impact on the relationship between CSR and safety behaviour (Zhang et al., 2021). Hameed et al. (2019) found that perceived organizational support moderates the relationship between CSR and organizational pride. Collectivism positively moderates the relationship between CSR initiatives and organizational identification, impact was stronger for the employees believing in collectivism than for employees believing in individualism (Farooq et al., 2014). However, these studies mainly investigated how different variables moderate the relationship between CSR and employee positive output.
The current research makes some significant contributions. First, the researcher conceptualized a new framework by introducing novel constructs through SIP, particularly in financial institutions, and empirically tested the framework. The researcher introduces SIP as a mechanism to investigate the relationship between CSR and negative emotions, thereby responding to the identified gaps associated with the mechanisms of CSR at a micro level (Hur et al., 2018; Aguinis and Glavas, 2012). Second, the current research captured the underlying gap of mediating effects, as suggested by Subba and Kumar (2018). Third, the current research tested the moderating effect of positive emotion between CSR and employees’ negative output. Finally, the current research may be useful for firms to reshape the behaviour of employees. Due to the stiff competition and high employee turnover rate faced by all industries, reshaping behaviour and attitudes becomes imperative for all organizations (Montani et al., 2020).
2. Literature review
2.1 Corporate social responsibility
CSR was first defined by Bowen (1953) as “the obligation to pursue those policies, to make decisions, or to follow those lines of action, which are desirable in terms of the objectives and values of our society” (p. 6). Other scholars have also attempted to define CSR. Davis (1960) defined CSR as:
“some socially responsible business decisions can be justified by […] having a good chance of bringing long-run economic gain to the firm, thus paying it back for its socially responsible outlook” (p. 70).
Carroll (1979) and Turker (2009) made significant contributions to the concept of CSR. Carroll (1979) conceptualized the framework of CSR based on four aspects, i.e. economic, legal, ethical and discretionary, while Turker (2009) conceptualized the CSR framework based on four components, namely, CSR to social and non-social stakeholders (CSR1), employees (CSR2), customers (CSR3) and government (CSR4). Backhaus et al. (2002) shared a similar view by demonstrating CSR as a multidimensional construct. The researcher used a four-dimensional construct of Turker (2009) to investigate the relationship between CSR and negative emotions.
2.2 Corporate social responsibility and negative emotions
To reduce negative behaviour among employees, organizations must become socially responsible. Researchers are rigorously attempting to understand CSR, as well as its causes and effects (Abbasi et al., 2024). A mounting amount of literature on CSR shows that organizations which are not socially responsible can have adverse effects on stakeholders. Some research measured the impact of irresponsible behaviour of an organization on employee negative output such as punishing behaviour (Abbasi et al., 2024), vengeance and boycotting behaviour (Krishna and Kim, 2022). Similarly, Hericher and Bridoux (2022) discovered that an organization’s irresponsible behaviour leads to punishing behaviour among employees. Employees may breach the data if they find their organization socially irresponsible (Xu et al., 2022). Irresponsible behaviour such as organisational injustice may be a major predictor of negative emotions, i.e. anger (Deng et al., 2023). Barclay et al. (2005) suggested that irresponsible attitudes of firms elicit anger among employees.
Similarly, the study by Lim and Shim (2019) discovered the same findings that socially irresponsible organizations elicit anger among employees. Employees may experience other negative emotions such as guilt in response to their organization’s irresponsible behaviour (Miranda et al., 2020). It is evident from the literature that the relationship between irresponsible behaviour and negative emotion has been established but to the best of the knowledge of the researcher the relationship between responsible behaviour of the organization and negative emotions has not been established.
2.3 Compassion
Sprecher and Fehr (2005) defined compassion as:
“an attitude toward other(s), either close others or strangers or all of humanity; containing feelings, cognitions, and behaviours that are focused on caring, concern, tenderness, and an orientation toward supporting, helping, and understanding the other(s), particularly when the other(s) is (are) perceived to be suffering or in need” (p. 630).
It brings intimacy among individuals, which in turn shapes and builds our social connection, both at home and in the workplace. Hence, it plays a significant role in creating and sustaining a conducive environment in our society (Nussbaum, 1996). If individuals become compassionate at the workplace, emotional connections among employees may build up and increase productivity (Dutton et al., 2002).
2.4 Emotions
According to Oatley (1992), “Emotions are mental states that usually occur in a particular context” (p. 76). Since, 1990, emotions have drastically attracted the attention of researchers (Fineman, 2004; Fineman and Gabriel, 1996). Larsen et al. (2001) categorized emotional states as positive and negative. Positive emotions constitute gratitude, sexual desire, contentment, hope, love, awe, amusement and pride, whereas negative emotions constitute disgust, embarrassment, anger, sadness, repentance, contempt and shamefulness (Fredrickson et al., 2003). In his study, Fredrickson (1998, 2000) suggested that negative emotions cultivate undesirable behaviour among individuals. Negative emotions may induce disidentification (Stanley et al., 2008), stress (Lilius et al., 2003) and conflict (Allen, 2001).
2.5 Theoretical background
2.5.1 Corporate social responsibility and negative emotions.
It is vital to understand employee emotions as they influence their work behaviour. Studies have found that effective management of emotions reduces negative employee outputs (Cho et al., 2017), so companies are keen to understand and manage employees’ emotions. The present study used SIP (Salancik and Pfeffer, 1978) to establish the linkage between CSR and negative emotions. The theorists stated that individuals work in a social environment, and their behaviour is shaped by deriving information from the social environment. A firm’s CSR actions imply their concerns for its stakeholders; some research showed that these CSR actions can be used as social information that individuals may use to interpret the principles of their employer (Gao and He, 2017; Rupp et al., 2013; Jones et al., 2014). CSR signals a company’s concerns about social commitment (Rupp and Mallory, 2015), which makes employees identify with their organization (Farooq et al., 2017). Lee et al. (2014) suggested that employees perceiving their organization as socially responsible are highly engaged in their work. Through CSR actions, firms send clear information cues (Gao and He, 2017). Indulging in several CSR programmes signals employees that employers are highly concerned about their employees (Ilkhanizadeh and Karatepe, 2017). Under these circumstances, CSR has a negative impact on employees’ negative behaviour. The findings of Evans et al. (2010) suggested that corporate citizenship reduces employee deviance via organizational cynicism. CSR actions significantly reduce employees’ perception of supervisor moral decoupling (Luan et al., 2022).
Furthermore, CSR reduces cheating behaviour via organizational identification (Luan et al., 2022). The study’s results suggested that CSR significantly reduces employee turnover intentions (Lin et al., 2022). Moreover, Spencer and Rupp (2009) argued that if firms are not socially responsible, employees may experience negative emotions, e.g. moral anger arising from perceived mistreatment by employers towards other stakeholders. Drawing on SIP the researcher anticipates that:
CSR influences negative emotions negatively.
2.6 Corporate social responsibility and compassion
Cropanzano and Wright (2001) revealed that employees’ attitudes and behaviour are greatly influenced by actions taken by firms in the workplace. These actions can be firms’ CSR activities taken voluntarily towards their stakeholders. The social actions taken by a firm towards its stakeholders make its employees feel connected to the firm (Maignan and Ferrell, 2001). Employees of socially responsible firms display positive work behaviour, which includes high-quality work (Dutton et al., 2002), creativity (Hur et al., 2018) and positive emotions (Subba and Kumar, 2018). Some of the studies have revealed that socially responsible firms greatly influence compassion. Moon et al. (2014) indicated that employees’ perceptions of CSR develop compassion directly and indirectly through organizational commitment and organizational justice perceptions. The study by Nazir and Islam (2020) identified how CSR actions bring out helping behaviour in the form of compassionate acts in the hospitality industry. The findings of Subba and Rao (2016) revealed that CSR to different stakeholders emerged as a significant antecedent of compassion in the workplace.
Thus, it can be postulated that:
CSR influences compassion positively.
2.7 Compassion and negative emotions
Compassion helps alleviate others’ agony by giving support (Fredrickson et al., 2003) and building a conducive environment in the workplace (Frost and Ho, 2006). Compassionate individuals prefer collective to individual gains (Miller et al., 2012). Employees receiving compassion at the workplace promote well-being (Frost, 2003; Kahn, 1998). Compassionate individuals accept positive and negative aspects of life and regulate negative effects by confronting adverse situations (Leary et al., 2007). These compassionate individuals reduce negative work behaviours to a greater extent (Shuler and Sypher, 2000; Hayward and Tuckey, 2011). In the longitudinal study of Eldor (2018), the findings indicated that employees receiving compassion from supervisors in the workplace reduce their burnout to a greater extent. The study conducted by Arimitsu and Hofmann (2015) indicated that compassionate acts shown by colleagues at the workplace influence employees’ negative emotions. It is evident from the literature that when employees demonstrate compassionate acts to each other, their negative work behaviour may be reduced. Based on the previous findings, it can be postulated:
Compassion influences the negative emotions of employees.
2.8 Mediating role of compassion
Prior research has identified the role of CSR in influencing employees’ positive work behaviour (Rupp et al., 2006; Turker, 2009; Peterson, 2004). Some studies have revealed that CSR initiatives induce compassion among the members in the workplace (Nazir and Islam, 2020; Abdelmotaleb et al., 2018; Subba and Kumar, 2018). These studies revealed that when firms indulge in CSR initiatives and show concerns about different stakeholders by going beyond profit maximization, employees demonstrate compassionate acts in the workplace towards the other members of the firm. Employees receiving compassion at the workplace promote a sense of well-being at the workplace (Frost, 2003; Kahn, 1998). A sense of being loved and loving others led to a decline in depressive symptoms (Kahana et al., 2021). Compassionate acts shown by colleagues at the workplace, in turn, influence employees’ negative emotions (Arimitsu and Hofmann, 2015). It is evident from the literature that CSR develops compassionate acts, and when employees show compassion to each other, their negative work behaviour may be reduced. Based on the previous, it is postulated that:
Compassion mediates the relationship between CSR and the negative emotions of employees.
2.9 Moderating role of positive emotions
Fredrickson (2013) and Shiota et al. (2011) observed that positive emotions surge vagal tone and oxytocin levels. Kok et al. (2013) argued that an upsurge in vagal tone helps in maintaining physical and mental health, whereas Fredrickson (2013) argued that an upsurge in oxytocin is associated with a lower amount of fear and a higher amount of trust towards others. Positive emotions help to tackle negative emotions and stress. It prevents individuals from experiencing rage and worries (Demorest, 2020), sadness (Brummett et al., 2009) and stress (Zander-Schellenberg et al., 2020). Further, positive emotions lead to a reduction in various physical stressors, including inflammation (Steptoe et al., 2009), agony (Ong et al., 2015) and stressful situations (Ong and Isen, 2010). Positive emotions lower the amount of negative emotions in reaction to everyday life events (Zautra et al., 2005). Fredrickson and Levenson (1998) suggested that individuals with positive emotions, such as amusement and contentment, recovered sooner than individuals in sadness from distress. Accordingly, the researcher assumes that positive emotions would reduce the effect of CSR on negative emotions. Thus, it is postulated that:
Positive emotions moderate the relationship between CSR and employees’ negative emotions.
3. Research methods
A cross-sectional study is conducted in which data are collected at a single point in time, while a longitudinal study is carried out to study people or phenomena at two or more points in time (Zikmund et al., 2010). However, financial and time constraints prohibit this study from being conducted longitudinally across a time. As a result, a cross-sectional, rather than a longitudinal design, was chosen in this study. There are some advantages of a cross-sectional design. First, cross-sectional research is useful for a hypothesis-testing study, which is the research design of this study. This is because cross-sectional research allows a researcher to analyse the cross-sectional sample carefully by studying a sample of elements from the population of interest that are measured at a single point in time. Second, cross-sectional data can be collected from a large number of people, and these data are comparable because they are not influenced by changes over time. In these regards, the use of a cross-sectional design appears to be an appropriate research method. The sample size was designed by following the guidelines of a minimum of 200 samples required to run SEM (Kline, 2011) and five samples per each parameter examined (Barclay et al., 1995). A self-administered questionnaire was distributed to 240 employees of financial institutions in the North East of India. In a self-administered questionnaire, there could be biases such as social response, non-response bias, prestige, etc. The researcher tried to control this bias through remedies such as keeping respondents’ information confidential, improving item wording, etc. The researcher used convenience sampling in quantitative research to collect data. This technique is suitable for collecting the data from a pool of large respondents (Passmore and Baker, 2005). A total of 240 samples were collected from the financial institutions. All the constructs were measured using existing scales presented in Table 1. A five-point Likert scale varying from 1 = “strongly disagree” to 5 = “strongly agree” was used to collect responses. The researcher also conducted a pilot study among the professionals working with financial institutions to check the reliability of the instruments. During the analysis, one item measuring CSR3 and items measuring CSR4 had Cronbach’s alpha below the threshold level, 0.7 (Hair et al., 2013), hence deleted for further analysis. The results of the pilot study are presented in Table 1.
Results of the pilot study
| S. no. | Constructs | Sub-constructs | Cronbach’s alpha | Scale adapted from | No. of items |
|---|---|---|---|---|---|
| 1 | CSR | CSR1 | 0.89 | Turker (2009) | 6 |
| CSR2 | 0.88 | 6 | |||
| CSR3 | 0.71 | 3 | |||
| 2 | PE | − | 0.78 | Fredrickson et al. (2003) | 10 |
| 3 | CO | − | 0.87 | Lilius et al. (2008) | 3 |
| 4 | NE | 0.85 | Fredrickson et al. (2003) | 8 |
CO = compassion; PE = positive emotions; and NE = negative emotions
Employees aged 20–29 are highest, 53.75%, followed by 30–39, 28.75%, 50 and above constitute 11.25%, while 40–49 constitutes a meagre of 6.25%. Men’s workforce is in the majority, constituting 72.08%, while women’s workforce is 27.92%. Regarding work experience, a maximum number of employees have 0–5 work experience in the same company, i.e. 62.50%, followed by 6–10, which constitutes 26.67%, while employees having work experience 11–15 is 10.83%.
4. Data analysis
IBM SPSS and AMOS 21 were used to analyse data.
4.1 Missing data and outliers
Before analysing the data, it is important to detect outliers and missing data, as the results obtained in the presence of these issues could lead to skewed outcomes. Hence, Mahalanobis D2 statistics were calculated to detect outliers in the data, and the result indicated no outliers. Thus, all data were included for further analysis. Furthermore, while examining the filled responses, it was found that there were five missing values. Thus, to carry out the data analysis, these missing values were replaced by using the expected maximization method.
4.2 Descriptive statistics
Descriptive statistics are presented in Table 2. The mean of dimensions of CSR ranges from 3 to 3.85, indicating that employees’ perception of CSR policies of the organization is high. The mean score of negative emotions is below the mid value, i.e. 2.13, indicating employees relatively experience low negative emotions at the workplace. The mean score of compassion is 3.95, indicating that employees are highly experiencing compassion at the workplace. The mean of the moderating variable, i.e. positive emotions, is 3.86, indicating employees are experiencing high positive emotions. The correlation table indicates that the dimensions of CSR2 and CSR3 showed a negative correlation with negative emotions and are statistically significant, but CSR1 with negative emotions is insignificant. The three dimensions of CSR showed a positive correlation with compassion and positive emotions and were statistically significant. The correlation of positive emotions with compassion is positive and negative with negative emotions and is statistically significant. Finally, the correlation between compassion and negative emotions is insignificant.
Descriptive statistics
| Variables | Mean | Std. dev | CSR1 | CSR2 | CSR3 | PE | NE | CO |
|---|---|---|---|---|---|---|---|---|
| CSR1 | 3.85 | 0.58 | 1 | |||||
| CSR2 | 3.73 | 0.73 | 556** | 1 | ||||
| CSR3 | 3 | 0.36 | 0.397** | 0.376** | 1 | |||
| PE | 3.86 | 0.42 | 0.301** | 0.367** | 0.209** | 1 | ||
| NE | 2.13 | 0.58 | −0.120 | −0.189* | −0.172* | −0.368** | 1 | |
| CO | 3.95 | 0.60 | 0.315** | 0.429** | 0.183** | 0.479** | −0.112 | 1 |
P < 0.05**; P < 0.01*
4.3 Common method bias
Harman’s single-factor test was performed by a researcher to test the presence of common method bias (CMB) in the data set. The principal component analysis was run, and the data set was free from CMB because the first factor accounted for 42.97% of the total variance, <50% (Podsakoff and Organ, 1986).
4.4 Confirmatory factor analysis
The measurement model was verified by running confirmatory factor analysis (CFA). Since CSR is a multi-dimensional construct, a second-order CFA was performed, and the result is presented in Table 3. The standardized factor loadings of all items were above the threshold of 0.5 (Hair et al., 2006) except four items measuring positive emotions, hence dropped for further analysis. The model exhibited an excellent fit to the data (CFI = 0.918, TLI = 0.910, IFI = 0.905, RMSEA = 0.053), and the normed chi-square is 1.679, within the threshold of 1–3 (Hair et al., 2013). The composite reliability (CR) was above the threshold of 0.7 (Hair et al., 2006), confirming data reliability. The average variance extracted (AVE) for each factor was >0.5 (Fornell and Larcker, 1981), thus confirming convergent validity (CV). The squared root of AVE was greater than inter-construct correlations, which confirmed the presence of discriminant validity (DV) (Hair et al., 2013). The results of CV and DV are displayed in Table 4.
Results of CFA
| Items | Factor loadings | CR | AVE |
|---|---|---|---|
| “Our company involves in the activities that aim to protect and improve the quality of the natural environment” | 0.56 | 0.78 | 0.552 |
| “Our company invests to make a better life for the upcoming generations” | 0.66 | ||
| “Our company indulges in special programmes to reduce its negative effect on natural environment” | 0.57 | ||
| “Our company targets a sustainable growth keeping in mind the future generations” | 0.68 | ||
| “Our company supports the non-governmental organizations working in the problematic areas” | 0.51 | ||
| “Our company contributes to the campaigns and projects that promote well-being of the society” | 0.65 | ||
| “Our company supports its employees to involve in voluntarily activities” | 0.54 | ||
| “Our company policies give encouragement to employees to enhance their careers and skills” | 0.65 | ||
| “The management of our company is concerned with employees’ wants and needs” | 0.78 | ||
| “Our company implements flexible policies to provide good work and life balance for its employees” | 0.78 | ||
| “The managerial decisions related with the employees are usually fair” | 0.76 | ||
| “Our company supports employees who want to acquire additional education” | 0.61 | ||
| “Our company gives accurate and full information to its customers about its products” | 0.73 | ||
| “Satisfaction of customers is very important for our company” | 0.76 | ||
| “I frequently experienced compassion on the job” | 0.67 | 0.84 | 0.648 |
| “I frequently experienced compassion from supervisor” | 0.91 | ||
| “I frequently experienced compassion from co-workers” | 0.80 | ||
| “I frequently experienced anger/ irritation/ annoyance” | 0.65 | 0.91 | 0.559 |
| “I frequently experienced sadness/ disappointment/ unhappiness” | 0.70 | ||
| “I frequently felt scared/ frightened/ terrified” | 0.75 | ||
| “I frequently experienced disgust/ distaste/ revulsion” | 0.82 | ||
| “I frequently felt contemptuous/ scornful/ disdainful” | 0.73 | ||
| “I frequently felt shy / embarrassed/ self-conscious” | 0.79 | ||
| “I frequently felt repentant/ guilty/ blameworthy” | 0.75 | ||
| “I frequently experienced reluctance/ humiliation/ disgrace” | 0.73 | ||
| “I frequently experienced gratefulness/ appreciation/ thankfulness” | 0.71 | 0.85 | 0.505 |
| “I frequently experienced attention/ vigilance/ curiosity” | 0.67 | ||
| “I frequently experienced love/ closeness/ trust” | 0.73 | ||
| “I frequently experienced gladness/joyfulness/happiness” | 0.75 | ||
| “I frequently experienced hope/courage/optimism” | 0.73 | ||
| “I frequently experienced awe/ wonder/ amazement” | 0.64 |
4.5 Structural equation modelling
The model fit the data well (x2/df = 1.957, IFI = 0.909, TLI = 0.898, CFI = 0.908, RMSEA = 0.06). Figure 1 presents the path estimates of the structural equation modelling (SEM). Concerning hypotheses testing, the results support H1 and H2 only. CSR has a significant negative influence on negative emotions (β = −0.23) and a significant positive influence on compassion (β = 0.51). However, compassion has no significant impact on negative emotions at >0.05. The bootstrapping (Shrout and Bolger, 2002) was performed to test the mediating effect of compassion in the relationship between CSR and negative emotions, and the result showed that compassion does not play a mediating factor between CSR and negative emotions, as depicted in Table 5. Furthermore, positive emotions do not moderate the relationship between CSR and negative emotions, insignificant at p > 0.05, presented in Table 6.
The structural path diagram shows latent variables C S R, C S R 1, C S R 2, C S R 3, C O, and N E with observed indicators and error terms, C S R 1 connects to C S 1, C S 2, C S 3, C S 4, and C S 5 with error terms e 1, e 2, e 3, e 4, and e 40, C S R 2 connects to C E 1, C E 2, C E 3, C E 4, C E 5, and C E 6 with error terms e 41, e 8, e 9, e 10, e 11, and e 12, C S R 3 connects to C C 2 and C C 3 with error terms e 13 and e 14, C S R connects to C S 6 with error term e 6, C S R receives paths from C S R 1, C S R 2, and C S R 3 with error terms e 37, e 38, and e 39, C S R connects to C O and N E, C O connects to C O 1, C O 2, and C O 3 with error terms e 16, e 17, and e 18, N E connects to N E 1, N E 2, N E 3, N E 4, N E 5, N E 6, N E 7, and N E 8 with error terms e 19, e 20, e 21, e 22, e 23, e 24, e 25, and e 26, additional error terms e 42 and e 43 connect to N E and C O respectively, path coefficients of 1 appear on indicator loadings throughout the model.Path estimates
The structural path diagram shows latent variables C S R, C S R 1, C S R 2, C S R 3, C O, and N E with observed indicators and error terms, C S R 1 connects to C S 1, C S 2, C S 3, C S 4, and C S 5 with error terms e 1, e 2, e 3, e 4, and e 40, C S R 2 connects to C E 1, C E 2, C E 3, C E 4, C E 5, and C E 6 with error terms e 41, e 8, e 9, e 10, e 11, and e 12, C S R 3 connects to C C 2 and C C 3 with error terms e 13 and e 14, C S R connects to C S 6 with error term e 6, C S R receives paths from C S R 1, C S R 2, and C S R 3 with error terms e 37, e 38, and e 39, C S R connects to C O and N E, C O connects to C O 1, C O 2, and C O 3 with error terms e 16, e 17, and e 18, N E connects to N E 1, N E 2, N E 3, N E 4, N E 5, N E 6, N E 7, and N E 8 with error terms e 19, e 20, e 21, e 22, e 23, e 24, e 25, and e 26, additional error terms e 42 and e 43 connect to N E and C O respectively, path coefficients of 1 appear on indicator loadings throughout the model.Path estimates
5. Discussion
After the extensive literature review, some relationships were hypothesized and verified by performing CFA and SEM. The current research findings indicated that only two out of five hypothesized relationships were significant. First, CSR was found to significantly influence negative emotions negatively (H1, p < 0.05), thus indicating that if employees perceive that their employer is very active in fulfilling their commitment to society and employees then employees regulate their emotions to a greater extent, which coincides with the previous study of Subba and Kumar (2018). Earlier research revealed that CSR actions of firms instil positive emotions among employees; the current research revealed that CSR actions help employees reduce their negative emotions to a great extent. Second, CSR also significantly influenced compassion (H2, p < 0.05), which coincides with previous findings (Moon et al., 2014; Nazir and Islam, 2020), which indicate that if employees perceive that firms are meeting the needs of their stakeholders by becoming socially responsible, then employees may exhibit compassionate acts towards their co-workers in the workplace and eventually may attempt to alleviate the suffering of colleagues in the workplace. Third, compassion does not impact negative emotions (H3, p > 0.05), thereby indicating that compassionate employees in financial institutions may not regulate their emotions and reduce their negative emotions, which is inconsistent with the previous findings (Arimitsu and Hofmann, 2015). There could be two reasons: first, compassionate employees feel distressed and cannot regulate their emotions when they see the agony of others. Second, when the same compassionate individuals do not receive compassion from other members at the workplace they may show negative emotions like anger and frustration. Fourth, it was observed that compassion does not mediate the relationship between CSR and negative emotions (H4, p > 0.05). The bootstrapping technique investigated the mediating effect of compassion in the relationship between CSR and negative emotions. The results showed that CSR does not influence negative emotions indirectly through compassion. Employees who become compassionate due to the social actions of firms do not reduce their negative emotions in the workplace. Finally, positive emotions do not increase/decrease the impact of the relationship between CSR and negative emotions.
5.1 Theoretical contribution
The study has addressed many gaps; earlier research has mainly investigated how CSR influences employees’ positive work behaviour (Aguinis and Glavas, 2012; Raub and Blunschi, 2014; Newman et al., 2015). Very few studies have focused on how CSR influences negative work behaviour (Panagopoulos et al., 2016; Lin and Liu, 2017). The prior research has motivated researcher to investigate how CSR reduces employees’ negative behaviour, particularly negative emotions. Previous CSR research focuses on employees’ emotions (Khan et al., 2021; Budhiraja and Yadav, 2020; Hericher and Bridoux, 2022). However, these studies have overlooked the way CSR influences negative emotions. The current research fills this gap by studying how CSR influences employees’ negative emotions.
Second, the findings of the current research have indicated that a firm’s CSR strategies influence the employee’s negative emotions through the SIP mechanism (Salancik and Pfeffer, 1978). The theorists stated that individuals work in a social environment, and their behaviour is shaped by deriving information from the social environment. A firm’s CSR actions imply their concerns for its stakeholders; some research showed that these CSR actions can be used as social information that individuals may use to interpret the principles of their employer (Gao and He, 2017; Rupp et al., 2013; Jones et al., 2014). Through CSR actions, firms send clear information cues (Gao and He, 2017). Indulging in several CSR programmes signals employees that employers are highly concerned about their employees (Ilkhanizadeh and Karatepe, 2017). Under these circumstances, CSR has a negative impact on employees’ negative behaviour.
Thirdly, some studies have investigated the mediating effect in the relationship between CSR and employee outcomes. For instance, well-being (Ahmed et al., 2020), organizational justice (Sarfraz et al., 2018), identification and trust (Su and Swanson, 2019). However, these studies mainly investigated the mediating effects in the relationship between CSR and positive work behaviour. Surprisingly, limited empirical research has tested the mediating effects of CSR and negative output. No evidence has investigated the mediating effects of compassion in the relationship between CSR and negative emotions.
Fourthly, based on the gaps identified, the researcher conceptualized a new framework by introducing a combination of novel constructs (CSR, compassion, negative emotions and positive emotions) through SIP in the Indian context, particularly in the financial institutions. The model showed that CSR influences negative emotions, adding to the existing management studies literature.
Finally, the literature revealed that the relationship between CSR and employee output is moderated by numerous factors. For instance, corporate ability (Brammer et al., 2015), strong responsible leadership (Bouichou et al., 2022) and restoration of belief (Zhang et al., 2021). However, these studies mainly investigated how some variables moderate the relationship between CSR and positive work behaviour. No evidence has investigated the moderating effects of compassion in the relationship between CSR and negative emotions.
5.2 Practical contribution
The contribution to CSR activities by financial institutions in India is paramount; they are consistently playing a crucial role in the country’s socioeconomic development. The workforce plays a significant role in making financial institutions one of the prominent sectors in India, yet there is no research on how CSR influences employees’ negative emotions. Negative emotions may bring undesirable behaviour in the workplace, which may be detrimental to the organization. The findings suggest that to reduce negative emotions in the workplace, managers should closely observe how employees respond to their employers concerning social responsibilities. Employees may evaluate their employer in different aspects such as the employer’s various initiatives for the socio-economic development of society, treatment of the employer in the form of ethical practices, trust, fair distribution of compensation and benefits and top management support in the forming and implementation of CSR programmes.
Second, the findings provide valuable insight for the practitioners who determine the CSR actions of their employer. Employees’ perception of CSR is a motivating force for expediting compassionate behaviour in the workplace. A firm’s CSR strategies may make employees perceive that they have been treated well by their employer without any self-motive, ultimately making employees demonstrate compassion towards their colleagues encountering hard times. Compassionate employees empathise with their colleagues and may try to alleviate their suffering to a certain extent without any self-motive.
Thus, while forming and implementing CSR policies employees’ perception towards CSR must be considered. The study findings suggest that firms invest more in CSR spending to induce positive work behaviour and decrease negative work behaviour. The findings further suggest that firms monitor and execute CSR policies properly. The firm should also create awareness about the CSR programme among the employees. As employees are key stakeholders their participation in CSR programmes must be considered. Employees’ perception concerning the CSR actions of firms can greatly impact their employees.
6. Conclusion
It was observed that CSR to different stakeholders, namely, society, customers and employees, influences employees’ work behaviour. Future researchers are suggested to incorporate other stakeholders, such as creditors and shareholders, and see the influence of CSR on employees. The current research investigated the impact of CSR on negative emotions. Future investigation may focus on how CSR reduces other negative work-related outputs that may reduce the firm’s productivity. The data was collected from the public sector unit and private firms. The employees’ experience in the workplace may be different due to variations in work culture. Thus, future researchers can carry out a comparative study. Methodically, the current study uses a cross-sectional design. The cross-sectional design is not time-consuming, but it fails to draw inferences about causality. Further research should use a longitudinal design to draw inferences about casualty. The current research used the SIP to explain the relationship between CSR and negative emotions, by observing the mediating role of compassion. Future SIP research should introduce other mediators in the relationship between CSR and negative emotions. Despite the limitations, this study would be very useful in conducting future research in the domain of CSR.
