key: cord-0047635-5eptdegv
authors: Zhang, Zhe; Wang, Juan; Jia, Ming
title: Integrating the Bright and Dark Sides of Corporate Volunteering Climate: Is Corporate Volunteering Climate a Burden or Boost to Employees?
date: 2020-06-29
journal: nan
DOI: 10.1111/1467-8551.12418
sha: 24889302a1a89e14259f28aa71427fc5e56a0aeb
doc_id: 47635
cord_uid: 5eptdegv

Conventional wisdom is that a corporate volunteering climate should be beneficial to employees. We challenge this perspective by integrating person–environment fit and conservation of resources theories and providing an integrative model that examines how and when corporate volunteering climate produces positive and negative consequences for employees. Using a two‐wave time‐lagged study of 283 employees in 42 companies, our multilevel model shows that positive emotions mediate the relationship between corporate volunteering climate and work engagement, and perceived role overload mediates the relationship between corporate volunteering climate and work–family conflict. Employees’ communal orientation moderates the positive relationship between corporate volunteering climate and positive emotions, whereas competitive orientation moderates the positive relationship between corporate volunteering climate and perceived role overload. Communal orientation moderates the indirect effect between corporate volunteering climate and work engagement via positive emotions, but competitive orientation moderates the indirect effect between corporate volunteering climate and work–family conflict via perceived role overload. Our study provides balanced insights to help understand the benefits and costs of corporate volunteering climate in work and non‐work domains.

As a corporate social responsibility (CSR) activity, corporate volunteering is a vital channel for showing concern and compassion for causes and communities in need (Grant, 2012) . Approximately 90% of Fortune Global 500 companies have corporate volunteering programmes (Rodell et al., 2017) . Corporate volunteering provides numerous benefits for personnel, communities and companies, but also presents several challenges (Dreesbach-Bundy and Scheck, 2017; Hu et al., 2016) . For example, some employees can benefit from such volunteering by improving their morale (Basil et al., 2009) , but others may have trouble balancing volunteering, work and family demands (Cowlishaw, McLennan and Evans, 2008) . Accordingly, the question of why the same corporate volunteering increases the work engagement of some employees but triggers work-family conflict among others is raised.

To date, limited studies on volunteering explore mainly the effects of volunteering on volunteers at the individual level (Booth, Park and Glomb, 2009; Jones, 2010; Rodell, 2013) . However, how employees react to organization-level corporate volunteering remains unclear. Recent research extends the idea of volunteering to the organization level by introducing a corporate volunteering climate, defined as 'employees' shared perception about the extent of employee volunteering through their corporate volunteering programs' (Rodell et al., 2017, 2 Z. Zhang, J. Wang and M. Jia p. 1665) . This kind of volunteering climate reflects the extent to which employees as a whole donate their time and skills to volunteer organizations (Grant, 2012; . For example, employees at Intel share their time and talent to help volunteer groups through Intel's 'Volunteer Matching Grant Program' . Similarly, every employee of Alibaba needs to participate in volunteer service for 3 hours each year through Alibaba's 'Three Hours of Public Welfare' programme. Corporate volunteering climate serves as a conduit through which the construct of volunteering influences all employees, regardless of whether they participate (Rodell et al., 2017) .

Is corporate volunteering climate a benefit or a burden for employees? Existing studies explore why corporate volunteering climate has a positive effect on affective commitment and volunteering intentions (Rodell et al., 2017) . However, these studies remain limited in two ways. First, previous studies have not provided a clear explanation of how and when corporate volunteering climate influences employees' positive work behaviours, which limits the theoretical understanding of corporate volunteering climate. Second, previous studies regarding volunteering are overwhelmingly positive. Recent studies offer evidence that corporate volunteering climate may have a 'dark side' in certain situations (Hu et al., 2016; Rodell and Lynch, 2016; . Scholars also call for adopting an even-handed view to understand human behaviour (Alvesson and Sandberg, 2011) . In line with this balanced perspective, we simultaneously explore the rewards and potential risks of corporate volunteering climate. We focus on employees' work engagement and work-family conflict because they reflect favourable and unfavourable behavioural reactions to corporate volunteering climates in work and non-work domains. Addressing these issues is important because it not only offers a better comprehensive framework to understand the double-edged nature of corporate volunteering climate and provides opportunities to shift consensus among scholars, but also demonstrates that the role of corporate volunteering climate may extend beyond the four walls of their workplaces.

How does corporate volunteering climate generate positive and negative effects? To answer this question, we draw on conservation of resources (COR) theory (Bono et al., 2013; Hobfoll, 1989) , which provides resource-building and resource-depleting mechanisms that account for the effects of corporate volunteering climate on employees' work engagements and work-family conflict. As a kind of affective response, positive emotions capture the resource-building dynamic because it reflects psychological and emotional gains that arise in the corporate volunteering climate (Baer et al., 2015; Bono et al., 2013; Koopman, Lanaj and Scott, 2016) . By contrast, as a kind of cognitive response, perceived role overload captures the resource-depleting dynamic as it reflects a perception that there is a lack of resources to meet role demands (Brown, Jones and Leigh, 2005) . We introduce COR theory into the field of volunteering and explain how corporate volunteering climate can lead to benefits and burdens for employees in work and non-work domains.

When does corporate volunteering climate have divergent effects? We draw on person-environment (P-E) fit theory (Brown, Jones and Leigh, 2005; Edwards, Caplan and Van Harrison, 2001) to answer this question. We focus on individual orientations (in particular, communal and competitive orientations) as factors that influence the effect of corporate volunteering climate because scholars call for exploring why individual factors (e.g. communal values and work orientation) influence how individuals experience CSR differently (Aguinis and Glavas, 2019) . Communal orientation as a kind of communal value (Rodell and Lynch, 2016) causes employees to perceive value fit in corporate volunteering climate (Jones, Willness and Madey, 2014) , thereby producing a positive effect. However, competitive orientation as a kind of work orientation and self-focused value (Chen, Xie and Chang, 2011) causes employees to tend to perceive value misfit in corporate volunteering climate and yields negative effects. In doing so, we also answer the call to explore the boundary conditions under which corporate volunteering climates are beneficial or destructive for employees (Rodell et al., 2017) . Figure 1 presents our multilevel theoretical framework.

The core tenet of COR theory (Bono et al., 2013; Hobfoll, 1989) is that individuals consistently attempt to protect their current resources and build new resources. According to the 'resource caravan passageways' of COR, 'people's resources

Work-family Conflict exist in ecological conditions that either foster and nurture or limit and block resource creation and sustenance. Organizations and the broader culture play a major role in this process' (Hobfoll et al., 2018, p. 107) . In line with this theory, the organizational environment -such as corporate volunteering climate -may generate and consume personal resources.

Positive emotions refer to the relatively intense affective experiences that are focused on specific situations (e.g. corporate volunteering climate) in workplaces (Ouweneel et al., 2012) . Scholars extend the positive psychology perspective into the COR model and find that positive emotions generate various resources that employees can use (Bono et al., 2013) . Based on COR theory, we argue that corporate volunteering climate can influence employees' work engagement through their positive emotions. First, a corporate volunteering climate can improve employees' positive emotions because it can trigger the resource-building process. COR theory indicates that resources are matters that people value, for example, positive feelings, feeling valuable to others and meaningful existence (Halbesleben et al., 2014; Hobfoll, 2001; Troester et al., 2019) . According to COR theory (Bono et al., 2013; Hobfoll, 1989) , positive work events or organizational environments can trigger the resources-building (e.g. positive emotions) process by satisfying basic human needs (Hobfoll et al., 2018) .

In line with this theory, a corporate volunteering climate, as a kind of positive work environment, can trigger positive emotions by satisfying basic human needs (e.g. meaningful existence). Corporate volunteering climate reflects the sense that volunteering is 'something people do here', which signals that the company is responsible for greater societal good and considers the social consequences of its success (Dreesbach-Bundy and Scheck, 2017) . It provides opportunities for employees to benefit others, connect with others and improve self-esteem (Grant, 2012; . It also makes employees derive a sense of meaning (Aguinis and Glavas, 2019; Caligiuri, Mencin and Jiang, 2013) . Accordingly, experiencing a corporate volunteering climate likely fulfils employees' basic human needs and triggers employees' positive emotions. In line with this perspective, empirical evidence indicates that when employees perceive their work as having a 4 Z. Zhang, J. Wang and M. Jia positive impact on others' well-being, employees' positive emotional reactions (e.g. enthusiasm and pride) can be derived from work (Bauman and Skitka, 2012; Grant, 2007; Rodell et al., 2017) .

Second, positive emotions increase work engagement. Work engagement refers to 'an active, work-related positive psychological state operationalized by the intensity and direction of cognitive, emotional, and behavioral energy' (Shuck, Adelson and Reio, 2017, p. 959) . According to the resource investment principle of COR theory, employees must invest resources as protection against resource loss and acquire additional resources (Hobfoll et al., 2018) . Work resources are often reinvested in workplaces (Hobfoll, 2001) . Studies show that positive emotions generate physical, intellectual and psychological resources (Bono et al., 2013) . Thus, employees with positive emotions tend to reinvest those resources back into their work by boosting their work engagement.

Ultimately, corporate volunteering climate builds resources that can be reinvested into work engagement by improving employees' positive emotions.

H1: Positive emotions mediate the relationship between corporate volunteering climate and work engagement.

Despite the potential for resource gain, a corporate volunteering climate can make individuals feel stressed and deplete personal resources (Rodell, 2013) . As a kind of mental strain, perceived role overload occurs when employees have multiple roles within a domain and perceive that they do not have the necessary resources to meet all that the role demands (Brown, Jones and Leigh, 2005; Jensen, Patel and Messersmith, 2013; Matthews, Winkel and Wayne, 2014) . Based on the COR theory, we argue that corporate volunteering climate can influence employees' work-family conflict through their perceived role overload. First, corporate volunteering climate increases employees' perceived role overload. As a kind of time-dependent activity, volunteering distracts employees' energy and time from the pursuit of job goals (Rodell, 2013; Rodell and Lynch, 2016) . COR theory acknowledges that energy and time are among the scarcest resources of employees (Hobfoll, 2001) . According to COR theory, when environmental conditions deplete or threaten finite and valued resources of individuals (i.e. actual or expected resources losses), they experience strain and suffer from resource depletion (Baer et al., 2018; Lanaj et al., 2018; Lapointe, Vandenberghe and Panaccio, 2011; Pingel, Fay and Urbach, 2019) . In line with this theory, perceived role overload reflects a state of depleted resources (Baer et al., 2015; Campbell et al., 2013) . Corporate volunteering climate expects and encourages employees to devote their scarce resources (e.g. time and energy) to perform additional tasks (e.g. volunteering). Meeting these expectations requires employees to invest additional valued resources. Thus, corporate volunteering climate threatens and drains employees' finite resources and makes employees perceive role overload.

Furthermore, employees generally enact two roles: job-holder role (i.e. formal responsibilities of the employee) and organizational member role (i.e. individuals' expectations to be a good employee; Vullinghs et al., 2018; Welbourne, Johnson and Erez, 1998) . Corporate volunteering climate also expects employees to be 'good citizens'. This kind of climate may create conformity pressure for employees and make employees have to engage in volunteering behaviours (Hu et al., 2016; Rodell and Lynch, 2016) . Thus, excessive demands imposed on employees by companies make employees perceive role overload.

Second, perceived role overload can increase work-family conflict. Work-family conflict is 'a form of inter role conflict in which the role pressures from the work and family domains are mutually incompatible in some respect' (Greenhaus and Beutell, 1985, p. 77) . COR theory provides some insights into the spillover process and can be used to develop a rationale for a proposed spillover relationship between role overload and work-family conflict (Hobfoll et al., 2018) . According to this theory, when individuals experience resource loss, they try to engage in defensive attempts to protect their remaining resources from further loss (Hobfoll et al., 2018) . Addressing perceived role overload can consume employees' psychological resources (Bolino and Turnley, 2005) . In line with COR perspective, when employees perceive role overload, they will try to protect their resources by decreasing resource investment in the family domain, therefore increasing work-family conflict. Supporting this argument, previous studies also indicate that expending greater resources to meet heavy workloads leaves one with fewer resources to devote to meeting family demands, thus leading to work-family conflict (Carnes, 2017; Goh, Ilies and Wilson, 2015; Ilies et al., 2007; Van Emmerik and Jawahar, 2006) .

Accordingly, corporate volunteering climate may make employees perceive role overload and leave them with fewer resources for non-work activities, all of which trigger work-family conflict.

H2: Perceived role overload mediates the relationship between corporate volunteering climate and work-family conflict.

Although COR theory outlines how corporate volunteering climate results in benefits and burdens for employees, this theory is silent about when and why the corporate volunteering climate can trigger resource-building and resource-depleting processes. To further address this issue, we draw on P-E fit theory and integrate this theory with COR theory. P-E fit theory indicates that personal characteristics that (mis)fit with organizational characteristics influence individual reactions to work environments, such as corporate volunteering climate (Edwards, Caplan and Van Harrison, 2001; Kristof-Brown, Zimmerman and Johnson, 2005) . This theory is similar to COR theory in that it also recognizes the important role of personality traits in influencing individuals' reactions to the work environment (Koopman, Lanaj and Scott, 2016) . Integrating these two theories, we argue that communal and competitive orientations are particularly relevant to whether work contexts are interpreted as accelerating resource gain or loss, and can have implications for the effect of corporate volunteering climate. First, organizations with a corporate volunteering climate and employees with communal orientation share consistent values. The value fit between employees and organizations makes employees pay greater attention to resource gain and then amplifies the positive effect of corporate volunteering climate on positive emotions. From the standpoint of P-E fit theory, the match between characteristics (e.g. values) of persons and their work environment has positive implications for individual outcomes, including their positive emotions (Edwards, Caplan and Van Harrison, 2001; Kristof-Brown, Zimmerman and Johnson, 2005; Yu, 2009 ). In line with this theory, the match between corporate volunteering climate and employees' communal orientation can trigger more positive emotions among employees.

On the one hand, communal orientation refers to an individual orientation -feeling a sense of responsibility towards the community (Rodell and Lynch, 2016) . Employees with communal orientation tend to assist people in need and pay greater attention to increasing social well-being (Jones, Willness and Madey, 2014) . For these employees, feeling valuable to others and having a meaningful existence are more valued resources (Hobfoll, 2001) . On the other hand, companies that have a volunteering climate can satisfy the basic needs (e.g. feeling valuable to others and having a meaningful existence) of these employees (Bono et al., 2013) . When employees with stronger communal orientation encounter companies that have volunteering climates, they experience greater value fit with those companies (Jones, Willness and Madey, 2014) , and pay greater attention to the resource gain effect of corporate volunteering climate and thus produce greater positive emotions.

H3: Communal orientation moderates the positive relationship between corporate volunteering climate and employees' positive emotions such that the relationship is more positive when communal orientation is high than it is low.

Second, the value misfit between employees with competitive orientation and organizations with corporate volunteering climate makes employees tend to focus on resource loss and then amplifies the detrimental impact of corporate volunteering climate on perceived role overload. Competitive orientation refers to the individual view regarding others as a means of self-development and to show self-worth (Chen, Xie and Chang, 2011) . Employees with competitive orientation tend to focus on self-development and aspire to do better than others at work (Chen, Xie and Chang, 2011) . These employees pay less attention to the well-being of others and to societies, and pay greater attention to pursue their self-interests and expect to be 'good soldiers' (Brik, Rettab and Mellahi, 2011) . They even 'defect' from collective interests to pursue self-interests. However, a corporate volunteering 6 Z. Zhang, J. Wang and M. Jia climate encourages employees to participate in volunteering activities, which can interfere with employees' work goal progress and compel employees to become involved in communities (Rodell, 2013; Rodell and Lynch, 2016; Rodell et al., 2017) . Based on the P-E fit theory (Edwards et al., 2001; Kristof-Brown, Zimmerman and Johnson, 2005) , the value misfit between employees and work environments produces mental strain in employees, and negative outcomes (Edwards and Cable, 2009) . Accordingly, employees with competitive orientation may feel that a corporate volunteering climate mismatches with their values; they may, in turn, consume resources to handle those inconsistent values and goals, bringing them greater mental strain (e.g. perceived role overload).

Meanwhile, employees with competitive orientation are also greatly sensitive to scarce resource loss in corporate volunteering climate, and therefore perceive greater role overload. For employees with competitive orientation, energy resources (e.g. time and skills) are better-valued resources (Hobfoll, 2001) . Excessive demands imposed on employees by companies (e.g. through the corporate volunteering climate) and by employees with competitive orientation themselves can make these employees perceive greater role overload (Vullinghs et al., 2018) . In line with this perspective, scholars also show that employees who focus on their work tend to have difficulty balancing the demands of multiple roles (Halbesleben, Harvey and Bolino, 2009 ). For employees with competitive orientation, corporate volunteering climate constitutes a threatening and stressful situation, which makes these employees perceive greater strain in the form of role overload (Jensen and Patel, 2013) . Thus, employees with competitive orientation pay greater attention to the resources consumption effect of corporate volunteering climate and therefore perceive greater role overload.

H4: Competitive orientation moderates the relationship between corporate volunteering climate and employees' perceived role overload, such that the relationship is more positive when competitive orientation is high than when it is low.

In line with COR theory, we argue that positive emotions mediate the relationship between corpo-rate volunteering climate and work engagement. According to the P-E fit theory, we posit that communal orientation strengthens the relationship between CSR and positive emotions. In turn, we suggest that the higher the level of communal orientation, the stronger the positive effect corporate volunteering climate has on positive emotions and the greater the work engagement. Through the moderated mediation hypothesis, we can test and compare whether the magnitudes of indirect effects of corporate volunteering climate on work engagement through positive emotions vary as communal orientation. Thus, we propose the following moderated mediation hypothesis.

H5: Positive emotions mediate the interactive effects of corporate volunteering climate and communal orientation on work engagement.

Similarly, we argue that perceived role overload mediates the relationship between corporate volunteering climate and work-family conflict. We posit that competitive orientation strengthens the relationship between CSR and perceived role overload. In turn, we suggest that the higher the level of competitive orientation, the stronger the positive effect corporate volunteering climate has on perceived role overload and the greater the workfamily conflict. Through the moderated mediation hypothesis, we can test and compare whether the magnitudes of indirect effects of corporate volunteering climate on work-family conflict through perceived role overload vary as competitive orientation. Thus, we propose the following moderated mediation hypothesis.

H6: Perceived role overload mediates the interactive effects of corporate volunteering climate and competitive orientation on work-family conflict.

To reduce potential common-method variance, we collected two-phase data from 283 employees in 42 companies in China. 1 We obtained strong management support for our research through personal contacts with leaders of the companies, a method that is feasible in China because of its unique Guanxi culture (Zhang and Jia, 2010) . The participating companies have their own corporate volunteering programmes to some extent. These companies represent a range of industries: technology (19%), financial (14%), manufacturing (12%), transportation (12%), construction (7%), energy (7%), medicine and health (10%), petrochemical engineering (7%) and services (12%). These companies are in Shaanxi, Beijing, Shanghai, Guangdong, Jiangsu, Shandong, Tianjin and Henan. Thus, the samples of companies appear to represent broadly the business organizations in China. Table 1 shows the profile of organizational characteristics. According to Rodell et al. (2017) , we asked each liaison in these companies to identify approximately 10 employees to participate in our study, including a mix of employees who participate in corporate volunteering programmes and employees who do not. In Phase 1, we asked employees to evaluate their companies' corporate volunteering climate and describe their communal orientation, competitive orientation, positive emotions and perceived role overload. We received 310 responses from employees in 45 companies with a Table 2 shows the profile of individual characteristics.

We measure all variables on a five-point Likerttype response scale, where 1 = strongly disagree and 5 = strongly agree (see Appendix 2). 8 Z. Zhang, J. Wang and M. Jia Corporate volunteering climate. We adopt a fiveitem scale developed by Rodell et al. (2017) . This construct reflects a referent-shift composition model, which measures employees' shared beliefs regarding engagement in corporate volunteering programmes. Following the prompt of 'Through the company's volunteering programme…', sample items include 'Employees at my company give their time to help a volunteer group'. The Cronbach's alpha is 0.87.

The referent-shift nature of corporate volunteering climate is supported by an examination of within-group agreement of individual ratings using this scale (Rodell et al., 2017) . Thus, we calculate RWG and ICC scores for each company to establish the appropriateness of aggregating employee responses from the individual level to the company level. The mean RWG for corporate volunteering climate is 0.82 (range 0.70-0.94), which is above the conventionally acceptable value of 0.70 and shows strong interrater agreement (Le-Breton and Senter, 2008) . For the scale, ICC(1) is 0.35, ICC(2) is 0.76 and there is significant between-group variance: F value = 4.26, p < 0.01. These results provide sufficient support for aggregating corporate volunteering climate at the organizational level, so we calculate the average value of employee responses within each company to create the corporate volunteering climate.

Communal orientation. We adopt a four-item scale developed by Rodell and Lynch (2016) . Sample items include, 'I feel a sense of responsibility towards the community'. The Cronbach's alpha is 0.86.

Competitive orientation. We adopt a six-item scale developed by Chen, Xie and Chang (2011) . Sample items include, 'I feel envious when my competitors receive awards for their performance'. The Cronbach's alpha is 0.82.

Positive emotions. We adopt a six-item scale developed by previous studies (Cole, Bruch and Vogel, 2006; Fisher, 2000) . We ask participants to rate the extent to which they experience each emotion state in their workplace, through their engagement in (or witnessing of) corporate volunteering programmes. Sample items include 'enthusiastic', 'energetic' and 'proud'. The Cronbach's alpha is 0.92.

Perceived role overload. We adopt a five-item scale developed by Thiagarajan, Chakrabarty and Taylor (2006) . Sample items include, 'I have to do things that I do not really have the time and energy for'. The Cronbach's alpha is 0.89.

Work engagement. We adopt a 12-item scale developed by Shuck, Adelson and Reio (2017) . Sample items include, 'I am really focused on my job when I am working'. The Cronbach's alpha is 0.90.

Work-family conflict. We adopt a nine-item scale developed by Carlson, Kacmar and Williams (2000) . Sample items include, 'I have to miss family activities due to the amount of time I must spend on work responsibilities'. The Cronbach's alpha is 0.90.

Following previous studies of volunteering (Rodell, 2013) , age, gender, education level and tenure are significantly related to employees' attitudes about volunteering. We control for these demographic variables.

Analyses. Because our study utilizes a nested design (multiple employees nested within companies), we test our model using multilevel path analysis modelling in MPlus 7.11. To test our hypotheses, we follow recommendations from Preacher, Zyphur and Zhang (2010) and utilize a parametric bootstrap to estimate the significance of indirect effects.

We conducted confirmatory factor analyses to assess the discriminant validity of the measurement model. Table 3 shows that the hypothesized sevenfactor model with χ 2 (1,037) = 1,915.32, p < 0.01, CFI = 0.910, TLI = 0.900 and RMSEA = 0.055 fitted the data significantly better than all other alternative models, such as the six-factor model, where communal orientation and competitive orientation are combined, with χ 2 (1,043) = 2,188.70, p < 0.01, CFI = 0.882, TLI = 0.872 and RMSEA = 0.062, and the one-factor model, with χ 2 (1,058) = 5,790.829, p < 0.01, CFI = 0.513, TLI = 0.480 and RMSEA = 0.126. The results indicate that the seven variables are distinctive constructs and are therefore retained in the subsequent analysis.

Although we collected two-phase data, five variables (i.e. independent variable, moderators and 

Eight-factor model: seven theoretical constructs with the latent common-methods variance factor. Seven-factor model: hypothesized model. Six-factor model: communal orientation and competitive orientation combined. Five-factor model: communal orientation and competitive orientation combined; positive emotions and perceived role overload combined. Four-factor model: communal orientation and competitive orientation combined; positive emotions and perceived role overload combined; work engagement and work-family conflict combined. One-factor model: all variables combined. mediators) are from the same source at Time 1. We used multiple methods, namely Harman's singlefactor test and controlling for the effects of an unmeasured latent methods factor to test for the presence of CMV (Podsakoff et al., 2003) . A onefactor model does not fit well (χ 2 = 2,368.11, df = 294, χ 2 /df = 8.06, p < 0.01, RMSEA = 0.16, CFI = 0.48, TLI = 0.43). However, the five-factor model has a better fit (χ 2 = 564.33, df = 284, χ 2 /df = 1.99, p < 0.01, RMSEA = 0.059, CFI = 0.93, TLI = 0.92). The χ 2 comparison shows that the one-factor model is significantly worse than the five-factor model. We also control for the effects of an unmeasured latent methods factor (Podsakoff et al., 2003) . We construct one latent variable, CMV, by loading all observed indicators of the five theoretical variables. As such, we develop a six-factor model that includes five theoretical variables and CMV. The results reveal that the six-factor model (χ 2 = 526.68, df = 261, χ 2 /df = 1.80, p < 0.01, RMSEA = 0.053, CFI = 0.94, TLI = 0.93) does not substantially improve the goodness of fit of the fivefactor model because the CFI change is only 0.01, which is less than the 0.05 rule of thumb (Bagozzi and Yi, 1990) . Common-method variance is thus not a serious threat for our single-source data.

In this step, we drew paths from positive emotions to work-family conflict and from perceived role overload to work engagement to test for cross-target effects and to compare these effects with the target-specific effects proposed herein. None of the cross-target paths are statistically significant, and in comparing a model with only target-similar paths to a model including crosstarget paths, we found no improvement in fit (χ 2 = 1,908.54, df = 1,035, χ 2 /df = 1.84, p < 0.01, RMSEA = 0.055, CFI = 0.910, TLI = 0.90). Consequently, we remove the crosstarget paths from the model (Farooq, Rupp and Farooq, 2017) .

The descriptive statistics and zero-order correlations for our variables are in Table 4 . Figure 2 illustrates the results of our analysis.

As shown in Figure 2 , corporate volunteering climate increases positive emotions (β = 0.16, p < 0.05). Positive emotions increase work engagement (β = 0.24, p < 0.01). The 95% confidence interval for the indirect effect excludes zero (indirect effect = 0.039, 95% CI = [0.006, 0.072]). Hypothesis 1 is thus supported.

Similarly, corporate volunteering climate is positively related to perceived role overload (β = 0.43, p < 0.01). Perceived role overload is positively related to work-family conflict (β = 0.22, p < 0.01). The 95% confidence interval for the indirect effect excludes zero (indirect effect = 0.095, 95% CI = [0.037, 0.154]). Hypothesis 2 is thus supported. 10 Z. Zhang, J. Wang and M. Jia Figure 2 , the interaction of corporate volunteering climate and communal orientation has a positive and significant effect on positive emotions (β = 0.21, p < 0.05). As shown in Figure 3 , the relationship between corporate volunteering climate and positive emotions is positive and significant when communal orientation is high (β = 0.37, p < 0.01) but not when communal orientation is low (β = −0.05, n.s.), supporting Hypothesis 3.

As shown in Figure 2 , the interaction of corporate volunteering climate and competitive orientation has a positive and significant effect on perceived role overload (β = 0.30, p < 0.05). As shown in Figure 4 , the relationship between corporate volunteering climate and perceived role overload is positive and stronger when competitive orientation is high (β = 0.53, p < 0.01) but weaker when competitive orientation is low (β = 0.31, p < 0.05), supporting Hypothesis 4. The results in Table 5 show that the indirect effect of corporate volunteering climate on work engagement via positive emotions is stronger (β = 0.111, 95% CI = [0.037, 0.186]) when communal orientation is high, compared to when it is low (β = 0.023, 95% CI = [−0.028, 0.073], n.s.). The indirect effects differ significantly when communal orientation is high versus low ( β = 0.089, 95% CI = [0.005, 0.172]), and those confidence intervals do not include zero. Thus, Hypothesis 5 is supported.

Supporting Hypothesis 6, the results in Table 5 show that the indirect effect of corporate volunteering climate on work-family conflict via perceived role overload is stronger (β = 0.146, 95% CI = [0.052, 0.240]) when competitive orientation is high, compared to when it is low (β = 0.044, 95% CI = [−0.028, 0.116], n.s.). The indirect effects differ significantly when competitive orientation is high versus low ( β = 0.102, 95% CI = [0.018, 12 Z. Zhang, J. Wang and M. Jia 

Although we collected the data in two phases, our independent variable and mediators are measured simultaneously. According to previous literature (Rego et al., 2019) , we tested the models wherein we reversed the order of independent variables and mediators. Neither of these models is significant. When we consider positive emotions as the independent variable, corporate volunteering climate as the mediator and work engagement as the outcome variable, the mediation path is not significant (95% CI = [−0.092, 0.773], n.s.). However, when we consider the perceived role overload as the independent variable, corporate volunteering climate as the mediator and work-family conflict as the outcome variable, the mediation path is not significant (95% CI = [−0.084, 0.372], n.s.). These analyses provide strong support for Hypotheses 1 and 2.

Corporate volunteering climate is a relatively new topic in the corporate world and academic research (Rodell et al., 2017) . Drawing from COR theory, we find that corporate volunteering climate increases employees' work engagement via positive emotions. Simultaneously, this kind of climate triggers work-family conflict via perceived role overload. Integrating COR with P-E fit theories, we find that communal orientation strengthens the positive relationship between corporate volunteering climate and positive emotions, whereas competitive orientation strengthens the positive relationship between corporate volunteering climate and perceived role overload. Furthermore, we find that individual orientation moderates the indirect effects of corporate volunteering climate on employees' positive and negative outcomes. Overall, our study serves as a key illustration of the paradox faced by corporate volunteering climate.

By viewing corporate volunteering climate through the lenses of COR and P-E fit theories, we provide balanced insights to facilitate a better understanding of the bright and dark sides of corporate volunteering climate. Our study provides important theoretical contributions in several ways. First, our study is among the first to link corporate volunteering climate with positive and negative employee outcomes (work engagement and work-family conflict), which expands the extremely limited understanding of corporate volunteering climate. Despite existing studies that focus on the individual volunteer, we still know little about corporate volunteering climate. Thus far, only Rodell et al. (2017) have developed the construct of 'corporate volunteering climate', and explored how corporate volunteering climate influences employees' positive attitude. However, to date, few studies have explored whether corporate volunteering climate leads to negative outcomes and why corporate volunteering climate produces favourable and harmful outcomes. We contribute to the literature by challenging the consensus that corporate volunteering climate is universally positive. By taking the 'mixed blessing' approach, our study extends the current understanding of whether corporate volunteering climate helps or hurts employees and shows that the effects of corporate volunteering climate are more nuanced and important than previously expected. Scholars have given limited attention to the 'crossover' effect that corporate volunteering climate influences both volunteers and non-volunteers (Rodell et al., 2017) . We theorize that such a 'crossover' effect significantly extends the scope and importance of corporate volunteering.

We further contribute to the CSR literature by adopting a multilevel approach to help bridge the micro-macro gap and identify the positive and dark sides of CSR simultaneously. To date, CSR studies have focused mainly on macro-level (i.e. institutional and organizational) analysis (e.g. Aguinis and Glavas, 2012; Cumming, Hou and Lee, 2016; Hillenbrand, Money and Ghobadian, 2013; Tang and Tang, 2018; Walker, Zhang and Ni, 2019) . Recently, scholars have focused on micro-level CSR studies, with particular focus on general CSR, and regard CSR as a vehicle to improve employees' positive attitudinal and behavioural outcomes in workplaces (e.g. Aguinis and Glavas, 2019; Glavas, 2016; Morgeson et al., 2013; Voegtlin and Greenwood, 2016) . This reflects 'an ideological pro-CSR bias in management research' (Gond et al., 2017, p. 239) . However, little theory or research has explored the effect of CSR on employees' negative outcomes (especially work-family conflict) in the non-work domain (De Roeck and Maon, 2018) . In addition, few scholars have explored the multilevel studies that integrate the separate micro and macro streams (Aguinis and Glavas, 2012) . Our study adopts a cross-level approach by considering the effects of organization-level corporate volunteering climate and individual-level employee orientation on the individual-level outcome, thereby helping bridge the micro-macro gap. Our findings help CSR studies move beyond the traditional focus on the business case towards a systematic examination of 'softer' criteria, especially employees' work-family conflict in the non-work domain, and enhance our understanding of the 'dark side' of specific CSR (corporate volunteering).

Second, we contribute to COR theory by proposing a framework of resource-building and resource-depleting processes to explore the different ways that organizational-level corporate volunteering climate affects individual-level employees' work and non-work behaviours. Previous studies on the relationship between corporate volunteering and employee outcomes have focused mainly on social identity theory (Gatignon-Turnau and Mignonac, 2015) . However, this theory fails to help explain why corporate volunteering climate can trigger employees' positive and negative reactions simultaneously.

Our study is grounded in a theoretical lens new to corporate volunteering climate literature -COR theory. Specifically, previous studies have overwhelmingly emphasized how individuals protect and gain resources at the individual level, but often ignore how organizations and the broader culture influence individuals' resources (Hobfoll et al., 2018) . Our study enriches the scope of COR theory by extending it into the corporate volunteering domain and identifying corporate volunteering climate as an important organization context involving resource-building and resourcedepleting processes. Existing studies using COR theory generally focus on either resource-building or resource-depleting processes (Koopman, Lanaj and Scott, 2016) . By modelling these two processes related to COR simultaneously, we respond to the call for modelling both mechanisms (Halbesleben et al., 2014) . Our findings confirm the affective (i.e. positive emotions) and cognitive responses (i.e. perceived role overload) as resource-building and resource-depleting processes through which corporate volunteering climate affects employee outcomes. Thus, we contribute to COR theory by revealing how corporate volunteering climate affects employees' resources and outcomes in work and non-work domains, and forming a deeply comprehensive understanding of the full scope of risks and rewards of corporate volunteering climate.

Third, we contribute to the P-E fit theory by integrating COR theory to shed light on for whom corporate volunteering climate can trigger resource-building and resource-depleting processes, through which it affects employees' positive and negative outcomes. Previous studies have focused mainly on the effect of corporate volunteering climate on employee outcomes and its mediating mechanism (Rodell et al., 2017) . However, they have focused limited attention on the boundary condition. By introducing the P-E fit theory into the corporate volunteering field and identifying communal and competitive orientations as key boundary conditions to make the corporate volunteering climate produce different effects, our study expands the scope of the P-E fit theory. To the best of our knowledge, our study is among the first to highlight the fact that different effects of corporate volunteering climate are contingent on employees' communal or competitive orientations, and so responds to scholars' calls to explore the interaction between situational and individual characteristics (Rodell, 2013) .

By integrating P-E fit and COR theories, we reveal how and when the same corporate volunteering climate produces different effects. In particular, P-E fit theory offers a multilevel and individualenvironment interaction framework to help explain whether the interactive effects of contextual and individual factors can affect employee outcomes. However, this theory is silent on how the interaction between employees and the environment generates benefits and burdens for employees. COR theory provides resource-building and resource-depleting processes to help us understand how and why corporate volunteering climates can bring benefits and costs for employees. The integration of these two theories sheds light on why the same corporate volunteering climates can help or negatively affect different employees. Thus, our study serves as a catalyst for the further exploration of the interaction between corporate volunteering climate and individual characteristics.

Finally, we contribute to the multiple domain literature by taking a contingency approach to expand extant corporate volunteering research and provide preliminary insight for scholars regarding when corporate volunteering benefits or harms employees in work and non-work domains. Such spillover effects significantly broaden the importance and scope of corporate volunteering climate (Rodell et al., 2017) . Our study takes multiple domain perspectives to explore the intersection of volunteering, work and family domains, and contributes to the work and non-work spillover literature.

Although the practice of corporate volunteering prevails globally, managers are unclear on the consequences of such practices. Implementing such a volunteering strategy in practice remains an ongoing challenge for many organizations, specifically when encountering public health emergencies (e.g. the challenge of . Managers tend to be very interested in understanding how to focus their efforts and investments to create the most successful environment. Our study has several implications.

First, our study shows that the corporate volunteering climate improves the work engagement of employees, including volunteers and nonvolunteers. This finding shows that companies do not need all employees to participate in volunteering activities, and only cultivate this kind of climate where the benefits of volunteering can be obtained. Given the possible benefits (e.g. addressing societal issues and improving employees' work engagement) of corporate volunteering climate, companies should cultivate this kind of climate. Specifically, managers should develop volunteering programmes that are in line with employees' values and interests, and implement various practices to encourage employees' volunteering behaviour. For example, managers should consider person-environment value fit during recruiting and selection, and reward volunteering behaviour in performance appraisals, promotions and rewards. Managers should communicate information regarding corporate volunteering activities to employees in various ways (e.g. forums, internal reports and training), and provide feedback to employees on how their companies' efforts make significant changes in beneficiaries' lives (e.g. by creating opportunities for employees to contact beneficiaries).

Second, although corporate volunteering climate entails benefits, our findings call attention to the potential risk of corporate volunteering climate in the non-work domain. To clarify, we are not suggesting that managers should avoid creating corporate volunteering climates in workplaces. Instead, we suggest that managers must recognize and manage the risk of work-family conflict related to corporate volunteering climate. For example, during the epidemic prevention of the COVID-19 period, many employees of taxi companies in Wuhan joined the volunteering team to send medical workers home voluntarily. However, they then had to be isolated from their families, and could not take care of their children. Thus, in a corporate volunteering climate, companies should better implement programmes that monitor employees' perceived role overload and develop greater effective mitigation strategies to reduce work-family conflict. For example, companies should adopt targeted training interventions and effective stress management (e.g. providing psychological assistance and mindfulness meditation practices), together with systemic measures (e.g. flexible work hours, family-friendly schedules or telecommuting). To reduce work-family conflict, companies should avoid forcing employees to participate in volunteering activities, encourage employees to communicate the meaningfulness of volunteering to their spouses and families, and provide the necessary help and support to the families of employees participating in volunteering service.

Third, our study provides important guidance for managers attempting to implement corporate volunteering. The most straightforward of these suggestions is that corporate volunteering climate may facilitate work engagement for employees with communal orientation. One area in which this finding may be beneficial is selection. Thus, companies should thoroughly consider personorganization value fit during recruitment and selection, and take full advantage of the strengths of the different individuals in their staff. For example, organizations that value volunteering programmes may benefit from hiring based on individual differences, such as communal orientation.

Our study also shows that employees' competitive orientation influences the effects of corporate volunteering climate on perceived role overload and work-family conflict. Companies should take precautions against the 'backfire effect' of corporate volunteering climate for employees with competitive orientation. For competitive employees, managers may seek to minimize the potential detriment to those employees before the resulting negative effects emerge. For example, managers should design corporate volunteering programmes that provide learning opportunities for those employees to develop their work skills. Moreover, to improve the benefits of corporate volunteering climate, managers should avoid inducing employees' competitive orientation by emphasizing the importance of collaboration rather than being a 'personal hero', and tolerating failure at work.

Our study has several limitations. First, we conducted our research in China. Hence, we do not know whether our study applies in other cultural settings (e.g. Western culture). Individuals from different cultures may respond to corporate volunteering climate differently, because they hold different values. Future research can test the role of cultural values in employees' perceptions of corporate volunteering climate and use cross-cultural data to increase the generalizability of our findings. Second, all measures are self-reported, which may result in common-method variance. As mentioned previously, we address this potential problem by collecting data twice. Future studies can collect data from different sources to reduce common-method bias.

Third, we also use the self-rating of positive emotions. Although this method is consistent with other emotion studies, this kind of subjective measure of emotion has some disadvantages. We encourage future research to measure emotions using subjective and objective measures (e.g. using Fac-eReader) to detect employees' real-time emotional states and discrete emotions.

Fourth, we only examine the cognitive and emotional mediated effects of corporate volunteering climate on employee behaviour, and thus fail to explore whether other underlying mechanisms influence the proposed relationships. Future research can broaden our model by considering motivational mechanisms.

Finally, our study focuses only on the moderating effect of individual differences (i.e. communal and competitive orientations). As such, further study is warranted to explore other variables that can be included in the conceptual framework. For example, testing whether individual ethics moderate the relationship between corporate volunteering climate and employees' behaviour would be a valuable undertaking.

Integrating COR theory with P-E fit theory, we create a multilevel and individual-environment interaction framework to explain how and when corporate volunteering climate can influence employees' work engagement and work-family conflict. Corporate volunteering climate is like two sides of the same coin, and different kinds of employees may see different sides of that coin. We hope that our paper will contribute to the conversation on the positive and negative consequences of corporate volunteering climate.

This study was funded by the National Natural Science Foundation of China (Grants No.

Jia 71672139 and 71932007), and was sponsored by the Humanities and Social Science

What we know and don't know about corporate social responsibility: a review and research agenda

On corporate social responsibility, sensemaking, and the search for meaningfulness through work

Generating research questions through problematization

Uneasy lies the head that bears the trust: the effects of feeling trusted on emotional exhaustion

The benefits and burdens of organizational reputation for employee well-being: a conservation of resources approach

Assessing method variance in multitrait-multimethod matrices: the case of self-reported affect and perceptions at work

Company support for employee volunteering: a national survey of companies in Canada

Corporate social responsibility as a source of employee satisfaction', Research in Organizational Behavior

The personal costs of citizenship behavior: the relationship between individual initiative and role overload, job stress, and work-family conflict

Building positive resources: effects of positive events and positive reflection on work stress and health

Employersupported volunteering benefits: gift exchange among employers, employees, and volunteer organizations

Market orientation, corporate social responsibility, and business performance

The attenuating effect of role overload on relationships linking self-efficacy and goal level to work performance

Win-win-win: the influence of company-sponsored volunteerism programs on employees, NGOs, and business units

All you need is… resources: the effects of justice and support on burnout and turnover

Construction and initial validation of a multidimensional measure of work-family conflict

Bringing work stress home: the impact of role conflict and role overload on spousal marital satisfaction

Cooperative and competitive orientation among Chinese people: scale development and validation

Emotion as mediators of the relations between perceived supervisor support and psychological hardiness on employee cynicism

Volunteer firefighting and family life: an organisational perspective on conflict between volunteer and family roles

Business ethics and finance in greater China: synthesis and future directions in sustainability, CSR, and fraud

Building the theoretical puzzle of employees' reactions to corporate social responsibility: an integrative conceptual framework and research agenda

Corporate volunteering: a bibliometric analysis from 1990 to

The value of value congruence

Theories of Organizational Stress

The multiple pathways through which internal and external corporate social responsibility influence organizational identification and multifoci outcomes: the moderating role of cultural and social orientations

Mood and emotions while working: missing pieces of job satisfaction?

'(Mis)using employee volunteering for public relations: implications for corporate volunteers' organizational commitment

Corporate social responsibility and employee engagement: enabling employees to employ more of their whole selves at work'

Supportive supervisors improve employees' daily lives: the role supervisors play in the impact of daily workload on life satisfaction via work-family conflict

The psychological micro-foundations of corporate social responsibility: a person-centric systematic review

Relational job design and the motivation to make a prosocial difference

Giving time, time after time: work design and sustained employee participation in corporate volunteering

Sources of conflict between work and family roles

Too engaged? A conservation of resources view of the relationship between work engagement and work interference with family

Getting to the "COR" understanding the role of resources in conservation of resources theory

Unpacking the mechanism by which corporate responsibility impacts stakeholder relationships

Conservation of resources: a new attempt at conceptualizing stress

The influence of culture, community, and the nested-self in the stress process: advancing conservation of resources theory

Conservation of resources in the organizational context: the reality of resources and their consequences

The motivational antecedents and performance consequences of corporate volunteering: when do employees volunteer and when does volunteering help versus harm work performance?', Organizational Behavior and Human Decision Processes

When can employees have a family life? The effects of daily workload and affect on workfamily conflict and social behaviors at home'

Highperformance work systems and job control: consequences for anxiety, role overload, and turnover intentions

Does serving the community also serve the company? Using organizational identification and social exchange theories to understand employee responses to a volunteerism programme

Why are job seekers attracted by corporate social performance? Experimental and field tests of three signal-based mechanisms

Integrating the bright and dark sides of OCB: a daily investigation of the benefits and costs of helping others

Consequences of individuals' fit at work: a metaanalysis of person-job, person-organization, person-group, and person-supervisor fit

Daily mistrust: a resource perspective and its implications for work and home

Organizational commitment, organization-based self-esteem, emotional exhaustion and turnover: a conservation of resources perspective

Answers to 20 questions about interrater reliability and interrater agreement

A longitudinal examination of role overload and work-family conflict: the mediating role of interdomain transitions

Extending corporate social responsibility research to the human resource management and organizational behavior domains: a look to the future

Good morning, good day: a diary study on positive emotions, hope, and work engagement

A resources perspective on when and how proactive work behaviour leads to employee withdrawal

Common method biases in behavioral research: a critical review of the literature and recommended remedies

A general multilevel SEM framework for assessing multilevel mediation

Leader humility and team performance: exploring the mediating mechanisms of team psycap and task allocation effectiveness

Finding meaning through volunteering: why do employees volunteer and what does it mean for their jobs?

Perceptions of employee volunteering: is it "credited" or "stigmatized" by colleagues?

Employee volunteering: a review and framework for future research

Corporate volunteering climate: mobilizing employee Z. Zhang, J. Wang and M. Jia passion for societal causes and inspiring future charitable action

The employee engagement scale: initial evidence for construct validity and implications for theory and practice', Human Resource Management

Stakeholder corporate social responsibility orientation congruence, entrepreneurial orientation and environmental performance of Chinese small and medium-sized enterprises

A confirmatory factor analysis of Reilly's role overload scale

The coevolution of social networks and thoughts of quitting

The independent relationships of objective and subjective workload with couples' mood

Corporate social responsibility and human resource management: a systematic review and conceptual analysis

The mirror effect: corporate social responsibility, corporate social irresponsibility and firm performance in coordinated market economies and liberal market economies

The rolebased performance scale: validity analysis of a theory-based measure

Affective influences in person-environment fit theory: exploring the role of affect as both cause and outcome of P-E fit

Using social exchange theory to predict the effects of high-performance human resource practices on corporate entrepreneurship: evidence from China

The authors declare that they have no conflict of interest.Appendix 1 The results of robustness analysis Appendix 2 All measurement items