Introduction

Although compassion at work (from the Latin, com-: together, and -passio: to suffer) has increasingly received attention from organizational scholars in the past decade, few studies have identified the factors that can elicit the appearance of compassion among staff in the hospitality industry. Kanov et al. (2004) define compassion by identifying its three sub-processes: noticing, feeling, and responding to others’ suffering. Paying attention to or noticing suffering is a critical first step that involves becoming aware of the suffering of the other. A compassionate feeling is the second step and resembles empathic concern (Batson 1994; Davis 1983), that is, a relation with the other that involves “suffering with.” Lastly, responding compassionately means taking actions to lessen or relieve the other person’s suffering (Clark 1997; Frost et al. 2000).

Compassion can be a beneficial behavior for organizations. Lilius et al. (2008) corroborate how organizational commitment is generally increased among staff members when they receive compassionate support during critical incidents, leading them to citizenship behavior, higher quality co-worker relations, pro-social behavior, and a reduction in costly staff absenteeism and turnover (Dutton et al. 2007). Other surveys also estimate, for example, that by alleviating the pain and suffering of others, compassionate employees can reduce financial costs significantly, as in the case of grief, which in the US industry costs upwards of $75 billion annually (Zaslow 2002), or job stress and burnout, which cost hundreds of billions of dollars annually (Butts 1997).

Compassion can also have beneficial effects for hospitality organizations. Thus, hotel staff members often act as third parties in a hotel because they have many opportunities to observe peers’ performance during service encounters. Given that suffering is a fundamental and inevitable aspect of the human condition (Barasch 2005; Lama 1995; Nussbaum 1996; Wuthnow 1991), it is quite likely that the staff in hospitality organizations frequently becomes aware of a large number of situations, where co-workers suffer. Furthermore, specificities of hotel performance suggest that, above all, staff members in the hospitality industry are especially aware of events where peers suffer. In effect, staff members in hospitality organizations, in contrast to other businesses, have a very different relationship with each other. They are arranged in homogeneous work teams (generally, closer in education and work activities, dependent task orientation, to name a few) where staff members have to be supported by each other since their work is measured by customer responses (Gittell and Douglass 2012; Grant and Patil 2012). This fact can cause the pain or suffering of one staff member to have an exaggerated effect on the others, and compassion to play a key role in the effective functioning of hospitality organizations.

The aforementioned benefits of compassion at work lead to a logical interest in identifying the organizational factors that are able to promote it. Particularly, in hospitality industry contexts, knowing why compassionate actions occur is extremely significant in promoting the support that members of work teams need from others to achieve team success. Since compassionate behavior cannot be enforced at work, as it greatly exceeds role demands, the existing research on compassion suggests that monitoring certain workplace conditions can create a mindset that stimulates compassion in organizations (Lilius et al. 2012). Prior research suggests a number of work conditions in which compassion is generally likely to unfold. Trust in sharing pain (Dutton et al. 2010), high-quality relationships (Eisenberg 2000; Parker and Axtell 2001), organizational and work group norms (Ashforth et al. 2000; Clark 2000), and leaders who establish the value and legitimacy of compassion (Dutton et al. 2002, 2006), among others, are factors that can generally foster compassion at work.

Due to the higher levels of task dependence existing in the hospitality industry, the causes and the way employees react to the pain or suffering of peers may have some specific characteristics. This paper argues that the hotel staff is particularly morally aware of the need for compassion toward peers because they basically want to ensure the team’s success, and supporting each other plays a key role in this regard. This approach stresses the ethical dynamics of compassion. They are consistent with positive organizational scholarship (POS) (Verbos et al. 2007) and, specifically, positive organizational ethics (POE), which is committed to compassion as a primary area of research and theory (Dutton and Glynn 2008; Dutton et al. 2007). They also fit the ethics of care literature, which suggests the study of compassion as a moral practice in the organizational context (Bowden 1997; Fine 2007; Gilligan 1982; Held 2006; Noddings 2002; Waerness 1996). One frequent instrument to study behavioral ethics in business is the organizational ethical climate (Victor and Cullen 1987). The ethical climate in an organization comprises shared perceptions of what is ethically correct, which involves a ‘team’ psychology. Thus, an ethical climate could provide work teams in hospitality settings with positive conditions to facilitate compassion events that contribute to team and hotel success. Therefore, the present study aims to explore which aspects of a positive ethical climate lead the staff in the hospitality industry to display compassion at work.

An intuitive understanding of the definition of compassion provided above suggests that it is a complex process that is usually measured in a wide range of ways (for a review, see Lilius et al. 2012). Positive organizational scholarship (POS) suggests that compassion measurement should focus on a combination of individual states and traits supporting interpersonal dealings (Cassell 2002; Neff et al. 2007). Based on this idea, the present study operationalizes compassion by invoking four relevant individual factors firmly fixed in the compassionate process, as outlined by prior theory and research (Cassell 2002; Neff et al. 2007). These individual factors are empathic concern or “other-oriented emotional response elicited by and congruent with the perceived welfare of a person in need” (Batson and Ahmad 2009, p. 6); mindfulness or “a state of consciousness in which attention is focused on present-moment phenomena occurring both externally and internally” (Dane 2011, p. 1000); kindness or opening one’s awareness to others’ pain and not avoiding or disconnecting from it, so that feelings of kindness toward others and the desire to alleviate their suffering emerge (Neff 2003; Wispe 1991); and common humanity or “offering non-judgmental understanding to those who fail or do wrong, so that their actions and behaviors are seen in the context of shared human fallibility” (Neff 2003, p. 87).

Before testing the predicted influence of organizational ethical climate on the studied factors of compassion—empathic concern (Ha), mindfulness (Hb), kindness (Hc), and common humanity (Hd), ethical climate will be factor analyzed and the components are extracted. Those elements found to be interpretable and related to the previous ethical climate literature will be examined as predictors of staff compassion in hospitality organizations. Finally, the paper will discuss the theoretical and managerial implications of the findings.

Theoretical Background and Hypotheses

Positive organizational ethics (hereinafter, POE) refer to a movement away from the traditional focus on merely reducing or eliminating unethical practices (Handelsman et al. 2002). Verbos et al. (2007) propose characteristics of the organizational context that can lead organizations to be ethically “positive,” that is, “contexts that cultivate and sustain individual and collective ethical strength to achieve successful and durable moral performance in organizations” (Sekerka et al. 2014, p. 439). Emphasizing the POE approach, this paper models ethical climate and compassion together to shed light on how to shape open and instructive work teams in the hospitality industry that is capable of encouraging compassion and, hence, performance.

As a component of the general organizational climate, the ethical climate of an organization refers to the shared perceptions of what behavior is ethically correct and how ethical issues should be addressed in the organization (Victor and Cullen 1987). The Ethical Climate Questionnaire (ECQ) by Victor and Cullen (1987) classifies ethical climate in organizations into categories that are analogous to Kohlberg’s (1981) ethical standards, where egoism would focus on hedonism, benevolence on utilitarianism, and principle on deontology. The referent level (individual, local, and cosmopolitan) indicates the source of moral reasoning (Victor and Cullen 1988). The individual level refers to employees’ self-determined ethical beliefs, the local level refers to the organization’s standards and policies, and the cosmopolitan level involves external sources of moral reasoning beyond the immediate organization or group. Victor and Cullen (1987) then combined these levels (individual, local, and cosmopolitan) and categories (egoism, benevolence, and principle) to propose an ethical climate classification with nine theoretical types.

In a subsequent empirical study, however, Victor and Cullen (1988) found that only five of these nine ethical climate types existed in organizations. The five types were called: (a) an instrumental climate, which emphasizes maximizing self-interest, i.e., an egoistic concern at the individual or local level; (b) a caring climate, which emphasizes the well-being of others, i.e., a benevolent concern at the individual or local level; (c) an independence climate, which focuses on adherence to one’s personal ethical beliefs, i.e., a principled concern at the individual level; (d) a rules climate, which focuses on following the company’s policies and procedures, i.e., a principled concern at the local level; and (e) a law and code climate, which emphasizes complying with the law and professional standards, i.e., a principled concern at the cosmopolitan level. Each of these types of climate refers to an ethical standard that provides a norm (with its source stemming from the individual, the organization and beyond) for guiding organizational members’ decision-making and team performance.

Across the individual, local, and cosmopolitan sources, the compassionate states and traits seem to display ‘transversal’ influences on compassion, affecting targets of compassion rather than compassion itself (Upchurch and Ruhland 1995). Therefore, this paper does not expect this distinction between level sources to significantly affect the occurrence of compassion itself (Upchurch and Ruhland 1995). For this reason, only the discussion of ECQ ethical categories will provide the basis for the argument in this study; these are egoistic (hedonistic), benevolent (utilitarian), and principled (deontic) climate perceptions (see Fig. 1).

Fig. 1
figure 1

Hypothesized model of organizational ethical climate as a predictor of employee compassion

Full size image

Some prior theory suggests that staff efforts to maximize efficiency or self-interest can generally reduce the likelihood that they will notice peers’ suffering and diminish their capacity to connect with them (Hallowell 1999; Frost 2003). However, specificities of hotel performance suggest the presence of a very different relationship among the members of hotel work teams compared to most businesses. Teams in hospitality organizations usually comprise homogeneous staff members who have an ‘urgent’ need to support others in order to maximize efficiency, since their work is collectively measured by customer responses. This unique relationship among staff members (e.g., closer in education and work activities, and dependent on task orientation) is likely to lead them to be more morally aware of the need for compassion in order to assure that the team succeeds. Hence, driven by an egoism-related climate, in trying to maximize efficiency or self-interest, they will probably increase their empathic concern and kindness (Kalshoven et al. 2013). Moreover, since staff performance in a hotel is usually ultimately measured by customer satisfaction with service, provided that the team succeeds, the staff is likely to be tolerant toward peers’ failures. This effort may lead staff to take peers’ suffering as part of the larger human experience and make positive judgments about that suffering (common humanity).

Finally, POE suggests that when staff members have egoistic attitudes, they are generally less able to understand their own values and motives (self-awareness), which would keep them from developing ethical sensitivity (Handelsman et al. 2002). However, specificities of hotel performance suggest that, apart from efforts to maximize efficiency or self-interest, egoistic ethical considerations in decision-making to assure that their team succeeds can lead hotel staff to be more morally aware of peers’ pain or suffering. Prior research has found that mindfulness is linked to moral awareness (Ruedy and Schweitzer 2010); therefore, the more egoistic ethical considerations in decision-making lead staff to be morally aware of peers’ pain or suffering, the more likely they are to experience mindfulness. Therefore,

H1a

Egoism-related climates affect empathic concern positively.

H1b

Egoism-related climates affect mindfulness positively.

H1c

Egoism-related climates affect kindness positively.

H1d

Egoism-related climates affect common humanity positively.

As Fehr and Gelfand (2012) propose, the self-transcendent values of benevolence and universalism facilitate compassion, whereas the self-enhancing values of individual achievement and power inhibit it. One reason for these relationships lies in the intricacies of a benevolent climate. Karakas and Sarigollu (2013) propose that in a benevolent climate, individuals are more likely to create observable benefits, actions, or results for the ‘common good.’ Therefore, under a benevolence-related climate, in trying to assure team success, the hotel staff is likely to elicit a sense of attachment to common humanity and deliver compassion (Bryson and Crosby 1992; Daly and Cobb 1989). Moreover, to create observable benefits, actions, or results for the ‘common good,’ a benevolent climate may also exemplify whole-hearted and genuine actions at work that benefit the people around them (Karakas and Sarigollu 2013). Therefore, a benevolent climate may lead hotel staff to exhibit compassion even beyond team achievement. Because they may feel inclined to use their attributes of love and charity specifically (Karakas and Sarigollu 2013), this compassion could take the form of kindness. Moreover, based on Rest’s (1994) theory, Patient and Skarlicki (2010) argued that empathic concern motivates individuals who have (benevolent) moral standards toward others to act based on moral decisions (Jones 1991). This argument suggests that a benevolent climate and empathic concern are related, since the latter seems to expand one’s circle of moral considerations (Aquino et al. 2005). Finally, prior research has shown that individual action triggered by automatic goal activation is usually executed with little to no conscious deliberation or awareness (e.g., Bargh et al. 2001; Fitzsimons and Bargh 2003; George 2009). Certainly, this idea does not suggest that benevolence has effects on mindfulness. However, as benevolence focuses on ‘utilitarianism,’ particularly in the hospitality context, in assuring that the team succeeds, hotel staff may engage in mindfulness as well. In effect, ‘utilitarianism’ seeks the best overall results as goal-based action, and the dynamics of goal-based action may imply the presence of mindfulness, in that the staff ‘needs’ to be aware of the need for compassion in the team. Therefore,

H2a

Benevolence-related climates affect empathic concern positively.

H2b

Benevolence-related climates affect mindfulness.

H2c

Benevolence-related climates affect kindness positively.

H2d

Benevolence-related climates affect common humanity positively.

Specificities of hotel performance suggest the presence of singularities, compared to most businesses, in the causes and the way the staff reacts to the pain or suffering of peers. As stated above, our argument is that the staff is particularly morally aware of peers’ need for compassion because they have to assure that the team succeeds. Although research on reactions to others’ suffering has generally assumed this individualistic and rationally self-interested focus, i.e., ‘what’s in it for me?’ (Treviño et al. 2006), some prior work suggests that hotel staff on a team may also react to the pain and suffering of peers based on principle or deontically, that is, through an automatic and affect-based process, regardless of the cost (Folger et al. 2005). This particular third-party intervention suggests that hotel staff members, deontically concerned about peers’ suffering, are able to display compassionate responses, such as empathic concern and kindness, regardless of whether or not they ultimately impact positive measures of customer response (e.g., Lind et al. 1998; Sheppard et al. 1992; Tyler and Smith 1998; Walster et al. 1978). Since this particular automatic and affect-based intervention often highlights a high degree of automaticity and less conscious reasoning or internal mental processes (Lapsley and Narvaez 2004; Narvaez and Lapsley 2005), principled climate perceptions should include internal decision-making, where mindfulness is less likely to occur (Blasi 2005; Colby and Damon 1992; Oliner and Oliner 1988). This is not the case of common humanity, which, as it does not follow moral exclusion patterns, is not a cognitive process (Opotow 1990) and, hence, may occur through an (deontic) affective process. Therefore,

H3a

Principle-related climates affect empathic concern positively.

H3b

Principle-related climates do not affect mindfulness.

H3c

Principle-related climates affect kindness positively.

H3d

Principle-related climates affect common humanity positively.

Method

Procedure and Sample Characteristics

To test the hypotheses, data were collected from employees at ten hotels in Gran Canaria (Spain’s Canary Islands). Gran Canaria receives about 3.23 million foreign tourists a year, with European countries being its principal markets. British and German tourists jointly represent 41 percent of the total, and Scandinavians make up 28 percent. In all, 304 questionnaires were distributed personally in two sampled two-star hotels, two three-star hotels, four four-star hotels, and two five-star hotels, in percentages ranging from 13 to 49 %.

The research project received official approval. The selected employees met the criterion of working 6 months or more, so that they had a socialization period at the hotel. Fieldwork was performed with random respondents during their time at work, and surveyors asked them to fill out the questionnaires in different places and situations within the hotel, in order to avoid response biases due to uncontrolled contextual conditions. The sample comprised 46.8 % men and 53.2 % women, 32.6 % were 35 years of age or younger, and 11.8 % were 55 years of age or older. In addition, 64.5 % were permanent employees, and the rest were temporary staff. Finally, 29.1 % of the respondents had only finished primary school. Eventually, there were 280 valid responses, after 24 were rejected due to incorrect completion and/or incoherent information.

The data analyses for this study include descriptive analyses, exploratory factor analyses (EFA), and multiple regression analyses, conducted by using the statistical package for social science (SPSS). Descriptive statistics include the mean and standard deviation of empathic concern, mindfulness, kindness, and common humanity, as well as the extracted components of ethical climate, once it is factor analyzed. Cronbach’s alpha is calculated to assess the reliability of the scales, EFA to assess the validity of the measures, and multiple linear regressions to test the hypothesized relationships. To ensure that the criterion variables (empathic concern, mindfulness, kindness, and common humanity) are four separate constructs, exploratory factor analyses (EFA) determine whether all the data load according to the expected four-factor structure. Similarly, ethical climate will be factor analyzed and interpretable components are extracted for our analyses.

Measures

All items were scored on a 7-point Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree), and in the case of empathic concern, mindfulness, kindness, and common humanity, from 1 (never) to 7 (constantly). Items are presented in Tables 1 and 2. Cronbach’s alpha values appear on the main diagonal of the correlations matrix (Table 3).

Table 1 Exploratory factor analysis of the criterion variables in this study
Full size table
Table 2 Factor analysis and reliability of organizational ethical climate
Full size table
Table 3 Means, standard deviations, correlations, and reliabilities
Full size table

Ethical Climate Questionnaire (ECQ)

Individual employees’ ethical climate was measured using the 36-item ECQ developed by Victor and Cullen (1988). The ECQ consisted of measures related to five types of ethical climates: instrumental, caring, independence, rules, and law and code (Cullen et al. 1993; Victor and Cullen 1988).

Mindfulness

Mindfulness was assessed using the 4-item Mindfulness subscale (e.g., “When something painful happens I try to take a balanced view of the situation”) from the Self-Compassion Scale (SCS; Neff 2003). The item “When I’m feeling down I try to approach my feelings with curiosity and openness” was reworded to focus on the suffering of others. Therefore, the opening expression “When I’m feeling down […]” was substituted by “When others are feeling down […].”

Empathic Concern

Empathic concern was measured by the 7-item Empathic Concern subscale of the Interpersonal Reactivity Index (Davis 1980), which assesses feelings of warmth, concern, and sympathy for others. We reversed the scoring of the three items, since they are worded in the opposite direction to empathic concern. These items are, “Other people’s misfortunes do not usually disturb me a great deal,” “Sometimes I don’t feel sorry for other people when they are having problems,” and “When I see someone being treated unfairly, I sometimes don’t feel very much pity for them.”

Kindness and Common Humanity

We adapted the 5-item Self-kindness and 4-item Common Humanity subscales of the Self-Compassion Scale (SCS; Neff 2003). We edited their items, refocusing compassion for the self on compassionate actions and feelings for others. For example, the item “I’m kind to myself when I’m suffering” was replaced with “I’m kind to others when they are suffering,” while the item “I try to see my failings as part of the human condition,” related to Common Humanity, was substituted by “I try to see others’ failings as part of the human condition.”

Results

Tables 1 and 2 display the EFA results. As Table 1 shows, since they did not load properly in their related factors, or their factor loadings were below the cutoff of .2, the EFA results for the four criterion variables suggest rejecting one item on empathic concern, one on mindfulness, and another on common humanity. The remaining items loaded as predicted in the expected factors, confirming four factors with eigenvalues greater than 1 and no cross-loadings over .2, and explaining 70.29 % of the total variance. The coefficient alphas ranged from .691 to .886, around the recommended alpha of 0.70 (Nunnally 1978). We used principal component analysis with Varimax rotation (Table 1 shows this EFA in detail). These patterns provide support for the distinctiveness of the four constructs used as criterion variables in this study.

As Table 2 shows in detail, a second EFA was conducted to determine the underlying structure of the ethical climate measures. The factor structure of the ECQ was analyzed using principal component analysis with Varimax rotation, which yielded six factors with eigenvalues greater than 1 that explained 64.43 % of the total variance, and coefficient alphas ranging from .696 to .823, around the recommended alpha of 0.70 (Nunnally 1978). The factors extracted in this EFA coincide to a large degree with the five types of ethical climate perceptions found by Victor and Cullen (1988): (a) instrumental, (b) caring, (c) law and code, (d) rules, and (e) independent. As Table 2 shows, our analyses of the ECQ factor structure found six types of ethical climate, which either coincide with Victor and Cullen’s (1988) factors or are breakdowns or combinations of them. Thus, our ‘rules and codes’ factor includes Victor and Cullen’s ‘law and code’ and ‘rules’ factors. The ‘social responsibility’ factor is somewhat similar to Victor and Cullen’s ‘caring,’ except that the former focuses exclusively on the cosmopolitan level. Finally, the ‘moral caring’ and ‘personal morality’ climate factors found in this study seem to be a breakdown of Victor and Cullen’s ‘independence’ climate perception, whereas the ‘efficiency’ and ‘self-interest’ factors are breakdowns of the ‘instrumental’ perception.

Table 3 displays the means, standard deviations, reliabilities, and correlations (r) among all the variables. Results show significant inter-correlations in the expected directions between the ethical climate factors and the variables making up compassion, indicating general support for the basic thesis of this study. To test the hypothesized relationships, multiple regression analyses were conducted (Aiken and West 1991), considering models with empathic concern (Model A), mindfulness (Model B), kindness (Model C), and common humanity (Model D) as the criterion variables (Table 4). Since the ‘efficiency’ and ‘self-interest’ factors represent egoism-related climates, both were used to test H1. Only support for H1a, 1b, (but not 1c and 1d) is found (see Table 4), due to the positive significant links from the ‘efficiency’ climate factor to empathic concern (B = .155; p < .05) and mindfulness (B = .222; p < .01). The null effect of ‘efficiency’ on kindness (B = .068; p ns.) and common humanity (B = .116; p ns.), along with the fact that the ‘self-interest’ climate factor did not show effects on any of the compassion variables (see Table 4), did not add further support to H1. In the case of benevolence-related climate, only ‘social responsibility’ perceptions were considered to test H2. ‘Social responsibility’ is significantly and positively associated with kindness (B = .245; p < .001) and negatively with common humanity (B = −.239; p < .01), and it has a null effect on mindfulness (B = .019; p ns.) and empathic concern (B = .138; p ns.). These patterns only support H2c. Finally, the ‘rules and codes’ and ‘personal morality’ factors are used to test H3, that is, how principle-related ethical climate influences employee compassion. H3a and H3c are supported, while H3b is rejected in the case of mindfulness, by the significant positive effects of ‘rules and codes’ on empathic concern (B = .178; p < .05), kindness (B = .278; p < .001), and mindfulness (B = .215; p < .01). Furthermore, principle-related ethical climate positively and significantly influences common humanity (H3d) through ‘personal morality.’ These results support H3d.

Table 4 Results of multiple linear regression analyses
Full size table

Discussion

The purpose of this study was to explore the usefulness of organizational ethical climate in predicting compassion at work in the hospitality industry. To the extent that ethical climate perceptions encourage employees to be compassionate at work, these perceptions inspire psychological processes from a compassionate approach, such as empathic concern, mindfulness, kindness, and common humanity. Specificities of hotel performance also suggest that proper ethical climate perceptions encourage hotel staff to be more willing to respond to the need for compassion within their work teams, helping them to be successful in terms of guest responses. Overall, as expected, benevolence-related and, to a higher extent, egoism-related and principle-related climates support our predicted influences on compassion. Based on these study results, we will offer several theoretical and practical implications for behavioral ethics in hospitality organizations. Finally, the paper opens up several avenues for future research.

A comprehensive first look at the results of this study shows some disparity in the way the ethical climate factors influenced the four studied psychological processes involved in compassion. Thus, while the positive relationships between the egoism-related ‘efficiency’ factor and empathic concern and mindfulness seem to overtly support our predictions, ‘efficiency’ did not show significant effects on kindness or common humanity, and ‘self-interest’ showed no effects on any of the processes. The benevolence-related ‘social responsibility’ factor also seems to perform with certain inconsistency, and while it significantly predicts kindness, contrary to expectations it showed a surprising significant negative effect on common humanity. Lastly, although the principle-related factors, either concerning ‘rules and codes’ or ‘personal morality,’ were capable of predicting all the studied elements of compassion, in the case of mindfulness, it performed contrary to expectations. Therefore, there is a need for further discussion to shed light on the raison d’être for these results.

First, an explanation for the existing differences in the significant influences of egoism-related climates on compassion may stem from our premises about the specific context in which this study was conducted. As mentioned earlier, hospitality industry organizations have specific characteristics that suggest that staff members are particularly morally aware of events involving peers’ suffering, and that they deliver compassion to them basically out of a principle of efficiency. The results in general support this idea. Thus, the egoism-related ‘efficiency’ factor showed significant effects on empathic concern and mindfulness, suggesting that staff members act in this way because they need to support to each other to achieve the reward derived from potential positive responses of customers. However, surprisingly, ‘self-interest’ did not show significant effects on any criterion variable. In our view, although ‘self-interest’ is also guided by egoism-related perceptions, since it performs at the individual level it may foster self-interested norms instead. Self-interested norms are known to inhibit helping behaviors (Ehrhart and Naumann 2004; Miller 1999), which would explain why self-interest did not affect compassion.

Second, the results point to principled climate perceptions as more consistent and intense predictors of compassion at work than others, as operationalized. In effect, concerning either ‘rules and codes’ or ‘personal morality,’ we found that the principle-related factors predicted all the studied elements of compassion and were probably emotionally driven by moral obligation. In other words, influenced by a principle-related climate, staff members seem to be moved to act out of moral obligation, turning quid pro quo compassionate responses to peers’ pain or suffering (seeking customer responses above all) into a ‘deontic’ compassion (customer responses move to the background). Thus, in making sure that suffering peers are supported, the principle-related factors appear to increase their compassion out of a moral imperative, intrinsically motivating them to support peers regardless of the cost. As the mechanisms used by egoism-related and principle-related climates appear to be different in impacting compassion in hotel settings, they could explain why ‘personal morality’ and ‘efficiency’ did not affect mindfulness, kindness, and common humanity in the same way, while ‘rules and codes’ and ‘efficiency’ did not affect kindness either.

In addition, we originally argued that principled climate perceptions should include internal decision-making (Blasi 2005; Colby and Damon 1992; Oliner and Oliner 1988), thus predicting negative influences on mindfulness. However, the positive effects of principled ethical climate on mindfulness imply acting contrary to these expectations. One way to explain this result may be to explore the intricacies that mindfulness shows as a phenomenon closely related to the process of compassion. Mindfulness is a so-called psychological flexibility factor that mainly seems to intervene in (rather than being part of) the compassionate process (Tirch 2010). Thus, mindfulness is a psychological state that seems to lubricate compassion’s sub-processes (noticing, feeling, and responding), by reducing, for example, the immediate reaction and sense of threat, or by increasing self-regulation (Atkins and Parker 2012). However, mindfulness is not full compassion itself, and it can be present in other individual and organizational processes apart from compassion. Dane (2011) found, for example, that mindfulness and positive measures of task performance could be directly linked, suggesting that the principle-related climate may influence mindfulness by indirectly boosting other aspects of psychological flexibility which are necessary for hotel staff to engage in task performance.

Third, regarding the effects of ‘social responsibility’ on kindness, our study results also supported benevolence-related climates as significantly fostering compassion. This finding is consistent with Swaen and Maignan’s (2003) argument that corporate social responsibility (CSR) can affect employees’ adoption of socially responsible behavior within organizations. Moreover, Moon et al. (2014) found that corporate social responsibility (CSR) predicts compassionate acts, with organizational justice acting as a mediator. However, ‘social responsibility’ failed to significantly affect empathic concern or display a significant positive influence on common humanity. The results showed, moreover, that ‘social responsibility’ and common humanity are negatively related. Simpson et al. (2014) may provide a reason for this unexpected result when they argue that not just any suffering person is seen as a legitimate and worthy recipient of compassion (Simpson et al. 2014). Impregnated with a benevolent climate, when achieving proper customer responses is at stake, socio-responsible staff can find it difficult to see peers’ failures in the context of shared human fallibility. Specificities of hospitality industry settings can support this because, compared to most businesses, the hospitality industry sector is highly competitive and creates tensions and conditions that can culturally permeate a hotel, making compassion much less likely to occur (George 2014). Thus, influenced by misunderstood social responsibility, the staff may opt to delegitimize (and even blame) the peer-sufferer. Although common humanity advocates diluting the moral strength of social judgments rooted in hospitably industry contexts, benevolent climates, as utilitarian situations, can make it easier for socio-responsible hotel staff to enforce a strict socio-moral code on peers, where proper customer responses are not at stake.

The findings can offer some practical implications. The inclusion of both egoism-related and principle-related climates as reasons for the compassionate process combines the importance of compassion as a moral imperative response with regulated responses involving cognition. The findings seem to suggest, for example, that hospitality organizations should place an emphasis on both egoism-related and principle-related climates, and that each of them motivates employee compassion by following different patterns. Practical implications derived from an egoism-related approach suggest, for example, that transactional leaders should be welcomed because they emphasize an adherence to standards that favor cognitively regulated quid pro quo dynamics that trigger the compassionate process. After all, under an egoism-related climate, compassionate responses automatically follow from noticing peers’ suffering because they stem from work teams where staff members have to be supported by each other because their work is rewarded by customer responses (Gittell and Douglass 2012; Grant and Patil 2012). In principle-related climates, however, the results show that peers’ suffering becomes internalized by staff to the point of abandoning quid pro quo dynamics to now follow deontic patterns, i.e., an automatic and affect-based process developed by the deonance theory of fairness (Folger 2001; Folger et al. 2005). In this case, altruistic and ethical supervision in the workplace should be welcomed, due to the influential key role that this type of leadership plays in creating deontic contexts (Karakas and Sarigollu 2013; Zoghbi-Manrique-de-Lara and Suárez-Acosta 2014). Finally, as the results for social responsibility (representing benevolent climate) inconsistently affected kindness and common humanity, they do not suggest clear patterns with which to delineate an ethical strategy that can foster compassion at work.

Limitations, Future Research, and Conclusions

Some questions remain that could form the basis for future research. First, although the psychological states and traits used in this study to measure the compassionate process are well grounded in the existing literature on compassion, future research is welcome to go deeper into the intricacies and appropriateness of the ‘elements of compassion’ used in this study, considering other compassionate variables, indicators, and psychological processes of interest. Second, future research on compassion can also include the impact of noticing suffering on the compassion factors studied here. For example, staff members’ observations of events of injustice for others (peers, customers, etc.,) can be studied as an operationalized measure of noticing suffering. Finally, future research on compassion can not only examine the impact of the studied compassion factors on measures of compassionate response in the hospitality industry, which can include different types of helping behavior toward peers, but also citizenship and pro-social behavior directed at hotel guests.

Regarding the limitations of our study, we acknowledge that the study contains weaknesses. First, it might suffer from mono-method/source bias. Second, this study was conducted in a hospitality industry context, and the data stem from a limited universe, raising concerns about the generalizability of the findings to other types of businesses. Since the surveyed hotel employees have certain job conditions that are often inherent to their particular role in the hospitality industry, the performance of the constructs used in the present research, as well as their implications, could vary.

This paper, on the other hand, contributes to better understanding how the ethical climate can influence hotel employees’ awareness of peers’ pain and suffering, and their motivation to participate in expressing compassion to them. The conclusions, narrowed to the hospitality industry, highlight that, through both moral imperative and as regulated responses involving cognition, hospitality organizations can develop ethical climate strategies that elicit compassion. Promoting this twofold ethical way of acting compassionately, in turn, hotel managers can promote staff members’ success across their work teams, either moving them to act compassionately out of a principle of efficiency, or encouraging them to support each other out of moral obligation.