We conducted a systematic keyword search using the search engines PsycINFO and ABI/INFORM Global. Keywords included the following terms: Employee Green Behavior, Employee Pro-Environmental Behavior, Organizational Citizenship Behavior for the Environment (OCBE), and Employee Eco-Initiative, as well as combinations of the terms Organizational Citizenship Behavior, Employee, and Behavior, with Environment, Green, and Sustainable (see Table 1 for the complete overview of search terms). We specifically searched for articles written in English, in peer-reviewed scholarly journals, covering adult populations. This resulted in a total of 1,031 identified articles on PsycINFO and 1,515 articles on ABI/Inform Global.

We then scanned titles and abstracts of the identified articles to ascertain their relevance for this review. We excluded all articles that did not focus on (a) individual behavior, (b) behavior performed at work, and/or (c) behavior related to sustainability- or green goals. Furthermore, we excluded all articles that did not discuss antecedents of EGB. Finally, we excluded all double articles, resulting in a final total of 63 articles via PsycINFO and 69 articles via ABI/INFORM Global.

In the next step, we carefully read and coded these 132 articles in a schematic overview, extracting data on each article's study design, sample, sample size, measurement, studied predictors, and core theories, among others. We further narrowed down the selection of relevant articles by removing articles that did not report results of quantitative or mixed-method empirical studies. We also had to remove two articles that were retracted by their authors. Finally, and importantly, we removed articles that studied negative or neutral predictors of EGB (e.g., barriers to EGB). Note that we did not include the term positive psychology as keyword in our systematic literature search because the majority of articles on EGB cover constructs related to positive psychology without explicitly referring to them. In this step, we therefore carefully reviewed the predictors studied per article and selected only those studies that investigated constructs with clear links to the three pillars of positive psychology (i.e., positive subjective experiences, positive individual characteristics, and positive institutions) (Seligman and Csikszentmihalyi, 2000; Donaldson and Ko, 2010).

This resulted in a final sample of 94 articles (89 quantitative/5 mixed method) included in this review. Out of the 94 identified articles, 71 draw on cross-sectional data, 19 draw on cross-sectional, but time-lagged data, and four draw on longitudinal data. Out of the longitudinal studies, two are daily diary studies and two are (quasi-) experimental studies. To investigate EGB, most studies relied on self-report surveys, with a minority (n = 19) making use of other-ratings, mostly supervisor-ratings.

The samples included in these studies consisted of workers from all over the globe, with China (k = 13), Vietnam (k = 11), Pakistan (k = 8), and the UK (k = 6) as the four most represented countries. Overall, it appears that the empirical literature on employee green behavior is predominantly originating from Asian countries (k = 50), followed by European countries (k = 21), and Northern American countries (k = 12). Workers included in the samples worked in a variety of industries, but most studies were either conducted in the tourism and hospitality industry (k = 31), or with (convenience) samples of workers from different industries (k = 28). Sample sizes ranged from N = 47 to N = 1,230, with an average sample size of N = 367.

Following Seligman and Csikszentmihalyi (2000) and Donaldson and Ko (2010), we categorized predictors of EGB as belonging to the first pillar of positive subjective experiences when they pertained to positive feelings and other relatively fluctuating positive states, including happiness, wellbeing, flow, pleasure, hope, optimism, and positive emotions (Seligman and Csikszentmihalyi, 2000; Donaldson and Ko, 2010). We categorized variables under the second pillar of positive individual characteristic when they described character traits of individuals and other relatively stable (trait-like) attributes of a person, such as talents (knowledge, skills, or abilities), interests, creativity, wisdom, values or attitudes, character strengths, and courage (Seligman and Csikszentmihalyi, 2000; Donaldson and Ko, 2010).

The categorization of predictors under the third pillar of positive institutions was slightly more complex. In their original work on the different pillars of positive psychology, Seligman and Csikszentmihalyi (2000), described the third pillar as follows: “At the group level, it is about the civic virtues and the institutions that move individuals toward better citizenship: responsibility, nurturance, altruism, civility, moderation, tolerance, and work ethic” (p. 5). In line with this, van Rensburg and Rothmann (2020) define positive institutions as entities that:

These definitions of positive institutions highlight that institutions are typically “shared” understandings or expectations, and manifest at the group level. Following these definitions, we categorized under general positive institutions different features of the organization or work group that move individuals within these organizations or groups toward better citizenship, often by entertaining a shared vision (think, for instance, of green human resource management), or by honoring common values such as safety, fairness, humanity, and dignity (think, for instance, of ethical leadership). At the same time, we found in the literature operationalizations of positive institutions as internalized by individuals. This is maybe not surprising given that institutions can be understood as formal rules or informal norms that govern the behavior of organizations, as well as individuals (Allard and Small, 2013). In the remainder of the manuscript, we therefore use the term internalized positive institutions to refer to the understood responsibilities or duties of individuals (think, for instance, of an individual's values or norms). Note that internalized institutions capture an individual's understanding of how a person should be or should behave, with reference to behavior that is considered good vs. bad, or right vs. wrong. In contrast to that, positive individual traits (Pillar 2) are simply about how a person is (free of judgment of whether this is right or wrong/good or bad).

Forty-eight articles investigated one or more predictors (k = 58 predictors in total) to EGB that we considered positive subjective experiences based on their respective definition and operationalization. These were commitment (k = 17), fit (k = 11), motivation (k = 7), autonomy (k = 5), job satisfaction (k = 4), passion (k = 4), meaningfulness (k = 2), state-like self-efficacy (k = 2), trust (k = 2), wellbeing (k = 2), daily affect (k = 1), and environmental engagement (k = 1).

Overall, those articles widely supported the positive relationship between positive subjective experiences and employee green behavior. For instance, 16 of the 17 articles investigating commitment—be it to the organization (k = 8), the environment (k = 7), colleagues (k = 1), or goals (k = 1) (e.g., Paillé et al., 2016)—found support for a positive relationship with EGB. Only one study found mixed results, with a positive and significant correlation coefficient, but a non-significant regression coefficient for organizational commitment (Afsar et al., 2020). In addition, all 11 articles investigating individual perceptions of fit, or other forms of perceived value congruence between a person and their environment such as organizational identification, supported a positive relationship with EGB (e.g., Su and Swanson, 2019). Similarly, intrinsic motivation (k = 7), autonomy (k = 5), passion (k = 4), meaningfulness (k = 2), self-efficacy (k = 2), trust (k = 2), employee wellbeing (k = 2), and environmental engagement (k = 1) were all found to be positively related to EGB.

Only few of the investigated predictors showed less conclusive results. First, while two studies investigating job satisfaction found a positive relationship with EGB (Paillé and Boiral, 2013; Kim et al., 2019a), two other studies found either no, or a negative relationship (Paillé and Mejía-Morelos, 2014; Paillé et al., 2016). Second, a daily diary study on positive affect in relation to EGB (Bissing-Olson et al., 2013) indicated that daily unactivated positive affect (i.e., “contentment”) was positively related to EGB, whereas daily activated positive affect (i.e., “excitement”) was not.

Thirty-four articles investigated one or more predictors of EGB that were classified as positive individual traits or trait-like characteristics (k = 36 traits in total). These included a person's pro-environmental attitude (k = 23), green identity (k = 4), green competency or ability (k = 2), empathy (k = 2), internal locus of control (k = 2), future time perspective (k = 1), and generalized self-efficacy (k = 1).

Again, most articles supported the positive relationship between the investigated positive traits and EGB, in particular, for traits that directly relate to green or pro-environmental concerns. For instance, the most widely studied trait (-like) variable was pro-environmental attitude, referring to an individual's general awareness of environmental problems or belief in humanity's negative impact on the environment. It has been studied under several terms including workers' environmental concern (k = 9), -knowledge (k = 7), -awareness (k = 6), and -beliefs (k = 1). Overall, results of 21 out of 23 studies indicate a positive relationship between pro-environmental attitude and EGB (e.g., Zientara and Zamojska, 2018; Paillé et al., 2019). One study reported mixed results (Ahmed et al., 2020), meaning that the relationship between environmental knowledge and -concern and EGB was significant and positive, while the relationship between environmental awareness and EGB was not. One other study found a negative relationship between environmental beliefs and EGB (Chou, 2014).

Similarly, all four studies focusing on green identity, indicating that being environmentally friendly is integrated into a person's self-view (Luu, 2020a), found positive relationships between green identity and EGB. Note that one of the four studies only supported this relationship for two out of three dimensions of EGB (e.g., Luu, 2020b). Finally, green competence and—ability were also found unanimously to be positively related to EGB (Subramanian et al., 2016; Rayner and Morgan, 2018).

The results for more general positive traits such as empathy and internal locus of control were mixed. Empathy was found to be unrelated to EGB in two studies (Islam et al., 2019; Tian and Robertson, 2019). Internal locus of control (as part of an individual's core self-evaluation) was found to be positively related to EGB in one study (Robertson and Carleton, 2018), yet unrelated in another (Afsar et al., 2020). Results for generalized job-related self-efficacy pointed to a non-significant correlation between self-efficacy and EGB (Paillé et al., 2019). By contrast, future time perspective (having a more open-ended perception of the future) was positively related to EGB (Jiang et al., 2019).

In total, 76 articles were identified within the third pillar, out of which 7 covered internalized institutions, 60 covered general positive institutions, and 9 covered both types of institutions. This means that a total of 16 articles covered internalized institutions, while 69 articles covered general institutions.

While the above text focuses on direct relationships between various predictors and EGB, many of the articles included in our review develop and test complex research models, including mediators, moderators, or both. We were able to derive some general patterns in results regarding moderators and mediators.

First, variables pertaining to pillar 1 or positive subjective experiences are commonly studied as mediators in the included articles. For example, commitment (k = 13) and fit (k = 10) have consistently been studied as a mediator, serving as the mechanism that links general positive institutions (e.g., perceived organizational support, green HRM, or positive leadership) and EGB (see, e.g., Paillé and Morelos, 2017; Afsar and Umrani, 2020). Other variables that were found to mediate the link between general positive institutions and EGB are environmental engagement (Luu, 2019a), passion (Robertson and Barling, 2013; Afsar et al., 2016), self-efficacy (Testa et al., 2020), trust (Su and Swanson, 2019), and wellbeing (Ahmed et al., 2020).

Second, in contrast to pillar 1 variables, pillar 2 variables (positive individual traits) have often been included as moderators in the selected articles. For example, pro-environmental attitude was included as a moderator in several studies (k = 5). Among others, it was found to weaken the relationship between daily activated positive affect and EGB (Bissing-Olson et al., 2013). Another example is green identity, which was found to strengthen the relationship between both leadership (Wang et al., 2018; Luu, 2020b) and green communication as part of HRM (Luu, 2020a), and EGB. Other positive individual trait variables that have been investigated as moderators were empathy (Tian and Robertson, 2019), internal locus of control (Robertson and Carleton, 2018; Afsar et al., 2020), future time perspective (Jiang et al., 2019), and general self-efficacy (Paillé et al., 2019).

Third, internalized positive institutions (Pillar 3a) have been studied as both moderators and mediators in the included articles. On the one hand, values have been studied as moderator in two studies, yielding mixed results. The first study found that green values strengthen the relationship between green climate and voluntary EGB, but do not influence the link between green climate and task EGB (Dumont et al., 2017). By contrast, another study found that strong green values might weaken the relationship between Corporate Environmental Responsibility (CER) and EGB (Ruepert et al., 2017). On the other hand, morality has been found to mediate the relationship between both personal characteristics such as conscientiousness (Kim et al., 2017) and EGB, and organizational initiatives such as CSR (Afsar and Umrani, 2020) and HRM (Luu, 2020a) and EGB (k = 3). Because there are comparatively fewer studies that include variables pertaining to this pillar, we cannot derive any definite patterns based on these results yet.

Fourth, general positive institutions (Pillar 3b) have often been studied as predictors in the included articles, exerting their influence on EGB via, for instance, positive subjective experiences (see Pillar 1). However, several selected articles also pointed out potential synergies between different general positive institutions in moderation analyses. For instance, both ethical (De Roeck and Farooq, 2018) and servant leadership (Luu, 2017, 2019b) were found to strengthen the positive relationship between CSR and EGB (k = 3). Moreover, green climate was found to strengthen the positive relationships between, respectively, green HRM (Luu, 2018), leadership (Khan et al., 2019), and training (Pham et al., 2018), and EGB (k = 5). In stark contrast to these synergetic effects stand the results of studies investigating green climate as a moderator of the relationships between positive individual traits or internalized institutions and EGB. In these studies, green climate was found to weaken the link between both personal green norms (Chou, 2014) and attitudes (Tian et al., 2020) and EGB (k = 3). Finally, studies have also recognized that one positive institution may influence EGB via another positive institution in mediation analyses. Most notably, green climate has been supported as a mediator (k = 8) in the relationships between organization-level variables such as leadership (e.g., Kim et al., 2017; Khan et al., 2019) and green policies (Norton et al., 2014), and EGB.

Various theories may explain these positive relationships. In the reviewed articles, the most commonly referenced theory is the Theory of Planned Behavior (TPB; Ajzen, 1991) (n = 17), followed by Social Identity Theory (Tajfel and Turner, 1986) (n = 13), and Social Exchange Theory (SET; Blau, 1964) (n = 12). Other theories that were drawn on repeatedly are the Value Beliefs Norms Theory (VBN; Stern, 2000) (n = 7) and Self-Determination Theory (SDT; Ryan and Deci, 2000; Deci and Ryan, 2008) (n = 6). Finally, due to its special place in positive psychology, we also refer to the broaden-and-build theory of positive emotions (Fredrickson, 1998, 2001) mentioned in n = 2 of the reviewed articles. Below, we outline how these theories can be used to explain the main results of our review per pillar of positive psychology.

First, our literature review unveiled consistent positive relationships between positive subjective experiences (Pillar 1; e.g., commitment, motivation, job satisfaction, positive affect) and EGB. These relationships have been commonly explained by SDT (Ryan and Deci, 2000; Deci and Ryan, 2008), proposing that autonomous motivation (including intrinsic motivation) does not only stimulate individual wellbeing and performance, but also individual perseverance and consistency in the display of behaviors. The theory may therefore explain the positive relationships between (green) intrinsic motivation and EGB, in particular, concerning voluntary (extra-role) EGB (see, e.g., Tian et al., 2020). Furthermore, two studies draw on the broaden-and-build theory of positive emotions (Fredrickson, 1998, 2001) to explain this relationship. This theory suggests that subjective experiences of positive emotions broaden an individual's momentary thought-action repertoire, leading to the exploration of new and creative action alternatives. Literature suggests that green behaviors are examples of such new and creative actions (Bissing-Olson et al., 2013; Kim et al., 2019a). However, Bissing-Olson et al. (2013) also note that these behaviors can only be considered new and creative for people who are normally not very concerned about the environment. People who are concerned about it, will habitually display green behaviors at work. This contingency on environmental concern may explain why the expected positive relationship between satisfaction or affect and EGB was not corroborated in a minority of studies (Bissing-Olson et al., 2013; Paillé and Mejía-Morelos, 2014; Paillé et al., 2016).

Second, our review corroborated positive relationships between positive individual characteristics (Pillar 2; e.g., pro-environmental attitude, green identity, locus of control, future time perspective) and EGB. These relationships have often been explained by two related theories. The first is TPB (Ajzen, 1991), postulating that behavior—including green behavior—is strongly determined by a person's behavioral intentions. Behavioral intentions, in turn, result from attitudes toward the behavior in question, subjective norms, and perceived behavioral control. In relation to the second pillar, this theory has been used to explain the relationships between positive attitudes toward green behavior and EGB (e.g., Lamm et al., 2013; Wang, 2016), as well as between self-efficacy (as an indicator of behavioral control) and EGB (Testa et al., 2020). The second related theory is the VBN Theory (Stern, 2000), proposing that values influence pro-environmental behavior via pro-environmental beliefs and personal norms. In line with this, VBN theory has been used to explain that employees are more likely to engage in EGB when they have pro-environmental attitudes or beliefs (see, e.g., Verplanken et al., 2008; Rezapouraghdam et al., 2018). These theories can also be used to explain why more general positive traits such as empathy are not always found to be positively related to EGB. These more general traits are not specifically linked to attitudes or norms regarding the environment, and would thus not necessarily translate into intentions regarding green behavior (Ajzen, 1991).

Third, the same two theories (TPB and VBN) have been used to explain the relationship between a person's internalized positive institutions (Pillar 3a; e.g., values, morality, norms, and stages of consciousness) and EGB. These theories argue that internalized institutions, such as values and norms, are strong predictors of displayed behaviors. Within the reviewed articles, TPB and VBN have been used conjointly to explain the relationship between, respectively, work ethic (Peng and Lee, 2019) and norms (Chou, 2014), and EGB. In addition, TPB has been used to explain the link between values and EGB (Boiral et al., 2015). Interestingly, the use of the same theories under pillar 2 and 3a may indicate that the distinction between positive individual traits and internalized positive institutions is a subtle one, and that the positive characteristics that matter for EGB may be closely tied to internalized institutions. However, we still see an important distinction between the two in that positive individual traits capture how a person is (how would you describe that person?) without passing judgment, whereas internalized positive institutions capture beliefs or convictions about how a person should be or behave (how should the person behave?), which entails a normative judgment of right and wrong, or good and bad.

Moreover, our review corroborated positive relationships between general positive institutions (Pillar 3b; e.g., support, positive leadership, GHRM, green climate, CSR, and green strategy) and EGB, which have commonly been explained by social exchange theory (SET; Blau, 1964) in the reviewed literature. This theory suggests that social exchange relationships depend on a careful balance of giving and taking among the involved parties. One party's investment in the relationship thus compels the other party to reciprocate with a similar investment (Gouldner, 1960). Social exchange relationships can also form between organizations and their employees (Rhoades and Eisenberger, 2002), where organizational investments in employees, which go beyond investments that are determined by the employment contract (e.g., extra HR initiatives or support), compel employees to return the favor. The social exchange reasoning suggests that employees, who perceive initiatives such as green HRM, green climate, or CSR as an extra investment in them, will engage in green behaviors to fulfill their perceived social obligations toward the organization (see, e.g., Gkorezis, 2015; Bhatnagar and Aggarwal, 2020). However, note that the lack of written agreements in social exchange relationships may also explain why some studies did not find the expected relationship between general positive institutions and EGB (e.g., Norton et al., 2017; Maziriri and Saurombe, 2018). While investments by the organization or leader will compel employees to reciprocate, the employees will still have to make an informed guess on the type of reciprocation that is most valued by the other party. If employees do not perceive strong signals that green behavior is highly valued by the other party, or if they perceive other signals to be stronger (e.g., that cost savings are valued), they may reciprocate with other behaviors than green behaviors. At the same time, the relationship between general positive institutions and EGB may also be explained by economic exchange relationships between organizations and their employees. Employees will engage in green behaviors if their contractual or economic exchange relationship with the organization mandates it and some organization clearly steer employees toward green behavior through their green policies, procedures, or practices (e.g., through rewards for green behavior as part of GHRM).

Finally, to explain the mediating roles of positive subjective experiences (most notably, commitment and fit) in the relationship between general positive institutions and EGB, the reviewed articles often refer to social identity theory (Tajfel and Turner, 1986). This theory suggests that individuals' self-concepts are highly contingent on their social group memberships. To bolster their self-concepts, individuals strive to be members of social groups with a high public standing. Pursuing green or sustainable initiatives (e.g., CSR, green HRM, green climate, green strategy) improves the public standing of organizations. Consequently, these initiatives entice employees to solidify their membership of these organizations, by strongly attaching themselves to the organizations (e.g., through commitment) or by integrating organizational values into their self-view (e.g., through organizational identification or fit) (Kim et al., 2019b; Su and Swanson, 2019). Subsequently, employees will be eager to display behavior (e.g., green behavior) that further benefits the standing of the organization they are committed to or identify with. By extension, this will also benefit their self-concept (Su and Swanson, 2019).

Our literature review suggests that pursuing positive avenues toward stimulating EGB is worthwhile because the review uncovered consistent positive relationships between variables relating to the three pillars of positive psychology and EGB. At the same time, the current state of theoretical development indicates that there is little explicit emphasis on positive psychology theorizing to date. We thus foresee many relevant avenues for future research at the intersection of positive organizational psychology and EGB (for an overview, see Figure 1).

To pursue the above avenues for future research, we suggest that scholars move away from the cross-sectional designs that have dominated EGB research (Norton et al., 2015) toward other, more advanced, research designs including longitudinal and multi-level designs. Specifically, daily diary studies such as in Bissing-Olson et al. (2013) or experience sampling method (ESM) studies (Larson and Csikszentmihalyi, 2014) are needed to investigate links between momentary experiences (e.g., emotions) and EGB. Moreover, intensive longitudinal studies, with several waves of daily diary- or ESM data collection separated by several weeks or months (for an example, see Quintus et al., 2020), allow researchers to link day-to-day or momentary experiences to longer term developments. These studies would be uniquely suitable to test not only the broadening-, but also the building hypothesis of positive emotions (Fredrickson, 2001), as well as processes of green identity development (Dutton et al., 2010). Furthermore, scholars may use (quasi-) experimental research to test the effects of targeted, positive workplace interventions on EGB, via constructs such as PsyCap (Luthans and Youssef-Morgan, 2017), strengths use (Miglianico et al., 2020), or gratitude (Emmons and McCullough, 2003). It would be interesting to consider boundary conditions of effects because workplace interventions may not always work for all employees under all circumstances (Unsworth et al., 2013; Nielsen and Miraglia, 2017). To conduct research on general positive institutions (e.g., organizational virtuousness) and EGB, researchers should conduct multi-level studies accounting for the fact that employees are nested in work teams or organizations (Norton et al., 2015). In addition, to capture the complex social dynamics that may contribute to EGB (e.g., amplifying dynamics, spread of energy, being inspired by leaders), we require dyadic, or even better, social network studies (Scott, 1988). The latter would allow to explore how and where positive movements originate, as well as which people are involved in which role (as energizers, followers, gatekeepers, etc.).

While this review opens up novel research directions at the intersection of positive organizational psychology and EGB, it is also subject to some limitations. First, we only searched for articles using overarching key terms (e.g., EGB) and not specific, single green behaviors (e.g., recycling, using public transport). In that, we followed Ones and Dilchert (2012) assuming that single indicators are too narrowly focused to be useful in building a scientific understanding of employee green behaviors. Second, we only considered articles that explored green behavior in the work context, but not other contexts, most notably the context of a person's home. Research has shown that green behavior are not always transferable across contexts, as each context entails specific incentives, obstacles, and opportunities (Lo et al., 2012). For instance, the wish or need to cut down own expenses may act as a strong incentive to conserve resources (e.g., water, electricity, gas) at home, that is not present at work. This calls for context-specific examinations. Third, the finding that almost all included studies reported positive relationships between the investigated variables and EGB needs to be interpreted with some caution. Although the open science movement has led to reductions in the file drawer problem (the tendency to only publish signifcant results; Rosenthal, 1979), the relationship between positive constructs and EGB may still be overestimated based on the included studies. Relatedly, there still are considerable shortcomings in the scientific rigor in several of the studies included in this review (e.g., cross-sectional designs; ambiguous definition or operationalization of key variables) that necessitate us to interpret the findings of this review with some caution.