Introduction

In the twenty-first century, globalization has accelerated the contact between cultures. Competition and collaboration now extend across geographical boundaries. Increasing mobility translates into growing interaction between people from diverse cultural backgrounds, both within and among countries. This places tremendous challenges on organizations to manage diverse perspectives of what justice is and how it should be applied.

Much is known about organizational justice perceptions, yet there is limited knowledge about how the sociocultural context affects them. Cross-cultural research has focused mainly on national differences in event justice, attributed to isolated cultural dimensions. Literature reviews have focused on particular aspects of the justice experience, such as conceptualization (Cole, 2009), international business (Dixon, Caldwell, Chatchutimakorn, Gradney, & Rattanametangkul, 2010), or expatriates (Toh & Denisi, 2003). This has advanced our knowledge, but not led to the integration of cultural, situational, and individual influences on the development of and reactions to justice judgements.

A systematic review of cross-cultural organizational justice literature is important as it can give us a clearer sense of where we have been and where we are going. On the one hand, throughout the years, many have stated the importance of contextual aspects to understand how organizational justice works in different cultures (e.g. Leung, 1987; Shao, Rupp, Skarlicki, & Jones, 2013). Country differences have been attributed to numerous factors at different levels, such as national values, institutional, or organizational context. However, approaches tended to focus only on one level. In order to promote a broader more holistic view of cross-cultural influences on organizational justice across cultures, it is important to account for what factors operate on different levels and how they interact. On the other hand, critical voices have been raised regarding how organizational justice is studied across cultures (e.g. Cole, 2009; Greenberg, 2001). This criticism relates to methodological and conceptual issues, such as what is the cross-cultural validity of justice constructs, how are cultural values measured, or how to integrate factors at different contextual levels? A systematic review may help to better understand how these challenges are being meet and suggest some ideas on how to better overcome them in the future.

Accordingly, we systematically reviewed the cross-cultural literature on organizational justice, in order to assess the range of conceptual and methodological frameworks and propose a future research agenda. In this review, we describe how cultural, situational, and individual contextual levels influence the relationship between justice perceptions and its antecedents and consequences across countries. Namely, we explore their influence over (1) what justice criteria people use; (2) how they apply them to develop justice perceptions; (3) what reactions people have to (un)fairness; and (4) how they express them.

Systematic reviews, unlike selective reviews, use reproducible procedures to search and critically appraise the literature and limit researchers’ biased preconceptions (Crossan & Apaydin, 2010; Tranfield, Denyer, & Smart, 2003). Therefore, boundary conditions are established and all elements that fit those conditions are analysed. We searched academic databasesFootnote 1 and included all English-language peer-reviewed articles up to 1 January 2014 containing the search terms: “organizational justice” or “organizational fairness”, and “culture” or “country”, or synonyms. We included only empirical studies that drew samples from at least two different countries. Another boundary condition of our review is that we focused on findings pertaining to antecedents and consequences of justice from the receiver perspective. Hence, our final sample consists of 74 papers.Footnote 2

In the following section, we describe the methodologies and conceptualizations used in cross-cultural organizational justice literature. Then, we review findings regarding the main sociocultural contextual levels that affect justice across cultures. Lastly, we discuss the implications.

Methodological Overview

Of the studies collected 14 are literature reviews, 60 are cross-cultural empirical studies, of which five are meta-analysis. Table 1 presents the designs, sample, and topics covered in the 74 papers collected. The studies were published from 1978 up to 2014. “Distributive criteria” was the most investigated topic (24 %), followed by “Procedural criteria” and “Values and justice” (16 %, respectively). About 39 % of empirical studies use experimental scenarios and mainly student samples. Cross-sectional studies make up about 32 % and use workers’ samples.

Table 1 Topics, methodological design, and samples of the studies collected
Full size table

The USA was the country most often included, present in 73 % of the empirical studies (N = 60). Together, China and Hong Kong were included in 60 %. At 72 % o, there is a clear preponderance of Asian countries. Most studies compare Asian countries (52 %), particularly China (25 %), with the USA. European countries are present in 43 %, South American countries and Canada are included in 23 %, and countries of Oceania are included in 20 % of studies. The least investigated regions are the Middle East (12 %) and sub-Saharan Africa (3 %). A large portion (47 %) includes only two countries, and only about 20 % include more than eight countries.

Conceptualizations and Measures of Organizational Justice

Most organizational justice research has examined the justice of events, pertaining either to specific situations or to how situations are generally handled. The fairness of outcome allocations was the first to draw researchers’ interest. Equity theory states that an allocation is perceived as fair, if a balance between perceivers’ contributions and rewards as compared to those of a referent other is attained (Adams, 1965). Other principles such as equality and need were also investigated (Deutsh, 1975). Yet, equity remains the dominant theory, serving as the criteria used for most indirect, criteria-based, distributive measures.

The preference for adversarialFootnote 3 over inquisitorial procedures for conflict resolution drew attention to the implications of procedural characteristics, such as process control and voice, for justice perceptions (Thibaut & Walker, 1975). The finding that, even faced with adverse outcomes, receivers’ attitudes remain positive if the procedures used to decide are considered fair (Lind & Tyler, 1988), inspired research about procedural fairness criteria. Accordingly, a fair procedure is one that is: consistent, representative, non-biased, accurate, correctable, and moral (Leventhal, Karuza, & Fry, 1980); those criteria are used in most indirect measures of procedural justice.

Interactional justice concerns the way procedures are enacted by authority figures, such as the supervisor (e.g. Greenberg, Bies, & Eskew, 1991). Some (e.g. Colquitt, Conlon, Wesson, Porter, & Ng, 2001) have proposed that it comprises two elements: informational—degree of honesty and timing of the information given—and interpersonal—degree of respect and sympathy showed; those criteria are used in most indirect measures of interactional justice.

Of all studies collected, 66 % examine distributive, 61 % examine procedural, and around 24 % of the studies examined interactional justice. Of those, 24 % (N = 4) distinguish interpersonal and informational justice. Some studies combine, either distributive and procedural (19 %) or distributive, procedural, and interactional justice (19 %).

The event paradigm has been contrasted to holistic, overall entity justice. Claiming that people experience justice as a holistic perception, and not a multi-dimensioned evaluation, some have studied overall justice appraisals (e.g. Kim & Leung, 2007). Over time, employees form representations of the degree of fairness they may expect from different entities. This entity approach tends to use direct measures. Around 7 % of studies use entity justice.

About 77 % of the empirical studies investigate antecedents such as allocation (20 %) and procedural criteria (15 %), or situational (17 %) and organizational (12 %) characteristics. Some include both antecedent and outcomes. About 60 % investigate outcomes of justice judgments, namely satisfaction (17 %), commitment (15 %), turnover and trust (12 %), respectively. Across cultures, more positive justice perspectives correspond to more positive outcomes.

Table 2 shows that of 53 empirical studies collectedFootnote 4 some develop their own measures of justice or experimentally manipulate it (43 %). Only 53 % of empirical studies apply validated justice measures. The majority (66 %) uses indirect measures; that is, they assess the application of the criteria previously found to be related to different dimensions of event justice. Only 21 % directly assess justice evaluations, and they ask participants how fair they consider a given event or entity to be. Others combine indirect and direct measures.

Table 2 Type of justice measures, scales, and perspectives used in the empirical studies
Full size table

The Colquitt and colleagues’ scale (2001; n = 5) is the most frequently cited, followed by those of Niehoff and Moorman (1993; n = 4) and Leung and colleagues (i.e. Leung, 1987; Leung, Smith, Wang, & Sun, 1996; n = 4). Most use some form of reliability analyses (74 %). Cronbach’s alpha is the most common, either alone (33 %) or combined with exploratory factor analysis (17 %). Few use more sophisticated techniques, such as confirmatory (15 %; e.g. Blader & Chen, 2012) and multi-group factor analyses (9 %; e.g. Scheer, Kumar, & Steenkamp, 2003). Only one study investigated cross-cultural measurement invariability; Colquitt et al.’s (2001) four-dimensional scale of event organizational justice works well across 13 countries; however, factor inter-correlations and reliabilities were found to systematically vary (Fischer et al., 2011).

In the next section, we review findings regarding the main sociocultural context levels that affect justice across cultures, namely national culture; socioeconomic development; organizational; situational; and perceiver’s characteristics. Figure 1 presents a road map of the factors identified in our review and examined within each contextual level. The findings reviewed in the next sections show that these factors dynamically interact to influence: (1) what justice criteria people from different countries use to evaluate a justice stimulus and (2) how they apply them to develop justice perceptions; and also, (3) what reactions those people have to (un)fairness and (4) how they express them.

Fig. 1
figure 1

Relationship between sociocultural context influence levels and organizational justice judgements

Full size image

Sociocultural Influence Levels

National Culture

Tables 3, 4, and 5 present the models and main cultural dimensions investigated. Only 19 % of empirical studies measured cultural variables, and only 14 % include individual-level measures in the analysis. The remainder use individual scores to confirm expectations about the countries’ culture. The majority (53 %) use nationality as a proxy for culture and base predictions on archive data. Most (62 %) use culture as an explanatory variable, and the remainder use it as a moderator between justice perceptions and antecedents (22 %) or outcomes (17 %).

Table 3 Studies that include only individualism–collectivism-related constructs (N = 29)
Full size table
Table 4 Studies that include only individualism–collectivism and power distance-related constructs (N = 22)
Full size table
Table 5 Studies that include only power distance-related constructs (N = 6), only other cultural dimensions (N = 9), or only socioeconomic factors (N = 7)
Full size table

As illustrated in Tables 4 and 5, Hofstede’s model (1980, 2001) is the most widely used (58 %). Others such as House, Hanges, Javidan, Dorfman, and Grupt (2004, 7 %), Triandis (1995, 7 %), Schwartz (1992, 7 %), and Inglehart (1971), Inglehart and Baker (2000, 4 %) models are also used. Collectivism and power distance are the most investigated. Collectivism or similar constructs are present in 69 %, and power distance or similar constructs in 36 % of studies.

Distributive Justice

Collectivism has been found to affect distributive criteria. Several authors (e.g. Chiang & Birtch, 2005; Giacobbe-Miller, Miller, & Victorov, 1998; Hui, Triandis, & Yee, 1991; Hysom & Fisek, 2011) have studied the fairness of different allocation procedures based on the Deutsch (1975) framework of allocation rules and groups’ goals. Equity—when the goal is performance, rewards are distributed according to each member’s relative contribution; equality—when the goal is interpersonal harmony, each group member receives the same; finally, need—when the goal is members’ welfare and development, each member receives enough to satisfy their needs.

Marín (1985) found that: when Indonesian and US students assumed the role of receivers—irrespective of sex, nationality, and level of friendship with the allocator—they preferred an equitable allocator to an egalitarian one and considered them fairer. Though equity is the preferred rule across cultures (e.g. Kim, Park, & Suzuki, 1990), collectivists tend to follow equality and need more than individualists do (Chen, Meindl, & Hui, 1998; Murphy-Berman, Berman, Singh, Pachauri, & Kuma, 1984). When rewards and self-contributions were high, or the partner was a friend, collectivistic Chinese were more egalitarian, than people in the USA (Hui et al., 1991). Yet, when allocating money cutbacks, across-culture people distribute more to needy than to meritorious recipients (Murphy-Berman et al., 1984).

Power distance has also been related to allocation preferences. Fischer and Smith (2003) conducted a meta-analysis comparing equity and equality across 14 countries. In cultures with greater hierarchical differences, people demonstrate a greater preference for equity and more positive perceptions of allocators using equity. Collectivism was not a significant predictor; the more hierarchical nature of collectivistic cultures may have buffered its effect.

Different cultural dimensions may interact. Students in the USA (i.e. individualistic and masculine) and Japan (i.e. collectivistic and masculine) have a stronger preference for equity than students from South Korea (i.e. collectivistic and feminine). The importance of achievement in masculine cultures could contribute to emphasis on equity, so Japan’s masculine culture may have compensated for its collectivism (Kim, Park, & Suzuki, 1990).

In addition to the principles of justice used, the way they are applied also varies. Even when using equity, collectivistic vertical societies tend to consider aspects such as tenure, quanxi, or social skills (Fischer & Smith, 2003; Mueller, Iverson, & Jo, 1999) as adequate contributions, whereas individualistic horizontal societies may consider only performance and efficiency. When calculating needs, collectivist vertical societies may be more likely to consider assistance to the extended family or employee relations as lifelong obligations, whereas individualistic societies may focus on individual short-term obligations (Leung & Kwong, 2003).

Procedural Justice

Collectivism has been found to affect procedural criteria preferences. Early studies (e.g. Lind, Erickson, Friedland, & Dickenberger, 1978) report a preference across cultures for adversarial disputant-controlled models, over more inquisitorial adjudicator-controlled models. These findings were attributed to a universal preference for adversarial procedures, which allow individuals to retain a higher process control (Thibaut & Walker, 1975).

The relational model (e.g. Lind & Tyler, 1988), on the other hand, proposes that relational concerns outweigh instrumental concerns in determining procedural perceptions. Procedural justice carries information about individuals’ standing within the group, so a positive evaluation of membership is associated with a greater level of group identification, which leads to a compliance with authorities and more positive attitudes towards the group.

In two studies, Lind, Tyler, and Huo (1997) examined the effect of relational criteria such as status recognition, trust in benevolence, and neutrality on procedural justice judgments. In the first study, students from the USA, Germany, and Hong Kong recalled a conflict and reported their reactions. In the second study, the US and Japanese students rated procedures for resolving a hypothetical dispute. In both studies, relational criteria mediated the effects of voice on procedural judgments. These findings showed that relational concerns are important across cultures and mediate the effect of instrumental criteria on justice judgements.

Comparing control (i.e. voice, consistency, outcome fairness and favourability) and relational (i.e. status recognition, benevolence, and neutrality) procedural criteria, Blader, Chang, and Tyler (2001) showed that in Taiwan, high on power distance, workers demonstrated a balanced influence of relational and instrumental criteria, while in the USA, low on power distance, workers defined procedural justice primarily through relational criteria. Procedural justice was less predictive of retaliation in Taiwan, and the relationship was fully, as opposed to partially, mediated by organizational identity in the USA. In high power distance cultures, workers accept strong hierarchical structures and status differences between supervisors and subordinates. Relational factors may be de-emphasized making treatment considered disrespectful in low power distance cultures more acceptable (Blader et al., 2001).

Most research has compared Western and East Asian societies, so findings may be due to East–West differences other than collectivism. To address this limitation, Leung, Au, Fernández-Dols, and Iwawaki (1992) compared two collectivist societies, Japan and Spain. In keeping with the collectivism framework, both preferred harmony-enhancing procedures. Expectations regarding process control and animosity reduction predict preferences.

The expectancy-valence model (e.g. Leung, Bond, Carment, Krishnan, & Liebrand, 1990) examines differences between what people want to accomplish and why. Different cultures may want different things—animosity reduction versus process control—or they may want the same things, but have different expectations regarding the best way to achieve them—harmony-enhancing procedures versus confrontation procedures. The adversarial system is favoured by both collectivists and individualists (Leung, 1987; Leung & Lind, 1986), but collectivists may favour bargaining and mediation more because of perceived potential for animosity reduction (Leung, 1987; Leung & Lind, 1986). Collectivists maintain fewer, enduring group memberships, so voice may not be as important for signalling identification as it is for individualists, and instead, it may be desired as a sign of relative status. Indeed, cultures may not only desire different outcomes, but their expectations may vary on how different procedures will achieve them (Leung et al., 1992).

Interactional Justice and Reaction to (In)Justice Perceptions

Interactional justice perceptions are affected by collectivistic cultural values. Tata, Fu, and Wu (2003) studied social sensitivity in the context of performance evaluations and showed that social sensitivity had a greater effect on overall justice in collectivistic China than in the USA. Also, compensation strategies for handling incidents of service failure are viewed more positively in individualistic cultures, while explanation, apology, and voice are viewed more positively in collectivistic cultures (Hui & Au, 2001; Mattila & Patterson, 2004). These findings may be attributed to the importance of face saving, harmony, social status, and conflict avoidance—all of which increase the relevance of interpersonal treatment for collectivists.

Interactional justice perceptions are also affected by power distance. Compared to low power distance Americans, Chinese employees seem to react less negatively to supervisory criticism (Leung, Su, & Morris, 2001a). In scenarios of less interpersonal fairness, Americans expressed more negative justice perceptions and less trust and satisfaction than the Chinese because of the lower tolerance for iniquity in low power distance cultures. Yet in both, criticism delivered with interpersonal fairness reduced negative attributions, increased feedback acceptance, and improved attitudes towards supervisor and organization (Leung et al., 2001a).

Culture affects the importance of the event justice dimensions. In collectivistic Turkey, support seeking was higher and more strongly related to justice than in France (Finkelstein, Minibas-Poussard, & Bastounis, 2009). Compared to the USA and Germany, distributive justice had a stronger effect on trust, satisfaction, and commitment in collectivistic countries (i.e. India and Hong Kong) than procedural justice did (Pillai, Williams, & Tan, 2001).

Across cultures, specific event justice dimensions appear to be related to reactions directed at specific entities. More and Tzafir (2009) investigated the mediational role of trust, in the relationship of event justice dimensions, turnover, and organizational citizenship behaviour in Israel, the UK, and Hungary. Supporting the social exchange theory, both organizational and supervisor trust were negatively associated with turnover intentions. Distributive, procedural, and informational, but not interpersonal justice, were found to be positively related to organizational trust, while both informational justice and interpersonal justice were related to supervisor trust.

Socioeconomic Development

Socioeconomic factors are mentioned in 20 % of the studies. Religious and philosophical traditions play a big role in people’s lives and affect the ways they conceptualize and apply justice (Neusner, Sonn, & Brockopp, 2000). Cole (2009) proposed a conceptual model that relates them to the importance and focus of distinct event justice dimensions. Accordingly, Christianity values equity and self-focused justice, while Islam values equity and need and group-focused procedural justice; both value informational more than interpersonal aspects. Confucians and Buddhists value equality and need, as well as interpersonal, more than informational or procedural justice, because of the emphasis on group harmony and paternalistic leadership. Oral Chinese traditions about ways to deal with injustice focus on the re-evaluation and devaluation of incidences and discourage confrontation, impact justice beliefs even among Westernized Hong Kong students (Chiu, 1991). Political traditions have been associated with differences in justice perceptions. Leung and Kwong (2003) emphasize the importance of communism to understand the perspective of Chinese partners in joint ventures. Similar arguments have been made for ex-soviet managers (e.g. Giacobbe-Miller et al., 1998; Giacobbe-Miller, Miller, Zhang, & Victorov, 2003).

The wealth available to a countries’ population (i.e. domestic product per capita or income inequality) influences which principles of justice are used, and the way they are applied. Societies with higher inequity allocate rewards more differentially—on the basis of equity and performance inputs, whereas societies with lower inequity prefer equality over equity (Fischer & Smith, 2003). High unemployment is related to a more negative attitude towards need as a valid allocation rule (Fischer et al., 2007). Instrumental, distributive, and uncertainty avoidance aspects of justice seem to be more valued in conditions of higher unemployment and social inequity, when the achievement of material necessities is at risk (Shao et al., 2013).

Organizational Characteristics

Institutional Context

Chen (1995) proposed that nations and organizations may have different goals at different moments in time, and those goals may override traditional cultural norms. Accordingly, Chinese employees from companies undergoing reforms aimed at increasing productivity emphasized economic goals, whereas US employees, from companies undergoing reforms aimed at improving team collaboration, emphasized humanistic goals (Chen, 1995). Also, in the context of transition to a free market, Russian managers emphasized equity and individual performance, over equality and co-worker relations (Giacobbe-Miller et al., 1998).

Fischer et al. (2007) demonstrated that reliance on equity is higher in the private sector and in cultures high on mastery values that encourage achievement and domination. Reliance on equality is higher in organizations that are performing better. Reliance on need is predicted by low unemployment rates and high embeddedness values, which encourage striving towards shared goals and maintaining the status quo. Public organizations in Germany are more hierarchical and bureaucratic and therefore rely more on seniority, yet private organizations rely less on seniority than in the UK (Fischer, 2004). Organization type (i.e. joint ventures, foreign-owned enterprises, or state-owned enterprises) is also important. Russian and US managers, in joint ventures but not state-owned enterprises, show higher levels of distributive rule convergence than Chinese managers (Giacobbe-Miller et al., 2003). Resistance to change may be higher in relation to core, as opposed to peripheral, values, and collectivists are believed to be more resistant to change than individualists (Giacobbe-Miller et al., 2003).

Organizational Culture and Climate

The relationship between leadership style, justice, leader–member exchange (LMX), and job performance is moderated by organizational culture (Tziner, Kaufmann, Vasiliu, & Tordera, 2011). Low justice may not lead to low LMX if employees attribute the inequity to organizational culture, rather than to deliberately unfair supervisor behaviour. If organizational culture is perceived to perpetuate injustice, employees may be more likely to excuse inequity on the part of their superior (Tziner et al., 2011) and direct their reaction towards the organization. Occupational culture may also impact justice perceptions. The relationships between justice, commitment, and turnover were examined across two employee samples (i.e. faculty and business managers) each from the USA and Bangladesh; more differences were found among employee samples than among countries (Rahim, Magner, Antonioni, & Rahman, 2001).

While organizational culture concerns the firm has a whole, climate concerns specific aspects of organizational life. Fehr and Gelfand (2012) propose that a forgiveness climate—tendency to manage conflicts ways that increments harmony and prevents future conflicts—is more likely to emerge from self-transcendent and restorative justice values. Others (e.g. Cole, 2009; Leung & Kwong, 2003) call attention to the importance of restorative justice and advise that the Western-style retributive aspect may not be seen as fair in other societies.

Leadership

Specific aspects of leadership may be differently interpreted in different cultures. In a meta-analysis of LMX in 23 countries, Rockstuhl, Dulebohn, Ang, and Shore (2012) found that relationships between LMX, transformational leadership, justice, satisfaction, turnover, and trust are stronger in horizontal-individualistic contexts. Transformational leadership is related to satisfaction only in Western countries (Pillai, Scandura, & Williams, 1999). When power distance is high, the leader acting as a coach may be viewed negatively. Subordinates may feel stressed by attempts to involve them in problem-solving. Transformational leadership may be more effective in countries where it complements existing values (Pillai et al., 1999).

The influence of supervisors’ communication tactics on trust was stronger for Japanese workers (Yamaguchi, 2009). Belonging to a high-context and uncertainty avoidance culture, Japanese workers tend to emphasize good human relations and working conditions that promote feelings of security and trust. On the other hand, the influence of procedural justice on trust was stronger for US workers. Because of low-context communication styles and uncertainty avoidance, they may link communication to practical aspects, such as the procedures of decision-making, rather than emotional responses such as trust.

Human Resources Management Practices

There are two perspectives regarding the effect of human resource practices across cultures. The situational approach (e.g. Mahajan, 2011) suggests practices are context dependent and should be designed and implemented differently according to the specific culture. The convergent approach (e.g. Chiang & Birtch, 2005; Ryan et al., 2009) suggests practices are effective across cultures.

In line with the convergent approach, Ryan et al. (2009) showed that the fairness perception of selection processes was similar across countries and across individuals holding different cultural values. Steiner and Gilliland (1996) demonstrated that interviews, work samples, and résumés were the most favoured techniques in the USA and France. Across countries, face validity, widespread use, employer’s right to obtain information, and opportunity to perform were the strongest correlates of favourable justice perceptions and reactions.

In line with the divergent approach, Lunnan and Traavik (2009) show in a scenario study that managers from Lithuania, China, and Norway differ in fairness assessment of appraisal tools. Those high on power distance had more positive perceptions. Standardized tools were judged more fair in countries low on self-expression, undergoing economic change and (i.e. China and Lithuania), than in a stable country high on self-expression (i.e. Norway).

When deciding how to deal with expatriates, most organizations chose to maintain a standard of living and working conditions comparable to those of the home country (Black, Gregersen, Mendenhall, & Stroh, 1998). A different set of practices may be applied to expatriates to motivate them, attend to their expectations, and avoid negative job-related outcomes (Toh & Denisi, 2003). But organizations should also be concerned with perceptions of the host country’s nationals. They constitute the majority of employees, are crucial to organizational performance, and have a big influence on the success of expatriates’ social and work integration (Toh & Denisi, 2003). If host country nationals see no additional value for themselves or the company, it may lead to feelings of relative deprivation (Runciman, 1966), particularly if expatriates creates a “glass ceiling effect” for the professional development of national employees (Leung, Lin, & Lu, 2014; Mahajan, 2011).

Leung et al. (1996) surveyed host country nationals in joint venture hotels in Shanghai. They found that comparison with expatriates did not add to the prediction of job satisfaction, but comparison with locals did. Later, Leung, Wang, and Smith (2001b) repeated the survey in the same setting and obtained strikingly different results. Comparison with expatriates not only impacted the job attitudes of host country nationals, but also caused them to regard their salary as unfair in the light of the huge disparity with expatriates. When the second survey was conducted, joint venture hotels had become common, and host country nationals had more experience working alongside expatriates (Leung & Kwong, 2003). As a result, host country nationals considered expatriates more similar to them and so used them as a reference group in social comparison. Justice perceptions are dynamic and can change rapidly in a fast-moving economy (Leung & Kwong, 2003). So, providing opportunities for professional advancement among host country nationals may be especially important in multi-cultural contexts.

Situational Characteristics

About 21 % of the studies examine situational aspects that impact justice perspectives. Aspects of the justice context such as in-group out-group belonging, task characteristics, and status have been analysed in relation to allocations and procedures. The Chinese tend to sacrifice self-gain to benefit in-group members when allocating rewards, yet when dealing with out-group members they tend to favour equity (Leung & Iwawaki, 1988). Analysing out-group versus in-group belonging and relative task contribution in a scenario study, Hui et al. (1991) found that: both were more egalitarian when self-contribution was high; the Chinese were more generous than Americans and used an allocation rule that favoured the partner, especially dealing with friends; and when there was no limit to the amount being distributed, the Chinese were more egalitarian than Americans.

In an examination of task interdependence, productive versus solidarity goals and cultural values, Chen et al. (1998) found that US and Hong Kong students respond similarly to situational demands. Equity is preferred for low interdependence and productivity goals, whereas parity is preferred for high interdependence and solidarity goals. Comparing reward types, Kim, Park, and Suzuki (1990) showed that the allocation of primary (i.e. grades) and social rewards (i.e. desire for future interaction outside and at work) were contingent upon masculine and collectivistic values. High collectivistic values predict the use of equality for allocating both primary and social rewards, while high masculine values predict the use of equity.

Chiang and Birtch (2005) proposed a type–system–criterion model which consisted of: (1) reward type—financial versus non-financial, extrinsic versus intrinsic; (2) reward systems—procedures by which the rewards are allocated, which can be performance based (i.e. yearly promotions, performance appraisals) or non-performance based (i.e. seniority adjustment to cost of living); and (3) reward criteria—equality, equality, and need. Financial rewards are more important in Honk Kong, Canada, and the UK, than in Finland, because of masculinity values that emphasize achievement of material accomplishments. The expectation that collectivistic cultures, as Honk Kong, tend to value material rewards less was not supported (Chiang & Birtch, 2005).

Hysom and Fisek (2011) proposed the equity–equality equilibrium model. In a scenario study, with students from the USA and Turkey, the authors found that in both groups, as status differentiation increased, equity was less used and allocators preferred equality as a way to maintain social harmony among group members. They suggest that people address the strain between equity, as a motivator of task performance or a potential motivator of social tension, by balancing equity and equality norms. Situational factors may stress the importance of competence and productivity, or of the maintenance of positive social relations among members.

Fewer studies have analysed the impact of situational aspects in procedural differences. Leung et al. (1992) tested whether group versus individual conflict would lead to different procedural preferences in distinct cultural environments. The author reported that in an earlier study (Leung et al., 1990) comparing the USA and China, a marginally significant effect had been found. Across cultures, when groups rather than individuals were involved in conflicts, participants preferred to use conflict reduction procedures more, and confrontational procedures less. Yet, the authors could not replicate this finding in Spain and Japan (Leung et al., 1992).

The fair process effect—a compensatory relationship between outcome favourability and process fairness—is contingent upon the relative status of the exchange parties. In both China and the USA, the positive relationship between outcome favourability and higher-status parties’ positive reactions was stronger when procedural fairness was high, particularly among those high in self-esteem, low in need to belong, or high in power distance orientation (Blader & Chen, 2012). So, for high-status individuals, both outcome favourability and procedural fairness appear to be necessary to maintain a positive group identity.

Perceivers’ Characteristics

Sociodemographic Differences

Gender interacts with country to predict justice preferences. Differences between men’s and women’s roles are larger in cultures low in gender egalitarianism, compared to cultures high in egalitarianism. Findings regarding allocation and interpersonal justice indicate that in cultures where gender socialization patterns and behavioural norms differ widely, differences in the development of and reactions to (in)justice judgments are also greater (Hysom & Fisek, 2011; Leung & Lind, 1986; Murphy-Berman et al., 1984; Tata, 2000). Age has also been investigated. As people progress through the stages of the life cycle, they become less satisfied with the fairness of service encounters (Nance & White, 2009). In high power distance societies and ascription cultures, younger or lower status employees are less likely to make social comparisons with older, higher-status workers (e.g. Toh & Denisi, 2003).

Individual Cultural Orientations Differences

Most studies reviewed base predictions on national-level values collected in previous studies, but about 25 % of 53 empirical studies do collect individual-level cultural data. Some use the data to confirm expectations regarding countries’ culture and use country as a proxy for culture. Others use individual values in the analysis, but do not compare their effect to that of country belonging. Few compare the relative effect of country and individuals’ values.

Openness to change versus conservatism and self-enhancement versus transcendence individual orientations influences the use of distributive criteria, such as performance or seniority. Employees in Germany endorsed self-transcendence and conservatism more than employees in the UK did, which translated into more positive attitudes towards seniority (Fischer & Smith, 2004). Individuals’ values were aligned with expectations regarding the countries’ cultures derived from previous studies and were not directly included in the analysis.

In a series of studies, Brockner, Chen, Mannix, Leung, and Skarlicki (2000) examined whether differences in interdependent versus independent self-construals moderated the interaction between procedural justice and outcome favourability. In the first study, country belonging was used as a proxy for self-construal. In the second, both country belonging and self-construal were included. In the third study, participants were classified on the basis of their self-construals. Consistent with expectations, US participants showed more independent self-construals than participants in China. The interaction between procedural fairness and outcome favourability was stronger among those with more interdependent self-construal. Both this and Fischer and Smith’s (2004) study did not directly compare the relative effect of country versus individual orientation.

Hui et al. (1991) found that the Chinese were more generous than Americans when dealing with friends, by using allocation rules that favoured the partner in a scenario study. Also, they were more egalitarian when there was no limit to the amount being distributed. These tendencies have traditionally been attributed to higher collectivism. However, when comparing the effect of country versus individual orientation, collectivistic values only explained the tendency of the Chinese towards equality, not their tendency towards generosity.

Analysing workers reaction in the USA and Hong Kong, Lam, Schaubroeck, and Aryee (2002) found that power distance moderated the relationships between justice, satisfaction, performance, and absenteeism. Justice effects were stronger among individuals scoring lower on power distance. Power distance was lower for participants from the USA. However, when comparing the relative effect of country and individual orientation, only individual power distance orientation, not country, moderated the effects. In a similar vein, Fischer and Smith (2006) showed that irrespective of country, employees from British and German who endorsed openness to change exhibited a stronger relationship between procedural justice and commitment or extra-role behaviours. Together, these results indicate that individual-level value orientations have an impact on justice that may extend beyond country belonging.

Expectations and Attribution Differences

Exploring the valence-expectancy model, Bond, Leung, and Schwartz (1992) examined the effect of expectations on the allocation and procedural preferences of Chinese and Israeli students. In both cultures, expectations tapping harmony and performance were important for resource allocation, and those tapping animosity reduction and process control were important for conflict resolution. Cultures may differ not only in their preference for different procedures but in the extent they expect specific procedures to lead to certain outcomes (Leung, 1987).

Culture conveys different expectations regarding the value of diverse job aspects. The more one’s expectation about job-related rewards are met, the more positive are justice perceptions. Accordingly, in the USA, where individualism is predominant, met expectations about autonomy are more important than in South Korea where expectations regarding advancement opportunities were more important (Mueller et al., 1999).

Judging one’s opponent to be stubborn and emotional increases the preference for formal procedures as a way to guard against antagonism (Morris, Leung & Iyengar, 2004). A scenario containing gossip about an opponent’s past actions in a dispute resulted in negative attributions, which led to a greater preference for formal procedures for both Hong Kong and US students. However, in the no-gossip condition, Hong Kong students were more inclined to use informal procedures. This may help explain contradictory findings regarding collectivists’ procedural preferences. Collectivists may feel an obligation to cooperate with another person once close contact is initiated, thus favouring informal strategies. When the other party appears disagreeable, they may try to keep their distance by favouring formal procedures (Morris et al., 2004). Responsibility attribution has also been shown to affect customer satisfaction and post-complaint behaviour (Au, Hui, & Leung, 2001).

Zourrig, Chebat, and Toffoli (2009) propose a conceptual model of customer unfairness coping. Individualistic preferences trigger the appraisals of idiocentrics, while collectivistic preferences trigger the appraisals of allocentrics. When attributing blame, idiocentrics emphasize individual control, while allocentrics emphasize the social role accountability of the offender. Because allocentrics are more concerned with group harmony, they tend to experience more inwardly focused emotions. Thus, allocentrics tend to favour avoidance, while idiocentrics favour confrontational strategies. Bias towards attribution of ambiguous behaviour to personal dispositions, instead of environmental pressures, is more salient in the Western cultures (e.g. Morris et al., 2004) that exaggerate autonomy. Collectivists, on the other hand, tend to take a more situationist approach to attribution (Au et al., 2001).

Motivation Differences

Different models have been suggested to explain why people care about justice, namely (1) instrumental—control over outcomes (e.g. McFarlin & Sweeney, 1993). (2) relational—status and respect within the group (e.g. Blader & Tyler, 2009; Tyler & Lind, 1992); (3) uncertainty avoidance—knowing what to expect (e.g. Lind & Van den Boss, 2002); and (4) ethical or deontic—the “correct way” to treat people (e.g. Rupp & Bell, 2010).

In line with the instrumental perspective, Li and Cropanzano (2009) reviewed studies that suggest the effects of justice on outcomes is greater in North America, than East Asia, where it is eclipsed by concerns of maintenance of social harmony. Income inequity has also been related to greater importance of instrumental rather than relational motives (Fischer et al., 2007). Some consider the importance of relational procedural criteria as evidence of the intercultural preponderance of relational concerns (e.g. Tyler & Lind, 1992; Lind et al., 1997). Others (e.g. Blader et al., 2001) have shown that in high power distance cultures, even collectivistic ones, relational concerns are balanced with instrumental concerns. Kim and Leung (2007) found that for Americans and Japanese, distributive justice was less strongly related to overall fairness, and interactional justice was more strongly related to overall fairness than for Chinese and Koreans. This could be attributed to higher materialistic values among Chinese and Koreans. Procedural justice was related to overall justice across cultures, possibly because it is affected by both relational and instrumental concerns. The relationship between collectivistic emphasis on social interconnections and the greater importance of relational concerns may not be straightforward, because other factors may also play a role.

Two meta-analyses have compared justice motivations across cultures. Fischer (2013) focused on the relative effects of distributive, procedural, and interpersonal justice on outcomes in over 36 countries, based on the GLOBE project’s cultural scores (House et al., 2004). Relationships were stronger when there was greater income inequality and collectivism, suggesting that belongingness, esteem, and control motives are all important to justice judgements, and also that their relevance is affected by socioeconomic and cultural characteristics. Shao et al. (2013) explored the moderating influence of Hofstede’s dimensions on the relationship between justice and reactions to the supervisor, or the organization, in 32 countries. Findings are in line with instrumental (individualism), relational (femininity), uncertainty management (high uncertainty avoidance), and moral (low power distance) motives. Justice effects are strongest among nations high on individualism, femininity, uncertainty avoidance, and low power distance. Findings show the impact of values on justice motives, and that motives seem to operate simultaneously to predict perceptions.

Discussion

Theoretical Implication and Avenues for Future Research Development

We have reviewed findings that show that national culture, socioeconomic development, organizational, situational, and individual characteristics interact to predict what principles people use and how they apply them to develop justice perceptions; and also, what reactions people have to (un)fairness and how they express them.

The main implication of the findings reviewed is that multiple levels of sociocultural influences impact justice perceptions and reactions in different societies. Despite interaction between sociocultural aspects, with a few exceptions (e.g. Fischer et al., 2007) studies have focused only on single contextual factors and their isolated influence on justice. Research tends to have a narrow view of the factors that explain differences between countries. Further integration is warranted to understand how the sociocultural influence levels work together to shape the way people conceptualize, appraise, and react to (in)justice in organizations.

The way culture is operationalized is important. Collectivism and power distance are the most frequently investigated factors. Yet data show that despite being positively correlated, they sometimes have opposite effects (Fischer & Smith, 2003). Many times, researchers do not measure cultural values but simply assume that countries vary with regard to dimensions based on previous studies. When findings are attributed post hoc to one dimension, it is impossible to differentiate between their effect and that of other cultural dimensions and sociocultural factors (e.g. Morris & Leung, 2000). When analysing only a few countries, additional caution should be taken to ensure the measurement of relevant characteristics that explains country differences. In these cases, the examination of micro-processes might be more valuable for justice research, than post hoc explanations based on assumed cultural syndromes.

Relying on country aggregated cultural data based on previous studies to compare countries has several limitations. First, culture may be an enduring feature of group life and yet, despite being relatively stable, it is not immutable. As a socially constructed representation, culture is adaptive, and subject to environmental, as well as internal pressures (Giacobbe-Miller et al., 2003). Scores collected long ago may not be accurate representations of the cultural dynamics present in societies. Second, studies often don’t use representative samples or check for agreement levels before aggregating responses to country level. Global cultural classifications may mask intra-cultural diversity, intra-individual variability between social relationships across different spheres, as well as commonalities among people of different cultural backgrounds.

Studies that use a large sample of countries face specific methodological challenges. Score aggregation implies country values are independent and do not account for cultural distance. This is commonly referred to as Galto’s problem (e.g. Cole, 2009). Geographical proximity leads to cultural similarity because of increased interactions. One way to account for this is to control for cultural clusters (e.g. House et al., 2004).

A greater integration of country and individual-level cultural processes is important. If culture is conceived as a shared meaning system, at the individual level, sufficient within-group agreement and between-group variability are needed; at the group level, when using a small country sample there is the need to test for structural equivalence, or when using a larger country sample, to test for isomorphism between individual and cultural levels structures (Fischer, 2009). Only if these measurement issues are addressed, can researchers use culture to explain differences.

Additionally, there is the issue of justice conceptualization and measure. Research has focused on event justice judgements, but alternative conceptualizations such as overall entity perceptions offer avenues for new theoretical advancements. There is limited use of previously validated justice measures, and a cross-cultural equivalence check is rare; this may pose threats to the cross-cultural validity of findings. Finally, regarding contexts investigated, most studies compare only two countries. The majority compare China with the USA. Also, there is a lack of representation of highly populated, and increasingly economically important, regions of the world, such as South America, the Middle East, and sub-Saharan Africa. A more complete view of cultural diversity might be obtained from research that extends beyond the most commonly investigated contexts.

Practical Implications

Predicting how organizational fairness works across cultures can help understand motivational processes (Tziner et al., 2011), design and implement policy options (Morris, Leung, Ames, & Lickel, 1999), manage conflict, avoid turnover, and increase satisfaction, collaboration, and productivity in inter-cultural work settings (Conner, 2003; Mahajan, 2011). It can even improve the success of strategies for handling service failures (e.g. Wang & Mattila, 2011). Across cultures, positive justice perceptions are associated with more positive outcomes for employees and the organization.

Treating employees respectfully and providing honest information increased acceptance of negative feedback and decreased negative attributions directed at the supervisor (Leung et al., 2001a). Also, justice perceptions are positively influenced by supervisor communication tactics that provide rationales and explanations and negatively influenced by tactics that resort to persuasion (Yamaguchi, 2009). Even if highly collectivist and power distant cultures react less negatively to injustice, more positive reactions are always associated with positive, rather than negative justice perceptions. So, supervisor justice and interactional aspects may be especially important in individualistic low power distance cultures, but are also valued in other cultures.

The expectations different cultures have as to what a leader should be vary widely (House et al., 2004). In high power distance countries individuals may have a more paternalistic view of leadership. In those countries, the application of participative and empowering techniques, subjacent to most indirect justice criteria, should be preceded by some groundwork like communication and training. Evidence shows that sometimes organizational goals can have a primacy effect over cultural values (Chen, 1995). Participative practices may be implemented in high distance cultures, provided that they respect the cultural sensibility of employees.

International experiences are increasingly important for employees and organizations. Organizations that wish to maximize the success of those experiences should strive to maintain positive justice perceptions among expatriates and HCNs. Demographical aspects such as gender, age and ethnicity, or organizational category proportion, and composition may affect the salience of social identities (Toh & Denisi, 2003). When structural differences, such as pay differentials, are in line with a salient social category the identity is reinforced and attains greater salience (Ashforth & Johnson, 2001). Developing a culturally aligned model and providing equal opportunities may contribute towards reducing social schism and promoting the benefits of international mobility (Mahajan, 2011).

Finally, inequality and unemployment are associated with different motivational orientations. Instrumental and uncertainty avoidance aspects of justice are valued in conditions of higher unemployment and social inequity, when material needs are at risk (Shao et al., 2013). Balanced resource distribution and low unemployment increase the importance of relational, interactional, and deontological aspects (Shao et al., 2013). Organizations should be aware of the potential impact of the socioeconomic development. These factors might provide useful insights with regard to planning organizational practices in different countries and among multi-cultural work teams.

Limitations

Because systematic reviews aim to provide a compressive illustration of a research field, the scope of the review is a critical quality criterion. To balance the number and quality of sources, and for methodological and conceptual reasons, we chose to include only papers, which include more than one country. We decided to focus on comparative studies for a number of reasons. First, for practical reasons, the collection procedure including single nation studies wielded a too large collection of materials, which would be unmanageable from a systematic review point of view. Also, it is unclear what could be the exclusion criteria for research conducted in single cultural settings; in extreme, all research outside cultural settings where most models were developed (i.e. USA) could potentially be included. Second for conceptual reasons, the use of single nation studies has long been criticized in cross-cultural research (e.g. Morris & Leung, 2000). By definition, cross-cultural research is the study of behaviours and processes in which national cultural characteristics play a major role as independent or moderating variables (Tsui, Nifadkar, & Ou, 2007). Single nation studies tend to use nation that has a proxy for culture which does not allow for a methodologically sound comparison of findings between countries (e.g. Fischer et al., 2007). Finally, findings from single nation studies were integrated in the literature reviews, so were unlikely to wield additional insights.

We have thoroughly applied the review protocol. We don’t exclude the possibility that there might be additional cross-cultural organizational justice studies that were not included because they fall outside this protocol, but we believe we collected a broad and representative sample of cross-cultural justice research to date.

Conclusion

Continued research on cross-cultural justice stands to offer much to uncover what lies behind justice perspectives and reactions. This review take-home message is that cross-cultural differences in organizational justice are influenced by multiple sociocultural context factors at different levels. Therefore, research should move beyond simplistic explanations to integrate multiple levels. Guided by a more comprehensive view of sociocultural country differences and empirically armed with alternative approaches to the use of country proxy and archive data on country-level cultural dimensions, future research on cross-cultural organizational justice looks very promising.