MAKING JUSTICE SENSE OF LOCAL-EXPATRIATE COMPENSATION
DISPARITY: MITIGATION BY LOCAL REFERENTS, IDEOLOGICAL
EXPLANATIONS, AND INTERPERSONAL SENSITIVITY IN
CHINA-FOREIGN JOINT VENTURES
CHAO C. CHEN Rutgers University New York University
JAEPIL CHOI
Hong Kong University of Science and Technology
SHU-CHENG CHI-National Taiwan University
We examined how local employees of international joint ventures (IJVs) perceived disparity between their compensation and foreign expatriates' compensation from equity theory and social justice perspectives. Chinese locals perceived less fairness when comparing their compensation with expatriates' than when comparing it with other locals'. However, fairness vis-a-vis expatriates increased if the locals were compensated higher than their peers in other IJVs or endorsed ideological explana-tions for expatriates' advantage. Furthermore, expatriates' interpersonal sensitivity toward locals reduced the effect of disparity on perceived fairness. Finally, perceived compensation fairness was related positively to compensation satisfaction hut nega-tively to intentions to quit.
Distributive justice gains significance in today's globalized business world. As multinational com-panies operate across nations and continents at vastly different levels of economic development, disparity in employee compensation is unavoid-able. This disparity is most salient in the remark-able gap between the compensation received by the local employees of international joint ventures (IJVs) in developing countries and foreign expatri-ates from developed countries.
To encourage employees from developed nations to work in nations where pay and living standards are significantly lower, United States-based multi-national companies have offered generous compen-sation packages to expatriates (Reynolds, 1997). In addition to hardship allowances, a 15 percent over-seas premium is common, plus other possible per-quisites such as cars, club memberships, servants, and vacation homes (DeLisle & Chin, 1994). The average compensation packet of U.S. expatriates, according to Reynolds (1997), is two to five times as
We thank Brian Becker, Michele Gelfand, Kenneth Law, Kwok Leung, Jim Meindl, and Phyllis Siegel for their constructive comments on drafts. Part of the data collection was supported by a grant from the National Science Council in Taiwan (NSC 89-2416-H-002-019) to the third author.
much as that received by their home country coun-terparts and a great deal more than that received by the local nationals in the developing countries. Such disparity in multinational companies is sim-ilar to the two-tier wage systems of some compa-nies in the United States, in which the pay scale is determined by tenure or date of hiring. Research has shown that low-tier workers are more likely to hold negative perceptions of pay equity than are high-tier workers, especially when workplace in-teractions between tiers are relatively frequent (Lee & Martin, 1991; Martin & Peterson, 1987; McFarlin & Frone, 1990).
In addition to the structural inequity between locals and expatriates within a company, structural inequity among employees of different companies results from the establishment of IJVs and the hir-ing of foreign expatriates. For instance, IJVs in de-veloping countries on the average offer better em-ployee compensation than do domestic companies in those countries. Furthermore, some IJVs offer better compensation than others as the foreign part-ners from different countries follow different re-ward and compensation systems.
These highly disparate reward systems within and between business operations in developing countries present interesting issues of distributive 807
justice. For instance, how do local employees make "justice sense" of the structural inequity? Is there a sense of distributive injustice when locals compare their compensation with that of expatriates? With multiple diverse and salient referents, how does one set of comparisons affect the evaluation of other comparisons? Are there factors that mitigate or exacerbate potential negative reactions by locals toward the compensation disparity between them and expatriates? And finally, does the locals' per-ception of distributive justice affect their attitudes toward their compensation and their organization? Chinese-foreign IJVs are an appropriate setting in which to address the justice issues discussed above. Various characteristics of the setting stiggest that several theories will offer relevant predictions. Expatriates in China earn 20 to 50 times what the local Chinese earn (Leung, Smith, Wang, & Sun, 1996) and, in some cases, the housing allowance of a foreign employee is more than the salary of a local employee of similar rank (DeLisle & Chin, 1994: 19). Chinese employees in IJVs have been found to use multiple social referents to evaluate the fair-ness of their compensation (Leung et al., 1996). Many IJVs in China justify the compensation ad-vantages of expatriates ideologically (Chen, Meindl, & Hunt, 1997; Child, 1994). The local Chi-nese employees are sensitive to interpersonal treat-ment in social interactions, which affects their jus-tice perceptions (Brockner, Chen, Mannix, Leung, & Skarlicki, 2000). Applying equity and social jus-tice theories, in the current study we examined factors that affect Chinese IJV employees' percep-tions of distributive justice regarding local-expatri-ate compensation disparity and how such percep-tions would affect compensation satisfaction and intentions to quit.
THEORETICAL BACKGROUND AND HYPOTHESES
Equity Theory and Social Comparisons
Equity theorists (e.g., Adams, 1965) posit that individuals evaluate organizational rewards such as pay and benefits by comparing the outcome/ input ratios of themselves and social referents. Equal ratios lead to perceived fairness, but unequal ones may lead to perceived unfairness, as the com-parer feels either under- or overcompensated. How-ever, people generally have an egocentric bias, so the threshold for perceiving overcompensation as unfair is higher and perceptions of fairness are gen-erally positively related to how favorable an out-come has been to the self. This effect is known as outcome favorability (Brockner & Wiesenfeld,
1996). Social comparison is essential for assessing distributive justice. Social comparison research has primarily focused on similar social referents (Kulik & Ambrose, 1992). Similarity typically refers to per-forming similar tasks at a similar level of the same organization (Coodman, 1974), but it can also refer to tenure, age, sex, race, and so forth (Major, 1994).
Foreign expatriates as social referents. Equity
theory both provides a foundation and raises cer-tain questions for studying distributive justice in international contexts. To start with, are expatriates so different in background from local nationals that they may not be meaningful social referents to the latter? To be sure, the expatriates are dissimilar in many ways. Nevertheless, although similar others are preferred for social comparisons, dissimilar others can be used, provided they are "minimally relevant" and information about their compensa-tion is available or accessible (Kulik & Ambrose, 1992: 219).
A global work environment may also stimulate social comparisons with otherwise dissimilar co-workers when employees from diverse social back-grounds interact with each other to perform joint tasks. In China, foreign expatriates are potential social referents for the local IJV Chinese for a num-ber of reasons. First, although the expatriates may be dissimilar in nationality, race, and ethnicity, they are similar in that they work for the same joint venture and are regarded as partners and col-leagues. Second, only a few local Chinese generally work with foreign expatriates, but those who do tend to have frequent interactions with their for-eign colleagues. Third, local-expatriate pay dispar-ity is widely known because it has had publicdispar-ity in the press. Previous reports have shown that locals are aware and often resentful of the compensation advantages of expatriates (e.g., DeLisle & Chin, 1994).
Social referents and compensation fairness perceptions. According to both social comparison
and outcome favorability arguments, a perception of distributive justice largely depends on the social referent chosen, as this choice determines the rel-ative favorability of outcomes. As referents change, the same outcome becomes more or less favorable, leading to a stronger or weaker sense of fairness. Accordingly, as long as there are outcome differ-ences among different social referents and between them and the comparer, different levels of per-ceived distributive justice will be associated with different social referents. In this research, we fo-cused on perceived distributive justice regarding compensation relative to specific social referents.
For a joint venture Chinese who works regularly with foreign expatriates, there are three other
groups of social referents in addition to the expa-triates: other local Chinese working in the same IJV (internal IJV locals), local Chinese from other sim-ilar IJVs (external IJV locals), and local Chinese from domestic companies in the same industry. On the average, Chinese employees of an international joint venture working with expatriates receive a great deal less compensation than the expatriates but significantly more than the average Chinese locals in and out of the IJV company. Although the magnitude of compensation disparity between a focal IJV Chinese employee and external IJV locals should be smaller than that with expatriates or domestic locals, there still exists some disparity owing to the different compensation systems of IJVs. In other words, the focal Chinese generally have a compensation disadvantage as compared with the expatriates but a compensation advantage over other locals. On the basis of the social com-parison and outcome favorability arguments that using different social referents should lead to dif-ferent levels of distributive justice, we hypothesize:
Hypothesis 1. Local Chinese employees of in-ternational joint ventures will perceive less compensation fairness vis-a-vis foreign expa-triates than vis-a-vis other local Chinese.
Although the argument of outcome favorability suggests a straightforward prediction that the greater the compensation disadvantage as com-pared with an expatriate, the lower the perceived fairness, our interest here was to explore whether the negative impact of such a disadvantage on per-ceived fairness could be mitigated by other factors. Specifically, we explored three types of mitigators: compensation advantage over other locals, ideolog-ical explanations, and interpersonal sensitivity.
Mitigating the Effect of Compensation Disadvantage vis-a-vis Expatriates
Compensation advantage over other locals.
Lo-cal Chinese employees' remarkable compensation disadvantage vis-a-vis expatriates has the potential to generate great tension for the Chinese. According to equity theory, one way to reduce the tension of inequity from undercompensation without chang-ing outcomes or inputs is to shift comparison to social referents for whom outcomes are similar or less favorable. However, very little is known about whether and how information about one referent affects the perceived fairness of one's rewards rel-ative to those of a different referent. We suggest that this information about other referents moderates the relationship between compensation disparity with a previous referent and the perceived fairness
of compensation. Specifically, we believe the rela-tive advantage over other local Chinese will help focal Chinese employees feel less negative about their relative disadvantage as compared with for-eigners. The tactic of referent shifting is well known in Chinese folk psychology. A favorite Chi-nese saying for consoling those who feel deprived goes like this: "Even though you may feel deprived in comparison with those who are above you, you'll feel abundant if you compare with those who are below you" (bi shang bu zu, bi xia you yu). Thus, we hypothesize:
Hypothesis 2. A compensation advantage over other local Chinese will reduce the negative effect of compensation disadvantage vis-a-vis expatriates on the compensation fairness per-ceptions of local Chinese IJV employees.
Ideological explanations. Although unfavorable
outcomes generally produce lower perceived fair-ness, to the extent that people find explanations reasonable and convincing, hence justifiable, the negative perceptions are mitigated (Brockner & Wiesenfeld, 1996). Greenberg (1993a) called atten-tion to the role of social justice in affecting em-ployee reactions to unfavorable decisions. Social justice consists of interpersonal justice, which re-fers to the interpersonal caring and sensitivity shown to recipients of negative outcomes or deci-sions, and informational justice, which refers to providing adequate explanations for a given out-come or decision. Explanations go beyond merely providing objective information. Bies's (1987) anal-ysis of social accounts (that is, verbal strategies) showed that, in an attempt to justify negative or controversial decisions, managers use ideological social accounts, specifically, explaining outcomes within a framework of "moral reasoning."
As the open-door policy and market-oriented economic reforms require major ideological justifi-cations (Chen, 1995; Chen et al., 1997; Child, 1994), there have been conscientious efforts on the part of the reformists to legitimize the establishment of IJVs in China. For example, joint venturing and hiring foreign expatriates have been defended on the grounds that they are necessary for global mar-ket penetration, for technology transfer, and for learning advanced management techniques from the industrialized nations (Child, 1994). In addi-tion, expatriates' compensation advantage can be defended in part by a common observation that foreign expatriates come from regions and coun-tries in which the general living standards are higher and that they have to deal with many hard-ships, such as being far away from families and relatives.
The above explanations appeal to a coUectivist and global way of thinking. That is, even if the contribution of particular individual expatriates may not match what they are compensated for, their collective participation in the Chinese econ-omy is beneficial to China in a global business environment. Hence, to the extent that the local Chinese endorse the above explanations, the com-pensation disparity becomes more justifiable. Thus, we expected ideological explanations to moderate the relationship between compensation disadvantage as compared with expatriates and compensation fairness perceptions. Specifically, we hypothesized:
Hypothesis 3. Ideological explanations will re-duce the negative effect of compensation dis-advantage vis-a-vis expatriates on the compen-sation fairness perceptions of local Chinese IfV employees.
Interpersonal sensitivity. Interpersonal
sensitiv-ity refers to the care and sensitivsensitiv-ity that managers show to employees in enacting organizational de-cisions (Bies & Moag, 1986). Research has demon-strated that when a decision maker is interperson-ally sensitive, perceptions of decision fairness and acceptance increase (Creenberg, 1993a, 1993b, 1994). In this study, we were interested in whether the interpersonal sensitivity shown by expatriates affected local Chinese employees' perceptions of the fairness of their compensation in comparison with that of the expatriates. Our focus on interper-sonal sensitivity differs from the typical focus in interactional justice research in two ways. First, the current study addresses the interpersonal sensitiv-ity of expatriates who typically were not the makers of decisions about compensation allocations but instead, the beneficiaries of such decisions. Sec-ond, whereas justice researchers have usually ex-amined interpersonal sensitivity within the context of particular decisions, we extended our examina-tion to expatriates' daily interacexamina-tions with local employees.
Research on social exchange relationships throws light on the mitigating effect of interper-sonal sensitivity on the relationship between com-pensation disparity with expatriates and fairness perceptions. Interpersonal relationships in organi-zations are of two primary types: (1) social, and largely informal, relationships involving exchang-ing socioemotional resources and (2) economic, and largely formal, relationships restricted to the exchange of economic resources (Blau, 1964). So-cial exchange relationships are in general of higher quality in terms of trust than are economic ex-change relationships. As a socioemotional
re-source, the interpersonal sensitivity shown by ex-patriates should facilitate the formulation of or indicate the existence of a social exchange relation-ship between themselves and the locals. One im-portant implication of different exchange relation-ships is that they affect the salience and the interpretation of incoming information (Cropan-zano, Rupp, Mohler, & Schminke, forthcoming; Schminke, Cropanzano, & Rupp, in press). Brock-ner and colleagues (2000) argued that the parties to relational or social exchanges tended to pay more attention to social and psychological aspects of the relationships, whereas the parties to transactional or economic exchanges tended to pay more atten-tion to the economic aspects. These authors further demonstrated that members of coUectivist cultures such as the Chinese were more likely to be affected by relational aspects than were members of indi-vidualist cultures.
In line with the above argumentation and find-ings, the interpersonal sensitivity of the expatriates should have a cumulative effect on the local Chi-nese. Those to whom expatriates have displayed higher interpersonal sensitivity should develop a more positive affect toward the expatriates. Ac-cordingly, they will be less sensitive to the local-expatriate compensation disparity than those to whom expatriates have been less sensitive. In sum-mary, niceness on the part of expatriates mutes Chinese concern about the compensation disparity, but meanness on their part exacerbates such con-cern. Hence:
Hypothesis 4. Interpersonal sensitivity dis-played by expatriates will reduce the negative effect of compensation disadvantage vis-a-vis expatriates on the compensation fairness per-ceptions of local Chinese IfV employees.
Effects of Perceived Compensation Eairness on Attitudes
To the extent that perceptions about how fair their compensation is as compared with that of expatriates are a real psychological experience of a joint venture's Chinese employees, these compen-sation fairness perceptions should affect employee attitudes such as compensation satisfaction and in-tentions to leave (turnover inin-tentions). Previous or-ganizational justice research has consistently shown that distributive justice perceptions signifi-cantly affect pay satisfaction (Folger & Konovsky, 1989; Sweeney & McFarlin, 1993) and turnover in-tentions (Alexander & Ruderman, 1987). Extending these results to our study, we hypothesized:
Hypothesis 5. Compensation fairness as com-pared with expatriates will be positively re-lated to satisfaction with compensation. Hypothesis 6. Compensation fairness as com-pared with expatriates will be negatively re-lated to turnover intentions.
METHODS Sample and Procedures
One hundred and sixty-one local Chinese em-ployees of IJVs who had worked with foreign expa-triates filled out our questionnaires. Sixty-two per-cent of them were men, and most (86%) were in their early 20s and 30s and college-educated (96%). Average tenure with the current employer was 4.5 years. Three percent were top managers; 18 per-cent, middle managers; 48 perper-cent, professionals; and the rest were staff members. The respondents were from nine Sino-foreign joint venture compa-nies in Jiangsu province (37%), Hunan province (16%), and Beijing (47%). The foreign parent com-panies were from Japan (29%), Europe (58%), and North America (13%), and most were in manufac-turing or trading businesses.
Respondents were ensured that individual re-sponses were anonymous and would not be avail-able to their employing companies. All completed questionnaires were returned to us in sealed enve-lopes either directly or indirectly, through the hu-man resource (HR) office of a participating IJV or the office of a municipal coordinator of joint ven-ture companies. The response rate ranged from 60 to 80 percent.
Measures
Independent variables. Compensation included
salary, bonus, allowances, and various benefits, and all compensation comparisons were made be-tween individuals with similar levels of education and similar job responsibilities. Compensation
dis-advantage vis-a-vis expatriates was measiu-ed by
two items assessing by what multiple the expatri-ates compensation exceeded a respondent's (5-8, 9-12, 13-16, 17-20, 21-24, 25-28, 29 and above).
Compensation advantage vis-a-vis other Chinese
was also measured by two items, which asked the respondents to rate their compensation relative to each of the three groups of Chinese referents, with 1 being "very low" and 7 being "very high." We were unable to use a unified measure for compen-sation advantage and disadvantage because of the huge discrepancy in the magnitude of disparity. Compensation advantage over other locals could
range from zero to a small percentage of a focal respondent's pay, but disadvantage vis-a-vis expa-triates ranged from 5 times to 30 or more times more than a respondent's pay.
Drawing on interviews conducted by the first author with ten Chinese employees and two HR managers working in IJVs in China, we identified
ideological explanations for paying foreign
expatri-ates high compensation and put them into five items with which respondents expressed their degree of agreement (1, "very much disagree"; 6, "very much agree"). Interpersonal sensitivity was assessed with five items developed by Niehoff and Moorman (1993) and using same response scale noted above. To increase the response ease and the accuracy of assessing interpersonal sensitivity, we asked respondents to think of one expatriate with whom they had the most frequent interaction or one with whom they had worked the longest time. An exploratory factor analysis showed that all items loaded on the expected factors, except those measuring compensation advantage over other lo-cals. Table 1 shows the results of this factor analy-sis. Advantage over the local Chinese employees of domestic companies was distinct from that over local Chinese employees of a respondent's own or another (an internal or external) IJV. In view of this empirical distinction and the conceptual internal-external distinction (Oldham, Kulik, Stepina, & Ambrose, 1986), we separated compensation ad-vantage into three types in subsequent analyses. All of the factors showed reasonably high inter-nal reliability coefficients (see Table 2, below, for reliabilities).
Dependent variables. Perceived compensation fairness was measured by one item asking
respon-dents about the fairness of their compensation (1, "very unfair"; 6, "very fair") in reference to each of the four types of social referents. Compensation
satisfaction (Sweeney & McFarlin, 1993) and turn-over intentions (Wayne, Shore, & Liden, 1997) were
both also measured on a six-point scale (1, "very much disagree"; 6, "very much agree"). The reli-ability coefficients were .92 and .83, respectively.
Control variables. We controlled for the
individ-ual demographic variables of sex, age, education, tenure, and job category, and the company vari-ables of size, foreign parent, geographic region, and company-level compensation disparity with expa-triates. Table 3, below, gives the coding schemes for these variables.
RESULTS
Table 2 presents descriptive statistics for the ma-jor variables in this study. The average
compensa-TABLE 1
Results of Factor Analysis for Independent Variables"
Factors and Items
1. Interpersonal sensitivity Kind and considerate Respects locals
Sensitive to locals' needs Treats locals trustfully Cares for employment rights 2. Compensation advantage over locals in
Other IJVs, with similar responsibilities Other IJVs, with similar educational levels Same IJV, with similar responsibilities Same IJV, with similar educational levels 3. Ideological explanations
Hiring highly paid expatriates is beneficial for Absorbing advanced managerial experiences Absorbing advanced technology
Developing markets at home and abroad Expatriates' high pay compensates for
Different living standards
Living away from family and friends 4. Compensation disadvantage vis-&-vis expatriates
With similar responsibilities With similar educational levels 5. Compensation advantage over locals in
Domestic firms, with similar responsibilities Domestic firms, with similar educational levels Eigenvalues .88 .92 .92 .89 .77 .04 .15 .12 .14 .14 .19 .23 -.15 -.05 .13 .04 .19 .15 .08 .08 .09 .16 .13 .79 .83 .79 .70 .02 .04 -.04 .15 .13 -.08 -.09 .25 .23 .14 .04 .03 .04 .16 .18 .14 .00 .01 .74 .77 .72 .63 .61 .07 .07 .13 .18 .11 .03 -.01 .02 .07 -.20 -.23 .10 .07 .04 -.02 -.11 .07 .18 .95 .95 -.04 -.03 .11 .14 .11 .16 -.07 .14 .03 .16 .17 .09 .01 .02 .17 .05 -.03 -.03 .87 .90 5.30 2.73 2.27 1.63 1.18
' Significant loadings are shown in boldface type.
tion disadvantage as compared with expatriates was 3.67, which means that by the locals' estimate, the expatriates on the average were compensated about 15 times more than the locals in this study. The estimated average company-level gap was 3.74 (s.d. = 1.19), which is similar to the average com-pensation gap at the individual level.
The means and standard deviations of perceived compensation fairness vis-a-vis the four referent groups were as follows: 2.31 (1.03) for expatriates; 3.43 (0.92) for internal IJV locals; 3.03 (0.98) for external IJV locals; and 3.72 (0.97) for domestic firm locals. In a series of paired Mests, we found that perceived compensation fairness as compared with expatriates was significantly lower than that vis-a-vis each of the three Chinese referent groups; local Chinese employees of the same IJV a respon-dent worked for (internal IJV locals, t = -12.04,
p < .001); local Chinese employees of another IJV
(external IJV locals, t = -9.66, p < .001); and local Chinese employees of a domestic firm (domestic firm locals, t = -14.89, p < .001). This pattern of values supports Hypothesis 1, which predicts that perceived compensation fairness vis-a-vis
expatri-ates will be lower than that vis-a-vis other local Chinese.
Results of hierarchical regression analyses on perceived compensation fairness are in .Table 3. The control variables together did not significantly explain the variance in perceived compensation fairness. As can be seen in model 2, compensation advantages over domestic company locals and in-ternal IJV locals were not significant predictors of perceived compensation fairness as compared with the expatriates. Disadvantage relative to expatriates had a marginally negative effect, whereas advan-tage over external IJV locals had a highly positive effect. Both ideological explanations and interper-sonal sensitivity had significant, positive effects. In model 3, all of the interaction terms are presented. Hypothesis 2 was not supported, since none of the interactions with advantage over locals was signif-icant. Hypothesis 3 was not supported either, be-cause the moderating effect of ideological explana-tions did not reach a significant level.
Interpersonal sensitivity showed a significant, moderating effect on the relationship between com-pensation disadvantage and perceived
compensa-TABLE 2
Correlation Matrix of Major Variables"
Variable
1. Perceived compensation fairness vis-i expatriates
2. Compensation disadvantage vis-^-vis expatriates
3. Compensation advantage over locals employed by tbe same IJV
4. Compensation advantage over locals employed by another IJV
5. Compensation advantage over locals employed by a domestic firm 6. Ideological explanations 7. Interpersonal sensitivity 8. Compensation satisfaction 9. Turnover intentions Mean i-vis 2.31 3.67 3.60 2.86 4.11 3.79 3.64 2.96 3.21 s.d. 1.03 2.00 0.72 0.85 1.05 0.83 1.01 0.99 1.02 1 -.05 .23** .50*** .28*** .36*** .42*** .56*** -.38*** 2 (.95) -.02 -.17* -.04 .09 .15 -.04 .00 3 (.77) .49*** .37*** .15 .25** .38*** - . 2 1 * * 4 (.91) .32*** .24** .22** .55*** - . 2 1 * * 5 (.89) .28*** .35*** .44*** -.34*** 6 (.74) .21** .34*** -.28*** 7 8 9 (.94) .47*** (.92) -.38*** -.55*** (.83) ' n = 143. Reliability coefficients are in parentheses along the diagonal.
* p < .05 **p < .01 ***p < .001
tion fairness as compared with the expatriates. We plotted the interaction by taking values one stan-dard deviation ahove and below the mean of inter-personal sensitivity (Aiken & West, 1991). The form of the interaction showed that the negative effect of compensation disadvantage on perceived fairness was less pronounced for those locals who reported higher interpersonal sensitivity from the expatri-ates. (Interested readers can request the full inter-action results from the first author.) This result supports Hypothesis 4, which predicts that inter-personal sensitivity will have such a mitigating effect.
Results of the hierarchical regression analyses on compensation satisfaction and turnover intentions showed some significant effects for the control vari-ables. Older workers had higher compensation sat-isfaction (/3 = 0.18, p < .05) and lower turnover intentions (/3 = -0.18, p < .05) than did younger workers; employment at a Japanese firm had a neg-ative effect on compensation satisfaction (j3 = -0.33, p < .001); and respondents from Jiangsu and Hunan provinces had lower turnover intentions (j3 = -0.45, p < .01; P = -0.33, p < .01) than those from Beijing.
After we had controlled for individual and com-pany background variables and other types of perceived compensation fairness, perceived com-pensation fairness vis-a-vis expatriates had a signif-icant, positive effect on compensation satisfaction (^ = 0.24, p < .001) and a negative one on turnover intentions (j3 = -0.22, p < .01). These findings support Hypotheses 5 and 6. It is worth noting that compensation fairness in relation to external IJV
locals also had significant effects on both compen-sation satisfaction (j3 = 0.29, p < .001) and turnover intentions (/3 = -0.32, p < .01). Perceived compen-sation fairness in relation to internal IJV locals and locals employed by domestic Chinese companies had a significant effect on compensation satisfac-tion (/3 = 0.26, p < .001; )3 = 0.15, p < .05) but not on turnover intentions.
DISCUSSION AND CONCLUSION
We found that a local Chinese IJV employee's perceived compensation advantage over the local Chinese employees of other IJVs had a "main ef-fect" on perceived compensation fairness vis-a-vis expatriates, although we hypothesized a moderat-ing effect. The main effect indicates that, regardless of how large or small the disadvantage as compared with expatriates, greater advantage over external IJV locals led to higher fairness perceptions. In other words, the positive effect of perceiving a com-pensation advantage over external IJV locals seemed to have offset rather than merely moderated the negative effect of a compensation disadvantage in relation to expatriates.
We offer two possible explanations off this find-ing. First, compensation advantage over other Chi-nese locals employed by external IJVs confers higher status to a local Chinese employee, even though all locals are below the expatriates in status. To maintain their sense of elite status, those with greater compensation advantage over similar others were more likely to view use of differentials as a fair principle of compensation and thus to endorse
TABLE 3
Results of Regression Analyses for Perceived Compensation Fairness^
Predictors Model 1 Model 2 Model 3
Step 1: Control variables'' Age
Sex Education Tenure Professional job Middle management job Top management job Company size Japanese company Company disparity Jiangsu province Hunan province
Step 2: Compensation disadvantage vis k vis expatriates Compensation advantage, internal IJV locals
Compensation advantage, external IJV locals Compensation advantage, domestic firm locals Ideological explanations
Interpersonal sensitivity Step 4: Interactions
Compensation disadvantage X internal IJV locals Compensation disadvantage X external IJV locals Compensation disadvantage X domestic firm locals Compensation disadvantage X ideological explanations Compensation disadvantage X interpersonal sensitivity Change in F Change in adjusted R^ 0.05 0.06 0.00 0.01 -0.09 -0.11 -0.04 -0.12 -0.31** 0.22 -0.09 -0.18 1.30 .02 143 0.04 0.04 0.03 0.03 -0.04 -0.08 -0.01 0.00 0.05 0.32** -0.20 -0.15 -0.16^ -0.03 0.48*** 0.00 0.16* 0.34*** 14.28*** .37 143 0.09 0.03 0.06 0.03 -0.02 -0.07 -0.03 0.09 0.07 0.34* -0.23 -0.11 -1.69** -0.07 0.56* -0.05 0.02 -0.09 0.25 -0.17 0.15 0.50 1.00** 2.91** .05 143
" Regression coefficients are standardized.
** The coding scheme was as follows: age: 1 = 20s, 2 = 30s, 3 = 40s, and 4 = 50s and above; sex: 0 = woman, 1 = man; education: 1 = 9 years, 2 = 12 years, 3 = 14-15 years, 4 = 16 years, and 5 = 18 years or above; tenure: years with the current company; company size: 1 = fewer than 100 employees, 2 = 100 to 499, 3 = 500 to 1,000, and 4 = more than 1,000; job: the omitted category is staff members; Japanese company: Western parent companies are the omitted category; company disparity: the average score of all respondents from a company on the estimated average local-expatriate compensation gap within the company; and region: Beijing is the omitted region,
^ p < .10 * p < .05 **p < .01 ***p < .001
both local-expatriate disparity and disparity among locals. Second, we speculate that the main effect of compensation advantage over external IJV employ-ees was a result of the dominance of that group as the most influential referents for the people we studied. To explore this conjecture, we examined the relative impacts of all compensation compari-sons (that is, advantage/disadvantage vis-a-vis ev-ery referent group) by running a series of regression analyses with perceived compensation fairness with a given referent group as the dependent vari-able and all compensation comparisons as the in-dependent variables. We found that external IJV locals were the only group whose compensation information affected perceived compensation fair-ness vis-a-vis not only the designated group, but
also, all other referent groups. In other words, com-pensation advantage over the local employees of other IJVs had a positive effect on perceived fair-ness in relation to these external IJV locals and, in addition, the positive effect spilled over to posi-tively influence perceived fairness in relation to each and every of the other three referent groups. Such a generalizing spillover effect is similar to past research findings regarding pay satisfaction. For example, Berkowitz and his colleagues (Berkowitz, Fraser, Treasure, & Cochran, 1987) found that employees' satisfaction with other aspects of their jobs, such as the jobs' intrinsic natures, generalized to pay satisfac-tion. We would like to add, though, that where there are multiple sources, spillover is more likely to orig-inate from the most influential source, which in our
study of social referents happens to be the external IJV locals.
We also found that ideological explanations had a main effect on perceptions of the fairness of com-pensation as compared ,with that of expatriates, instead of the hypothesized moderating effect. This result suggests that ideological explanations for hir-ing highly compensated foreign expatriates might have significantly affected the perceptions of the local Chinese IJV employees about the contribu-tions (inputs) made by foreign expatriates. The ex-isting literature on justice has emphasized ideolog-ical explanations in light of their effects on procedural and interactional justice (e.g., Bies, 1987; Greenberg, 1990). The finding of this study shows that ideological explanations have a significant effect on distributive justice perceptions as well.
Limitations
A number of limitations of the present st;udy should be addressed in future research. First, com-pensation fairness as compared with expatriates was measured by one item, which may be suscep-tible to low reliability. Second, as we described in the methods section, compensation advantage was measured qualitatively, on a seven-point low-to-high Likert scale, whereas disadvantage was mea-sured quantitatively, with seven categories of mul-tiples. The qualitative scale allowed us to measure perceived advantage across different Chinese refer-ent groups without having to develop differrefer-ent quantitative units for each group. Its drawback is a loss of precision. Furthermore, the qualitative an-chors were inconsistent with the quantitative cate-gories of disadvantage vis-a-vis the expatriates, making comparisons between the two less straight-forward, especially if one is interested in the im-pact of compensation advantages and disadvan-tages in financial units.
A third limitation lies in the cross-sectional sign and in measuring both independent and de-pendent variables by self-reports, raising the possi-bility of common method bias producing the results. Although this possibility cannot be entirely ruled out, we believe that it is unlikely. Common methods would have triggered significant correla-tions among the major variables. Yet, the major independent variable, compensation disadvantage in relation to expatriates, had insignificant correla-tions with most of other variables. The significant interactive effect it has with interpersonal sensitiv-ity further increases our confidence in the validsensitiv-ity of the findings.
Theoretical and Research Implications
Our study is among the first to examine the role of social comparisons in an international context. Identifying relevant social referents in such a con-text is by no means straightforward because rele-vance seems to cut across objective criteria of similarity/dissimilarity in terms of organizational boundary and national citizenship. For example, judging from the impact of social referents on the dependent variables, it is the external, not the in-ternal, IJV locals who were the most influential social referent group, and the perceived fairness of compensation as compared with expatriates' and external IJV locals' compensation had greater im-pacts on turnover intentions than did the compar-ison with the internal IJV locals and domestic company locals. It appears that managers and pro-fessionals who are exposed to cross-national inter-actions are emerging as a special group that is ex-ternally and globally oriented in their choice of social referents. This idea is consistent with Good-man's (1974) finding that the level of professional-ism increases the likelihood of selecting referents outside the focal organization.
Second, how do employees perceive compensa-tion fairness when compensacompensa-tion advantage, neu-trality, and disadvantage vis a vis different social referents are present? Instead of the interaction that we hypothesized, we found a generalized spillover effect for the most influential referent, namely, the external IJV locals. We also speculated about a pos-sible consequent rigid endorsement of the differen-tiation principle of compensation. Future studies could explicitly test for the effects of the relevance of various referent groups and the endorsement of various allocation principles.
The dominance of the external IJV locals does not negate the relevance of expatriates as social refer-ents, as can be seen in the marginal main effect and significant interactive effect with interpersonal sensitivity. Had we found a more rigorous main effect for perceived compensation disadvantage in comparison with the expatriates, an additional sig-nificant, positive effect of advantage over locals would suggest an additive model—that is, that peo-ple sum up both advantage and disadvantage infor-mation to arrive at a fairness judgment. To fully explore interactive or additive effects of equally salient referents, researchers may want to study overall distributive justice perceptions instead of the referent-specific ones that we used in this study. The overall perceptions contain room in the dependent variable to allow different combinations of advantages and disadvantages associated with different social referents.
The third research implication lies in ideological explanations as applied to the international con-text. This study enriches notions of informational justice and social accounts. Whereas Creenberg's explanations (1993b, 1994) are more informational, we empbasize the ideological nature of explana-tions. In addition, we further expand individual-ized social accounts to collective explanations at the societal level; such explanations are particu-larly important for transformational changes in so-cieties and organizations, changes that often call for different frameworks of entitlement and fairness (Sampson, 1989). Finally, although interpersonal sensitivity has previously been,known primarily for its role in perceptions of interactional justice, our study shows an accumulative effect of general inter-personal sensitivity on social relationships in IJVs, affecting compensation fairness perceptions involv-ing comparisons with dissimilar social referents.
Practical Implications
This research has important practical implica-tions for IJVs. First, it brought social comparisons between expatriates and host nationals to the atten-tion of internaatten-tional HR managers. Although it would be unrealistic or even unfair to eliminate local-expatriate disparity, the magnitude of such disparity could be reduced to some extent. For example, alternative nonfinancial incentives, such as a fast track to senior management, could be as effective in promoting global mobility (Reynolds, 1997). Furthermore, companies could improve lo-cals' perceptions of distributive justice by develop-ing reasonable explanations and justifications for structural inequity. Finally, a positive message to expatriate managers and professionals is that al-though compensation issues are largely out of their control, being caring and sensitive to locals goes a long way in cross-national interactions, not only toward obtaining cooperation from culturally dif-ferent others, but also for remedying some of the hard feelings that result from the inherent and con-spicuous disparities between expatriate and local employees of international joint ventures.
REFERENCES
Adams, J. S. 1965. Inequity in social exchange. In L. L. Berkowitz (Ed.). Advances in experimental social
psycbology, vol. 2: 267-299. New York: Academic
Press.
Aiken, L. A., & West, S. G. 1991. Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage.
Alexander, S., & Ruderman, M. 1987. The role of
proce-dural and distributive justice in organizational be-havior. Social Justice Researcb, 1: 177-198. Berkowitz, L., Fraser, C, Treasure, F. P., & Cochran, S.
1987. Pay, equity, job gratifications, and compari-sons in pay satisfaction. Journal of Applied Psy-cbology, 72: 544-551.
Bies, R. J. 1987. The predicament of injustice: The man-agement of moral outrage. In L. L. Cummings & B. M. Staw (Eds.), Researcb in organizational bebavior, vol. 9: 289-319. Greenwich, CT: JAI Press.
Bies, R. J., & Moag, J. S. 1986. Interactional justice: Com-munication criteria for fairness. In R. J. Lewicki, B. Sheppard, & M. H. Bazerman (Eds.), Researcb on negotiation in organizations, vol. 1: 43-55. Green-wich, CT: JAI Press.
Blau, P. M. 1964. Excbange and power in social life.
New York: Wiley.
Brockner, J., Chen, Y.-R., Mannix, E. A., Leung, K., & Skarlicki, D. 2000. Gultiue and procedural fairness: When the effects of what you do depend on how you
do it. Administrative Science Quarterly, 45:
138-159.
Brockner, J., & Wiesenfeld, B. M. 1996. An integrative framework for explaining reactions to decisions: The interactive effects of outcomes and procedures. Psy-cbological Bulletin, 120: 189-208.
Chen, C. C. 1995. New trends in rewards allocation pref-erences: A Sino-U.S. comparison. Academy of Man-agement Journal, 38: 408-428.
Chen, G. C, Meindl, J. R., & Hunt, R. G. 1997. Testing the effects of vertical and horizontal collectivism: A study of reward allocation preferences in Ghina. Journal of Cross-Cultural Psycbology, 28: 44-70. Ghild, J. 1994. Management in Cbina during tbe age of
reform. New York: Cambridge University Press. Gropanzano, R., Rupp, D. E., Mohler, C. J., & Schminke,
M. Forthcoming. Three roads to organizational jus-tice. In J. Ferris (Ed.), Researcb in personnel and buman resources management. Greenwich, GT: JAI Press.
DeLisle, P., & Ghin, S. 1994. Remunerating employees in Ghina—The complicated task faced by foreign firms. Benefits & Compensation International, Septem-ber: 16-20.
Folger, R., & Konovsky, M. A. 1989. Effects of procedural and distributive justice on reactions to pay raise decisions. Academy of Management Journal, 32: 115-130.
Goodman, P. S. 1974. An examination of referents used in the evaluation of pay. Organizational Bebavior and Human Performance, 12: 170-195.
Greenberg, J. 1990. Looking fair vs. being fair: Managing impressions of organizational justice. In B. M. Staw
& L. L. Gummings (Eds.), Researcb in
organiza-tional bebavior, vol. 12: 111-157. Greenwich, GT: JAI Press.
Greenberg, J. 1993a. The social side of fairness: Interper-sonal and informational classes of organizational justice. In R. Gropanzano (Ed.), Justice in tbe work-place: 79-103. Hillsdale, NJ: Erlbaum.
Greenberg, J. 1993b. Stealing in the name of justice: In-formational and interpersonal moderators of theft reactions to underpayment inequity. Organiza-tional Bebavior and Human Decision Processes, 54: 81-103.
Greenberg, J. 1994. Using socially fair treatment to pro-mote acceptance of a work site smoking ban. Journal of Applied Psycbology, 79: 288-297.
Kulik, G. T., & Ambrose, M. L. 1992. Personal and situa-tional determinants of referent choice. Academy of Management Review, 17: 212-237.
Lee, R. T., & Martin, J. E. 1991. Internal and external referents as predictors of pay satisfaction among em-ployees in a two-tier wage setting. Journal of Occu-pational Psycbology, 64: 57-66.
Leung, K., Smith, P. B., Wang, Z.-M., & Sun, H. 1996. Job satisfaction in joint venture hotels in Ghina: An or-ganizational justice analysis. Journal of Interna-tional Business Studies, 27: 947-962.
Major, B. 1994. From social inequality to personal enti-tlement: The role of social comparisons, legitimacy appraisals, and group membership. In M. P. Zanna (Ed.), Advances in experimental social psycbology, vol. 26: 293-355. New York: Academic Press. Martin, J. E., & Peterson, M. M. 1987. Two-tier wage
structures: Implications for equity theory. Academy of Management Journal, 30: 297-315.
McFarlin, D. B., & Frone, M. R. 1990. A two-tier wage structure in a nonunion firm. Industrial Relations, 29: 145-153.
Niehoff, B. P., & Moorman, R. H. 1993. Justice as a me-diator of the relationship between methods of mon-itoring and organizational citizenship behavior. Academy of Management Journal, 36: 527-556. Oldham, G. R., Kulik, G. T., Stepina, L. P., & Ambrose,
M. L., 1986. Relations between situational factors and the comparative referents used by employees. Academy of Management Journal, 29: 599-608. Reynolds, G. 1997. Expatriate compensation in historical
perspective. Journal of World Business, 32(2): 118-132.
Sampson, E. E. 1989. The challenge of social change for psychology. American Psycbologist, 44: 914-921. Schminke, M., Gropanzano, R., & Rupp, D. E. In press.
Organization structure and fairness perceptions: The moderating effects of organizational level. Organi-zational Bebavior and Human Decision Processes. Sweeney, P. D., & McFarlin, D. B. 1993. Workers'
evalu-ations of the "ends" and the "means": An examina-tion of four models of distributive and procedural justice. Organizational Bebavior and Human Deci-sion Processes, 55: 23-40.
Wayne, S. J., Shore, L. M., & Liden, R. G. 1997. Perceived organizational support and leader-member ex-change: A social exchange perspective. Academy of Management Journal, 40: 82-111.
_A)5\
Chao C. Chen (cchen@stem.nyu.edu), who received his
Ph.D. from the State University of New York at Buffalo, is an associate professor of organization management at Rutgers University. He is currently visiting the Stern School of Business at New York University. His research interests include culture and organizational justice. A recent interest is the functionality of personal guanxi in Ghinese organizations.
Jaepil Choi is an assistant professor of management of
organizations at the Hong Kong University of Science and Technology. He received his Ph.D. in organizational behavior from Rutgers University. His current research focuses on organizational justice, cross-cultural issues, and leadership.
Shu-cheng Chi is a professor of organizational behavior
at National Taiwan University. He received his Ph.D. from the State University of New York at Buffalo. His current research includes managerial decision processes, conflict resolution, cross-cultural management, and Ghi-nese organizational behavior.
