Procedural justice as modernism: Placing industrial/organisational psychology in context

Une thCorie particulierement ethnocentrique-1’6quitC de traitement-est placCe dans un contexte thCorique plus large tirC des theories des institutions comparCes. Toutes les hypotheses ont CtC confirmCes grsce au traitement d’un Cchantillon de 1604 ingCnieurs et cadres d’entreprises lithuaniennes et arnkricaines du secteur Clectronique: (a) Les salariCs de 1’Ccononomie politique “nCo-traditionnelle” conqoivent leur organisation comme significativement (et considCrablement) rnoins mkritocratique que leurs pairs de l’bcononomie politique “moderne”. (b) La relation entre 1’Cconomie politique et la perception par les salariCs de 1’Cquitt de traitement est mCdiatisCe par l’usage que fait l’organisation des pratiques mkritocratiques. (c) L’CquitC de traitement est like a l’implication organisationnelle des salariCs et a la confiance envers les collbgues, m&me en tenant constante 1’Cconomie politique. C’est en situant 1’CquitC de traitement dans son contexte social que l’on a progress6 dans la comprkhension de son r61e dans le renforcement de la confiance faite aux collbgues; on a egalement pu mettre en Cvidence son importance aux yeux des salariCs, y compris dans les organisations ou elle Ctait inattendue. A particularly ethnocentric theory-procedural justice-is placed within a larger theoretical context drawn theories of comparative institutions. impersonality and anonymity, the relative political and economic independence of workers from management, and the resulting prominence of group conflict, bargaining, contract, and the relatively tight bureaucratic restriction of personal discretion of immediate supervisors. The employees’ perceptions of procedural justice, the greater employees’ trust in co-workers, even after controlling for political economy.

of the hypotheses (tested in a sample of 1604 engineers and managers in Lithuanian and American electronics companies) were supported: (a) Employees in the "neo-traditional" political economy perceived theirorganisations as significantly (and substantially) less meritocratic than did their peers in the "modern" political economy. (b) The relationship between political economy and employee perceptions of procedural justice was mediated by the organisation's use of rneritocratic practices. (c) Procedural justice was associated with employee organisational commitment and coworker trust, even when controlling for political economy. By placing procedural justice into its societal context, insight was gained into its role in fostering peer trust and its value to employees even in those organisations where it is unexpected.
I NTR 0 D U CTI 0 N IndustriaVorganisational psychology has developed into a particularly non-contextual applied social science. This is in contrast to some other branches of applied psychology that long have sought to explain behaviour in social, political, and economic environments (e.g. Atkinson, 1958;Bettelheim, 1943). When explanations for industriaUorganisationa1 psychological phenomena have been sought within any context larger than a particular employer, they have usually consisted of either an atheoretical description of different human resources practices in different countries or a study of links between national and ethnic cultures and the values and preferences of individual employees. These works have provided valuable documentation of differences in cultures and organisational practices across nations; yet they provide little explanation of how social structures and processes at the societal level may be linked to employee psychology and behaviour. For example, there has been speculation about the cultural specificity of particular human resources practices and theories (cf. Brewster & Burnois, 1991), but there have been limited attempts to explain such differences and only scant systematic research indicating that practices that work well in one national culture could not work as well if imported into a different culture. Therefore, we wish to add to the considerable insight gained from cross-cultural research by drawing on selected comparative institutions frameworks. Several prominent theories of comparative institutions seek to explain why different institutional arrangements have evolved (cf. Jacoby, 1985;North 1990;Zucker, 1986) and how it is that new institutional practices are or are not adopted (cf. Fallers, 1965;Putnam, 1993;Walder, 1986). As the borrowing and adaptation of organisational practices across national (and cultural) borders is an important and growing phenomenon, insight into ways in which institutions may affect individual behaviours and attitudes in organisations is timely.
However, there are risks in applying this scholarly work to industrial/ organisational psychology. Comparative institutions theorists rarely extend their arguments to hypotheses framed and tested at the individual level; industrial/organisational psychology researchers seldom test theories of the effects of societal context on organisational practices and individual psychology, and it is not difficult to understand why. First, components of such theories operate at very different levels of analysis, which leads to serious concerns about the validity of any observed relationships. Although correlates may be discovered between societal-level phenomena, organisational practices, and employee psychology, eliminating all of the alternative explanations is never possible; consequently, one cannot be sure that the causal arguments are complete. In addition, as a practical matter, it is more difficult to obtain samples of societies than to obtain samples of employees. Nevertheless, we believe the richness of insights that come from assuming a new perspective on industrial/organisational psychology merits this initial attempt at theory building and testing. As Staw (1995) suggested, studies wherein there is the greatest distance between independent and dependent variables, while controversial, are the potential "jackpots of social science".
The present argument consists of two parts. First, we have chosen a well-established concept from industriaUorganisationa1 psychology that is particularly ethnocentric-procedural justice. While noting the clear cultural origins of this concept, we demonstrate how a richer understanding of its role in organisations can be gained by drawing on several theories of social institutions. We suggest that procedures intended to foster fairness perceptions are inherent to different types of authority structures. Nevertheless, in the second part, we propose that historical experience with different authority structures should not be used to claim that employees will not respond positively to organisational practices that will benefit them, such as procedural justice. We test both arguments in a sample of organisations operating in the same industry in two countries posited to reflect different political economies-a newly independent Lithuania and the United States. This test thus forms the basis for extending our understanding of the effects of theories of societal factors on employee attitudes and perceptions, while it simultaneously examines the generalisability of one of our important (if highly ethnocentric) theories.

Procedural Justice and Modernism
Much of the research in the area of procedural justice has focused on organisational rules and policies as antecedents of fairness perceptions, which in turn are posited to affect employee attitudes and behaviour. Early studies of procedural justice examined responses to simulated dispute resolution processes (Thibaut & Kelley, 1975). One major finding of these initial investigations was that decisions following from procedures that offered disputants "process control" (i.e. the disputants could collect and present relevant information) were perceived as more fair than identical decisions that denied these people such control.
The process control element of procedures subsequently was labelled "voice" (Folger, 1977) and has been the subject of a substantial number of investigations over the last two decades (e.g. Folger, 1977;Folger, Rosenfield, & Robinson, 1983;Kanfer, Sawyer, Earley, & Lind, 1987;Lind, Kanfer, & Earley, 1990;Tyler, 1987;Tyler, Rasinski, & Spodick, 1985). Studies have found that procedures offering individuals voice consistently are perceived as fairer than those that do not, even after the effects of the decision outcomes have been taken into account. In addition, some evidence suggests these voice effects manifest themselves in Hong Kong as well as in the United States (e.g. Leung, 1987;Leung & Lind, 1986).
While voice has been the most prominent procedural characteristic in the organisational justice literature, other dimensions have been advanced. For example, Leventhal (1980;Leventhal, Karuza, & Fry, 1980) posited that people evaluate the fairness of a procedure on the basis of whether and to what extent the procedures meet any of six criteria: bias suppression, consistency, accuracy of information, correctability, representativeness, and ethicality. Research involving these dimensions has generally been supportive (e.g. Barrett-Howard & Tyler, 1986;Greenberg, 1986Greenberg, , 1987. Procedural justice research initially centred exclusively on the structural features of rules and policies as the basis for fairness evaluations. However, more recently the scope of inquiry has been expanded so that scholars are now examining "relational" factors as predictors of perceptions of procedural fairness (e.g. Bies & Shapiro, 1987;Lind & Tyler, 1988;Tyler, 1989Tyler, ,1990. For example, Lind and Tyler (1988) argued that the opportunity for voice leads to perceptions of procedural justice because it not only allows people to influence outcomes but also conveys the idea that they are respected and valued members of the group or organisation. As individuals may have many reasons to value long-term relationships with groups, they view evidence of acceptance as psychologically rewarding and are troubled by signs of rejection. In other words, certain procedures not only help to protect individuals from arbitrary treatment but may also serve to support social relations among participants.
Echoes of American cultural influences are certainly apparent in the study of procedural justice in organisations. The litigiousness of and, particularly, the overabundance of lawyers in the United States long has been noted and decried. Given this context, it is not surprising that so much of the industrial/oganisational psychology research that has originated in the United States is concerned with protecting the organisation from lawsuits by importing due-process procedures into organisational policy (cf. Guttman, 1993). The dominance of legal preoccupations is, if anything, increasing in American industriaVorganisationa1 psychology-cf. Sitkin and Bies's recent (1994) book, The legalistic organization. While ethnocentric to American preoccupations with legalisms, procedural justice research nevertheless reflects a more universal type of organisational form that comparative institutions theorists have called "modernism" or bureaucracy.
Many of the organisational factors that have been linked to employee perceptions of procedural justice (e.g. voice, bias suppression) reflect the meritocratic principles of universalism and impersonal decision-making that Weber (1947) associated with "bureaucracy". Bureaucracy is a kind of social organisation intended to enforce a merit-based, rule-governed, universalistic order (Clegg, 1990;Perrow, 1979), one in which all members are evaluated based on demonstrated performance, identically, regardless of personal characteristics or group membership. It uses formal procedures, Coleman's (1993) "purposive organisation", to foster impersonal treatment and perceptions of fairness when the scale of interdependent activity becomes too large for particularistic social control-that based on individual characteristics or group membership (Coleman, 1993;Parsons, 1951;Weber, 1947;Zucker, 1986). Certain procedural characteristics that have been shown to foster perceptions of justice, such as formal mechanisms for voice and bias suppression (e.g. Leventhal, 1980;Thibaut & Kelley, 1975), have long been considered surrogates for level of bureaucratisation by scholars of institutional development (cf. Coleman, 1993;Jacoby, 1985).
Yet, bureaucracy itself is largely a result of the interplay of a larger set of forces that here we call political economy. Political economy is a broad term referring to the study of the interaction between government and economic behaviour. It can include analyses of the role of government as an economic actor (e.g. Musgrave & Musgrave, 1975) and comparisons of the economic effects of different governmental policies (e.g. Gieger, 1979) among other phenomena. Here, our focus is on the organisational authority relations that may be fostered by governmental policies and practices.
One of the best descriptions of how societal context affects the actual use of organisational practices in different societies is an analysis by Fallers (1965) of what he called "Bantu Bureaucracy". In his study of the African Soga people (in today's Uganda) during the colonial period in the mid-twentieth century, Fallers discovered what he described as a conflict arising from an incompatibility between the Soga's traditional practices and the European-imposed "modern" bureaucracy. Whereas the former, based on lineage and chiefdoms, is particularistic, the latter is universalistic. These incompatible objectives resulted in interpersonal conflicts and instability in administration. For example, universalistic colonial rules dictated that employees should be selected based on merit, yet tribal chiefs also had strong moral obligations to distribute resources to kin and loyal dependents. Fallers (1965, p.243) indicated that universalistic meritocratic organisational practices were adopted only nominally: In the lower courts in Busoga, . . . procedure often assumes that the community knows, or can discover, the circumstances of the case. There is not the same obligation to tell the truth which one finds at higher levels, where anonymity is possible. Litigants and witnesses are more free to lie, since it is assumed that members of the court, knowing the persons involved, can nevertheless discover the truth. The problem is rather to find the most satisfactory means of closing the breach in the social relations of the community. The court applies substantive justice rather than strictly-defined rule of law.
Particularism and concern for substantive justice (that is, a just outcome, regardless of how that outcome is obtained) is apparent in all societies, yet the difference in degree is an important one. Of course, Weber never intended his ideal-type bureaucracy to be a description of actual organisational functioning; rather, he attempted to present certain organising principles in unambiguous form. Yet, differences in the extent to which authority is intendedly universalistic (however imperfectly implemented; cf. Perrow, 1979) or intendedly particularistic can have a substantial impact on organisational practices and participants' expectations. This is illustrated by Putnam's (1993) contrast of the evolution of new regional governmental institutions in the relatively more universalistic northern Italian regions with the ways in which the more particularistic southern Italian regions constructed these mandated organisations. Despite the same formal national requirements for these new regional governments, different civic traditions in the north and the south resulted in the development of quite different organisations. In the south, with vertical dependence (dependence on superiors) rather than the horizontal dependence (on collective action) of political and mutual aid associations and the south's long tradition of amoral familialism, which Putnam characterised as "clientelism", the new regional governments were captured by local autocrats. With the widespread distrust characteristic of social institutions in southern Italy (Gambetta, 1988), local citizens never expected an organisation to operate in any other way. In the north, these new governments became relatively more universalistic and meritocratic in their treatment of local citizens. It is not that particularism was unknown in the northern Italian regions so much as that it was constrained by existing horizontal associations and expectations for impersonal rule-of-law.
The best example of the use of traditional authority structures in what are nominally bureaucratic organisations can be found in communist countries. Walder (1986, p.10) described what he called the "neo-traditionalism" of the workplace in the People's Republic of China. He suggests that all communist states rely on neo-traditionalism, a system characterised by: dependence, deference, and particularism . . . in contrast with the more familiar modern forms of industrial authority that are notable for their relative impersonality and anonymity, the relative political and economic independence of workers from management, and the resulting prominence of group conflict, bargaining, contract, and the relatively tight bureaucratic restriction of personal discretion of immediate supervisors.
In communist neo-traditionalism, the workplace is the focal point for the delivery of public goods and services not available from other sources, the party tries to eliminate all competing informal political organisations of employees, and the discretion of supervisors is relatively unrestrained by enforceable regulations (Pearce, Branyiczki, & Bakacsi, 1994). Despite these organisations' bureaucratic formalism, particularism was necessary to communist political control. In order to maintain political control, wide discretion was delegated to managers and party officials (usually the same people) who then had wide latitude to exercise this discretion as they saw fit. Thus, they had influence over promotions, raises, and the non-wage benefits supplied by the employer. Walder writes (1986, p.22) that in China: Shop officials screen requests for factory housing and special distributions of consumer items. They review and approve requests for benefits under state labor insurance guidelines: vacations, annual home leave, personal leave, visits to sanatoria, special medications, and welfare and loan payments. Shop supervisors are also responsible for writing character reports, relaying information to the party and security apparatus, securing permission for workers to travel, and deciding on the application of fines and other punishments for breach of factory rules.
Although such supervisory personal discretion may have been more extensive in Asian than in European communist workplaces, in all of these societies the local managers and party officials retained wide discretion to make particularistic decisions about more matters of importance to employees than in any developed-world workplace (Gregory, 1989;Pearce, 1991;Pearce, Branyiczki, & Bigley, 1995;Voslensky, 1984). This resulted in the central feature of communist-society workplace culture-4ientelism (Walder, 1986, p.25): Party-clientelism is not comprised of personal ties that exist separately from the formal organization of relationships and roles: it emerges from standard party recruitment and leadership practices-indeed, it is created by them. The party and management seek to control the workforce and elicit its active cooperation by developing stable ties with a minority of loyal and cooperative workers.
Such neo-traditionalist organisations have many of the visible forms of bureaucracy-hierarchy, paperwork, written rules-but these are in the service of particularistic decisions by those in positions of power, whose primary objective is the maintenance of their own power, not efficient commercial aviation or the manufacture of reliable buses or attractive shoes. Of course, non-meritocratic organisations exist in the developed world, as many a terrorised employee can attest. However, large employers in developed societies, compared to their counterparts in traditionalist and neo-traditionalist political economies, find themselves relatively more constrained by labour laws, trade union contracts, social welfare legislation, the threat of litigation, and requirements to file public reports of their finances and activities.
These two authority forms have been called by many different names. Weber (1947) distinguished bureaucracy from traditionalism; Riggs (1964), substantive bureaucracy from formalistic bureaucracy; Fallers (1965), modern bureaucracy from Bantu bureaucracy; Putnam (1993), civic communities from patron-clientelism; and Coleman (1993), modern from traditional forms of authority. Here, because we are reporting the reactions of employees in a political economy just emerging from communism, we shall use Walder's term "neo-traditionalist" to characterise the particularistic and vertically dependent form of social organisation we expect to find there. The comparison country is one of the wealthiest developed-world economies, with businesses among the earliest to adopt bureaucratic organisational forms. Thus, Coleman's term "modern" would be the closest parallel to "neo-traditionalist".
Although scholarship on modernism and traditionalism has a long history, and the distinction noted earlier is widely accepted in social science, we do not know whether these differences are reflected in employees' perceptions of the extent to which their organisations are more or less meritocratic. We are familiar with numerous examples of particularistic and clientelistic practices in modern political economies (cf. Kanter, 1977;Van Maanen & Pentland, 1994). It could well be that this distinction is a fine one, visible to academic specialists but having no practical meaning to employees. If employees themselves do not perceive differences in their organisations' practices, generalisations from these scholarly bodies of knowledge to organisational behaviour would be moot. Therefore, we begin by testing whether differences in the use of meritocratic organisational practices in neo-traditionalist and modern political economies, as described by Pearce (1991), Voslensky (1984, and Walder (1986) are reflected in employees' reports: H,: Employees in a neo-traditionalist political economy will report that their employing organisation is less meritocratic than will be reported by employees in a modern political economy. Political economy is proposed to affect procedural justice, not directly but indirectly, via its effect on the organisation's use of meritocratic practices. Organisational practices intended to foster procedural justice constrain managers. These are internal organisational rules and policies by which the organisation's management agrees to constrain its own power by requiring such practices as grievance procedures and rules regarding written warnings and the use of objective criteria (Taylor et al., 1995). We suggest that managers and owners do not willingly limit their own power. This is based on the observations of historians that managers have rarely chosen to constrain themselves with procedural niceties. Jacoby (1985) described the evolution of professional personnel policies and the specialists who developed and enforced them as an attempt to preempt worker protective legislation and trade union organising in the United States. He discussed how trade unions and professional personnel specialists both worked to bureaucratise the personnel systems in organisations as the best defence against arbitrary supervisory power. Workers in democratic polities have some means to protect themselves from arbitrary actions, and they often seek to do so by imposing bureaucratic procedures of objective assessments of merit and due process guarantees on their employers (Jacoby, 1985). Thus, for example, political pressure in the United States to eliminate hiring practices that discriminated against women and minorities fostered wider adoption of more meritocratic selection practices based on job-relevant criteria (Guttman, 1993). Over time, employers were compelled to use more meritocratic grounds to defend their decisions.
In contrast, under traditionalism, the only reliable form of protection is to become the valued dependent of a powerful person or family (Gambetta, 1988;Putnam, 1993). Those in power may use or may not use meritocratic criteria to select their dependents; for example, they might prefer to use loyalty or the ability to do useful favours. In any case, they are not constrained to provide public defences of their choices based on meritocratic grounds: H,: A relatively more modern political economy will have a positive effect on employees' reports of procedural justice via the relatively greater use of meritocratic practices by its organisations.

Reactions to Perceived Procedural Justice
In a society in which employees have learned what protections impersonal universalism can provide, we would expect employees to respond positively to procedural justice in their workplaces. Empirical research in modern societies has found a positive relationship between procedural justice and various work attitudes and behaviours (e.g. Alexander & Ruderman, 1987;Brockner, Tyler, & Cooper-Schneider, 1992;Folger & Konovsky, 1989;Lind et al., 1990;McFarlin & Sweeney, 1992;Moorman, 1991;Neihoff & Moorman, 1993;Tyler, 1991). This research suggests that perceptions of injustice will lead to a range of negative outcomes, including low levels of organisational commitment and job satisfaction, reduced task performance, and employee withholding of organisational citizenship behaviours.
Yet, what about polities in which particularism and clientelism are the norm? Here, we will argue that the actual use of these procedures will matter to employees, whether or not they reflect normative, taken-for-granted expectations. That is, the arguments of scholars writing on procedural justice are not dependent on established employee expectations. If actually practised, due process safeguards (e.g. procedures that are open, suppress bias, encourage the collection of accurate and complete information, have mechanisms for appeal) will have positive consequences even among those who have learned not to expect them. Employees in all political economies would be expected to value an opportunity to escape from arbitrary treatment and to receive the social respect such procedures imply. This reflects an important distinction between employees' accommodation to a bad situation and their embrace of it. That is, we suggest employees' past experiences and expectations should not be used as the basis for an argument that they will not respond positively to practices that are favourable to them, such as procedural justice. To test this idea, we propose that, despite the ethnocentric origins of the concept of organisational procedural justice, employees in settings in which just procedures are far from a reasonable expectation will react in much the same way to perceptions of organisational justice as employees who expect no less. In this study, we have been able to test two expected outcomes of perceived procedural justice: organisational commitment and trust in co-workers.
For employees working in modern political economies, procedural justice has been shown to be positively related to organisational commitment (e.g. Folger & Konovsky, 1989;McFarlin & Sweeney, 1992). Therefore, it seems appropriate to begin the test of the generalisability of procedural justice with a test of its relationship with organisational commitment within a neotraditional political economy: H,: The greater the perception of procedural justice, the greater the employees' organisational commitment, even after controlling for political economy.
Further, following Gambetta's and Putnam's work on clientelism and trust, we suggest that procedural justice has an effect on trust at work. Several studies have demonstrated a positive relationship between procedural justice and trust (e.g. Alexander & Ruderman, 1987;Folger & Konovsky, 1989;McFarlin & Sweeney, 1992;Tyler, 1989). In this work, though, the target of employee trust was the supervisor rather than co-workers. We propose that perceptions of procedural justice will foster more general workplace trust in others. Zucker (1986) has argued that one form of trust between individuals and groups is based on expectations held in common about the terms of an exchange relationship, thus just procedures are a mechanism that encourages shared expectations among employees. Employees who understand the "rules" under which they are all governed have less to fear from others.  reported results of research in Hungary indicating that arbitrary actions of those in power, what they called person-based reward systems, were associated with widespread employee distrust. Personbased reward systems are a combination of highly valued organisational rewards and personalistic (particularistic) criteria for reward distribution. They found that employees working in organisations with person-based reward systems reported that they were more anxious, that their employers were more unfair, and that their co-workers were less competent than did employees working in "performance-based reward systems".
These evaluations by the Hungarian employees reflected greater distrust of both their employers and co-workers. That employees would distrust the employers who distribute rewards based on particularistic criteria may be expected. Yet, that employees in these neo-traditional workplaces should distrust their peers is less intuitively obvious. Therefore, we again turn to those who have studied traditionalist societies. Gambetta (1988, pp.162-163) described this phenomenon in patron-clientelist southern Italy: . . . distrust percolates through the social ladder and the unpredictability of sanctions generates uncertainty in agreements, stagnation in commerce and industry, and a general reluctance towards impersonal and extensive forms of cooperation. Sicilians-as everyone knows-do not trust the state; beyond the boundaries of limited clusters, they often end up distrusting each other as well.
In arbitrary authority relations, it is hard for peers to know where they stand with one another. Because the criteria for rewards are unclear, no one knows whether or not they have done enough or whether someone else has done more . Further, powerful figures sometimes purposely sow dissension among their subordinates in order to strengthen their own positions (Putnam, 1993;Voslensky, 1984). As Zucker (1986) noted, clear rules reduce the fear of arbitrary attacks from those outside one's own "limited cluster". As power in traditionalist societies is based more on connections than on office, even organisational peers can pose a powerful threat. Peers may be well-connected themselves and even if they are not particularly well-connected, one can never be sure that they won't seek to ingratiate themselves to the powerful by denouncing or informing on you (Haraszti, 1977;Pearce, 1991;Walder, 1986). Despite these persuasive descriptions of the effect of procedural injustice on peer distrust, these possible effects of justice perceptions have not received adequate empirical attention. Therefore, we offer the following hypothesis regarding the relationship between perceptions of procedural justice and trust in coworkers: H4: The greater employees' perceptions of procedural justice, the greater employees' trust in co-workers, even after controlling for political economy.

METHOD Sample
The sample consists of (non-executive) managerial and professional employees in five large electronics manufacturing organisations. All companies were considered "high technology" in their markets and employed large research and engineering staffs to maintain their preeminent technological positions. One of the five was in Lithuania, classified as a neo-traditionalist political economy at the time of data collection in 1993. The other four elctronics companies were American offices of large international electronics companies (all American-owned but with worldwide operations and sales). These were classified as operating in a modern political economy at the time of data collection in 1992. Summary information about the sampled companies, sample sizes, and response rates are provided in Table 1. As comparative-institutions scholars commonly categorise the United States as a relatively modern political economy, there is little dispute in the classification of these American electronics companies, but the classification of the Lithuanian company may be less straightforward.
Lithuania has a unique national language and identity despite its occupation and control by Germans, Poles, and Russians, in rotating order, over the centuries. At the time of data collection Lithuania had only recently succeeded in its long struggle to be independent of the Soviet Union; yet in 1993 it had inherited Soviet institutions and was still deeply interdependent with other components of the old Soviet command economy (an economy now spread across independent republics). Following the writings of Walder (1986), we would expect Lithuania to be a neo-traditionalist society.
An attempt to confirm this preliminary classification was conducted via interviews with members of management. We found, consistent with the arguments of Walder (1986), that there was little expectation of impersonal treatment. Further, the breakdown of the Soviet state had led to an initial escalation of favouritism and corruption. To cite an example: according to newly passed privatisation legislation, anyone was able to apply to purchase a building owned by the state, paying only a nominal fee. A partner in an entrepreneurial group described their actual experience: the partnership applied to the building privatisation office for a building for their business and were told to find a building that suited their purposes and return to apply for it. (They were given no indication of which buildings were available.) The partners selected a building and went back and applied for it, only to be told "That one is not available; go select another one." No further explanation was offered. They came back with another choice, only to be told the same thing. After several similar rounds, the entrepreneurs finally understood that there would never be any building available to them unless they paid a bribe. As they didn't have enough money for a bribe, they could not obtain a building. Of course, the defining feature of bureaucracy is the application of written rules in an attempt to reach a decision based on the impersonal merits of the case. Weber (1947) and his many followers note that using offices as an opportunity to "sell decisions" for personal gain is a feature of traditionalism. This was hardly an isolated incident in Lithuania. At the time of data collection, a local entrepreneur had become so disgusted with government corruption that he put up the equivalent of about US$lO,OOO for a newspaper to hold the following contest. A government official could win the prize money by being the first official to nominate him or herself as having been honest during his or her term of government service and surviving 10 days of investigation by the newspaper. If the newspaper found evidence of wrongdoing it would publish the self-nomination letter with the uncovered evidence that the nominee had not been honest. If nothing incriminating could be uncovered in 10 days, the nominee would win the prize. By the date of the first author's departure from the country, dozens of government officials had nominated themselves (sometimes with pathetically selfserving letters detailing their sacrifices for their country), and every single one had failed to survive a 10-day investigation. The entire country was riveted to this uproarious series, which boosted the newspaper's circulation beyond the publisher's dreams. This account not only illustrates the pervasive corruption in Lithuanian government at the time of data collection but also reflects a society in which privacy rights and libel laws were not as constraining as they would be in the United States.
The lack of a functioning legal structure in Lithuania had a pervasive impact on the electronics company. The Personnel Director gave the following account of his difficulties: the new Lithuanian laws recognised the right of employees to unionise and sign collective bargaining agreements with their employers, but it did not specify how bargaining units were to be determined. Further, if unions were present, they had to approve any layoffs of their units' members. Several months prior to data collection, the company had sought to lay off 30 engineers from a unionised group of several hundred engineers and the union had approved the layoff. However, when the 30 engineers received their layoff notices, they met together, declared themselves to be a new union (of the 30), and refused permission to lay off its members. The issue was pending before the courts-no cause for optimism. A recent court case initiated by an employee of the company disputing her layoff had dragged on for two years before there was finally a hearing before a judge. The judge asked, "Has this woman done anything wrong?" "No," the Personnel Director replied, "She was not fired for poor performance. She was laid off because we have no work for her to do." The judge replied, "Well, if she hasn't done anything wrong, why are you picking on her? You are a big company and she is a poor woman. You should take her back." Thus, although the Lithuanian and American companies are in the same industry (electronics manufacturing for other manufacturers and consumers) and the same occupations were sampled in all organisations, the Lithuanian and American companies found themselves operating in vastly different political economies. While the Americans could assume a stable political and legal system built over centuries that was generally seen to be constraining, the Lithuanians found themselves in a revolutionary period, leaving a system regarded by scholars such as Simis (1982) and Voslensky (1984), among others, as subject to rule-by-party rather than rule-of-lawnow without either party or adequate law. This certainly isn't the only difference in the environments faced by these companies-for example, the Lithuanian company was a former oligopoly in a command economy and was learning how to compete for customers, while the American companies had always faced severe competition. Nevertheless, we believe the political economies are consistent with the hypothesised constructs and allow a test of the impact of political economy on employee perceptions and attitudes. Yet, readers should recognise that this one study of one company in one neo-traditionalist society cannot be definitive.

Measures
All data used for these tests were taken from anonymous questionnaires, with the study identified as a university research project undertaken by the first author. All companies were provided with summary questionnaire feedback (provided in a way that no one in the company could be individually identified). This feedback was accompanied by three days of managerial training by the first author in the Lithuanian company. As can be seen in Table 1, all of the response rates were acceptable.
All questionnaire items were constructed first in English. They were initially translated into Lithuanian by Dr. Arunas Kuras. This translation was back-translated into English by a professional translator following the procedure recommended by Brislin (1986). Dr. Kuras and the first author reviewed and discussed the back translation in depth, making any final decisions regarding wording. As Poortinga and Van de Vijver (1987) state, the question of separating substantive differences from bias in comparing cross-cultural responses poses a serious dilemma. When comparing only two cultures (as is done here) any systematic component can affect bias, and item analysis alone cannot untangle the two. Nevertheless, we attempted to at least eliminate serious problems in scale meaning by factor analysing the items separately for each country-sample (Adler, Campbell, & Laurent, 1979). As all items loaded on the same scales in both language-samples, we maintain some confidence that the scales retained similar core meanings to the two language groups. Poortinga and Van de Vijver (1987) suggest that the best solution is to "unpack culture" by specifying context variables that may account for the differences. We view the present study as an attempt to establish what we believe will be an important context variable: political economy. All measures are 5-point Likert-type scales with 1 = strongly disagree to 5 = strongly agree.
The Organisational Commitment scale is the short form of Mowday, Steers, and Porter's (1979) scale that has been used extensively in organisational behaviour research; it obtained as of .87 in the US sample and .82 in the Lithuanian one. See Table 2 for the means, standard deviations, intercorrelations, and internal consistency coefficients, separately by country, for the variables used in this study. The scale assessing trust in co-workers has been used by  and Pearce et al. (1995). A sample item is "I can rely on those I work with in this group", and the scale had an a = 3 9 in the US sample and an a = .85 in the Lithuanian sample.
The scales assessing meritocratic practices and procedural justice are relatively new. The meritocratic practices scale was developed in a large study of human resources practices reported in Pearce and Tsui (1994). This particular scale was selected because the based rewarding of individuals is a fundamental feature of meritocracy. The meritocratic practices scale had an a = .77 for the US employees and a = .76 for the Lithuanians. Although this scale has the same good convergent and discriminant validity in the Lithuanian sample as in the American one, it should be noted that in Lithuania performance assessment was not formalised in written performance appraisal systems as was the case in all American companies for employees at this level. The procedural justice scale is a compilation of items based on the work of Folger and Konovsky (1989), Leventhal(1980), Lind and Tyler (1988), and Thibaut and Kelly (1975) and had as of .94 for the US and .85 for Lithuania. These scales showed good discriminant validity from one another in this sample for both language samples separately. Finally, although we were unable to control for possible company effects in Lithuania, in the US sample we were able to analyse possible companyspecific effects. First, we computed ANOVAs for all dependent variables by company in the American sample. We found no cross-company differences for procedural justice or co-worker trust. However, as employees in "Company A" reported significantly higher use of meritocratic practices while "Company B" employees reported more organisational commitment, these two companies were entered as covariates for the tests of these dependent variables. Taken as a whole, these tests suggest that the scales have good convergent and discriminant validity, and the internal consistency coefficients indicate acceptable scale reliability.

RESULTS
Hypothesis 1 proposed that employees in the neo-traditional political economy will perceive their organisations to be less meritocratic than will employees in the modern political economy. This hypothesis was tested by regressing meritocratic practices on political economy, with the results reported in Table 3. We see that this hypothesis was supported, with a substantial 21% explained variance in employee reports of their organisation's use of meritocratic practices accounted for by the differences in the political economies in which these electronics companies operate. Thus the employees working in these electronics companies confirm the observations of social scientists that the workplaces in neo-traditionalist political economies reflect the arbitrary polities in which they exist. As employees identify the differences in use of meritocratic forms of organisational practice described by scholars, it is possible to test subsequent hypotheses concerned with whether or not they react to these differences in the ways predicted. Hypothesis 2 proposed that the influence of political economy on employees' reports of procedural justice would be mediated by the companies' use of meritocratic personnel practices. In Table 3, we can see that despite the significant direct relationship between political economy and procedural justice (see Table 2), when meritocratic practices is added to the equation, the beta for procedural justice drops from a significant .27 to a non-significant .05. Thus, these data are consistent with a causal sequence which runs from political economy to the use of meritocratic practices in organisations to employee perceptions that the organisation's procedures are just. That is, despite the substantial cultural and other differences in the societies in which these electronics employees are working, these politylevel differences affect perceptions of procedural justice via their effects on their employers' use of meritocratic practices.
There were two hypotheses proposing to test whether or not the rather ethnocentric concept of procedural justice would also have the same employee attitudinal correlates in a society in which employees had little reason to expect or have confidence that administrative procedures could promote justice. The first, Hypothesis 3, stated that procedural justice would be associated with employee organisational commitment, even after controlling for differences in the employees' political economies. Table 4  reports the results of the test and indicates that the hypothesis was supported. Even when political economy is controlled, procedural justice and organisational commitment still share a significant 21% variance. Finally, Hypothesis 4 posited that procedural justice would be associated with greater trust in co-workers, after controlling for political economy. In Table 4, the results indicate confirmation of the hypothesis. Thus, we find that procedural justice, despite its origins in a particular nation's legalistic culture, is significantly associated with employee commitment and trust even in political economies in which employees have learned not to expect it. Given all of the other factors that might affect an attitude such as commitment (such as, for example, the quality of supervision) or trust in co-workers (such as the personalities of the employees), that 25% and 13% of the variance (respectively) can be explained by societal-and organisational-level phenomena is noteworthy.

DISCUSSION
The results of the hypothesis testing support the argument that meritocratic procedures are a feature of modern political economies. Yet, despite procedural justice's ethnocentric origins, employees react negatively to unjust procedures, whether or not they have any realistic expectation of (or experience with) this form of justice. We found that employees in the neo-traditional political economy perceived that their organisations were significantly (and substantially) less meritocratic, as reflected in rewarding individual performance, than did their professional peers in the modern one. Further, the relationship between political economy and employee perceptions of procedural justice was mediated by the organisation's use of meritocratic reward practices, as expected. Finally, the results were consistent with the hypotheses that perceptions of procedural justice are associated with employee organisational commitment and co-worker trust, even after control for differences in political economy.
Before discussing these results, the limitations of this initial theory testing should be reiterated. First, the measures of the theories (modernisdneotraditionalism and meritocracy) capture only a few facets of these complex phenomena, and so these theoretical interpretations must proceed with caution. Further, it is never possible to be completely confident that all relevant causal variables have been controlled when phenomena are linked at such disparate levels of analysis. Although these data were consistent with the arguments and we were able to establish the pivotal role of organisational practices as mediating between political economy and individual perceptions, the results should be interpreted with caution. New research might profitably posit alternative causal reasons for these observed relationships and then test the competing models. With these cautions in mind, our results do suggest that placing organisational practices into a larger explanatory context can provide new insights. We found the meritocratic procedures of rewarding individual performance indicative of the modern form of political economy. Hitherto the literature in industrial/organisational psychology has reported analyses of the effects of different concepts and programmes as decontextualised tool^''. The implication has been that a formal performance appraisal system might be needed here and a job analysis there, with little recognition that such techniques themselves reflect a particular form of authority relation. The work of comparative institutions scholars, such as Fallers and Putnam, suggests that attempts to impose isolated human resources programmes (for example, a formal grievance procedure) would not be likely to be effective if the larger organisation were not one based on meritocratic universalism. In organisations dominated by patronclientelism, such a system simply would be expropriated as a vehicle for displaying loyalty to the "patron". Putnam documented these processes in southern Italy, and Walder provided a rich description of the conversion of procedures intended to improve factory performance in China into opportunities to display loyalty to the powerful. A better understanding of the context of organisational practices could prove useful in predicting "implementation failures".
Yet, organisational practices do change over time. One advantage of a focus on theories of comparative institutions is that it allows us to draw on the rich scholarship on institutional change. Why do fundamental authority relations change? According to Zucker (1986), it is when the scale and scope of interdependent activity becomes too complex to be accomplished by primary groups and when mobility and cross-cultural contact break down these face-to-face groups. All of these historical forces continue to extend their influence in more areas of the world, suggesting, as Weber argued over a century ago, that the bureaucratic form will continue its spread. This has implications for both theory building and practice. An example of the latter is that sophisticated human resources practices, an important means for implementing modernism, may spread with increasing cross-cultural economic interdependence.
Still, the comparative institutions literature also provides evidence against the inexorable march of bureaucracy. Redding (1990) studied the (so far) unique group, which has had inordinate economic success using traditional primary-group authority structures-the "overseas Chinese". He provides evidence that overseas Chinese have had to rely on the primary groups of family and clan because they have operated in hostile political jurisdictions, but they have accommodated to this situation by selecting technologies simple enough to be controlled personally (e.g. trade and simple manufacturing). Further, like all arbitrary fiefdoms, these organisations face succession problems, rarely outlasting their founders. Most interesting for students of industriaUorganisationa1 psychology is Redding's description of the employee attitudes, which seem very close to the unhappy descriptions of employee attitudes in neo-traditionalist organisations reported by Pearce et al. ( ,1995, Walder (1986) and here. Redding's (1990) work suggests that traditionalist and neo-traditionalist organisational authority structures may not necessarily become bureaucratised, but rather may be applied to those tasks where they face no technical disadvantage.
This study also extends our understanding of procedural justice. We found that perceptions of procedural justice were associated not only with a greater trust in supervisors but also with a greater trust of peers. This particular finding helps to illustrate the utility of placing procedural justice in a larger context as a component of the bureaucratic organisational form characteristic of modern political economies. With a focus on micro-level bivariate relationships among individuals, this relationship might appear inexplicable, particularly to those with little experience of traditional and neo-traditional societies. Yet, when we draw on societal-level studies, such as work by the political scientist Putnam (1993), the anthropologist Yang (1994), and sociologists like Gambetta (1988), Haraszti (1977, and Walder (1986), the relationship between perceived injustice and peer distrust becomes clearer. Particularistic vertical relations often foster rivalry for the favour of the powerful. There are many rich insights in the other social sciences, and we hope this study illustrates the profit from reading beyond cross-cultural psychology.
Finally, we found that a perception of procedural justice was important to employees in Lithuania, where they had little reason to expect that formal meritocratic practices could afford them just treatment. This suggests that the unambiguous ethnocentrism of research on procedural justice should not imply that employees in other cultures would not respond to procedural justice in similar ways, if these practices benefit them. Hofstede (1980) in his classic article revealed the cultural origins (and distortions) of prominent theories of management. We conclude by noting that the theories he discussed probably would not have become so prominent if they did not contain at least a modicum of universal appeal.

Manuscript received March 1996
Revised manuscript received August 1996