Impact of Employees Turnover

A Conceptual Model of Expatriate Turnover Author(s): Earl Naumann Reviewed work(s): Source: Journal of International Business Studies, Vol. 23, No. 3 (3rd Qtr. , 1992), pp. 499-531 Published by: Palgrave Macmillan Journals Stable URL: http://www. jstor. org/stable/155094 . Accessed: 09/10/2012 02:27 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www. jstor. org/page/info/about/policies/terms. jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive.We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email  protected] org. . Palgrave Macmillan Journals is collaborating with JSTOR to digitize, preserve and extend access to Journal of International Business Studies. http://www. jstor. org A CONCEPTUAL MODEL OF EXPATRIATETURNOVER Earl Naumann* Boise State University Abstract. Many factorsmay contributeto the high rate of expatriate turnover experienced by U. S. ultinational corporations. The objective of this article is to present a comprehensive model of the expatriateturnoverprocessby identifyingthe key contributing factors and suggesting intermediate linkages and relationships. The model appears to have both managerial and theoretical implications. The majority of U. S. multinationalcorporations(MNCs) suffer from an abnormally igh turnover ateamongexpatriate anagers, articularly hen h r w m p t T compared o bothforeign-based NCsanddomesticoperations. he turnover M ate commonly falls in the 20%-50%range for expatriatetransfers[Black 1988; Black and Stephens 1989; Copelandand Griggs 1985; Mendenhall and Oddou 1985;Tung 1988;ZeiraandBanai 1985], althoughturnover ay m be as high as 70% for some firns, particularlyn less developed countries i [Desatnick and Bennett 1978; Lanier 1979]. The problem of expatriate i turnovers more evidentamongU. S. -basedMNCs thanamongforeign-based MNCs. Tung [1982] noted that turnoverrates among U. S. MNCs are two to three times higher than foreign MNCs. For example, Tung's research M ndicatedthatonly 3%of European NCs and 14%of JapaneseMNCs had turnoverrates greaterthan 10% of expatriatetransferswhile 76% of U. S. MNCs had turnoverrates exceeding that level. b t Althoughdirectcomparisons etweendomestic(U. S. ) andexpatriateurnover ratesaredifficult,expatriateurnover atesappear o be at leasttwice domestic t t r rates. â€Å"Expatriateturnover†or â€Å"expatriatefailure† typically includes all individualswho quit or transferback to the U. S. priorto the completionof theirexpectedforeignassignments. ome researchersontendthat†expatriate S c turnover†may also occur up to a year or more after repatriation[Adler 1986; Harvey 1989].Conversely,virtuallyall domestic studies of turnover refer to the separationof the emp loyee from the organizationwith the o o t T preponderancef the studiesconcentratingn voluntaryurnover. o illustrate, in McEvoy ndCascio's 1987]meta-analytic a reviewof turnoverndperformance, a [ *Earl Naumannis Professorof Marketingat Boise State University. He received his Ph. D. from ArizonaState University. The authorwould like to thankthe reviewersfor their effort and contributions hich helped signifiw cantly improvesome crudeideas. Received: September1990; Revised: February,June, September1991 & January1992; Accepted: February1992. 499 500JOURNAL OF INTERNATIONALBUSINESS STUDIES, THIRD QUARTER 1992 twenty-four studies involving 7,717 individuals were cited. External turnover rates varied from 3% to 106% annually with a median of 22%. However, the subjects in the studies were drawn from nineteen rather generic job s categories including aerospace employees, pharmaceutical cientists, engineers, navy enlisted personnel, nurses, bank tellers, and small business new hires, for examp le. None of the subjects are directly comparable to expatriate managers. If the assumption can be made that expatriate managers are more similar to individuals such as engineers, scientists, and exempt lectronic employees than to individuals such as nurses, enlisted personnel, or small business new hires, the â€Å"normal domestic† turnover rate is more likely to approximate 10% annually [McEvoy and Cascio 1987]. This figure is consistent with the average of 12% for 303 firms reported by Mercer [1988]. Further, turnover rates typically decline at higher organizational levels so managerial turnover rates are likely to be less than the overall average figure. Thus, while no empirical research exists that directly compares turnover, the expatriate turnover rate appears to be at least twice the domestic rate.The extreme expatriate turnover rate results in high direct and indirect costs to U. S. MNCs. The direct costs associated with each expatriate turnover are estimated to be be tween $55,000 and $150,000 [Copeland and Griggs 1985; Harvey 1985; Mendenhall, Dunbar and Oddou 1987; Misa and Fabricatore 1979; Zeira and Banai 1985]. Therefore, the aggregate direct costs for expatriate turnover for U. S. MNCs are quite high [Copeland and Griggs 1985]. Additionally, the indirect costs associated with expatriate turnover are estimated to be even greater [Harvey 1985].The indirect costs include items such as reduced productivity and efficiencies, lost sales, market share, competitive position, unstable corporate image, and tarnished corporate reputation. While turnover is costly in domestic operations, turnover is apparently even more costly and troublesome in overseas operations. Due to the relatively high frequency of expatriateturnoverand the associated costs, the international business literature has been punctuated with efforts to isolate the factors causing difficulty in cross-cultural-adjustment.The three areas receiving the greatestinteresthave been the sele ction of the â€Å"ideal† expatriate manager [Abe and Wiseman 1983; Church 1982; Mendenhall and Oddou 1985; Mendenhall, Dunbar and Oddou 1987; Tung 1982, 1988; Zeira and Banai 1985], the underdeveloped state of expatriation and cross-cultural training programs [Black 1988; Black and Mendenhall 1990; Brislin 1979; Mendenhall and Oddou 1985; Kohls 1985; Schwind 1985; Torbiorn 1982; Tung 1982, 1984], and the difficulties encountered by the expatriate's spouse and/or family [Black 1988; Black and Stephens 1989; Grain and Cooper 1981; Harvey 1985; Tung 1982].While these factors are undoubtedly contributors to the turnover problem, study of these variables in isolation or jointly is likely to result in very little advancement of the understanding of the expatriate turnover problem. The reason for the low probability of significant improvements flowing from this stream of research is that there are CONCEPTUAL MODEL OF EXPATRIATE TURNOVER 501 t t manyothervariableshatinfluence heturn over rocess. pecifically, onceptual p S c a models of the domesticturnover rocesshave pursued more comprehensive p pproachthan that found in the international usiness literature,and these b r i modelshavebeen generally alidated y subsequent esearchn organizational v b behaviorand appliedpsychology. While there have been numerousattemptsto model the domestic turnover process, two turnovermodels have received the strongestsupportbased on the frequency of citations in subsequentliteratureand on the number of attemptsto empiricallyvalidatethe constructsandrelationships. he model T of Mobley, Griffeth,Handand Meglino [1979] suggestedthatcharacteristics of the organization,he individual, ndthe environmenthapean individual's t s perceptionsand satisfactionleading to the formationof intentionsto stay or quit. The Steers and Mowday [1981] model addedadditionalconstructs of job performancelevel, efforts to change the situation, and non-work influences. In addition, he SteersandMowdaymodel exp andedthe affective t responsesto the job to includejob satisfaction,organizational ommitment, c andjob involvement. The satisfaction,commitment,and involvementof an w t individual ould lead to the formation f intentions o stay or quit. Together, o these models have guided or influencedturnoverresearchfor a decade.In thattime, both models have generallystood the test of empiricalvalidation. Thus,it appearsthatconstructs entralto these models may be generalizable c to the internationalnvironment ndhelp explainexpatriateurnover. y using e a t B theoretical odelsfromthe areaof appliedpsychology,this approach ttempts m a to address the criticisms of the internationalliteraturefrequentlyvoiced [BlackandMendenhall 990;Kyi 1988;Newmann, hattandGutteridge 978]. B 1 1 These authors ave notedthatthe internationalusinessliterature ften is not h b o integratedinto a theoreticalframeworkand appearspiecemeal and ad hoc.The purposeof this articleis to presenta model of expatriate urnover ased t b o n constructs entralto the Mobley,et al. [1979] andthe Steersand Mowday c [1981]models. Additionally,he expatriateurnover odelattemptso integrate t t m t concepts from the international usiness literature here possible. By purb w suing a more comprehensive heoreticalapproach nd recognizingthe comt a plex interactiveand longitudinaleffects, an improvedunderstanding f the o expatriateturnoverprocess may result. This improvedunderstanding ay m contributeto the developmentof retentionstrategiesthat would reduce the high turnoverrate.Lee and Mowday [1987] provided several reasons for the value of comprehensive odelsof turnover. irst,comprehensive odels m F m identify and categorizethe variousfactorsthatpreviousresearchhad found to be important. Second, models provide directionfor empiricalresearch, thus imposing discipline on researchers nd reducingthe occurrenceof an a ad hoc researchagenda. Finally, models help managersthink heuristically about a broaderrange of causes of turnoverrather than concentratingon one or two factors. Therefore,the model of expatriateturnovermay be of value to both researchers nd practicingmanagers. 502 JOURNAL OF INTERNATIONALBUSINESS STUDIES, THIRD QUARTER 1992 THE MODEL The basic rationale that underlies this model is that the fundamentalconcepts associated with the turnover process in domestic situations can be adapted to model the expatriate turnover process. Thus, the discussion of major constructs will summarize the existent body of research where possible although there is an absence of empirical studies of a variety of unique international factors. Also, the model of expatriate turnover is implicitly oriented toward middle to upper level managers.These organizational levels appear to be most commonly represented among expatriates [Tung 1981]. While many of the constructs will also apply to operatives, the propositions and discussions adopt a managerial orientation. The discussion of the model (Figure 1) will flow from the predicto rvariables through intermediate linkages to the turnover decision. The sequence of discussion is intended to follow the direction of predominant causality in this turnover model. Three categories of predictor variables have received general empirical support domestically as being important elements of the o mployee turnoverprocess:job/task characteristics, rganizationcharacteristics, three groups of variables collectively and worker characteristics. These influence an expatriate's degree of job satisfaction, commitment to the organization, and involvement in the achievement of the organization's goals. These attitudes are formed with respect to the parent organization, and, since most expatriates are initially transfers from domestic positions, the attitudes have probably been formulated predominantly in a domestic environment.The expatriate's general satisfaction with, commitment to, and involvement in the organization may be moderated by perceptions of the career path resulting fr om the international assignment or by the overall level of cross-culturaladjustment. The expatriate's satisfaction, commitment, and involvement may also be moderated by the employee's family situation and the family's overall satisfaction with the international experience or by o the characteristics f the countryto which the expatriateis assigned.Collectively these factors will lead to the modification of satisfaction, commitment, and involvement with respect to the organization in the foreign assignment. Due to the developmental state of the international business literature and the small percentage of international human resource articles appearing in leading business journals [Black and Mendenhall 1990; Boyacigiller and Adler 1991; Rosenzweig and Singh 1991], the specific international variables influencing expatriateattitudes are largely unknown.However, these attitudes toward the organization in the foreign assignment may result in the expatriate clarifying the intent to change employers, stay with the same employer but transfer â€Å"home,† or stay in the international assignment. The expatriate's intentions may be modified by perceptions of both external and internal employment alternatives. The intentionsmay result in explicit search behavior, ultimately resulting in an initial turnover decision. The turnover decision, even if the choice is to stay, may result in changes in the employee's job CONCEPTUAL MODEL OF EXPATRIATETURNOVER 03 0~~~~~~~~~~~~~~~~~ 0~~~~~~~~~~~~~~~0c U 0~~~~~~~0 0 .. 1 I=- I z0 LL †¦Ã¢â‚¬ ¦ E 0 4) ll l la E L, k . ; ~~~~0 , 504 JOURNAL OF INTERNATIONALBUSINESS STUDIES, THIRD QUARTER 1992 and career expectations and influence subsequent performance. A decision to leave may not be manifested for an extended time period and may result in other withdrawal cognitions. To assist in clarifying the expected relationships in the model, research propositions are presented after each discussion. The degree of detail in each propos ition is largely a reflection of the existent body of research.If research is available that suggests specific relationships and the direction of influence, the propositions are relatively detailed. If research is lacking and the expected relationships are based instead on intuitive reasoning, the propositions are stated very generally. Also, although this is a model of expatriate turnover, only a few of the propositions deal explicitly with turnover. The reason for this is that turnover is the behavioral outcome of a complex process and the propositions address the specific antecedent relationships among variables. Each of these variables will be addressed individually.However, clarifying the relatively ambiguous concept of â€Å"turnover† may be useful first. Turnover Turnover among domestic operations typically refers to the separation of the individual employee from the firm. This definition is too narrow for the international environment and must be broadened to include several other dimensions of turnover since expatriate turnover often involves transfers internal to the organization. Turnover may be categorized as external or internal, voluntary or involuntary, and functional or dysfunctional. External turnover occurs when an individual leaves an organization to seek employment elsewhere.While this type of turnoveris most common domestically, there is no research indicating the frequency of external turnover as a component of total expatriate turnover. By any measure, external turnover is costly and results in reduced short-term efficiency, at the very least. Internal turnover occurs when an individual changes positions but remains in the same firm. This situation is apparently more common in international operations than in domestic situations [Tung 1984]. Essentially this occurs when an expatriate manager is transferred back to the U. S. or another internationallocation.Research has indicated that many expatriate managers find the repatriation process much more stressful and frustrating than the initial expatriationand repatriationexperiences may be a cause of subsequent turnover [Adler 1980; Harvey 1989]. Also, many expatriates may develop an intention to quit while on foreign assignment and view the transfer â€Å"home† as simply an intermediate step to leaving the firm [Harvey 1989]. Turnover may also be voluntary or involuntary (from the employee's viewpoint). Voluntary turnover occurs when the employee quits or requests and receives a transfer.Involuntary turnover occurs when an employee is fired or transferred at the will of the organization. By integratingthese concepts, a four-celledmatrixcan be developed (Figure 2). The turnover models of Mobley, et al. [1979] and Steers and Mowday CONCEPTUAL MODEL OF EXPATRIATETURNOVER 505 r [1981] and much of the subsequentdomesticturnover esearchin organizational behavior and applied psychology concentrateon cell 1, voluntary external turnover,and to a lesser degree, on cell 2, involuntaryexternal f turnover. The international usiness literature ocuses primarilyon cells 3 b i a and4, voluntary ndinvoluntarynternal urnover, ith little or no distinction. t e Very little discussionin the international usiness literature xists of cell 2, b involuntaryexternal turnover,or cell 1, voluntaryexternal turnover. The model presentedhere will apply generally to all four types of turnover. f While the model is conceptuallymost appropriateor both types of voluntary turnover,the model also has implicationsfor both types of involuntary w turnover. he implicationsor the varioustypes of turnover ill be addressed T f in subsequentsections of this article.Turnovermay also be conceptualizedas functionalor dysfunctionalto the firm. Functionalumoveris typicallyviewed as beneficialto the organization. t An example might be when a low-performingexpatriatequits or is fired, t thus creatingan opportunity o staff the position with a more capable indit vidual. Anotherexampleo f functional urnover ightbe the internaltransfer m and/or promotionof a high-performing xpatriate. There appearsto be a e dark side of functional turnoverthat may be easily overlooked. When a t â€Å"low performing†expatriate eaves the organization, he departure ay be l m he organization s a resultof the individual'snability o effectively a t viewedby i o make the cross-cultural djustment r as a result of spouse or family proba lems [Tung 1982]. However, it appearslikely that the cause of â€Å"low performance† may be mistakes made by the organizationin many cases. To illustrate,the selection criteriafor expatriatemanagersmay be inappropriate, expatriation raininginadequate or nonexistent),the job may be poorly t ( designed, the performanceexpectationsunreasonable,or the performance W appraisal rocessinadequate. hilethe expatriate ay be a â€Å"low performer,† p m he low performance ay be due to the organization'smismanagednternam i tional efforts,not the expatr iate'sncompetence. ennings[1985] noted that i J companies are often the cause of their high turnoverrates due to poor managementdevelopment and misuse of talented people. Failure by the t organization o use the skills developedin the international nvironment y e b a the expatriates apparently commonsourceof dissatisfactionmongrecently i a repatriated anagers [Harvey 1989]. If an expatriateis viewed as a low m t t perforner,identifying he causesof low performancesppears o be important. a oViewing the departure f a low perfonneras a good thing may, indeed, be very simplistic. Dysfunctionalturnoverdamagesthe firm in some way by having a valued employee quit or requestan early transfer. Dysfunctionalturnoveris most as commonly onceptualized beinginitiated y theemployee lthoughnvoluntary c b a i internaltransfersare also undoubtedly ysfunctionalin the short tenn. d The functional/dysfunctionalonceptualization ould apply to all four types c c of turnoverin Figure 2. The point to be made h ere is that not all turnover 506 JOURNAL OF INTERNATIONALBUSINESS STUDIES, THIRD QUARTER 1992FIGURE 2 Types of Turnover Voluntary (For the Employee) External (To the Organization) Involuntary (For the Employee) I Quit! You're Fired! 12 34 Internal (To the Organization) Transfer Me! You're Transferred! is bad; some turnover may actually be beneficial and afford the organization an opportunityto improve. However, since a significant portion of expatriate managers are initially â€Å"internal transfers† from a domestic assignment rather than â€Å"new hires,† the expatriates are likely to be high performers in the domestic environmentdue to the expatriateselection process.Specifically, a manager's â€Å"domestic track record† is often a major factor in the expatriate selection process [Mendenhall and Oddou 1987; Tung 1981]. Thus, it may be hypothesizedthat dysfunctionalturnoveris more common among expatriates than among domestic managers (i. e. , losing valued empl oyees). A further complication to the traditional approach to turnover research also t appearsnecessary regardingthe appropriate ime horizon. Turnovermay occur significantly after repatriation but be caused by the international experience or the repatriationprocess.An expatriatemanager may become disenchanted while on the international assignment and actually formulate an intention to quit [Adler 1986]. However, the expatriate may simply â€Å"gut it out† until b transferred ack to the U. S. , and a new job is found. Alternatively, a expatriate may become disenchanted with the organization due to events occurring during the repatriation process and subsequently leave the organization. However, turnover after repatriation introduces many issues not addressed in the expatriate turnover model presented here.Therefore, the discussion in this article is limited to turnover that occurs while an expatriate is actually on the foreign assignment. The model does adopt a decidedly tempo ral dimension, however. Steers and Mowday [1981] contend that the intention to quit is usually manifested behaviorally within a year and that the relationship between intention and turnover subsequently weakens. The reason that an extended time horizon is necessary in turnover studies is the existence of gradual changes in worker attitudesleading to behavioralchanges, a phenomenareferredto as â€Å"progression theory† [Rusbult, Farell, Rogers and Mainous 1988].Specifically, declining CONCEPTUAL MODEL OF EXPATRIATETURNOVER 507 workerattitudesdo not instantlyresultin the formationof intentionsto quit. For example,an expatriate ay experiencelow levels of attitudes or several f m monthsbeforethinking boutquitting ndthenultimatelyormingthe intention a a f to quit. Likewise, an employee may have low levels of attitudes,but the t situationmay improve,resultingin an intention o stay with the organization. Since the expatriateturnover ate appearsto be roughlytwice the domestic r ate, a variety of unique internationalfactors apparentlycontributeto a gradualdeterioration f worker attitudesover a period of time for many o expatriates. anyexpatriates ay have positive attitudesinitiallybut graduM m ally develop more negative attitudes. Thus, a longitudinalperspective is t important o the model. t Regardlessof the type or timing of turnover, he ultimateturnoverdecision is precededby antecedent ariablesandintermediateinkages. The predictor v l variablesof job/task,organization, nd workercharacteristics re the initial a a startingpoint in many studies of employee turnover.Job/Task Characteristics Of the threebroadcategoriesof predictorvariables,job/taskcharacteristics have received the most researchattentiondomestically[Glisson and Durick 1988] while receiving very little attention internationally. eceiving the R strongest support as predictorsof employee attitudes are role ambiguity [Teely, French and Scott 1971; Lyons 1971; Abdel-Halim 1981; Bedeian and Armenakis 1981 ] and skill variety [Bartel 1982; Marsh and Manari 1977; Price and Mueller 1981; Dewar and Werbel 1979; Gerhart 1987; Glisson and Durick 1988; Blau and Boal 1989].Thus, the greaterthe role clarity and the more diverse the skills needed, the more likely the worker is to be satisfied,committed,and involved. Additionally,the characteristics of role conflict, task identity, and task significance have received support as predictor ariables. It appears,therefore, hatissues relatedto the specific v t task environmentcan influence the satisfaction,commitment,and involvement of workersat a variety of organizationalevels. l The uniquenessand dynamismof international nvironments ay lead both e m to more role ambiguity and task variety among expatriates.The volatile externalenvironment ay cause MNCs to have less clarityin theirposition m descriptionsas managersrequiremore flexibility to respond to changes. m Therefore, orerole ambiguity ay exist whichmay lead to reducedsatisfacm tion amongexpat riate anagers. owever,overseasassignments ftenrequire m H o managersto use a broaderrange of general managementskills than their domestic counterpartsCzinkota,Rivoli and Ronkainen1989; Edstromand [ Galbraith1977]. This implies that increasedtask varietywill contributeto a higher level of satisfaction among expatriatemanagers.Thus, job/task characteristics ay offset one anotherin theircontributiono an expatriate's m t workattitudes. Withone exception,these constructs enerallyhave not been g investigatedin the international nvironment. e 508 JOURNAL OF INTERNATIONALBUSINESS STUDIES, THIRD QUARTER 1992 While most of the studies cited above utilized domestic U. S. subjects, Black [1988] found that role ambiguity, conflict, overload, and discretion were related to the cross-cultural adjustment of American expatriates in Japan. The direction of Black's findings was generally consistent with the domestic literature.A potentially importantconsideration in the international situation is the similarity of the domestic and foreign jobs performed by the expatriate [Dowling and Schuler 1990]. If the foreign job requirementsare quite similar to the domestic position previously held, the transition and work adjustment should be easier. If the domestic and foreign positions are highly divergent, the expatriate must adjust to both a new work environment as well as a new cultural and physical environment. Another job/task characteristic that may be of importance internationally is the concept of work grouphomogeneity. Althoughusing a domestic U.S. sample frame, O'Reilly, Caldwell, and Barnett [1989] found that aspects of work group homogeneity were related to the degree of social integration which was negatively related to turnover. Since foreign work groups in U. S. MNCs are often composed of U. S. expatriates, host country nationals, and third country nationals, a good deal of work group heterogeneity is likely to exist. This diversity is likely to inhibit the individual soci alization process and result in lower levels of cohesiveness. The lower level of social integration may contributeto reduced attitudesthat would ultimately lead to increasedturnover.Work group homogeneity may also be conceptualized on a broader, more macro level. In addition to the fit between the expatriate and other job associates, the degree of homogeneity between the foreign subsidiary and parent corporation may be important [Brittain and Freeman 1980; Lincoln, Olson and Hanada 1978]. The more similar the composition and functioning of the foreign work group to the parent, the more positive the expatriate's attitudes are likely to be. Milliman, Von Glinow and Nathan [1991] contend that MNCs attempt to apply isomorphism to foreign subsidiaries as a mechanism of control and continuity.Based on the literature that addressed job/task characteristics, both domestically and internationally, the following propositions were developed. Proposition 1 There is a positive relationshipbetwee n skill variety, task identity, autonomy, and task significance and expatriate satisfaction, commitment, and involvement. b Proposition 2 Thereis a negativerelationship etweenrole ambiguity and role conflict and expatriate satisfaction, commitment, and involvement. Proposition 3 The degree of similarity between the expatriate's domestic and foreign jobs is positively related to xpatriate atisfaction,commitment,and involvement. s CONCEPTUAL MODEL OF EXPATRIATETURNOVER 509 Proposition4 There is a positive relationship between work group homogeneity and expatriate satisfaction, commitment,and involvement. OrganizationCharacteristics Behavioral, structural,and demographiccharacteristicsof organizations have received researchattentionas predictorsof satisfaction,commitment, and involvement. Behavioralissues such as participationn decisionmaking i have received strong support [Gladstein 1984; Glisson and Durick 1988; Millerand Monge 1986].Leadership onsideration as also receivedsupport c h [Morrisand Sherman1981;Batemanand Strasser1984; Glisson andDurick 1988; Blau and Boal 1989]. Dimensions of organizationstructuresuch as centralization Child 1972; Hage and Aiken 1969], differentiation Redding [ [ and Hicks 1983; Bimbaum and Wong 1985], and formalization[Redding andHicks 1983;Child1972;HageandAiken 1969]havealso beenempirically validatedas predictors f attitudes. emographic rganizationalactorssuch o D o f as organization ge, workgroupsize, andtype of industry ave receivedweak a h upportas predictive variables. Cotton and Tuttle's [1986] meta analytic review providesa more comprehensive eview of these variables,and their r findings indicated U. S. studies differed from non-U. S. studies in several ways. In U. S. studies, an employee's genderhad less impacton satisfaction and unions had more influence on turnover. Satisfactionwas less reliably tled to turnoverin non-U. S. firms. In the internationalnvironment varietyof othervariables ppear otentially e a a p A import ant. s Dowling and Schuler[1990] noted, a firm's structure, ontrol c echanisms, and human resource policies are linked to the evolutionary P processof globalization. ucik[1985] also notedthatthe evolutionary rocess p is continuousas organizationscontinuallyadapt to dynamic international environments. Although there are several organizationchange models, the concept of organizationlife cycles (OLC)has been applied internationally Adler and [ Ghadar1989; Milliman, Von Glinow and Nathan 1991]. As organizations progressfrom one stage to another fouror five stages are typicallyincluded) ( an MNC's strategy, structure,and human resource policies also need to change.Milliman, Von Glinow and Nathan [1991] noted that change is needed so that MNCs can optimizetheir â€Å"fit. † There is a need to optimize the externalfit of the foreign subsidiaryto the local environmentand the parent ompanyto the domesticenvironment. hereis also a need to optimize c T the internalfit between the parent and foreign subsidiary. Complicatingthe optimizationof both externaland internalfit is the need for flexibility due to environmental ifferencesand changes. For example,Black, Mendenhall d and Oddou [1991] noted that flexibility in the work environmentis related to aspects of cross-cultural djustment. illustratethe importance f these T a o concepts, when an organization becomes more experienced internationally, 510 JOURNAL OF INTERNATIONALBUSINESS STUDIES, THIRD QUARTER 1992 the amount of role conflict and role ambiguity may be reduced as the a s organization trategically daptsmore effectivelyto the foreignenvironment. ( c m Also, the adaptation ay lead to a differentstructural onfiguration foreign sales agent versus foreign sales office versus vertically integratedforeign operation)that could enhance satisfaction,commitment,and involvement though improvedjob design. t uThe open systems approach sed by manyorganizationalheoryresearchers [Lawrenceand Lorsch 1967; Pfeffer and Salancik 1978; Thompson 1967] d clearly implies that the natureof the environment ictates the appropriate e w S structure. ince most researchers ould agreethatinternationalnvironments a t relative o the U. S. , thenmoreflexible,decentralized,utonomous aredynamic, T foreign structureswould be appropriate. his view of more decentralized foreignsubsidiariesis also consistentwith the trendtowardflatterstructures in the U. S. and with the trend of workerempowerment. Specifically, as a rganizations ttemptto become more innovativeand responsive,decisionl making authorityis often forced to lower organization evels. While these variablesare intuitivelylogical predictorsof satisfaction,commitment,and e involvementand may be relatedto a firm's multinational volution in the been very little empiricalinvestigationor international ontext, there has c validationof these constructs. h training ave receiveda good deal of attention a Expatriationnd cross-cultural D business literature. espite the rapidincreas ein world in the international o e tradeflows in the pasttwentyyearsandtherelated mergence f multinational f e corporations, xpatriation rainingfor managersin U. S. multinationalirms is still in its infancy. Specifically, in multinationalcorporations,training a programs re generallysuperficialand incompleteor nonexistent. In Tung's [1981] study, only 32% of respondingfirms had formalizedtrainingprot t grams. Furthermore,he focus of the formaltrainingprograms endedto be t c on environmental oncerns. Country-level rainingin cultureand language was uncommon. DunbarandEhrlich[1986] and Schwind[1985] also found t that the majorityof firms had no formalexpatriation rainingfor managers. i thatthe stateof expatriation rainingn U. S. multinational Tung [1981]suggests corporationsis a major reason for expatriateturnoverrates in U. S. firms a being two to threetimes the level of those in European nd Japanesefirms. That is, the resultof underdeveloped xpatriationrainingprograms ay be m t e that expatriate anagers xperiencevery high levels of initialrole amtbiguity m e f P s and conflict which in turnreduces workattitudes. roviding upport or this t position is Black and Mendenhall's[1990] review of cross-culturalraining i research. They noted that cross-culturalraining s positivelyrelatedto skill t a development, djustment,ndperformance. t c An additionalfactorthatmay be an important ontributoro turnoveris the c t conceptof corporate ulture. Organizationsypicallydevelopa predominant, identifiableculturealthoughit is rarelystatedexplicitly [Wilkinsand Ouchi 1983]. Also, an organization'sculture can vary across functional areas, CONCEPTUAL MODEL OF EXPATRIATETURNOVER 511 operatingdivisions, or geographicallocations [Gregory1983; Wilkins and Ouchi 1983]. Since most expatriatemanagersare internaltransfersrather than new hires, the assumptionprobably can be safely made that most i xpatriates ave been socially integrated nto the parentfirm's predominant h culture. If the work gro up at the foreign location has developed a unique and differentcorporateculture,the expatriatemay experience adjustment c difficultiesdue to the divergent otporate ulture. pecifically,the expatriate's c S w valuesandorganizationaleliefs may be incongruent ith the local operation, b hence decreasedwork attitudesmay result. While this issue is conceptually similar to work group homogeneity, subunit corporateculture is a more p b training rograms. pervasiveconceptandcouldbe influenced y cross-cultural bProposition5 There is a significantrelationship etween dimena sions of a foreign subsidiary'sstructure nd expatriatesatisfaction,commitment,and involvement. t The moredecentralizedndautonomoushe foreign a subsidiary,the more positive the work attitudes. b Proposition 6 Thereis a positiverelationship etweenthe quality of expatriationrainingand expatriate atisfaction, t s commitment,and involvement. b Proposition 7 Thereis a positive relationship etween an organization'sstageof multination al evolution, articipation p in decisionmaking,leadershipconsideration,and s a expatriateatisfaction,ommitment,ndinvolvement. Proposition8 There is a positive relationshipbetween the similarity of a firn's predominantcorporateculture and local foreign corporatecultureand expatriate satisfaction,commitment,and involvement. Worker Characteristics o Empiricalinvestigationdomesticallyof characteristics f the workerhave o a c yielded mixed resultsas predictors f satisfaction, ommitment, nd involvement. Personality[Staw and Ross 1985; Staw, Bell and Clausen 1986], age [Dewar and Werbel 1979], being female [McNeely 1984], and job tenure [Coverdaleand Terborg1980] have received weak supportas predictorsof satisfaction.However, personality [Hulin and Blood 1968; Steers and Spencer 1977], marital status [Porterand Steers 1973], age [Morris and Sherman1981], tenurein the organization Stevens,Beyer and Trice 1978], [ and education[Steers 1977] have received strong supportas predictorsof commitm ent. Age, tenure, and maritalstatus (being married)are usually y w positivelyassociatedwith commitment hile yearsof education, earsin the samejob, and alternative mployment pportunitiesre negativelyassociated o a e with commitment. Althoughmicro level contingencytheory models would contend that individual characteristicswould be importantpredictorsof 12 JOURNAL OF INTERNATIONALBUSINESS STUDIES, THIRD QUARTER 1992 attitudes in different work contexts, there is apparently little research that provides consistent direction in this area [Glisson and Durick 1988]. The level of performance of the employee is also a possible predictor of satisfaction, commitment, and involvement. Research has generally supported the contention that performance leads to satisfaction [Ivancevich 1978; Lawler and Porter 1967; Sheridanand Slocum 1975]. Therefore,an individual's performance level is generally thought to be positively associated with satisfaction, commitment, and involvement.In the international context, a variety of worker characteristics have been discussed or empirically investigated. Worker characteristics postulated as important to expatriates are technical competence [Hawes and Kealey 1981; Tung 1982], relational abilities [Tung 1982; Mendenhall and Oddou 1986], and stress reduction [Abe and Wiseman 1983; Hammer, Gundykunst and Wiseman 1978; Barrettand Bass 1976; Ratiu 1983]. Other possibly important workercharacteristics re personality[Torbiom 1982; Dapsin 1985; Mendenhall a and Oddou 1986] and tolerance for ambiguity [Hammer, Gundykunst and Wiseman 1978; Ratiu 1983].Another worker characteristicthat may be positively associated with intemational success is the individual's years of previous international experience [Black and Stephens 1989; Church 1982]. An individual who was unsuccessful in a previous international ssignmentwould likely limit subsequentinternational a assignments. Conversely, an individual who has experienced success internationally may be more re ceptive to subsequent international assignments. Thus, simple trial and errormay result in an expatriate's years of international experience being positively related to satisfaction, commitment, and involvement.Somewhat related to internationalexperience, particularlyin a specific country, is an expatriate's linguistic ability. The more fluent the expatriate in the language of the host country, the easier the social integration to both the work and general environment. Thus, linguistic ability should be positively related to work attitudes. Two other rather controversial types of worker characteristics exist, an expatriate's sex and marital status. The majority of expatriates are male, 97% in Adler's [1984] study, and most are married [Black 1988].Since most expatriatesare male, does the high turnoverrate caused by an expatriate's gender, coincidentally vary with it, or is it totally unrelated to it? Unfortunately, this topic has not been empirically investigated in an international environment. Thai and Cateora [1979] noted that being a woman expatriate can be a significant liability due to the culturalbias in some foreign countries (i. e. , in the Middle East, Latin America, and Japan). However, Adler [1984, 1986] and Adler and Izraeli [1988] contend that cultural bias against women in management often does not apply to expatriates.Cultural views of women are often restricted primarily to women of that country. A woman expatriateis usually viewed predominantlyas a foreigner,who also incidentally happens to be a woman. In some cases, being a woman was enough of a CONCEPTUAL MODEL OF EXPATRIATETURNOVER 513 novelty to be viewed as a positive asset. As Adler and lzraeli [1988] noted, the overwhelming conclusion about women expatriatesis their scarcity. While there are more U. S. women managersand expatriates(as a percentage of the total workforce) thanin most countries,the proportion f women o anagersin almost all countriesis low due to the same general reasons: culturalsanctions, educationalbarriers,legal restrictions,corporateobstacles, and women's disinterestin pursuingmanagerialcareers. Due largely to the scarcity of women expatriates,there is no data that suggests that women would have more, or less, positive work attitudesthan men. Maritalstatus has also received researchattention,but with mixed results. i Some authors ontendthatbeing marrieds a stabilizing actorfor expatriates, f c while otherscontendthatspouse and family problemsare the leading cause of expatriate ailure[Tung 1984].While some MNCs develop cross-cultural f f i trainingprograms or an expatriate's pouse, the assumption s usuallymade s thatthe expatriate s male andthe spouse is female [Adlerand Izraeli 1988]. i t o Unfortunately, he literature f the impact of maritalstatus is scarce so no directionof relationshipcan be inferred. SteersandMowday[1981]contended hatworker xpectations ay be related t e m to subsequentattitudes. Their discussion focused on â€Å"met expect ations,† or the extent to which pre-employmentexpectations were subsequently fulfilled by job expenrences. n arguingfor realismin pre-employmentnterI i iews, Steersand Mowdayfelt that â€Å"met expectations† nd workerattitudes a were positively related. The same concept may apply internationally. f a I manager's pre-international xpectations differ greatly from subsequent e experiences,an expatriate'sattitudeswould likely decline. While expatriation trainingmay provide an opportunity or a realisticjob preview, other f techniquessuch as preview trips, may also help clarify expectations. Proposition9 Thereis a positive relationship etween an expab triate's tenure, organizationlevel, performance, yearsof previousinternationalxperience, inguise l tic ability,age, andrelational bilityand expatriate satisfaction,commitment,and involvement. Proposition10 Thereis a positiverelationship etweenthe degree b to whichan expatriates9sre-intemational p expectaaremetin subsequentnterna tional tions i experiences and satisfaction,commitment,and involvement. a T t C Satisfaction, ommitmen4nd Involvement owardhe ParentOrganization a i l Satisfaction,commitment, nd involvementare importantntermediateinkages between predictorvariablesand employee tumoverdecisions. Because of theirimportance, hese constructs ave receivedextensiveconceptualand t h empiricalattention. Models of employee turnover ave explicitly addressed h heirrole [Blau and Boal 1987; Bluedom 1982; Mobley, et al. 1979; Steers 514 JOURNAL OF INTERNATIONALBUSINESS STUDIES, THIRD QUARTER 1992 and Mowday 1981; Price 1977]. Additionally, over one hundred empirical studies have validated these constructs and have indicated the strength and direction of intermediate linkages (see Miller and Monge [1986] and Cotton and Tuttle [1986] for meta analytic reviews). Therefore, only a brief definition and discussion of these variables will be presented here. Job satisfaction has been defined as â€Å"the positive emoti onal state resulting from the appraisalof one's job or job experiences† [Locke 1976].Organization commitment is defined as the worker's â€Å"strong belief in the organization's goals and values, a willingness to exert considerable effort on behalf of the organization, and a strong desire to remain a member of the organization† [Mowday, Porter and Steers 1982]. Job involvement is defined as â€Å"the extent to which an individual identifies psychologically with her/her job† [Blau 1985]. Although these constructs are related, each is distinct due to measurement of attitudes about the individual (satisfaction), the job (involvement), or the organization (commitment) [Blau and Boal 1987; Morrow 1983].For example, employees with high job satisfaction may feel positive about their jobs due to fulfillment of personal needs and values [Miller and Monge 1986]. For individuals with high levels of job involvement, the job is important to the worker's self-image [Kanungo 19 82]. Workers with a high level of organizational commitment feel positive about their employer, identify with the organization,and wish to maintain membership in it [Porter, Crampton and Smith 1976]. While these constructs are conceptually distinct, interactive effects have been shown to exist between commitment and involvement [Blau and Boal 1989], nd satisfaction and commitment [Glisson and Durick 1988] and satisfaction, commitment, and involvement [Lee and Mowday 1987]. However, satisfaction, commitment, and involvement generally complement one another as intermediate linkages between predictor variables and turnover [Blau and Boal 1987]. Attempts have been made to identify a causal orderingbetween these attitudinal variables. The implicit assumptionof these attemptsis thatsince satisfactioncan be formulated quickly, satisfaction leads to commitment and involvement which are more long lasting and enduring [Steers 1977; Stevens, Beyer and Trice 1978].Attempting to test this conten tion, Williams and Hazer [1986] found strong interaction between satisfaction and commitment but could not infer causality due to the cross-sectional nature of the data. However, Farkas and Tetrick [1989] used a longitudinal design but were unable to identify causal direction. They did identify that satisfaction and commitment are differentially related over time, and the relationship may be cyclical or reciprocal. Therefore, the model presented here assumes that these variables are strongly and positively related to one another, but does not assume any causal relationship.Since the majority of expatriates are internal transfers ratherthan new hires, the workers have probably achieved some degree of social integration into CONCEPTUAL MODEL OF EXPATRIATETURNOVER 515 the organization. Accordingly,the argumentis made that expatriatesstart theirinternational ssignmentwith attitudesthatare reasonablywell formua lated. Also, as transfers,expatriates ay have reasonablypositive attitudes m initially. Althoughthereis no empiricalsupportfor this contention,it seems intuitivelylogical that a disgruntled mployee would be an unlikely candie t date for an internationalransfer.The situationfor new hires is more complex. While new hires apparently constitutea small portionof expatriates,new hires' attitudesare probably more formative than internaltransfers. Since new hires would have little i b t opportunityo achievesocial integrationntothe organization eforedeparting for the foreignassignment, he local foreignorganization ould be relatively t w i more importantn shapingworkattitudes. Also, pursuingthe logic of Steers and Mowday [1981], pre-employmentexpectations would be relatively more importantin shaping new hires' work attitudessince expectations w a regarding oththe organization ndthe foreignassignment ouldbe clarified. Whetherthe expatriateis an internaltransferor a new hire, the expatriate initially arrives in the foreign assignmentwith some level of satisfaction, b co mmitment nd involvement. dditionally, ased on the domesticresearch, a A these three attitudesare positively relatedto one another. Proposition 11 Thereis a positive elationshipmongan expatriate's r a satisfaction,commitment,and involvementwith respectto the organization. Attitudestowardthe Organizationin the Foreign AssignmentThe traditionalturnovermodels of Mobley, et al. [1979] and Steers and Mowday [1981] postulatethat satisfaction,commitment,and involvement collectively result in an employee formulatingintentions to stay in the organization r to quit. In the international ontext,an expatriate anager's o c m a family situation,the careerpathingperceptions,countrycharacteristics, nd the degree of cross-cultural djustment ppearto be moderatingvariables. a a Thus, the employees' attitudetowardthe parentorganization ay be modim fied to resultin attitudestowardthe organization n the foreign assignment. These attitudeswould still conceptuallyinclude satisfaction,commitment, and involve ment,but the constructswould be more directlyinfluencedby the international xperience. e Supportfor the changes in an expatriate'sattitudestowardthe organization in the international ontext is providedby Gregersonand Black [1990a]. c Their researchindicatedthat expatriatesare often differentiallycommitted to the organization nd local foreignoperations. urthermore,on-jobfactors a F n were significantly related to local commitment. These findings appearto f t and c rovide upportorthecontentionhatcountry haractenrstics an expatriate's s family situationwill affect the expatriate'ssatisfaction,commitment,and involvementwith respectto the organizationn the international ssignment. i a 516 JOURNAL OF INTERNATIONALBUSINESS STUDIES, THIRD QUARTER 1992 Unfortunately,the concept of dual commitments, or dual attitudes in general, has received very little research attention internationally. Therefore, it is unknown whether attitudes toward the parent organization or attitudes toward the local f oreign organization would dominate in the formation of behavioral intentions.However, the model explicitly assumes that the expatriate's initial attitudes do not remain constant. As suggested by the high expatriate turnover rate, the expatriate's attitudes generally deteriorate due to a variety of unique internationalfactors. Some of the factors were discussed previously within the three categories of predictor variables. The remaining unique internationalfactors of family situation, country characteristics,career pathing, and cross-cultural adjustment appear quite important and will be addressed individually in the following sections.The assumption is made, however, that there is a â€Å"spill-over effect† between an expatriate's attitudes toward the parent organization (probably formulated domestically) and attitudes toward the international assignment. Expatriates with very positive attitudes toward the organization in general are also likely to have more positive attitude s internationally. Expatriates who possess lower levels of attitudes toward the organization in general are likely also to have lower levels of affective responses internationally.This explicitly assumes that each of the three attitudes can be formulated at a more macro level toward the organization in general or at a micro level that is more closely related to the foreign assignment. While the dual levels of commitment and satisfaction have been well identified in the domestic literature, it seems intuitively logical that the same could be said for job involvement. For some expatriates, attitudes toward the organization may remain positive while negative international attitudes may develop. Such a situation may result in an internal transfer.For other expatriates, initially positive attitudes toward the organization may deteriorate internationally leading to external turnover. Proposition 12 There is a positive relationship between job satisfaction,commitment,and involvement (with respect to the organization) and expatriate satisfaction, commitment, and involvement (with respect to the international assignment). Expatriate's Family While an employee's family situation may not be central to the domestic turnover process, the family situation is apparently a critically important oderatingvariable in the expatriateturnoverprocess. Harvey [1985] contends that the family situation may be the most important contributorto expatriate turnover. Supportfor Harvey's contentionis providedby Tung's [1982] research that indicated that two of three most frequently cited causes for expatriate failure were family related. Specifically, â€Å"the inability of the expatriate's family/spouse to adjust to a different physical or cultural environment† and CONCEPTUAL MODEL OF EXPATRIATETURNOVER 517 â€Å"otherfamilyrelatedproblems†receivedstrongsupport. espitethe apparent D mportanceof the family, less than half of MNCs interviewthe spouse in the expatriate election p rocess [Black and Stephens1989; Tung 1981] and s expatriation rainingfor the family is very rare [Black and Stephens 1989]. t If a spouse or family memberis undergoingcultureshock or experiencing t difficulty in makingthe cross-cultural djustment, he morale,performance a and work attitudesof the expatriatemanagermay be adversely affected [Harvey 1985; Mendenhall,Dunbarand Oddou 1987; Tung 1982]. Since the majority of expatriatemanagers are male [Adler 1984], the greatest impact of a foreign relocationmay be experiencedby the wife [Gaylord 1979].The wife may experiencehigh levels of stress due to a disruptionof i children'seducation,loss of self-worthand identity,particularlyf she was previouslyemployed, lack of contactwith friends and relatives, and social or culturalostracism in the foreign country [Harvey 1985]. Furthermore, children are often resistantto moving, even domestically. The problems faced by childrenregarding ducation,linguisticdifferences,social relatione ships, and c ulturalvalues also can be the sourceof stress and conflict. Thus, the greaterthe numberof children,the more likely adjustment roblemsare p o occur. Also, older children such as teenagers are more likely to have t developedstrongpeer social relationships nddisrupting hose relationships a for an internationalransfer ay have negativeresults. Despite the apparent t m importanceof these variables,there is apparentlyittle empiricalinvestigal on expatriateturnover. tion of the impactof the family situation However,as noted by Black and Stephens[1989], the family situationis an e t intuitivelyimportant lementof the expatriate urnover rocess. They noted p t hatthereappearso be botha positiveandnegative†spillovereffect†between expatriatesand their spouses' perceptionsof an international ssignment. a w Specifically, expatriateand spouse adjustment ere significantlyand positively correlated, nd were relatedto the expatriate'sintentionto stay in the a foreign assignment. Even if an e xpatriatemanageris very positive about the foreign assignment, a transfer â€Å"home† may be the result of family dissatisfaction. Conversely,a positive family situationis likely to enhance an expatriate's attitudes and cross-culturaladjustment,and increase the probabilityof a successful foreign experience.Theredoes appearto be a potentialproblemin evaluatingthe impactof the spouse or family situationon turnover. Attributingexpatriateturnoverto the failure of the spouse to adjustmay be simplistic [Dowling and Welch 1988]. Expatriatesmay use their spouses as scapegoats,ratherthan admit theirown failuresto adjust. This may reducethe negativecareerimplications of an early transferhome. Likewise, top level corporateexecutives may attributeexpatriatefailure to the expatriate'sfamily ratherthan critically evaluatetheir own firm's expatriation rogramswhich may be deficient. p There are many dimensions of an expatriate'sfamily situation that may w oderate orkattitudes. nfortunately, ost of these variables ave received h m U little or no researchattention. Therefore,the following propositionhas been 518 JOURNAL OF INTERNATIONALBUSINESS STUDIES, THIRD QUARTER 1992 stated very generally as a global measure of â€Å"family satisfaction,† which is undoubtedly a multifaceted concept. Proposition 13 The expatriate's family satisfaction with the international experience has a positive moderating influence on the expatriate's job satisfaction, commitment, and involvement with respect to the organization in the foreign assignment. The more ositive the expatriate's family attitudes, the more likely the expatriate is to develop positive work attitudes internationally. Proposition 14 The fewer the number of children that relocate with the expatriate, the more positive the expatriate's work attitudes. Proposition 15 The younger the children that relocate with the expatriate, the more positive the expatriate's work attitudes. CountryCharacteristics All foreign assignme nts are not created equal. Dramatic differences exist both between and within countries that can influence expatriates', and their families', cross-cultural adjustment and their subsequent attitudes.In many cultures, foreigners are viewed with distrust and skepticism. As a result, building social relationships with local nationals may be impossible. For example, few western women find the Arabic countries enjoyable due to the constrained role of women in those societies [Thai and Cateora 1979; Dowling and Schuler 1990]. As a result, some companies may prefer to select managers who are unmarried for foreign assignments in the belief that a single person will have fewer adjustments.However, many single male expatriates are socially ostracized, and they may have a more difficult social adjustment than a married couple who can provide each other mutual support. In addition to social and cultural factors, expatriates can experience dissatisfaction due to living conditions and health prob lems. A foreign oil company executive with extensive international experience expressed his preference for having Americans on international projects because they were always the first to get sick if a health problem existed. The Americans were his â€Å"miner's canary† that served as an early warning of potential health problems.Issues such as these have been referred to as â€Å"cultural toughness† and â€Å"cultural novelty† [Black and Stephens 1989; Mendenhall and Oddou 1985; Torbiorn 1982]. Cultural toughness describes the difficulty that a western expatriate would have adapting to certain cultures. For example, Torbiorn [1982] noted that western expatriates experienced higher levels of dissatisfaction with assignments in India, the Middle East, North Africa, CONCEPTUAL MODEL OF EXPATRIATE TURNOVER 519 East Africa, and Liberia. Thus, some cultures are very different from western cultures and are also very difficult to adjust to.Other countries may have a ve ry different culture, but are much easier to adapt to for expatriates and their families. The concept of culturaldifferences influencing the cross-cultural adjustment process is intuitively logical. However, there appears to be a lack of consensus regarding conceptualizations of this factor. For example, the terms â€Å"cultural toughness,† and â€Å"cultural novelty,† appearin the international business literature. Unfortunately, tightly developed definitions and consistent research operationalizations are generally lacking. Hofstede [1980] perationalized â€Å"cultural distance† along four indices of work-related values, but most researchershave used a broader,more macro level approach. Harbison and Myers [1959] conceptualized cultural differences in stages of economic development and the role of the firm in that process. Farmer and Richman [1980] conceptualizeddifferences along socio-cultural,legal-political, economic, and educational dimensions. Nath [1988] described behavioral differences that included cultural factors such as attitudes, beliefs, value systems, behavioral patterns, and management philosophies.Negandhi and Prasad [1971] evaluated differences in the task environment such as distributors, suppliers, employees, consumers, government, and community. Consensus regarding the appropriatedimensions to use in measuring cultural differences is lacking, but there is apparently strong support for the general concepts of cultural toughness, cultural novelty, and/or cultural distance. Each country presents the expatriate, and the expatriate's family, with a unique set of adjustmentproblems. The more divergent a foreign assignment and location from the home country environment, the greater the potential of experiencing culture shock.Indirectly supporting this contention was Beamish's [1985] finding that MNCs are more likely to be dissatisfied with the performance of operating units located in less developed countries that are likely to be â€Å"culturally tough. † The more â€Å"culturally tough† a foreign country, the more likely the expatriate's work attitudes will be negatively affected. Therefore, the following proposition was developed. Proposition 16 The culturaland environmentalsimilarity of home and host countries has a positive, moderating influence on expatriate satisfaction, commitment, and involvement (with respect to the international assignment).The more similar the foreign culture and environmentto the home environment, the more positive the impact on expatriate satisfaction, commitment, and involvement with respect to the organization in the foreign assignment. Career Pathing Research on the career value of foreign assignments in MNCs is inconsistent. The inconsistency may result from the fact that the human resources function constitutes the weakest link in the overall strategic planning process in 520 JOURNAL OF INTERNATIONALBUSINESS STUDIES, THIRD QUARTER 1992 most U. S. MNCs [L orange and Murphy 1983; Tung 1984].Thus, many foreign assignments appear haphazard rather than part of a planned process to develop certain managerial skills. As a result, international assignments appear to be avoided by some high performing managers due to potential negative career consequences [Adler 1980]. For example, Edstrom and Galbraith [1977] suggested that expatriates often viewed the purpose of their overseas assignment as development for future executive responsibility. However, Gonzalez and Negandhi [1966] found that about half of former expatriates were in low level positions fifteen years later.Howard [1973] and Harvey [1989] found that a good deal of uncertainty and conflict surrounded the expatriation and repatriation process and skills developed by expatriates were seldom used after their return home. These results led Mendenhall, Dunbar and Oddou [1987] to note that the â€Å"overseas assignment is a haphazard, ill-planned affair that is usually accompanied with vertical advancement. † They concluded that the impetus for overseas staffing seems to be more to meet immediate manpower needs than to create an integrated career development strategy for future corporate executives.This lack of career path clarity may cause expatriate managers to reevaluate their own career goals and the congruity of those goals with the organization. Since career counseling is very rare regarding international transfers, many expatriates are probably unaware of some career issues until they are actually on foreign assignment.