for magz64 help, from Tina Li

Converging on IHRM Best Practices • 123 CONVERGING ON IHRM BEST PRACTICES:

LESSONS LEARNED FROM A GLOBALLY DISTRIBUTED CONSORTIUM ON THEORY AND PRACTICE Human Resource Management, Spring 2002, Vol. 41, No. 1, Pp. 123–140 © 2002 Wiley Periodicals, Inc. This article published in Wiley InterScience (www.interscience.wiley.com).

DOI: 10.1002/hrm.10023 Mary Ann Von Glinow, Ellen A. Drost, and Mary B. Teagarden This article captures the results of the “Best Practices” Project and attempts to identify which IHRM practices are universally embraced in our ten-country/region sample. Here we present a compelling argument for understanding cultural contexts by seeking and establishing derived etics. No longer content with the traditional conduct of cross-cultural research, we challenge “how” we do research, and encourage gatekeepers to broaden their research lenses with multiple embedded contexts—polycontextuality—as they search for answers. We find anomalies, and counterintuitive findings, and through our “gap analysis,” we discovered several universally em- braced etics or best practices. We believe we have not only made a significant contribution to research, but, in particular, we offer a solution methodology for conducting globally distributed IHRM research. These findings signal new directions for all deeply involved in managing within and across different cultures. © 2002 Wiley Periodicals, Inc. Introduction This special issue has addressed a complicated and somewhat maligned term of late—best practices—in International Human Resource Management. In brief, the challenge of the Best Practices Project reported on in this spe- cial issue was to identify which HR practices stood out across which countries and why.

Were there universal “best practices,” or purely context-specific practices that were meaning- less out of cultural context? Such analyses de facto demand consideration of a number of contextual, cultural, and organizational vari- ables. The research quandaries associated within the massive data sets reported here were exacerbated by philosophical challenges which lie at the root of what researchers are trained not to go beyond: the scientific method. Despite it all, the Best Practices Project (the multiresearcher, multicultural, multinational, interdisciplinary research con-sortium described in the previous articles) has succeeded in shedding light on theoretical and practical arguments for how IHRM practices and the effectiveness of their use, can and do vary in response to cultural and other contex- tual factors. In short, this project’s findings have revealed that there are IHRM “best prac- tices.” However, they are not as universal as we had hoped. This is not a semantic teaser.

We believe that, in total, we have been able to derive some IHRM best practices common to the cultures we have compared. However, we haven’t analyzed the population of all cultures; thus the term “universal” cannot be fully em- braced. Some of this is simply not knowable at this point in the research process: not from these data nor from any other large scale, glo- bally distributed research consortium such as this. Some of this is due to the state of our “theory” around IHRM; some of it is due to methodological issues. We will begin with what we do not know now (nor could we expect to 124 • H UMAN RESOURCE M ANAGEMENT , Spring 2002 know as a result of this large-scale study) and then iterate up the knowledge chain. We will then identify what these data have told us about “best practices” and subsequently dis- cuss some of the unusual and/or counterintuitive findings. Practitioners and academics alike will not be surprised if we suggest here that we have not, as yet, a unify- ing theory of IHRM which drives questions such as those posed throughout this special issue. Ultimately it hangs on the theoretical underpinnings, which we address next.

What Theory Drives a Multicultural Study of “Best Practices?” Huo, Huang, and Napier (in press) remind us that IHRM research is best described as pa- rochial, and this parochialism notwithstand- ing, has found its way into the mainstream even after its patrons were chastised. 1 Witness today just how much emphasis is on the costs and benefits of various HR practices for the entire Multinational Enterprise (MNE) rather than the core characteristics of the HR prac- tices in each country. In other words, we know a great deal about compensating the expatri- ate manager, and how expatriates fare in the U.S., Europe, and Japan. We know next to nothing about an international compensation theory, or properties of compensation systems in other countries (to which those expatriates are sent!) (Lowe, Milliman, De Cieri, & Dowling, in press) that can generalize with any degree of accuracy. That said, there has been scant research on “best IHRM practices” in different countries. Possibly this is just be- cause there is no IHRM theory that guides us, or helps to meaningfully inform practice. Why?

There has been a long-standing debate in most of the literatures of the social sciences, from anthropology, psychology to organiza- tional science about the usefulness of work- ing intensively within one country to try to understand the issues purely pertinent to that culture (emic) versus those that believe work- ing across cultures can reveal universal gen- eralizations (etic) (Berry, 1990; Drost & Von Glinow, 1998; Teagarden & Von Glinow, 1997). However, all such a debate ultimately does is cloud the differences toward the means, and the argument reheats. It is in thatspirit that we have had debates around uni- versalism vs. particularism, or etic vs. emic, or convergence vs. divergence. It is not our purpose here to resolve that debate; however, it is our purpose to try to understand HR prac- tices uniquely within a culture (e.g., U.S.) and compare them to some other culture’s practices such that the overlap can be considered to be a “derived etic” (Berry, 1990), or a set of results that permit valid comparisons within the lim- ited set of cultures compared. Berry (1990, p.

92) notes: “If this procedure were to be ex- tended to all cultures and a derived etic thereby attained, then there would be a universal es- tablished. In other words, a universal is truly ‘universal to everyone’ while a derived etic is limited to that set of cultures for which it has been established.” Visually, it would look like Figure 1 (adapted from Berry, 1990).

Several researchers have used this frame- work, or variations on it, to try to get clarity on cultural differences. We think it is useful for two reasons: One is our discussion of “best practices” across different cultures. Here we have painstakingly sought to compare HR practices derived in North America [i.e., the U.S. (emic A)], then diffused to another cul- ture, [i.e., Australia (imposed etic)] in an at- tempt to discover HR practices in Australia’s culture (emic B). This allows for possible com- parison of emic A and emic B, and where com- parison reveals overlaps between the two, we have our “derived etic.” This is to say, (1) if both the U.S. and Australia utilize the same IHRM practices, (2) there is understanding of the cultural context within which these IHRM practices are applied, and (3) there is evidence that these practices are effective in both coun- tries, then it can be stipulated that the over- lapping part of the Venn diagram represents a “best practice,” or a “derived etic.” The second reason we believe the “derived etic” discussion is relevant here is that it may structure our thinking about doing cross-cul- tural comparisons by allowing us to use a com- mon solution methodology as we aim for theory development in IHRM.

Most of the work done within the field of IHRM generally tries to compare a few prac- tices across two countries. As a result, simi- larities and differences are assiduously reported, but there is little explanation given … it is our purpose to try to understand HR practices uniquely within a culture (e.g., U.S.) and compare them to some other culture’s practices such that the overlap can be considered to be a “derived etic” (Berry, 1990), or a set of results that permit valid comparisons within the limited set of cultures compared. Converging on IHRM Best Practices • 125 as to why and under what conditions the simi- larities and differences exist. 2 There is no at- tempt made to derive any “etics” or, for that matter, to determine whether the practices of the two countries are viewed from the same cultural lenses. Thus, context needs to be embedded into the discussion of “valid” ob- served similarities and differences in HR prac- tices, and it is to this methodological discussion we now turn our attention.

Traditionally, the use of valid and reliable standardized measurement scales 3 is a prereq- uisite for publication within any good research outlet, thus the measurement scale becomes the driver, and not the research question. Let us leave aside the issue of relevance for the moment, but take for example a selection tech- nique common in France: graphology, or hand- writing analysis. Any “measurement” device that fails to assess this selection criterion may be “standard/reliable” but not at all useful to the French. No amount of statistical manipu- lation or sophistication in structural equation modeling will make it so either. And while the “derived etics” that may emerge when it comes to selection criteria in the U.S. and France can exist, they will undoubtedly be somewhat limited and more measurement-bound than descriptive of selection practices. This is to say that much of what we know about IHRM today has unfortunately been bounded by our measurement parochialism or dependency.

There is an overtendency to rely on measure- ment precision in the reporting of differences to the exclusion of rich, thick contextual de-scriptions of cultures and their HR practices, and the extent to which they overlap with other countries. In other words, we have consider- able sophistication with respect to measure- ment devices, much like stomping out cockroaches with weapons of mass destruc- tion. Our intent here is not to criticize exist- ing measurement theory, nor use it as a ruse to obscure meaningful discussions as we at- tempt to devise theory on IHRM derived etics.

Our intent here is to make meaningful cul- tural comparisons and not be held prisoner to our measurement guns in the interest of per- mitting valid comparisons—a highly unortho- dox research approach to be sure.

As Geringer, Frayne, and Milliman (in press) more articulately note, we have a rather unusual research instrument: We do not use many standardized measurements. The LBDQ Short Form was the only one used. Single items (such as the graphology item discussed earlier) were used to assess the extent to which these practices are used now, and if the re- spondents thought they should be used more (or less) in the future. We looked at the rela- tionships between “is now” and “should be”— another taboo methodologically—as a proxy for ideological/within-culture gaps. So, al- though we tried to avoid comparing mean scores across nations, we looked very closely at the ideological gaps within a culture, since they do share a common frame of reference.

Huo, Huang, and Napier (in press) note that it is conceivable that a country could have a large gap between “is now” and “should be” Figure 1. Visual Model of Understanding of HR Practices within a Culture 1. Begin research in one’s own culture.

2. Transport to another culture.

3. Discover other culture.

4. Compare the two cultures.

5. Comparisons not possible?

6. Comparisons possible.

(Adapted from Berry, 1990) EMIC A IMPOSED ETIC EMIC B EMIC A vs EMIC B EMIC A ≠ EMIC B EMIC A EMIC B DERIVED ETIC 126 • H UMAN RESOURCE M ANAGEMENT , Spring 2002 means, but have a relatively low “should be” mean score; thus, the “should-be” mean re- veals much about how indigenous people view IHRM practices. In addition, comparing “is- now” practices with responses to how these practices “should be” used in the future, may be of some use in determining whether cur- rent pervasive practices are likely to be an ef- fective source of added value for the firm in the long term (Jennings, Cyr, & Moore, 1995).

Therefore, managerial attitudes about future practices may actually characterize the pref- erence functions that will be used to main- tain or adopt practices in organizations across cultures. In our data analysis among ten coun- tries/region, we have found the derived etics featured in Table I to be perceived as univer- sally important future IHRM practices.

Further, random sampling was often re- jected, in favor of snowball, and convenience sampling as well as personal contacts. We aimed for functional equivalence, and could hope for nothing more when it came to “roughly equivalent items.” We did not get hard data on organizational performance be- cause it was impossible in many SOEs, and, in any case, we took other proxies for organi- zational effectiveness. We also asked everyone to comment as to whether the HR practices contributed to high performing employees, satisfied employees, and organizational effec- tiveness. We did this on the survey knowing about common method bias, and attempted to triangulate on these items in other mean- ingful ways. We sought insights on IHRM practices within cultural contexts, and, even more importantly, we used the overlap of prac- tices within those different cultural contexts to help us with our derived etic, or in our ter- minology, a best practice. Ultimately, we see this as a first-stage attempt at theory develop- ment of IHRM. The data analysis of the ten countries and one region is still limited with respect to the set of cultures upon which it is now established. However, this derived etic has allowed us to finally permit valid comparisons within the set of cultures compared.

So, What Do We Know?

The results of this preliminary but massive data collection effort permit one general ob-servation and three categorizations. The gen- eral observation demands that we address methodological and contextual issues in cross- cultural research much more carefully. It is no longer enough to simply enter a new cul- ture with the “extant” survey instrument de- veloped and implemented in North America.

We were reminded of this years ago when Chris Brewster (Brewster, Hegewisch, Holden, & Lockhart, 1992) pummeled IHRM re- searchers with the indictment of America’s hegemony, 4 particularly as pertains to the use of measurements derived in the North Ameri- can context, and then generalized all over the world. The world in which we now live is much more complicated, and even the word “rigor- ous” must be questioned within this context.

It is not our intent to raise the specter of the rigor vs. relevance argument here; our goal is to sensitize researchers to key dilemmas en- countered when rigor is pushed to its utmost in order to conduct competent, publishable research. We reported on some of the dilem- mas encountered on these methodological travels earlier (cf. Teagarden et al., 1995).

However, we must now reinforce this with new and stronger cautions: Most of what cross-cul- tural and international research does is empha- size the imposed etic; from here on, we should be striving toward derived etics, and using con- structs that make sense within and across the cultures under study regardless of how it seems to those that would police such work.

To use an example mentioned throughout, one can question the selection of a senior-level French employee without acknowledging the use of graphology. Can it be done? Of course.

Should it? Probably not. And since there are no standardized instruments for assessing gra- phology, does this mean we should not ask the question in the first place? Our contention is no! On the contrary, we may be forced to vio- late some of the methodological tenets of “good” research in order to hone in on the right issues and to do it in such a way as to add value to our knowledge of IHRM. This is not to say the Best Practices Project has done this throughout. In some cases, we were lucky; in other cases, we were not as fortunate; how- ever, we must in the future consider our in- ternational methodological issues even more judiciously. We can ill afford to conduct re- It is no longer enough to simply enter a new culture with the “extant” survey instrument developed and implemented in North America. Converging on IHRM Best Practices • 127 search in the next millennium using the same “scientific approaches” we have perfected in the twentieth century. We have learned that the rig- orous execution of a research design and in- strument that masks the derived etic in favor of the imposed etic is simply error of the third kind or, solving the wrong problem well! We must embed context into the explanation of any uni- versals or derived etics; failure to do so will re- sult in marginal understandings of meaningful cross-cultural differences, or overly simplified explanations of similarities and differences in imposed etic HR practices.

In addition to the general observations around methodology that pervades our think- ing, we also observed three general catego- ries, to which we now turn. These are Universal Practices, Regional Practices, and Country-Specific Practices for each of the IHRM practices we probed: compensation, selection, appraisal, training and development, and strategic orientation. These comparisons are vital for moving the field ahead, and es- tablishing a theory of IHRM Best Practices.

However, this is not enough; it is necessary but not sufficient. After years of data collec- tion, and huge sums of research monies spent, we wanted more than what seemed “interest- ing” or even vital. What we discovered were many anomalies and counterintuitive findings that now empirically challenge much of what we think we know about HR practices and their generalizability. Let us examine each practice in turn, for universal, regional, or country-spe- cific practices and, in so doing, uncover some disturbing new twists.

Compensation Based on what we know from the literature, our data provide some support, but more con- tradictory evidence about how national cul- ture might be expected to impact employee preferences for compensation practices. In terms of pay incentives, many countries within our sample do not currently use pay incentives very much. Of the ten countries/ region we surveyed, only Taiwan and the PRC indicated above-average scores on the use of incentives. That means individualis- tic countries like the U.S. and Canada sur- prisingly had low mean scores on payincentives (which seems remarkable in view of the incredible emphasis on pay for perfor- mance, merit pay, etc., in the literature as well as the popular press). However, it also means that Taiwan and China (both collec- tively oriented societies) whose compensa- tion systems have historically emphasized collectivism and egalitarianism, now ironi- cally prefer reward differentials “determined primarily according to individual contributions”(Zhao, 1995, p. 127). There is also greater acceptance of wider reward dis- parities based on individual performance (Aiello, 1991, Ding, Fields, & Akhtar, 1997).

It is inevitable that compensation systems based on individual performance and indi- vidual incentives will become much more common and entrenched within the Chinese context, with further reforms on the horizon (Zhu, 1997). This is important because for China and Taiwan, much of the extant lit- erature speaks to the “iron rice bowl” con- cept. We have discovered just the reverse here in terms of pay incentives with an overarching emphasis on individual-level incentives!

Equally provocative is the more Anglo-ori- ented countries’ (U.S./Canada/Australia) lack of emphasis on pay incentives, individual or otherwise! We expected from the literature here that these types of individualistic coun- tries would place a much higher emphasis on “incentives” as a means of differentiating em- ployees from each another, while more col- lectively oriented countries would deemphasize these incentives in order to pre- serve group harmony (Milliman, Nason, Zhu, & De Cieri, 1998; Schuler & Rogovsky, 1998).

The results indicate a nearly universal agree- ment that incentives should comprise a lim- ited portion of the compensation package. The concept of “some, but not too much,” chal- lenges much of what we know from the merit pay and pay-for-performance literature. Per- haps this is because most employees are risk- averse, with cultural differences explaining degrees of risk aversion.

Additionally, the data indicate some agree- ment among our diverse countries that there should be a larger emphasis placed on indi- vidual job performance goals than specifically on group or organizational goals. This also contradicts much of the way global organiza- That means individualistic countries like the U.S. and Canada surprisingly had low mean scores on pay incentives (which seems remarkable in view of the incredible emphasis on pay for performance, merit pay, etc., in the literature as well as the popular press). 128 • H UMAN RESOURCE M ANAGEMENT , Spring 2002 tions are heading—via the use of teams rang- ing from globally distributed to co-located.

Again, this may also reflect risk aversion to some extent, given that most employees see individual performance as more controllable than group performance; however, it does not bode well for global organizations making wide use of teams, whether virtual or traditional!

Similarly, our study challenges the role that seniority plays in global organizations.

Seniority is a bona-fide occupational qualifi- cation in the U.S. for many jobs, and is as- siduously protected under the law. As well, seniority has been prized in many parts of the world for centuries. In much of Asian culture, Confucian values emphasize the role of the senior family member. Similarly, in Mexico and much of Latin America, “el patron” is also held with high respect. Collectivist cultures are generally thought to emphasize seniority in HR decisions to a much larger extent than indi- vidualistic cultures (Milliman et al., 1998).

Only three countries in our sample (Mexico, Korea, and the PRC) indicated that seniority should be emphasized more in the future, and, in a few cases, more individually oriented countries scored similarly to the collectively oriented countries. These findings may sug- gest that some type of social change is occur- ring, including a reduced emphasis on seniority in societies that traditionally have honored their elders (Stening & Ngan, 1997; Zhu et al., 1998).

In terms of benefits, this was seen as uni- versally important to compensation. Almost all respondents indicated that it was not so much the generosity of the benefits package, but that benefits should be an important part of the total compensation package.

Overall, the collectivist countries in our sample have only a slight, to moderately higher emphasis than the individualist coun- tries on seniority-based pay, pay based on group or organizational performance/goals, and pay based on future goals. Even though the pay systems between these various coun- tries do have a number of important differ- ences, they seem to share more similarities than we had anticipated.

Our ideological difference analysis (differ- ence between “is-now” and “should-be” scores within a country) reveals that the countries inthe Americas had the largest gaps between current practices and desired future practice on pay policies. There were also large gaps on many of the pay practices items. It is puzzling as to why the gaps are greater in the Western- ized countries rather than the Asian countries, and we think it plausible that one reason is what Kluckhohn and Strodbeck (1961) re- ferred to as a culture’s “orientation toward time.” It is generally assumed that most coun- tries in Asia tend to share a time orientation that looks to the past—emphasizing the sta- tus quo, stability, and slow change. In con- trast, countries such as the U.S. tend to have a future time orientation, which emphasizes flexibility, and more change to meet the de- mands of the future (Milliman et al., 1998).

Another explanation would relate to harmony; in this case the Asian countries seeking less change or departure from the past to ensure harmonious relations.

What appear to be universal trends for the future include the use of benefits, the empha- sis on individual performance rather than group/team or organizational performance, and a reduced emphasis on seniority.

Selection Our data revealed a significant number of cross-national differences in the recruiting criteria and the procedures used to select employees. By and large, the more individual- istic countries of the U.S., Australia, and Canada all considered the job interview, one’s technical skills, and work experience as the most important selection criteria. In fact, two of those items were universally rated high in all countries with one exception: China. The items that seemed to be high in almost all countries except China were the personal in- terview and technical requirements of the job.

All of the other items seemed to have a high degree of culture-driven, contextual diver- gence. Indeed, there seems to be more diver- gence here than convergence in recruiting criteria with the earlier exception noted. There is some similarity between Japan, Taiwan, and Korea in selection criteria. However, this is limited to a person’s potential to do a good job and get along. What appears fascinating, how- ever, is the ideological gap analysis. This sug- What appear to be universal trends for the future include the use of benefits, the emphasis on individual performance rather than group/team or organizational performance, and a reduced emphasis on seniority. Converging on IHRM Best Practices • 129 gests that individualistic countries are deemphasizing the “proven work experience” in favor of “one’s ability to get along with oth- ers,” and “how well the person will fit the company’s values,” as part of the top three hiring criteria for future hiring. This dovetails with some more recent observations that a person’s past work experience is no longer as highly valued in the U.S. as it was in the past, while the person’s potential to become a com- petent employee in the future has become much more important (Munk, 1999).

There is a striking similarity between Ja- pan and Taiwan in terms of their current and future selection practices. Some country-level differences warrant comment here. As was mentioned earlier, the French use of graphol- ogy is critical to the French, so is the person’s potential and ability to get along with others critical to the Japanese. Japanese placed a rela- tively low weight on job-related skills as a se- lection criterion. One can only surmise why this is so, and to what extent this need for harmony and potential over technical/work related skills has contributed to economic dif- ficulties at the country level over the last de- cade. The same could be said of Korea, which shares with Japan a similarity in their hiring practices with both deemphasizing proven work experience in favor of hiring new gradu- ates. However, the Koreans also favor use of employment tests, and, when it comes to those employment tests, many of the elements con- tained within speak to the issue of harmony and getting along with others. Taiwan is dif- ferent here—the Taiwanese favor the job in- terview, together with proven work experience as the key selection criteria. So, while there is some similarity at the country level there are many differences observed between the coun- tries in our sample. Overall, the results can suggest a slight movement toward convergence in global hiring practices in the future, but a strong country emphasis dominates the land- scape today. Huo, Huang, and Napier (in press) speculate that the globalization of busi- nesses, coupled with the communication and IT revolution, will undoubtedly affect how recruiting practices evolve toward the irresist- ible—global convergence! This finding is less controversial than our findings on compensa- tion, but raise important issues that will bediscussed in the final section. Overall, it can be said that future universal trends empha- size “West meets East.” There is some conver- gence in selection criteria away from the use of experience (emphasized in the West) to ability to get along with others and how well the person fits with the company’s values (em- phasized in the East).

Performance Appraisal Just what are the purposes of performance appraisal? Worldwide, it has been stated in the (mostly North American) literature that ap- praisals have been used for several different purposes: (1) documentation (documenting a subordinate’s performance, evaluation of the subordinate’s strengths/weaknesses, etc.), (2) future development, (3) administrative pur- poses (for pay, salary, promotion decisions) and (4) allowing subordinates to express their views (Milliman, Nason, Zhu, & De Cieri, in press).

However, our data reveal a mixed picture on the extent to which performance appraisals are used to accomplish these purposes. The good news is that virtually every country conducted performance appraisals for these purposes; however, the bad news is that this was only on a low-to-moderate basis. Why else would per- formance appraisals be practiced if not for the aforementioned purposes? Is it because it was expected on the part of someone, or a parent company? Is it because it just seems to be the right thing to do? Is it for fear of legal ramifications encountered by many MNEs? Furthermore, the is-now data contra- dict our predictions from the literature review on two of the purposes (documentation and development) in many of the countries. The current documentation scores were slightly lower than we expected in both the U.S. and Canada, very low in most other countries, but higher than expected for Japan. The U.S. case is intriguing, because most of the literature stems from U.S.-based studies, and the U.S.

is-now scores were lower than what we had expected for pay and promotion purposes. It was also predicted that the development-re- lated items would be highly emphasized in many of the countries, but the data indicate these purposes are emphasized only to a small extent in most of the countries. Just what are the purposes of performance appraisal? 130 • H UMAN RESOURCE M ANAGEMENT , Spring 2002 In terms of the ideological gaps, the should- be means were uniformly higher than the is- now means on every purpose in every country, suggesting that the purposes of appraisal are not implemented as satisfactorily as they were intended to be implemented across the board.

Thus, it may be fair to say that the purposes of performance appraisal have fallen short not only in the U.S., but in every country we have sampled. This borders on a universal finding; however, it is surprisingly the reverse of what we would have predicted based solely upon the performance appraisal literature!

Even though it is difficult to draw defi- nite conclusions relative to best practices with- out assessing the contextual factors, there are some interesting patterns that emerge region- ally, based on the ideological gap analysis. The countries of the Americas (Canada, the U.S., Mexico, and Latin America) in concert with Australia and Taiwan revealed large should-be means, on development and documentation appraisal purposes. In particular, recognizing the subordinate and evaluating the achieve- ment of the subordinate’s goals received the highest scores in documentation purposes, along with two development items focusing on improving performance. A similar pattern is noted on two subordinate expression items among the same countries with the exception of Taiwan. It is noteworthy here that these countries consider an important purpose of the performance appraisal is to allow subor- dinates’ freedon to express their thoughts, whereas the rest of the countries indicated a more moderate view of subordinate expression in appraisal reviews. Perhaps, as Milliman and his colleagues argue (1998), this is due to the high power distance orientation of most Asian countries which places less value on subordi- nate input and participation in the appraisal process (Huo, Huang, & Napier, in press).

Similar findings were observed for the ad- ministrative purposes of pay and promotion, which received only moderate emphases across most of the countries. The exceptions were Mexico, Taiwan, and Latin America where it was generally seen that there should be more emphasis on discussing promotions rather than pay in appraisals. Perhaps the delicate if not tenuous balance between discussions of pay linked to appraisal is a dilemma worldwide.Taken as a whole, development and docu- mentation were viewed as more important than administration and subordinate expres- sion. But there are some important geographic similarities between the four countries in the American continent, Australia, and Taiwan, on the one hand, and the four Asian countries of Indonesia, Japan, Korea, and China, which grouped together, on the other hand. Inter- estingly, Taiwan follows much more closely the U.S. than it does China perhaps because of its strong market orientation and democratic norms. Equally important is the controversial nature of the performance appraisal process worldwide. Whether we discuss appraisal prac- tices as they are enacted now, or as they should be enacted, there is considerable controversy and contradiction based on what we could have expected from the existing literature on performance appraisal. Our ideological gap analysis reveals greater “should-be” scores on every purpose in every country, a stunning universal trend for performance appraisal.

Training and Development No universal or common T&D practices across the countries and region examined were found; however, significant similarities in practices within country clusters were observed. The managers from the Asian countries stood out as the most consistently satisfied with their organizations’ training and development prac- tices. By contrast, the managers from Mexico and Latin America observed their organiza- tions’ training and development practices as largely inadequate and in need of improve- ment. The American, Canadian, and Austra- lian managers perceived what we observe to be “softer” T&D practices (team building, understanding business practices, and corpo- rate values) as well as the more “proactive” T&D practices (preparation for future assign- ment and cross-training) to be sadly lacking, and in need of substantial improvement. The common practices found within these three country clusters are strongly influenced by cultural values and type of industry, two im- portant contextual elements.

Cultural Values. In Asia, the work culture has been characterized as collectivist. In col- lectivist societies, workers are integrated into Our ideological gap analysis reveals greater “should-be” scores on every purpose in every country, a stunning universal trend for performance appraisal. Converging on IHRM Best Practices • 131 strong, cohesive in-groups, which continue to protect them in exchange for unquestionable loyalty throughout their lives (Hofstede, 1991). Consistent with this philosophy is the commitment to training as well as develop- ment across the Asian countries evidenced by the small ideological gap between the is-now and should-be T&D practices. In contrast with T&D practices in the U.S. which deemphasize T&D in times of economic difficulties, many of the Asian countries in our sample have ex- perienced considerable economic difficulties and yet they still emphasize T&D practices.

In the U.S., Canada, and Australia, work cultures have been characterized as individu- alistic in nature. In individualistic societies, the relationship between employer and em- ployee is utilitarian, or similar to a business transaction (Hofstede, 1991). Here, a worker acts to optimize his/her own needs, and the employer expects him/her to perform the job accordingly. Consistent with this philosophy is a nonproactive, reactive, and task-oriented approach to T&D practices in these Anglo countries. This is seen as a small ideological gap in technical training, and a large ideologi- cal gap in all of the future-oriented training practices, and the “softer” management prac- tices such as team building, understanding business practices and corporate values.

Industry Differences. Our results indicate that T&D practices accommodate shifts in industry requirements and relate to needs for enhanced competitiveness. The Mexican and Latin American country cluster is character- ized by the continuous need to upgrade basic skills in labor-intensive industries, as seen in the large ideological gap in the “softer” man- agement practices such as team building, un- derstanding business practices and corporate values. This could imply that the Mexican and Latin organizations sampled were in a height- ened learning state with respect to modern management techniques, and in need of ap- propriate training and development.

In the U.S., Canada, and Australia, where industry and technology tend to be more ser- vice-driven, worker productivity can be in- creased by resourceful T&D practices (Jennings, Cyr, & Moore, 1995). In such en- vironments, companies require different skill sets such as critical thinking, team skills, andlearning on the job. Another indication of the trends facing different industries, and the ideo- logical gaps in training is the pressure on com- panies to cut costs and outsource training. As training moves outside the organization, par- ticularly in the U.S., there are myriad prob- lems associated with its relevance. Witness the recent aphorisms “training du jour” or “pro- gram of the month” discussed in much of the more recent literature on training and HR ini- tiatives gone haywire (Kerr & Von Glinow, 1997; Salas, Cannon-Bowers, Rhodenizer, & Bowers, 1999). The obvious conclusion here is that employees perceive that training is nei- ther valued nor effective; hence the large ideo- logical gap between all T&D practices with the exception of technical training.

In Asia, globalization and the expansion of trade have altered the structure of many industries. As a result, labor-intensive indus- tries will shift locations to reduce costs (Jennings, Cyr, & Moore, 1995). For example, Taiwan and South Korea have traditionally been the preferred low-cost sites for manu- facturing operations. However, as wages and other cost-related factors began to escalate in these countries, multinationals began to look toward Indonesia, Malaysia, and Thai- land for their manufacturing operations (Teagarden, Butler, & Von Glinow, 1992).

Therefore, industries in Taiwan and Korea will, and have, become more capital-intensive and technology-driven, creating different training needs, as well as employee develop- ment needs. There are very few training needs assessments performed in the U.S. and else- where, raising new challenges. Despite the 1997–1999 economic crisis which high- lighted many training needs, the trend in T&D for Asian countries has remained surprisingly stable and effective. These results may reflect the collectivist mind-set of the Asian manag- ers, perhaps indicating an unwillingness to report inadequacies in their organizations’ training and development practices.

Role and Status of HRM Departments: Is the HR Department Linked to Company Strategy?

Bowen, Galang, and Pillai (in press) note that there are few differences in the status of HRM … industries in Taiwan and Korea will, and have, become more capital- intensive and technology- driven, creating different training needs, as well as employee development needs. 132 • H UMAN RESOURCE M ANAGEMENT , Spring 2002 departments in this empirical investigation: All countries view the HR Department’s status as fairly high. Ironically, this is in direct contrast with the evidently “old” assumption in the U.S.

of low status of the HR Department, and there is evidence now that there is a strategic focus for firms that see HRM status linked to orga- nizational capability as a source of competi- tive advantage (Becker & Gerhart, 1996). The focus on organizational capability implies that a company can develop a set of processes that are capable of sustaining competitive advan- tage. However, the HRM practices that were highly correlated with increased organizational capability in our study are somewhat surpris- ing: Both training and development, as well as performance appraisal practices, were viewed as contributing to enhanced organiza- tional capability. This is less surprising for T&D than for performance appraisal, in view of the finding on how poorly PA was imple- mented worldwide (Milliman, Nason, Zhu, & De Cieri, in press) as reported earlier in this special issue. In some sense, the proliferation of U.S.-based training models around the world, especially in the English-speaking cul- tures as well as widescale adoption of U.S.

models of performance appraisal in these cul- tures (regardless of how poorly executed they might be) could account for these findings.

It is more than intuitively appealing to suggest that an organization’s strategy depends upon certain contextual factors, as well as ef- fective HRM practices. The finding that there is a strong link between organizational capa- bility, the collective abilities that an organiza- tion needs to execute its business strategy, and HR practices in a number of countries, is worthy of further exploration. In companies trained to view human resources as a distinct source of competitive advantage, finding the right cocktail mix of people, culture, process, structure, and assets is an exercise in strength- ening their organizational capabilities (Turnbull, 1996). It is the sustainability of the competitive advantage at stake here, and the reason so many companies focus on organiza- tional capability and increasingly seek HR so- lutions to strategic goals. Specifically, the U.S., Australia, Canada, China, and Indonesia all see the importance of linking HR practices to strategy. Is this a passing fancy, or a findingthat can withstand the test of time? This is an empirical question, however, there is increas- ing theoretical and empirical support that aligning a firm’s strategy with its HR practices is critical to organizational performance (Becker & Gerhart, 1996; Bird & Beechler, 1995). One study focused on “bundles” of HR practices that included team-based work sys- tems, extensive training, and performance- based compensation systems rather than individual human resource practices, and sug- gested that the organization create mutually reinforcing conditions that support productiv- ity (Phipps, 1995). Our study finds a similar connection between strategy and effective HR practices and fuels future explorations as to how this linkage translates into productivity improvements. Bowen, Galang, and Pillai (in press) caution that countries embroiled in political turmoil, or that are lower on the in- dustrialization or globalization chain, or those that are moving from centrally planned toward more market-driven economies are all subject to dramatic shifts influencing the role and sta- tus of HR Departments. What is becoming abundantly clear is that rapid advances in com- munication and information technology, knowledge management, and global competi- tion for markets has made it imperative that companies of the future work toward adopt- ing best practices that have been documented within a polycontextual study, such as the one reported on here. In other words, those best practices that are truly universals are based on derived etics and not simply on imposed etics using North American standards.

Implications for Theory, Research, and Practice Table I presents the trends we have high- lighted in this special issue. We have at- tempted to identify those trends that appear to have universal applicability, either now or in the future as a desired end state. Further- more, there are some clusters of countries or regions that appear to group together and not necessarily along immediately apparent clus- ter lines. For example, sometimes Taiwan aligns more with Australia and the U.S. than it does with China, and as such, may seem counterintuitive relative to the Asian region. The finding that there is a strong link between organizational capability, the collective abilities that an organization needs to execute its business strategy, and HR practices in a number of countries, is worthy of further exploration. Converging on IHRM Best Practices • 133 Practice COMPENSATION SELECTIONUniversals Derived ETICS “Best Practices” Pay incentives should not comprise too much of an employee’s compensation package.

Compensation should be based on individual job performance.

There should be a reduced emphasis on seniority.

Benefits should comprise an important part of a compensation package.

“Getting along with others”, and “Fit with the Corporate Values” signals a shift in selection from “West meets East.”Regional or Country Clusters Seniority-based pay, pay based on group/team or organiza- tional goals, and pay based on future goals—all are used to a larger extent in the Asian and Latin countries now.

Selection practices were remarkably similar among the Anglo countries. Specifically, job interview, technical skill, and work experience are the most important selection criteria. How well the person fits the company’s values replaces work experience as one of the top selection criteria for future selection practices.

Selection practices are quite similar in Korea, Japan, and Taiwan. Specifically, proven work experience is de- emphasized as a selection practice in these countries.

In the Anglo and Latin American countries, allowing subordinates to express themselves is perceived as an important future appraisal practice.Country Specific U.S. and Canada had less use of pay incentives than expected.

China and Taiwan had above-average use of pay incentives, and wanted more based on individual contributions.

In Japan, a heavy emphasis is placed on a person’s potential (thus hiring new graduates) and his/her ability to get along with others. A relatively low weight was given to job- related skills, and experi- ence as a selection criterion.

In Korea, employment tests are considered crucial and are used to a large extent as a selection tool, as well as hiring new graduates. Koreans de- emphasize experience.

In Taiwan, the job interview is considered the most important criterion in the selection process. TABLE I Trends in International Human Resource Management Practices Across Selected Countries and Regions. However, Table I also attempts to identify country-specific findings that will be useful for IHRM researchers and practitioners alike, which have been highlighted throughout this special issue. Because our findings were puz- zling, counterintuitive, and contrary to what the research literature suggests on each of the HR practices in the ten countries/region, we have attempted to illustrate how the useof derived etics/universals/best practices can indeed add value to theory, research meth- odology as well as practice.

Earlier, we indicated that we had no overarching theory of IHRM that would help to guide our thinking. Even worse, our meth- ods for conducting so-called traditional 5 cross- cultural research masked discovery of universal practices constituting a “derived (continued on next page) 134 • H UMAN RESOURCE M ANAGEMENT , Spring 2002 etic.” Our discussion of just how derived etics might help us master this dilemma throws us squarely into the realm of what we call polycontextuality, or the embeddedness of multiple contextual elements. This in turn influences more how researchers think aboutuniversal best practices rather than what they think about them. In other words, culture re- searchers discuss national culture, strategy researchers discuss contingencies, and HRM researchers discuss specific HR practices; however, we submit that the way in which we Universals Derived ETICS “Best Practices” In all countries, the “should- be” scores were higher on every purpose, suggesting that the pur- poses of PA have fallen short in every country.

All countries indicated that a greater emphasis be placed on development and documentation in future PA practices. In particular, recognizing subordinates, evaluating their goal achievement, planning their development activities, and (ways to) improving their performance are considered the most important appraisal practices for the future.

In most countries, T&D practices are used to improve employees’ technical skills.

There is a growing trend toward using T&D for team building and “soft manage- ment practices.”Regional or Country Clusters In contrast, in the Asian countries expression is used to a low extent, particularly in Korea.

In the Latin American countries, the administrative purposes of performance appraisal are considered important in future practice.

In the Anglo countries, the softer T&D practices such as team building, understanding business practices and corporate culture, and the pro-active T&D practices such as preparation for future assignment and cross-training are used moderately; however, a significant increase in these practices is desired.

In the Latin countries, an increase in the extent to which all T&D practices are used is desired.

In the Asian countries, most T&D practices are used moderately and are consis- tently considered satisfactory. Practice PERFORMANCE APPRAISAL T&D TABLE I (continued.) Country Specific In Taiwan, the administrative purposes of performance appraisal are considered important in future practice.

In Mexico, T&D as a reward to employees is considered a highly desirable practice.

In the U.S. and Korea, preparing employees for future job assignments is used to a lesser extent.

U.S. is using outsourcing more.

In Japan, remedying past performance is used to a small extent, however, a significant increase in this practice is desired.

In Korea, team building is used extensively and emphasized in all T&D practices. (continued on next page) Converging on IHRM Best Practices • 135 should think about the embeddedness of mul- tiple contextual elements, or polycontextuality, is critical to understanding universality. We do not believe that much value can be added by examining single HR practices without exam- ining them within their multiple embedded contexts. Thus, it is imperative to seek derived etics where possible, and to do so with full understanding of specific practices within context. Let us revisit the French graphology discussion to elaborate this point. We now believe we have some universal best practices on selection criteria; those should be employed in France in concert with practices (graphol- ogy tests) made salient by the French context.

This allows us to bring context into the equa- tion more openly. We are not suggesting a vari- ant on configurational approaches for IHRM theory building, but instead a strong empha- sis on embedding context into specific coun- try practices so that emic practices (graphology) can become embedded in etic theories (selection). Does this mean that we have universally derived etics on all aspects of selection? No; however, it does allow us to suggest a solution methodology to the research around global selection. Recall that a derived etic hinges on both cultures A and B using some of the same IHRM practices. However, recall also that there must be understandingof each culture’s context, and there must be some evidence of effectiveness in both A and B; then it can be stipulated that the overlap of A and B is the derived etic. This is not a solu- tion, but a solution methodology that allows us to finally permit valid comparisons within the set of cultures compared. Therefore, al- though we suspect we have not made head- way on a theory of IHRM with this large-scale study, we do think we have contributed to a solution methodology by which we can now examine derived etics, not just imposed etics, or emic practices, a common practice in IHRM research. After all, best practices, or universals do not exist independently of cul- tural context, and it may well be that best prac- tices or these derived etics we have been discussing may act in combination with one another for an even greater impact: The whole may indeed be greater than the sum of the parts.

One noteworthy area is the debate re- ported on by Becker and Gerhart (1996) of what was termed “best practice-fit” which fo- cuses on a within-company levels of analysis argument. They discuss the overarching ru- bric within which best practice can be best discussed as the architecture of the system.

Thus, a best practice effect might be attrib- uted to some overarching architecture, which Practice RELATION TO BUSINESS STRATEGY STATUS OF HRM FUNCTIONUniversals Derived ETICS “Best Practices” Across most countries, the HRM practices most closely linked to organiza- tional capability are training and development and performance appraisal.Regional or Country Clusters In the Asian countries, linkages were indi- cated between both low cost and differen- tiation strategies and HRM practices.Country Specific In Mexico, no linkages were indicated between organizational capabil- ity and HRM practices.

In Japan and Taiwan few linkages were indicated between organizational capabil- ity and HRM practices.

Status of HRM was highest in Australia and lowest in Indonesia.

TABLE I (continued.) 136 • H UMAN RESOURCE M ANAGEMENT , Spring 2002 would account for differences in practices.

They focus on the strategic element of best practice in this argument, whereas we are at- tempting to derive etics, or best practices that have universally been endorsed at the level of practice taking polycontextuality into account.

That said, we also believe that there are implications for practice as well here. Clearly the previous discussion suggests much greater attention to the multiple contexts within which international business is done. These local contextual issues should not be dismissed as irrelevant to the “corporate culture” of the parent company; in fact, they are critical and can influence, as we have seen, all HR prac- tices that are “imposed” on the local enter- prise, subsidiary or its indigenous peoples. We briefly highlight these implications now:

Compensation. Overall, our research re- veals that, in the future, compensa- tion systems should be based more on individual level job performance, and not team or organizational perfor- mance. Global firms should also re- duce their reliance on seniority, and pay incentives should not comprise very much of the compensation pack- age at all. Benefits should continue to play an important role in all com- pany compensation packages. Does context matter in these findings? Of course! In some cultures, compensa- tion is set at the level of the “state” which mandates the conditions. If seniority were automatically dismissed in the U.S. where seniority can be a bona-fide occupational qualification, even though a finding here suggests worldwide movement away from se- niority-based systems, there could be debilitating labor relations problems encountered. Thus, any of these find- ings must examine the contexts within which these best practices emerge as universally endorsed.

Selection. The universal/derived etics on selection reinforce a shift, which we term “West meets East.” For example, in most selection decisions, there was universal endorsement of how well the person “fits” with the corporate/com-pany values, and how well the person “gets along” with others in the corpo- ration. There were regional or coun- try-cluster findings showing striking similarities among Anglo countries.

Specifically, the job interview, techni- cal skill proficiency, and work experi- ence were all seen as the most important selection criteria today; however, the shift toward the more Eastern notion of getting along with others and fit to company values was noted for the future. If these best prac- tice selection/recruiting practices vary only slightly all over the world, but, perhaps along critical context lines, the company could still encounter dif- ficulties. For example, if an American MNE goes into the Korean market vowing to select personnel by using a “fit to the corporate values” criterion along with their other top three crite- ria, using personal interviews and fail- ing to use employment tests can be just as disastrous as going into France without the graphology test. Unfortu- nately, a little knowledge can be a dan- gerous thing, and if American MNEs want to navigate the Korean terrain using American preferred selection criteria, and are inattentive to Korean contextual elements, it is no wonder that they encounter selection difficul- ties.

Performance Appraisal. Given the ubiquity of U.S.-based performance appraisal instruments, and the presumed popu- larity of these devices to “change” per- formance, it was surprising to us just how poorly executed most perfor- mance appraisal events were in our entire sample. However, our research revealed that PA should be used to rec- ognize subordinates, evaluate subordinate’s goal achievement, plan subordinate’s development activities, and focus on ways to improve subordinate’s performance. Our sense here is that context-specific issues were not taken into consideration when executing a performance ap- praisal activity. For example, we have Overall, our research reveals that, in the future, compensation systems should be based more on individual level job performance, and not team or organizational performance. Converging on IHRM Best Practices • 137 generally operated under the fairly old assumption of “praise in public, criti- cize in private.” However, there are some cultural contexts wherein pub- lic praise presumes the employee could not perform the activity in the first instance, and thus praise can be seen as unintentionally demeaning.

There are other cultures where criti- cism should not be delivered verbally; clearly context matters.

Training and Development. Training and development activities globally seem to be moving toward team building and “soft management practices.” However, T&D continues to be used currently to improve employees’ tech- nical skills. Hence, a gap exists be- tween what employees have now, and what is desired in the future, particu- larly for Anglo and Latin American employees. Small gaps also exist in Asia, where T&D is highly valued.

Ironically, for all countries, gaps exist despite an appalling lack of training “needs assessments” or training “evaluations.” In summary, we found some significant global or converging trends in IHRM practices across the cultures, industries, and organiza- tions we examined. These trends are derived etics and to some extent reflect the diffusion of HRM practices wherein traditional West- ern approaches are converging to more East- ern HRM practices and vice versa. We believe that we have made a significant contribution to the IHRM literature, particularly as regards the development of a shared methodology to unearth vexing issues such as universal bestpractices, and derived etics, and hone in on IHRM practices that seem to withstand the “test of time.” In so doing, we have a height- ened awareness of the polycontextuality that surrounds all of our IHRM practices.

Similarly, we also believe we have some- thing very important to say to practitioners, HR professionals, and those concerned with managing globally distributed teams. We be- lieve that what we have discovered global managers desire more (or less) of (compared to what they have presently) reflects their per- ception of effective practices in the future.

This might be construed as a preference func- tion for how selection, training and develop- ment, compensation, and performance appraisal should be done for an organization to increase their capabilities for enhanced competitiveness in tomorrow’s pressure- packed global environment. These global man- agers have been quite clear on this as important universal “best practices” which can enhance a firm’s capability. Our gap analysis reveals some universal “best practices” that may well contribute to sustainable competi- tive advantage. Although sustainability (from a resource-based view) is based on inimitable practices, understanding the polycontextuality of an IHRM practice might just provide the organization with the inimitability of such a practice, ultimately resulting in its competi- tive advantage. We strongly believe that this ten-country/nation study has been able to con- tribute to our overall understanding of how context affects IHRM practices. New ways of thinking about IHRM practices can ultimately allow us to think more broadly about how best to conduct our business in the 21st century for positive results well beyond a firm’s com- petitive advantage. 138 • H UMAN RESOURCE M ANAGEMENT , Spring 2002 REFERENCES Aiello, P. (1991). Building a joint venture in China:

The case of Chrysler and the Beijing Jeep Cor- poration. Journal of General Management, 17(2), 47–64.

Becker, B., & Gerhart, B. (1996). The impact of hu- man resource management on organizational performance: Progress and prospects. Academy of Management Journal, 39(4), 779–801.

Berry, J.W. (1990). Imposed etics, emics and derived emics: Their conceptual and operational status in cross-cultural psychology. In T.N. Headland, K.L. Pike, & M. Harris (Eds.), Emics and etics:

Insider/outsider debate (pp. 84–99). Newbury Park, CA: Sage Publications.

Bird, A., & Beechler, S. (1995). Links between busi-ness strategy and human resources management strategy in U.S. based Japanese subsidiaries: An empirical investigation. Journal of International Business Studies, 26, 23–47.

Bowen, D.E., Galang, C., & Pillai, R. (in press). The role of human resource management: An explor- atory study of cross-country variance. Human Resource Management.

Brewster, C., Hegewisch, A., Holden, L., & Lockhart T. (1992). The European human resource man- agement guide. London: Academic Press.

Ding, D., Fields, D., & Akhtar, S. (1997). An empiri- cal study of human resource management poli- cies and practices in foreign-invested enterprises in China: The case of Shenzhen Special Eco- nomic Zone. The International Journal of Hu- man Resource Management, 8(5), 595–613.

Drost, E.A., Frayne, C., Lowe, K.B., & Geringer, J. Mary Ann Von Glinow is CIBER Director and Professor of Management and Inter- national Business at Florida International University. Previously at USC, she was the 1994–1995 President of the Academy of Management. She has written nine books, many articles, serves on 11 editorial review boards, and several international panels. Dr.

Von Glinow consults and teaches in Latin America, Asia, the U.S., and Europe. Her work on IHRM as head of the “Best Practices” consortium won an award from the American Society for Competitiveness. In 2001, she received an NSF grant on globally distributed software development projects. She serves on the boards of Fielding Gradu- ate Institute, Friends of Bay Oaks, and Animal Alliance, and won the Humanitarian Award of the Year from Miami’s Adopt-a-Pet.

Ellen A. Drost is a visiting professor at American University, where she teaches inter- national business. She was previously on the faculty of Thunderbird and San Diego State University. Ellen has authored numerous articles, manuscripts, and conference papers in international strategic human resource management and cross-cultural lead- ership. Her most recent research focuses on strategic decision-making and entrepre- neurship. Ellen is a principal researcher in an international research consortium, which investigates best international human resource management practices. She has col- laborated as a consultant, benchmarked organizations in Mexico, and given lecture tours in Mexico. Ellen received her doctorate degree in international business and man- agement from Florida International University. She received her management experi- ence in Amsterdam and Brussels with the International Gold Corporation, an international marketing organization. Ellen was born in the Netherlands and speaks, reads, and writes Dutch, English, and French, fluently.

Mary B. Teagarden is Professor of Global Strategy at Thunderbird, The American Graduate School of International Management, where she teaches Global Strategy and Strategic Human Resource Management. Her research interest focuses on com- petitiveness, strategic alignment, and capability building with an emphasis on the management of technology-intensive firms, off-shore manufacturing, high technol- ogy transfer, and strategic human resource management in the People’s Republic of China and Mexico. Her current research focuses on the localization of leadership development for transformation and strategic alignment in the telecommunications industry in China. She received her Ph.D. in strategy implementation from the Uni- versity of Southern California. Converging on IHRM Best Practices • 139 M. (in press). Benchmarking training and devel- opment practices: A multi-country comparative analysis. Human Resource Management.

Drost, E.A., & Von Glinow, M.A. (1998) Leadership behavior in Mexico: Etic philosophies/Emic prac- tices. Research in International Business and International Relations, 7, 3–28.

Geringer, J. M., Frayne, C., & Milliman, J. (in press).

In search of “best practices” in international hu- man resource management: Research design and methodology. Human Resource Management.

Hofstede, G. (1991) Cultures and organizations: Soft- ware of the mind. Berkshire, England: McGraw- Hill Europe.

Huo, Y.P., Huang, J.H., & Napier, N.K. (in press).

Divergence or convergence: A cross-national com- parison of personnel selection practices. Human Resource Management.

Jennings, P.D., Cyr, D., & Moore, L.F. (1995). Hu- man resource management on the Pacific Rim:

An integration In L.F. Moore & P.D. Jennings (Eds.), Human resource management on the Pa- cific Rim: Institutions, practices, and attitudes (pp. 351–379). Berlin: de Gruyter.

Kanter, R.M. (1995). World Class: Thriving Locally in the Global Economy. NY: Simon & Schuster.

Kerr, S., & Von Glinow, M.A. (1997) The future of human resources: Plus ça change, plus c’est la même chose. In D. Ulrich, M. Losey, & G. Lake (Eds.), Tomorrow’s HR management (pp. 175– 181). New York: John Wiley & Sons, Inc.

Kluckhohn, C., & Strodtbeck, F.L. (1961). Variations in value orientations. Evanston, IL: Row, Peterson & Company.

Lowe, K.B., Milliman, J., De Cieri, H., & Dowling, P.J. (in press). International compensation prac- tices: A ten-country comparative analysis. Human Resource Management.

Milliman, J.F., Nason, S., Gallagher, E., Huo, Y.P., Von Glinow, M.A., & Lowe, K.B. (1998). The impact of national culture on human resource manage- ment practices: The case of performance ap- praisal. In G.R. Ferris & K.M. Rowland (Eds.), Research in personnel and human resources management: Supplement 2 (pp. 201–219).

Greenwich, CT: JAI Press.

Milliman, J.F., Nason, S., Zhu, C., & De Cieri, H. (in press). An exploratory assessment of the purposes of performance appraisals in North and Central America and the Pacific Rim. Human Resource Management.

Munk, N (1999, February 1). Finished at forty. For- tune, 139(2), 50–66.

Phipps, P.A. (1995) Firm productivity and competi- tiveness. Monthly Labor Review, 118 (5), 64–74.

Salas, E., Cannon-Bowers, J.A., Rhodenizer, L., & Bowers, C. (1999). Training in organizations:

Myths, misconceptions and mistaken assump- tions. Research in Personnel and Human Re- sources Management, 17, 123–161.Schuler, R.S., & Rogovsky, N (1998). Understanding compensation practice variations across firms:

The impact of national culture. Journal of Inter- national Business Studies, 29(1), 159–177.

Stening,. B.W., & Ngan, E.F. (1997). The cultural context of human resource management in East Asia. Asia Pacific Journal of Human Resources, 35(2), 3–15.

Teagarden, M.B., Butler, M., & Von Glinow, M.A.

(1992). Mexico’s maquiladora industry: Where strategic human resource management makes a difference. Organizational Dynamics, 20(3) (Winter), 34–47.

Teagarden, M.B., & Von Glinow, M.A. (1997). Hu- man resource management in cross-cultural con- texts: Emic practices vs. etic philosophies.

Management International Review, 37, 7–20.

Teagarden, M.B., Von Glinow, M.A., Bowen, D., Frayne, C., Nason, S., Huo, P., Milliman, J., Arias, M., Butler, M., Kim, N., Scullion, H., & Lowe, K. (1995). Toward building a theory of comparative management research methodol- ogy: An idiographic case study of the best inter- national human resources management project.

Academy of Management Journal, 38(5), 1261– 1287.

Turnbull, A.M. (1996) Redesigning the organization and making it work with people strategy. Cana- dian Management Accounting, 70, 19–20.

Von Glinow, M.A. (1993). Disgnosing “best practice” in human resource management practices. In J.B.

Shaw, P.S. Kirkbridge, K.M. Rowland (Eds.), Research in perosnnel and human resource man- agement, Supplement 3. International Human Resources Management. Greenwich, CT: JAT Press.

Zhao, Y. (1995). Chinese motivation theory and ap- plication in China: An overview. In H.S.R. Kao, D. Sinha, & S.H. Ng (Eds.), Effective organisations and social values (p. 127). New Delhi: Sage.

Zhu, C.J. (1997). Human resource development in China during the transition to a new economic system. Asia Pacific Journal of Human Resources, 35(3), 19–44. ENDNOTES 1. The indictment of IHRM as atheoretical could well be applied to many fields of research; how- ever, for illustrative purposes, we discuss this in the context of IHRM.

2. In traditional SHRM research, the discussion might revolve around horizontal or vertical fit, or alignment of HR practices with internal or external factors; however, fit has dominated the landscape. In practice most research has been focused on a simple interaction between two vari- ables at a time, i.e., pay and business strategy (Becker & Gerhart, 1996). 140 • H UMAN RESOURCE M ANAGEMENT , Spring 2002 3. While it is true that valid and reliable measures are a necessary condition for comparison be- tween groups, validity must be established only once, and this is generally established using measures developed in North America on North American populations. If the reliability coeffi- cients are less than desirable, or less than ideal, researchers fall back on the validity reported elsewhere. This gives rise to the “one way fits all” modality in most international research ini- tiatives, and this is what we wish to challenge in the ensuing argument.4. Brewster’s contention was that most research on IHRM automatically assumed that what- ever practices were in place, they must have been invented in the U.S. and were therefore ubiquitous. He challenged this with his Cranfield project, which suggested that not all IHRM practices were in fact invented in the U.S.!

5. “Traditional” here defined as using standardized instruments developed in the North American context, and examining a few select “HR prac- tices” in two or three countries.