HRMperformance link in the Eritrean civil service Mussie Teclemichael Tessema and Joseph L. Soeters Abstract In this article, the authors examine how, when and to what extent HR practices affect performance at the employee level. As performance is a multi-faceted and complicated concept, HRM outcomes were used as mediating factors between HR practices and employee performance. The data were collected among civil servants in Eritrea, Africas youngest and poorest country. Although the results generally are in line with previous studies using Western data, their implications in this particular country may be different. Therefore, the challenges and prospects of HR practices in Eritrean civil service organizations are critically analysed and discussed. In the authors opinion, that the Eritrean economic and political environment within which HR practices operate has not been conducive in maximizing the impact of HR practices on performance. These ndings highlight the situation of most developing countries. Keywords Human resource management; HRMperformance link; civil service organizations; Eritrea; developing countries. Introduction Every organization, whether it be a public, private or NGO, must operate with and through people. Public organizations in particular are judged on the basis of the performance of their human resources. Ingraham and Kneedler (2000: 245) underline that government activities are typically highly personnel intensive. And thus, Human Resource Management (HRM) practices are central to improving the quality of services offered by the governments. In the words of Pfeffer (1994: 33), having good HRM is likely to generate much loyalty, commitment, or willingness to expend extra effort for the organizations objectives. Moreover, Stone (1998: 4) remarks that HRM is either part of the problem or part of the solution in gaining the productive contribution of people. The above quotes suggest that organizations need to effectively manage their human resources if they are to get the maximum contribution of their employees. But the question is, how and when does HRM affect performance? Recently, the dominant focus The International Journal of Human Resource Management ISSN 0958-5192 print/ISSN 1466-4399 online q 2006 Taylor & Francis http://www.tandf.co.uk/journals DOI: 10.1080/09585190500366532 Mussie Teclemichael Tessema, Tilburg University (IVO), the Netherlands and University of Asmara, Eritrea: POBox 90153, 5000LE, Tilburg, The Netherlands (e-mail: M.T.Tessema@uvt.nl). Joseph L. Soeters, Tilburg University and Royal Netherlands Military Academy, the Netherlands: PO Box 90002, 4800 PA Breda, the Netherlands (jmlm.soeters@mindef.nl). Int. J. of Human Resource Management 17:1 January 2006 86105 on HRM literature has been to demonstrate the importance of effectively managing human resources of organizations. Management scholars and practitioners alike have become increasingly interested in learning more about HR practices to enhance employee and organization performance (e.g. Boselie et al., 2001; Den Hartog and Verburg, 2004; Ferris et al., 1999; Guest, 1997; Huselid, 1995; Paauwe, 1998, Power and Boselie, 2003; Pfeffer, 1994). In particular, the last ten years have seen an increasing research interest in the HRMperformance relationship, although the focus of the research is in the developed world as well as in the manufacturing industry. The current study aims to test empirically the impact of eight HR practices on employee performance in a developing country: the Eritrean civil service organizations. Eritrea, being the youngest African nation, became a sovereign nation in 1993. Since 1993, Eritrea has embarked on a multifaceted nation-building and reconstruction process in which the civil service is one aspect. The critical challenge that faces Eritrea today is the establishment of economic, social, administrative and political institutions and the development and utilization of human resources to enable these institutions to operate effectively (Gafer, 1996; Haregot et al., 1993; UNDP, 2002; UOA, 1997). Thus, Eritrea today is experiencing a growing need for civil servants who are capable of efciently, effectively and creatively mobilizing the available scarce resources to achieve national objectives. Eritrea, just like most developing countries (DCs), introduced civil service reforms (19957) that led to (1) streamlining of about 34 per cent of the Eritrean civil servants (UOA, 1997); (2) the establishment of the Eritrean Institute of Management (EIM) in 1995; (3) the introduction of a new salary scale in 1997; and (4) the launching of the Eritrean HRD Project (19982003) (EHRDP, 2003). However, the critical challenge that faces Eritrea today is the utilization of human resources to enable civil service organizations to operate effectively. Generally speaking, the environment within which Eritrean civil servants are employed does not seem to attract, motivate and retain competent civil servants. In this connection, the availability and utilization of capable civil servants is of utmost importance. The main research questions of this study are therefore: . How, when and to what extent do HR practices affect HRM outcomes (HR competence, motivation, role clarity and retention)? . How do HRM outcomes in turn affect employee performance in the context of Eritrean civil service? Literature review Effective HRM now more than ever before is a crucial ingredient in the development process of DCs. However, HRM has come under strong criticism in many DCs with their effectiveness thrown into considerable doubt (Bennell, 1994; Budhwar and Debrah, 2001; Hilderbrand and Grindle, 1997; Kiggundu, 1989; Praha, 2004; World Bank, 1994b). Decades of declining real incomes, deplorable working conditions, political interference and poor management have created cadres of civil servants in many DCs who are chronically demoralized and de-motivated (e.g. Das, 1998; Jaeger et al., 1995; Kiggundu, 1989). The civil service, on the one hand, is increasingly unable to retain trained personnel wherever other employment opportunities exist. On the other hand, it utilizes poorly the expertise of those civil servants who do not leave. Even worse, moonlighting and corrupt rent-seeking practices have become a way of life for civil Teclemichael Tessema and Soeters: HRM in developing countries 87 servants in many DCs (Bennell, 1994; Budhwar and Debrah, 2001; Das, 1998; Grindle, 1997; Prah, 2004). More than anything else, thus, it is the personnel crisis in the civil service organizations in DCs that has to be addressed if meaningful improvements in service delivery are to be realized. Cohen and Wheeler (1997) as well as Hilderbrand and Grindle (1997) summarized the current situation of HRM in many DCs as follows: low salary levels, lack of effective performance standards, inability to re people, too few rewards for good performance, recruitment procedures that do not attract appropriately trained people, promotion patterns based too much on seniority or patronage and too little on performance, slow promotion and lack of reward for hard work and initiative, inadequate and demoralizing management by supervisors (ineffective leadership), underemployment and lack of stimulating assignments. Moreover, Bennell (1994) and Budhwar and Debrah (2001) disclosed that many DCs are trapped by outdated and ineffective HRM systems that put unintended roadblocks in the way. The situation discussed above being the existing situation of many DCs, one may wonder how HR practices affect performance under such conditions. Recently, the dominant focus on HRM literature has been to demonstrate the importance of effectively managing human resources of organizations (e.g. Ahmad and Schroeder, 2003; Delaney and Huselid, 1996; Ichniowski et al., 1997). Many scholars have identied a number of HRM-related practices that greatly affect performance. For example, Pfeffer (1994) advocated the use of 16 HRM-related practices to achieve higher performance; Delery and Doty (1996) identied seven HRM-related practices. More recently, researchers have found that bundles, or systems, of HR practices had more inuence on performance than individual practices working in isolation (Arthur, 1994; Baron and Kreps, 1999; Huselid, 1995; MacDufe, 1995). In other words, a greater use of those specied practices results in higher performance across all types of organizations or countries. As will be discussed, in this study, we used eight HR practices that we assume affect performance. Performance, as underlined by Paauwe (1998) and Guest (1997), is a multifaceted and complex phenomenon. It is difcult to clearly know to what extent HRM affects performance. When it comes to measuring public sector performance, the problem is even more difcult (Guest, 2001: 1100; Hays and Reeves, 1984: 290; Ozgediz, 1983: 59). Performance is, therefore, not the direct result of any one factor such as HRM. Rather, HRM is only one important component of a diverse set of inuences that determine performance level. If we are to speak with any certainty about the extent (net effect) to which HRM affects performance, one would rst need to isolate HR practices effects by controlling the rest of the variables bearing on performance. Only when we have made progress in measuring the independent and dependent variables can we begin to give full attention to the way in which they are linked (Guest, 1997: 274). As remarked by Legge (2001: 30), there is a need to open up the black box of the process that links HRM and organizational performance. This is mainly due to the existence of intervening variables. Unless very careful controls are used to take account of all factors affecting performance, it is possible that the results may overstate or understate the inuence of HRM on performance (Fey et al., 2000; Guest, 1997; Wright et al., 2003). What has so far been achieved is a skeletal nding and we need to put a lot of esh on the bones (Boxall, 2003; Guest, 1997; Paauwe, 1998). Although our knowledge of the net impact of HRM functions on performance is still incomplete, there is no shortage of assumptions and theories regarding the role that HRM plays in positively affecting performance. In other words, despite the absence of unambiguous proof or net impact of HRM on performance, there is evidence that a 88 The International Journal of Human Resource Management positive relationship does exist between the two (e.g. Becker and Gerhart, 1996; Huselid et al., 1997; Wright et al., 2003; Youndt et al., 1996). Recently, there has been a growing body of research that seeks to examine the impact, if any, of HR practices on organizational performance. The most notable studies are those that use large data sets and interrogate data using sophisticated statistical techniques. In general, the available studies appear to reveal impressive evidence of robust impacts and outcomes. The now classical studies in the United States include those by Becker and Gerhart (1996); Huselid (1995); Huselid et al. (1997); Icheniowski et al. (1997); and MacDufe (1995). In Britain, the major studies include those by Guest (1997) and Patterson et al. (1997). In the Netherlands, studies include those of Boselie et al. (2001); Paauwe (1998); and Den Hartog and Verburg (2004). Recently, studies have begun to use HRM outcomes such as employee competence, motivation and role clarity in explaining the HRMperformance relationship. As can be seen from Figure 1, HR practices lead to HRM outcomes, which subsequently affect employee and organization performance (e:g: Fey et al., 2000; Guest, 1997; Paauwe, 1998). The assumption here is that HRM outcomes serve as mediating variables between HRM and performance. HR practices give rise to HRM outcomes, which will inuence performance in and of the organization (Paauwe, 1998: 5). Basically, there are a number of theoretical arguments (e.g. expectancy theory, resource-based theory, human capital theory) supporting the idea that HR practices affect performance (e.g. Guest, 1997; Paauwe, 1998; Pauuwe and Boselie, 2003; Youndt et al., 1996). These theoretical arguments provide some insights into howHRpractices translate into higher performance. Thus, the logic connecting the HR practices and performance is intuitively appealing and supported by theoretical arguments froma number of disciplines. Below, a brief discussion is given on the theoretical arguments of expectancy theory mainly due to the fact that there is a growing consensus that the concept of expectancy theory can provide one possible route to an explanation of how HR practices have an impact on performance (Becker and Gerhart, 1996; Guest, 1997; MacDufe, 1995). Expectancy theory assumes that if an employee is to be productive, three elements must be in place: competence, motivation and role clarity (MacDufe, 1995). Put somewhat differently, it has been proposed that HRM outcomes (employee competence, motivation and role clarity) mediate the relationship between HR practices and employee Figure 1 Conceptual framework for analysing the challenges and prospects of effective HRM in civil service organizations Teclemichael Tessema and Soeters: HRM in developing countries 89 performance. Employee performance, in turn, positively affects performance at the organization level. However, given the situation of most DCs, like Eritrea, facing the problem of retention of professionals, our conceptual framework adds employee retention as the fourth HRM outcome that affects performance (see also Aredo, 2002; Fey et al., 2000; Hilderbrand and Grindle, 1997; World Bank, 1994b, 1997). The argument is that if civil service organizations are to provide the needed quality of services (high performance), they must be able to retain professionals (competent employees who are willing to stay and do not have the intention to leave). As contended by Taormina (1999: 1060), employee turnover has high potential to negatively affect an organization since the loss of trained employees would mean a reduction in organization performance. Employee retention is not a frequently cited HRM outcome in the developed world as contrasted with the situation in most DCs. Employee retention may be particularly important in DCs like Eritrea in which, on the one hand, the country has been experiencing acute shortages of qualied and experienced human resources and, on the other hand, a high brain drain of qualied civil servants exists (Aredo, 2002; Das, 1998; ECA, 1989; Kiggundu, 1989, 1998; ILO, 1998). That is why, recently, it has been argued that it is a paradox of todays world that the DCs (such as Eritrea) that have the greatest need to use professionals to help them along the road to economic and social development are not using them efciently and in fact are losing many of their best educated young men and women to the developed societies. To facilitate understanding, different parts of the conceptual framework (Figure 1) are presented in the following manner: environmental factors such as economic and political (A1A2) affect HR practices. HR practices (B1B8) affect HRM outcomes (C1C4), which subsequently affect employee performance (D). The conceptual framework reveals that HRM outcomes serve as mediating variables between HRM and performance. Our conceptual framework provides a clear chain of causation. It is assumed that other things being equal, employees who are motivated are more effective and productive than apathetic employees; competent employees are more effective and productive than incompetent employees; employees with clear duties and responsibilities are more productive than employees with role ambiguity and confusion; and employees who do not have an intention to leave are more productive than those who have an intention to leave the organization. Thus, the higher the above HRM outcomes in place, the better HR contribution would be (e.g. Becker and Gerhart, 1996; Den Hartog and Verburg, 2004; Guest, 1997, 2001; Paauwe, 1998). Given the scheme in Figure 1, one may ask which HR practices impact on the aforementioned HRM outcomes: employee competence (ability to work), motivation (willingness to work), role clarity (clear duties and responsibilities), retention (willingness to stay)? A considerable amount of research has been done concerning the relationships between HR practices and the above-mentioned HRM outcomes. Although, the research so far conducted is unable to recognize clearly the net impact of each HR practice on HRM outcomes, many of the studies found that different HR practices are positively correlated with the HRM outcomes. For example, stable connections have been found between compensation programmes and motivation/ job-satisfaction (e.g. Huselid, 1995; Lienert, 1998), and between promotion opportunities and motivation (e.g. Ahmad and Schroeder, 2003; Hilderbrand and Grindle, 1997). Others have also found unequivocal relations between investment in training and employee competence (e.g. Kalleberg and Moody, 1994), between recruitment and selection and employee competence (e.g. Hsu and Leat, 2000; Huselid, 1995), between job descriptions and placement and role clarity (e.g. Becker and Gerhart, 1996; Fey et al., 2000) and between compensation and retention (Hilderbrand and Grindle, 90 The International Journal of Human Resource Management 1997; Aredo, 2002; Taormina, 1999). The above-mentioned studies, therefore, support our main argument stating that HRM outcomes can serve as mediating variables between HR practices and performance. In this study, we empirically tested the impacts of eight HR practices on HRM outcomes. We also tested the impacts of both HR practices and HRM outcomes together on performance, as perceived by the respondents. Research methodology This study is based on a sample of ten Eritrean ministries. The data were collected using two methods: 1 A questionnaire distributed to 400 civil servants: in this survey-study, a multi-stage sampling procedure was used: rst, ten out of 16 ministries were randomly included in the sample; next 40 civil servants from each ministry were randomly selected. In order to create a representative sample, the respondents were equally drawn from the following ve services: administrative service, professional and scientic service, sub-professional service, clerical and scal service, and crafts and manual service. All in all, 400 questionnaires were distributed, of which 313 usable questionnaires were returned, leading to a response rate of 78 per cent. In total, 23 per cent of the respondents were from the administrative service, 24 per cent of them were from the professional and scientic service, 25 per cent of them were from the sub-professional service, 16 per cent of them were from the clerical and scal service, and the remaining 12 per cent were from the crafts and manual service. Hence, the non-response rate is somewhat larger in clerical/scal as well as crafts and manual services. The questionnaires were developed in the English language and subsequently translated from English into the local language, Tigrigna. Back-translation was not deemed necessary, since the translation was done by the rst author who has mastered both languages. Questionnaire data were collected in April and May 2003. 2 Interviews with ordinary civil servants and management bodies: specically, in-depth interviews were conducted with: (a) ten HR managers, (b) ten HRD managers, (c) the Eritrean Civil Service Commissioner, (d) the Director General of the Eritrean HRD Project, and (e) 50 ordinary civil servants. As to the process of conducting the interviews, rst we identied relevant concepts from the literature to be used as a signpost for us in order to focus the interviews on the most important issues. The interviewees responded in the local language to facilitate communications and we then transcribed the interviews into English on the same day the interview was conducted while observations and information were fresh in our memory. The transcribed interviews were read and the essential issues reported were annotated. The questionnaire we administered contained eight HR practices, which included the following items: recruitment and selection practices with ve items, placement practices with three items, training practices with six items, compensation practices with six items, employees performance evaluation practices with six items, promotion practices with three items, grievance procedures with three items, pension programme (social security) with three items. In addition, four items with regard to HRM outcomes and three items with regard to performance as perceived by the respondents were included (see Appendix 1 for the specic measurements). The model therefore consists of eight HR practices as predictor variables and HRM outcomes as well as performance as dependent variables. Respondents were asked to indicate the degree to which they agreed Teclemichael Tessema and Soeters: HRM in developing countries 91 or disagreed with the items related to the above HRM issues (see Appendix 1). All the items were measured on a ve-point scale ranging from 1, strongly disagree, to 5, strongly agree (see Table 1 for explanation and scale reliability). The analysis also includes three control variables. Except for the variables tenure and monthly salary, which are simply entered in the equation as continuous variables, gender is measured with a dummy variable. Table 1 portrays the alphas to check the reliability of the eight HR practices and HRM outcomes as well as the perception of performance. All alphas range from .71 to .92, which can be considered satisfactory. We also performed ANOVA tests to discover whether there are differences among the ten sample ministries; the tests, however, demonstrate that the perceptions of the respondents are found to be statistically not signicant among the ten sample ministries. This is mainly due to the fact that the ten sample ministries are part and parcel of the Eritrean Ministries (civil service organizations) operating under fairly similar conditions. Results As Table 1 discloses, the overwhelming majority of the variables were rated fairly low (mean values ranging between 2.1 and 2.9 on a ve-point scale), which in turn signals the magnitude of the problems associated with HR practices, as will be discussed later. Table 2 presents the correlations between the variables included in the analysis. There are a number of itemcorrelations; hence, potential multicollinearity had to be considered. We computed variance ination factors (VIFs), which indicated that multicollinearity was not a problem for the variables in the regressions whose results appear in Tables 3 and 4. As predicted, with the exception of placement practices, the other seven HR practices are signicantly positively correlated with HRM outcomes (see Table 2). Such correlations may disclose that HR practices affect HRM outcomes, as demonstrated in our model (see Figure 1). Overall, the pattern of results reported in Table 2 is consistent with our studys underlying theoretical premise, that HR practices are positively correlated with HRM outcomes (see also Ahmad and Schroeder, 2003; Becker and Huselid, 1998; Guest, 2001; Ichniowski et al., 1997; Paauwe, 1998). One of the goals of the present study was to test to what extent the eight HR practices explains the HRM outcomes, which are the mediating variable between HR practices and performance. To that end, we performed a regression analysis (see Table 3). Table 3 demonstrates that ve out of the eight HR practices identied in this study show a statistically signicant positive impact in explaining the change in HRM outcomes and are greater than or equal to b .10. The ve signicant HR practices are recruitment and selection practices, training practices, compensation practices, grievance procedures, and pension or social security programmes. In addition, the eight HR practices altogether explain about 54 per cent of the change in HRM outcomes (R 2 .54). Generally speaking, we found reasonably good support for the propositions of our model. This may suggest that the more the HR practices are in place, the more HRM outcomes would be up to standard, which subsequently would positively affect performance (e.g. Guest, 1997; Ichniowski et al., 1997; MacDufe, 1995; Paauwe, 1998; Wright et al., 2003). Hence, we have found that the more HR practices are in place, the more the civil servants are competent, satised with the existing HR practices, have sufcient role clarity in their job and have no intention to leave the organization. These ndings are in line with the predicted relationships and thus support our model. Results presented in Table 3 also show that compensation and pension practices are the major contributors to the HRM outcomes (b . 24 and .21). One of the main reasons 92 The International Journal of Human Resource Management Table 1 Descriptive statistics and Cronbachs alphas Variable in model High value means Alpha Mean Min. Maxi. Recruitment and selection practices (5 items) Strongly agree that effective recruitment & selection practices exist (.75) 2.7 1.2 5.0 Placement practices (3 items) Strongly agree that effective placement practices exist (.92) 2.7 1.0 5.0 Training practices (6 items) Strongly agree that effective training practices exist (.75) 2.7 1.0 5.0 Compensation practices (6 items) Strongly agree that effective compensation exists (.82) 2.4 1.0 4.5 Employee performance evaluation practices (6 items) Strongly agree that effective employee performance evaluation practices exist (.73) 2.9 1.0 4.3 Promotion practices (3 items) Strongly agree that effective promotion practices exist (.74) 2.4 1.0 4.7 Grievance procedure (3 items) Strongly agree that effective grievance procedures exist (.72) 2.7 1.0 5.0 Pension programme (3 items) Strongly agree that effective pension programme exists (.86) 2.1 1.0 4.0 HRM outcomes (4 items) Strongly agree that high HRM outcomes exist (.71) 2.6 1.0 4.5 Performance (3 items) Strongly agree that respondents own performance is relatively good (.74) 2.9 1.0 5.0 Gender a M/F Tenure Years of work experience 17.2 1 43 Salary b Monthly salary level in Nackfa 1463 245 3020 Notes: a 33% were female; b monthly salary is in terms of Nackfa. US$1 13.5 Nackfa as of September 2003. T e c l e m i c h a e l T e s s e m a a n d S o e t e r s : H R M i n d e v e l o p i n g c o u n t r i e s 9 3 Table 2 Correlations matrix No Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 1 Gender 2 Tenure 2.045 3 Salary 2.24** .27** 4 Recruitment and selection practices .13* 2.05 2.10 5 Placement practices .00 .04 .06 2.07 6 Training practices .08 .03 2.10 .63** 2.15* 7 Compensation practices .05 2.08 2.12 .41** .06 .42** 8 Employee performance evaluation .04 .11 .01 .43** 2.10 .52** .43** 9 Promotion practices .09 .06 2.10 .55** 2.04 .72** .49** .47** 10 Grievance procedure .07 2.08 2.10 .46** 2.01 .50** .59** .37** .50** 11 Pension programme .04 .03 2.11 .50** 2.08 .49** .52** .45** .55** .43** 12 HRM outcomes .10 .01 .00 .57** 2.06 .59** .59** .48** .56** .54** .58** 13 Performance level .15** .04 2.07 .57** 2.09 .66** .38** .65** .48** .42** .45** .66** Notes: **Correlation is signicant at the 0.01 level; and *correlation is signicant at the 0.05 level (two-tailed). 9 4 T h e I n t e r n a t i o n a l J o u r n a l o f H u m a n R e s o u r c e M a n a g e m e n t why the contributions of compensation and pension practices are found to be the highest is that, given the poor economy and volatile environment of most DCs including Eritrea, one may expect a greater impact of these two HR practices on HRM outcomes. As previously discussed, compensation issues affect most HR practices or functions either directly or indirectly. For instance, compensation affects employee motivation (Huselid, 1995; Lienert, 1998) and retentionintention to stay in the organization (Becker and Gerhart, 1996; Grindle, 1997; Pfeffer, 1994). In model 2 in Table 3, when gender, tenure and salary are added, the R 2 change that we found was only .01. However, in connection with the importance of compensation Table 3 Results of regression analysis on HRM outcomes Variables Model 1 a Model 2 a Recruitment and selection practices .17 ** .16 ** Placement practices .00 2.02 Training practices .20 ** .20 ** Compensation practices .24 *** .25 *** Employee performance evaluation practices .06 .05 Promotion practices .01 .00 Grievance procedure practices .10* .10* Pension practices .21 *** .22 *** Gender .06 Tenure .00 Salary .11 ** F 44.0 *** 33.3 *** R 2 .54 .55 R 2 change .01* Notes: a Standardized regression coefcients are reported; * p , .05; ** p , .01; *** p , .001; n 313. Table 4 Results of regression analysis on performance a Variables Model 1 b Model 2 b Model 3b Recruitmentment and selection practices .19 *** .18 *** .12 * Placement practices .03 .03 .03 Training practices .37 *** .37 *** .30 *** Compensation practices 2 .03 2 .03 2 .12* Employee performance evaluation practices .40 *** .40 *** .38 *** Promotion practices 2 .12* 2 .12* 2 .12* Grievance procedure .06 .06 .02 Pension or social security .07 .07 2 .01 Gender .08* .06 Tenure .00 .00 Salary 2 .01 2 .05 HRM outcomes .35 *** F 49.1 *** 36.4 *** 42.0 *** R 2 .56 .57 .63 R 2 change .01 .06 *** Notes: a Performance as perceived by the respondents themselves. b Standardized regression coefcients are reported; *p , .05; ** p , .01; *** p , .001; n 313. Teclemichael Tessema and Soeters: HRM in developing countries 95 practices, salary was also found to be signicant, which in turn suggests that compensation-related issues considerably affect HRM outcomes such as employees motivation and retention, as already mentioned. As indicated in the introduction, we have gone further to inspect the impact of the eight HR practices plus HRM outcomes together on performance as perceived by the respondents (see Table 4). Table 4 also depicts some important results: three of the eight HR practices show a statistically signicant positive impact in explaining the change in performance, namely recruitment and selection practices, training practices, and employee performance evaluation practices. In addition, the eight HR practices altogether explain about 56 per cent of the change in performance (R 2 .56). As revealed in Table 4, employee performance evaluation is found to be a major contributor to performance (b .40, p , 0.001). This may be because employee performance evaluation is crucial for most HR practices in that, if it is done properly, it could serve several purposes: rst, for taking proper personnel actions such as salary increment, promotion and other incentives; second, for assessing the strengths and weaknesses of employees and taking the appropriate training and development measures; and third, for assessing the effectiveness of HR practices such as recruitment, selection and training (Baron and Kreps, 1999; Bowen and Ostroff, 2004; Kalleberg and Moody, 1994; Pfeffer, 1994). An interesting nding of our study is that, although we expected that the impact of HR practices would weaken as we go to the right side of our conceptual framework i.e. towards performance D in Figure 1 (Fey et al., 2000; Guest, 1997; Legge, 2001; Paauwe, 1998) we found that the eight HR practices altogether explain 56 per cent of the change in performance (R 2 .56). Thus, contrary to our expectation, the impact of the eight HRpractices on performance (R 2 .56) exceeds that of HRMoutcomes (R 2
.54) (see Tables 3 and 4 and Figure 1).
In model 3 in Table 4, when gender, tenure, salary and HRM outcomes are added, the R 2 change that we found was .06, which is signicant at p , .001. Besides, HRM outcomes are found to show a considerable positive impact, which in turn may suggest the importance of HRM outcomes in affecting performance. All these results conrm our expectations based on theoretical insights. Clearly, the HRMperformance link that has been demonstrated in the Western hemisphere also exists in the Eritrean context. However, there is one exception. In all three models of Table 4 we found signicant negative correlations to exist between the existing promotion practices and employee performance. This nding suggests that the respondents with a (claimed) better performance record are less convinced that the existing promotion practices are based on written, formal policies and that priority is given to merit and seniority when decisions on promotions are taken (see Appendix 1). Obviously, those respondents feel that decisions on promotions are too much inuenced by other, mainly political considerations, which is a well-known point of concern in the Eritrean civil service (e.g. Soeters and Tessema, 2004; Tessema and Soeters, 2005). Discussion One of the main goals of this study has been to discuss the HRM challenges facing civil service organizations in DCs such as Eritrea. To that end, we have presented the impacts of HR practices on HRM outcomes, which subsequently affect performance at the employee level. Again we have to stress that our results have demonstrated a strong HRMperformance link in the Eritrean civil service. As such, our study aligns to a large degree with other studies on the HRMperformance link in Western countries. However, 96 The International Journal of Human Resource Management the situation is not fully comparable to the situation in Western parts of the world. In this section, we need to go further to relate the empirical ndings to the actual economic and political environment of Eritrea. One important issue is the role of the environmental factors (economic and political), which are decisive in inuencing the impact of HR practices on the performance of civil servants in DCs (see also Austin, 1990; Budhwar and Debrah, 2001; Grindle, 1997; Jaeger et al., 1995; Kiggundu, 1989; Wasti, 1998). Environmental factors, according to Kiggundu (1989: 75), can be very instrumental either in facilitating or in hindering the operation of an organization. However, as previously described, the Eritrean economic and political environment within which the eight HR practices operate has not been conducive in maximizing the impact of HR practices on performance. As a consequence, these eight HR practices were rated fairly low (between 2.1 and 2.9 on a ve-point scale) by the respondents. Most DCs are known for their economic problem in nancing different projects and programmes, including HRM programmes (e.g. Freeman and Lindauer, 1999; Jaeger et al., 1995; Kiggundu, 1989; Prah, 2004). Almost all of the senior civil service managers interviewed pointed out that because of the deteriorating Eritrean economy (see also UNDP, 2002; National Statistics Ofce, 2003), they were unable to allocate the required resource for the eight HR (mainly compensation) practices. This demonstrates that, in a period of budget constraints, the ability to pay has become the key criterion in pay determination, overshadowing any broader conceptions of comparability. For instance, according to the CPA (2002a), as of 2002, employees of international NGOs and the private sector were earning a salary that is about three times higher than that of the civil servants. This implies that there is a wide gap between the public and international NGOs and private-sector pay, in favour of the latter. The existing compensation practices have been adversely affecting most HR practices (see also Amanuel and Tesfagabir, 2002; CPA, 2002b; Soeters and Tessema, 2004). The prevailing situation has made it increasingly difcult for Eritrean civil service organizations to attract qualied and suitable applicants, which in turn affects their ability to be selective in their hiring exercise. Moreover, it has adversely affected the impact of training in that although many civil servants have been given training opportunities by donor agencies and countries, they have been unable to utilize their expertise effectively, mainly due to the unattractive compensation programmes. As a consequence, civil servants tend, once they have obtained better qualications through training, to move over to the private sector or International NGOs or overseas. This phenomenon is widespread in numerous developing countries (e.g. Cohen and Wheeler, 1997; Das, 1998; Grindle, 1997; ILO, 1998; Tessema et al., 2005; World Bank, 1994b). This is mainly due to the fact that if training is not supplemented with necessary incentives, it may not increase employees motivation as well as their commitment to the organization. Prah (2004: 3) argues that in most SSA countries, middle-level ofcials can still barely feed, let alone adequately house, clothe and educate their family. It was also underlined that temporary solutions like increasing salaries on a project basis or earmarking foreign assistance to enhance remuneration were not found to be sustainable for the aforementioned problem (Das, 1998; Maggregor et al., 1998). Unattractive compensation has been important in undermining the previously existing capacity (see also Cohen and Wheeler, 1997: 137; Hilderbrand and Grindle, 1997: 42). The prevailing economic context, within which HRM is taking place in most DCs like Eritrea, has therefore adversely affected the impacts of HR practices (Analoui, 1998; Lienert, 1998). Furthermore, like many DCs, the Eritrean political environment does not seem to maximize the impact of the eight HR practices. It has been underscored that organizations in developing countries often face a highly volatile and unstable political Teclemichael Tessema and Soeters: HRM in developing countries 97 environment (see also Austin, 1990; Kanungo and Jaeger, 1990; Kiggundu, 1989). For example, according to Kiggundu (1989: 62), many developing countries are characterized by unexpected political changes, which create so much environmental uncertainty and complexity that managers consider almost any planning or strategic management responses impossible. DCs in general and sub-Saharan African (SSA) countries in particular are characterized by low predictability of events, unstable political climates and corrupt legal practices (see also Chabal and Daloz, 1999; Prah, 2004; Smith, 2003). Continuity of government policies is also hindered by frequent changes in ruling groups or individual government ofcials. In some instances, these political groups strive to gain and consolidate their power through coercive means. Patrimonial tendencies are strong (Beugre and Offodile, 2001; Chabal and Daloz, 1999). Patrimonialism grew out of the single-party, military or one-man-rule states, which have characterized many DCs. By and large, the consequences of political instability for HR practices are signicant. Indeed, it is frequently cited as one of the main reasons for qualied civil servants to migrate to the West (ILO, 1998; Aredo, 2002). It is also interesting to note that the civil service of many DCs has been politicized. As noted by Das (1998: 19), politicization of the civil service has resulted in total erosion of traditional civil service values such as political neutrality, probity, rectitude, and objectivity. Interviews conducted with the civil servants uncovered that the politicization of the Eritrean civil service has affected many HR practices, such as recruitment and selection, placement and promotion, and compensation. For instance, about 34 per cent of the interviewees believed that recruitment and selection practices are inconsistent with modern principles such as open, fair and merit based. That is why many scholars noted that favouritism, nepotism and political loyalty have been affecting recruitment, selection and promotion practices in many developing countries (e.g. Bennell, 1994; Beugre and Offodile, 2001; Das, 1998; Soeters and Tessema, 2004; World Bank, 1994b). About 38 per cent of the interviewed civil servants also believed that the existing environment diminished the link between performance and reward, which in turn suggests that employee performance evaluation is often largely based on compliance rather than productivity. The nding of our study also reects the general situation of the civil service in most developing countries (e.g. Analoui, 1998; Bennell, 1994; Beugre and Offodile, 2001; Soeters and Tessema, 2004). As a result, many HRM decisions have been taken subjectively. As Waiguchu (Waiguchu et al., 1999: 198) puts it, in an authoritarian setting, an appraisal system is unavoidably one-sided. In such a case, the supervisors view of the subordinates performance prevails. Interviewed senior civil service managers noted that placement and promotion to upper middle and top civil service positions are very much politicized in that the majority of the senior positions in the Eritrean Civil Service (Minister, General Director, Director, and Unit Head) are held by individuals who are loyal to the government. Although a civil service is composed of both political appointees and career civil servants, of whom the political appointees are few in number and occupy top or most senior positions (Heady, 1996), the Eritrean case reveals that the political appointees have even been holding positions that conventionally are occupied by career civil servants (see also Pool, 2001; Soeters and Tessema, 2004). Hence, non-merit considerations have been very much affecting promotion decisions, which in fact we have shown in our regression analyses (see Table 4). Up to now, there is no clear dividing line between administration and politics. As Heady (1996) argues, some systems are civil service in name only and actually function as spoils systems. That is, the governments personnel systems may be nominally merit based but practically politically based. 98 The International Journal of Human Resource Management The foregoing brief discussion demonstrates that in an environment where there are no promising economic and political factors (peace and stability as well as good governance), it is unlikely that a higher impact of HR practices on employee performance will be achieved. For instance, the Eritrean case shows that, although many scholars and government leaders like Bill Clinton cited it as a beacon of Africa, after seven years of economic growth and political stability, it was again involved in a border war with Ethiopia, which in turn has adversely affected the economy as well as the political condition of the country, which further diverted the governments priority to national defence. It should be noted that the border war between Ethiopia and Eritrea (1998 2000) has affected either directly or indirectly most HR practices. For instance, during the border war, about 300,000 Eritreans were mobilized, that is, about 10 per cent of the entire population or more than 50 per cent of the population of working age (Amanuel, 2002). Thus, the war had made the normal continuation and effective implementation of many HR practices a very difcult task indeed. One can argue that such an environment is likely to adversely affect the impact of HR practices on performance. Heady (1996: 217), after analysing the management of civil servants in developing countries, underlines that: In single or dominant party political systems, party claims to a monopoly on policy making and executive make it imperative that civil servants pass whatever test of party loyalty may be imposed and that they suffer not only loss of status or position for falling short, but suffer in other ways as well, including imprisonment, banishment to the countryside, or death . . . in such circumstances, the behaviour of civil servants who want to remain in the civil service is overwhelmingly determined by what is required of them by the current party line as to their roles. What Heady underlines has great relevance to the actual situation of many DCs and particularly that of SSA (including Eritrea) where the administrative crisis is most critical and the enforcement of personnel rules and procedures has been broken down (e.g. Chabal and Daloz, 1999; Prah, 2004; Smith, 2003). For the above reason, one may argue that the rational, hierarchical and meritocratic model has failed in many DCs. One can also suggest that there is nothing wrong with the civil servants of developing countries, rather there is a problem in the management of the civil servants. The existing situation has also led to widespread absenteeism, petty corruption, moonlighting (to use ofce hours and equipment for private purposes), and a general breakdown in morale and discipline (see also Lienert, 1998: 43; Prah, 2004: 3; World Bank, 1997: 95). One may then argue that civil service organizations in developing countries such as Eritrea, are not prestigious places that can attract, motivate and retain qualied and experienced personnel (see also Das, 1998; ECA, 1989; Hilderbrand and Grindle, 1997). The question is, what would be the prospects of HRM in the Eritrean civil service organizations as well as those of other developing countries? One can reasonably argue that the prospects of effective HRM would be contingent mainly upon the countrys economic and political condition. That is, if there are improving economy as well as political conditions, there is a high probability of successfully putting into effect all HR practices thereby positively affecting performance (Dia, 1996; World Bank, 1997). As previously indicated, given the above-stated environmental factors, the impact of the eight HR practices on HRM outcomes in the Eritrean civil service organizations is more or less in line with the predicted relationships (see Tables 1 to 4). This may suggest that our ndings support the underlining premise of expectancy theory in particular and other theories such as human capital theory and resource-based theory in general (e.g. Paauwe and Boselie, 2003; Youndt et al., 1996). In particular, the ndings of this study support the idea advocated by expectancy theory in that HRM outcomes could be mediating variables between HR practices and performance at the employee level. Thus, Teclemichael Tessema and Soeters: HRM in developing countries 99 an important nding of this study is that the assumption of expectancy theory may work not only in the developed world but also in developing countries such as Eritrea. Our ndings can also be viewed in the context of the growing body of research on HRM and performance relationship (e.g. Ahmad and Schroeder, 2003; Delaney and Huselid, 1996; Guest, 1997, 2001; Ichniowski et al., 1997), and the role of the economic and political environment in that respect. The present study can be seen as adding to this literature in that, unlike most prior research, we empirically tested the model in the public sector as well as in the context of DCs. This is because prior research has relied mostly on manufacturing industry and other parts of the world. In this regard, the key contributions of the present research are: rst its attempt to open the black box (the process aspect) by answering the following questions: how, when and to what extent do HR practices affect HRM outcomes, which subsequently affect performance? (Becker and Huselid, 1998; Guest, 1997, 2001; Legge, 2001; Paauwe, 1998); and second, its attempt to relate the ndings to the economic and political environment, in explaining the HRM challenges facing civil service organizations in DCs. Hence, our study has important implications for the theoretical and practical debate in the area of the HRMperformance relationship. Conclusion This paper concludes that if the civil service organizations in developing countries like Eritrea are able to successfully implement HR practices, they could achieve the maximum contribution of their employees, although, at present, the economic and political environment within which HR practices operate is not that conducive (see also Hilderbrand and Grindle, 1997; Jaeger and Kanungo, 1990; Kiggundu, 1989; Prah, 2004; Wasti, 1998). This study provides further evidence with regard to the link between HR practices and performance and relates the ndings to environmental factors such as economic and political inuences. This study joins a growing body of research that attempts to open the black box by explaining how, when and to what extent HR practices affect performance at employee level. And thus, it is believed that this study contributes to research on the HRM and performance relationship in general and that of the civil service of developing countries in particular. While this study is an important step forward in understanding how, when and to what extent HR practices affect performance, as well as the challenges and prospects for effective HRM in developing countries, it also leaves some questions open for future research. First, the sample size may not be very large to generalize the ndings; second, this study is the rst of its kind and thus we have not found prior studies against which to compare our ndings. This is because similar studies in a similar environment have not been conducted. Hence, in order to generalize and validate the ndings of this study, we suggest that the same study is conducted with a larger sample size in other developing countries. References Ahmad, S. and Schroeder, R. (2003) The Impact of HR Practices on Operational Performance: Recognizing Country and Industry Differences, Journal of Operations Management, 21: 1943. Amanuel, G. and Tesfagabir, Z. (2002) A Report on Labour Market Issues and Remuneration Schemes. Eritrea: Asmara. 100 The International Journal of Human Resource Management Amanuel, M. (2002) Veteran Combatants Do Not Fade Away: A Comparative Study on Two Demobilisation and Reintegration Exercises in Eritrea. Bonn: BICC. Analoui, F. (ed.) (1998) Human Resource Management Issues in Developing Countries. Aldershot: Ashgate. Aredo, D. (2002) Brain Drain Reportedly Costing $4 Billion a Year. Addis Ababa: UN Integrated Regional Information Networks. Arthur, J.B. (1994) Effects of Human Resource Systems on Manufacturing Performance and Turnover, Academy of Management Journal, 37(3): 67087. Austin, J. (1990) Managing in Developing Countries: Strategic Analysis and Operating Techniques. New York: Free Press. Baron, J. and Kreps, D. (1999) Strategic Human Resources: Frameworks for General Managers. New York: Wiley. Becker, B.E. and Gerhart, M.A. (1996) The Impact of Human Resource Management on Organisational Performance: Progress and Prospects, Academy of Management Journal, 39(4): 799801. Becker, B.E. and Huselid, M.A. (1998) High Performance Work Systems and Firm Performance: A Synthesis of Research and Managerial Implications, Research in Personnel and Human Resource Management, 16: 53101. Benkhoff, B. (1996) Catching up on Competitors: How Organisations can Motivate Employees to Work Harder, International Journal of Human Resource Management, 7(3): 73652. Bennell, P. (1994) Improving the Performance of the Public Sector in LDCs: New Approaches to Human Resource Planning and Management. Occasional Paper no. 25. Geneva: ILO. Beugre, C.D. and Offodile, O.F. (2001) Managing for Organisational Effectiveness in Sub-Saharan Africa: A CulturalFit Model, International Journal of Human Resource Management, 10(4): 53550. Boselie, P. and Dietz, G. (2003) Commonalities and Contradictions in Research on Human Resource Management and Performance. paper presented at the Academy of Management Meeting, Seattle, August. Boselie, P., Paauwe, J. and Jansen, P. (2001) Human Resource Management and Performance: Lessons from the Netherlands, International Journal of Human Resource Management, 12(7): 110725. Bowen, D. and Ostroff, C. (2004) Understanding HRMFirm Performance Linkages: the Role of the Strengths of the HRM System, Academy of Management Journal, 29(2): 20321. Boxall, P. (2003) HRStrategy and Competitive Advantage in the Service Sector, Human Resource Management Journal, 13(3): 520. Brudney, J., OToole, L. and Rainey, H. (2000) Advancing Public Management: NewDevelopments in Theory, Methods, and Practice. Washington, DC: Georgetown University Press. Budhwar, P. and Debrah, Y. (2001) Introduction. In Budhwar, P. and Debrah, Y. (eds) Human Resource Management in Developing Countries. London: Routledge, pp. 115. Central Personnel Administration (CPA) (2002a) Statistical Data of Eritrean Civil Service. Eritrea: Asmara. Central Personnel Administration (CPA) (2002b) Civil Service Functional Reviews Report. Eritrea: Asmara. Chabal, P. and Daloz, J. (1999) African Issues: Disorder as Political Instrument. Oxford: James Currey for the International African Institute. Cohen, J. and Wheeler, J. (1997) Training and Retention in African Public Sectors: Capacity- Building Lessons from Kenya. In Grindle, M. (ed.) Getting Good Government: Capacity Building in the Public Sector of Developing Countries. Boston, MA: Harvard Institute for International Development, pp. 12553. Das, S.K. (1998) Civil Service Reform and Structural Adjustment. New York: Oxford University Press. Delaney, J.T. and Huselid, M.A. (1996) The Impact of Human Resource Management Practices on Perceptions of Organisational Performance, Academy of Management Journal, 39(4): 94969. Teclemichael Tessema and Soeters: HRM in developing countries 101 Delery, J.E. and Doty, D.H. (1996) Modes of Theorizing in Strategic Human Resource Management: Tests of Universalistic, Contingency and Congurational Performance Predictions, Academy of Management Journal, 39(4): 80235. Den Hartog, D. and Verburg, R. (2004) High Performance Work Systems, Organisational Culture and Firm Effectiveness, Human Resource Management Journal, 14(1): 5578. Dia, M. (1996) Africas Management in the 1990s and Beyond: Reconciling Indigenous and Transplanted Institutions. Washington, DC: World Bank. Economic Commission for Africa (ECA) (1989) Handbook for Manpower Planners in Africa. Addis Ababa: ECA. Eritrean Human Resources Development Project (2003) Terminal Report and Exit Strategy. Eritrea: Asmara, 14 May. Ferris, G.R., Hochwater, W.A., Buckley, M.R., Harrell-Cook, G. and Frink, D.D. (1999) Human Resource Management: Some New Directions, Journal of Management, 25: 385415. Fey, C., Bjorkman, I. and Pavlovskaya, A. (2000) The Effect of Human Resource Management Practices on Firm Performance in Russia, International Journal of Human Resource Management, 10(4): 53550. Freeman, R.H. and Lindaer (1999) Why not Africa? Working Paper No. 694. Cambridge, MA: National Bureau of Economic Research. Frezghi, A. (2003) The Development of a Performance Appraisal System For the Eritrean Civil Service. Unpublished MA thesis, University of Natal, South Africa. Gafer, A. (1996) The Role of the Civil Servant in the Development of Eritrea. Eritrea: Asmara. Gerhart, B., Wright, P.M. and McMahan, G.C. (2000) Measuring Error and Estimates of the HR Firm Performance Relationship: Further Evidence and Analysis, Personnel Psychology, 53: 85572. Grindle, M. (1997) The Good Governance Imperative: Human Resources, Organisations, and Institutions. In Merilee, G. (ed.) Getting Good Government: Capacity Building in the Public Sector of Developing Countries. Boston, MA: Harvard Institute for International Development, pp. 328. Guest, D.E. (1997) Human Resource Management and Performance: A Review and Research Agenda, International Journal of Human Resource Management, 8(3): 26376. Guest, D.E. (2001) Human Resource Management: When Research Confronts Theory, International Journal of Human Resource Management, 12(7): 1092106. Haregot, S., Stanley, J. Doctoroff, M. and Yosief, G. (1993) Public Sector Management Program for Eritrea: A Diagnostic. Asmara: United Nations Development Program. Hays, S. and Reeves, T. (1984) Personnel Management in the Public Sector. Massachusetts: Allyn and Bacon Inc. Heady, F. (1996) Congurations of Civil Service Systems. In Bekke, H., Perry, J. and Toonen, T. (eds) Civil Service Systems in Comparative Perspective. Bloomington: Indiana University Press, pp. 20766. Hilderbrand, M. and Grindle, M. (1997) Building Sustainable Capacity in the Public Sector: What Can Be Done? In Grindle, M. (ed.) Getting Good Government: Capacity Building in the Public Sector of Developing Countries. Boston, MA: Harvard Institute for International Development, pp. 3161. Hsu, Y. and Leat, M. (2000) A Study of HRM and Recruitment and Selection Policies and Practices in Taiwan, International Journal of Human Resource Management, 11(2): 41335. Huselid, M.A. (1995) The Impact of Human Resource Management Practices on Turnover, Productivity and Corporate Financial Performance, Academy of Management Journal, 38(3): 63572. Huselid, M.A., Jackson, S.E. and Schuler, R.S. (1997) Technical and Strategic Human Resource Management Effectiveness as Determinants of Firm Performance, Academy of Management Journal, 40(1): 17188. Ichniowski, C., Shaw, K. and Prennushi, G. (1997) The Effect of Human Resource Management Practices on Productivity: A Study of Steel Finishing Lines, American Economic Review, 87(3): 291313. 102 The International Journal of Human Resource Management Ingraham, P. and Kneedler, A. (2000) Dissecting the Black Box: Toward a Model and Measures of Government Management Performace. In Brudney, J., OToole, L. and Rainey, H. (eds) Advancing Public Management: New Developments in Theory, Methods and Practice. Washington, DC: Georgetown University Press. International Labour Organisation (ILO) (1998) Human Resource Development in the Public Service in the Context of Structural Adjustment and Transition. Geneva: ILO. Jackson, S.E. and Schuler, R.S. (1999) Understanding Human Resource Management in the Context of Organisations and their Environment. In Schuler, R. and Jackson, S. (eds) Strategic Human Resource Management. Oxford: Blackwell, pp. 428. Jaeger, A.M. and Kanungo, R.N. (eds) (1990) Management in Developing Countries. London and New York: Routledge. Jaeger, A.M., Kanungo, R.N. and Srinvas, N. (1995) A Review of Human Resource Management Success in Developing Countries. In Kanungo, R.N. and Saunders, D.M. (eds) Employee Management in Developing Countries. Greenwich, CT: JAI Press. Kalleberg, A.L. and Moody, J.W. (1994) Human Resource Management and Organizational Performance, American Behavioural Scientist, 37(7): 94862. Kamoche, K. (1993) Towards a Model of HRM in Africa. In Shaw, J.B., Kirkbride, P.S. and Rowlands, K.M. (eds) Research in Personnel and Human Resource Management. Suppl. 3. Greenwich, CT: JAI Press. Kamoche, K. (1997) Managing Human Resources in Africa: Strategic, Organisational and Epistemological Issues, International Business Review, 6(5): 547. Kamoche, K. (2000) Sociological Paradigms and Human Resources: An African Context. Aldershot: Ashgate. Kanungo, R.N. (ed.) (1995) Employee Management in Developing Countries. Greenwich, CT: JAI Press. Kanungo, R.N and Jaeger, A.M. (1990) Introduction: The Need for Indigenous Management in Developing Countries. In Kanungo, R.N. and Jaeger, A.M. (eds) Management in Developing Countries. London: Routledge, pp. 119. Kiggundu, M. (1989) Managing Organisations in Developing Countries: An Operational and Strategic Approach. Bloomeld, CT: Kumarian Press. Legge, K. (2001) Silver Bullet or Spent Round? Assessing the Meaning of the High Commitment Performance Relationship. In Storey, J. (ed.) Human Resource Management: A Critical Test. Stamford, CT: Thomson Learning, pp. 2136. Lienert, I. (1998) Civil Service Reform in Africa: Mixed Result after 10 Years. In IMF Finance & Development. Washington, DC: IMF. Maggregor, J., Peterson, S. and Schuftan, C. (1998) Downsizing the Civil Service in Developing Countries: The Golden Handshake Option Revisited, Public Administration and Development, 19(2): 4563. MacDufe, J.P. (1995) Human Resource Bundles and Manufacturing Performance: Organisational Logic and Flexible Production Systems in the World Auto Industry, Industrial and Labour Relations Reviews, 48(2): 197221. Ministry of Education (MOE) [Eritrea] (2001) Our People are our Future: A Framework for the Development of Human Resources in the Education Sector. Eritrea: Asmara/MOE. National Statistics and Evaluation Ofce (2003) Eritrea: Demographic and Health Survey 2002. Eritrea: Asmara. Noorderhaven, N. and Tidjani, B. (2001) Culture, Governance, and Economic Performance: An Explorative Study with a Special Focus on Africa, International Journal of Cross Cultural Management, 27(1): 3152. Ozgediz, S. (1983) Managing the Public Service in Developing Countries: Issues and Prospects. Staff Working Paper No. 583. Washington, DC: World Bank. Paauwe, J. (1998) HRM and Performance: The Linkage between Resources and Institutional Context. Rotterdam: Erasmus University. Paauwe, J. and Boselie, P. (2003) Challenging Strategic HRM and the Relevance of the Institutional Setting, Human Resource Management Journal, 13(3): 5670. Teclemichael Tessema and Soeters: HRM in developing countries 103 Patterson, G., Johnson, W. and Spreng, A. (1997) Modeling the Determinants of Customer Satisfaction for Business-to-Business Professional Services, Journal of the Academy of Marketing Science. Pfeffer, J. (1994) Competitive Advantage through People: Unleashing the Power of the Work Force. Boston, MA: Harvard Business School Press. Pool, D. (2001) From Guerrillas to Government: The Eritrean Peoples Liberation Front. Eastern African Studies London: Long House Publishing Service. Prah, K. (2004) African Wars and Ethnic Conicts: Rebuilding Failed States. Human Development Report 2004 African Regional Background Paper, UNDP. Smith, B. (2003) Understanding Third World Politics: Theories of Political Change & Development. New York: Palgrave Macmillan. Soeters, J. and Tessema, M. (2004) Public Management in Developing Countries: Some Notes on Eritrea, International Review of Administrative Sciences, 70(4): 62335. Stone, R. (1998) Human Resource Management. New York: Wiley. Taormina, R. (1999) Predicting Employees Commitment and Satisfaction: The Relative Effects of Socialisation and Demographics, International Journal of Human Resource Management, 10(6): 106076. Tessema, M. and Soeters, J. (2005) Practices and Challenges of Converting Former Fighters into Civil Servants: The Case of Eritrea, Public Administration and Development. Tessema, M., Soeters, J. and Abraham, K. (2005) Practices and Challenges of Training and Labour Utilisation in Sub-Saharan Africa: The Case of Eritrea, International Journal of Training and Development, 9(4). University of Asmara (1997) Implementation Plan of the Eritrean Human Resource Development Project. Asmara: University of Asmara. United Nations Development Program (UNDP) (2002) United Nations Development Assistance Framework for Eritrea: 20022006. Eritrea: Asmara. Waiguchu, M.J., Tiagha, E. and Mwaura, M. (eds) (1999) Management of Organizations in Africa: A Handbook of Reference. Westport, CT: Quorum Books, pp. 99123. Wasti, S.A. (1998) Cultural Barriers in the Transferability of Japanese and American Human Resources Practices to Developing Countries: The Turkish Case, International Journal of Human Resource Management, 9(4): 60931. World Bank (1994a) Eritrea: Options and Strategies for Growth. Washington, DC: Country Operations Division Eastern Africa Department, Africa Region. World Bank (1994b) Adjustment in Africa: Reforms, Results, and the Road Ahead. A World Bank Policy Research Report. New York: Oxford University Press. World Bank (1997) The State in a Changing World. World Development Report, 1997. New York: Oxford University Press. Wright, P., Gardner, T. and Moynihan, L. (2003) The Impact of HR Practices on the Performance of Business Units, Human Resource Management Journal, 13(3): 2136. Youndt, M., Snell, S., Dean, J. and Lepak, D. (1996) Human Resource Management, Manufacturing Strategy and Firm Performance, Academy of Management Journal, 39(4): 83666. Appendix 1: Measurement of items Respondents were asked to indicate the degree to which they agree or disagree with the following items related to different HRM issues. All the items were measured on a ve- point scale ranging from 1, strongly disagree to 5, strongly agree. I Recruitment and selection practices 1 Presence of written and operational recruitment and selection policy 2 Presence of clear job description and specication 3 Presence of attractive salary scales that can attract qualied applicants 4 High role of merit in recruitment and selection exercise* 5 Presence of a good image that attract qualied applicants 104 The International Journal of Human Resource Management II Placement practices 1 Assignment to the best position 2 Presence of clear duties and responsibilities* 3 Presence of good supervision III Training practices 1 Presence of clearly written and operational training policy 2 Presence of continuous training needs assessment* 3 Presence of written & operational trainee-selection procedure 4 Presence of governments commitment to training* 5 Presence of linkages of training to other HR programmes 6 Continuity of monitoring and evaluation of training programmes IV Compensation management practices 1 Presence of attractive compensation system 2 Presence of equitable internal salary 3 Presence of equitable external salary 4 Presence of salary that reects performance* 5 Presence of salary that encourages better performance 6 Presence of salary that reects the standard of living V Employee performance evaluation (EPE) practices 1 Presence of written and operational EPE 2 EPE results has a lot to do with salary* 3 EPE results has a lot to do with personnel decisions 4 Provision of feedback of EPE results 5 EPE is considered as important task by superiors 6 Performance evaluators are knowledgeable VI Promotion practices 1 Presence of written and operational promotion policy 2 Provision of priority to merit in promotion decision 3 Provision of priority to seniority in promotion decision VII Grievance procedure 1 Presence of written and operational grievance procedure 2 Presence of written and operational disciplinary programme 3 Presence of superiors who are knowledgeable in solving personnel-related problems VIII Pension programme or social security 1 Presence of attractive pension programme 2 I feel secure nancially in the future* 3 I save money in a bank for the future IX HRM outcomes 1 I have got training that makes me competent (competence) 2 I am satised with HR practices (satisfaction/motivation) 3 I do not have role ambiguity and confusion (clarity of role) 4 I do have an intention to leave the organization (retention)* X Performance as perceived by the respondents 1 My performance is better than that of my colleagues with similar qualications 2 My performance is better than that of employees with similar qualications in other ministries 3 The performance of my ministry is better than that of other ministries *Items that were phrased as negative are reversed in the analysis in order to be in the same direction. Thus, all items are measured on the same underlying dimension. Teclemichael Tessema and Soeters: HRM in developing countries 105