High performance work systems in Ireland: human resource and organizational outcomes

Abstract

From the perspective of the resource based view (RBV), although traditional sources of competitive advantage such as natural resources, technology, economies of scales, and so forth, create value, these resources are increasingly easy to imitate. As such, human capital and human resource management may be an especially important source of sustainable competitive advantage. Over the past few years, there has been growing research interest in the competitive advantage associated with high performance work systems (HPWS), a set of management policies and practices thought to endow employees with greater levels of skill, information, motivation and discretion. In this study, we describe a study examining the relative effectiveness of HPWS in the Irish context. Results suggest that greater use of high performance work systems is associated with positive human resource and organizational outcomes. Specifically, firms utilizing higher levels of HPWS tend to have lower rates of employee absenteeism and voluntary turnover along with higher labor productivity and lower labor costs.

Keywords:

• high performance work systems
• Ireland

Acknowledgements

We wish to acknowledge the financial support of the National Centre for Partnership and Performance (NCPP) and the helpful advice of Ms. Lucy Fallon Byrne and Mr. Larry O'Connell at the NCPP. We also acknowledge the support of the Centre for International Business Education and Research (CIBER) at the University of Kansas. The third author acknowledges funding received from the Irish Research Council for the Humanities and Social Sciences.

Notes

1. The 43 manufacturing firms in our sample are a more direct comparison with Datta et al. (2005). The average HPWS score for these firms is 46.34.

2. In general, labor intensive industries had sharply higher labor costs (e.g., finance with 60%) whereas capital intensive industries had lower costs (e.g. manufacturing with 27%).

3. This is indicated in the standardized beta coefficients. For employee turnover, the coefficients for the two groups were − .213 (Group A) and − .012 (Group B). For absenteeism, the comparable coefficients were − .228 and − .169.

4. This suggests disordinal interaction effects for retail industry ^* HPWS. Moderated regression analysis confirmed this to be the case.

5. At the suggestion of a reviewer, we tested whether or not the relationship between HPWS and organizational outcomes (productivity and labor expense) was mediated by either of our HR outcome variables (absenteeism and turnover rates). We used Baron and Kenny's (1986) procedure to test for mediating effects. In essence, this procedure tests whether the influence of HPWS (the IV) on productivity and labor expense (the DVs) is attenuated by the introduction of hypothesized mediators. We directly test this attenuation using a modification of the Sobel (1982) test statistic as suggested by Baron and Kenny (1986). This procedure follows Goodman (1960) in computing a standard error for the mediating effect and then calculating a z-score as the test statistic. Using this approach, of the four possibilities, we found support for one mediating relationship: voluntary turnover partially mediates the relationship between HPWS and productivity and this mediating effect is significant (p = .047). This finding agrees with other work suggesting deleterious effects for voluntary turnover on firm outcomes (Shaw, Gupta and Delery 2005).