Human resource management and firm performance in Europe through the lens of business systems: best fit, best practice or both?

Abstract

This study uses the lens of Business Systems theory to explore the importance of geographic context on the link between human resource management and organizational performance. Basing the analysis on ‘HRM bundles of competitive advantage’, drawing evidence from a large-scale survey of European private sector businesses, and using multiple methodologies, we find three distinct geographic regions and 21 ‘HRM bundles of competitive advantage’. Of those bundles 10 were significantly related to performance in one or more regions. The results raise issues about the universal applicability of HRM-performance research and have implications for the standardization of HRM policies and practices within internationally operating organizations.

Keywords:

• comparative HRM
• Europe
• performance

Acknowledgements

This research was supported by a University of Cyprus grant on HRM and Competitiveness. We especially appreciate the assistance of Stellios Spiliotis, Christos Kilaniotis and Irene Karamanou (University of Cyprus) for their input in this study. We would like to thank our many Cranet colleagues for allowing us to use their country data.

Notes

1. Further studies of the HRM-performance link: Addison, Siebert, Wagner and Wei (2000) in Germany and Britain; d'Arcimoles (1997) in France; Bae and Lawler (2000) in Korea; Boselie (2002), and Den Hartog and Verburg (2004) in the Netherlands; Faems, Sels, De Winne and Maes (2005) in Belgium; Guest, Michie, Conway and Sheehan (2003); Purcell, Hutchinson, Kinnie, Rayton and Swart (2003); Khatri (2000), in Singapore; Lee and Miller (1999) in Korea; Truss (2001); Wood (1999) in the UK; Fey and Bjorkman (2001) and Fey and Dennison (2003) in Russia; Godard (1998) in Canada; Guthrie (2001) in New Zealand; Lahteenmaki et al. (1998) in Finland; Mitsuhashi, Park, Wright and Chua (2000) in China; Teo (2002) in Australia; and Ngo et al. (1998) in Asia.

2. Empirical evidence on the link: Becker and Gerhart (1996); Becker and Huselid (1998); Guest et al. (2003); Wall and Wood (2005); Wood (1999) and a range of special issues in the Academy of Management Journal, 39, 4, 1996; Human Resource Management Journal, Fall, 1999, Fall, 2003; Human Resource Management, Fall, 1997; Human Resource Management Review, 8, 1998; Industrial Relations, 35, 3, 1996; International Journal of Human Resource Management, 8, 3, 1997; 12, 7, 2001; Personnel Psychology, 53, 2000; and Personnel Psychology, 58, 2005.

3. Launched in 1989, the Cranfield Network on Human Resource Management (Cranet) has been established to meet the need for ready access to information on best practice and comparative performance within HRM. The Network itself is a collaboration between 41 universities and business schools worldwide.

4. EU member states prior to May 2004 were: Austria, Belgium, Denmark, Finland, France, Germany, Greece, the Netherlands, Ireland, Italy, Portugal, Spain, Sweden, UK and Luxembourg. Since Luxembourg did not have a sufficient sample of organizations of the requisite size it was excluded from our analysis.

5. Since the dependent variable is categorical, with more than two categories, we need to ensure that the beta coefficients are the same for all categories of the dependent variable (performance). The test of parallel lines (based on the chi-square distribution) compares a model with one set of beta coefficients for all categories of the dependent variable to a model with a separate set of beta coefficients for each category.

6. By unit increase we imply a displacement of unit length of the sub-bundles within a specific bundle.