Human resource flexibility, organizational culture and firm performance: an investigation of multinational firms in Hong Kong

Abstract

This article explores the relationships between human resource (HR) flexibility, organizational culture, and organizational performance. Drawing on previous works, we develop a conceptual model that links the constructs together. It is hypothesized that the three sub-dimensions of HR flexibility (i.e. employee skill flexibility, employee behaviour flexibility, and HR practice flexibility) affect adaptability culture, which in turn impacts HR-related and market-related outcomes. The data were collected via a questionnaire survey of HR directors/managers in multinational corporations (MNCs) in Hong Kong. The results of structural equation modelling (SEM) reveal the positive effects of employee behaviour flexibility and HR practice flexibility on adaptability culture. Employee skill flexibility, however, exerts no such effect. Further, adaptability culture is found to affect both HR-related and market-related performance in a positive manner.

Keywords:

• adaptability culture
• firm performance
• Hong Kong
• human resource flexibility
• multinational corporations

Notes

1. The measurement model shows the links between the latent variables (i.e. the three sub-dimensions of HR flexibility, adaptability culture and two performance indices) and their measurement items, while the structural model depicts the links among the latent variables under study. The acceptability of the model can be evaluated on the basis of the various fit indices as stipulated in the text. In conducting the confirmatory factor analysis, we ran the measurement model only. In testing the hypotheses, we ran a full model which comprises of both the measurement model and the structural model.

2. In structural equation modelling, if an initial model does not fit the data well, the model can be modified and tested again using the same data. Modification index shows the degree of improvement of the model fit when one parameter of the model is freely estimated instead of fixed. It helps to detect the misfitting parameters in the original hypothesized model (Joreskog 1993; Byrne 1998). In Figure 2 (i.e. modified model), we allowed the path from HR-related performance to market-related performance to be freely estimated. Compared with the original model in Figure 1, the RMSEA value of the new model reduced and fell within the acceptable level. In addition, the chi-square difference test (Δ χ²(1) = 39.5, p < 0.001) also evidenced that the new model fits the data better than the original one.