Abstract
Unlike most of human resource management (HRM) research attempting to identify the set of practices that are likely to improve organizational performance, this paper focuses on two key aspects of the organization of HRM: 1) the integration of HRM and business strategy; and 2) the distribution of roles and influences between line managers LMs) and HRM specialists. Building on the resource-based view, we suggest that HRM integration is a necessary but not sufficient condition for HRM positively to impact organizational performance. An equally necessary condition is to provide HRM specialists with a prominent role compared to LMs in order to ensure the required proper quality of implementation of decided HRM policies. Using data from the Cranet Survey, we employed a series of structural equation models to test the moderating effect of the HRM/LM relationship on the link between HRM strategic integration and organizational performance. This technique allowed us to estimate measurement models and structural relations among latent variables, which reinforces to a great extent the robustness of our empirical findings compared to previous studies, which have relied merely on standard OLS regression analysis. Our empirical findings lead us to call for less emphasis on 'instrumental' approaches to researching HRM, emphasizing 'what practices must be implemented?', in favour of an approach which HRM examines the question of who is in charge of defining and implementing HRM practices.