HRM implementation by line managers: explaining the discrepancy in HR-line perceptions of HR devolution

Abstract

Previous studies already established the idea of a partnership in which HR professionals and line managers share an organisation’s HRM responsibility. Yet, this relationship is often plagued by conflicts and other obstacles. As such, a perceptual discrepancy is likely to exist between both parties on the degree of HR devolution, which may eventually lead to bad performance. Using survey data, we empirically analyse which factors may explain a perceptual discrepancy between HR professionals and line managers on the latter group’s role in HRM. Results show that the HR-line discrepancy on the degree of HR devolution is rooted in differences in perception on several other factors, including organisational support, (personnel) red tape, the line’s individual capacity and age. Overall, though, it is a matter of understanding both HR’s expectations and the line’s experiences in all aspects of their partnership.

Keywords:

• HR devolution
• HR partnership
• line management
• perceptual differences
• Tobit approach

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. We use the term ‘HR professional’ to refer to the position of those working in the field of personnel management, people management or HRM, and who possess general knowledge and experience in this area.

2. For some items, the formulation was slightly adjusted to the participants, i.e. line managers or HR professionals.

3. In order to construct the dependent and independent variables, both HR professionals and line managers completed the same questions. Hence, we ended up with two sets of paired items. For each variable and per organisation, we calculated the difference in the scores of HR and line managers. Since our data included 901 line managers and only 26 HR professionals, this difference was calculated by randomly selecting HR professionals per organisation with replacement. The responses of HR professionals were thus used repeatedly. As such, a new observation is created for each line manager, indicating the difference between his/her response and a randomly selected (with replacement) HR professional. Overall, this construction may result in a positive difference when HR managers gave a higher score than line managers, and vice versa. The values included in our regression, however, are formulated in absolute terms. This process was followed for the dependent, the independent and the control variables included in our research (unless indicated otherwise).

4. 1 = no responsibility at all; 4 = full responsibility. Respondents who answered ‘not applicable’ on one or more of the HR tasks, were not included in our study.

5. Using a polychoric correlation matrix, an exploratory factor analysis (EFA) was performed on the items making up the independent variables. Five factors were retained, with reliability of the final scales demonstrated by Cronbach’s alphas all ranging between .68 and .94.

6. We also conducted a ‘regular’ Ordinary Least Squares regression. This only affected the size of the coefficients, the significance of the variables remained the same. As the Tobit coefficients are considered to be more reliable, OLS results have been omitted. They are available upon request from the authors.