Yannis Georgellis, Ph.D, Economics and Finance, Brunel University, School of Social Sciences, West London, UB8 3PH, United Kingdom
Although empirical evidence confirms the importance of comparison earnings as a determinant of worker's well-being, the issue of whether recent studies have exaggerated the prevalence of reference-dependent preferences is yet to be settled. The purpose of this article is to re-evaluate the hypothesis of reference-dependent preferences by investigating the dynamics of adjustment of actual to comparison earnings in Britain. The investigation is based on the premise that a wedge between actual and reference earnings is associated with a reduction in workers' well-being that spurs them into action to change their status quo, the robustness of which is linked to the size of the wedge in a non-linear fashion. The presence of non-linearities characterizing the dynamics of adaptation is consistent with stronger evidence of reference-dependent preferences when actual earnings are substantially lower than reference earnings. In contrast, evidence of reference-dependent preferences is much weaker, or non-existent, when actual earnings are close to reference earnings. Significant differences in adaptation speeds across the private and public sectors, moderated by gender and educational attainment, are also evident. We attribute such differences to the differential returns to education across the two sectors, to the prominence of wage compression in the public sector, and to the higher incidence of public service motivation among public sector workers.