The Dependent Variable Problem within the Study of Welfare State Retrenchment: Defining the Problem and Looking for Solutions

Since the publication of Pierson's seminal work, a scholarly debate about welfare state retrenchment has emerged. One of the debated issues has been the “dependent variable problem”: what is welfare state retrenchment and how can it be measured. In particular the pros and cons of different types of data have been discussed. The argument of this article is that the “dependent variable problem” is a problem of theoretical conceptualization rather than a problem of data. It is crucial to be aware that different theoretical perspectives on retrenchment should lead to different conceptualizations of retrenchment. Furthermore, different conceptualizations lead to different evaluations of the same changes in welfare schemes, just as the question of which data to use depends very much on the theoretical conceptualization of retrenchment.