And Nothing else Matters? Revisiting the Partisan Theory of the Welfare State

While being one of the central explanatory approaches to welfare state politics in Western industrialized countries, partisan theory arguing that welfare state change is to be explained by the ideology of governing parties has been under attack in recent decades. First, contesting the explanatory power of the thesis for the period of welfare state expansion, scholars from the Varieties-of-Capitalism tradition have criticized the focus on the influence of social democracy. Instead, Voc-theorists argue that conservative parties who represent the interest of capitalists can also have a strong incentive to support a generous welfare state because it enables them to achieve comparative advantages via the investment into the specific skills of their employees (e.g. Swenson 2002). Second, following Paul Pierson’ New-Politics-argument, an important strand of the literature has argued that the impact of partisan ideology on social policies vanishes or even turns around when governments set out to cut the welfare state in an era of austerity (e.g. Kitschelt 2001; Ross 2000). Empirical results on that matter are, however, inconclusive: Depending on the measurement of welfare state change, researchers come to very different conclusions corroborating or contradicting the basic claim of partisan theory (Allan and Scruggs 2004; Kittel and Obinger 2003; Korpi and Palme 2003). The proposed paper revisits the much disputed claim that partisan ideology is a major explanation of welfare state change. We contribute to the literature in three respects: First, we present new hand-coded data on single welfare state reform events in five countries – Denmark, Finland, France, Germany and the United Kingdom – which enables us to identify legislative changes to the welfare state. Second, with this fine-grained data at hand, we explore whether we find an impact of the partisan ideology of the government on welfare state change. And third, our data which is based on the identification of individual policy instruments enables us to see whether the impact of partisan ideology is conditioned by the policy instruments used to cut or expand the welfare state. This is particularly interesting with respect to the visibility of a policy change (compare, for instance, a visible change to the nominal benefit level or an invisible change to the indexation formula).