This paper investigates the time-honoured “Do Parties matter?” question. It starts by setting two criteria for measuring partisanship: (1) a differentiation of policy pledges (if parties offer same policy choices which party gets elected does not make any difference); and (2) a fulfilment of these pledges (if parties in government do not implement the policies they promised, which party gets elected does not make any difference either).
We then identify a series of conditions (party system, institutional power structure, issue salience, public opinion moves, policy type) under which these two criteria are most likely to be met. The explanatory power of our hypotheses is tested with respect to the degree of partisan distinction regarding one specific policy domain, the regulation of copyright protections in relation to the Internet in more than twenty countries since the early 2000s using ResponsiveGov and Comparative Agendas Project data.
