The evolution of partisan research policy positions in Western Europe

Did partisan politics matter for research policy in Western Europe over the last decades, and if so how? A possible answer is that partisan actors do not matter much to research policy at all, due to low saliency and widespread agreement on goals such as high-quality research conducted efficiently, providing new knowledge that contributes to growth and societal improvement. Under such conditions, it seems unlikely that political parties can reap major electoral benefits from their research policies. Consequently, partisan actors may not devote resources to developing research-policy positions, instead leaving the details of policymaking to bureaucracy and stakeholders. Another possible answer is that partisan political interest in research policy has strengthened over time, with the increasing share of GDP devoted to research funding and the growing importance of research to an increasing number of other, salient policy areas making, potentially leading to partisan disagreement on which research goals that should be prioritized. Previous literature do not systematically investigate this question empirically. This paper seeks to start filling these gaps by asking how variation in partisan research policy-positions expressed in party manifestos evolve from the 70s and onward. Theoretically, I argue that partisan political actors are likely to care about research policy to the extent that the parties and their constituents believe research is important to other, more salient, policy issues. Drawing on literature on partisan politics, I provide an account of when and how various partisan actors may conceive of research as important to other policy areas. Based on this account, I develop a set of propositions on variation in partisan research policy-positions in terms of saliency and other policy issues tied to research policy. These propositions structure my empirical analysis, but due to the lack of strong expectations and previous research on partisan political variation, I keep my analysis exploratory. Empirically, I utilize existing data on political manifestos in a novel fashion. First, I apply a simple dictionary of research words consisting of different versions of research, science, researcher and scientist to party manifestos. These keywords may or may not entail a research policy position. Therefore, the second step of my initial data collection is a content analysis of sentences surrounding a sample of the words detected in the first step. I use these data in three ways: 1) Quantitative estimation of saliency through the relative frequency of research policy positions 2) Quantitative estimation of the number of other policy issues partisan actors tie research policy to. Drawing on content analyses of party manifesto provided by the Comparative Manifesto project and the Comparative Agendas project, I am also able to get some leverage on which policy-issues partisan political actors tie research policy to from the late 90s and onward. 3) Qualitative content analysis of the content of partisan research policy positions allowing me to probe for other forms of variation in the content of partisan research policy positions in further detail.