PARTISAN ELECTORAL CYCLES AND GOVERNMENT RESPONSE TO THE COVID-19 PANDEMIC. A quantitative study of subnational governments in the US and Canada

Abstract

Although decisive government response to the COVID-19 pandemic have been critical to keep down infection rates such policies have also brought with it substantial social and economic costs. Much scholarly attention has been directed towards how political determinants have influenced elected leaders’ decision making on this. While these studies have provided important insights regarding the role of for example centralization, electoral system, and government ideology, less is known about more direct elements of political competition like elections. In this thesis, I investigate whether upcoming elections have influenced elected governments’ containment policy choices, and whether this effect have varied by partisanship. Drawing on the Electoral Cycles framework, I develop an argument that approaching elections will in general lead governments to relax containment measures as election approaches, but that this effect is mainly driven by politically conservative governments. I test these expectations on subnational elections in two similar countries that are likely to exhibit the effect: the US and Canada. Employing Two-way Fixed Effects models, I find that approaching elections are associated with a weak general decrease of containment policies. However, contrary to expectations, this pattern is most prevalent within liberal states and provinces. Analysing the countries separately does however not lend support to the main findings. I therefore conclude that approaching elections have had little to no effect on containment policy choices. These results are, however, interesting in themselves and add both to the literature on government COVID-19 response, as well as to the studies of Electoral Cycles.