Abstract
Policy priorities, which are set by political leaders, determine the economic-institutional environment in which financial actors operate. How investors react to changes in policies is, however, still an open issue. This thesis sets out to investigate how stock markets react to changes to the political leadership of a country. Few political events better illustrate uncertainty around future policies than changes in the executive leadership. Regardless, few political scientists have set out to describe and interpret the complex relationship between politics and finance. Using theories and methods from both economics and political science, this thesis seek to bridge the gap and bring a more holistic understanding of the link between politics and finance. The main argument in the thesis is that change in the political leadership affects markets by adding uncertainty around future policy, and therefore future cost and gain for investors. I assume that this will lead to lower returns and higher volatility in periods around when the de facto incumbent of a country is changed. In order to formalize my argument, I develop a game-theoretical model to explain the assumed behavior for investors in a framework of policy uncertainty. To measure financial markets performance I use national stock market indexes, and the data on political leaders are taken for the Archigos-dataset. The research design is a large-N quantitative study. In order to capture two different ways in which markets might react to these events, I test for effects of leadership changes on stock markets both in terms of their mean value and volatility. To do this I use two different statistical techniques: (1) an event study model and (2) a GARCH (1,1). The results from the analysis show that markets indeed react to political changes, but first and foremost with higher volatility. Higher volatility means that financial market values vary more in these periods of political transition. While the theoretical expectation is that also mean returns should be affected – as investors generally were expected to be fearful of political leadership changes - no statistical significant result are found in regard to this in the empirical analysis. Both of these results are stable and robust through different specifications.