PRESIDENTIALISM IN DEVELOPING DEMOCRACIES: A SOURCE FOR BAD GOVERNANCE? A quantitative study on the effect of presidentialism on corruption in developing democracies

Abstract: Researchers within the field of political science have been analyzing the effects of legislative-executive relations for several decades and both parliamentarism and presidentialism have their critics and proponents. However, this topic needs revisiting, especially after Sri Lanka and Turkey recently experienced a shift in legislative-executive relations. This thesis aims to examine the effect of presidentialism’s interaction with lack of judicial and legislative control on the executive on control of corruption in developing democracies. The hypothesis claims that the higher the degree of presidentialism in a developing democracy, the more likely is the state to exhibit higher levels of corruption, due to the lack of judicial and legislative constraints on the executive. The hypothesis and the theory it is based on is tested quantitatively by employing an OLS multivariate regression analysis. Presidentialism was measured by taking the mean value of several VDem indicators describing presidential prerogatives. An interaction variable was composed of the presidentialism variable and the mean value of legislative and judicial control on the executive. Furthermore, six control variables were introduced to the multivariate regression. According to the results, the hypothesis had to be rejected, as presidentialism interacting with lack of judicial and legislative control on the executive is not correlated with higher levels of corruption at a level of significance. This result is important, as it shows that when employing a different measurement of presidentialism, taking a smaller sample size and adding an interaction effect can show a completely different result than that discussed in the literature review.