The first chapter studies the relationship between democratization and production of knowledge. Using bibliographic and patents data, we show that there is a positive and strong impact of democratization on the formation of knowledge in social sciences and business but not in other fields and patenting activity. We confirm these findings using an instrumental variable approach to correct for the endogeneity problems, originating from the unobservables that affect both innovation and democratization and from measurement errors in quantifying democracy indices. Our instrumental variable results are in line with our baseline results. In fact, they indicate that there is a downward bias in the baseline result, which is likely to stem from measurement errors in quantifying democracy indices and unobservables. Finally, our results are robust to a number of estimation methods, outliers, an alternative construction of our IV, and different measures of human capital.

In the second chapter, we examine the effect of mortgage credit market conditions on U.S. elections. During the financial crisis of 2008, the U.S economy experienced a sudden drop in mortgage credit supply. According to the previous research, voters responded to the financial crisis of 2007-2008 by punishing the incumbent party in the presidential election, meaning that the vote share of the incumbent party decreased. To further investigate the effects of the financial crisis on elections, we employ an individual-level dataset of loan application outcomes to examine the effects of the contraction in the mortgage credit market on the House and Gubernatorial elections of 2008. A two-stage approach is employed in order to estimate the impact of the mortgage market conditions on election outcomes. In the first-stage regression, a measure of the change in mortgage credit supply from 2004 to 2008 is derived by taking into account the demand for credit. In the second stage, we estimate the effects of the change in mortgage credit supply on the change in votes for the candidate of the democratic party as well as the candidate of the challenger party. We find no significant impact of the shrinkage in mortgage credits on House and Gubernatorial elections' outcomes. This finding suggests that voters only punish the president for the change in mortgage credits as they may believe lower-level officials are not responsible for this shift.

In the third chapter, we study the effects of elections on the changes in the supply of mortgage credits around elections. According to the literature, politicians have incentives to change economic policies in order to attract voters. We consider a particular type of credit offered through financial institutions and a specific kind of election: mortgage credits supply and Gubernatorial elections. We conduct a spatial regression discontinuity design and explore the financial consequences of gubernatorial elections. We focus on census tracts adjacent to one another yet in two different states. We find that census tracts in states where gubernatorial elections are held and governors have full control over both chambers of state legislatures, lending growth rates increase dramatically. Our results are robust to different specifications.