This dissertation analyzes how political behavior influences the allocation of public goods. The first chapter studies the effect of conventional political participation (electoral competitiveness) on natural disaster preparedness spending in the United States. The second chapter analyzes the effect of non-conventional political participation (protest) on water policy outcomes using Mexico's data. I discuss that contentious actions such as protest (from sit-ins to sabotage) are pervasive in many democratic countries, yet their effect on public policy has been overlooked, leaving a gap in our knowledge of how the public may influence the policy-making process. From the social welfare perspective, the crucial question is the following: What is the overall effect of protest (performed by a few) on policy outcomes (affecting the utility of a wider group)? I show theoretically that even a social planner with complete information will allocate water sub-optimally and, empirically --using an estimation of propensity to protest in Mexico-- I show that protest affects the provision of water water at the municipality level. The last two chapters analyze why citizens engage in protest, contributing to a better understanding of a behavior that has consequential effects for social welfare. I address why citizens engage in protest as an alternative or as a complement to electoral participation. The results show that the chosen actions depend of the relative costs and benefits of each strategy. The fourth chapter explores why some citizens engage in illegal protest specifically. The empirical model shows that affiliation to some organizations increases the probability of that type of protest. As a whole, the paper expands the knowledge of political participation in Mexico and addresses the implications for policymaking in democratic countries.