Electoral systems and forms of abstention

Abstract

Esta tesis es de carácter interdisciplinario, entre economía y ciencias políticas. Usando métodos bien explotados en economía explico como las instituciones electorales afectan las decisiones de los individuos en varios contextos políticos. El capítulo 2 es normativo y propone dos modificaciones en los sistemas electorales. Ambos resultan en una participación más alta y dan incentivos a los partidos políticos a que sean de mejor calidad. Primero, analizo un sistema electoral donde el número de los candidatos que obtienen representación en el parlamento es endógeno y depende del nivel de la participación. Segundo, analizo el efecto de un quorum de participación en elecciones parlamentarias. Según mis resultados, las dos modificaciones implican beneficios más bajos para los partidos, y bienestar más alto para los ciudadanos. Dichos resultados explican porqué los partidos usan otros tipos de medidas para incrementar la participación en las elecciones. Además, mis resultados explican la evolución del uso de un quorum en algunos países. El capítulo 3 es un trabajo junto con Sabine Flamand. Estudiamos el efecto de un quorum de participación en reuniones de pocos miembros. Usando un modelo de reuniones repetitivas, demostramos que la decisión se retrasa cuando el quorum es alto, y que los miembros no pierden utilidad por el hecho de retrasar la decisión. Finalmente, la presencia de un quorum puede tanto aumentar como reducir el número de miembros tomando la decisión final. El capítulo 4 ofrece un análisis empírico de las decisiones de votar en blanco y votar nulo. El análisis es innovador porque por primera vez se tratan las dos decisiones por separado. Demuestro que el voto blanco contiene un mensaje de insatisfacción contra los partidos políticos, mientras que el voto nulo contiene insatisfacción contra las instituciones y los niveles de democracia.

The present thesis lies at the intersection of economics and political science. Using methods, well developed in economics, I try to understand how electoral institutions shape individuals' behavior in different political contexts. Chapter 2 has a normative flavor. Focusing on the interplay between low quality parties and citizens' apathy, I try to break the vicious cycle that links the two, by proposing two electoral rules that increase turnout in PR elections, and at the same time give incentives to parties to be of better quality. First, I propose an electoral rule where the number of candidates elected depends on the level of participation. Second, I propose the introduction of a participation quorum that has to be met in order for the election to be valid. The common feature and innovation of these rules is that turnout affects the electoral outcome, and as a consequence these rules incentivize parties to care about the level of turnout. I show that both rules, while they increase turnout they imply lower profits for parties. My results explain why parties target to increase turnout through a certain type of measures that do not necessary improve the quality of the vote. Moreover, I also explain the evolution of the use of the participation quorum in certain countries. Chapter 3 is coauthored with Sabine Flamand, and tries to understand the effect of such a participation requirement on individuals' behavior and the decision outcome. To this end, we model a setup of repeated meetings, where a small group of individuals has to take a decision. We show that the decision is delayed when the quorum requirement is high and members are not harmed by postponing the decision. Surprisingly, the presence of the quorum may decrease the number of attendees taking the decision, while we show that in order to avoid policy distortions, the required number of participants must be even. Apart from abstaining, voters that are not willing to support any of the candidates in most parliamentary elections, are given the choice to participate in the election and cast a blank or a null vote. A blank vote is a disapproval vote of all competing candidates, while a null vote is a vote cast erroneously or deliberately in a way not conforming with the legal voting procedure. Political scientists were treating blank and null votes in an identical way. My attempt in chapter 4 is to study these two protest actions on a separate basis, in order to understand, why in some elections blank votes are many more than null votes and vice versa. After constructing a database considering the percentages of blank and null votes separately, I show that the amount of blank and null votes cast in an election are not affected by the same factors. Null votes convey dissatisfaction towards the electoral and democratic institutions, while blank votes convey dissatisfaction towards the parties. More important, my results go against one of the prevailing criticisms of compulsory voting. The latter has no significant effect on the amount of uninformative votes since it has no significant effect on the amount of null votes. On the contrary, it increases only the amount of blank votes, which by definition disclose information, and in particular voters' disapproval of all competing parties. Although abstention is one of the most studied issues both by political scientists and economists, the current thesis extends our knowledge, by giving insight into some of abstention's unexplored but widely observed aspects.