L’intervention du pouvoir judiciaire, pour arbitrer un scrutin en dernier recours, pose certains problèmes de déroulement dans la démocratie représentative, voire un paradoxe : le principe de la représentation démocratique par vote direct des citoyens peut être mis en cause par l’interprétation judiciaire de l’expression des voix. En même temps, l’intervention du judiciaire comme pouvoir indépendant constitue une garantie d’impartialité sur les résolutions politiques cherchant à renforcer la confiance des principaux acteurs sociaux et politiques sur l’administration électorale. D’autre part, le problème de la rationalité du politique versus la rationalité juridique reste confronté en permanence en cas de processus électoraux conflictuels. Celles-ci s’avèrent des approches indispensables bien que pratiquement antagoniques, du fait qu’elles poursuivent des intérêts proches, mais en soit très distincts. / The intervention of the judiciary power to solve a ballot in last resort, arise several problems concerning the progress of the representative democracy, and even a paradox: the principle of the democratic representation by an indirect vote of the citizens can be put forward by the judiciary interpretation on the meaning of the ballot's votes. At the same time, the intervention of the judiciary as an independent power constitutes a guarantee of impartiality for the political resolutions, seeking to reinforce the trust of the main social and political protagonists about the electoral administration. We are also faced with the problem of the political rationality versus the legal rationality, constantly put forward during electoral conflict contemporary processes as essential protagonists but nearly antagonistic ones, by the fact that they pursue close but distinct interests.

This thesis identifies the effect of traditional political coalitions on the left-right spectrum in Swedish municipal politics on economic outcomes such as Municipal Revenues, Expenditures, Net-expenditures, Municipal Tax-rates, Unemployment, and the share of Municipal Employment. To do so, varying time spans of Swedish municipal data from 1994-2017 are used in a regression discontinuity design, basing its identification on quasi-random variation created by close municipal elections. The results indicate that the left-leaning Red-Green coalition has no isolated impact on economic variables when considering the entire sample, but seems to have a significant impact on some economic variables compared to others when the sample is restricted to decrease the number of mixed coalitions in the sample. These results indicate that the increase in mixed governing coalitions in recent years could have watered down the clear left-right dimension in Swedish politics found in previous research.