從直選總統與民主化談政黨體系變遷：歐洲國家的比較研究

Directly Elected President, Democratization and Party System Change: A Comparative Study on European Countries

政黨政治向來為研究憲政運作、民主深化的核心議題。過往的研究，多將政黨政治視為影響憲政與民主的自變項，但憲政特性以及民主轉型的過程，也可能系統性的影響政黨體系，對於政黨體系變遷的規模、機率，存在一定的關連性。基於此，本文以政黨體系變遷為依變數，從制度特性以及民主化背景來觀察政黨體系變遷的特性。核心假設有三：首先，就制度特性而言，直選總統的國家，比起議會內閣制的國家，政黨體系變遷的機率較高、規模較大；第二，以選舉時程而言，也是制度面向的因素，在直選總統的國家中，同時選舉或蜜月選舉，因為總統衣尾效應較強，引起國會內政黨體系變遷的機率也較高。此外，因為小黨在總統選舉容易邊緣化，因此同時選舉或蜜月選舉下，小黨泡沫化的機率也較高；第三，非制度面向的部分，從民主化時程與轉型遺緒來看，歐洲國家中，後共轉型的新興民主，相較於歐洲中的老牌民主而言，政黨體系變遷的機率也較高。

Party politics is an important issue to explain the constitutional work and consolidation of a democracy, and especially for newly democracies. If the party system would be affected by different types of constitutional systems and historical legacy, the party system should also be changed. We have three hypotheses: first, we suppose those countries with a directly elected president will have a higher probability for party system change. And the smaller parties are more likely out of parliament. Second, based on the coattail effect, countries with the concurrent election or the honeymoon election will have a higher probability for party system change. Third, the degree of democratization will also be an important variable in measuring party system change. The party system change will have a higher probability in new democracies, such as in Central and Eastern Europe. Therefore, this paper goes to show that among these three hypotheses, directly elected president and electoral timing are institutional variables, while the degree of democratization is historical legacy.