The roots of partisan effect: party support and cabinet support under the coalition governments in Japan in the 1990s

This dissertation examines the determinants of the effect of partisanship on support for a parliamentary government. In doing so, I address a set of related questions, using Japan as an example. I begin with a descriptive question: Is the effect of partisanship on the job approval of the administration changing over time? To answer this question, using 1960-2001 time series data, I demonstrate the changing impact of the job approval rate of the cabinets over this period. Then I turn to explanation for the change and ask: Why does the effect change over time? I hypothesize that supporters of newly established parties in the government are less likely to be influenced by their partisanship when evaluating the cabinet' performance. Partisanship, defined here as a predisposition to support a particular political party, grows with the cumulative effect of political experience and learning. There is, however, less opportunity for newly established and political parties to have such loyal supporters. My second hypothesis holds that supporters of ruling parties to which the prime minister does not belong are less likely to make partisan judgments in appraising the cabinet's performance. Party identification extends to the government in which the party participates, the partisan effect on the appraisal of the government's performance emerges. The party affiliation of the prime minister influences to what extent people associate the government with the party.