Can Oversized Governments Be Stable? Exploring Coalition Politics in Japan

Coalition government becomes far from unusual in Japan, especially since 1990s. This paper investigates why some oversized coalitions could be stable in Japan. Although this coalition form is theoretically understood as an exotic and unstable in comparison with single-party governments or minimum winning coalitions. However most of Japanese cabinets until 2018 preferred to include surplus junior partner(s), furthermore some of the oversized coalitions can be seen more stable than other types of Japanese cabinets. This paper attempts to explore different patterns of oversized coalitions from Japanese cabinets ranging from 1993 to 2018. Although several studies have already tackled the puzzle of Japanese coalition politics, but this study makes key contributions in three respects. First, prior research has paid little attention to mechanisms of oversized coalitions in Japan. While empirical studies in Japan often lack a systematic comparative perspective of the different mechanisms, theoretically informed studies focused mainly on intra-party politics of Liberal Democratic Party (LDP) under the predominant-party system. Since non-LDP government was formed by new prime minister Hosokawa in 1993, we can observe various coalition strategies based on the institutional settings and/or intra-party motivations such as office-seeking, policy-seeking, and/or vote-seeking. The second point concerns a methodological pitfall in coalition studies. Recent statistical methods for investigating coalition governments such as conditional logit can be seen as a solution to the configurative problem of institution-centred and actor-centred explanations by considering possible coalitions of political parties. Although these statistical methods yield more observational implications than other discrete choice models, it is not the case for the conjecture of X-combinations, i.e. possible conjunction of conditions. Therefore, we suggest applying a configurative method to coalition research in Japan. The case-based and set-theoretic approach is a welcome tool for considering comparative ways of the oversized coalition patterns. Finally, the above research problems are linked to the issue of what research on Japan can theoretically add to coalition studies. While a general theory of coalitions has been developed through Western European studies, this cannot be automatically transferred to Japanese case, principally because we should consider a low level of disciplined party organisations. Furthermore, opportunistic coalitions as a result of ‘free-style’ negotiations between political parties would facilitate fragile coalition formations in Japan. Theoretically we may consider that the smallest majority is not plausible, consequently the formation of oversized coalitions is more likely to secure the winning coalition. However, this theoretical explanation will face a difficulty of the application for relatively stable oversized coalitions in Japan. To fill the gaps this paper attempts to extract patterns of oversized coalitions from case studies of Japanese cabinets ranging from 1993 to 2018. Specifically, we trace coalition processes under two different stages: While resigning LDP and ‘power shuffles’ in the 1990s would have an impact on party behavior, we can consider that returning LDP modified rules of coalition politics from the 2000s against the background of the new electoral system and the bicameral structure. From the configurative analyses, this paper finally aims to identify comparable components of stable oversized coalitions.