US-Japan Security Alliance in Crisis and Japanese Citizens’ Support for the SDF

• In Event: Poster Session: Political Economy II & Public Opinion
  In Poster Session: Poster Session: Public Opinion

Abstract

In light of the heightened political tensions in East Asia, there is a deepened necessity for a broader understanding of perceptions about security issues among citizens of an important ally for the United States---Japan. We administer a survey experiment in Japan to unpack Japanese citizens’ honest and nuanced feelings toward the complex nature of the Self-Defense Forces (SDF), Japan’s de facto military forces, and Japan’s security policy.

Specifically, to delve into research on Japanese citizens’ support for the SDF and the U.S.-Japan security alliance in crisis, we field a survey experiment in Japan with a national sample. We first explain a hypothetical situation abroad in which the SDF and the U.S. militaries are operating together; specifically, either a U.S.-Japan joint naval patrol operation in South China Sea or a joint amphibious landing operation in Senkaku islands. Given this background, we randomly present various contingency situations (e.g. variation in who the attacker is), and study participants are informed that a certain number of Japanese and/or American personnel are killed. The number of casualties is also randomized. We then ask participants how soon the SDF should be withdrawn from the mission and return to Japan. We also ask for the participants’ views in relation to the U.S.-Japan security alliance in situations where the SDF have casualties.

An important fact is that since its establishment in 1954, no member of the SDF has been killed in action. In Japan, it is often claimed in political debates that any potential risk that may cause the death of SDF personnel should be removed before dispatching the SDF in overseas missions. In this context, we expect that Japanese citizens will be hyper-sensitized to the death of Japanese citizens. When any member of the SDF is killed, but perhaps not necessarily when U.S. soldiers are killed (regardless of the number of U.S. casualties), Japanese opinion on the immediate withdrawal may be strengthened. Japanese citizens may even question the effectiveness of the U.S.-Japan security alliance and consider the alliance as a possible source of danger for Japanese personnel. All of these results, as discovered, will question the sustainability of the U.S-Japan alliance.

This experiment is also expected to make an important contribution to the scholarly literature on casualty sensitivity (Larson, 1996, Gelpi, Feaver and Reifler 2005, 2009). The existing studies on alliance/coalition reliability suggest that the size of casualty in a joint military operation positively affects the probability of an early withdrawal by the affected country. Such an early departure from a coalition would be a serious defective action to its alliance/coalition partner and risk its credibility. Tago (2009), by using the original data of the U.S.-led coalition in Iraq, shows that one additional combat casualty in the troops increases the chance of an early withdrawal from the coalition by 60%. Studies by Tago (2009) and others on public opinion and war (e.g. Mueller 1973; Weisiger 2016), however, do not empirically demonstrate the mechanism linking the size of causality and the likelihood of early departure. They only assume that the public is sensitive to combat casualties and thereby pressure the government to terminate a military action even if it could mean ruin for its alliance and the coalition’s reliability. To the best of our knowledge, this important assumption has never been tested. We address this by manipulating the number of casualties and measuring respondents’ attitudes toward the troops’ withdrawal and toward the U.S.-Japan alliance within our experimental setup.

Authors

• Takako Hikotani, Gakushuin University

• Yusaku Horiuchi, Dartmouth College

• Atsushi Tago, Waseda University