Empirical analysis of the defense interdependence between Japan and the United States

Abstract

The purpose of this paper is to analyze the defense interdependence between Japan and the United States (US) in the short and long run and to investigate whether Japan is a follower or a free rider of the “US” over the 1975–2009 period. Given that mutual aid between Japan and the US has been maintained, the empirical results suggest that Japanese defense has a stable relationship with US defense. Furthermore, the results reveal that Japan is not a free rider but rather a follower in the period under consideration.

Keywords:

• Japan
• Defense expenditure
• Japan–US alliance

JEL Codes:

• D74
• H56

Acknowledgments

I would like to thank the editor Christos Kollias, two anonymous referees of this journal, Ryokichi Chida, Kenichi Fukumiya, and Toshiya Hatano for their many helpful and insightful comments. Any remaining errors are the author’s responsibility.

Notes

1 As Morimoto (2008) mentions, for Japan, the aim of the reorganization of US forces is to strengthen the deterrent function between both countries and to reduce the burden of maintaining military bases, including Okinawa. On the other hand, the aims of the US are to strengthen Japan’s strategic function, similar to Guam, in order to expand the access to bases and to reinforce the alliance between both countries cooperatively.

2 For more discussion and technical details with respect to the alliance models a comprehensive exposition can be found in Sandler and Hartley (1995).

3 The data used here are drawn from the following sources: a. The Center for International Comparisons at the University of Pennsylvania (various years), The Penn World Table, http://pwt.econ.upenn.edu/php_site/pwt_index.php. b. SIPRI (various years), Yearbook: World Armaments and Disarmament, Stockholm International Peace Research Institute, Oxford University Press. c. US Department of Defense (various years), DoD Personnel and Military Casualty Statistics http://siadapp.dmdc.osd.mil/personnel/MMIDHOME.HTM

⁴The 14 countries taken into account in this paper are Australia, Bangladesh, China, India, Indonesia, Korea, Malaysia, New Zealand, Pakistan, Philippines, Singapore, Sri Lanka, Taiwan, and Thailand. The data are in US $m, at constant 2000 prices.

⁵For GDP, government expenditure, population, real GDP per capita, and government share of real GDP data are drawn from ‘The Penn World Table’. Japanese, US, and Asia-Pacific defense expenditure data are obtained from SIPRI Yearbooks (various years) and deflated using the government expenditure deflator. Data for US Forces stationed in Japan are from US Department of Defense ‘Personnel and Military Casualty Statistics’.

⁶The panel unit-root test by the Levin, Lin, and Chu test (no time trend) for the level data shows that the null hypothesis is rejected at the 1% significance level in all data series.

⁷For more details, see Pesaran and Shin (1999) and Pesaran et al. (2001).