The US demand for defense spending: An empirical investigation for the post-cold war era

Based on Smith (1989)'s neoclassical framework, the US demand for defense spending as a share of GDP (defense burden) is estimated by employing a relatively newly developed method called Me-boot rolling windows analysis using quarterly BEA and SIPRI data for the Post-Cold War times. While institutional inertia plays an important role in determining the US defense burden in general, there is no conclusive result concerning the growth rate since positive, negative, and insignificant effects were all observed for different time periods. The US defense burden was found to be well correlated with the price ratio of defense goods to civilian goods during times of high military mobilization probably due to the increasing demand for defense goods. The price effect in Smith's original theory that has been omitted in previous empirical studies due to data unavailability was, thus, confirmed. Both Russian and Chinese defense burdens were found to be important determinants of the US defense decisions after the 2nd millennium. However, since the implementation of the US new Asia-Pacific rebalance policy in 2012, no significant result was reported concerning the Russian threat while there was evidence towards a rising rivalry between the US and China in great extent. This, in turn, suggests that the US military policy follows its foreign policy closely in a very dynamic way.