Causes and effects of United States military expenditures (time-series models and applications)

The study is composed of three main parts. The first part identifies limitations of various techniques used to empirically model arms races, and proposes an alternative, an error correction model by which we can test (1) whether two military expenditure series exhibit a long-run equilibrium relationship (i.e., are cointegrated), and (2) whether one or both series adjust to disturbances in this equilibrium state (i.e., show error correcting behavior). Results of this part suggest that the US-Soviet arms race can be modeled as an error correction process.; The second part, as a supplement to the first, is designed to empirically investigate the proposition that a true arms race should exhibit a strong degree of autoregression. A model of the autoregression (AR model) was evaluated using various diagnostic tools such as tests for CUSUM, or for constancy in a model's coefficients and recursive residuals. The AR model reveals a good performance in accounting for U.S. military expenditures. Such results thus can be taken as an another evidence supporting the existence of the US-Soviet arms race during the cold war period.; The last part of the study, given various arguments on the nature and direction of the causal relationship between military expenditures and economic growth, adopts a specific approach (VAR) not only confirmatory but also exploratory. But, unlike previous researchers, the author modified the VAR model by accommodating both theoretical and statistical assumptions (i.e., SVAR model with Bayesian prior) as an effort to reach a coherent findings on economic motivations for and consequences of the heavy military expenditures. Results show that U.S. decision makers have not used defense spending as a fiscal stimulus to offset economic downturns, and no substantial effects of the defense spending on economic growth was found.; Overall, given all of those findings, it may be concluded that U.S. military expenditures have been determining by external security considerations rather than domestic economic considerations during the cold war era.