26 Pages Posted: 27 Apr 2009 Last revised: 5 Jul 2017
Date Written: January 2010
The causal relationship between GDP growth and military spending is a long lasting debate. Earlier studies employing cross-sectional, pooled or fixed effects panel data, report mixed findings. More recent studies employing causality tests also find mixed results. We argue that structural changes in data (i.e. significant economic or military events) play an important role and should not be ignored. We also argue that treatment of stationarity around mean and around time trend needs special attention. Structural changes, if ignored, have the potential to render long-run tests through VECM biased. Mistreatment of trend stationarity may result in spurious results. With a sample of 65 countries, for the time period between 1975 and 2004, we analyze the relationship between military spending and GDP growth with special emphasis to structural change and stationarity around time trend. For the Granger causation through VAR, we test GDP, military spending, government size and openness for stationarity around mean and around time trend using augmented Dickey-Fuller (GLS) test. While trend stationarity is a common trait of the variables, we also find that there are structural changes in the variables' time trends. Considering the bias towards rejecting stationarity in the presence of structural changes, we also employ Zivot-Andrews unit root test. We find causal relationship between military spending and GDP growth for 54 (of the 65) countries. Based on this evidence, we estimate panel VAR estimation for Granger causation for the entire sample and find bi-directional positive causal relationship.
Keywords: Military spending, defense burden, economic growth
JEL Classification: P16
Suggested Citation: Suggested Citation
Dicle, Betul and Dicle, Mehmet F., Military Spending and GDP Growth: Is There a General Causal Relationship? (January 2010). Journal of Comparative Policy Analysis, Vol. 12, No. 3, pp. 311-345, 2010. Available at SSRN: https://ssrn.com/abstract=1395007