Empirical Test of Non-Linear Relationship between Military Spending and Unemployment

Empirical literature on the relationship between military spending and unemployment presents heterogeneous and different findings. A group of studies show that military spending can help reduce unemployment, while another group has concluded that military spending exacerbates unemployment. A group also shows that there is no relationship between military spending and unemployment. According to this, Lee (2021) using a theoretical model including supply-side effect, demand-side effect and security effect caused by military spending in the form of a random endogenous model shows that the relationship between military burden and unemployment (employment) is non-linear and U-shaped (U reverse). In this regard, the main goal of this article is the empirical test of this hypothesis in 171 countries of the world during the years 2001-2020. For this purpose, the effect of military burden on the unemployment rate has been modeled by using the theoretical literature and applying the threshold panel method.
The results show that the effect of military burden on unemployment is non-linear and in the form of a U-shaped relationship. The threshold level value of military spending is also estimated at 1.61% of GDP. In other words, before the threshold level, the military burden had a negative and significant effect on the unemployment rate; But after crossing this threshold level, the military burden has a positive and significant effect on the unemployment rate, which confirms the hypothesis of Lee (2021).