ABSTRACT
Second, lower income groups are the main people who carry the burden of defense and the heavy defense spending hinders the establishment of more welfare programs. By contrast, some studies argue that military spending may render a positive effect on
investment, economic growth, and welfare (Benoit, 1973, 1978; Lindgren, 1984; Ram, 1995). The defense spending is believed to be helpful to human capital formation in education and health because defense personnel and conscripts are well trained physically and receive good skills education. Besides, military spending may lead to technological innovations and even spin-offs in defense field. Therefore, the positive trade-off relationship (i.e. complementary relationship) has also been found in the literature; e.g. Verner (1983), Harris and Pranowo (1988), and Kollias and Paleologou (2011). It is noted that the type of welfare spending may play an important role in the trade-off relationship. Yildirim and Sezgin (2002) and Ali (2011) suggest that the trade-off is positive between education and defense spending, while the relationship is negative between health and military expenditures. Lastly, in addition to the negative or positive guns-butter argument, no trade-off relation-
ship between military and welfare spending has also been found in the literature as well (Caputo, 1975; Russett, 1982; Domke et al., 1983; Eichenberg, 1984; Hess and Mullan, 1988; Mintz, 1989; Davis and Chan, 1990; Frederiksen and Looney, 1994). For instance, Russett (1982) argues that no systematic trade-off between military spending and federal health and education expenditures is found, nor there is any significant depressing effect on health and education expenditures. Domke et al. (1983) utilize three-equation model controlling for the variety of possible determining factors of public resource allocation, and finds that no patterns of trade-off emerges. Mintz (1989) concludes that, although there may be years in which R&D or procurement spending increased and welfare spending decreased, the pattern is not common enough to indicate a trade-off between these spending categories. Those studies suggest that the budgetary allocation of defense and welfare expenditures seems to be driven by separate determinants. An increase in defense spending does not necessarily lead to a sacrifice in health or education expenditure. To sum up, numerous studies have attempted to investigate the trade-off relationship
between defense and welfare spending, while there is no consensus on the empirical findings. As indicated in Section 1, the existing empirical studies are rather limited in terms of data and econometric methods, that hampered the efforts on empirical discussion of the validity of guns-butter argument. This study revisits the warfare-welfare trade-off by taking advantage of a panel of OECD countries to control for unobserved country-specific effects and employing the panel GMM method to allow for potential simultaneity issue.
