Don’t just click 'download': The case of U.S. military expenditure data

Abstract

With a view toward two purposes, the article examines measures of United States military expenditure. It first discusses what types of data would be most suitable for the analysis of the economic burden of such expenditure and it explains, second, why existing databases with U.S. (and global) military expenditure coverage have limited validity, and therefore utility, for the first purpose. The article advances the concepts of minimal and maximal augmented military expenditure measures to better capture the full economic resource burden imposed on an economy than do the presently available measures. The difference of current measures to the proposed augmented measures is not primarily one of an intercept change but mostly of a slope change. The article claims that, by fiscal year 2018, the economic burden imposed by military expenditure on the U.S. economy when using the augmented measures is about double the size of the burden as measured by all of the currently available measures. Since, to date, most empirical work on the effects of military expenditure on economies tends to rely on inappropriate measures, misleading findings may have resulted.