In the 1980s much scholarly attention was paid to estimating the impact of Third World security sectors on domestic economic performance. Generally speaking, three branches of the pertinent literature can be identified. One branch consists of case studies of arms-producing less developed countries (LDC APs) to assess the level of sophistication and the volume of arms produced in LDCs as well as to explore the economic consequences of'indigenous arms production.' A second branch investigates the impact of LDC military spending on specific macroeconomic aspects, such as the impact on inflation, the impact on employment, budgetary trade-offs, military research and development (R & D) and technology transfer issues, and so on.2 The third branch, covering possibly the most prominent and most widely researched topic, directly relates military expenditures to rates of economic growth, sometimes as time-series for specific countries, more often as cross-sectional studies for a given year. The dominant finding is that increases in military expenditures are causally and negatively related to rates of economy-wide growth.3 A number of scholars report contrary findings, however. Some, for example, Chan, Hsiao, and Keng, Evans, as well as Biswas and Ram, find that defense spending and economic growth appear unrelated. Others, most prominently Frederiksen and Looney (summarized, e.g., in Looney), discover that the defense-growth relation differs across country groups of interest.4 Frederiksen and Looney's work in particular submits that the negative impact of military expenditures on economic growth is mitigated in the case of those LDCs that engage in some arms-production activities. But why does this effect occur? I suggest in this article that in the case of LDC APs some part of military expenditures must be invested in the domestic economy for