Military spending as another cause of the failure of the PPP

Abstract

In a recent article, with a strong theoretical argument but poor empirical support, military spending was identified to be a determinant of real exchange rate or a factor causing the PPP to deviate from equilibrium exchange rate. The poor empirical results were mostly due to a small number of observations (21 cross-sectional). This article employs pooled data from across 37 developing countries over the 1985–1998 period and provides very strong empirical support for the notion that indeed increased military spending could cause the PPP to deviate from equilibrium exchange rate.

Notes

Among all these factors, productivity differentials originally introduced by Balassa (1964) has received the most attention in the empirical literature. Numerous studies have tested the so called ‘productivity bias hypothesis’ and provided mixed results. For review of these studies see Bahmani-Oskooee and Niroomand (1996) and Bahmani-Oskooee and Nasir (2001, 2002).

Note that we are following Bahmani-Oskooee and Nasir's (2002) approach who investigated the impact of corruption on the real exchange rate. We are replacing their measure of corruption by a measure of military spending. Since military spending is already in percentage terms, one enters it at its level rather than log level.

For list of countries see Table 1.

Such measures are also used by Davoodi et al. (1999: 13) and Gupta et al. (2001) in different research topics.

The authors thank Gupta et al. (2001), especially Mello for providing military spending data.