Citation

Hess PN. J. Peace Res. 1989; 26(4): 399-412.

Copyright

(Copyright © 1989, SAGE Publishing)

DOI

10.1177/0022343389026004006

PMID

unavailable

Abstract

A simultaneous model for the force ratio, arms imports, and foreign aid receipts is estimated for a sample of seventy-six developing nations and the period 1978-84. Regression diagnostics are applied to identify outlier nations and to assess the sensitivity of the estimated coefficients to multicollinearity and influential observations. The influence diagnostics reveal that the inferences drawn from the empirical analysis are conditional on the composition of nations in the sample. In particular, the statistically significant, direct relationship between real per capita arms imports and foreign aid receipts found for the full sample of seventy-six countries largely reflects the influence of the half-dozen Middle Eastern nations in the sample. The key determinants of the force ratio (armed forces per thousand population) are per capita income and security concerns. Moreover, there are significant differences in the response of the force ratio to income across several of the major regions of the developing world. The force ratio, in turn, is the primary determinant of the level of arms imports. There is evidence of a strong import substitution effect for those developing nations exporting arms. And, whether the US or the USSR is the principal supplier of arms appears to make a difference for the volume of arms transfers. Foreign aid receipts do correspond to need, as measured by an international human suffering index. Nevertheless, outside of the Middle East, where aid flows are correlated with arms imports, a considerable portion of the aid is unexplained by the regression model.