Military Spending, Conflict and External Debt in Sub-Saharan Africa

ABSTRACT

The damaging economic effects of the debt crises on Africa in the late 1980s encouraged considerable research on the determinants of external debt in developing economies. Although sub-Saharan Africa's (SSA) debt was cut by two-thirds by 2008, through two debt relief programmes, debt in the region has since been rising at an increasingly rapid pace. This study provides an empirical analysis of the determinants of external debt in SSA over the period 1960–2016, using dynamic panel methods. It also considers two potentially important factors that have received relatively little attention. One is military spending, rarely considered, despite a number of well-publicised scandals over the procurement of unnecessary and expensive high-tech weapons systems. A second, is the possibility that the countries studied have been involved in conflict. The empirical results point to a positive impact of military spending on external debt, but with some evidence of heterogeneity across the countries. Furthermore, findings indicate that the positive effect of military expenditure on debt becomes more marked in countries that have been affected by conflict. These results imply that policies to improve security and reduce military spending could be beneficial in reducing external debt and, potentially, improving economic performance in the region.

KEYWORDS:

• Military spending
• external debt
• SSA
• conflict
• dynamic panel methods

JEL CLASSIFICATION:

• H63
• H56
• O55

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. The list of the 47 SSA countries included in the full sample can be found in Table A1 in the Appendix.

2. There were some missing observations in the SIPRI series in the body of the series. Rather than lose these runs of data, where there were 3 or less these were extrapolated from the before and after values.

3. Uppsala Conflict Data Program (UCDP) and International Peace Research Institute Oslo. See http://ucdp.uu.se/#/

4. The list of SSA countries that have been in conflict over the period 1960–2016 can be found in Table A1 in the Appendix.

5. The calculated long run coefficients on the military burden variable increase from about 0.3 to 0.4 for the conflict affected countries when the full model is considered and from 0.88 to 0.93 when the restricted model (without exports and reserves) is considered.

6. The list of SSA countries that are included in the balanced panel appear in Table A1 in the Appendix.

7. The list of SSA countries that are included in the balanced panel can be found in Table A1 in the Appendix.