Selling to Both Sides: The Effects of Major Conventional Weapons Transfers on Civil War Severity and Duration

Abstract

Civil wars are primarily fought with small and light arms, but the availability of major conventional weapons to states and rebels can alter the nature of the war being fought. This study explores the impact of major conventional weapons transfers on civil war severity and duration. By using a recipient based approach to arms transfers, I find rebel acquisition of major conventional weapons from international sources leads to conflict escalation and deadlier conflicts. State importation of major conventional weapons is associated with longer conflicts. These findings provide researchers a means to account for rebel capabilities in civil war research and policy makers insight to limit the destructiveness of civil wars.

Keywords:

• arms transfers
• battle deaths
• civil war

Notes

Replication data can be found at http://$$$$dvn.iq.harvard.edu/dvn/dv/internationalinteractions.

¹SIPRI trade registers were generated for each rebel group used in this study which show the types of systems acquired as well as the sources of those systems. The registries include cases where it is known that deliveries have begun or taken place (SIPRI 2010).

²Of the groups covered in the SIPRI data only six were able to acquire some form of aircraft. The aircraft tended to be basic transport aircraft and few in numbers.

³Portable missiles are sometimes also classified as small and light weapons, but this categorization is more focused on their size and portability rather than their technological sophistication. On the battlefield antitank and aircraft missile systems are more akin to major conventional weapons. First, these weapon systems are very sophisticated and cannot be acquired by rebel groups unless they are stolen or provided by an external supplier. Second, they are designed to be used in conventional military engagements against military vehicles as opposed to most small and light arms that are anti-personnel weapons. These weapons systems meet SIPRI's definition for major conventional weapons and are coded accordingly.

⁴Models were also run using a Weibull specification with no discernable difference in the results.

⁵Information from the SIPRI Arms Transfers Database (http://armstrade.sipri.org/).

⁶Diagnostics were run to ensure that the missing cases were not influencing the outcomes. Running the original replication model with only the cases used in this study produces similar results. Since several of the conflicts start within five years of 1950 they are excluded from the third model.