Arming the Peace: Foreign Security Assistance and Human Rights Conditions in Post-Conflict Countries

ABSTRACT

What are the effects of foreign security assistance on the quality of the peace in post-conflict countries? Despite the stakes, and the tremendous amount of weaponry and other forms of foreign military aid flowing to governments of post-conflict countries, the academic literature provides little guidance as to what effects policymakers and practitioners should expect from this type of aid. Military assistance provided to the government of a country emerging from the turmoil of civil war could enable the state to establish a monopoly on the legitimate use of force, leading to a more durable peace and greater human security. However, we contend that significant flows of military aid and weapons from foreign governments may encourage regimes to adopt more repressive approaches to governance. We investigate the impact of security assistance on human rights conditions after 171 internal armed conflicts that ended between 1956 and 2012 using a novel measure of military aid and an instrumented measure of weapons transfers. We find strong evidence that both military aid and arms transfers to post-conflict governments increase state repression.

KEYWORDS:

• Foreign aid
• military aid
• arms transfers
• human rights
• repression
• H11
• H41
• H56
• F35

Acknowledgements

Earlier versions of this paper were presented at the 2017 meeting of the International Studies Association and the “Political Economy of Power Projection” conference organized by Jonathan Markowitz and Erik Gartzke at the USC Center for International Studies. We thank the participants and, in particular,  Michael Kenwick, Marina Henke, Paul Poast, Chris Fariss, Eugene Gholz, Therese Anders, and Harris Mylonas for helpful comments. The paper was also improved by thorough and insightful feedback from the DPE Editorial Board and anonymous reviewers. We are grateful to Frances Duffy, Menevis Cilizoglu, and Ghazal Dezfuli for exceptional research assistance and Oliver Pamp for providing early access to replication data. This study stems from a project on the long-term impacts of lethal aid Sullivan began as an Andrew Carnegie Fellow (2015-2017).

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. CIA World Factbook. ‘South Korea’. Accessed at: https://www.cia.gov/library/publications/the-world-factbook/geos/ks.html

2. It is important to note that all aid is fungible to some extent (McGuire 1978) and leaders could use foreign security assistance to reduce state spending on the military and reallocate the savings to public goods provision. However, scholars have found evidence that differences in the fungibility of different types of aid can have divergent effects on government spending. Feyzioglu, Swaroop, and Zhu (1998), for example, find that governments receiving aid for education, agriculture, or energy reduce the share of their own resources allocated to these sectors and spend the savings elsewhere. Less fungible aid earmarked for transportation and communications projects tends to be spent on those sectors. We follow a similar path here, arguing that aid to a state’s security sector may create stronger incentives for leaders to choose more repressive strategies for retaining power than other types of aid because of its direct effect on repressive capacity. We believe it is unlikely that post-conflict governments will reduce their own security sector spending in response for foreign security assistance because leaders are particularly vulnerable to domestic threats after internal armed conflicts, non-tax sources of revenue tend to reduce incentives to invest in public goods, and military spending can be used to reward military loyalty. In fact, most research to date has been concerned about the reverse – the risk that countries will use economic aid to increase military spending (Langlotz and Potrafke 2016, Kono and Montinola 2013).

3. The volume of arms transfers to a country may also be more reflective of domestic industrial policy or commercial interests in weapon-producing states than of their foreign policy concerns. This is an important issue addressed in other scholarship. We do not think the motivation of arms providers should change the impact of these transfers on recipients except in the sense that arms exporters may be less selective about the governments they transfer weapons to if they are not viewing arms exports as a foreign policy tool.

4. Each variable is the natural log of the TIV multiplied by 100. SIPRI only records arms trades where the TIV of the weapons is at least .5 (U.S.$500,000).

5. The Historical Phoenix Event Data are described in more detail in Althaus, Scott, Joseph Bajjalieh, John F. Carter, Buddy Peyton, and Dan A. Shalmon. 2017. ‘Cline Center Historical Phoenix Event Data Variable Descriptions’. Cline Center Historical Phoenix Event Data. v.1.0.0. Cline Center for Democracy, June 30. http://www.clinecenter.illinois.edu/data/event/phoenix/.

6. We use a two-year period (rather than simply a one-year lag) because the training, equipment, weapons, and other capacity-building goods and services provided by an external state should continue to benefit the security sector for some number of years after aid is delivered.

7. Using only the executive constraints measure from the Polity IV dataset, rather than the full democracy index, allows us to avoid potential endogeneity. Scholars have noted that several components of the democracy index implicitly capture human rights protections in the measure (Hill and Jones 2014, 677). We hold executive constraints constant at its level in the year after conflict termination because including a time-varying measure runs the risk of introducing post-treatment bias. We predict that security assistance will increase state repression, but it may also impede democratization by similar mechanisms.

8. Our instrumental variables regression models are estimated with the ivreg2 Stata module in Stata 14 (Baum, Schaffer, and Stillman 2010). As recommended by Baum, Schaffer, and Stillman (2007), we use the Continuous Updating Estimator (CUE) to improve the efficiency of IV-2SLS estimates with clustered standard errors.

9. Models including control variables not appearing in these models are included in the online appendix.

10. Although the executive constraints measure is a 7-point scale, not a continuous variable, including the measure as a series of category dummies does not significantly improve the fit of the model or change our substantive results. We treat the variable as continuous for parsimony. Similarly, models representing elapsed time as a count of post-conflict years, as we do in the models in Table 2, performed on par with models employing various transformation of elapsed time. There is no evidence of a nonlinear relationship.

11. Recent scholarship suggests the impact of peacekeeping operations varies with the number of troops committed, the nature of the operation (e.g. Chapter VI or Chapter VII, traditional peacekeeping or multidimensional peace enforcement), and whether the intervention is conducted by the United Nations. See Doyle and Sambanis (2000), Hultman, Kathman, and Shannon (2013), and Solomon (2007) for discussions of salient differences among peacekeeping operations.

12. The PTS measures levels of state-sanctioned killings, torture, disappearances and political imprisonment. The CIRI physical integrity index measures the prevalence of torture, extrajudicial killing, disappearances, and political imprisonment.

13. This figure includes Foreign Military Financing (FMF), International Military Education and Training (IMET), and several smaller funds (U.S. Department of State. ‘Foreign Military Financing Account Summary.’ 30 June 2016). The majority of this aid goes to approximately 60 countries in the form of financing for the purchase of American weapons. In 2016, the U.S. purchased $10 billion worth of weapons for other countries. (DefenseNews.com, 29 November 2017).

14. See, for example, Dan Lamothe. ‘The Afghan Air Force is Growing. So Are Questions about Its Actions in Combat.’ Washington Post. 30 July 2018; Shane Bauer. ‘Iraq’s New Death Squad: America has built an elite and lethal counterterrorism force. But who’s calling the shots?’ The Nation. 3 June 2009; Human Rights Watch, ‘Iraq: Secret Jail Uncovered in Baghdad,’ Human Rights Watch, 1 February 2011, http://www.hrw.org/news/2011/02/01/iraq-secret-jail-uncovered-baghdad.

15. Loveday Morris. 2016. ‘The Force Leading the Fight against ISIS Went from “Dirty Division” to Golden Boys.’ Washington Post 26 July 2016.

16. Approximate inflation-adjusted production costs in constant-1990 U.S. dollars.

Additional information

Funding

Research for this article was made possible in part by a grant from the Carnegie Corporation of New York [D 15126]. The statements made and views expressed are solely the responsibility of the authors.