Burden Sharing in CSDP Military Operations

Abstract

Military burden sharing has been a subject of repeated debates in NATO and the UN. Despite more modest goals, the European Union’s (EU) Common Security and Defense Policy (CSDP) has experienced no fewer difficulties in garnering men, money, and materiel. While this may not come as a surprise, the fact that some EU member states have carried disproportionate shares of the burden of CSDP operations is a puzzle that remains unaccounted for. We address this gap by analyzing determinants of contribution levels to CSDP operations. In employing an innovative multi-method design that combines insights from collection action theory with those from integrated theories of military burden sharing, our results indicate that EU countries tend to contribute in positive disproportion with their capabilities when they have a strong peacekeeping tradition and elections are distant. In contrast, they undercontribute when small trade volumes with the area of operations combine with a weak peacekeeping tradition.

Keywords:

• Burden sharing
• CSDP
• defense policy
• ESDP
• military operations
• multi-value Qualitative Comparative Analysis
• peacekeeping

Acknowledgement

Previous versions of this article have been presented at the ECPR Joint Sessions of Workshops, University of Antwerp (Belgium), April 2012; the Annual Conference of the Swiss Political Science Association, University of Zurich (Switzerland), January 2013; the conference “Applications and Methodological Challenges in QCA”, Goethe University Frankfurt (Germany), November 2013; the General Conference of the European Consortium for Political Research, University of Glasgow (UK), September 2014; and the Annual Conference of the University Association for Contemporary European Studies, University of Deusto (Spain), September 2015. We thank Ruth Beckmann, Klaus Brummer, Markus Jachtenfuchs, Christiane Kraft-Kasack, the participants at the above-mentioned events as well as the anonymous reviewers and the editor of Defence and Peace Economics for their very helpful comments and suggestions.

Notes

1. For reviews of literature on NATO and ad hoc coalition burden sharing, see Sandler (1993) and Oma (2012). Major works that include UN peacekeeping operations are Bove and Elia (2011), Gaibulloev, Sandler, and Shimizu (2009), Khanna, Sandler, and Shimizu (1998), Lebovic (2004) and Shimizu and Sandler (2002).

2. The European Security and Defense Policy was renamed to Common Security and Defense Policy in the Treaty of Lisbon. The new name is used throughout this article.

3. The Petersberg tasks include joint disarmament operations, humanitarian and rescue tasks, military advice and assistance tasks, conflict prevention and peace-keeping tasks, and tasks of combat forces in crisis management.

4. EUFOR RD Congo will be abbreviated to EUFOR Congo, EUFOR Chad/CAR to EUFOR Chad.

5. A detailed description of these burden-sharing problems is provided in the online appendix.

6. We focus on ground-troop missions because all member states possess the capabilities to contribute, unlike the naval assets required for EU NAVFOR Atalanta, for example.

7. Military personnel in Concordia was 350, in EUTM – a military training mission for Somali security forces – 45, and in the EU support mission to AMIS only 17.

8. Analyses of the reasons why some CSDP operations were of a very limited size, why France assumed the bulk of the costs of operation Artemis instead of trying to share the burden or why the CSDP took over from NATO in Bosnia would also be worthwhile, but fall outside the scope of this article.

9. To the best of our knowledge, the only study so far that has made use of QCA in the context of European security and defense cooperation is Thiem (2011).

10. For example, some high-ranking officials in the German defense ministry privately feared that if Germany took on the leading role in EUFOR Congo and the mission would fail for reasons of insufficient strength, they would have to endure Washington’s scornful laughter. See ‘Die Kongo Falle’, Der Spiegel, 26 March 2006, 30.

11. We do not summarize the tenets of either model herein as this has been done in previous work. See Hartley and Sandler (1999, 666–668); Khanna, Sandler, and Shimizu (1998, 181–183); Sandler (1993, 448–474), Sandler and Murdoch (2000, 300–304), Shimizu and Sandler (2002, 655–657) and Shimizu and Sandler (2010, 1480–1481).

12. See the online appendix for details. In the context of the Iraq War, Mello (2012) also takes account of the deployment phase. This criterion is irrelevant for our purposes as the only major late deployment or early redeployment occurred when Sweden pulled most of its troops out of EUFOR Chad before the end of the mandate.

13. In line with Forster and Cimbala (2005, 22), we consider leadership responsibility to be a cost, not a benefit. Especially in the case of EUFOR Congo, states were very reluctant to assume the position of operational commander (see also online appendix).

14. We use the official start of the mandate as stipulated in the Council Joint Actions rather than the date of initial or full operational capability since troop mobilization and transportation already add to a country’s burden.

15. Other schemes may be applied, but we consider ours a plausible baseline. In both operations, logistical units were mainly responsible for air transport and medical units for setting up field hospitals. In contrast, combat units could participate in military interventions during EUFOR Congo and took on risky patrolling missions in EUFOR Chad (see online appendix).

16. Turkey contributed 16 troops to EUFOR Congo, Albania and Russia 130 troops to EUFOR Chad.

17. Dorussen, Kirchner, and Sperling (2009), Khanna and Sandler (1996), Khanna, Sandler, and Shimizu (1998), Shimizu and Sandler (2002, 2010) and Siegel (2009). In contrast, Sandler and Murdoch (2000) and Sandler and Shimizu (2014) use Spearman coefficients, while Solomon (2004) conducts Wilcoxon rank tests. As to software, we use the ppcor package for the R environment by Kim (2011).

18. A state with a long tradition of international peacekeeping provides a public good to the global community of states but may derive private benefits through its reputational status within that community (Shimizu and Sandler 2010, 1481).

19. In contrast, Andersson (2002, 380), doubts that trade considerations have influenced commitment to UN peacekeeping operations, and Gegout (2009, 204) concludes that French trade interests have been irrelevant in EUFOR Congo.

20. Cumming (2011, 566–567). Decisions to contribute can also be based on current expectations of future trade opportunities. For example, conservative parliamentarians in the German Bundestag argued that participation was important as Congo possessed some of the largest copper and tantalite reserves worldwide. See Bundestag plenary protocol 16/36, 19 May 2006, p. 3105.

21. For a review of the extensive literature in this area, see Aldrich et al. (2006).

22. In contrast, Domke (1991) argues, based on data for six advanced industrial democracies over the years 1955–1985, that a state’s budget situation does not influence its defense spending.

23. Constitutional restrictions are irrelevant in the context of the CSDP, since all EU member states participate, except Denmark.

24. Detailed descriptions are provided in the online appendix.

25. For binary factors, each level represents the Boolean negation of the other level.

26. NATO usability targets for ground forces consist of a 40% deployability target and an 8% sustainability target.

27. Foreign direct investment (FDI) may be more important than trade volume given that FDI cannot be easily diverted to safer markets (Fordham 2008, 745). However, given that corresponding data are not easily available for the bulk of our cases, and that trade and FDI are likely to be highly correlated, we consider trade volume the best available proxy.

28. For operation EUFOR Chad, trade volume figures take into account exports and imports for both Chad as well as the Central African Republic.