Managing risks, side payments, and multi-institutional enlargement: the role of US defence, big four investment agreements and candidate risks on NATO and EU enlargement

ABSTRACT

Risky investments under incomplete information (i.e. about future behaviour, future state of the world) characterize international institution enlargements. In clubs, a logic of lowest common denominator accompanies member heterogeneity as weaker states reduce collective good quality to their suboptimal level. Disturbances in a good’s provision are attenuated if a club provides diverse goods, however reductions are problematic when it produces fewer/scarce goods. Heavily invested, stronger partners may offset candidate risk through bilateral contracting. NATO & EU enlargement after the cold war provides a set of states to examine the effects of risks and bilateral (security/economic) investments on membership offers in either or both organizations providing a test of rationalist claims. Despite distinct criteria, perceptions of candidate’s (political or investment) risk shaped when offers were given. Finally, research treats enlargements as separate despite simultaneous occurrence. This research contributes by (1) specifying a simultaneous model of enlargements linking the processes indirectly and (2) examining risks and side payments. Results confirm security side payments increased chances of a NATO offer, but economic equivalents had no effect on EU offers. Reducing risks increased the likelihood of offers. Finally, the unmeasured factors correlated with offers in both were positive and substantively significant.

KEYWORDS:

• Enlargement
• EU
• NATO
• bargaining
• rational institutionalism
• bivariate probit model

Acknowledgements

Professor Kimball is an associate professor of political science and director of the Center for international security at the École supérieure d’études internationals at Université Laval and co-director, Security of the Canadian Defence and Security Network. The author thanks Christian Picard for data assistance. Helpful comments were integrated from a previous version presented at the 2019 Canadian Political Science Association meeting, Vancouver, BC.

Notes

1 Imagine the strongest club partner increased investment in a candidate, increasing from 3 contracts in 1999 to 5 by 2003 and negotiating 9 additional contracts (Kavanagh, 2014). Big four equals US, UK, France and Germany.

2 An oligopoly with one major supplier of defence investments, the US.

3 Formal cooperation consists of a formal treaty agreement requiring legislative ratification; see Abbott and Snidal (2000).

4 See Kimball (2017).

5 The US defence investment contract ‘playbook’ included Status of Forces, technical/classified defence information sharing, military training and exchange, defence investment, RDT&E, and other DSA together constituting a public declaration of candidate quality.

6 A dozen states that negotiated to join both institutions are studied herein.

7 Leaders were concerned with political survival as well. Snider argued, ‘the political arrangement and institutions that help leaders stay in office are not necessarily the ones that promote growth and prosperity’ concluding, ‘the increased chances of an economic crisis are the acceptable price to pay if it means avoiding a political crisis which challenges a leader’s hold on power’ (2007, p. 206).

8 On the EU, Schimmelfennig and Sedelmeier (2005), Schwarz (2016); on NATO, Kay (1998); Crisen, 2000; Reiter, 2001.

9 Piedrafita and Torreblanca present the rationalist among three logics of EU enlargement with a collective identity logic (i.e. democratic character) following a logic of appropriateness, and a deliberative framework where actors justify ‘policy positions in terms of some universally valid principles, rather than in terms of relative power or costs and benefits,’ (2005, p. 31).

10 That model covers a limited timeframe (1997–2010) and the choice of estimator (regression) is suboptimal with the dependent variable as a count ranging from 0 to 2; a Poisson model is recommended though employing a count variable is debatable.

11 ‘If strategic complementarity is sufficiently strong enough in treaty ratification,’ then spillover causes clustering (Wagner, 2016).

12 One way to distinguish mandates is the EU considers ‘engaging in war’ NATO’s domain (Smith & Timmins, 2000, p. 85).

13 Defence and security are not in the policy areas considered the exclusive competences of the EU. Security and justice are shared competences. See European Commission, 2019 at https://ec.europa.eu/info/about-european-commission/what-european-commission-does/law/areas-euaction_en, Accessed: 5 March 2019.

14 NATO. 2020. ‘Partnership for Peace,’ https://www.nato.int/cps/en/natohq/topics_50349.htm. Accessed 29 July 2020.

15 Georgia’s commitments under the Individual Partnership Plan, 2004–2006, Ukraine IPAP 2002–2004, Montenegro IPAP 2008–2010; copies available upon request.

16 Redistributing EU decision-making power in anticipation of new partners was a bargaining issue (as its decision-making apparatus is more decentralized) which NATO avoids with consensus decision-making and power centralization on the North Atlantic Council.

17 Weak economic performance produced non-membership combined with bad reform policy (Schwarz, 2016).

18 Slovakia and Bulgaria’s investment profiles equaled the Czech Republic’s (PRS Group, 2019) though only Slovenia and Slovakia were lower political risks by less than 1 point and a 5-point gap to the next less politically risky, Estonia (75 per cent).

19 Smith and Timmons clarified NATO was both deterrent and reassurance commitment, i.e. the US would be there ‘just in case’ similar to a caretaker over a child or an invalid. ‘It is precisely this kind of vague but vital military reassurance that CEE states are seeking from the West, specifically from the US, through NATO,’ (2000, p. 86).

20 Realist and materialist approaches contend state survival through security as a political priority dominates prosperity. Other reasons were guaranteeing fragile democracies and sociopsychological issues, i.e. perception of being functional equals in Europe (Wiarda, 2002, p. 179).

21 Meeting the criteria resulted in both tying hands and sinking costs for candidates though the signals are distinct.

22 Most models of EU enlargement include wealth (GDP/capita) though the required legal and economic reforms promoted wealth resulting in endogeneity; employing investment risk measures mitigates such concerns.

23 The de jure and de facto delegation of budgeting and military command authority to the Executive in democracies facilitated the implementation of the reforms required for NATO criteria with minimal legislative intervention increasing speed and efficiency. Whereas, the collaborative process across policy domains for EU liberalization was delayed from complex multi-actor bargaining.

24 Zorn et al. (1997) model the timing of Senator support for the NAFTA agreement highlighting the role of side payments.

25 The managerial school of compliance (von Stein, 2017) argues some involuntary failures result from poor management or low capacity. In cases of high uncertainty regarding future compliance, side payments are useful to increase capacity.

26 DSA by the executive are implemented via delegation to the actors involved in the ministry and firms/projects may be identified for investment as well resulting in lower implementation costs and a ‘direct’ route to the economy among other positive externalities. See Kimball, 2018.

27 For investment risks, Shanks (1986), Jakobson (2010), Baas (2010); on political risk, Snider (2005), Jarvis and Griffiths (2007).

28 Albania, Bulgaria, Croatia, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Slovakia, & Slovenia (i.e. E12). NATO’s accession institution, the Partnership for Peace, was established in 1993. Based on formal accession dates.

29 Independent variables lagged. See appendix for details.

30 See PRS Group, 2019. https://epub.prsgroup.com/list-of-all-variable-definitions. Data from 1994 to 2016. Accessed: 25 February 2019.

31 PRS data employed due to its transparency, ease of access, temporal coverage, and intent to measure distinct aspects of risk.

32 There is moderate correlation (.553) between the risk measures. Comparative testing shows each contribute distinctly to each model. The correlation between the measures of democracy and investment profile is <.5. Model 8 shows stable performance of risk measures when controlling for the direct influence NATO offers on EU entry.

33 Excludes the US and Canada.

34 Data cover 1949–2007 updated to 2016 by author.

35 Kimball (2006)’s application of the bivariate probit model examines alliance formation and conflict initiation. Recalling the temporal period is <30 years, no cubic splines appear because there are no temporal knots.

36 Risk data limit observations to 250.

37 NATO entry preceded EU entry in all cases.

38 Interpretations of results are calculated from model 9.

39 Panel b of Figure 3 includes confidence intervals.

40 For example, Ukraine’s status despite its capacity was due to the risk of provoking Russia.

41 US Congressional Budget Office. 2001. ‘NATO burden-sharing after enlargement,’ CBO Papers: August, Washington, DC: 30–31.

42 The 2004 EU round added 8 states in total.

43 The US acted as a mortgage-based security organisation reducing risks and absorbing expected externalities of risky partners.