Small states shelter diplomacy: Balancing costs of entrapment and abandonment in the alliance dilemma

Introduction

It is widely acknowledged that small states approach International Relations (IR) from a position of structural weakness (Handel, 1981). Small states have often been defined as the weaker part in an asymmetric relationship and are therefore ‘stuck with the power configuration and its institutional expression’ (Mouritzen and Wivel, 2005: 4). While recognizing that small states enjoy benefits from international cooperation, the small state literature has also identified various ‘costs’ when they become involved in international organizations and alliances. The literature has therefore suggested that small states face a dilemma in international cooperation, where they have to prioritize between the costs of entrapment and abandonment (Petersen, 1998; Snyder, 1984, 1997). To manage the costs, it has been suggested that small states can adopt a number of strategies to ease the consequences of the power asymmetries. Two strategies have gained prominence in the literature on the so-called alliance dilemma. First, small states can make use of the strategy of hiding, seeking to maximize autonomy by opting out of specific aspects of the cooperation or by setting up ‘bastions’ in the cooperation. Second, small states can utilize ‘bandwagon’ strategies, pursuing adaptation strategies to the more powerful actors in the cooperation (Wivel and Thorhallsson, 2018). ¹ When hiding, it is often implicitly assumed that small states prioritize the abandonment cost over the entrapment cost, while in the second strategy, they prioritize the entrapment cost over the abandonment cost. Until recently, the small state literature has failed to theorize how small states can utilize strategies to balance between the types of costs and have paid little theoretical attention to the actual diplomatic strategies that have guided the small states between the Scylla and Charybdis in the dilemma. This article argues that recent insights from the shelter theory (Bailes et al., 2016; Thorhallsson, 2011, 2018; Wivel and Thorhallsson, 2018) have the unexplored potential to explain how small states are able to negotiate from a position of weakness and, through carefully designed shelter diplomacy, can navigate, balance and reduce the costs of the dilemma. Shelter theory has been designed to explore the consequences of the inhered relational power asymmetry that exists between great powers and small states and has been presented as an alternative to the existing theories of bandwagon and hiding in the IR literature (Wivel and Thorhallsson, 2018). The theory posits that smaller and weaker states can utilize the asymmetric relations to greater powers to their own advantage by using carefully designed diplomatic strategies. For small states, the cost of dependency is considered as a necessary cost that allows small states to enjoy the benefits from bilateral and multilateral relations with stronger powers. However, in opposition to bandwagon theories, shelter theory suggests that small states can maintain a relatively high degree of sovereignty and control in relation to the greater powers as the benefits from cooperation are not one-sided, as often assumed in defensive realist bandwagon theory. At present, the concept of alliance shelter is promising, but it is still theoretically underdeveloped and still needs to be applied and tested empirically. The shortcoming relates to both the theory’s overall relation to the IR-inspired alliance and bargaining theory and European Union (EU) studies (Bailes et al., 2016; Thorhallsson and Steinsson, 2017; Wivel and Ingebritsen, 2019). At present, the actual application of the theory at the midrange level is unclear as it is underspecified how shelter diplomacy works in practice. How can small states utilize the inhered power differences in asymmetric relations to their own advantage in a manner that separates it from purely free riding on the structural powers of others? In short, what are the effects of shelter diplomacy, what type of costs can shelter diplomacy protect small states against and to what degree are shelters useful for small states if the distance in preference between the small state and the larger power widens?

This article addresses these questions and aims to contribute to the further development of shelter theory and the literature on small-state diplomacy by integrating shelter diplomacy in the wider IR literature. The main argument is that shelter strategies can most forcefully be analysed through the lenses of realist alliance theory, as shelter diplomacy shares light on how small states can utilize asymmetric power relations in alliances to balance the costs on international cooperation to reduce abandonment and entrapment costs. The article demonstrates how small states can apply shelter strategies in the European cooperation by reducing and balancing the entrapment and abandonment costs associated with the integration process. Shelter theory thereby has the potential to offer a new middle position between the bandwagon and the hiding strategies that are better equipped to account for how small states balance the cost of the dilemma rather than just prioritizing one cost type over the other. The article also discusses the usefulness of the concept in situations after Brexit, where most of the smaller northernmost member states in the cooperation lost their traditional shelter partner and explore whether we see a transformation from asymmetric to symmetric shelter strategies.

The costs associated with small states’ international engagement

The small state literature has often utilized Snyder’s (1984: 466–468, 1997: 180–192) alliance dilemma as a theoretical tool to explain the costs and risks for small states when engaging in multilateral or bilateral cooperation with greater powers. In the dilemma, (small) states face a situation in which they have decided to join an alliance or institutional arrangement because of the benefits associated with the cooperation. Involvement in such arrangements is difficult for small states because involvement also activates certain entrapment mechanisms as it can be difficult and costly to leave a cooperation such as the EU. In the cooperation, states are assumed to have shared but not identical preferences. To the extent they are shared, they may also be valued to various degrees between the small states and the greater allies. As most alliances and cooperation are asymmetric, this suggests that the larger member states have the capacity and the ability to set the pace and direction in the cooperation because of their structural powers. This direction might differ from the preferences of the smaller states. Smaller member states are therefore assumed to be caught in a ‘cost dilemma’ between abandonment and entrapment costs. Entrapment occurs when the development in an organization changes and turns detrimental to one’s interests – or moves too much in a direction not favoured by the small state. Entrapment therefore implies that small states can be forced to accept new agendas, policies and even further integration that they were not willing to accept originally. In a cooperation such as the EU, with a relative broad range of policy areas and agendas, this fear is often very visible for small states (Snyder, 1984: 466–467). Such a cost can take different forms, but for small states, it has often been argued that the cost is associated with the loss of sovereignty and self-determination. The other side of the dilemma relates to the cost of maintaining the asset of sovereignty at the expense of remaining a member of the international cooperation. This relates to the fear of abandonment. Abandonment is often referred to as the fear of being left behind, losing relevance or being marginalized in the international cooperation. For small states, such a cost can also relate to situations where they are left outside the cooperation, where they do not have the influence on the dimensioning of the cooperation nor have the opportunity to influence decisions that they depend on because of their outsider status. Such a cost relates to the situation that countries such as Denmark or Sweden face as non-members on central elements in the monetary area, where the countries’ economic dispositions hinge on the developments within the Eurozone. Abandonment costs can also relate to situations where states decide to stay in the cooperation only to find themselves placed in an isolated position. The decision to leave or remain will depend on the actual level of the exit costs. The cost of abandoning is often higher than the cost of remaining in the cooperation. Small states in the EU often seek to further their interests by trying to preserve as much autonomy as possible while influencing the actions of the great powers. They seek to expand their influence over the great powers mainly through international organizations, but participation typically reduces political autonomy. Thus, small states face a dilemma between autonomy and influence. The cooperation in the EU aggravates this dilemma by increasing the potential cost and benefits of institutionalization. Costs and benefits are likely to rise continuously as the process becomes still more binding and encompasses even more issue areas. A state facing this dilemma either surrenders autonomy and risks entrapment in a process leading to still more dependency or preserves autonomy and thereby runs the risk of abandonment, forgoing the chance of influence over other states and other gains from integration (Petersen, 1998). Small powers may seek to avoid entrapment by opting out of parts of the integration process, but only at a high risk of abandonment because they have no decisive effect on the integration process or its chance of success (Adler-Nissen, 2009: 57; Wivel, 2005: 396).

Balancing the costs in the dilemma: a model of shelter diplomacy

While the dynamics in the alliance dilemma are relatively well described in the international alliance literature (Snyder, 1984, see also Kelstrup, 2014; Miles and Wivel, 2014; Petersen, 1998), the question of how small states are able to reduce or manage the cost associated with the dilemma is less well described and understood.

The literature has traditionally argued that small states might pursue two overall types of strategies to reduce some of the costs associated with international cooperation. The first type is hiding strategies, which can take different forms as either opt outs from specific parts of international cooperation or ‘bastion building’ (Mouritzen, 2017: 78–80, 2022), where small states protect sovereignty in certain areas (Agius and Devine, 2011: 266). In relation to the costs of the dilemma, hiding strategies are per design aimed at avoiding further entrapment costs as small states gain an opt-out position in areas that clearly marginalize their position in the cooperation. Hiding strategies are thereby employed in situations where abandonment or marginalization costs are evaluated as more desirable than entrapment costs. Bandwagon strategies applied to the dilemma are utilized in situations where the evaluation of the inflicted costs is different, as bandwagon strategies are designed to match situations where small states prioritize entrapment over abandonment costs (Figure 1). In bandwagon, small states are generally assumed to accommodate to the great powers and the demands raised from the alliance hegemons, where the small states follow the agenda of the great powers in the alliance (Oma and Petersson, 2019; Walt, 1987).

OPEN IN VIEWER

Figure 1.

Hiding and bandwagon strategies and abandonment and entrapment costs.

The existing alliance literature has thereby worked from an assumption that the dominant strategies imply that small states must decide between the two types of costs but has offered little guidance on how small states can balance and create trade-offs between the cost types and avoid paying either the full entrapment or abandonment costs in asymmetric power relations, where the small state depends on the great power. The article suggests that shelter diplomacy can address this gap and provide new insights into how small states can reduce and balance the cost of the dilemma by utilizing their relations to the greater powers. Bailes et al. (2016) have suggested that shelters are created in asymmetric bilateral relations where the shelter is provided by a stronger state – or a multilateral organization. Small states are assumed to seek shelter as a means to increase the political, economic and societal resilience as they operate from a position of weakness where their vulnerability forces them to pursue shelter (Thorhallsson and Steinsson, 2017: 26). Small states are therefore assumed to design strategies that allow them to utilize the power asymmetries and their vulnerability to reduce some of the cost associated with their international engagement. Shelter theory is thereby assumed to refer to a series of strategies that small states adopt to alleviate the inherent vulnerabilities of being small and, at the same time, maintain control over its political decision-making (Bailes et al., 2016; Thorhallsson, 2000, 2018; Thorhallsson and Wivel, 2006). Accordingly shelter strategies involve accepting some degree of dependence as a necessary cost to enjoy the benefits of bilateral or multilateral relationships with stronger powers. To be sure, the small states and alliance literatures are aware of the importance of shelter but have so far failed to theorize it and often conflated it with either a bandwagon-for-profit strategy (Schweller, 1994) or considered it a form of free riding on the efforts of the greater member states of the cooperation. However, shelter seeking is distinct from bandwagoning in that it points to the prominence of small-state strategies combining different types of shelters (political, economic and societal) and in stressing the combination of defensive (protection) and offensive (influence) goals in small-state foreign policy (Wivel and Thorhallsson, 2018). Furthermore, shelter theory involves a higher degree of proactivity to be able to utilize the power distribution to their own advantage. Shelter theory is also different from other proactive small-state strategies like Grøn and Wivel’s (2011) smart state theory. The main difference between shelter (especially the asymmetric version) and smart state strategies is that shelter diplomacy is a bilaterally designed small-state strategy, where the small state depends on shelter partners to succeed in balancing the alliance dilemma. ‘Smart state strategies’ belong to the category of unilaterally designed small-state strategies, which in its core is more independent and ambitious because the small states have to rely on their own ability to form agendas, provide ideas and exercise entrepreneurship. Another difference is that smart state strategies are driven by positive incentives that aim to ‘maximize influence’ (at least according to Grøn and Wivel (2011)) rather than avoid costs and penalties, which is the main purpose of shelter theory in a realist interpretation.

Shelter theory is a potential alternative to the solutions of the trade-offs presented in the alliance dilemma compared with hiding and bandwagon strategies. In the following, the article aims to theorize on the more specific effects of shelter diplomacy and offer new insights into what type of costs shelter diplomacy can protect small states against and to what degree shelters are useful for small states when the distance in preference between the small state and the larger power widens over time. The contribution of the article is thereby to integrate shelter diplomacy in the realist alliance literature and demonstrate how the strategy has the potential to balance the costs associated with the dilemma and reduce entrapment and marginalization costs.

Figure 2 illustrates the potential of shelter strategies from the position of smaller states at it might help to elevate some of the challenges involved with international engagement for small states. The figure also illustrates the potential of shelter theory as an alternative to both strategies of hiding and bandwagon, as it opens op for a more nuanced examination of small states’ diplomatic practice.

OPEN IN VIEWER

Figure 2.

Shelter strategies as a mean to balance abandonment and entrapment costs.

Theoretical effects of shelter diplomacy

Shelter theory has not yet reached its full potential as the actual mechanism by which shelter strategies can help to achieve its goal is still underdeveloped. To enhance our understanding of how exactly shelter strategies can operate in practice, the article provides insights into how small states can utilize the structural power of great powers to their advantage by applying the concept of shelter on the dilemma that small states face in the European integration process. In the following, it is assumed that small states participate in the European cooperation because they fundamentally benefit from membership, but also that small states have a preference for maintaining sovereignty and autonomy that makes them less integrationist compared with some of the greater powers in the cooperation, who have a higher chance of controlling the development and direction in the cooperation.

To assess the potential effect of shelter diplomacy, we should first consider a situation of a bilateral negotiation between a small state and a great power. Figure 3 illustrates a situation of potential abandonment and entrapment costs in such an asymmetric bilateral negotiation with the small state and great power B, where the small state is isolated in the negotiations without allies with veto opportunities. The figure shows how a small state in an asymmetric bargaining situation can be forced to accept an outcome *1 close to B’s ideal point, which might entail a high proportion of entrapment costs if it is accepted, but also runs the risks of abandonment costs if they reject an agreement. As it is illustrated, the small state would be left in a situation where it has to choose between either abandonment or entrapment costs, as the integrationist, because of its structural power, would be able to determine the outcome that would reflect the great powers preference. The abandonment or entrapment cost would then be the spatial distance between the small states’ ideal point and the outcome, which, in this scenario, would inflict a high level of either type of cost on the small state depending on whether the small state chooses to bandwagon (full entrapment costs), to hide (some abandonment costs), or break with the cooperation (full abandonment costs).

OPEN IN VIEWER

Figure 3.

Costs associated with an agreement in an asymmetric bilateral negotiation without veto-options for the small state.

Figure 4 illustrates a situation where shelter diplomacy can be utilized to balance the costs of entrapment and marginalization. In this scenario, we can include a second great power (A), which has preferences closer to the small state than great power B. Great power also A holds credible veto power over the outcome. This suggests that the two greater powers have to find a ‘balanced’ compromise to settle for an agreement. In this scenario, we assume that a likely outcome (ceteris paribus) would be a symmetric compromise between the ideal points of the two great powers illustrated in *2. Compared with the situation in Figure 3, the outcome can be expected to be much closer to the preferred outcome of the small state. By sheltering the structural power of great power A, the small state therefore only has to ‘pay’ the reduced costs of the distance between *2 and its ideal point compared with Figure 4, where the spatial distance was between the ideal point and *1. The exact location of the compromise between great powers A and B of course depends on their evaluation of exit costs, but as illustrated, the small state can significantly benefit from seeking shelter by aligning its bargaining position with A as they share somewhat similar preferences, which allows the small state to utilize the structural powers of A to reduce and balance potential costs associated with the dilemma. The effects of shelter diplomacy thereby illustrates that the hypothesized costs of entrapment and abandonment for small states are lower when the country can negotiate in the shadow of the great powers. A basic element in shelter diplomacy is that the small states should be able to navigate and utilize the preference constellation between the two greater powers to their favour rather than rely on their own ability as assumed in the smart state theory (Grøn and Wivel, 2011). As the figure illustrates, we can expect that shelter diplomacy is most effectual in a scenario where the small states share preferences with one of the great powers in the cooperation. In the figure, great power A and the small state both share a prefer-less integration compared with the position of great power B, who has a preference for more integration. The small state does not have to have identical ideal points with the great power. What is important for shelter diplomacy is that the small-state ideal point is relatively closer to the one of the great powers compared with the other great powers in the negotiation. In the model, it is assumed that the two great powers formally and informally have veto power over the scope of the outcome because of their structural powers, while smaller powers in the cooperation do not hold the same type of veto power because of its small-state status (Bickerton et al., 2015; Panke, 2010; Puetter, 2014).

OPEN IN VIEWER

Figure 4.

Preference constellation between small state and two greater powers in a cooperation.

The limitation of this logic is that the position of great power A can only be an asset if a large heterogeneity is not developed between the two countries’ ideal positions. This also points to the vulnerability of the shelter strategy, as it is always contingent on the position of the actor that provides the shelter. From a theoretical point of view, the effects of shelter diplomacy and free riding can be difficult to separate, as both free riding and shelter strategy might entail the same type of benefits. From an empirical point of view, free riding would involve strategies of non-involvement or buck-passing in concrete negotiations, while shelter strategy can include more proactive means to support the shelter actor in the negotiations, without running the risk of paying marginalization costs.

Shelter diplomacy in practice

The relevance of the strategy is illustrated by utilizing the model to study the Danish strategies to manage and balance the cost of the dilemma in the EU after its accession to the cooperation in 1973 until 2019. The central argument throughout the analysis is that it is key to understand how the country historically has attempted to balance the cost of the dilemma and relate to the utility of shelter diplomacy. In the Danish case, this has traditionally been done by aligning the country’s position with larger countries in the negotiations on the direction in the integration process (Mariager, 2012; Pedersen, 2014). For instance, in the 1970s and early 1980s, Denmark often oriented itself towards the United Kingdom and France in questions related to constitutional integration (Pedersen, 2013; Petersen, 2006). They shared Denmark’s scepticism towards further constitutional integration and, thereby, constituted relevant allies in the cooperation. Especially, the ties to the United Kingdom played a central role in the preparation of the Danish bargaining strategy in the various intergovernmental conferences, not least after a shift towards pro-integration in the French positions around 1982 (Petersen, 2006).

The orientation towards the United Kingdom meant that the less integrationist British positions often came to serve as a shelter for Denmark, which tried to design bargaining strategies that allowed the country to negotiate in the shadow of the United Kingdom towards further institutional integration (Pedersen, 2013). By aligning with the British on central issues, the country has thereby avoided to pay the full costs of abandonment and marginalization due to the credibility of a potential British veto. In situations where the United Kingdom has not provided vetoes, they have, given their structural position in the cooperation, often been able to forge compromises that have been close to Danish positions and preferences. This means that the country could enjoy the relative benefits from the British opposition without upholding the full costs of opposing the French and German axis in the cooperation.

The relevance of the shelter has changed and developed over time as there have been changes in both the Danish and British positions. This has reduced the benefits of the shelter diplomacy, as the United Kingdom became too sceptical relative to the Danish positions. In the following, four different phases are illustrated that represent different situations where the relationship between Denmark and the United Kingdom has changed, resulting in a change in the cost associated with the Danish dilemma. In the following, it is argued that there are three central phases in the Danish shelter diplomacy and its relation to the United Kingdom. The first phase is the engagement phase, largely covering the period from 1973 to 1992, where there was a high degree of correspondence between British and Danish positions on issues related to constitutional integration. Denmark was able to rely on the British scepticism to avoid too high costs related to the integration process. The second phase is the partnership phase, running from 1993 to 2016, where a greater dissonance in the two countries’ positions gradually evolved. In this period, the United Kingdom continued to play an important role for Denmark regarding the reduction of the abandonment cost related to the Danish opt outs. Throughout the period, we saw a greater discrepancy between the British and Danish positions, which meant that the strategic relevance of the British was reduced throughout the period. The third phase is the separation phase, marking the period around Brexit from 2016 to 2020, where we have seen larger dissonance in the positions between Denmark and the United Kingdom. The final phase is described as a divorce phase, where the British positions no longer play a strategic role in balancing the costs related to the dilemma. The article argues that in this phase, we have seen the development of a more complex version of shelter diplomacy, where Denmark has engaged in a symmetric shelter strategy in which it has been forced to align itself with smaller member states on a more ad hoc basis to continuously balance entrapment and abandonment costs. We saw this, for instance, in the 2020 budget negotiations, where Denmark, as part of the Fugal Five group, was able to reduce entrapment costs without suffering too heavy abandonment costs. We will discuss the dynamics of the phases in more detail in the following.

Phase 1: engagement

In the early post–Second World War period, a central theme in Danish foreign policy was how the country should position itself towards Europe and the emerging bipolar order. Denmark chose to align itself with British positions, as the country was the most important market for Danish export, and decided to stay out of the early phases of the European cooperation and integration process in the 1950s (Wivel, 2018). Denmark gradually developed an interest in joining the EEC, partly to gain access to a larger market because of its large agricultural production, which would allow the country to enjoy the benefits from the intensified cooperation. The British decision to apply for membership became decisive since Denmark depended on the large export of Danish agricultural products to the United Kingdom. At the same time, the United Kingdom was considered a central ally in relation to Danish interests in free trade, a strong Atlantic tie, a continuous important actor for the national state in the cooperation and a guardian against the continental powers’ dominance over Europe (Pedersen, 2014). This is part of the reason that Denmark chose to apply for membership of the EEC in 1961 together with the United Kingdom, Norway and Ireland, but decided to withdraw the application after the French parliament refused to ratify the bill that would have secured British membership of the EEC. Denmark reapplied together with the United Kingdom and Ireland in 1970 and joined the EU on 1 January 1973, after a referendum in 1972. In the 1970s and early 1980s, Danish EEC policy was primarily aimed at keeping a low profile on institutional integration while actively exploring the political and economic benefits of gaining market access to the Common Agricultural Policy (Nedergaard, 2014; Petersen, 2006: 63–64, 195). On the question of constitutional integration, the Danish position was to secure status quo of the institutional framework from 1972 and, thereby, to guard the intergovernmental nature of the cooperation against supranational tendencies. The premise in the Danish negotiation strategy was that the country had accepted EEC membership under some certain ‘institutional preconditions’ that were not to be changed or altered (Pedersen, 2013). Theoretically speaking, the Danish positions were aimed at minimizing the risk of further entrapment costs, or at least keeping them at a tolerable level. Until the mid-1980s, shifting Danish governments regularly argued that there was no need to amend the Rome Treaty as long as the existing institutional framework was not fully utilized politically (Pedersen, 2013). Throughout the 1980s, such positions were sustainable because of the British (and initially also French) resistance towards further institutional integration (Pedersen, 2013, 2014). This ensured that Denmark had strong allies and could utilize shelter strategies or simply just free ride on their scepticism. In this period, Denmark often sought to coordinate its bargaining positions with the British to avoid becoming isolated in the EEC negotiations – or at least sought to resist institutional changes in the shadow of the British scepticism. For instance, Denmark used this strategy during the negotiations in the Dooge commission in 1984 (Pedersen, 2013; Petersen, 2006), where Denmark tried to avoid the risk of further institutional entrapment and, at the same time, minimize the risk of being isolated in the cooperation. Theoretically, the period represented a situation where Denmark and the British had more or less the same preferences regarding further institutional integration, which meant that Denmark had a close ally on the issues, helping to balance and reduce the entrapment and abandonment costs. This situation can be illustrated in Figure 5, in which the expected outcome of negotiations over further institutional agreement would reflect outcomes similar or close to the British – and Danish – ideal positions because of the credibility of a British veto in the period.

OPEN IN VIEWER

Figure 5.

Shelter diplomacy in the engagement phase.

The official Danish preference for maintaining the institutional status quo was put under pressure in the early 1980s, when several initiatives were attempted in an effort to relaunch the Single European Market and create an EU. The debate focussed on the need for institutional reforms, including changes in the unanimity principle and the balance between the Council and the European Parliament as it was seen as a precondition for advancing the political cooperation (Pedersen, 2013; Petersen, 2006). Traditionally, such pressure would not be problematic for Denmark’s status quo adherence, not least because it soon became clear in the early 1980s that Denmark’s traditional ally, France, began shifting in a more integrationist direction. Moreover, because of the potential economic benefits from the internal market, the British also began to consider being willing to accept further institutional integration to gain economics benefits (Petersen, 2006: 372). However, during the 1985 Intergovernmental conference (IGC), the Danish government was instructed by a majority in the Danish Folketing to continue to opt for maintaining the institutional status quo (Petersen, 2006). The government was instructed to create an alliance with the United Kingdom and Greece to ensure that Denmark was not left in a vulnerable and/or isolated position on the IGC and could continue to negotiate in the shadow of a potential British veto-. However, the outcome of the IGC was that the British and Greek negotiators chose to accept the Single European Act, leaving Denmark as the only country that failed to ratify the treaty after the IGC, which would potentially inflict massive abandonment and marginalization costs on Denmark. This led to a stalemate in the Danish parliament, which was resolved when the government launched a referendum on whether Denmark should accept the treaty, resulting in a Danish yes (Pedersen, 2013). Despite the turbulent negotiation process, the British positions still served the purpose of reducing the potential entrapment costs related to the introduction to majority voting and the increased role of the European Parliament as the treaty clearly reflected elements that were much closer to British (and Danish) positions compared with the initial proposals raised by Germany and France at the IGC (see Pedersen, 2013, 2014; Petersen, 2006) for a detailed analysis of the 1985 IGC). The situation can be illustrated in the Figure 6, in which Denmark under the IGC had positions closer to the status quo than the British had, but also that the outcome * reflected a position that would entail much fewer entrapment costs if the French or German ideal point had been realized in the negotiations.

OPEN IN VIEWER

Figure 6.

Shelter diplomacy in a situation with diverging preferences between shelter and small state.

Phase 2: partnership

In the period from 1993 to 2016, the official Danish positions gradually became more positive and the question of membership of the EU was less debated in the Danish public as there seemed to be an overall acceptance of the membership, and it became an even more central element in the Danish foreign policy strategy (Udenrigskommissionen, 1990). This was both a consequence of the changes in the political and security environments in Europe and of the new dynamism within the EU institutions (Wivel, 2018). However, the more pro-integrationist outlook became severely handicapped after the Danish rejection of the Maastricht Treaty in a referendum in 1992. The rejection once again meant that Denmark was put in a vulnerable position, where the country risked being isolated in the cooperation or even forced to leave it. However, in the aftermath of the Maastricht negotiations, Denmark was able to negotiate four opt outs in central areas such as defence, the economic and monetary union, home and justice affairs, and European citizenship in the Edinburgh Agreement in 1993, which was later ratified in a new referendum (Danish Institute for International Studies (DIIS), 2008, 2019).

As the European integration process began to move forward and entail both deeper and wider integration, particularly in the economic and the home and justice areas, Denmark’s relation to the EU in these areas was forced to take its point of departure in the Edinburgh Agreement. The Danish opt outs meant that Denmark was excluded from the decision-making process in these areas and, thereby, had no opportunity to influence the direction of the cooperation in these important areas (Adler-Nissen, 2009, 2014).

Because of the opt outs, Denmark pragmatically had to accept and accommodate to the deepening of the cooperation in these areas, as it was considered an unavoidable condition for the countries’ continued EU membership (Sørensen and Wivel, 2017). A key priority for Denmark in the negotiations of the Treaties of Amsterdam (1997), Nice (2001) and Lisbon (2007) was therefore to ensure the continued acceptance of the Danish opt outs in the cooperation (Wivel, 2018). Obviously, another reason was that the Danish policy makers viewed the treaties in the context of EU enlargement as necessitating more effective decision-making procedures and changed voting weights (Petersen, 2006: 524–532).

The widening of the Danish opt outs due to the intensified cooperation in the inflicted areas also meant that Denmark feared isolation and marginalization in central areas. The honeymoon of the engagement period in the Danish and British relations was over after the IGC in 1986. However, the position of the United Kingdom continued to play an important role for Denmark in the sense that both Denmark and Britain had opt outs in central areas in the cooperation. This shared position meant that Denmark was not completely isolated as an outsider in the period (Adler-Nissen, 2016). The British position as a fellow outsider helped Denmark to reduce some of the abandonment cost otherwise related to the marginalized Danish position. This primarily took place in the everyday discussions and negotiations in Brussels, where the structural position of the United Kingdom meant that the countries standing outside the Economic and Monetary Union, for instance, continued to have a certain ‘weight’, forcing the Euro insiders to show a certain amount of consideration to the states in the outsider group(s). In this sense, the British positions helped reduce the risks associated with the dilemma because the United Kingdom was able to absorb some of the isolation cost by insisting that the insider group took the outsiders’ positions into account. In this phase, the United Kingdom clearly played a central role in preventing the Eurozone from developing into a closed and exclusive club as the country helped keeping the Eurozone relatively open (Adler-Nissen, 2016; Sørensen and Wivel, 2017: 118). There are also examples that the British were able to slow the pace on the deepening of the integration on some of the inflicted policy areas by reducing the ambitions in the French and German proposals in the period. Theoretically, this can be illustrated in Figure 7, where *1 represents the likely outcome of a proposal from the French–German axis without the British positions and *2 represents a hypothesized outcome where the insider states have taken the positions of Britain into account, despite the United Kingdom not having formal veto opportunities. The hypothesized difference between *1 and *2 represents the reductions in the abandonment costs, which meant that the British positions were able to be a viable asset in the Danish bargaining strategy as it helped to reduce the cost associated with the dilemma.

OPEN IN VIEWER

Figure 7.

Shelter strategy in the partnership phase.

However, in the final stage of this phase, new dynamics in the Danish and British relationship came into force. This related to a gradual shift in the British positions, where the country began to adopt even more sceptical stances in its relation to Europe. In this period, the special relation clearly became ‘less special’ from the Danish perspective since the United Kingdom distanced itself from the official Danish position, which meant that the two countries gradually drifted apart in the cooperation. This meant that the British shelter became less relevant even though the British positions in the areas of the Danish opt outs remained important for the country to reduce the costs of isolation and marginalization.

Phase 3: separation

Brexit significantly impacted the small intergovernmentalist-oriented EU member states (Wivel and Thorhallsson, 2018). The fear was that the cooperation would intensify further when a central veto player such as the United Kingdom no longer acted as a formal or informal break in the cooperation. From a Danish perspective, these expected dynamics could be problematic because of its opt outs, where further developments would marginalize it even further, causing significant abandonment costs (Christoffersen, 2017). The theoretical situation is illustrated in Figure 8. The figure shows a situation in which there are no credible veto players close to the Danish positions in the negotiations. The implications could be that a likely outcome of a future bargain would be closer to more integrationist positions – such as the French and German ideal points. The costs of the dilemma would therefore be felt as much higher compared with earlier scenarios. This situation might apply to the overall direction of the cooperation, but for Denmark, the effects would probably be felt most intensively in the areas covered by the Danish opt outs. When areas change from the intergovernmental law base to the new supranational base in the Lisbon Treaty, the difference between Denmark and the rest of the European countries will widen and the country will risk paying the full price of marginalization.

OPEN IN VIEWER

Figure 8.

Increase in the entrapment cost in the divorce phase without shelter options.

The British exit will therefore potentially increase abandonment costs on the area of justice and home affairs, which includes several important policy areas. Ireland has an opt-in solution on the cooperation but also a clause and a protocol in the Lisbon Treaty where they pledge to cooperation as much as possible. The same was the case for the intensification of the cooperation in the security and defence area and the economic and monetary area, where Denmark felt it was faced with higher isolation and marginalization costs until the country decided to put the opt out for a referendum on the defence area. In the monetary area, it has been suggested that the Euro insiders have taken the outsider group’s position into account because of the structural British importance. After Brexit, the encouragement to take the outsider group’s interest into account will be reduced, which might affect the cost/benefit calculations for small states such as Denmark.

Phase 4: divorce

In the post-Brexit phase, Denmark and other small states stand to bear the full costs of their marginalized positions, with no others to pay the cost of rejecting further integration or share or reduce the isolation costs of standing outside the cooperation. This situation has encouraged small states to develop new strategies or forced them to adapt their modus operandi in the cooperation. In its original form, shelter theory is based on certain assumptions about the nature of shelters, as they are generally assumed to be asymmetric in the sense that they describe the relationship between a small state and a great(er) power. Empirically, the analysis of the three first phases of the Danish membership has illustrated the utility and rationale behind asymmetric shelter diplomacy as a means to reduce and balance entrapment and abandonment costs. Broadening the empirical scope would have shown that many of the northernmost EU members have been closely aligned, also on central issues, with the United Kingdom, have tried to utilize the British scepticism and relied on their occasional vetoes or threats to reduce potential abandonment and entrapment costs. In particular, Sweden has used shelter diplomacy to protect its interests as Euro outsiders now face somewhat similar challenges in the post-Brexit period. This has led to developments in the strategy, where there are numerous indications that small states have established new forms of more symmetric small-state shelters (Figure 9). The most prominent example of such a shelter was operational during the budget negotiations in 2020, where a new alliance between Denmark, the Netherlands, Sweden and Austria began to make its voice heard in the form of the Frugal Four. The budget negotiations demonstrated that the group was able to stay united in rejecting a larger EU budget, thereby avoiding the risk of marginalization costs, and it was also successful in exercising major influence on the outcome of the negotiations. The Frugal Four, in cooperation with Finland, negotiated a lower general EU budget and larger discounts for themselves and shifted the distribution between loans and grants in the recovery fund.

OPEN IN VIEWER

Figure 9.

Effects of symmetric shelter.

Consolidating this ad hoc coalition into a more permanent small-state shelter in the post-Brexit period would make life easier for small states, but consolidations can be difficult, as the glue that binds the shelter together was the ambition to balance entrapment and abandonment costs, rather than a more substantial coherent core of interests. Another more institutional explanation for the lack of consolidation of the block can be found in the fact that small-state shelters do not possess veto rights (Schulz, 2021; Schulz and Henökl, 2020), which might encourage more individual pursuit of smart state strategies, bilateral bandwagon or even hiding (Grøn and Wivel, 2011). However, their political function still offers the opportunity to reduce abandonment and entrapment costs. The empirical example holds another interesting theoretical aspect, namely that it highlights that the core in shelter diplomacy is not reactive or simply free riding, as shelter diplomacy – in both its asymmetric and its symmetric variant – seems to entail a proactive element, that is, small states engage more actively in attempts to forge coalitions by relying more on a pooling of powers among equals (Schulz, 2021). The post-Brexit situation therefore seems to have transformed Denmark’s as well as many smaller states’ shelter strategies, forced them to become more active and engage in alliances or small-state shelter building in the EU.

Perspectives and conclusions

Despite its promises, the article has argued that shelter theory has not fully reached its potential, as it has not been able to answer exactly what the effects of shelter diplomacy are, what type of costs shelter diplomacy can protect small states against and to what degree shelters are useful for small states if the distance in preference between the small state and the larger power widens. To gain theoretical advances, the argument is that shelter theory – or, more accurately, shelter diplomacy – can forcefully be integrated into the mainstream IR literature. Utilizing realist-inspired alliance theory and shelter theory, the argument here is that shelter theory has the potential to function as a middle position between the more well-known and -described strategies of hiding and bandwagon. The analytical benefit is that shelter diplomacy identifies the practices and mechanisms whereby small states can balance the entrapment and abandonment costs associated with the alliance dilemma. Empirically, the article has illustrated the potential of the concept of shelter diplomacy in the asymmetric shelter relation between the United Kingdom and Denmark in the European integration process, where it has been illustrated how a small state can utilize the structural powers of great powers to balance costs of abandonment and entrapment through shelter diplomacy. Furthermore, the article expands on the utility of the shelter concept and moves it beyond the predominant asymmetric interpretation by suggesting that small states can seek and form more symmetric shelters through formation of small-state coalitions, as, for example, the Fugal Five after the British exit from the cooperation. Developing the theory in this direction might help to levitate the theoretical ambition of shelter theory to become a more genuine alternative to theories of hiding and bandwagon in the small state literature by offering an explanation to how small states can compensate for their structural weaknesses and reduce the cost associated with international engagement. Broadening the shelter concept can furthermore help bring shelter logics and diplomacy into the game-theoretical and liberal oriented alliance and coalition literature, as a central driver of the establishment of small-state coalitions following shelter logics seems to be common aversions and cost reductions rather than prospects of + sum logics and absolute gains. The concept of shelter diplomacy is not without caveats, however. Despite its potential contribution to our knowledge and understanding of the conduct of small states’ foreign policies, diplomatic practices and strategies, the definition of asymmetric shelters risks conflating the effects of free riding and shelter diplomacy, as small states can also reap the benefits of others’ strength by buck passing or free riding. In line with the original shelter theory, the argument here is that shelter diplomacy is not just disguised celebrated free riding on the efforts of the great powers. Shelter diplomacy can rather be seen as the continuous reading of the power political realities of the bargaining games and the focussed ambition to utilize the power balance, the positions of others in combination with pragmatic bargaining strategies to reduce and balance the costs associated with the alliance dilemma. This can be done in asymmetric bilateral relations, where the small state depends on one shelter partner, or it can be done in symmetrical multilateral relations with a larger number of partners where the small state tries to coordinate and utilize the protection potential of strategic partnerships with likeminded small states to avoid certain cost types. Shelter diplomacy therefore contains a proactive element, as the benefit of the strategy requires constant awareness of the relative distance between the shelter great power and the small state, and often requires active participation in negotiations. Symmetric shelters might share some definitional overlap with insights from the coalition literature, where smaller states balancing behaviour in coalitions have certain similarities with this variant of shelter theory. As future studies might continue to pursue the ambition of developing shelter theory and explore its integration into mainstream IR alliance theory, some of these demarcation lines can be clearer as the shelter logic has numerous potentials to contribute to and supplement existing models of small-state diplomacy.