Differentiated disintegration in the Economic Community of West African States, the Eurasian Economic Community and the European Union: a comparative regionalism approach

ABSTRACT

The United Kingdom’s withdrawal from the European Union has sparked interest in voluntary withdrawals from regional organisations (RO). While Brexit is a well-researched subject, other exits from ROs around the globe have been somewhat neglected. We still know relatively little about states’ motivations to leave and how such exits play out in the short and long run. This article addresses both gaps. First, it conceptualises withdrawals from ROs as differentiated disintegration to better grasp the pre- and post-exit dynamics. Second, it puts forth three factor groups explaining states’ reasoning composed of (i) geopolitical and geoeconomic conditions, (ii) intra-regional tensions and (iii) domestic factors. Third, it applies this framework to Uzbekistan’s exit from the EurAsEC, Mauritania’s departure from ECOWAS and the UK’s withdrawal from the EU. Despite great differences in the regional setup and level of institutionalisation, we find that strong parallels can be drawn between all three cases.

KEYWORDS:

• Africa
• Central Asia
• comparative regionalism
• differentiation
• Europe
• flexible cooperation
• regional organisations

1. Introduction

Withdrawals from regional organisations, like Brexit, are not unusual within schemes of regional integration. In fact, there have been 49 such exits recorded between 1945 and 2024. The January 2024 pronouncements from the recently installed military governments in Burkina Faso, Mali and Niger, signalling their intention to exit ECOWAS, may soon add to this number. However, unlike Brexit, most of these exists have garnered little to no academic attention. This leaves a void in our general understanding of regional integration processes or the reversal thereof. In this article, we address two important questions raised by these phenomena. Why do states choose to leave regional organisations? What defines the post-withdrawal relationships between the exiting state, the remaining member states and the regional organisation itself? In order to do so, we draw on a comparative regionalism framework as well as theoretical insights from the literature on Brexit and, covering regional integration more generally, from the field of IR.

The burgeoning literature on Brexit demonstrates that the UK did not sever all ties with the EU and its member states. Consequently, scholars have framed Brexit as a form of differentiated disintegration (Gänzle 2019; Gänzle, Leruth, and Trondal 2020; Gänzle and Wunderlich 2022; Schimmelfennig 2018). Despite numerous withdrawals from other regional organisations, it remains unclear whether the findings from Brexit scholarship can be generalised to those other cases. Some pioneering studies situated more broadly in the field of IR have investigated state withdrawals from international governmental organisations (IGOs) (Closa and Casini 2016; Eilstrup-Sangiovanni 2020; von Borzyskowski and Vabulas 2019). Regional organisations (ROs) like the EU, however, deviate significantly from other IGOs in two crucial aspects. Firstly, RO membership has a geographic dimension. Secondly, ROs facilitate cooperation and integration across multiple policy domains, making them multifunctional entities that address various issues, including economic and security concerns. Thus, we focus on the comparison of regional, multi-purpose organisations. Notwithstanding the challenges of comparing regions and their respective ROs that differ in crucial aspects such as politics, culture, identity and geography, we believe that this endeavour promises to yield valuable insights concerning the general mechanisms underlying exits from regional organisations (for an overview, see Table 1 below).

Table 1. List of regional organisations with voluntary state exits (ordered by ‘regionness’).

No.	Acronym	Institution	Foundation – dissolution/replacement	Voluntary exits and dates
Regional society				Total: 45 exits
1	ACC	Arab Cooperation Council	1989–1990	Egypt (1990)
2	ALBA	Bolivarian Alliance for the Peoples of Our Americas	2004	Bolivia (2019) – re-joined 2020 Ecuador (2018) Honduras (2010)
3	CEEAC	Economic Community of Central African States	1983	Rwanda (2007) – re-joined 2016
4	CEFTA	Central European Free Trade Agreement	1992	Poland (2004) Hungary (2004) Czechia (2004) Slovakia (2004) Slovenia (2004) Romania (2007) Bulgaria (2007) Croatia (2013)
5	CELAC	Community of Latin American and Caribbean States	2011	Brazil (2020)
6	CIS	Commonwealth of Independent States	1991	Georgia (2009)
7	COE	Council of Europe	1949	Russia (2022)
8	COMESA	Common Market for Eastern and Southern Africa	1993	Lesotho (1997) Mozambique (1997) Tanzania (2000) Namibia (2004) Angola (2007)
9	CSTO	Collective Security Treaty Organization	1992	Azerbaijan (1999) Georgia (1999) Uzbekistan (1999) – re-joined 2006 Uzbekistan (2012)
10	EFTA	European Free Trade Association	1960	Austria (1994) Denmark (1972) Finland (1994) Portugal (1985) Sweden (1994) UK (1972)
11	EurAsEC	EurAsian Economic Community	2000–2014	Uzbekistan (2008)
12	GUAM	Organization for Democracy and Economic Development	1997	Uzbekistan (2005)
13	IGAD	Intergovernmental Authority on Development	1986	Eritrea (2007) – re-joined 2011
14	OAU	Organisation of African Unity	1963–2002	Morocco (1984)
15	PIF	Pacific Island Forum	1971	Kiribati (2022) – re-joined 2023
16	SADC	Southern African Development Community	1980	Seychelles (2004) – re-joined 2008
17	SEATO	Southeast Asia Treaty Organization	1954–1977	Pakistan (1972)
18	UNASUR	Union of South American Nations	2008	Argentina (2018) Brazil (2018) Chile (2018) Colombia (2018) Paraguay (2018) Peru (2018) Uruguay (2020)
Regional community				Total: 3 exits
19	ANDEAN	Andean Community	1969	Chile (1976) Venezuela (2006)
20	ECOWAS	Economic Community of West African States	1975	Mauritania (2000)
Regional (institutionalised) polity				Total: 1 exit
21	EU	European Union	1951	UK (2020)

Source: own compilation.²

The article unfolds as follows: Firstly, we contextualise the reasons for voluntary exits from regional organisations by drawing from the literature on regional integration and international relations. We identify three factor groups situated on the global, regional and domestic level. Secondly, we illuminate the process and outcome of such exits by invoking the Brexit literature and utilising the concept of differentiated disintegration. This demonstrates that the UK’s withdrawal from the EU was not ‘complete’ in the sense that not all institutional ties were severed and new ones had to be created. Thirdly, we integrate this concept into a comparative regionalism framework and apply the regionness spectrum to map ROs with different levels of institutionalisation and, thereby, facilitate comparison (Hettne and Söderbaum 2000). Fourthly, we apply the factor groups defining exits and the concept of differentiated integration to three illustrative mini-case studies: Brexit, Mauritania’s exit from the Economic Community of West African States (ECOWAS) and Uzbekistan’s withdrawal from the Eurasian Economic Community (EurAsEC). On this basis, we finally conclude that exits from ROs are best described as cases of differentiated disintegration. While certain institutional connections might be severed, others persist and new ones might emerge to manage relationships with neighbouring countries, potentially enabling re-joining under specific circumstances.

2. Why do states leave regional organisations? Domestic factors, regional contestation, geopolitical and geoeconomic environment

To understand the post-withdrawal relations between the exiting state, remaining member states and the regional organisation, one must pay attention to the reasons for exit. Numerous attempts have been made to theorise these (Nolte and Weiffen 2021; Vollaard 2014, 2018). Broadening the perspective to international organisations, von Borzyskowski and Vabulas (2019) categorised the key influencing factors for such withdrawals, encompassing political, institutional and geopolitical dimensions. Building on this scholarship, our analysis of exits from regional organisations is inspired by the literature on regional integration and International Relations and concentrates on domestic considerations, regional disputes and shifts in the geopolitical and geoeconomic landscape. Although there can be overlap, these three factor groups operate on distinct analytical levels: the domestic, regional and global.

Various theoretical frameworks, including neoclassical realism (Rose 1998), neofunctionalism (Haas 1958; Schmitter 1971, liberal intergovernmentalism (Moravcsik 1998), constructivism (Risse 2019) and post-functionalism (Hooghe and Marks 2009), highlight the significance of domestic factors. Among these factors, identity concerns, nationalism, cultural values and domestic leadership priorities play pivotal roles in shaping regional dynamics and withdrawal decisions. Understanding the influence of these factors on withdrawal decisions is crucial for a better comprehension of exit motivations and post-exit relations as broader phenomena. Therefore, we will closely examine these factors in our case studies.

Disputes between RO members can also prompt exits, with divergent views among member states regarding the future trajectory of the organisation contributing to tensions. Dominant states may opt to withdraw if they feel their interests are jeopardised, as noted by Schimmelfennig (2019), while smaller states might leave due to imbalances in power dynamics. The presence of long-standing rivalries (Krapohl and Vasileva-Dienes 2020) and perceptions of an ineffective RO can also factor into withdrawal decisions. Additionally, dissatisfaction among members stemming from an uneven distribution of benefits and costs from integration can be a catalyst for departures (Axline 1977).

On the global level, geopolitical and geoeconomic factors may also play a role. Realist theories highlight that regional institutions are shaped by considerations of power balance and containment (Fry 2000; Gilpin 1981; Hoffmann 1982). Hegemonic powers establish and support regional organisations to extend their influence (Gilpin 1987; Ikenberry 2001; Kindleberger 1986; Webber 2019) and shifts in power distribution or threat perception can alter participation calculations (Lobell, Jesse, and Williams 2015; Mearsheimer 1990). Economic interdependencies also drive regional integration according to liberal institutionalist theories, and economic stagnation or opportunities elsewhere can reshape integration dynamics (Hoffmann 1966; Keohane 1984; Moravcsik 1998).

Altogether, voluntary exits from ROs can be attributed to factors operating on the domestic, regional and global levels. However, it is important to note that the distinction between these levels and factors is not always crystal clear and is mainly made for analytical convenience. For example, the ideological shifts in many EU member states towards a more conservative and restrictive handling of immigration are to some extent related to a perceived mismanagement on the regional level. Furthermore, the boundaries between the regional and global may become blurred, as seen in geopolitical rivalries between two economic or political blocs involving states within and outside a particular region.

3. Conceptualising member state withdrawals from regional organisations as cases of differentiated disintegration

Integration and regionalism represent collaborative responses to challenges that extend beyond the confines of the nation state. These responses, however, can undergo shifts in perception over time leading to a dynamic spectrum of integration levels required to effectively address these challenges. The concept of selective interconnectivity has long been a central theme in European integration research, where the EU stands as an example of differentiated integration (Leuffen, Rittberger, and Schimmelfennig 2013). Through time, the landscape of European integration has evolved into an intricate system characterised by concentric circles, encompassing core members, states with opt-outs from certain policies, temporarily excluded nations, aspiring members and an array of neighbourhood partnerships (Gstöhl and Lannon 2018; Lavenex 2011).

Differentiated integration, at its essence, pertains to a set of processes in which certain states within an RO pursue closer cooperation while others who prefer less integration opt not to participate in specific initiatives, either temporarily or permanently. Notably, differentiated integration defines not only the EU but has been found also in other ROs such as ASEAN, NAFTA and APEC (Warleigh-Lack 2015). It can, therefore, be seen as a fundamental aspect of regional institution-building; differentiated integration acknowledges and embraces the diversity of interests, preferences and capacities of states, allowing for flexible cooperation.

Propelled by this premise, we posit that differentiated disintegration should also be embraced within the overarching context of flexible cooperation or selective interconnectivity. Research on Brexit has defined differentiated disintegration as ‘a process of unequal reduction in the level, scope, or membership of the EU’ (Schimmelfennig 2018). In this sense, just as gaining membership of an RO does not automatically mean adopting all policies to the same extent, withdrawals do not necessarily entail a ‘complete’ separation. As was demonstrated by the UK’s long and difficult separation process from the EU, the exiting state continues to maintain varying degrees of institutional ties with the organisation itself and its member states (Gänzle, Leruth, and Trondal 2020; Martill and Sus 2022; Schimmelfennig 2022). This is related to the very nature of ROs which are intrinsically linked to specific regions and designed to address the distinct needs of states sharing that space. This spatial dimension underscores the uniqueness of ROs: States possess the ability to disengage from a specific RO, but they remain part of the wider regional space. This necessitates the establishment of some form of ongoing institutional connections. Differentiated disintegration enables states to withdraw from an RO while preserving a certain level of integration or cooperation with it and the wider region. It acknowledges that not all relationships and aspects of cooperation need to be entirely severed, and some level of connection and cooperation will inevitably persist due to mutual interests and shared goals.

The complexities of regionalism and regional cooperation add to the differentiated nature of exiting an RO. ROs often exist within a larger network of multiple and overlapping institutional arrangements within a region (Panke and Stapel 2018). This complexity extends beyond the European context; in Africa, for example, regional integration has created a ‘spaghetti bowl’ of overlapping regional institutions (Hartmann 2021). Burundi, Kenya, Rwanda and Uganda, for example, are members of the continental African Union (AU), the East African Community (EAC) and the Common Market for Eastern and Southern Africa (COMESA). Even if a state broke all ties with one of these organisations, the larger web of ROs ensures that a significant level of institutionalised relations with neighbouring states in the region persists. This adds to the differentiated nature of RO withdrawals if seen from a broader, regional perspective.

In conclusion, we argue that exits from ROs are consistently characterised by differentiation within the broader regional context. Firstly, a withdrawal from an RO rarely constitutes complete disengagement, as institutional links often persist, suggesting that it is better understood as a reconfiguration of regional integration. Secondly, a departing member state’s connection is influenced by various formal and informal ties that continue to connect it to neighbouring countries and the wider regional system. These factors influence the significance of a state’s exit from an RO and its anticipated consequences, thereby likely informing the reasoning of decision-makers.

4. Comparing instances of differentiated disintegration

The comparative endeavour of this article follows a line of research which in recent years brought EU studies, International Relations and new regionalism research closer together (Acharya 2014; Acharya and Johnston 2007; Börzel and Risse 2016, 2019; Hooghe, Lenz, and Marks 2019; Sbragia 2008; Söderbaum 2009; Warleigh-Lack and Rosamond 2010). The regionness concept has been a particular milestone as a non-Eurocentric comparative analytical tool for understanding the emergence and construction of regions (Hettne and Söderbaum 2000). Regionness depicts regional cooperation as a set of non-hierarchical, flexible and not necessarily sequential processes and dynamics that create a region out of a grouping of geographically proximate states and other actors.

Regional cooperation is a flexible process that involves dynamic and adaptable cycles of institutionalisation, de-institutionalisation and re-institutionalisation. It is therefore best characterised as an ‘untidy’ system of multiple and, at times, overlapping institutions. The boundaries between various institutional arrangements within a wider region are not rigidly defined but rather exist as flexible buffer zones, where one institutional system overlaps with another, offering the option for gradual transition. This harks back to the previously described system of concentric circles. Within the European context, for example, Norway, Iceland and Lichtenstein, are part of the EU’s internal market but not EU member states. Switzerland’s EU relations are governed by a set of bilateral treaties, other non-member states enjoy candidate status or fall under the framework of EU neighbourhood policy. After leaving the EU, the UK is now orbiting the EU somewhere within these outer circles but still rather closely attached. The EU-UK Trade and Cooperation Agreement was put in place to manage market access. Meanwhile, Northern Ireland has effectively remained part of the EU’s single market, which created frictions for trade with Great Britain until the issue was settled under the so-called Windsor Framework together with the EU. Gibraltar has drawn even closer to the EU by becoming part of the Schengen zone after Brexit. This exemplifies the differentiated nature of membership withdrawals in the sense that one institutional solution is often replaced with another.

The regionness framework offers a typology of regions and regional institutionalisation, allowing us to (a) map and compare instances of withdrawal from diverse institutional arrangements and (b) conceptualise these exits as instances of differentiated disintegration. The regionness framework (see Table 2) posits five distinct levels of regional cooperation, distinguished by the quality, density and complexity of institutionalisation.

Table 2. Regionness typology.

level of regionness	characteristics	empirical examples
regional space	geographically delineated region; few relations among neighbouring societies	/
regional system	loose regional cooperation in security affairs	Eastern and Western Europe in the Cold War era
regional society	institution building; regional cooperation in several policy areas	EFTA, Eurasian Union
regional community	developing supranational authority; emerging regional identification	Andean Community, ASEAN, ECOWAS, MERCOSUR
regional (institutionalised) polity	Deep political, economic and institutional integration; pooled sovereignty; actor capabilities; elements of supranational identity	EU

Adapted from: Hettne & Söderbaum (2000).

In a regional space, geographically proximate sovereign states hardly interact. However, this is largely hypothetical as neighbouring societies tend to overlap in various aspects and require relationship management. At a minimum, neighbouring states must acknowledge each other and establish a system for border management and regulation.

Once neighbouring states establish basic institutions to facilitate coexistence, a regional system emerges. It is characterised by rudimentary institutional structures, akin to Bull’s international system (Bull 1977). These minimal institutions primarily focus on coordination, such as a regional balance of power (Söderbaum, 2026, 167). Economic interactions within the regional system may be driven by self-interest and short-term goals rather than cooperation. However, increased interaction can prompt the desire for more formal and informal institutional structures to ensure predictable and stable relations (Haas 1958).

Institution-building through formal and informal transnational rules can transform regional cooperation into a regional society, extending beyond mere coordination efforts. This involves a broader range of actors, including non-state actors like businesses, NGOs and social movements, operating within a regulated transnational environment (Hettne and Söderbaum 2000; Söderbaum 2016, 166). These processes contribute to the development of a transnational society, where institution-building encompasses active cooperation and the pursuit of common interests. This stage aligns closely with Hedley Bull’s (1977) notion of international society.

Further institutionalisation can foster a new level of regional cooperation in the form of a regional community with shared norms and identities. The convergence of various regional processes nurtures collective regional identities, bridging national communities and promoting transnational solidarity. This resembles a security community (Deutsch et al. 1957). A spectrum of regional identity emerges, ranging from reduced military threats among community members to collective representation vis-à-vis external actors under specific circumstances. Deeper institutionalisation may involve supranational institutions like a common parliament, a transnational court of justice or banking institutions.

A regional institutionalised polity signifies the most dense and formal set of institutional relationships. ROs in this category have a distinct identity and some actor capabilities. Mutual solidarity and identification connect diverse societies, with tight-knit institutional processes spanning across politics, economics and society. This is not a replication of the Westphalian state at a regional level, but an integrated, yet heterogeneous, pluralistic and multicultural polity. The famous EU motto ‘united in diversity’ reflects the aspiration for a political order of this kind, where member states retain sovereignty but pool state powers it in certain areas (Söderbaum 2016, 171).

The original regionness framework as introduced by Hettne and Söderbaum is a broad and heuristic tool for comparative purposes, explaining how the intensification of a broad range of regionalisation processes leads to the solidification and consolidation of a region. Our focus is on the dynamic changes of the institutionalised regional cooperation. This framework allows us to locate ROs at their respective level of regionness and to map withdrawals as instances of differentiated disintegration in that the exiting state does not fall out of the regional system but, at most, moves to a different level of institutionalised (regional) relations. This implies that movement along the regionness scale is possible in both directions (Warleigh-Lack and Robinson 2011). In other words, it offers a framework that incorporates both integration and disintegration: ‘Regionness can both increase and decrease … Integration and disintegration go hand in hand (albeit at different levels) and at each stage there is the possibility of spill-back’ (Söderbaum 2016, 163). Thus, the regionness continuum can be adapted to provide a more nuanced understanding of disintegration: The leaving state seeks to extricate itself from some shared rules or norms, effectively replacing them with different institutional solutions. In summary, the regionness framework offers a flexible typology to compare withdrawals from ROs in different regionals as instances of differentiated disintegration. We demonstrate this below.

5. Differentiated disintegration in ROs of a regional society, community and polity: the exits of Uzbekistan, Mauritania and the United Kingdom

To support our argument, we utilise the structured comparison method. This approach involves a limited set of illustrative case studies using a consistent analytical framework to examine both similarities and differences (George and Bennett 2005). These case studies are largely exploratory in scope, designed to develop an initial understanding of the reasons behind withdrawals from ROs and how they relate to the concept of differentiated disintegration.

We delve into three distinctive cases: Uzbekistan’s withdrawal from EurAsEC, Mauritania’s departure from ECOWAS and the UK’s exit from the EU. Each of these instances stands as a pivotal milestone for their respective ROs, representing the first and only instance of a full member state voluntarily leaving. Collectively, they exemplify differentiated disintegration in highly dissimilar regional settings, i.e. Central Asia, Western Africa and Europe. The respective ROs exhibit vastly different levels of institutionalisation, which positions them at different tiers of regionness – ranging from regional society and regional community to regional institutionalised polity.¹

Our case selection, thus, follows the most-different cases approach which aptly serves to qualitatively gauge the comparability of RO exits both with regards to the reasons behind them and the degree of differentiation witnessed in the process. While we acknowledge that the nature of our case studies does not allow for proper testing of causal relations, they do illustrate the plausibility of our arguments. Data has been gathered through document analysis, secondary literature and was complemented by expert interviews when appropriate. Our case studies are structured as follows: We commence by furnishing essential contextual information that aids in comprehending the region and the catalysts driving institutional development. Subsequently, we delineate the extent of differentiated disintegration by retracing the respective state’s withdrawal from the concerned RO. Lastly, we systematically scrutinise the rationales underpinning each state’s exit with respect to the three factor groups, while also considering the level of institutionalisation within each RO.

5.1. Regional society: Uzbekistan’s withdrawal from the Eurasian Economic Community

Our first case focuses on Uzbekistan’s disengagement from the Eurasian Economic Community (EurAsEC) which was subsequently transformed into the Eurasian Economic Union (EAEU). We classify EurAsEC as an RO at the regional society level. Despite cooperative pursuits and defined mutual goals, institutional development was limited. Unlike ECOWAS or the EU, robust supranationalism was absent in EurAsEC, resulting in intergovernmental cooperation primarily revolving around trade and other economic matters.

Eurasian regionalism aimed at revitalising the Central Asian republics – Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan and Uzbekistan – following the Soviet Union’s dissolution (Blockmans, Kostanyan, and Vorobiov 2012; Roberts and Moshes 2016). In 1994, Kazakhstan’s then-president, Nursultan Nazarbayev, advocated for a ‘Eurasian Union’ (Tarr 2016). At the sub-regional level, the Central Asian Economic Cooperation Organization, established in 2002, replaced the Central Asian Economic Union from 1998, focusing on promoting regional collaboration and stability in Central Asia until it merged with the Eurasian Economic Community in 2005.

The EurAsEC, formed in October 2000 by Belarus, Kazakhstan, Kyrgyzstan, Russia and Tajikistan, aimed to facilitate multilateral economic cooperation, striving for a Single Economic Space, common market mechanisms and coordinated approaches to global integration. Key focus areas included transportation, energy, agriculture and labour migration (Vinokurov 2018). Russian President Vladimir Putin later sought to deepen economic ties within EurAsEC, but these efforts were increasingly seen as Russia’s pursuit of hegemony (Roberts and Moshes 2016). Realising such ambitious goals necessitated the transformation of EurAsEC into the EAEU in 2015, marking a significant shift from its initially limited institutionalisation.

Uzbekistan joined EurAsEC in 2006 but withdrew in 2008, reflecting its generally complex history with Eurasian integration bodies (Vinokurov 2018, 123). Previously, it participated briefly in the Organization for Democracy and Economic Development (GUAM) (1999–2005) and the Collective Security Treaty Organization (CSTO) (1992–1999 and 2006–2012). These exits were voluntary and differentiated. When Uzbekistan left EurAsEC, it remained a member of the Shanghai Cooperation Organization (SCO) and the Commonwealth of Independent States (CIS). Both are but loosely institutionalised, also resembling the level of regional society and comprising Central Asian states with some overlap in membership and purpose.

5.1.1. Motivations driving the withdrawal

Several reasons for Uzbekistan’s choice to withdraw from EurAsEC can be advanced. The exit was influenced by geopolitical factors such as its location amidst overlapping Russian, Chinese and European interests. Russia’s hegemonic aspirations and acting as the de facto leader of EurAsEC informed the withdrawal. Uzbekistan’s president at the time, Islam Karimov, was vehemently opposed to any integration of the former Soviet republics (Madiyev 2021, 89; Tolipov 2019). The Uzbek government’s withdrawal from the EurAsEC is an example of the ‘Uzbek Path’, highlighting Uzbekistan’s tradition of independent policy-making and geopolitical orientation (Gleason 2008a). This approach can be described as pragmatic equidistance in managing great power relations with Russia and China and maintaining strategic autonomy (interview with an EEAS official, 13 April 2022).

Geoeconomic concerns, in contrast, seem not to have played a role. Economic interdependence among Central Asian states has been low and EurAsEC did little to foster intra-regional trade (Krapohl and Vasileva-Dienes 2020). However, as a double-landlocked country, Uzbekistan is dependent on favourable access to trade routes across its neighbours (interview with an EEAS official, 13 April 2022). Although leaving EurAsEC did away with institutionalised trade agreements in the region, it did not facilitate trade relations with other important partners such as China either.

These geopolitical convictions intersect with intra-regional differences regarding institutional depth and design of EurAsEC. At the time when Uzbekistan joined, the Community sought to establish a customs union (Gleason 2008b). Although participation therein was never imposed upon Uzbekistan, this and the overall objective of deeper integration did not resonate well with Tashkent’s objectives of independent policymaking. Moreover, unresolved border disputes with Kazakhstan and Kyrgyzstan further complicated Uzbekistan’s position in Central Asia and, by extension, EurAsEC. During the brief period of membership, Uzbekistan even slowed down initiatives of the Community. For example, Tashkent opposed Russian assistance to Kyrgyzstan and Tajikistan in developing or rehabilitating their hydro power generation with potentially negative impact on downstream riparian countries like Uzbekistan (written correspondence with Eurasian Development Bank official, St. Petersburg, 9 March 2021).

Finally, domestic factors related to national identity appear to have significantly informed Uzbekistan’s decision to leave EurAsEC as well. President Karimov had always advocated for a distinct Central Asian orientation, encapsulated in the idea of Turanism which emphasises Turkic and Muslim values (Fazendeiro 2020; Laruelle 2012). This, arguably, could not be realised in an organisation dominated by Russia. In light of Uzbekistan’s autocratic system of government, the executive power of President Karimov with relatively few checks and balances and his determination to follow the ‘Uzbek Path’ is another explanatory factor for the exit from EurAsEC.

5.1.2. Evidence for differentiated disintegration

Uzbekistan’s exit from EurAsEC mirrors the image of an overall protracted relationship with its regional partners and organisations and strongly underpins the concept of differentiated disintegration. The Karimov regime did not even consider pursuing its path of regional integration further despite the fact that it had been offered an opt-out on the projected customs union. Still, under the current President Mirziyoyev who replaced Karimov after his death in 2016, Uzbekistan engaged in a series of economic reforms and made some cautious steps towards the EAEU, signalling its desire to attain observer status and has even taken steps towards full membership (see Hashimova 2021). In December 2020, Uzbekistan was eventually granted EAEU observer status; in April 2021, the EAEU commission and the Uzbek government signed a Joint Action Plan for 2021–23. However, the future direction remains uncertain. Wedged between Chinese and Russian hegemonic ambitions, Uzbekistan remains cautious. Indeed, following the full-scale Russian war in Ukraine, Central Asian leaders, including the Uzbek government, have grown increasingly wary of Moscow’s foreign policy.

5.2. Regional community: Mauritania’s withdrawal from the Economic community of West African states

The Economic Community of West African States (ECOWAS) is a West-African RO established by the Treaty of Lagos in May 1975. Mauritania was one of its founding members. ECOWAS has embarked on a path towards some level of political integration, its primary objective is to create ‘an economic union in West Africa’ (ECOWAS Treaty, 1993, Article 3(1)). Today, it is the second largest RO in Africa, covers one of the most populated regions and has established a relatively high level of institutionalisation (Müller 2023). ECOWAS has gained considerable clout as a security actor in the region, having undertaken several regional peacekeeping interventions in the 1990s and 2000s (Coleman 2007).

We conceive of ECOWAS as a regional community, which is primarily characterised by displaying some forms of regional identity as well as a gradually developing supranational authority. The supranational core of ECOWAS has developed as a consequence of several structural reforms during the 1990s and 2000s, such as the revision of the ECOWAS Treaty in 1993 (Lavergne 1997; Mair and Peters-Berries 2001) and the more recent 2006 reforms of the secretariat which was then turned into an executive commission (Lokulo-Sodipe and Osuntogun 2013).

Mauritania withdrew from the organisation in 2000 under the authoritarian leadership of President Maaouya Ould Sid’Ahmed Taya who was in power from 1984 until he was ousted in the military coup of 2005. Although the withdrawal came rather unexpectedly to most observers and some still argue ‘the exit of Mauritania was without reasons’ (Azu 2020, 3), several rationales can be advanced.

5.2.1. Motivations driving the withdrawal

The geopolitical and geoeconomic framework conditions underwriting Mauritania’s exit paint a mixed picture. According to some analysts, Mauritania’s Taya regime preferred to concentrate on the Arab Maghreb Union AMU, of which it is a co-founder, at the expense of West African integration (Jeune Afrique 2017; Melly 2019, 6). Economically, this step made little sense, as intra-regional trade among the conflict-ridden AMU had been minimal (Martinez 2006). Moreover, the tilt towards the Maghreb was also unlikely to yield an increase in geopolitical influence. Within the AMU, Mauritania had little political or economic clout, being the smallest member state and viewed as ‘peripheral by much of the Arab world’ (Melly 2019, 9). Still, it was felt that Mauritania’s pan-Arabic foreign policy objectives which reflected i.a. in support for Iraq during the first gulf war (Seddon, 1996) could be realised more adequately within the AMU.

Intra-regional disagreements within the Western African bloc were cited by the Mauritanian government as a major motivating factor to leave. In a press statement, Prime Minister Khouna explained that the decision to leave ECOWAS was a response to the ‘latest decisions of the community’ (Independent Online South Africa 1999). Several assumptions can be advanced. The Mauritanian government objected to deepening ECOWAS integration, in particular ECOWAS’ calls ‘for a West African federation’ (Independent Online South Africa 1999). Mauritania also opposed ECOWAS’ plans to establish a currency union’ (BBC 1999), as the country seemed to be ‘deeply attached to its own currency, the ouguiya (Independent Online South Africa 1999). Finally, ECOWAS’ emerging regime of democratic accountability did not suit Mauritania’s repressive domestic policies towards its non-Arab population, which the government likely saw better accommodated within the AMU (N’Diaye 2006)).

The withdrawal from ECOWAS and shift of attention to the AMU was also motivated by domestic level factors. Cultural and identity-related considerations arguably played a key role in Mauritania’s decision to withdraw from ECOWAS. Like Uzbekistan, Mauritania’s departure from ECOWAS was facilitated by its location at the confluence of two regions: West Africa and the Maghreb. Melly (2019) states that Mauritania’s location between these two regions permeates every aspect of the country’s history, social composition and economic and political outlook. ECOWAS membership, he argues, was a result of historical links with West Africa, established when the country was subject to French suzerainty. At the same time, Mauritania never shed its Arab identity and institutionalised links with other Arab nations, for example via its membership of the Arab League or ties with the Gulf states. Indeed, Arab nationalism gained in importance during the authoritarian rule of President Taya (Melly 2019). Mauritania was the only Arabic-speaking country within ECOWAS and remains dominated by its Arab population. The country’s Arab identity informed Mauritania’s co-founding of the Arab Maghreb Union (AMU) in 1989, although the enduring rift between Morocco and Algeria hindered the anticipated evolution of the AMU into an integrated regional economic block.

5.2.2. Evidence for differentiated disintegration

The exit marks a clear case of differentiated disintegration. First, while no longer part of ECOWAS, Mauritania still continued to uphold bilateral agreements with several ECOWAS member states, such as Senegal, Mali, Ivory Coast, Guinea and Niger, thus effectively maintaining institutional relations with most its West African neighbours. Second, Mauritania soon embarked again on a pathway towards closer relations with ECOWAS itself. Under the regime of President Abdelaziz, assuming power in 2009, Mauritania tilted more towards its West African partners, including ECOWAS. Abdelaziz was instrumental in setting up the G5 Sahel grouping encompassing Burkina Faso, Chad, Mali, Niger and Mauritania in 2014 and then, in 2017, the government started to negotiate the terms of readmission to ECOWAS. Although these negotiations have not been concluded to date, the government has signalled its political will to implement several measures effectively leading to economic reintegration such as the implementation of a free trade agreement with ECOWAS (Journal du Cameroun 2020). Full economic integration would ensure the harmonisation of Mauritania’s tariff regime with that of ECOWAS, reducing costs and delays in formal cross-border trade and eventually boosting the country’s role as a gateway port for land-locked Sahel countries on the one hand and important transit area between Morocco and ECOWAS on the other. Yet, Abdelaziz still held back from a return to full membership in order to maintain national control over domestic developments since ECOWAS had embarked on an agenda for consolidating multi-party democracy in West Africa (see Melly, 2019, 9). Abdelaziz was also able to keep Mauritania out of the provisions for free movement of citizens allowing the country to regulate migration by way of bilateral agreements with individual countries. Mauritania, for example, has an accord with Senegal for the free movement of citizens between the two countries, but it does not extend this agreement to other West African countries (Melly, 2019, 9).

Mauritania’s post-exit relations with ECOWAS provide clear evidence for differentiated disintegration. The country’s pursuit of rapprochement has been characterised by a keen interest in fostering closer economic ties. Simultaneously, it has maintained a steadfast stance, keeping itself at a distance from several of the community’s political norms and values. Furthermore, Mauritania pursued relations with core partners from ECOWAS at a strictly bilateral level and engaged in the establishment of alternative intergovernmental organisations such as the G5 Sahel grouping.

5.3. Regional institutionalised polity: the United Kingdom’s withdrawal from the EU

We perceive the EU as an emerging regional polity characterised by deep political, economic and institutional integration as well as an evolving supranational identity. The EU represents a union of sovereign states that have created a unique political entity, where state sovereignty in certain policy areas has been pooled, resulting in a distinct level of regional actor capabilities (Wunderlich 2012).

The UK’s departure from the EU – based on Art. 50 TEU – stands as a unique occurrence, with no other full member state having formally left the union to date. Britain’s relationship with European integration has been historically complicated. Notably, in 1951 and 1958, the UK chose not to join the European Coal and Steel Community and the European Economic Community (EEC) respectively. Instead, in 1960, the British government was instrumental in creating the European Free Trade Agreement (EFTA) as an alternative trade bloc without supranational inclinations (Jenkins 1961). However, only one year later, the far higher growth rates of the EEC members prompted London to make its first bid for membership and after being vetoed twice by French President De Gaulle, Britain finally joined in 1973. But only two years later, a national referendum was held to determine whether the UK should remain in the EEC, which was confirmed by over two thirds of participating voters.

During its membership in both the EEC and EU, the UK consistently presented itself as a challenging partner, often finding itself in disagreement with other member states and EU institutions regarding various policies and issues. This contentious dynamic resulted in the UK maintaining a differentiated EU membership status, characterised by budget rebates and policy opt-outs (S. George 1990; Murray, Warleigh-Lack, and He 2014). In 2016, under the leadership of the conservative Cameron government, a momentous referendum on EU membership was held, with the outcome in favour of Brexit by a small but decisive margin. Consequently, on 31 January 2020, the UK formally left EU membership.

5.3.1. Motivations driving the withdrawal

The vast literature on Brexit has uncovered a large swathe of explanatory factors. The global financial crisis of 2007/2008 significantly altered the geopolitical and geoeconomic landscape, challenging the normative dominance of globalisation and liberal internationalism. Concerns about cultural homogenisation, identity issues and the marginalisation of certain groups, created a divide between those perceived as ‘winners’ and ‘losers’ of globalisation, fuelling particularism and identity politics at the domestic level (Hobolt 2016; Zürn and de Wilde 2016). The Brexit campaign capitalised on these sentiments by questioning the accountability of EU institutions, criticising free movement, portraying EU membership as costly and suggesting to redirect EU budget contributions to the National Health Service.

Deep-rooted disagreements over the direction of European integration and the handling of recent crises intensified regional tensions between the UK and its EU neighbours. Perceived failures in managing the eurozone and refugee crises undermined solidarity norms, leading to portrayals of the EU as a dysfunctional bloc (Hobolt and de Vries 2016). Suspicions regarding the EU’s path toward potential federalism were further exacerbated. Indeed, the prospect of a Brexit referendum was used as a bargaining chip by the Cameron government in its attempt to renegotiate the UK’s EU membership terms, specifically seeking exemptions from the concept of ‘ever closer Union’ and changes to the free movement for workers.

Thus, while external factors contributed to the deteriorating image of the EU and European integration within the UK, the Brexit vote was primarily driven by domestic factors such as widespread Euroscepticism, a lack of shared identity and the perception of a distinct British political culture (Glencross 2021). Britain, with its imperial legacy, sees itself as a nation on the fringes of Europe, positioned between Western Europe and global/transatlantic interests (Wodak 2016). Eurosceptic sentiments were not limited to political parties but were widespread among the population, as consistently indicated by the EU’s polling instrument Eurobarometer. Additionally, a weak commitment to a common European identity was observed among the British population (Carl, Dennison, and Evans 2019). Ultimately, Prime Minister Cameron’s decision to hold the Brexit referendum was motivated by an attempt to strengthen his domestic political position and to put an end to the endless debate over EU membership in his party.

5.3.2. Evidence for differentiated disintegration

Scholars have argued that the UK’s withdrawal from the EU and the subsequent process of defining post-Brexit relations should be seen as a case of differentiated disintegration (Gänzle, Leruth, and Trondal 2020; Leruth, Gänzle, and Trondal 2019; Schimmelfennig 2018). Despite having left the EU, the UK remains very much linked to the EU and its members via a set of overlapping regional relations, mini- and bilateralisms, institutional sectoral ties and interdependencies (Phinnemore 2022). This is exemplified by the UK’s recent partial rejoining of the EU’s Horizon research programme. Moreover, the UK’s geographical proximity to the EU, shared interests, economic ties, common values, cultural and political connections create an inescapable gravitational pull (McGowan 2023). Clearly, Brexit severed several important institutional links, but others remain and new ones have already been created.

The Withdrawal Agreement serves as the primary framework for managing relations with the EU, while the EU-UK Trade and Cooperation Agreement provides institutional foundations for ongoing economic ties, including trade, security and cooperation. The UK’s membership in various agencies and the EU’s regulatory influence over certain territories continue (Phinnemore 2022). Close cooperation persists in areas such as security, defence, intelligence sharing and multilateral institutions like NATO. Cross-border cooperation continues, encompassing shared concerns like border management, law enforcement and environmental protection. Additionally, the UK remains a member of the Council of Europe and the OSCE, reflecting its widespread integration in the web of European integration projects. The recently founded European Political Community (EPC) further bridges the gap between EU and non-EU countries with a vested interest in finding cooperative solutions to joint issues.

6. Conclusion

The three cases studies provide several important insights concerning the nature of voluntary exits from regional organisations. First, they indicate the possibility of conceptual travel by highlighting the application of differentiated disintegration beyond EU studies and Brexit. Second, whereas the study of the motivation for withdrawals from international organisations and cooperation agreements has received a great deal of attention in recent years, our findings prompt us to conceive of regional organisations as a somewhat different category. In contrast to von Borzyskowski and Vabulas (2019) who have emphasised the role of geopolitical factors in this context, our three case studies provide evidence for the significant role of domestic level factors, suggesting that matters related to identity and culture have strong bearing on the choice to leave an RO regardless of its level of regional institutionalisation. Uzbekistan saw both its Turanist policy and Turkic culture poorly represented in EurAsEC; Mauritania’s Arab elite left ECOWAS in order to deepen integration in the AMU; and the UK, eventually, harnessed its widespread Eurosceptic sentiments and exclusive identity against EU membership. However, exiting an RO does not mean that citizens and decision-makers suddenly rid themselves of any sense of shared regional identity built up in prior decades through close economic, political and cultural cooperation. This sense of community may in the long run be the catalyst facilitating the kind of reintegration witnessed to varying extent in all three cases.

At the domestic level, we also found that the role of executive leaders was considerable regardless of the withdrawing state’s political system and regional setting. The fates of Uzbekistan and Mauritania have clearly been at the whim of their authoritarian leaders, but even Brexit can ultimately be attributed to Cameron’s decision to hold a referendum – and his successors, in particular Johnson’s stout commitment to ‘get Brexit done’. Geopolitical factors, in contrast, seem less pronounced overall and more significant at the lower levels of regionness, i.e. regional societies and communities.

A third point worth noting is that Uzbekistan, Mauritania and the UK are all located at the confluence of overlapping regional forces. Uzbekistan is situated where Asia and Europe intersect and Mauritania is culturally much closer to the Arab world compared to its sub-Saharan West African neighbours. Brexit, too, can be interpreted as the UK fully embracing its long-running narrative as an island nation with a special global and transatlantic orientation. One of the stated intentions was to leave the confines of the EU in favour of reinvigorated trade and security relations with its former colonies on the other side of the Atlantic and in the Indo-Pacific. This may suggest that states located in between regions perceive exits from one RO as an opportunity to shift their attention to integration in the other adjacent region which may appear more economically advantageous or politically likeminded. In comparison with states centrally located within a certain region, the (perceived) availability of alternative regional integration projects may also make withdrawals seem less daunting.

Finally, the case studies demonstrate that irrespective of the level of institutionalisation, voluntary withdrawals from ROs resemble differentiated disintegration rather than a complete departure from regional cooperation. Although the lengthy and messy decision-making and, ultimately, exiting process of Brexit has certainly demonstrated that leaving a densely institutionalised organisation like the EU is highly complicated, this did not deter the Cameron government to initiate it by calling for a referendum. In contrast, leaving organisations like EurAsEC that are more loosely organised appears like a lower hurdle. The main difference between leaving ROs on different levels of regionness seems to lie in the degree of differentiation. The UK’s tight integration in the EU’s common market and political institutions as well as the close relations with its member states made it an economic and to some extent political necessity to leave in place certain institutionalised ties. In contrast, the more loose integration of Mauritania and Uzbekistan in the respective ROs they withdrew from and the region they are situated in allowed for a more clear-cut divorce, albeit also not cutting all ties.

To conclude the discussion of the case studies, it is important to acknowledge the limitations of generalising based on a restricted number of cases. Additionally, it is worth noting that the case studies in this analysis were designed primarily with an illustrative purpose in mind, resulting in a potential lack of the empirical depth typically associated with this method. Further research involving a larger sample size and more comprehensive empirical analysis is necessary for a more robust understanding of differentiated disintegration as a general phenomenon and its causes.

In sum, this article makes three general contributions. First, it presents a compelling argument for the application of differentiated disintegration as a conceptual tool to analyse voluntary state withdrawals from ROs utilising the regionness typology. This approach offers two distinct advantages as it brings attention to the intricate and interconnected nature of regional contexts, which are characterised by multiple layers of relationships and overlapping connections. Furthermore, it highlights the dynamic nature of regional institutions, including ROs. By conceptualising withdrawals as differentiated disintegration rather than mere ‘exits’, these cases are placed within their appropriate regional contexts, treating them as ongoing processes rather than isolated and finite events. This perspective allows for a more comprehensive understanding of the complexities involved in state withdrawals from ROs and provides valuable insights into the evolving nature of regional dynamics and institutions.

Secondly, the article showcases the potential of extending the concept of differentiated disintegration beyond the EU and Brexit, opening up new avenues for exploration in Comparative Regionalism. By applying differentiated disintegration to analyse withdrawals from various regional organisations, we can move beyond singular case studies and begin to unravel broader patterns and dynamics. This shift in perspective offers a more nuanced understanding of differentiated disintegration as a versatile analytical tool with applicability across diverse regional contexts, fostering debate and stimulating innovative research directions.

Thirdly, our contribution adds to the Comparative Regionalism literature that challenges the notion of the EU’s uniqueness, fostering comparative analyses with other regional cooperation frameworks. By treating the EU as one instance of a broader phenomenon, we can uncover commonalities and divergences across regional processes. This comparative approach not only enhances our understanding of European integration but also facilitates a deeper exploration of regional dynamics globally, encouraging scholars to delve into the specificities of various regional contexts and the implications for regional cooperation – and disintegration.

Acknowledgements:

We would like to thank the participants at the Norwegian Political Science Conference at the Universoty of Agder in January 2023, at the EUSA conference in Pittsburgh in May 2023 and at the UACES annual conference in Belfast in September 2023. We are particulary grateful to the JEI anonymous reviewers for their extremely thorough and constructive comments.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. We omit the levels of regional space and regional system because these kinds of regions do not involve a sufficient degree of integration.

2. This list is based on the publicly available ROCO dataset which comprises all regional organisations created between 1945 and 2015 (Panke and Starkmann 2015). It includes ROs that have been dissolved as well as such that have been subsumed in reform processes. We updated the list to 2024 and only consider voluntary exits of sovereign states. We omitted institutions in which membership was not entirely voluntary such as the Warsaw Treaty Organisation.