Strongholds of Liberalism? The Reaction of Regional Integration Institutions to the Pandemic Trade Crisis

ABSTRACT

Regional integration institutions play an important but ambiguous role in the liberal international system, especially when it comes to trade policy. The Covid-19 pandemic generated a trade crisis, which gave them an opportunity to expand their role in a liberal or illiberal direction. An analysis of the framing and policy measures of five key regional institutions in Europe, South America, Africa and Asia will demonstrate whether they were geared towards liberalisation and/or interventionism in early 2020. The European Union stood out for its strong collective action on trade, working out a new balance between interventionism and liberalism. Other institutions hinted at interventionism but did not take collective action. The trade response was overwhelmingly state-driven, and regional institutions outside Europe clearly have limited agency in the global trade system. While their behaviour is difficult to qualify ideologically, their framing supported liberal principles of international relations.

KEYWORDS:

• regional integration
• trade
• liberalism
• political economy
• pandemic

Regional integration institutions are often viewed as liberal forces in the international system. However, in an era of illiberalism, there is the potential for new protectionist forms of regional integration to emerge. The debate over ‘open’ versus ‘closed’ regionalism has grown somewhat stagnant, calling for a new approach to analysing the liberalism of regional institutions. This article focuses on the unique situation brought about by the onset of the Covid-19 crisis. This pandemic gravely impeded the free flow of goods and services and also inspired zero-sum competition between states over the acquisition of key medical equipment. Global multilateral cooperation was relatively weak as a result of geopolitical tensions (exacerbated by the pandemic) and in this context, the regional dimension was more important than ever. This article investigates what key regional institutions did and said in the first six months of the crisis, covering the European Union (EU) and a select group of African, American and Asian regional institutions. Specifically, it will look at the Economic Community of West African States (ECOWAS), the East African Community (EAC), the Common Market of the South (commonly known as Mercosur) and the Association of South East Asian Nations (ASEAN).

A lot has been written about the impact of Covid-19 on international trade and regions in general (Ferrera et al. 2021; Hoekman 2020; Oloruntoba 2021). Less attention has been paid to the trade policy and ideology of specific regional institutions in the crucial early period of the global crisis (January to June 2020). This article offers a theoretically informed analysis of how they framed the trade and economic dimensions of the crisis, together with a study of the different forms of trade measures (based on the official records of international organisations). Specifically, it analyses the institutions in terms of different forms of liberalism during this initial crisis, asking to what extent they supported (verbally or in deed) liberal economic and/or liberal internationalist values. At such a critical conjuncture, there is an opportunity for regional institutions to radically change policy or greatly increase their power (Debre and Dijkstra 2021). Thus, the research also examines the agency of different forms of regional institutions. In both respects (ideology and agency), a qualitative comparative approach is used to shed light on the context and meaning of the words and actions of these institutions. More generally, it aims to make a contribution to the broader understanding of the role of regional institutions in the changing international system.

While there are many forms of regionalism, the focus here is on regional institutions (hereafter RIs) that are concerned with integration. Regional integration is understood as the formal agreement by states to create common laws and institutions for their region. This is different from other forms of regional cooperation and broader socio-economic regionalisation patterns (Breslin and Higgott 2000). Each of these institutions has its own particular legal and political-economic contexts, with quite different economic structures and trading patterns. As such, it was by no means expected that their responses would be similar or that there would be a correct template to follow. Simplistic comparisons of RIs are not appropriate, but we can – and indeed must – examine how different regions develop in relation to each other (de Lombaerde et al. 2010). Likewise, there is no normative bias that these institutions should be ‘liberal’; rather, it is assumed here that it is important to understand the relationship of regional institutions to this foundational political concept, which has been hegemonic in the international system but is increasingly under threat. Starting with a theoretical and analytical overview of the relationship between regional integration and liberalism, the article will then explain the methodology and case studies, and analyse the responses of the five RIs to the crisis. The last section presents the main findings before drawing broader conclusions and suggesting avenues for future research.

Liberalism and comparative regionalism

RIs are composite, mutable and contested entities (van Langenhove 2020). Accordingly, this research is guided by constructivist principles, implying that these institutions are ‘constructed’ intersubjectively by their membership through discourse and action over time (Checkel 2005). A period of crisis can offer an opportunity to consolidate or reshape the RIs’ identity and ideological posture, and the language that is used (as well as the actions that are taken) is therefore crucial. The liberal constructivist thesis (or hope) suggests that international organisations spread liberal norms and values (Ibid). Liberalism embodies a vast complex of ideas and is a contested, historically contingent philosophy (Moravcsik 1992; Doyle 2005; Pitts 2018). This article will focus on two key elements that pertain to regional institutions: economic liberalism, specifically support for trade liberalisation, and broader liberal internationalism. Economic liberalism emphasises the separation of the economic from the political, foregrounds the power of private economic actors in the market and, on the international level, advocates free trade and country specialisation (Smith 1910; Patomäki 2022). The term ‘neoliberalism’ is more widely used to denote contemporary economic liberalism; it implies more systematic support for ‘market rule’ and the intrusion of market forces into all aspects of life (education, health) and for the depoliticisation of the economy. The meaning and continuing relevance of neoliberalism are much debated, and indeed some argue that the term has been so widely used as to lose its explanatory value (Peck 2013). Therefore, this article uses the simpler term of ‘economic liberalism’ as an analytical concept to analyse the role of RIs during the crisis.

The relationship between regional trade agreements and ‘regional integration’ and broader multilateral trade liberalisation is problematic. Regional integration contributes to the liberalisation of trade within the region, but its impact on outsiders is less clear. Even regional preferential trade agreements divert international trade, while deeper customs unions and economic integration arrangements have the potential to become highly protectionist (Bhagwati et al. 1999). In its formative period after WWII, the General Agreement on Tariffs and Trade (GATT) adopted a pragmatic approach allowing for regional trade agreements (even though they are contrary to the most favoured nation principle), but specified that they should cover all trade and operate transparently. Coalitions did become increasingly important in the GATT, but these were usually not regional alliances (Narlikar 2003). European integration, in particular, has caused controversy over its economic ideological essence from the beginning. It was supported by many neoliberals; the project of transnational economic laws binding states to allow the free circulation of goods is quintessential liberalism (Slobodian 2018, 183). Yet, many other features, from the operation of its external trade policies to its many political interventions in the economy, were not liberal (Ibid).

There have been several historical periods (notably the 1930s and 1980s) when fears were expressed that the world was degenerating into ‘regional blocs’. This vision never materialised and, indeed, in most regions, integration has not developed sufficiently for this to be possible, for better or for worse. Said fears re-emerged during the recent eruption of disorder and illiberalism. A World Economic Forum paper of 2019 identified various scenarios for future international trade relations, one of which was a global system resting primarily on ‘competing coalitions’ that could well be dominated by protectionist regional blocs (Tan and González 2019). The debate is often considered in terms of the basic binary of ‘open regionalism’ versus ‘closed regionalism’ (Bhagwati et al. 1999; Gamble 2019). In the immediate post-1989 era, regional institutions that had emerged or were being revitalised were very much geared towards engagement with the global economy and could be considered as ‘open’ (Gamble 2019).

In more recent years, the open versus closed binary has mostly been applied to RIs in Latin America, where starkly different regionalist schemes had been instigated, often with explicit ideological dimensions (Mejía Baños et al. 2020). This is in contrast to the rest of the world, where the ideological element of integration is usually downplayed in official discourse. Generally, the ‘open-closed’ binary is too simplistic a paradigm to characterise contemporary regional institutions, in particular when one considers the complexity of new trade issues (non-tariff barriers, services, data and environmental measures) and the fact that they are continually adapting/evolving. In Latin America, Mercosur moved from a strongly liberal/neoliberal position in the 1990s to a broader forum for economic cooperation and public power in the early 2000s (Kaltenthaler and Mora 2002). ECOWAS began as an effort to forge a New International Economic Order in the 1970s and then adapted to a more liberal pro-globalisation stance by the 1990s.¹ The question is what direction these institutions will take now that the global political economy is again undergoing a process of change – and the answer is likely to be nuanced.

Beyond the economically liberal focus on free trade, there is a broader liberal internationalist/liberal institutionalist vision of multilateral cooperation for progress, based on idealism or rationalism (Moravcsik 1992; Ikenberry 2020). This broader sense of liberalism is key to John Ikenberry’s (2020, 206-7) project to rescue the liberal international order more generally from excessive economic liberalism/hyper-liberalism, which is seen as politically unsustainable. Liberal internationalism involves the use of institutions to manage interdependence and reduce transaction costs and, on a more idealistic level, to support international law and constitutionalism. While economic policies are often the basis of cooperation, the broader meaning is a deeper sense of openness (to people, ideas and perspectives) and partnership between states (Ikenberry 2020). Most RIs embody liberal policies of cooperation for mutual benefit, with each other and outsiders, developing institutions to build trust and facilitate the mobility of people as well as goods. However, they are not truly multilateral in the global sense, being geographically restricted, and there is clearly the potential for RIs (such as the Gulf Cooperation Council) to act as powerful illiberal forces. In this regard, this article will analyse the RIs’ general contribution to liberal internationalism (international law, multilateralism and norms of cooperation) in this crisis.

Comparative regionalism and agency

In investigating the relationship between these regional institutions and liberalism, the obvious point to repeat is that they are all very different. The risks of any form of crude regional integration comparisons are well known, not least of which is an EU-centric approach to ‘evaluating’ other institutions (Acharya 2012; Börzel and Risse 2019). The general problem arises not just from the deep difference but also from the sheer complexity and different levels of analysis involved in studying regionalism (de Lombaerde et al. 2010). Even the basic vocabulary is highly contested, and an eclectic range of theories is used (Sbragia 2008). However, Philippe de Lombaerde et al. (2010) argue convincingly that comparison can still be fruitful, provided there is a suitably refined theoretical and analytical framework. This comparative analysis could be based on specific institutional features or core policies such as budgets (Ibid), socio-legal cultures (Duina 2005) or the different drivers of integration (Börzel and Risse 2019). This article uses the theoretical analytical lens of liberalism to compare and contrast these different institutions, over a relatively short period. This comparative analysis is conducted in full awareness of the RIs’ differences; there is no expectation that they should be similar. The aim is to understand the impact of the crisis on different RIs and contribute to our understanding of their role in the international system.

The intergovernmental versus supranational typology is a classic institutional comparator (Lenz and Marks 2016), which in turn is linked to the capacity for agency. Among our case studies, one institution – the EU – clearly has much stronger supranational institutions and much greater bureaucratic capacity and autonomy (as well as a supranational legal system) than the others. It is logical that stronger supranational institutions enable greater agency (see below); yet, it is noteworthy that even in the case of the EU, many propose an essential intergovernmental dynamic in reacting to crises, with member-state bargaining and collective decision-making as the key driving factor, albeit strongly facilitated by the supranational institutions (Bickerton et al. 2014). Theoretically, at least, it is possible for much looser institutions such as ASEAN and Mercosur to exert a form of agency via their member states taking action collectively, whether they use formal institutions or not; this broader form of regional agency is also considered here (but it must be noted that the latter is unlikely to have the same degree of agency as that of an institution with strong autonomous competences).

Literature review: the pandemic’s impact on international organisations

The pandemic dealt a multifaceted blow to global neoliberalism, which was already reeling from a number of factors (Mariotti 2022). Given the impact of the Covid-19 lockdowns and the zero-sum game struggle to acquire key medical goods, liberal notions of comparative advantage in a global marketplace seemed redundant. Much has been written about the impact of the pandemic on elements of trade policy and politics (Hoekman 2020), especially on the role of ideas and discourse (Orbie and de Ville 2020), given the aforementioned challenges to liberal norms of trade. A real-time, policy and intellectual debate took place as liberals rallied support for a free-market approach to resilience and stressed the self-defeating nature of protectionism (Baldwin and Evenett 2020). It is generally argued that the crisis led to the further supplanting of neoliberal practices and norms, but this is not universal (see Baranes and Hazen 2022). With regard to the role of international and regional institutions, notwithstanding the enormous problems, many academics saw the crisis as an opportunity for them to play a stronger role (Debre and Dijkstra 2021). As traditional modes of governance are clearly inadequate to deal with the scope and intensity of emerging problems, a space is opened up for different institutions to potentially increase their role and accumulate power. In this case, the transnational nature of the problem offered an opportunity for different kinds of international organisations to increase their roles (especially geographically based regional institutions, given the geospatial nature of the pandemic).

Nevertheless, the role of the World Trade Organization (WTO) and global institutions such as the G20 was regarded as relatively weak during the pandemic; for example, the G20 came up with declarations and commitments that were substantially weaker than their response to the global financial crisis (Hoekman et al. 2020, 15). Maria Josepha Debre and Hylke Dijkstra (2021) qualified and coded the response of over 80 international organisations (including major regional institutions) to the initial crisis, with a focus on whether their power had increased. Their approach is broadly similar to that used in this article, but their sample is much larger and they are less concerned with the substance of policy and ideology. Their main conclusion is that bureaucratic capacity and, to a lesser extent, the degree of delegated power explain why some international organisations could increase their role and innovate during this time. Steven van Hecke et al. (2021) also stress how institutional autonomy and the attitudes of member states determined the performance of organisations during the pandemic. The findings of this article bear this out, but – as noted in the previous section – it is at least theoretically possible for highly intergovernmental institutions to increase collaboration and coordinate trade measures even if the regional institutional bedrock is looser. A lot has been written about the EU’s reaction to the crisis, in particular (Ferrera et al. 2021; Alcaro and Tocci 2021). The EU is unique in terms of the range of policies over which it has substantial power (now including health policy), and this article will focus on the key aspects of its trade policy and what increased regional integration means for its international political role.

Although coverage of the role of different regional institutions around the world has generally been critical, some useful features have been noted in the case of Africa (Medinilla et al. 2020; Oloruntoba 2021). Avraham Izhar Baranes and Timothy Hazen (2022) have argued, perhaps counter-intuitively, that the responses of regional institutions have been very much shaped (and hampered) by neoliberalism; they have either worked through private sector actors or relied on neoliberal institutions such as the International Monetary Fund. The Caribbean Community and Common Market is an example of a regional institution that was seen as having played a useful role, in terms of health policy and representation in the WTO (Alden and Dunst 2020). The added value of this article to the above corpus is its focus on regional institutions (which are profoundly different from global multilateral institutions) as well as trade and ideology (rather than a holistic evaluation of ‘performance’).

Methodology and case studies

The method adopted here is a twofold approach. Firstly, a framing analysis of the language used by the RI leadership during the period in question is carried out. Framing refers to how political and policy issues are communicated and represented: what is emphasised and de-emphasised, what narrative is told, what kind of language is used and so forth (Entman 1993; Holden 2019). Official texts, press releases and speeches were stored on NVIVO (63 in total) and thematically coded. There is an asymmetry in the number of texts as some institutions, in particular the EU, produced a lot while others, such as Mercosur, said very little (which is interesting in itself). The codes were derived partly inductively and partly from the theoretical and analytical framework. The codes are:

• Economic liberalism: denoted by a focus on market solutions, market forces and the maintenance and further liberalisation of trade;

• Intraregional liberalism: compatible with the above, but only focused on the region in question;

• Liberal internationalism: a broader political emphasis on international cooperation beyond the region, supporting multilateralism, institutions such as the WTO and the rule of law;

• Interventionism/dirigisme: more political efforts to shape trade and production by regional institutions (action by the regional institution as an entity or the coordinated actions of member states);

• Protectionist regionalism: outright efforts to manage trade (reduce imports or exports in the case of scarce goods) at the regional level, without considering the impact on external actors.

These thematic codes are not necessarily mutually exclusive, although protectionist regionalism is clearly in opposition to economic liberalism. A strong form of interventionism is in tension with economic liberalism and intraregional liberalism, but a certain degree of intervention is almost inevitable in a crisis such as this. Liberal internationalism is compatible with different forms of economic liberalism and with some forms of interventionism but it is in tension with protectionist regionalism. Although quantitative content analysis was performed on the texts, the research is essentially qualitative as the words have to be understood in context. The coding scheme here is straightforward as the implications of the language are clear and do not require much interpretation. For example, a statement in support of liberalising trade is coded as economic liberalism, whereas a statement calling for more proactive regional-level control of trade flows is coded as interventionist, or it is coded as protectionist if the policy advocated is highly restrictive and the language includes terms such as import substitution and reshoring. Each institution is given a simple rating of “strong”, “moderate” or “absent” for each code. This rating is a qualitative judgement based not just on the frequency of such coded language, but also on its strength (prominence in texts, clarity and power of language).

The second approach consists of an empirical analysis of the trade policy actions and the institutional behaviour (the agency in sum) of the RIs. The source for the trade policy measures is a database by the International Trade Centre, which monitors the actions taken by states (covering those officially notified to the WTO at the time and some other actions).² There is a wide range of regional institutions to choose from. The selection of case studies was based on their geoeconomic significance and their competence in trade policy, but global coverage was also an overarching consideration. Given these criteria, the EU had to be covered. There are numerous institutions in Latin America, but it is generally agreed that Mercosur is the most important (being one of the most legally developed and centred on the two largest South American countries). In Africa, two of the institutions commonly considered to be the most effective (ECOWAS and EAC) were chosen. These are all customs unions of sorts, which have at least a legal basis for some kind of agency in trade. ASEAN is a useful example of an RI that is not a customs union but is active in international economic diplomacy.

Of the five institutions covered here, the EU is by far the most integrated. It is a customs union and a regulatory union (via its single market), while 20 of its 27 members are also in a currency union. The single market gives member states extensive access to each other’s markets (including in many services) and the quid pro quo for this is that they accept an extensive legal system, monitored and enforced by powerful supranational institutions such as the European Commission (EC). The EU’s customs union is comprehensive in scope and the EC has significant power to negotiate and manage trade defence measures (Young 2017). Given the relatively high degree of restrictions on state autonomy, the politics of burden sharing take place on a much more ‘advanced’ and institutionalised level compared to most other regions (Thielemann 2012). Due to the extensive, if somewhat uneven, presence of the EU in all aspects of social and political life, the expectations are high. By contrast, ASEAN (which includes both highly developed and least developed countries) is known for privileging a flexible, multi-speed and member-state-driven approach to regional cooperation. Notwithstanding the many agreements, institutions and initiatives, actual legal economic integration is limited (Dosch 2017). Furthermore, although ASEAN signs free trade agreements as a bloc, thus exhibiting a form of agency, it is not a customs union and would therefore not be legally bound to develop a common external trade policy on Covid-19-related issues.

As noted, the other three RIs are customs unions, but they have less scope and rigour of implementation than the EU. Although legalised, they all rely on intergovernmental institutions for key decisions (ECOWAS has a Commission but without the power of its European counterpart). Mercosur is a political and economic alliance based mainly on a rapprochement between Argentina and Brazil. Founded in 1991, it has outlasted dramatic changes in the politics and political economy of the region (Kaltenthaler and Mora 2002; Margheritis 2013). Owing to the very different political alignments of the two dominant states (Argentina and Brazil), it was already under strain at the time. Yet, it remains the most important regional institution (in combining critical mass and degree of integration) in Latin America. EAC is widely regarded as one of the more integrated African regional institutions, in terms of its original core membership of Kenya, Tanzania and Uganda, in particular. It has implemented an imperfect but substantial common market and customs union (Bach 2015, 109). However, a mini-crisis was triggered by Kenya’s trade agreement with the US in February 2019. ECOWAS covers a much larger territory and population than EAC; it includes the highly integrated West African Economic and Monetary Union of former French colonies. It has evolved into an important political and security actor in the region, which has severe security problems (Iwilade and Agbo 2012). It has pushed forward with its ambitious economic integration plans, but there are often tensions between Anglophone and Francophone countries as well as other conflicts, such as the disputes that led to Nigeria closing its borders with Benin and Niger in August 2019.

Regions, trade and the initial crisis

This crisis posed acute threats to ‘truly existing’ regional integration. Complex regional supply chains could be broken up by unilateral economic actions in one country; for example, Mexican shutdowns spilled over into the broader North American manufacturing sectors (Behsudi 2020). Moreover, existing geopolitical tensions risked being exacerbated under the pressure of the pandemic; its fissile effect on regions was therefore not to be dismissed. The following section will look at the responses of the five specific regional institutions in more detail.

The EU

The EU’s response to the crisis showed some tensions between open and closed regionalism, although the former was very dominant in its language. The response was led by the EC as the permanent institution with significant powers in trade (Holden 2019). Although the Council of the EU (ministers) and the EU Council (leaders) met virtually as well, their communications did not cover external trade. Faced with the challenge posed by Covid-19, the EC had to combine its liberal rhetoric with overwhelming interventionist logic. Some individual European ministers openly called for radically different and illiberal trade and economic solutions (Orbie and de Ville 2020), but the EU as a whole did not follow them. Instead, it came up with the concept of “open strategic autonomy” as a compromise. Although this has strongly protectionist connotations, the then Commissioner Hogan emphasised that this was more about the intelligent management of trade and production rather than the development of some form of autarchy (Hogan 2020c). Strategic autonomy would be achieved by intelligent stockpiling, diversification of supply chains and so forth, rather than by a fortress Europe approach (Hogan 2020c; Orbie and de Ville 2020). Europe, he stressed, could never be self-sufficient. Likewise, the revitalised call for a “Europe that Protects” was in no way a call for economic protectionism.

The EU’s response emphasised the unprecedented financial aid and flexibilities offered within Europe (European Commission 2020a). Protection of intra-EU trade was a priority, as “the EU must respect our Single Market and ensure that there are no internal barriers to intra-EU trade” (Hogan 2020a). Its solution to the crisis included classical liberal responses such as reducing tariffs on health and pharmaceutical tariffs globally (Hogan 2020b). Hogan (2020a) criticised “distortive state interventions or unjustified export restrictions across most economic sectors”. He went so far as to say: “The global integration of supply chains therefore provides a guarantee for the availability and diversity of supply” (Ibid). The subtext to this strong defence of liberal principles is that the EU itself– as the price for removing barriers to trade in Covid-19-related products within the EU – had imposed controls on the export of personal protective equipment (PPE) beyond the EU (see below). It was at pains to emphasise that these were controls, not a ban. Nevertheless, this move reflected a striking dynamic of intraregional liberalism leading to a form of closed regionalism and it received much criticism for the signal it sent (Bown 2020, 34). The Commission also issued guidelines and began to prepare legislation on the screening of “foreign” direct investments (from outside Europe) “to protect critical European assets and technology in the current crisis” (European Commission 2020a, 4).

As might be expected, the EU maintained a strong liberal internationalist tone in its language. It stressed the need for a rules-based trading system and the WTO in particular (encouraging others to use the WTO as a coordinating mechanism and reiterating its support for reform of the institution, including dispute settlement procedures), as well as broader support for global governance structures (Hogan 2020a). Its High Representative for Foreign and Security Policy framed the crisis as a transformation of the international system and as accelerating rivalry with authoritarian states and value systems (naming China as an example). The rhetoric suggests that there is a strong sense of EU agency. “Only by pulling together and cooperating across borders can we beat the virus and contain its consequences – and the EU has a central role to play” (Borrell 2020).

Let us now turn to the policy details. On 4 March 2020, the German government banned the export of many medical products (including to EU states). On the same day, the French government commandeered some key materials (masks mostly), effectively banning any exports (Reed Smith 2020). Such restrictions were an existential threat to European integration and the Commission worked quickly and effectively to bring them to an end. On 15 March, the EC introduced a regulation that would require licences for the export of a broad range of PPE products from the EU to third countries. When it was renewed on 26 April, the list of covered products was reduced and more exemptions from licensing were given to neighbouring countries, forcing member states to “swiftly grant authorisations for exports for humanitarian purposes” (European Commission 2020a). Although an example of closed regionalism, it is highly probable that in a counterfactual world without European integration, exports would have been restricted more. The controls ended on 26 May (European Commission 2020b). The Commission noted that 95 per cent of all applications were approved and more than 14 million PPE products were exported from the EU since 26 April (European Commission 2020a). As for the broader trade and economic response, the EU implemented green lanes within Europe to minimise the inevitable disruption to trade; it also initiated joint procurement of medical supplies and coordinated funding for vaccine research to generate economies of scale. Furthermore, it established an unprecedented mechanism of economic burden sharing and solidarity, including common debt, instruments of a type that had previously been denied (Jones 2020; Alcaro and Tocci 2021). Finally, it significantly loosened the national constraints imposed by EU law on competition policy and fiscal controls.

Mercosur

As we have seen, Mercosur was already under strain in the early 1990s as a result of the different political alignments of its two dominant states, Argentina (left-wing populists) and Brazil (a form of right-wing neoliberal populism). This misalignment and the idiosyncratic approach to the virus in Brazil (whose president refused to implement adequate health measures) shaped the regional response (as an intergovernmental organisation, Mercosur is prone to national politics). On 18 March, the four national presidents met virtually and signed a declaration in which members committed to keeping trade and transport links open and maintaining a free flow of information (MercoPress 2020). They also agreed to look into removing tariffs on medical equipment, but there was no follow-up on this. Member states took a range of bilateral actions on this issue. There was no discussion of Mercosur’s agency in the changing global system, perhaps unsurprisingly, given the extreme splits within the institution. This also explains why Mercosur’s official communications at the time lacked any support for liberal internationalism. On 24 April, Argentina announced that it would no longer participate in the bloc’s new trade agreements, which its government felt were too liberal and threatened the overall unity of the bloc (Arredondo 2020). The following month, Argentina revoked this decision but declared that it would maintain a protectionist approach to new agreements.

Table 1 shows the main trade measures taken by each member state.³ In the case of Argentina, for example, most of the direct trade measures were liberalising (on general imports or imports from specific countries). At the same time, however, the government took a range of measures in response to the crisis, including extensive price controls (boosting the price of oil and reducing the prices of food and other necessities) that were bound to affect its trading partners. Brazil took four liberalising measures and imposed one export ban and one set of export controls. At home, the government took a range of social welfare, employment and socio-economic measures, but without strong intervention in industry or anything like the price interventions of Argentina. The two smaller members also took their own moderate trade measures and interventions. Extensive border controls were introduced, especially with Brazil, which became a centre of the epidemic. For example, Uruguay set up 800 new border check-points during this period; all countries introduced quarantine measures and extensive travel bans. In Mercosur’s case, the pressures on the organisation predate the pandemic but it is clear that the pandemic did not have a unifying effect.

Table 1. Mercosur: trade measures from 1 January 2020 to 30 June 2020.

Country	Liberalising measures	Restrictive measures
Argentina	5: including tariff reductions, suspension of anti-dumping measures and VAT waivers for health-related products.	1: export licensing of medical ventilators.
Brazil	4: including tariff reductions, suspension of import licensing requirements and other control measures, suspension of anti-dumping measures for health-related products.	2: export prohibition on some medical products and export permits required for others.
Paraguay	2: changes in VAT for some health-related products, tariff reduction for others.	1: export licensing requirement for PPE.
Uruguay	1: tariff elimination for some health-related products.	0

ASEAN

Neighbouring ASEAN countries and China immediately felt the health and economic shock waves of Covid-19. As noted, ASEAN does not have the same integrative pretensions as other institutions covered in this article. Despite having no supranational element, it has been highly active in trade, economics, security and social fields. Observers had called for the moment to be seized to move integration in Asia “beyond free trade paradigms”, but this was not followed by ASEAN (Oba 2020). ASEAN economic ministers released an initial response to the virus in March 2020. They agreed to “remain committed in keeping the ASEAN market open for trade and investment” (hence making no internal/external distinction) and called to improve “long-term supply chain resilience and sustainability” (ASEAN 2020a). They also emphasised the need to enhance cooperation with international partners. A leaders’ summit reiterated this message the following month. They agreed to “[r]emain committed to keeping ASEAN’s markets open for trade and investment, and enhance cooperation among ASEAN Member States and also with ASEAN’s external partners” (ASEAN 2020b, 3). ASEAN was noteworthy for making little distinction between the internal and external dimensions. It also called to “refrain from imposing unnecessary restrictions on the flow of medical, food and essential supplies, in accordance with the rights and obligations under the WTO covered agreements … ”(ASEAN 2020b, 4). On 19 June, ASEAN announced a Plan of Action, although it typically stressed its non-legally binding nature, which was again strongly liberal in terms of international trade (ASEAN 2020c). In sum, ASEAN had a strong sense of agency, namely as a vehicle for its member states to retain a liberal trading system (not as an agent of change). Its framing was impeccably liberal internationalist, but others have noted how the crisis itself served to bolster authoritarianism within ASEAN states (Rüland 2021).

As for its actual policies, trade measures were taken by individual member states (see Table 2). These applied to fellow members as well as third countries and there was no sense that these were coordinated. ASEAN includes some major producers and exporters of PPE products, but some individual countries were net importers even before Covid-19 (Suvannaphakdy 2020). Indonesia, for example, imposed strict export bans on some medical equipment. Other countries also introduced export restrictions. Most strikingly, Vietnam applied export restrictions (quotas) on rice (of which it is a major exporter) as well as key medical products. This was a short-term measure that ended in May 2020, but it could have been highly significant as the country is a major source of rice for many developing states. ASEAN countries individually implemented many stimulus packages for different sectors (with no sign of real coordination), although given ASEAN’s limited legal economic integration, this is less problematic than for an entity such as the EU.

Table 2. ASEAN: trade measures from 1 January 2020 to 30 June 2020.

Country	Liberalising measures	Restrictive measures
Brunei	1: tariff reduction on personal hygiene products.	0
Cambodia	0	2: export prohibition of rice/fish, terminated 20 May; export prohibition on face masks, terminated 4 May.
Indonesia	3: tariff reduction on medical supply products; suspension of import certification requirements for PPE, terminated in June; suspension of import certification requirements for onions and garlic, terminated 31 May.	2: export prohibition of medical supply products and PPE; and on the raw materials needed to produce the latter, terminated in June.
Laos	1: tariff reduction exempting health-related items and medical equipment from import duties, taxes and other charges.	0
Malaysia	2: tariff reduction eliminating import duties and sales tax on PPE and medical supplies.	1: prohibiting export of masks.
Myanmar	1: tariff reduction waiving specific goods tax on medical supply products.	1: suspension of rice export permits (terminated 1 May).
Philippines	1: tariff reduction exempting health-care equipment and supplies from import duties, taxes and other fees, terminated 23 May.	2: export prohibition of rice from a province (Bukidnon). 10% tariff increase on petroleum and petroleum products.
Singapore	2: including relaxation of import licensing requirements for health-related products and elimination of import tariffs on essential medical, hygiene, pharmaceutical and agricultural products.	0
Thailand	3: including tariff reductions on medical supply products and equipment for producing medical devices; import VAT exemption for donated materials.	2: export prohibition of masks; export ban on eggs, terminated 1 May.
Vietnam	1: tariff reduction exempting PPE-related products from import tax.	4: prohibition of trade and consumption of wild animals (active); licensing of masks, terminated 29 April; export prohibition of 37 pharmaceutical products, terminated 7 May; rice export quota set, terminated 1 May.

ECOWAS

The ECOWAS Commission has far fewer delegated powers than the EC and action would depend on the member states. No references to Covid-19 were made during a summit of heads of state and government on 9 February 2020; instead, the main focus was on the closure of Nigeria’s land border. As the spread of the virus accelerated, a summit on 23 April looked at how to provide an adequate health and economic response. The Commission used a language that stood out for its internationalist approach, namechecking “the African Union, African Development Bank, United Nations, International Monetary Fund, World Bank, European Union, the G20” (ECOWAS 2020). This is partly because it clearly recognised that it needed assistance from international actors.

As a bloc, ECOWAS did not follow the call for more radical interventionism by the president of Niger, the group’s president in office at the time. In his summit speech, he called for a return to “the founding principles of our organization”, “solidarity and collective self-sufficiency” and the “fair and equitable distribution of the costs and benefits of cooperation and economic integration” (Mahamadou 2020). More precisely, he advocated “control of production factors: energy, transport, and NICT [New Information and Communication Technology] … – the creation of industrial centers aimed at transforming the comparative advantages of each country” (Ibid). The bloc as a whole agreed on much lighter targeted regional interventionism, as the summit called to “implement urgent measures to support the local production of consumer goods, including agricultural products, thereby reducing the import bill for these goods” and to “set up a support programme for the pharmaceutical and health protection equipment manufacturing sector, whose output covers barely 20% of the region’s current consumption needs” (ECOWAS 2020). It stressed the need for intraregional liberalism rather than global liberalism, calling to “avoid the imposition of import restrictions on other ECOWAS countries [author’s emphasis], particularly with regard to essential goods (drugs, food, etc.)”.

The actual trade measures taken within ECOWAS were very much member-state-driven (see Table 3) and, as in other cases, there is no evidence that they were coordinated; restrictions applied to fellow member states as well as third countries. Being the largest economy, Nigeria’s only official trade measure was to liberalise imports of some health-related products (see Table 3), but the impact of its economic shutdown and ongoing border disputes has been severe. Ivory Coast took two trade measures, one of which was a restriction on the export of medical material. The region is not a big supplier of PPE and only one other country (Mali) restricted its export. It could be seen as a positive result that only one country (again Mali) imposed any restrictions on food exports. The other formal measures taken were all moderately liberalising.

Table 3. ECOWAS: trade measures from 1 January 2020 to 30 June 2020.

Country	Liberalising measures	Restrictive measures
Benin	0	0
Burkina Faso	1: relief efforts to facilitate the importation of medical products and other consumer goods.	0
Cape Verde	1: exemption from customs duties and VAT for imported masks.	0
Gambia	0	0
Ghana	0	0
Guinea	0	0
Guinea-Bissau	0	0
Ivory Coast	1: temporary tariff reductions on PPE and medicines.	1: ban on the export of hand sanitisers.
Liberia	0	0
Mali	1: temporary tariff reductions on key foodstuffs.	1: ban on the export of PPE and key foodstuffs.
Mauritania	0	0
Niger	1: temporary exemption from tariffs for goods to combat Covid-19.	0
Nigeria	1: temporary suspension of import duties and VAT for medical supply products.	0
Senegal	1: temporary suspension of import duties and VAT for medical supply products.	0
Sierra Leone	0	0
Togo	1: temporary suspension of import duties and VAT for medical supply products.	0

EAC

The acute health and economic implications of Covid-19 were significant in East Africa, as in other parts of Africa. High-profile issues included the plight of the Kenyan flower sector (reliant on sales to Europe) and the collapse of Ugandan exports to China, as well as general concerns about food security. Despite this, EAC’s leadership took some time to meet. The secretariat worked up several versions of an EAC COVID-19 Response Plan while trying to rally a unified approach from the member states (EAC 2020a). However, it could do little given the political tensions between the countries. A virtual meeting of heads of state was held on 12 May after being postponed in April (EAC 2020c), but only four countries were included (Burundi was absent owing to elections and Tanzania because of a dispute with Rwanda). A meeting of health and EAC affairs ministers had been held on 25 March (EAC 2020b). Very few references to specific elements of trade policy were made. The March meeting simply “[r]esolved to facilitate free movements of goods and services in the region”(EAC 2020b, 2). In terms of regional collective action, it “directed partner States to support local companies, to ensure the local production and availability of key consumables/products used in COVID-19 response” (Ibid). The leaders’ summit did not reach a trade agreement. Likewise, EAC’s role in the broader international system (political or economic) was not discussed. Generally, liberal internationalist values are weak in EAC discourse at this time, but there is a strong focus on intraregional openness. The Response Plan went into more detail on practical logistical and customs issues, trying to balance improved health measures with maintaining the flow of goods. Regional coordination of the response was allocated USD 18.6 million. Maintaining free movement of goods and services in the region was clearly a high priority and was emphasised early and often. In reality, intraregional liberalism came under pressure. Tensions erupted between Tanzania and Kenya over the treatment of truck drivers after mutual border closures from 16 to 18 May (Ilako and Amadala 2020). This followed several trade disputes between the two neighbouring countries in recent years. On 22 May 22, the two governments agreed to bilateral talks and a testing regime. EAC had no formal role in the dispute. Meanwhile, each country took individual trade measures (export restrictions and import liberalisations for PPE, with Uganda adding an import restriction on second-hand clothes), again with no sign of coordination or special provisions for fellow member states (see Table 4). On the whole, EAC’s response was politically and legally weak.

Table 4. EAC: trade measures from 1 January 2020 to 30 June 2020.

Country	Liberalising measures	Restrictive measures
Burundi	0	0
Kenya	1: VAT cuts on all goods.	2: prohibition of import of used footwear and ban on the export of face masks.
Rwanda	0	0
South Sudan	1: tariff elimination for some health-related products.	0
Uganda	0	1: prohibition of import of used footwear and textiles (terminated on 22 May, one day after launch).
United Republic of Tanzania	1: tariff exemption and VAT cuts for selected medical products.	0

A holistic comparison of the RIs’ role in this period

Intraregional liberalism is very strong in the framing of all the institutions examined here, as it is in their legal and normative DNA (Table 5). They did not entirely follow through in practice: the EU was most rigorous in overcoming different member-state restrictions; in the other institutions, individual states took trade actions. Serious disputes took place in South America and East Africa. Their more global framing of economic liberalism was much more variable and ambivalent. Mercosur member states had an explicit debate about the pursuit of further trade liberalisation with the outside world but said little on this topic with regard to the pandemic. ECOWAS and EAC expressed a much greater desire for interventionism/protectionism at the regional level but were unable to substantially put this into practice in the form of any collective trade policies. Neither did they succeed in entirely coordinating member-state trade policies in one direction or another. ASEAN de-emphasised a response in terms of regional economic interventionism in favour of liberal economic values. In practice, some ASEAN member states imposed potentially important restrictions on trade. The EU put the strongest accent on economic liberalism, although this was partly to compensate for the illiberal connotations of some of its actions and rhetoric (the PPE controls and strategic autonomy).

Table 5. Framing summary.

	Economic liberalism	Liberal internationalism	Intraregional liberalism	Regional interventionism	Regional protectionism
EU	Strong	Strong	Strong	Moderate	Moderate
ASEAN	Strong	Strong	Strong	Moderate	Absent
EAC	Moderate	Moderate	Strong	Moderate	Moderate
ECOWAS	Moderate	Strong	Strong	Moderate	Moderate
Mercosur	Moderate	Absent	Moderate	Absent	Absent

To a certain degree, liberal internationalism is also in the DNA of these institutions, as they are multilateral organisations based on legal treaties. However, the strength of this liberal internationalism varied, and institutions with stronger self-confidence – such as the EU – or those in acute need of international aid – such as ECOWAS – used this language most. Mercosur and EAC, struggling with internal challenges, used it least. The sense in which the institution was framed as a major agent in international relations also varied. In the EU, this was clearly the case, as with ASEAN (albeit with much looser legal connotations in practice). ASEAN’s framing of international relations was perfectly liberal (although, as noted, this was not at all reflected in domestic politics). In evaluating the different forms of agency in practice, the (unsurprising) conclusion is that the RI with the strongest supranational institutions (with delegated powers and significant bureaucratic capacity) – namely the EU – was the most active and effective. However, as three of the other RIs are customs unions of different kinds, there was at least a strong legal and logical basis for collective action on trade issues on their part. This did not occur. Instead, a broadly realist and statist logic predominated in which national political and economic interests vastly outweighed regionalist tendencies. Other explanatory variables for the level of collective action include the degree of economic interdependence and the culture of political elites (Börzel and Risse 2019). It should be noted that this crisis came at a particularly bad time for an institution like Mercosur, as Brazil was under the populist leadership of Bolsonaro. This just goes to show that intergovernmental institutions are hostages to domestic political trends.

Conclusion

The crisis presented an opportunity for RIs to increase their role in trade that was not exploited in most cases studied here. In the crucial initial six months of the pandemic, action on trade was overwhelmingly state-driven (even in Europe, individual member states took immediate action until the EC took control). Outside of Europe, individual member states took concrete trade measures (tariffs, export controls), while the regional institutions offered a coordinating and/or monitoring function and obtained political commitments to minimise the disruption to intraregional trade. As such, the institutions covered here had a low level of agency in that they were more of a forum for state interaction. Of course, the state is usually the locus of decision-making in a crisis, but this pattern has continued as the pandemic has matured; it is clear that, globally, this was not the hour of the regional institution. As such, there was little evidence that the RIs were developing an exclusionary, macro-protectionist and closed kind of regionalism, but neither were they bulwarks of economic liberalism. Except for the EU, they did not play a major role in the trade response or, more generally, in the global political economy.

Developments since 2020 further confirm that the era of global neoliberalism is over, even if there is no clear paradigm to describe the current global system, which involves different forms of statism and geoeconomic positioning. This is significant for regional economic integration as increasing state intervention will, ceteris paribus, distort the neoliberal ideal of a single market with a level playing field. Only the EU has demonstrated the legal/political power and the concomitant supranational investment needed to potentially restrain statism. The global ideological muddle is reflected in the regional institutions studied here, which have always been pragmatic and ideologically fluid. In a sense, this means that they are promising receptacles for Ikenberry’s (2020) hope for a broader pragmatic liberalism beyond economic neoliberalism. They all seek to maintain a form of ‘openness’, sometimes just internally but usually more broadly. Their efforts to promote norms and, at least, coordinate behaviour may well have helped regions to avoid even more fragmentation and strife (the counterfactual is not provable). Even in cases like ASEAN, which includes authoritarian regimes, the institution is promoting a kind of thin liberalism (cooperation, multilateralism, information sharing and so forth). As such, they are crucial partners for Western policymakers trying to defend a liberal international order. It is hoped that the qualitative comparative approach presented here has shed light on the role of RIs at a crucial moment in recent history. As regional integration becomes increasingly diverse, the most pressing research agenda (one that this article contributes to) is to explore how individual regional institutions relate to different ideologies and broader patterns of regionalisation in their geographical space. However, there are still important insights to be drawn from a comparison of how they navigate these issues and how they interact with each other. In particular, more comparative research is needed on how the institutional design of RIs shapes or is shaped by ideological factors. More specifically, the distinction between the thin liberal internationalism of some regional institutions compared to the deeper liberalism developed by others could be a fruitful direction for future research.

Acknowledgements

The author would like to acknowledge the very helpful feedback of the journal’s editors and two anonymous reviewers. Thanks are also due to the Faculty of Arts, Humanities and Business at the University of Plymouth for funding the initial research, and to Michele Fox for her great work as a research assistant.

Additional information

Notes on contributors

Patrick Holden

Patrick Holden is an Associate Professor of International Relations and Politics at the University of Plymouth, Plymouth, United Kingdom.

Notes

1 ECOWAS was founded in 1975 with the aim of developing the collective bargaining power of the region (Francis 2005, 132) at a time when developing countries were campaigning for a radically different international trade and financial system. By the 1990s, this political project had failed utterly and regions like ECOWAS were (at least formally) committed to integrating into the global economy on free-market terms.

2 Officially, 149 measures were notified to the WTO during this period; the ITC counts 285. The latter’s database was consulted for the study of the RIs’ member states.

3 The source for all Tables is the ITC database: https://www.macmap.org/covid19.