African regionalism, economic nationalism and the contested politics of social purpose: the East African Community and the ‘new developmentalism’

Abstract

Over the last decade, a new developmentalism has taken root across Africa, centred on promoting local production and industrialisation. One unintended consequence of this has been the proliferation of economically nationalist policy measures that have increasingly come into tension with the aims of regional integration in Africa. This article sets out to offer insights as to why these tensions are emerging by focusing on the East African Community (EAC) and the growing trend of economic nationalism among its members. Contrary to what rationalist and structuralist accounts might presume, this article argues that this rise in economic nationalism is instead reflective of a weakening of the discursive imperative – or social purpose – that had initially converged various actors around the EAC's integration agenda when revived in 2000. While drawing from the EAC's experience, it concludes by highlighting a broader legitimacy dilemma facing African regional organisations within this ‘new developmentalism’.

INTRODUCTION

Over the last decade, economic policy agendas across Africa have increasingly come to be defined by a common concern around productive restructuring and industrialisation (Lopes & te Velde 2021). Two key pillars appear at the heart of this emerging policy agenda. On the one hand, policymakers have reiterated the role of regionalism in promoting local production on the continent through economies of scale, regional production networks and expanded markets for African producers (O'Reilly 2020). This narrative has been especially reiterated since negotiations started in 2012 for the creation of an African Continental Free Trade Area (AfCFTA) (UNCTAD 2019). On the other hand, there has also been a proliferation of national development plans across African states advocating targeted interventions to support the development of certain domestic industrial sectors (Odijie 2019; Walter 2021). Whilst the two pillars of this agenda are geared towards achieving the same end (i.e. promoting local production), it has been observed that there are evident tensions between the two regarding the means by which this is achieved (Odijie 2019, 2020). That is to say, there is a growing conflict between industrial policies being pursued at the national level in Africa and the objectives of regional economic integration.

In this paper, I set out to offer insights as to why these tensions and conflicts are emerging. I do so by narrowing my focus onto the East African Community (EAC), a regional economic community (REC) with an ambitious regional integration agenda that has faced growing challenges over the 2010s due to the proliferation of economically nationalist policy measures among its members. Two literatures offer potential insights into these trends. Rationalist accounts might place the tensions between regionalism and economic nationalism in Africa as a reflection of specific elite political survival strategies. Several studies, for instance, have asserted that African political leaders only engage with regional organisations performatively, merely as a way to boost their external sovereignty and legitimacy (Söderbaum 2004; Herbst 2007; Gibb 2009). It could be surmised then that African governments are pursuing national economic strategies that conflict with their regional commitments because they have no intention of abiding by these commitments in the first place. Meanwhile, structuralist accounts might place these observed tensions down to features of African economies, which tend to remain relatively undiversified and dominated by primary sectors (Gibb 2009; Odijie 2019, 2020). I argue, however, that both these explanations are incomplete when examining the rise of economic nationalism in the EAC. Notably, both fail to explain why EAC members had previously demonstrated commitment to aspects of the region's integration agenda and why subsequently, since the mid-2010s, they have pursued domestic economic policies which conflict with that agenda.

As such, I argue that we need to complement such accounts by focusing on both the discursive and institutional foundations of regional regimes in Africa. I do so by linking into constructivist writings that have examined the legitimacy and social purpose of international economic regimes (Ruggie 1982; Best 2003; Baker 2018). The assertion these works make is that international institutions are not given or self-evident, but are the work of intense discursive practice. Put differently, the legitimacy of international institutions is underpinned by an underlying social purpose – an assemblage of ideas and meanings, and the discourses through which they are articulated, which serve to specify the role, purpose and necessity of international institutions. Like international regimes, I argue that regional economic regimes are also underpinned by distinct social purposes. In this article, I further contribute this concept by identifying an important temporal dimension that draws attention to the ways in which particular articulations of social purpose can come to be contested over time.

Applied to the case study of the EAC, I argue that the rise of economic nationalism reflects certain discursive and institutional tensions which have developed within the EAC's regional regime. I argue that following the EAC's revival in 2000, its social purpose increasingly came to be articulated in policy discourses through an economic frame that cast its market-integration agenda as a necessary measure to respond to the perceived threats posed by globalisation at the time. Whilst such discursive appeals initially served to converge the perceptions of regional stakeholders around the necessity of the EAC's regionalist agenda, I highlight how the salience of such appeals increasingly came to diminish as regional integration entered its second decade. The consequence of this was that it opened discursive space for the articulation and practice of more developmentalist economic strategies within the region. Although some tentative attempts were made to coordinate such strategies at the regional level (Murray-Evans and O'Reilly 2022; O'Reilly & Heron 2022), it was within national policy spaces where they principally emerged. The reason for this I argue is twofold. The first is that the EAC was initially conceived as space to support market-led development and lacked the institutional capacity to pursue more ambitious coordinated trade and industrial strategies. The second is that as the discursive foundations of the EAC became more and more contested, it opened space for the articulation of more nationalistic (rather than regional) development visions.

The contribution of this article to the literature is thus twofold. First, it adds nuance and contingency to what can be viewed as the determinism underpinning existing accounts of African regionalism. Both rationalist and structuralist accounts tend to take a dim view of the prospects of regional integration in Africa given what they see as the prevailing political and economic conditions that exist across African states. Developing a constructivist analytic framework, this article instead highlights how under certain discursive contexts African actors can converge and commit to the aims of regional integration. As with other recent publications (Mumford 2021), therefore, this article emphasises the role of discursive subjectivities in shaping the trajectory of African regionalism. Second, utilising the EAC's integration experience, this article contributes further to the theorisation of regional integration. Existing scholarship has emphasised how the discursive invocation of external economic threats can create an imperative for regional integration (see Rosamond 1999, 2002). This article builds upon this scholarship by further identifying how the salience of such discursive imperatives can wax and wane over time. Drawing on the EAC's experience, this article highlights that such discursive imperatives remain salient to the extent that they can convincingly cast regional integration as a necessary and non-negotiable response to a given external economic threat and frame national policy responses as either unviable or insufficient. In the EAC's case, the imperative for regional integration initially centred upon a discourse of globalisation as non-negotiable economic constraint. This article, however, charts how this discourse of non-negotiability weakened by the early 2010s, opening space for the emergence of statist development alternatives which, in turn, came to conflict with the objectives of the EAC's integration agenda.

Methodologically, this article employs a discourse analysis to explore how the social purpose of the EAC, as articulated through official documents and speeches, has evolved over time. Put simply, a discourse can be understood as ‘a particular way of talking about and understanding the world (or an aspect of the world)’ (Jørgensen & Phillips 2002: 1). Discourse analysis, therefore, can help to unpack the prevailing (or hegemonic) ways in which actors talk about and interpret their environment at a particular moment in time and the limits this can place on what is deemed legitimate or viable policy (Diez 2014). For instance, this article explores how particular discourses of the global economy – and their change over time – have shaped rationales for regional integration in the EAC and delimited the range of development policy options that are perceived as viable and legitimate. The article avoids the more fine-grained approach to discourse analysis (e.g. Fairclough 1995) – considered more applicable for analysing a small number of texts – and instead opts for a more dynamic and parsimonious approach that traces continuity and change in the discursive practices of the EAC over time (see Siles-Brügge 2011; De Ville & Orbie 2014: 153). The source material for this article derives primarily from documentary materials sourced from the EAC's Information Resource Centre during a visit in August 2017 and through the EAC's online reports database thereafter.¹ This includes publicly oriented sources, such as EAC regional development strategies and speeches by regional political actors, alongside internal EAC meeting reports and workshop proceedings. It also draws directly and indirectly from over 40 interviews conducted in the region since 2017 with regional and national policy makers, private-sector representatives, civil society organisations, donors and policy experts.

Whilst the specific findings of this paper draw from the EAC's experience, I conclude by highlighting their broader implications for RECs across Africa. Like the EAC, most African RECs were established or revived during the 1990s with the intention of supporting market-led development and global integration (Söderbaum 2004). However, as this ‘neoliberal’ consensus shows tentative signs of receding in Africa (Hickey 2012; Harrison 2019), I argue that these RECs face a dilemma concerning how they frame their purpose and relevance moving into the future. The remainder of this article is organised as follows. In the next section, I chart the recent rise of economic nationalism among EAC members and the limits of rationalist and structuralist explanations in accounting for these. I then develop a theoretical framework around the concept of regional social purpose which I use to offer an alternate explanation for the rise of economic nationalism in the EAC that situates this state of affairs within specific discursive and institutional tensions that have emerged within its regional regime. I conclude with comments and remarks on the broader significance of these findings.

THE RISE OF ECONOMIC NATIONALISM IN THE EAC

The EAC was re-established as an REC in 2000, following its earlier collapse in 1977 (Hazlewood 1979). Founded originally between Kenya, Tanzania and Uganda, the EAC has since expanded to include Rwanda, Burundi, South Sudan and most recently the Democratic Republic of Congo (DRC).² At the time of its establishment, the EAC emerged into an already complex regional institutional landscape with its founding members already being spread across competing RECs with their own mandates for regional integration.³ However, despite the crowded field of regional organisations in eastern and southern Africa, the EAC soon established its credentials as one of the more successful RECs in Africa (Bach 2015: 97–8), reinforced through the various agreements the EAC members have agreed to since 2000. This has included the implementation of a customs union and common external tariff (CET) (EAC 2004a), alongside commitments to deepen economic integration under a common market (EAC 2010) and monetary union (EAC 2013). Throughout the 2010s, however, this image of success became somewhat punctuated. This was illustrated by the slow implementation of the common market and monetary union protocols, an increase in intra-regional trade barriers (Mugisha 2019), border closures (Neiman 2019), and the emergence of political instability in Burundi (Vandegiste 2015) and South Sudan (De Waal 2014).

One issue which has received less media attention – but is no less significant – has been the increase in ‘economically nationalist’ policy measures implemented by EAC members in recent years aimed at supporting domestic producers and industries. The Government of Uganda (2014) led the way in this regard when in 2014 it launched its ‘Buy Uganda, Build Uganda’ strategy. This was followed with similar strategies emerging in Rwanda (Government of Rwanda 2015) and Kenya (Government of Kenya 2017). In Tanzania, the country's Public Procurement Act was amended in 2016 to give preferential treatment to local goods and services in government tenders (Extractives Hub 2016). The broad intention of these national strategies was to attempt to leverage domestic demand and public procurement towards goods and services produced nationally and, in doing so, reduce their reliance on imports and support local producers.

Alongside, and as part of these local production strategies, governments in the region have also sought to alter tariff schedules on certain imported products in an effort to support local production. They have done this through the practice of ‘stays of application’ (SOAs), a policy mechanism that allows EAC members to unilaterally diverge from specific tariff lines in the regional CET for up to a year at a time. SOAs were originally introduced as a measure to assist the EAC states transition to operating under a regional CET and were initially used quite sparingly either to reduce tariffs on imported food staples or on finished goods which were not as readily manufactured within the region (Bünder 2018). Yet, as recent research by the International Growth Centre has shown, the number of SOAs has dramatically increased since the mid-2010s (Rauschendorfer & Twum 2020). The report notes that SOAs have been used increasingly by EAC states to protect or support domestic production. In Kenya, Tanzania and Uganda's case, SOAs have primarily been used to increase tariff rates on imported products. Meanwhile, Rwanda has largely used them to decrease tariff rates. However, as the report explicitly notes, this has not been with the intention of lowering the price of consumption goods, but to decrease the cost of intermediary inputs to support domestic producers. Despite being intended as a short-term measure, the report also notes that once implemented, SOAs are usually maintained in subsequent years, undermining the integrity and ‘commonality’ of the regional CET.

As outlined later in this paper, the emergence of these local production and industrialisation strategies reflected a shift in development priorities among governments in the region. What is worth noting here, however, is the consequence of these strategies and how they have come to conflict with the EAC's regional integration and cooperation agenda. For instance, as part of these strategies, governments in the region have indicated their intention to give preferential treatment to goods and services produced nationally. The Kenyan government has even committed 40% of its public procurement budget to purchasing local goods and services (Government of Kenya 2017: 29). But, by championing national goods and services, governments in the region are coming into direct conflict with the non-discrimination principles which lie at the heart of both the EAC customs union (EAC 2004a) and common market (EAC 2010) – i.e. that all goods and services originating in the region are treated equally. More specifically, they stand in direct conflict with Article 35 of the EAC's common market protocol (EAC 2010: 34) which notes that, in matters of public procurement, governments will ‘not discriminate against suppliers, products or services originating from other Partner States’. Furthermore, increasing national divergences from the regional CET means that the consistency of the EAC's customs union has been increasingly undermined and uncertainty deepened with regards to intra-regional trade and whether goods are eligible for duty-free treatment within the region.⁴

One possible explanation for this state of affairs lies with rationalist accounts which emphasise regionalism as a political survival strategy for African elites (Herbst 2007; Gibb 2009; Söderbaum 2004). These accounts assert that African leaders will make rhetorical commitments to the aims and goals of regional organisations, but will stop short of substantively implementing the actual content of regional agreements that may undermine their domestic authority. The premise of these accounts is that ‘weak’ African states will only performatively engage with regional organisations as a way to ‘boost’ their sovereignty and external legitimacy. From this perspective, therefore, it could be asserted that governments in the EAC have implemented national economic strategies which conflict with their regional commitments because they had no intention of abiding by these commitments in the first place.

The limitation of these accounts, however, is that they presuppose that the motivations of political actors towards regionalism are solely reducible to an ambition to bolster domestic sovereignty and external legitimacy. However, if African regionalism simply serves as a performative façade to bolster domestic sovereignty, it becomes difficult to explain instances where regional agreements have been both agreed to and substantively implemented. For instance, whilst the EAC customs union has been somewhat undermined since the mid-2010s by internal trade barriers (Mugisha 2019) and SOAs (Rauschendorfer & Twum 2020), it is still significant that EAC members committed to and implemented a common trade regime in the first place. What this suggests is that we need to consider the motivations and preferences of actors in their own right, rather than viewing them as fixed and immutable. As Death (2018: 426) notes, it is possible for political actors in Africa to be motivated by various worldviews and ideological projects that extend beyond a narrow quest for political power. Even if we accept that many African states – to greater and lesser extents – are defined by weak internal sovereignty, this does not mean that the motivations of political actors can be automatically inferred from this condition. While some political actors may view regionalism as a threat to their domestic authority, others might perceive it as a way to promote development and political stability and, therefore, their own political survival. The key point is that the motivations of African political actors towards regionalism should be viewed as contingent and subjective, rather than fixed and universal, and studied as such (see Murray-Evans 2015).

One further explanation worth considering here are structuralist accounts which highlight the challenges of implementing market integration in Africa (Gibb 2009; Odijie 2019, 2020). The assumption underpinning these accounts is that, for the most part, African economies remain undiversified and dominated by primary sectors, such as agriculture and mining. The consequence of this is that it makes market integration difficult because, firstly, African countries have very little to trade with each other and, secondly, African governments tend to protect nascent and strategic economic sectors from regional competition (Odijie 2019, 2020). Similar arguments have been advanced in other developing regions of the world. For instance, Chandra (2016) notes in the case of the ASEAN Economic Community that rising protectionism and economic nationalism among its members is rooted in a similar desire to protect strategic economic sectors. Such accounts certainly offer important insights, including the relatively low volumes of intra-regional trade in Africa (UNCTAD 2019) and the trade-offs African policymakers might face between domestic industrial policy and regional integration. However, they also remain incomplete in other aspects. Notably, the tensions between economic nationalism and regionalism in the EAC, and elsewhere in Africa (Odijie 2019), have been a fairly recent phenomenon. In other words, why are these tensions more present at certain moments in time and not others? This suggests that these tensions are not entirely reducible to the underlying structure of African economies, but that there are intermediary political dynamics and subjectivities shaping the timing of their emergence.

SOCIAL PURPOSE AND THE DISCURSIVE CONSTRUCTION OF REGIONAL REGIMES

In this section, I develop a constructivist account which captures more fully the discursive drivers of African regionalism. Within the field of comparative regionalism, the notion that regions are socially constructed, rather than being determined by material geographies, is now an uncontroversial proposition. There are, however, differences in regard to the analytic focus of these literatures. For instance, some have focused on the links between regional identity formation and the construction of regional space (Hettne & Söderbaum 2000). The approach taken in this paper, however, sits closer to those literatures that have explored the role of discourse in the imagination and institutionalisation of regional space (see Van Langenhove 2011: Ch. 3). In a study of the European Union (EU), for instance, Rosamond (2002) has shown how EU policy actors’ discursive appeals to external economic threats helped construct the idea of a European economy as a ‘self-evident’ economic space. More recently, Mumford (2021) has emphasised how discursive appeals to Pan-African norms served as a strategic device for proponents wanting to empower the regional parliament of the Economic Community of West African States (ECOWAS).

In effect, these studies examine how actors discursively appeal to certain norms or perceived threats in order to persuade or even coerce others towards their vision of regionalism. While not ignoring the role of discourse in this more strategic sense, in this article I emphasise more its constitutive dimension – i.e. the role of discourse in constituting the role, function and purpose of regionalism. In other words, I am less interested in the argumentative dimension of discourse and more in its role in shaping the meaning of particular issues or events and setting boundaries for legitimate political action (Diez 2014). Analytically, I link to debates within the international studies literature concerning the legitimacy and social purpose of international regimes. The origins of this literature can be traced to Ruggie's (1982) seminal article in the journal International Organization, which explored the origins and institutional makeup of the post-1945 global economic order. Ruggie's key conceptual contribution was to highlight that the strength and stability of this order was not entirely reducible to the structural power of the United States, but that it also drew strength from an underlying social purpose that served to give it meaning and legitimacy (Best 2003). While Ruggie was vague on the exact meaning of social purpose, others have since defined it more specifically. As Baker notes, social purpose can be understood as an assemblage of ‘analytical and normative reasoning communicated through a public discourse, that builds a widely shared intersubjective consensus … concerning the desirability and benefits of a given system’ (Baker 2018: 296). The key insight that this concept of social purpose brings forth, therefore, is the notion that institutionalised systems of governance (international, regional or national) are not self-evident, but underpinned by particular discourses that serve to mobilise and converge actors’ expectations around their legitimate function, purpose and necessity.

In this article, I emphasize on how the social purpose of institutional regimes are articulated in relation to external economic threats. Put differently, I am interested in how institutional regimes draw legitimacy and meaning from their perceived ability to mitigate external economic threats. In doing so, I link to a constructivist literature which has explored the discursive construction of external economic threats (Hay and Rosamond 2002; Watson & Hay 2003; Siles-Brügge 2011). The point of departure for this literature is the so-called ‘globalisation debates’ which occurred during the 1990s (for overview, see Hobson & Ramesh 2002). The essence of these debates was the degree to which globalisation – defined by declining trade barriers and loosening capital controls – had rendered the economic role of the state obsolete. For the constructivist literature cited above, however, the degree to which globalisation was constraining the domestic policy space was less important than the degree to which actors had acted in the belief that it had done so. For these writers, the actual constraints imposed by such a complex process as globalisation was difficult (if not impossible) to verify empirically. In political terms, therefore, what mattered more were the discourses that political actors internalised about globalisation, rather than its material actuality.

I further contribute to these debates around social purpose by identifying an important temporal dimension to its logic. This aspect builds upon two key features of constructivist social theory (see Hay 2016): namely, that reality is socially constructed (and, therefore, susceptible to change) and that actors are reflexive and will continually reflect upon their social context and its perceived constraints and opportunities. Accordingly, social purpose can be understood as a discursive vision that resonates with actors’ interpretation of their external environment at a particular moment in time. Over time, however, it is possible for actors’ understanding of their context to diverge from the one presented within such discourses. In these contexts, new possibilities open for actors to imagine and pursue alternative policy practices and agendas. Crucially, however, these policy practices can (often unintentionally) come into conflict with the norms and practices of an institutional system, resulting in its social purpose becoming increasingly contested. In the remainder of this article, I use these theoretical insights to explain the rise of economic nationalism within the EAC.

TRACING THE SOCIAL PURPOSE OF THE EAC: GLOBALISATION, NON-NEGOTIABILITY AND THE CONSTRUCTION OF REGIONAL SPACE

The history of regional integration and cooperation between Kenya, Tanzania and Uganda dates back to the early 20th century, when the three countries were British colonies (Ingham 1967; Nye 1966) and following their independence in the early 1960s, there were even discussions about them uniting under a political federation (Vaughan 2019). Ultimately, political unity did not emerge and regionalism did not survive long after independence, with the first iteration of the EAC having effectively collapsed by 1977 (Hazlewood 1979). By the 1990s, however, discussions once again turned in East Africa to the question of regional cooperation. In 1993, the three states established the Permanent Tripartite Commission for Cooperation (PTCC), a loose inter-governmental forum aimed at facilitating cooperation in areas such as infrastructure and fiscal policy. These discussions intensified over the 1990s, culminating in the 1999 ‘Treaty for the Establishment of the East African Community’ which laid the foundations for the EAC's revival in 2000 (Mangachi 2011). This treaty laid out an ambitious cooperation agenda that called for the phased implementation of a regional customs union, common market and monetary union. The treaty also invoked the ideals of the early 1960s, committing the three states, at least in principle, to political unity and the eventual creation of a regional federation.

Why did policy actors in East Africa come to see it as necessary to re-establish the EAC at this time? To understand this, it is important to analyse the broader discursive environment in which the EAC was revived. Whilst much of the EAC's agenda, as outlined within its founding treaty, was concerned with economic integration, it is important to note that there was also a professed political dimension underpinning its revival which appealed to the region's shared history and common bonds. For instance, in a 1996 media interview with the Financial Times, the then Secretary-General of East African Cooperation, Francis Muthaura, noted the following:

We have many other regional organisations in this part of the world – Comesa, SADC etc. The point is not to have a repeat of the existing institutions. The East African region is unique – we're talking about three countries which for a long time were managed as one federal state which more or less speak the same language, whose citizens went to the same schools. These countries feel they have to be united. Those kind of arrangements cannot include very many countries without losing their focus. (Wrong & Goldman 1996)

Moreover, in a speech delivered in 1996, Kenyan President Daniel Arap Moi invoked a similar exceptionalism when he noted that ‘We are here because we believe that the destinies of our three countries are strongly linked. This is because of our common history, our common culture and indeed our common aspiration’ (EAC 2002a: 8).

Although difficult to empirically verify the sincerity of such professed ideals of shared political community, it was certainly the case that they were somewhat detached from the realpolitik on the ground. While relations between Kenya, Tanzania and Uganda had improved from their low point following the collapse of the first EAC in 1977, it was clear that officials in the three governments were still cautious about the consequences of integration. One of the key reasons behind the collapse of the first EAC had been the perception that the benefits of regional integration were distributed unevenly and primarily to the benefit of Kenya, the most developed economy in the region (Mold 2015: 582). In the run-up to the EAC's revival, these underlying concerns were still present. For instance, in 1999 the Confederation of Tanzania Industries (CTI) released a report criticising proposals that would have seen the removal of all internal tariffs within the EAC by 2000 (CTI 1999). The report argued that the rapid removal of intra-regional tariffs had the potential to place Tanzanian industries under intense competitive pressures from within the region and, in particular, from Kenya. Moreover, that the EAC's treaty of establishment stressed the need for cooperation to be based upon the ‘equitable sharing of benefits’ is further testament to the fact that concerns around the distributive consequences of the regional integration were still present.

Given these concerns, and considering reports at the time which spoke of the frosty relations between the leaders of Kenya, Tanzania and Uganda,⁵ discursive appeals to supposed political bonds alone would arguably have been insufficient to mobilise key actors around the EAC's ambitious integration agenda. Instead, what appeared to be a more significant factor were threats of economic marginalisation, driven by a perception that globalisation was unleashing intense competitive pressures on the region's economies. An examination of documentary sources from the late 1990s and early 2000s reveals several key features regarding how globalisation was being discursively framed and understood by key policy actors within the region.

In the late 1990s and early 2000s, the EAC Secretariat organised a series of workshops that brought together relevant ministers, senior bureaucrats, donors and consultants from across East Africa to deliberate on the region's proposed integration and cooperation agenda (EAC 2000a, 2000b, 2000c, 2001). The proceedings of these workshops, therefore, offer a glimpse into internal policy discourses that were underpinning the EAC's revival, with the looming spectre of globalisation being a central feature of these discussions. As Table I shows, within these workshops there were numerous references to the non-negotiable nature of globalisation and the economic constraints and threats it posed to East Africa's economy. Participants spoke of globalisation as an exogenous and unavoidable process that was increasingly subordinating national markets in the region to global competitive disciplines. Furthermore, not only was globalisation seen to be placing economic constraints on the region, there also seemed to be a consensus that it was also an environment of limited policy manoeuvrability, where economic governance needed to conform with market logics. In this context, participants spoke of the need for national governments to avoid previously practiced development policies, such as state-led import-substitution and industrial policy, which were no longer seen as consistent with this new global economic order. This reflected a broader trend in the region since the 1990s, where the Kenyan, Tanzanian and Ugandan governments had – whether out of conviction or lack of alternatives – come to embrace market-conforming economic governance (Harrison 2001; Murunga 2007).

Table I. Thematic references to globalisation within EAC workshop proceedings.

Within this discursive context, regional integration and cooperation in effect came to be cast as a necessary and non-negotiable response to the perceived economic constraints being imposed by globalisation. As Table I highlights, participants at the EAC workshop series spoke of the need for regional unity and cooperation to withstand the pressures of globalisation. Such sentiments were also reflected in the public realm. At the signing of the EAC Treaty in November 1999, Tanzanian president Benjamin Mkapa noted in his speech that:

It is true our economies are not exactly on par with each other. But the truth is that none of us can individually be a significant player in the competitive environment ahead of us. There are multinational corporations … with much larger economic muscles than the economies of all our three countries combined … The only assured key to our survival is to integrate our economies and markets. (EAC 2002a: 9, 41)

Indeed, for East Africa's policy and epistemic community, the attraction of regional integration was that it would seemingly improve the region's economic competitiveness – via intra-regional competition and economies-of-scale – while remaining consistent with the market-conforming logic that the process of globalisation was perceived to demand.

This depiction of the global economy, as one of increasing and unavoidable competitive pressures, also served as an important discursive resource that served to converge regional stakeholders’ expectations around the necessity of the EAC's agenda. As pointed out earlier, while there were still hesitations within the region about the political and economic consequences of regional integration, it appeared as though the threat of economic marginalisation posed by globalisation eclipsed these concerns. This is reflected in the foregoing extract from President Mkapa's 1999 speech, which spoke of the need for regional integration even if the three economies were not exactly ‘on par’ with each other. It is also reflected in the degree to which sceptics (as well as proponents) of regional integration appeared to internalise this logic of non-negotiability. For instance, the same CTI report from 1999 that criticised proposals for the removal of intra-EAC tariffs by 2000, also argued that a phased process of regional integration would be needed for Tanzanian industries to become globally competitive (CTI 1999: 7). In short, CTI did not reject the necessity of regional integration, but merely questioned the planned pace of its implementation.

Following the EAC's launch in 2000, appeals to the threat of globalisation and the role of regional integration in mitigating these continued to underpin EAC policy discourse. In January 2000, negotiations began between the three EAC members to establish a regional customs union. During these negotiations we again see key figures and policy documents invoke the threat of globalisation as an imperative for finalising and implementing a customs union agreement. For instance, at an EAC heads of state meeting in 2002, Kenyan President Daniel Arap Moi stressed in a speech the need to maintain momentum towards the creation of a single trade and investment area to ‘withstand the adverse effects of globalisation’ (EAC 2002b: 2). That year also, the EAC Secretariat released an occasional paper entitled ‘East African Customs Union: Information and Implications’ (EAC 2002a). This paper reiterated narratives around the relative un-competitiveness of EAC members’ economies and emphasised that to avoid marginalisation within the world economy, it was necessary to integrate markets and foster intra-regional competition to improve the productivity and international competitiveness of firms operating within the region (EAC 2002a: 7).

Moreover, in 2004 the EAC heads of state also established the Committee on Fast Tracking East African Federation and tasked it to explore the possibility of ‘fast tracking’ the end goal of the region's integration agenda – the creation of single political federation. Published in November 2004 (though not publicly), the committee's report recommended that the EAC's integration timeframe be condensed allowing for a political federation between the three states to be established no later than 2013 (EAC 2004b). Given that the implementation of the EAC's common market and monetary union are still incomplete today, this timeframe was always highly unrealistic. However, the significance of this report lay not necessarily with these proposals, but in how it sought to recast the EAC's ambitions for political integration as an economic necessity. As the report noted:

Globalization dictates that poor and politically weak states come together in order to withstand its impact and to be able to gain from it … In order to reverse marginalization, we must, on a priority basis, work on effective and timely responses to globalization … The decision to fast-track the integration process towards East African Federation is therefore a pragmatic move, necessary for survival. (EAC 2004b: 14–15)

In other words, political unity was not framed around affective rhetoric and symbolism, which others have noted are essential to the imagination of political communities (Anderson 1983), but simply as a ‘pragmatic move, necessary for survival’.

The summative point to note here is that by the early 2000s, the social purpose of the EAC's regional regime had come to converge around a discourse of globalisation as a non-negotiable economic constraint. That is to say, the EAC's integration agenda drew its purpose and legitimacy from its perceived ability to mitigate the seemingly intense competitive threats posed by globalisation. As shown, this discourse served to construct a powerful imperative for regionalism among East Africa's policy community and other regional stakeholders (proponents and sceptics alike) by casting other national development policy alternatives (e.g. import-substitution, industrial policy) as unviable in the new era of economic globalisation. As we shall see next, however, the underlying dilemma facing the EAC was that by legitimating regional integration solely through the spectre of external economic threats, the salience of those same threats might come to weaken over time.

CONTESTING NON-NEGOTIABILITY: THE EAC AND THE ‘NEW DEVELOPMENTALISM’

Throughout the latter part of the 2000s, EAC policy discourses continued to invoke the spectre of globalisation as a central justification for regional integration and cooperation. As the third EAC development strategy released in 2006 noted, ‘the economic challenges facing EAC include the process and effects of globalisation … EAC Partner States will continue to view regional integration arrangements as giving them a range of opportunities to liberalise their economies and avoid marginalisation’ (EAC 2006: 41). That year also, the then EAC Secretary-General Juma Mwapachu noted in a speech that ‘nation states along cannot realize the economies of scale and scope … to be able to survive the acute economic pressures unleashed by globalization’ (Mwapachu 2012: 122).

To reiterate, two key axioms underpinned interpretations of globalisation among East Africa's policy community in the late 1990s and early 2000s. The first was an understanding of the global economy as one of growing competitive pressures brought about by global trade liberalisation and capital mobility. In essence, globalisation was viewed as a rapidly unfolding process which threatened to marginalise East Africa within the world economy. The second axiom was that globalisation was also perceived in the context of limited policy manoeuvrability. In an increasingly open and outward looking global economy, East Africa's policy community effectively saw it as no longer feasible to pursue the more inward development strategies which had defined previous eras, such as state-led import-substitution and industrial policy. Rather, it was perceived necessary for economic governance to be market-driven and private-sector led. Indeed, one of the attractions of regional integration was that it would promote economic competitiveness while remaining consistent with these free market dictates.

Moving into the 2010s, policy officials and regional stakeholders continued to hold onto the view of the world economy as one of intense competitiveness. For instance, in interviews conducted with policy officials and business representatives in 2017, many spoke of the growing threats posed to regional producers from emerging markets such as China, India and other Asian economies that had risen to prominence during the 2000s.⁶ In 2018, the Kenyan government also vetoed ongoing negotiations for an EAC–China free trade agreement to protect domestic firms from being overrun by Chinese producers (Otieno 2018). According to an official from the EAC Secretariat in 2017, these negotiations had been at a very preliminary stage.⁷ Even so, the fact that the Kenyan government was not even willing to engage in these talks speaks to growing fears in the region around China's rise and its domestic economic consequences. In short, external economic threats and competitive pressures (perceived or otherwise) continued to feature as strongly in the outlooks of policymakers and other regional stakeholders as they did when the EAC was re-established in the early 2000s.

In other respects, however, there were signs that the second axiom of this discourse of non-negotiability was beginning to loosen among actors within the region. Notably, by the turn of the decade, development priorities in the region had come to shift from an earlier focus on macroeconomic stability, poverty reduction and good governance, typical in the late 1990s and early 2000s, to new national agendas geared towards structural transformation, industrialisation and modernisation. Within this emerging agenda, there was also a reconsideration of the role of the state in guiding and shaping development outcomes. In 2008, the Kenyan government released its ‘Vision 2030’ economic strategy giving prioritisation to six strategic industrial sectors for investment, reflecting a greater acknowledgement within the Kenyan government that an official industrial policy was needed (Fourie 2014: 550). Meanwhile in 2010, the Ugandan government released its first ‘National Development Plan’ with a prioritisation of productive sectors (e.g. manufacturing, agriculture) over so-called social sectors (e.g. health, education) which had been the focus of its earlier development strategies (Hickey 2012). In Tanzania, successive governments have sought to re-assert government authority and ownership of its extractive industries over the last decade (Jacob & Pederson 2018) coupled with a discursive shift towards a national project of industrialisation (Paget 2020). In Rwanda, which joined the EAC in 2007, the previously circumscribed practice of import-substitution (or ‘domestic market recapturing’) was tentatively revived in 2011 to increase local production capacities (Behuria 2019).

Although this ‘new developmentalism’ in East Africa did not mark a return to the central planning policies typical in the region during the 1960s and 1970s, it did reflect a reorientation towards a more state-driven approach to industrial development. As pointed out earlier in this article, since the mid-2010s, East African governments have intervened domestically by unilaterally imposing higher tariffs and implementing local procurement policies to protect and nurture domestic producers. Whilst earlier regional discourses saw the role of government as strictly limited to supporting the ‘maintenance of law and order … and [the] expansion of market mechanisms’ (PTCC 1997: 11), this new developmentalism offered an expanded understanding of state intervention. Several factors appeared to underpin this discursive shift. Notably, the rise of China and other emerging markets called into question earlier presumptions that, in an era of globalisation, the economic role of the state should be limited to law and order and enabling markets (Heron 2016). As is well documented, China's industrial development rested on a close relationship between the state and the market. It was also clear that the economic liberalisation measures implemented across Africa during the 1990s failed to deliver expected structural transformation and economic diversification, leading African governments to tentatively begin looking at policy alternatives (Harrison 2019).

At the regional level, this discursive shift became apparent in 2012, with the release of the EAC's regional industrial strategy (EAC 2012a) and industrial policy (EAC 2012b). As an extract from the EAC's industrial policy notes:

Without underestimating the need for investments in infrastructure and private sector development, the presence of concrete policy reflects the acceptance of a larger governmental role in the promotion of productive restructuring … The implicit assumption is that the industrialisation process is essential for the transformation of the economy as a whole and that it is possible to influence this process through strategic, targeted interventions. (EAC 2012b: 4, 6)

In contrast to earlier regional policy schemas, which solely valorised market integration as a path towards development and economic diversification, the importance of these two documents was that they acknowledged the need also for government interventions in the market to guide this process. Significantly, however, neither document clearly articulated a role for the EAC within this ‘new developmentalism’, beyond facilitating expanded regional markets for producers and industrial policy coordination between its members. Although attempts were made during the 2010s to construct bespoke regional strategies for promoting sectors such as textiles and clothing and pharmaceuticals (Murray-Evans & O'Reilly 2022; O'Reilly & Heron 2022), these ultimately fell short of expectations, due to policy coordination issues and unilateralism by EAC members. Instead, EAC members opted for more economically nationalist visions of development which, as outlined earlier in this paper, have increasingly come to conflict with the principles and practices of regional integration.

At this point, it is worth returning to the question of why this situation has emerged. In other words, given that the process of regional integration was once conceived as essential to the region's ‘survival’ within the global economy, why is it that national governments in the EAC have increasingly come to pursue policies which conflict with that same process? The answer, I contend, lies in certain institutional and discursive tensions which have gradually emerged within the EAC's regional regime.

The first point here relates to the institutional design of the EAC. When re-established in 2000, the EAC was imagined and defined as a space that would support market-led development. Regionalism was, therefore, institutionalised around a process geared towards the removal of intra-regional market barriers. As a result, the architects of the EAC did not deem it necessary to construct a large regional bureaucracy to oversee this process. Instead, regional decisions and policies were to be domesticated through national ministries. For instance, in a 1996 media interview, the then Secretary-General of East African Cooperation, Francis Muthaura, noted that the EAC Secretariat would deliberately be ‘small, but highly professional’ (Wrong 1996). In other words, the EAC was premised on a ‘negative’ form of integration, that prioritised the removal of barriers to trade and investment, over more ‘positive’ forms of integration aimed at constructing dense regional institutional structures (see Scharpf 1998). The EAC, therefore, lacked the institutional capacities and centralised authority necessary to coordinate a regional industrial strategy. This was particularly evident with the implementation of the EAC's regional pharmaceutical production strategy since 2012, with stakeholders remarking on the limited capacity of the EAC Secretariat and regional institutions to coordinate such an ambitious and complex agenda (Murray-Evans & O'Reilly 2022).

The second point relates to the discursive foundations and social purpose of the EAC's regional regime. To borrow once again from the terminology of Scharpf (1998), not only was the EAC institutionally premised on a negative form of integration, but it was also discursively articulated as such. As we saw in the previous section, political actors and other stakeholders converged around the EAC's integration agenda in the late 1990s and early 2000s because they saw it as a necessary response to the competitive pressures perceived to be unleashed by globalisation. The imperative towards market integration was increased further by the perception that previously practiced national economic policies (i.e. import substitution, industrial policy etc.) were no longer feasible in an era of globalisation. The EAC's social purpose was, therefore, articulated in a negative sense, in terms of its ability to mitigate the impact of globalisation on its members’ national economies. In other words, policy discourses did not attempt to construct a ‘positive’ case for integration, based on collective ideas of development and common political community. Indeed, as we saw earlier, when debates turned in 2004 to discussions of fast-tracking political integration, this was framed simply as a ‘pragmatic’ necessity in light of the economising logic of globalisation. In short, EAC policy discourses did not attempt to overcome what former EAC Secretary General, Francis Muthaura, once described as the ‘entrenched’ notions of ‘national sovereignty’ that existed within the region (EAC 2001: 53–54).

Accordingly, as the discursive glue which held the EAC together (i.e. the non-negotiability of globalisation) came to weaken, it opened space for policy alternatives to emerge in the form of more statist development agendas. Therefore, the challenge in coordinating these strategies together not only derived from the institutional features of the EAC, but also arguably from the fact that development was always (and continues to be) imagined as a distinctly state-centric and nationalist endeavour (Mold 2015). In effect, the initial imperative for the EAC's integration agenda rested on the perception that no other development alternatives existed. As this discursive imperative has weakened, it has called into question how the EAC continues to frame its social purpose in this context. Although there is no sign that national governments are questioning their membership of the EAC, their increasing pursuit of economically nationalist policy measures suggests, at the very least, that regional integration is no longer as central in their vision of economic development. The dilemma, therefore, for proponents of EAC integration and cooperation is how they articulate the continued purpose and necessity of this agenda in a context where governments in the region have come to reorient their economic and development policy towards more state-oriented strategies. As recent contributions have argued, this may require a reimagining of African regionalism beyond the paradigm of market integration (Rudahindwa & van Huellen 2021).

CONCLUSION

Over the previous decade, growing tensions have been observed between long-standing ambitions for regional integration in Africa and national trade and industrial strategies being pursued by governments across the continent. This article has sought to account for these tensions by narrowing its focus onto the EAC – an REC where economic nationalism has become increasingly prevalent since the mid-2010s. In doing so, it offers a counterpoint to rationalist and structuralist accounts and instead emphasised the importance of discursive subjectivities in shaping the dynamics of regionalism in Africa. In short, the paper argues that regional institutions in Africa (and elsewhere) draw their legitimacy from an underlying social purpose that mobilises and converges actors around their function, purpose and necessity.

In the EAC's case, following its revival in 2000, its social purpose centred upon a discourse of globalisation as a non-negotiable economic constraint. This discourse served to construct a powerful imperative for regionalism by casting integration as a necessary and indispensable response to emerging external economic threats and by casting other national policy alternatives as unviable. The article went on to chart, however, how this discourse of non-negotiability came to weaken within the EAC, opening space for both the articulation and practice of previously circumscribed developmentalist policies, such as industrial policy and import substitution. While some efforts were made to coordinate these at the regional level, it was primarily within the national sphere where these initiatives emerged and, subsequently, came to conflict with the EAC's integration agenda. As argued, this situation stemmed in part from the limited capacity of the EAC's existing institutional structures to coordinate regional trade and industrial strategies, but also from the inability of EAC policy discourses to articulate the indispensable relevance of its regional integration agenda within this emerging context of statist trade and industrial strategies. In other words, EAC policy discourses can no longer – as they once did – cast regional integration as the only viable pathway to promote economic development when national governments are openly espousing and practising industrial policies and import-substitution.

The findings of this article contribute to the academic literature in at least two ways. First, as discussed in the previous paragraph, it emphasises the importance of discursive subjectivities in shaping the dynamics of regionalism in Africa (see also Mumford 2021). This contradicts rationalist and structuralist accounts which largely emphasise elite political interests and economic structures in explaining the limited success of regional integration projects in Africa. As this article has emphasised, under certain discursive contexts African actors can converge around and commit to the aims of regional integration. Second, this article further contributes to the theorisation of regional integration. Existing scholarship has emphasised how discursive imperatives for regional integration can be constructed through appeals to external economic threats (see Rosamond 1999, 2002). This article builds upon these insights by highlighting how such discursive imperatives can wax and wane over time and only remain salient to the extent that they can cast regional integration as a necessary and non-negotiable measure to respond to a given economic context.

Beyond these contributions to the academic literature, the findings of this article also have broader practical implications for understanding the drivers of African regionalism. Like the EAC, most other RECs in Africa were created or revived during the 1990s around a market-oriented (or ‘neoliberal’) economic integration agenda (Söderbaum 2004). Yet, as the neoliberal consensus, which has dominated African economic governance since the 1980s, shows some tentative signs of receding (Hickey 2012; Harrison 2019), it calls into question how African RECs continue to frame their purpose and relevance. This is particularly important as policymakers across Africa turn towards the implementation of the AfCFTA, following its signing in 2018. As the case of the EAC has shown, the success or failure of this ambitious initiative will rest in part on the ability of its proponents to not only articulate its relevance – but also its necessity – for addressing the various external challenges the continent faces.