Arguing a Political Economy Approach to Regional Integration

1

1. Introduction1.1 Motivation and purposeThere is long-standing, wide agreement on the need for greater economic connectedness in Africa. This is intended to “de-fragment” Africa (Brenton and Isik, 2012), thus bringing about economic transformation through expanded markets, investment and employment creation, and therefore inclusive growth and poverty reduction. Regional markets may further offer opportunities as a testing and learning ground for exports due to similar tastes and lower barriers to market access in the African region (Hoppe and Aidoo, 2012). Despite the strong rhetorical and political consensus across the continent in support of regional economic integration there is slow progress on implementing commitments to free movement of goods, services, people and capital between national markets.

There is also general accordance on the two most important implementation challenges: lack of political will and lack of capacity. Yet, there is less consensus on how these twin deficiencies come about, on which political actors and factors influence regional integration, on how effective regional integration looks like, and on the different institutions underpinning it. Rather than presenting a trajectory to an “ideal” or “best practice” regional integration model, this paper proposes an analytical approach that prioritises a better understanding of the political economy (PE) context in which the regional integration agenda is being pursued, either through the formal regional bodies or through the multitude of cross-border cooperation or “regionalism”.

This paper presents five PE lenses and introduces a skeleton PE approach for systematising knowledge development about the drivers and obstacles to regional integration. This should contribute to efforts by policy makers and other actors in the field to better identify opportunities for realistic or effective support strategies to regional integration and potentially inclusive growth.

In presenting an analytical framework to analyse regional integration through a political economy lens, the paper highlights the need to be clear about the purpose(s) of analysis. Regional integration involves a complex of inter-state processes, different levels of aspiration and ambition, a myriad of in-country measures, projects and processes ranging from formal trade agreements, formal and informal trade, regulatory mechanisms, macroeconomic policies, industrial and agricultural policies, and many more. Therefore, it will be important, to be spell out clearly what the unit of analysis is within the regional processes being examined.

The paper does not discriminate between the formal regional integration agenda and processes as embodied by the mandated regional institutions. It argues the importance of applying a political economy analysis to existing cross-border initiatives that may become central to the gradual build-up of more inclusive regional integration dynamics in areas such as infrastructure investments, the movement of goods, people and finance. Ultimately, regional processes are most likely to meet with successful implementation where they support national priorities, as revealed through actions rather than policies, and according to the interests of influential domestic actors and their interaction with ruling elites.

1.2 Regional integration contextThe blueprint for regional integration is commonly taken from Balassa’s seminal book of 1961. This describes five forms of economic integration: a Free Trade Area (FTA), a Customs Union (CU), a Common Market (CM), an Economic Union and complete economic integration. Although frequently understood as a linear sequence of deepening integration towards a regional ideal, as Baldwin (2011) points out, these were never intended to represent an ideal sequence but instead to describe different possible levels of integration. Nonetheless, most African regions have adopted this sequential model, with varying degrees of implementation, as shown in Table 1.

2

Table 1: Status of integration1

Notes: *achieved (green) *envisaged (blue) * not planned (grey)Source: AfDB (2013).

While this represents some progress in complying with formal commitments, progress can perhaps best be understood in terms of regional trade and ultimately, integration into the world economy. While the combined exports of all eight RECs grew at 5.2 percent over the period from 2007 to 2011, above average world export growth (4.8 percent), intra-African exports grew at only 3.2 percent over the same period. As such, while these more than doubled between 2005 and 2006 (AfDB, 2012), the average share of intra-African exports in total merchandise exports therefore remains low, estimated at 11 percent, compared with 50 percent in Asia, 21 percent in Latin America and the Caribbean and 70 percent in Europe (UNCTAD, 2013).

Further, while these averages hide considerable variation and trends in intra-Africa trade across RECs, half of intra-African trade takes place within the SADC region, between South Africa and its neighbours (AfDB, 2012). Indeed, 26 countries counted South Africa among their five main export destinations (UNCTAD, 2013) suggesting a considerable level of concentration in intra-African trade. Although these figures clearly exclude what may be vibrant informal trade, the implication is that regional trade in Africa has substantial room to grow.

1.3 Structure of the paperThe remainder of the present paper is organised as follows. Section 2 provides an overview of the literature on political economy analysis and regional integration. On the basis of that literature and the issues raised, Section 3 provides an analytical framework for a political economy approach before this is applied to aspects of regional integration through a number of cases in Section 4. Section 5 concludes and provides some indicative recommendations.

2. Political economy, development & regional integration1 The African Union recognises eight regional economic communities (RECs) in Africa. These are the Arab Maghreb Union (AMU), the Community of Sahel-Saharan States (CEN-SAD), the Common Market for Eastern and Southern Africa (COMESA), the East African Community (EAC), the Economic Community for Central African States (ECCAS), the Economic Community of West African States (ECOWAS), the Intergovernmental Authority on Development (IGAD) and the Southern African Development Community (SADC).

3

Since the 1990s there has been a surge in efforts to understand governance, institutions, and policies for inclusive and sustainable development. This growing literature underlines several useful concepts for understanding the success or failure of policy reforms. More fundamentally, they point to the fact that beyond economic considerations, policy reforms take place in a context of complex, long-term, and non-linear processes that are determined by the unique history and institutions, and the economic and political economy underpinnings of a country. Political economy analysis of policies to promote regional integration therefore brings an additional level of complexity, requiring analyses of the dynamics between the various centres of power and influence. The following literatures provide some conceptual guidance on what has been analysed in this regard, and what can be used to develop a political economy framework and approach for regional integration.

2.1 The political economy of developmentThere is now broad consensus that a well performing economy rests on a “foundation of good governance” that includes transparent and predictable decision making, oversight mechanisms, accountability in how resources are used, committed public officials, a political process viewed as legitimate, and the protection of property rights, a regulatory apparatus curbing the worst forms of fraud, anticompetitive behavior, and moral hazard, a moderately cohesive society exhibiting trust and social cooperation; social and political institutions that mitigate risk and manage social conflicts (Levy, 2010; Rodrik, 2007: p. 153). Yet, building effective and accountable institutions in developing countries based on such ‘best-practice’ formal institutions has proved a major challenge with debate around the causal relations and linkages with development outcomes and trajectories of change.

By moving away from a view of development as a linear process towards some ideal state, attention has shifted towards ‘good enough governance’ (e.g. Grindle, 2011) or ‘best fit’ policies, rather than ‘best practice’ (e.g. Levy, 2011; Booth, 2011). As Rodrik (2008) points out, this idea of the need for “appropriate” institutions goes at least as far back as Gerschenkron (1962). This involves a shift from emphasis on formal institutions, the so-called formal “rules of the game”, to the interactions between formal and informal institutions. Informal institutions are those rules that are not enforced by formal agencies. These include habits, customs, cultures and values (e.g. North, 1991). Taking a ‘best fit’ approach implies that external actors understand the interplay between formal and informal institutions, and take that as the starting point for developing support strategies for institutional reforms.

Political survival is assumed to be the key motivation for ruling elites (often ruling coalitions), including their interventions to support the productive sectors (Geddes, 1994; Bueno de Mesquita et al., 2005; Moore and Schmitz, 2008; Leftwich, 2011; Kahn, 2010; Whitfield and Therkildsen, 2011). When governments face the choice between their own political survival and longer-term interests in stability and collective development, whether at the national or regional level, they tend to prioritise the former in ways that may not align with the latter, and may even undermine or compromise it. Furthermore, political survival in a democracy depends on being able to maintain a ruling coalition and win elections. This may include the use by ruling coalitions of patron–client relations - in exchange for favours or patronage, ruling coalitions expect electoral or other support. This creates strong incentives for ‘rent-seeking’, including from aid, to cement alliances for maintaining or gaining a hold on power. As such, while patronage may play an important role in maintaining political stability and social cohesion, it can also undermine productive investment (Unsworth and Williams, 2011: p.7). It is therefore also important to have an understanding of the degree of political competition (Khan, 2005), the process of state-society bargaining (IDS, 2010) and the distribution and management of economic rents (Khan, 2006).

2.2 Political economy of regional integration

4

While the above literature focuses on processes within nation-states, applying the PE lens to regional processes brings both parallels and additional complexity. These can be examined in terms of the negotiation process and implementation of agreed measures. While the negotiation and agreement of regional trade or other agreements represent a rules-based formal agreement depending on inter-state interests and balances of power, implementation will depend very much on how it affects the current political settlement, the distribution of economic rents and other factors referred to above.

Mansfield observes that “while it is frequently acknowledged that political factors shape regionalism, surprisingly few systematic attempts have been made to address exactly which ones most heavily influence why states choose to pursue regional trade strategies and the precise nature of their effects”. This observation remains valid for regional integration in Africa at large.

While political economy analyses attempt to move away from a linear view of development, a more political approach to regional integration may similarly require that policymakers reconsider the focus on the linear process along Balassa’s (1961) five integration categories. The rules-based nature of regional trade agreements is at risk of what Pritchett et al. (2010) call isomorphic mimicry - institutions that superficially take the form of those in functional states but do not fully play the necessary roles and functions given the lack of supporting institutions, accountability and enforcement mechanisms. Taking this into account would therefore imply focussing less on designing policies associated with the “next stage of integration”, than identifying policies to strengthen integration within the given political-economic context. This leads some such as Amani et al. (2013) to question “the appropriateness of the current approach to regional integration in Southern Africa”.

The additional complexity brought by moving from national to regional processes may helpfully be considered as a “two-level game” of regional diplomacy and domestic politics (Putnam, 1988). While “fruitless to debate whether domestic politics really determine international relations, or the reverse”, in his model, domestic groups at the national level pressure the government on policies in favour of their interests through formal or informal means, while politicians seek power by working with coalitions among those groups. At the regional level, governments seek to satisfy domestic pressures while minimising adverse foreign consequences. To add to this framework, regional agreements may also result from state-level peer pressure, or what Andrews (2013) calls the “signalling’ of a “willingness to modernise” by agreeing to the principles of regional integration.

This then points to the inherent tensions that arise in regional processes, with the two levels reflecting the dilemma of “giving up sovereignty” through greater integration versus maintaining greater national independence. Outcomes in the form of successful agreement and implementation will only occur where the demands of international agreements align with organised domestic interests, relating back to the literature discussed above on political settlement and competition.

Inter-state bargaining around regional integration may also relate to how regional objectives are framed or conceptualised. Hentz (2005) sets out three views of regional integration in Southern Africa in the post-apartheid period: i) regional developmental cooperation, ii) regional market integration, and iii) ad-hoc cooperation. Each has a different vision and has found support among different interest groups at different times. To illustrate, regional developmental cooperation is reflected in the initial model of SADCC and in ANC statements, while regional market integration is more in line with SACU and business interests that envisage the more linear path to integration that was adopted in the Lagos Treaty and ultimately adopted by SADC, while ad-hoc cooperation comprises large investment projects, such as the Lesotho Highlands Water project, the Cahora Bassa Dam, and Maputo Port, where interest groups have been more narrowly related to project specifics. How these different interest groups organise and pressure governments and elites, and how these national elites then interact at a regional level may then determine the model applied in practice and the ultimate form of implementation.

5

Draper (2010) argues that African regional integration has de facto relied on “European” intellectual foundations, which has limited its ability to address Africa’s challenges. In this light, Hartzenberg (2012) argues that the model to be adopted by the TFTA is much more reflective of Africa’s needs and realities. Encompassing infrastructure and industrialization issues, the TFTA may provide a welcome alternative to models relying solely on liberalization or on “linear” integration, addressing supply side issues typically not covered by models emphasizing the development of common markets.

It is therefore important to map out the interactions of these more international dynamics with national-level dynamics. Studies such as Carère et al (2012) rely on public choice or game theoretic constructions to conceive how agents interact at the national level around regional integration, while Flatters (2002) focuses on the domestic political economy of SADC countries to explain the relatively restrictive rules of origin on textile. In what may be an overly sceptical view, Soderbaum and Taylor (2008) suggest that “instead of investing in regional projects that promote broad-based development as a regional collective good, the ruling elites will much more likely seek to control what material benefits of state sovereignty they can muster to strengthen their own political authority, as well as to benefit personally, often financially. This helps explain in part why so many regional projects fail in Africa...what remains in such circumstances are often ad hoc arrangements, invariantly financed by the donor community”. Again, this points us to the need to understand both political processes at the national level, and how these interact at a regional level.

Increasingly, research is undertaken at sector level with a focus on the political economy dimensions that help explain the under-provision of key regional public goods. The provision of public goods and services in a regional context - in areas ranging from infrastructure, trade facilitation, management of common pool resources, peace and security etc. is difficult as it requires a high level of trust and cooperation, not just in one country but between two or more (Sandler, 2007; DfID, 2011). DfID (2011) also draws attention to a category of literature about less visible and less formal forms of regionalism, “regionalism as experienced”. Authors such as Trémolières (2007), Soderbaum and Taylor (2007; 2008) point to the relevance of uncovering such on-the-ground regional realities as informal practices and the long-lasting prevalence of informal intra-regional activities interact with formal processes of regional integration.

Underlying these analyses linking national and regional dynamics, a key element is the role of the private sector as potential driver and beneficiary of greater regional integration. Looking at the private sector role in regional integration in the SADC region, Bertelsmann-Scott et al. (2013) conclude that the fact that “they chose to cooperate at the regional level in policy debates and push for certain outcomes, as in the case of telecommunications and banking, shows that benefits are to be had from active participation”.

2.3 Donor applications of political economy analysesAs Barma et al. (2012) state, “understanding the political economy aspects of policy interventions can mean the difference between a successful intervention compatible with political incentives and “first-best” technical fix that falls flat”. Building on the growing academic and policy insights summarised above, an initially small number of donors recognised the importance of country level political economy analyses, primarily for the purpose of informing country programming or strategy development, problem solving or feeding policy dialogue with partner countries.2 The depth and scope of donor-driven political economy diagnostics has varied as the purposes for undertaking them also

2 Initially DfID (Drivers of Change), NL (Strategic Governance and Anticorruption Analysis), SIDA (drivers of change), with other donors joining including the EC, Norway, Germany, the World Bank, etc.

6

varied, with PE analyses being undertaken at country and sector levels, and increasingly also at the level of particular problems or policy processes that the donor sought to understand and address.

The analytical frameworks developed by donors all combine an interest in the interactions between i) structural factors, ii) formal and informal rules of the game or institutions, and iii) the actors operating in the current context. The OECD has developed another analytical tool that helps understand how the domestic political economy is influenced by regional and global drivers. So far, there has been little systematic research of how these donor tools have been applied and their impact. But from what little there is (including: Fisher and Marquette 2013, Desai 2011), the record is mixed. There is apparently little change from 2008, when Unsworth (2008: p. 2) noted that “political analysis seems to be having virtually no impact at a corporate level”. Challenges include the time and energy required to diagnose diversity and complexity; accepting that donor plans may not have political traction locally; accepting the challenge that an political economy analysis may oblige donors to rethink the way they position themselves or operate in particular contexts; and accepting the fact that donor expectations may be too ambitious (Unsworth, 2011). Still, political economy analyses have generated a repository of country and sector level insights, and further draw the attention to a number of challenges and opportunities for donors.

Based on this overview of the literature, the following section presents a conceptual framework that builds on the above, offering a useful framework for systematising information on the political economy of regional integration processes.

3. A political economy framework and approach Political economy analyses can be used at the project, sectoral, national or regional level with the goal of better understanding how policy makers might seek to enhance regional integration processes in Africa. As such, it is important to define the unit of analysis, which might range from the political economy of a regional trade agreement, to the implementation of a one-stop border post. Once this is clearly established, the framework presented here is intended to help uncover some of the key political-economic factors that may influence regional integration processes.

Whatever the domain of application, the paper proposes five building blocks or “lenses” through which to analyse or asses formal regional integration dynamics or cross-border dynamics. These include three basic analytical categories: structural factors, institutions or rules of the game, and actors/agents. The interactions between these categories shape the interests and behaviour of key actors and their relations. In order to complete the picture, two additional components are considered in the framework: one reflects regional and global drivers that are external to the object or process being examined through the political economy lens, while the other reflects sector-level characteristics of national and regional public goods, particularly those that matter for regional integration. It is suggested to develop a political economy approach with these five analytical building blocks.

3.1 Structures, institutions and actorsAs the literature review suggests, bringing a political economy understanding to regional integration processes must recognise that each country has a unique set of characteristics shaped by structural features (including its history, geography, natural resource endowment, etc), institutions, and actors

7

and that these will inform the way domestic and regional integration processes, policy reforms and cross-border cooperation take place.

1 Structural factorsStructural or “foundational” factors relate to often invisible, contextual aspects, such as historical, geographical or economic characteristics. These are hard to transform in the medium and even long-term and include a country’s geography, whether it is landlocked, has access to ports, or is mountainous. This clearly can have long-term impacts on a country’s domestic political-economic dynamics and regional relations. The history of state formation and a country’s sources of income can clearly also influence current contexts, be it broad-based taxation, or extractive rents from a limited number of sectors or actors.3 The type of colonisation and decolonisation are historical processes that have also profoundly shaped the types of political, economic and socio-cultural institutions that create incentives on different sets of actors in the present. The economic structure of an economy can also be considered a structural factor, including the income level and size of an economy, the level of concentration of economic activity, and employment in key sectors such as agriculture and the extractive sector. Such structural factors will clearly influence regional processes of integration whether in terms of broad regional negotiation processes or more narrowly defined cross-border projects with implications for the political, economic or military positions of certain countries, thereby creating a degree of path dependency.

2 Formal and informal institutions As the literature review highlights, institutions or the “rules of the game” are crucial for understanding how economic growth, political and social development, and policies affecting the nature of regional integration come about or function. This then relates to the way politics play out, the type and degree of political competition, the form in which political coalitions are formed and maintained, and the way in which rents and government revenues are raised and distributed, whether through formal or informal bargaining processes. as Rodrik (2007) states, all the institutional components of a well-functioning society are usually taken for granted by economists. Rather than take these for granted either at the national or regional level, these must be examined to identify where opportunities lies for effective policy reform towards greater regional integration.

While formal and informal institutions interact and influence one another in every country, the influence of informal institutions is likely to be higher in developing countries. Andrews (2013) uses the graphic representation or metaphor of an iceberg (see Figure 1) to draw attention to two key observations: firstly, and most visibly, formal institutions tend to capture the eye but in reality they thrive on the basis of less visible beliefs, norms, and cultural mechanisms, or informal institutions; secondly, in order to obtain a fuller picture of how formal institutions function - and influence policies and decisions - one has to understand the inter-relatedness between formal and informal institutions.

Figure 1: institutional structures as icebergs (Andrews 2013: p. 44)

3 “Access to high levels of rents and unearned income can reduce elite incentives to bargain with citizens and encourage elite predation.” (OECD, 2011: 26). ‘Rents” can be understood as income flows that are additional to market-based profits and wages. “The category includes revenues from natural and administratively created monopolies, subsidies, natural resources and, not leas important, corruption and aid. Some of the rents are economically efficient or socially beneficial, others are not.” (Booth and Therkildsen 2012: p. 10).

8

As the literature review discussed, an examination of formal and informal institutions means, among many other things, examining the process through which competition for political power takes place, whether through formal legal frameworks or more informal arrangements. Winning or retaining power may depend on informal practices such as exclusive patronage benefits such as jobs, money, access to services or monopoly privileges to supporters or "clients". Or it may depend on appealing to ethnic or other sources of identity. Understanding the type of effects of these institutional combinations in a given context is therefore critically important and will impact on how decisions are taken and implemented vis-a-vis regional processes.

An institutional analysis would also examine regulatory processes governing the private sector and the degree to which contracts and the rule of law are upheld. All of these factors then influence how different actors will interact around specific policy issues.

3 Actors, political elites and agency A third component of the conceptual framework assesses how actors or agents interact within the structural and institutional context and how this affects policy choices and behaviour. Such interactions may, for example, create incentives for different types of actors to act collectively in support of development outcomes, or conversely, for rent seeking, or for ‘crony’ relationships with exclusive benefits for a few.

As the literature review underlined, when analysing development processes, three types of actors/agents matter particularly: political elites, state bureaucrats and sector actors (civil society, firms, farms and households) and should be the focus of this element of the analysis. The ruling elite or coalition is often too weak to adopt a longer time horizon or to centralise rents for productive or developmental purposes, including regional projects. may be under pressure not to undertake those actions that are required for economic transformation or regional integration for that matter given the impact they will have on specific actors and their economic interests.

3.2 Zooming-in and zooming-out: sector characteristics and external driversThe picture drawn by an analysis through the three lenses can usefully be complemented with two more aspects. The first of these “zooms in” on the sector and public goods characteristics, and

9

presents a taxonomy that helps assess some of the linkages with governance and accountability implications. The second tool we introduce “zooms out” at the regional and global level, looking particularly at how regional and global drivers that are external to the object of the political economy analysis may affect or influence institutional or political incentives.

4. Sector characteristics, public goods and governance implicationsIn analysing political economic processes at a national level, development practitioners tend to refer to sectors and public goods in generic terms or as a single output (transport, education, health, etc.). But each of these sector goods and services can be disaggregated into various sub components with particular characteristics and potentially differing governance implications. For those national and regional public goods that are crucial for regional integration such as infrastructure development, regulatory frameworks, etc., policy and development practitioners will benefit from a deeper understanding of how these sector or public goods characteristics shape the political opportunities for reforms and development outcomes.

As discussed above, Mcloughlin and Batley (2012) develop a useful framework for breaking down these services as “goods” (or tasks) that helps understand the market for these goods, as well as the political incentives and constraints that different actors face who are involved in their provision or consumption. Four types of technical sector characteristics can be distinguished (see Figure 2). For the purpose of this paper, we merely summarise these categories of characteristics that will be important to consider in analysing any particular aspect of regional integration.4

Figure 2: Core sector characteristics

Source: Batley and Mcloughlin (2012)

As enumerated by Harris et al. (2013) the political salience and visibility of a service or a sector at national level will influence the incentives of politicians and providers to commit to providing quality services to citizens and be accountable for this. Secondly, more visible policies imply more easily attributable credit or blame, depending on information asymmetries. Thirdly, the balance of power between policymakers and other actors involved in service delivery is also important, with monopoly-provided services implying less incentive for state oversight or improved performance. And finally, services that are frequent, predictable and area-based may make it easier for citizens to mobilise collectively and to make demands for service delivery. Understanding the nature of specific aspects of reform processes in this way may help to further inform on where or to what degree the incentives of relevant actors are likely to be affected.

The complexities and governance constraints in the supply of public goods and services at national level are considerable in developing countries. These complexities and constraints are even more

4 This pertinent analytical framework is further explained in: Batley and Mcloughlin (2012), Chambers et al (2012). Its application to the sub sectors of road construction and road rehabilitation in: Wales and Wild (2012).

10

pronounced when more than one country is involved in regional public goods delivery. Providing an optimal level of supply of regional public goods requires a high level of trust and cooperation between countries (DfID 2011). Moreover, many of such regional goods are not pure public goods and do not benefit all members of the regional equally. In order to demonstrate the importance of this more fine-grained and sector specific analytical approach, Section 4 uses examples from the transport sector and the road subsector. Ensuring successful provision of roads and their maintenance often requires complex negotiations in an environment that is characterised by numerous asymmetries of interest, information, endowments and power.

5. External factors influencing the political economy contextNational and sector level actors and factors influence regional integration dynamics, whether through informal or formal REC channels. But it is important to consider the way in which external factors affect domestic and regional actors in their resolve to engage in cross-country cooperation or regional integration. Certain external global, continental or regional drivers may impact on institutional and political incentives, depending on the context. International drivers may push in directions that result in positive development outcomes, for example through global consumer and media pressures for legally produced and certified goods (timber, diamonds, etc.).

But these drivers can also impact negatively on domestic institutions and governance arrangements. Sources of rents and unearned income offer a clear example of a driver with potentially extreme harmful effects on state stability or on economic growth. The sources of rents and of state revenues have drastically changed over the last decades through the changing nature of globalisation (Moore et al. 2009; Moore 2011). Two features in the global economy jump to mind. One is the increased demand for natural resources, which has enabled certain political elites to access rents from the export of these resources. Secondly, the revolution in communication technology has made it much easier to conceal, launder and transfer financial assets. So ruling elites with access to both may end up with reduced political pressures or incentives for bargaining processes that may inform about constructive approaches to public investments in research, innovation, delivery of public and regional goods and services, or other policy areas that are critical for economic transformation, or for inclusive regional integration.

Unraveling such influences is an intricate exercise as there are multiple drivers, influencing country and sector level actors and institutions simultaneously, with different impacts over time. Therefore the OECD (2011) has developed an analytical tool that helps to do this. Identified potential influential drivers include sources of rents and unearned income (such as aid), opportunities for concealing and moving illicit assets, foreign investments, global and regional security threats and responses, international legal measures and sanctions against domestic elites, and reputational pressures on political elites from regional and international actors. There are many more, but what matters is to appreciate the extent to which such drivers may impact on the incentive environment in which national political and economic elites operate.

3.3. Developing a political economy approachThese analytical categories, tools and frameworks can be used in a variety of ways, as part of a quick scan or review or for a more in-depth study that involves original research. The purpose of the PE analysis will vary, and so will the level and scope of analysis. There is, however, much value in the warning of the OECD when it introduced its analytical tool on international drivers of corruption (OECD 2011): “The inherently incomplete and contested sources of evidence will limit users’ ability to arrive at definitive or causal explanations.” Still, such a conceptual framework can help users make more sense of a complex reality, and help prioritise in the shorter run, while also identifying areas and approaches for further knowledge development over a longer time.

11

It is clear from the level of complexity in regional integration in terms of diversity of policy and sector arenas, multitude of stakeholders, divergence of regional institutions and the range of external players that there is not one political economy approach to regional integration. The five components of the analytical framework will have to be combined, applied and adapted according to purpose and subject matter. A broad distinction can be made between assessing and addressing institutional and governance constraints in formal regional integration structures and processes on the one hand, and on the other hand the cross-border, sector specific processes (usually involving multiple public goods). These processes and dynamics may also impact on the formal integration dynamics.

4. Applying the political economy approach to regional integrationThis section provides examples of the use of these five analytical lenses on regional integration as applied basically in two different units of analysis: one is about the use of these lenses when focusing on one or more of the African RECs5, the second category involves different examples of “regionalism” or cross-border processes. This last category of examples usually interact with REC processes, but are not principally driven by it, and as argued may influence the regional integration dynamics of RECs.

This section therefore does not constitute a full political economy analysis of each case. The examples illustrate the application of the conceptual framework to different aspects of regional integration, point to the relevance of the types of evidence, information and data it may reveal and here and there contains some indications of how a Political Economy Approach may look like. Such approach is informed by the purpose(s) and the unit of PE analysis and basically depends on the combination of the uses of these lenses, the scope and depth of analysis, and the time frame and processes related to the implementation of the analysis.

Much remains to be said about such an approach - and some useful lessons and recommendations can be distilled from the available literature (Desai 2011, Fisher and Marquette 2013, OECD DAC 2008, etc.) - but this falls beyond the scope of this paper. Often, PE analyses have been donor sponsored for the purpose of understanding successes and failures in support to institutional reforms, or for purposes related to programme planning and implementation. For now, the cases are introduced in order to both illustrate the use of the framework and to enable to draw conclusions on the importance of employing a PE approach in analysing regional integration processes.

4.1. Structures, institutions, actors The generic PE framework invites to use these three lenses - structures, institutions and actors - to gain insight into how structural features, formal and informal institutions interact with sector actors (in particular political elites, state bureaucrats and sector actors) and shape the incentive environment. This may take the form of a quick scan if the purpose is merely to to identify some of the key challenges at regional or national level, but it may also involve a more considerate use of the three

5 Comment of the authors: the eight RECs are not equally represented in this first draft - and given the time and space constraints, this may not be feasible for the purpose of this paper but was part of the original design.

12

lenses, eventually requiring a more serious study through the lens of sector specificity and/or involving the study of global drivers (see next subsection 4.2).

1 Structural factors

From SADCC to SADC: historical factors in SADC formationRegional integration in Southern Africa is deeply affected by the history of colonisation, colonisation and the struggle against apartheid. The creation by the Frontline States in 1980, after the first free and fair elections in Zimbabwe, of the Southern African Development Cooperation Conference (SADCC) was essentially a combination of security, political and economic concerns. The military and economic destabilisation by apartheid South Africa of its neighbours, the continuing war in Angola and the occupation of Namibia were all elements affecting the agenda setting and the institutional arrangements of SADCC. Economic considerations were closely linked to the political rationale since it was necessary for landlocked neighbouring countries to reduce economic and infrastructure dependency on South Africa. The independence of Namibia, peace in the region and the prospects of democratic elections in South Africa stimulated the creation of the Southern African Development Community (SADC) in 1992, with a more development and integration oriented mandate and institutional architecture. Such evolution coincided with a greater prominence to economic considerations in SADC‘s agenda.

Political economy of infrastructure development in South Africa

Geography and history are determining, structural factors for the infrastructure development in South and Southern Africa. Roads and railways were constructed early on to serve the interests of the capital and labour intensive mines in South Africa. The geographic pattern of infrastructural investments reflected the hunger of the expanding mine industry for cheap labour and transport of minerals (initially diamonds and gold, but early in the 20th century more transport dependent minerals and other industries). The mining conglomerates continued to play a dominant role in South Africa and imposed some of their preferences in terms of soft and hard infrastructure in later years. Even after South Africa joined SADC in the mid nineties, it proved very difficult to diverge from this path dependency. When the newly elected government and President Mandela engaged in early discussions with his Mozambican counterpart about the reconstruction and the upgrading of the Maputo Development Corridor, the infrastructural trajectory basically followed the same path as in the early nineteenth century, despite the changes in the field of economic and social players. Such structural factors, combined with political imperatives may help explain certain policy preferences, persistent patterns and arrangements. The fact that the private sector controls and manages all bulk terminals in South Africa - and not the public sector - can be explained by a political deal that was struck between the apartheid state and the private sector: in exchange for the private sector’s support for government’s survival strategy (through an Import-Substitution Industrialisation model buffering against the threat of economic sanctions) the mining conglomerates received certain favours relating to exporting their minerals.

2 Formal and Informal Institutions6

Migration and free movement of persons - ECOWAS

● Within the realm of the regional structures such as ECOWAS there has been considerable progress in terms of formal protocols, mechanisms and arrangements to facilitate the intra-regional labour flows policies free movement of persons and migration. Yet, the implementation of the 32 years old protocol the Free Movement of Persons, Goods and

6 the bullet text requires some further finalisation and are indicative of the types of topics that can be dealt with

13

Services and the Right of Residence lags behind the policies and stated commitment to implementation (see also Chambers et al. 2012). Specific impediments include lack of sufficiently resourced immigration systems and informal practices of rent seeking at borders

ECOWAS/UEMOA - and informal cross-border trade in West Africa

● The establishment of free-trade areas (FTA) and common external tariff (CET) regimes within the UEMOA and ECOWAS has not resulted in an effective and even implementation among or across the member states. The anticipated increase in formal intra-regional trade has not been realised. Yet, informal market activity across national borders within the ECOWAS region continues to thrive. For example, the border between Benin and Nigeria presents various informal institutional arrangements that continue to baffle policy makers. From a perspective of trying to gradually shift to formal arrangements, particular analytical diagnostics (see subsection 4.2) would have to be undertaken to inform policy measures according to the nature of the different informal transactions in cross-border trade. Such informal cross-border trade arrangements often predate colonial times and administration, and have proven to be very sticky (see also Chambers et al. 2011). Chambers et al argue that such a sector level approach would also complement more general work on the political economy of Nigerian trade policy.

3 Actors

Regional infrastructure initiatives: North South Corridor and Zambia● Zambia is landlocked, yet the Sata-government prioritised a national road construction

programme over the regional infrastructural flagship programme of the Tripartite Free Trade Area of SADC-COMESA-EAC, the North South Corridor linking Durban to Dar-es-Salaam. The political incentives and the sector characteristics help understand the preferences of powerful actors, and the likely implications for future investments.

Private sector dynamics as drivers for cross-border cooperation● role of private sector actors (role of Yara and Unilever) in taking forward the Southern

Agricultural Growth Corridor of Tanzania (SAGCOT) as an inclusive, multi-stakeholder partnership to rapidly develop the region’s agricultural potential - in partnership with other stakeholders such as the government of Tanzania. In fact, it is the first PPP in Tanzania.

● [role of private players and investors in the Maputo Development Corridor in terms of hard and soft infrastructure development dynamics can be an alternative example]

4.2. The relevance of sector characteristics and regional/international drivers

The provision of key regional public goods remains a considerable constraint on regional integration despite the significant efforts by RECs and at cross-border levels. These constraints in goods (such as infrastructure, trade facilitation, management of water/fisheries/land, research and development) can be better understood when diagnosed in relation to the broader political incentives and the specific characteristics of the sub sector or the public good. This requires more in-depth analysis and drilling down to the level of one and usually more subsectors.

14

Similarly, more in-depth analysis is often required when regional or international drivers affect the political incentives on the domestic political economy. In fact, the OECD proposed a four-step analytical approach to unraveling such influences. One step involves a more careful appreciation and selection out of the multitude of regional and global influences those that have a stronger influence in terms of the political incentives they create.

4. Characteristics of sub sectors or regional public goods

Rural roads construction and rehabilitation

Understanding sector characteristics within the broader political environment may help identify particular governance constraints. Wales and Wild (2012) provide useful examples for the relevance of this lens by analysing the political economy of both the sub sectors of rural road construction and rural road maintenance. Typically, roads have a tendency towards market failures with sunk costs in construction and the potential for substantial free riding that leads to non-provision, unless they are provided as a public good. Another important characteristic is the high degree of political salience, as politicians often view road provision as a major mechanisms to reward supporters and to create new patron-client relations. Road maintenance, on the other hand has a low degree of political salience, so patronage politics may create incentives for politicians to prioritize highly visible projects or one-off visible road construction over less visible tasks of regular maintenance. Kenya provides an example of such “political market imperfections” (Burgess et al. 2009), as ethnic polarization resulted in road construction being targeted for patronage purposes. Another example of a governance constraint is provided by the World Bank (2010) and relates to the moral hazard7 that is induced by donor funding in sub-Saharan Africa. In low-income countries (both fragile and non-fragile) donor funding of road sector expenditure may amount to about half of all road sector expenditures. This availability of aid reduces the incentives for road reforms and creates a bias in favour of road construction projects. Transport prices and transport costs

The relevance of drilling down to the level of specific sub-sectors has been well demonstrated by the approach of Raballand and Teravaninthorn (2009) in their study on transport prices and transport costs in Africa. The study, among many other things, looks in more detail at the regulatory environment of the transport market. In the case of West and Central Africa, the study reveals some of the formal and informal regulatory constraints that result in limited competition, poor services and prices.Such more fine-grained analysis may help target or re-direct interventions in support of feasible and effective reforms, as demonstrate for Western and Central Africa. In their approach, they also paid particular attention to stakeholders’ opinions through workshops (with trucking companies and owner-operators). By bringing in these views, such more detailed analysis of specific characteristics of a sub-sector this lens provided insights into particular blockages to reforms, but also into possible avenues for reducing the resistance of certain stakeholders. The study, for example, pointed out the importance of compensatory measures for the potential losers of particular regulatory reforms.

5 External drivers

Role of emerging players

● The recent upsurge in trade and investments from emerging economies, in which China is the major partner, can be assessed more carefully, as it is likely to affect the existing incentive environment on key actors in African countries. Khadilgala (2011) suggested that China and

7 Moral hazard occurs when actors are insulated from the consequences of their actions, for example when there is another actor that provides costly compensation for such actions

15

others could form part of a ‘second generation’ of regional integration, which is endogenously driven by African institutions and disengages from the first generation of ‘exogenous’ regionalism primarily driven and informed by the EU.

● A PE analysis (see OECD, 2009) would study the effects of particular drivers in this area on political processes and the domestic political economy. It would also look at the secondary effects on institutions and structures.

5. Conclusion The conclusions will deal with some of the

● generic lessons of applying such PE framework and developing a PE approach on regional integration dynamics - scope, unit of analysis, etc.

● the PE approach will also involve reflections on the type of process in which the knowledge development should be integrated

● and it will deal with the implications for different types of donors/external actors who seek to support regional integration dynamics for inclusive growth

Bibliography

Acclassato, D. 2013. Intra-West African Trade. DPS N°2, Center for European Integration Studies. AFDB. 2011. Southern Africa Regional Integration Strategy Paper 2011-2015. AfDB AFDB. 2011b. Eastern Africa Regional Integration Strategy Paper 2011-2015. AfDB

Andrews, M. 2013, The Limits of Institutional Reform in Development. Changing Rules for Realistic Solutions. Cambridge University Press

16

Balassa, B. 1961. The Theory of Economic Integration. Greenwood Press. Barka, Ben, H. 2012. [KCKM1] Border Posts, Checkpoints, and Intra-African Trade: Challenges and Solutions. Policy Brief. African Development Bank, 2012.

Barma, N., Kaiser, K., Minh Le, T., Viñuela, L., 2012, Rents to Riches?: The Political Economy of Natural Resource-Led Development, The World Bank. Batley, R. and Mcloughlin, C. 2012. The effects of sector characteristics on accountability relationships in service delivery. London: Working Paper 350 Bertelmann-Scott, T., Catherine Grant, Lesley Wentworth. 2013a. Southern African Private Sector Drivers of Regional Integration. Paper for Session 2. Political Economy of Regional Integration in Southern Africa (PERISA). ECDPM and SAIIA Dialogue Meeting, 2-3 July 2013, Pretoria. Bertelmann-Scott, T., Catherine Grant, Lesley Wentworth. 2013b. The Management of Regional Public Goods in Favour of Regional Integration. Paper for Session 4. Political Economy of Regional Integration in Southern Africa (PERISA). ECDPM and SAIIA Dialogue Meeting, 2-3 July 2013, Pretoria. Booth, D., and Frederick Golooba-Mutebi. 2009. Aiding economic growth in Africa: The political economy of roads reform in Uganda. ODI Working Papers Issue 307. September 2009. Booth, D. and Therkildsen, O. 2012. The political economy of development in Africa: a joint statement from five research programmes, Africa Power and Politics Programme; Developmental Leadership Programme; Elites, Production and Poverty; Future Agricultures; Tracking Development. Carère, C., Julien Gourdon, Marcello Olarreaga. 2012. Regional Integration and Natural Resources: Who Benefit, Evidence from MENA. Policy Research Working Paper 5970. World Bank.

Desai, M.R. 2011. An Evaluation of Political-Economic Analysis in Support of the World Bank’s Governance and Anticorruption Strategy. IEG Working Paper 2011/4, Washington: IEG DFID. 2009. How To Note. Political Economy Analysis. A DFID practice paper.

Draper, P. 2010. Rethinking the (European) Foundation of Sub-Saharan African Regional Economic Integration: A Political Economy Essay. OECD Working Paper N° 293. OECD Development Centre. ECDPM and SAIIA. 2012. Political Economy of Regional Integration in Southern Africa South Africa as a Catalyst of Regional Integration. Project Abstract. March 2012 – July 2013. Fisher, J. and Marquette, H. 2013, Donors Doing Political Economy AnalysisTM: From Process to Product (and back again?). Birmingham: University of Birmingham

Flatters, F. 2002. ‘SADC Rules of Origin: Undermining Regional Free Trade’ Paper Prepared for TIPS Forum, Johannesburg, South Africa, September 9-11 2002 Fouroutan , F. 1993. Regional integration in sub-Saharan Africa: experience and prospects. World Bank.

17

Harris, D., Mcloughlin C. and Wild, l. 2013. The technical is political. Why understanding the political implications of technical characteristics can help improve service delivery. ODI Hartzenberg, T. 2012. The Tripartite Free Trade Area – shaping a new paradigm for African integration? Foreign Voices. No. 02 April 2012. Hoppe, M., and Francis Aidoo. 2012. Removing Barriers to Trade between Ghana and Nigeria: Strengthening Regional Integration by Implementing ECOWAS Commitments. Africa Trade Policy Note No. 30. March. Khan, M. 2010. ‘Political Settlements and the Governance of Growth-Enhancing Institutions’, Draft Paper in Research Paper Series on 'Growth-Enhancing Governance'. Countries include Thailand, India (Maharashtra and West Bengal), Bangladesh and Tanzania.

Khan, M. 2012. ‘The Political Economy of Inclusive Growth’, in OECD and the World Bank, Promoting Inclusive Growth: Challenges and Policies. Paris: OECD.

Lamy, P. 2010. Intégrations régionales en Afrique: Ambitions et Vicissitude. In Europe et Gouvernance Mondiale. Notre Europe.[KCKM7] Lawrence, R. 1996. Regionalisms, Multilateralism and Deeper Integration. Washington: TheBookings Institution

OECD, 2009, Donor Approaches to Governance Assessments. A Sourcebook, Paris: OECD OECD. 2011. International Drivers of Corruption: A Tool for Analysis. Paris: OECD Putnam, R.D.1988. “Diplomacy and Domestic Politics”, International Organization, Vol.42, No.3. Raballand, G. and Teravaninthorn, S. 2009. Transport Prices and Costs in Africa. A Review of International Corridors. Washington: the World Bank Rodrik, Dani. 2008. “Second-Best Institutions” American Economic Review: Papers and Proceedings 98(2): 100-104 Unsworth, Sue and Gareth Williams. 2011. Using Political Economy Analysis to improve EU Development Effectiveness. A DEVCO Concept Paper. Wales, J. with Wild, L. 2012. The political economy of roads. An overview and analysis of existing literature. London: ODI West African Trade Hub. 2010. ECOWAS Market Integration: Results of A Preliminary Gap Analysis. West African Trade Hub Technical Report No. 33. Whitfield, L. and Therkildsen, O. 2011. ‘What Drives States to Support the Development of Productive Sectors? Strategies ruling elites pursue for political survival and their policy implications’. DIIS Working Paper No 15. Copenhagen: Danish Institute for International Studies.

World Bank. 2010. Roads: Broadening the Agenda, in V. Foster and C. Briceno-Garmendia (eds) Africa’s Infrastructure: A Time for Transformation. Washington: the World Bank

18