Article en français : Passerelles, Volume 16 - Number 5.
The Tripartite FTA (TFTA) and the proposed Continental FTA (CFTA) are the latest African initiatives towards regional cooperation. To succeed, these have to confront a very uneven distribution of resources that have sharpened the trade-off between the benefits of common policies needed to tackle cross-border externalities and their costs which are heightened by the sharp differences in policy preferences across members. Abandoning the linear model of integration and integrating in small groups should help.
Following the implementation of the Economic Community of West African States’ (ECOWAS) common external tariff (CET) in January 2015, this June saw the launch of the Tripartite Free Trade Area among 26 countries, accounting for over half of Africa’s GDP and, with 632 million people, 56 percent of the continental population. A Continental FTA is also to be launched in or around 2017. Phase I of the TFTA suggests modest efforts at integration as it is built on the principles of variable geometry eschewing a more ambitious “single undertaking” and the acquis (go forward but not backwards) with modest tariff reductions on the table, a list (rather than an economy-wide criterion) for rules of origin, trade remedies to address dumping, and import surges. The agenda for phase II is to be decided but should include services and harmonization of rules on competition policy. The TFTA is expected to be ratified by at least half of the members within a year, at which point it will come to life.
Is the attempt at rationalizing the multiple regional integration arrangements (RIAs) across the continent a milestone towards greater cooperation across the continent? Drawing on observations and analysis of the recent experience, I argue that, in spite of the unfavorable geography that makes it difficult to deal with the high costs of heterogeneity, integration initiatives in small member groups will produce the highest benefits.
Following the founding of the Organization of African Unity (OAU) in 1963, a first wave of RIAs took place along “regional economic communities” (RECs) behind high tariff walls. These RECs were to be the “building blocks” of the hoped-for African union in the immediate post-colonial era. Now, they are central for implementing the New Partnership for Africa’s Development (NEPAD). In short, the RECs were and continue to be the glue that will cement African unity. The first wave failed not only for economic reasons, but also because the leaders of these young, post-independence African states were reluctant to encourage the emergence of a supra-national authority necessary to deepen cooperation and coordinate and manage the affairs of the hoped-for African union. Great diversity within the RIAs translated into different interests that strengthened countries’ insistence on the “respect for the sovereignty and territorial integrity of each State and the inalienable right to independent existence,” as written in the Organization of African Unity charter of 1963. Commitment to pan-Africanism was weakened, leading to vagueness and a multitude of declared objectives in these RIAs that helped states gloss over the issues that divided them.
A second wave of RIAs took place after the Abuja Treaty of 1991. A look at the 10 major RIAs started in this second wave of RIAs shows that only three have aimed for FTA status, and all others aiming for deeper integration, with integration moving along the linear model following a stepwise integration of goods, labor, and capital markets, and eventually monetary and fiscal integration. Goods market integration would start with an FTA, then move on to a customs union (CU) with a common external tariff and to a common market. Along this linear sequence, except for the Southern African Customs Union (SACU), none have really reached full CU status because exceptions to the 4-5 CET tariff band structure are so numerous. For example, the ECOWAS CET includes an “exceptions list” of about 300 products eligible for exemption from the new tariffs that includes 200 products from the former Nigerian Import Ban list.
The disappointing trade performance of this model of integration has been widely discussed. Among others, estimates of the volume of intra-regional trade in African RIAs suggest that trade is, on average, 40 percent less than potential trade, and that the ratio of actual trade to potential (i.e., to frictionless) trade among partners has fallen by about 10 percentage points from 0.63 two years before signature to 0.53 seven years after signature, suggesting that trade costs among partners have fallen less rapidly than trade costs with outside partners. This persisting thickness of borders not only reflects the geography of African trade, the low trade complementarity across partners, poor logistics, and border delays, but also the neglect of services in the African linear integration model, which is no longer adapted to 21st century trade.
So far, negotiations for the TFTA and CFTA are following this model of linear integration that neglects the fact that 21st century production is increasingly taking the form of trade in tasks (i.e., services) as opposed to trade in products. In this new environment, services play an input function through space (transport, telecommunications) and time (financial services) as well as direct inputs into economic activity as they generate knowledge and human capital. Recent developments in the study of global value chains by the OECD show that services may account for more than 50 percent of exports when measured in value added. Because services do not meet customs for registration, and regulations are, at best, imperfectly captured, services—except for labor and FDI flows—are not directly observed crossing borders. Measures of the restrictiveness of trade services are only very approximate, though estimates of trade costs for mode I (cross-border services trade) and mode II (movement of consumers) could be two to three times higher than trade costs for trade in goods measured by the same approach (the “gravity trade model”).
Breaking away from the linear model of integration by emphasizing trade facilitation measures at the border that have full support of the business community is a first step now under way. However, even in the case of the East African Community Common Market, there has been little progress at removing restrictions for professional services, telecommunications, and transport either unilaterally or on a regional basis. Likewise, progress with liberalization of services through harmonization and mutual recognition has been slow where opting for “mutual equivalence,” the route that was followed by the European Union Services Directive, might have worked better, as this approach is less demanding on trust than mutual recognition or harmonization.
In 2013, all of Africa’s GDP at PPP was less than Germany’s, and the median GDP size of African countries was $12.3 billion, about 10 percent the size of the canton of Zurich’s. The potential benefits of economies of scale and of diluted monopoly power present a case for RIAs to have large memberships, as seen in the TFTA and the CFTA. But a large membership also implies more heterogeneity and greater sources of potential conflicts (more ethnic groups, large and small countries, and landlocked and coastal states belonging in the same regional group) with higher political costs in the provision of public goods. In large membership groups, integration is shallow because it is difficult to reach agreement, and it is likely that the interests of the more powerful members that are naturally less open to the outside world will prevail. Take ECOWAS, where Liberia and Nigeria are both members. Adopting the CET took close to 10 years of negotiations as Nigeria insisted on a 5-band CET (0-35 percent) while West African Economic and Monetary Union (UEMOA) and others were in favor of a 4-band CET (0-20 percent). For Liberia, the move to the CET could double the average tariff and raise the current costs of living for rural and urban households by 6 percent and 3 percent, respectively, with temporary special protection measures only envisaged for products currently above their respective band, but no consideration has been given for tariffs below their respective band . The costs of integration to a customs union for small countries in a large membership group with large partners are likely to be high.
This experience poses a challenge for the 26 member TFTA because 21st century regionalism is no longer about an exchange of market access at the expense of non-members but about implementing reforms that will attract FDI, which brings to the region the service activities necessary to participate in the outsourcing of production. In this new environment, where trade is trade in tasks and increasingly involves an exchange of intermediate goods, protection (or exchange of market access) amounts to depriving oneself participation in global outsourcing. Not only is deep integration (which is necessary to attract FDI) likely to be hard to carry out within a large membership, but there is also the risk that protection towards non-members could remain high.
Deep integration requires some delegation of authority to a supranational level. This is easier to carry out under small membership. The five-member EAC, which in 2010 started implementing a common market in capital, goods, and services, uses a scorecard approach to measure progress (violations of the protocol provisions in services are made public on the EAC website, which is far more informative on progress at integration than websites for other African RIAs). The EAC is also promoting competition in telecommunication by banning roaming charges within the region and issues single tourist visas for northern corridor countries (Rwanda, Kenya, and Uganda). The EAC is the only RIA where the ratio of actual to potential intra-regional trade has risen following integration.
In order to break the curse of small markets, the large group approach appears to be the most appropriate for exploiting economies of scale, but cooperation associated with public goods like a common currency, a common judicial and legal framework, as well as appropriate regulatory policies also bring benefits. For the franc zone members, sharing a common currency is associated with more intense bilateral trade attributable to less volatility in bilateral exchange rates. Thus SACU, UEMOA, and Economic and Monetary Community of Central Africa (CEMAC) have benefitted from deep integration, albeit with the costs of institutional development covered by the colonizers and the EAC is moving in that direction. On the other hand, arrangements with larger memberships and more heterogeneous populations like the TFTA, face higher political costs in the provision of public goods. The European experience shows that the trade-off between economies of scale and heterogeneity of preferences can only be partially addressed through decentralization at different layers of administration.
In Africa, regional spillovers are important as transport and communications infrastructure are underprovided, but the ethno-linguistic diversity across “artificial” borders indicate strong differences in policy preferences that will continue to hinder the future supply of public goods through the adoption of common regional policies in large groupings. Common decision making internalizes the spillovers but moves the common policy away from its preferred national policy (i.e., a loss of national sovereignty). Are initiatives like the TFTA and the CFTA the start of institutional and political cooperation along intergovernmental lines, where regional institutions pursue the economic interests of domestic constituencies as has largely been the case of the EU?  Or, more optimistically, as hoped for by the African Union (formerly the OAU), is this a start along functionalist lines where supranational institutions and agents develop autonomous roles leading to further integration?