2

Federalism and Decentralization

Rotimi T. Suberu

Although Africa boasts only three relatively established federations – Nigeria, Ethiopia, and South Africa – federalist ideas and institutions for combining self-rule with shared rule in multilevel governance systems have tremendous appeal, resonance, and relevance across most of the continent’s 54 states. Federal principles for the intergovernmental division of powers have been used explicitly or implicitly in Africa not only to construct multi-ethnic or multi-racial federations, but also to foster political decentralization in unitary states, and to promote inter-African or supranational relations and institutions among different states.

At the same time, federalist experiments in Africa, not unlike other formal political institutions in the continent, have an unimpressive and uninspiring track record. The federal systems in Nigeria, Ethiopia, and South Africa, for instance, face enormous political and economic strains that challenge the federal character of these polities. In those three federations, and in the less established federal system of Comoros, as well as in decentralizing unitary states like Senegal and Uganda, the relations between national and sub-national orders of government remain fundamentally lopsided, precarious, or conflicted. Similarly, the intergovernmental institutions designed to promote international (continental/regional and sub-regional) integration in Africa are mostly weak, rudimentary, and problematic. Indeed, a hiatus exists between the obvious attractions and imperatives of federalism in Africa and the absence and paucity of the structural and political conditions that are conducive to the development of federal institutions (Kymlicka 2006).

This chapter explores Africa’s chequered experiences with federalism. It begins by discussing the federal idea and the principles and properties that have made it paradoxically both appealing and unviable in the African context. A consideration of the implementation and performance of federalist principles and arrangements in Africa’s federations and decentralized unitary states is then offered, and the final section draws together the main implications of my argument: that the limited success of federal structures thus far is unlikely to dampen their allure in the future.

The Architecture and Attractions of Federalism

Like many other important social science concepts, federalism is, analytically, a highly contested, elusive, and ambiguous term. The concept has been persuasively used to describe radically different political systems, including confederations, federations, devolutionary arrangements in unitary states, international unions, and hybrids combining characteristics of different governmental designs (Watts 2008: 10–11). This conceptual nightmare may be mitigated by distinguishing between federalism as a generic paradigm and philosophy of political design, on the one hand, and the practical applications of this design in specific governmental systems and institutions – notably federations, decentralized unitary states, confederations, and international unions – on the other.

As a broad genus, federalism is a philosophy of governance that advocates and promotes the combination of shared rule and self-rule in voluntary, multi-tiered political systems that can advance such normative principles as subsidiarity (the idea that a higher authority should only undertake those functions that cannot be accomplished by lower levels of authority), unity in diversity, power-sharing, liberty, competitive innovation and experimentation, prosperity, democracy, conflict management, peace, and security. The classic governmental expression of federalism is federation, which is a political system involving the constitutionally entrenched division of powers between regional governments and a central government as well as the representation of the regional governments in the machinery of the central government.

Aside from a written and rigid constitution specifying the governmental division of powers, the key institutional properties associated with federations include: a bicameral legislature, one of which represents the people of the federation as a whole and the other the constituent units; an umpire, usually a supreme or constitutional court, that rules on intergovernmental constitutional disputes; allocation of revenue resources between regional and central governments in order to ensure genuine autonomy for each order of government; and ‘processes and institutions to facilitate intergovernmental collaboration for those areas where governmental responsibilities are shared or inevitably overlap’ (Watts 2008: 9). According to two leading experts on federalism, 28 countries – accounting for over 40 per cent of the world’s population – currently either call themselves federations or are generally considered to be so (Anderson 2008; Watts 2008). These include six African countries, namely, Comoros, Democratic Republic of Congo (DRC), Ethiopia, Nigeria, South Africa, and the former Sudan (before the July 2011 independence of South Sudan). However, in the Comoros, as in the so-called Transitional Federal Government (TFG) of the collapsed state of Somalia, the federal characteristics of the polity have been rendered inchoate by chronic political instability and chaos. Similarly, the federal credentials of the DRC are dubious not only because of the incomplete reconstitution of the failed Congolese state and the non-implementation of some of the more decentralist features of its 2006 Constitution, but also because the country has officially avoided an explicit federal identity. Another African country with a contested or ambiguous federalist identity is the federacy of Tanzania, where an otherwise unitary (albeit decentralized) state has developed a (con)federal relationship with approximately 3 per cent of its population living in Zanzibar, which has its own distinctive constitution, presidency, legislature, judicature, bureaucracy, local government system, and electoral political dynamics. Described as the only variation between federal and unitary states, a federacy involves the asymmetrical grant of political autonomy to a distinct, usually peripheral, community in a state, while the core population of the state remains under unitary rule (Stepan 1999).

Indeed, in an overwhelming majority of African countries, devolution or decentralization within a unitary framework, rather than federation, has been the preferred mode for institutionalizing self-rule and shared rule since the third wave of democratization inspired a corresponding wave of territorial political reforms across the continent in the 1990s. This is partly because federation entails a degree of territorial autonomy that is considered by central authorities in Africa and most of the developing world to be potentially subversive of state continuity, stability, and unity. In most decentralized unitary arrangements, unlike in federations, the powers of lower levels of government are not constitutionally entrenched. Rather, sub-national powers are devolved by a central government, the authority of which to decentralize also includes the ability to re-centralize power. Furthermore, unitary arrangements do not typically provide for the representation of sub-national authorities in the machinery of the central government.

Yet, the distinction between federations and decentralized unitary arrangements can be very imprecise and ambiguous for several reasons. For one, because a country’s self-identity as federal or unitary may have ‘deep historical and symbolic meanings’, identical labels ‘may resonate differently in different contexts’ (USAID 2009: 18). South Africa, for instance, has adopted all the basic features of federations while formally rejecting a federal identity because of the label’s association with the country’s history of racial apartheid. Second, several polities in Africa, as elsewhere, include features from different constitutional systems, thereby defying any rigid or rigorous classification into unitary and federal types. Third, and related to the above, decentralization or the shift of powers and resources to sub-national orders of government can theoretically take place in both unitary polities and federations. Consequently, although federations are usually ‘more decentralized than unitary countries, federations can be quite centralized and unitary countries can in fact be highly decentralized’ (ibid.: 19).

The formally decentralized polities highlighted in this chapter include Kenya, Uganda, Cameroon, and the DRC, where decentralization was preceded by failed attempts to experiment with federation or quasi-federation. Also noteworthy are Mozambique, Mali, and Senegal, where decentralization processes have included significant elements of devolution (the transfer of autonomous competencies and resources from higher orders of government to elected sub-national provincial and/or local governments), as distinct from mere deconcentration (the execution of central government functions by the centre’s own field staff) or delegation (the implementation of central directives and functions by sub-national authorities).

Federalism’s appeal in Africa derives from its presumed capacity to facilitate the three overarching aspirations of the developing world – unity, democratic development, and socioeconomic progress. Federalism can be especially apt for managing the multi-ethnicity of most African states, enabling these countries to diffuse and defuse ethnic conflict. Specifically, the grant of autonomy to territorial sub-units in multi-ethnic states can empower or appease ethnic minorities, transforming them from national minorities to sub-national majorities. Such territorial autonomy can also: compartmentalize conflict so that the conflicts of one sub-unit do not polarize or destabilize the rest of the state; generate potentially cross-cutting, intra-ethnic competition in ethnically homogeneous sub-units; promote inter-ethnic cooperation and political socialization in more ethnically heterogeneous sub-units; stimulate alignment on non-ethnic issues as regions controlled by different ethnic groups forge functional lines of cooperation or coalesce to defend their collective interests; and reduce ethnic or horizontal inequalities through schemes for inter-regional redistribution or equalization (Horowitz 1985).

Federalism can deepen and broaden democracy in a number of ways: by facilitating citizen engagement at sub-national or local levels; by enhancing the accountability and responsiveness of government to citizen needs; by making government more representative of the diversity of the population; by furnishing checks and balances on the powers of the central state; and by providing opportunities for the political opposition at the national level to exercise power at the sub-national level and thus, to acquire a direct stake in the development of the democratic political system (Diamond 1999). Federalism can also engender economic efficiency and prosperity through inter-jurisdictional or inter-governmental competition. Indeed, the most economically dynamic and prosperous countries in the world today are either fully-fledged federations or, like China, highly decentralized unitary states. This is because, as claimed by the theory of market-preserving federalism, competition between sub-units for capital and labour can reduce corruption and mismanagement, stimulate innovation and experimentation, and generate a credible political commitment to ‘limited government’ and the preservation of productive private economic activity (ibid.: 153).

Yet, especially in the developing world, federalism and decentralized governance has often failed to fulfil its advertised advantages. Rather, a substantial countervailing literature exists linking federal institutions to the exacerbation of inter-ethnic discrimination and ethno-secessionist conflict, the creation of authoritarian enclaves, and the complication of macro-economic management. More pertinently, however, recent scholarly writings on federalism have moved beyond simplistic generalizations regarding the benign or malign consequences of federalism, and focus instead on the institutional and structural conditions that may produce federalism’s beneficial or baneful effects.

Constraints on Federalism in Africa

Six interrelated factors have undermined the successful implementation of federalism and decentralization in Africa. First, the arbitrariness, artificiality, and perceived illegitimacy of Africa’s colonially imposed boundaries are at odds with federalism, which ultimately ‘requires a voluntary will to federate’ (Adamolekun and Kincaid 1991: 180). Africa’s multi-ethnic states continue to struggle fundamentally with this issue of their territorial legitimacy, as underscored by the constant demands for a ‘sovereign national conference’ to discuss the desirability and modalities of federal union in Nigeria, or by the formal insertion of a voluntary exit or secession clause in the Ethiopian Constitution, and in the Comprehensive Peace Agreement (CPA) for Sudan. Yet, this issue has generally been avoided in Africa, where the prevailing norm is to maintain existing state boundaries at all costs, rather than to accept their negotiability or reversibility through democratic processes.

Second, African states are inherently weak economically and politically, and therefore are largely bereft of the critical capacities required for sustaining federalism (see also Abrahamsen, this volume). These capacities not only enable central authorities to promote a ‘common roof’ of rights and liberties, and ‘minimum floor’ of socioeconomic opportunities throughout the state, but also give sub-national administrations the resources to establish their own autonomy; exercise legislative, administrative, and fiscal powers; and engage constructively in intergovernmental relationships (Diamond 1999; Simeon and Murray 2004). Without these capacities, federalism can compromise rather than aid democracy, stability, and development. Thus, although decentralization is often promoted as a pragmatic response to the inability of African states effectively to control their hinterlands or peripheries, ‘only governments that [are] strong at the [centre], both politically and administratively, could take the risk and find the resources to decentralize important programs or functions’ (Stren and Eyoh 2007: 6). Not surprisingly, Africa’s insecure central elites have rarely promoted genuine local autonomy. Rather, they have often manipulated the rhetoric and institutions of decentralization to subvert, ‘coopt, demobilize, usurp, bypass, or modify’ sub-national institutions (Boone 2003: 369).

Third, although Africa’s diverse territorially based ethnic constituencies are considered to make federalism a ‘natural option’ for the continent’s states (Kymlicka 2006: 46), the often fragmentary, fluid, and multi-layered – yet parochial and discriminatory – nature of those identities can pose difficult challenges for federal design. Where groups lack internal homogeneity and stable external boundaries, the establishment of viable units of sub-national governance may be problematic. Rather, an inclusive multi-ethnic coalition in the central state apparatus, as distinct from territorial self-rule at the sub-state level, may be a more sustainable strategy of ethnic conflict management. The case against territorial autonomy is strengthened by the pervasiveness of discourses and practices of discriminatory autochthony or indigenousness in most African communities. These practices promote inter-group conflict at the sub-national level, militate against the development of a common civic state citizenship, and undermine the inter-jurisdictional mobility that is required to engender ‘market-preserving’ intergovernmental competition, experimentation, and innovation (Ndegwa 1997).

Fourth, basic flaws in the design and implementation of federalist schemes have discredited and undermined federalism in Africa. Some of the more common design flaws include sub-optimal numbers of constituent units, dramatic inequalities in the sizes of sub-units, unsustainable asymmetries in the powers of sub-units, the over-concentration of functions and/or revenues in the central government, the failure to incorporate strong incentives for sub-national units to generate their own resources, and the belated implementation of federalist schemes after ethnic feelings of injustice and exclusion have hardened, leading to the development of federal arrangements that are more divisive than integrative. As underscored by the Sudanese and Ethiopian federal experiments, the price for such belated separatist federalism may include the suicidal enshrinement of secession as an explicit principle of constitutional design rather than as something that may implicitly evolve ‘out of piecemeal democratic negotiations’ (Kymlicka 2006: 55).

Fifth, while federalism is democracy’s territorial correlate, most African countries have simply lacked the basic democratic architecture required for federalism to take root and flourish. In particular, federalism in Africa has been severely degraded and truncated by electoral processes lacking in transparency, credibility, and fair, inter-party competition for control of central and sub-central government. Thus, the credibility and vitality of federalism and decentralization in Africa has been undermined by ‘the prevalence of dominant-party states’ in countries as diverse as Botswana, Burkina Faso, Ethiopia, Mozambique, Nigeria, South Africa, Tanzania, and Uganda (Dickovick and Riedl 2010: 10). Finally, the African regional context or system of inter-state relations has been particularly inauspicious for the development of federalism. Essentially, the combination of artificial colonial boundaries, weak states, and insecure state elites have militated against the development of a pan-African consensus on values – like democracy and the rights of ethnic and political minorities – that are necessary for federalism to flourish both nationally and internationally.

Federations

Nigeria is Africa’s most established federation, having instituted a three-region federation under British colonial auspices in 1954 to hold together its three major ethnic groups, hundreds of smaller ethno-linguistic communities, and roughly equal numbers of Muslims and Christians. This tri-regional federal system, however, exacerbated Nigeria’s ethnic divisions, leading to the destabilization and collapse of the country’s first post-independence republic (1960–66), bloody ethno-military infighting, and three years of civil war (1967–70). Since the country’s first military coup in 1966, Nigeria’s most important achievement as a federation consists of the progressive transformation of this centrifugal three-region structure into a much more centripetal, multi-unit federal system, which currently consists of a federal government, 36 states, and 776 constitutionally designated local governments.

Despite its remarkable success in preventing a recurrence of ethno-secessionist warfare and state collapse, Nigeria’s current federal system suffers from several pathologies and contradictions that graphically illustrate the limitations of federalism in African contexts. Since the inception of the country’s fourth civilian republic in 1999, for instance, the Nigerian federation has witnessed thousands of deaths in ethnic clashes arising from discrimination against so-called non-indigenes (Nigerians resident in sub-units outside their presumed ancestral communal roots) by ethnic indigenes in control of state and local governments. The communal bloodletting has been aggravated by the terrorist insurgency of the Islamic sect Boko Haram, which has killed and campaigned for the expulsion of southern Nigerian Christians living in the predominantly Muslim north, partly in retaliation for violent attacks against Muslims by indigenous non-Muslim groups in Jos and other cities in Nigeria’s ethnically and religiously mixed Middle Belt. Nigeria’s federal constitution abets the conflict by mandating the equitable representation of indigenes of each constituent state in the central government and then defining such indigenes in terms of ethnic autochthony and genealogy rather than geographical residence (Geschiere, this volume).

Nigerian federalism has also been degraded by violent insurgency in the Niger Delta, the ecologically and economically neglected and ethnic minority-populated region that produces all the oil revenues on which Nigeria’s federal, state, and local governments are completely dependent. In addition to engendering a sense of economic grievance and deprivation in the Delta, the centralized collection and redistribution of oil revenues in Nigeria has undermined economic decentralization and diversification, promoted fiscal irresponsibility and political corruption (including massive electoral fraud perpetrated to gain control of central, state, and local governments), and spawned significant agitations for a ‘true federalism’ that would dismantle some of the military’s centrist reforms and establish more financially viable and democratically accountable sub-national authorities.

The formal design of the current Ethiopian federal system, following failed experiments in the Ethiopia-Eritrea Federation (1952–62) and in unitary nationalization (1962–91), contains several features that many Nigerian advocates of ‘true federalism’ will endorse. This includes the organization of its nine constituent regional states on an explicitly ethnic basis, the endowment of constituent units with their own constitutions and police agencies, the promotion of ethnic political parties, the formal granting of a right to self-determination (up to and including secession) to ethnic communities, and the demarcation of the federation into a small number of relatively large states (rather than a large number of mostly small or weak states).

Yet, Ethiopian federalism is vexed by contradictions that are not unlike the failings of the Nigerian system. While contributing significantly to the regulation of national-level ethno-political conflict in Ethiopia, for instance, the policy of ethnic federalism has aggravated such conflict at the regional and local level. This is especially evident in states like Beninshangul-Gumuz, Gambela, Oromia, and the Southern Nations, Nationalities, and People’s Region (SNNPR), where ethno-linguistic communities considered non-autochthonous to these states suffer discriminatory practices that effectively reduce them to second-class citizens (Fiseha and Habib 2010: 154–55).

The Ethiopian federal system is also heavily centralized financially, with the federal government controlling more than 80 per cent of aggregate revenue in the country, leading to high and sustained regional dependency on central financial subsidy. This vertical fiscal imbalance contributes to a centrist political and policy process that ‘enhances central political leaders’ domination, kills regional incentives to innovate and be flexible, discourages challenge to the status quo, and makes lower level politicians and office holders insecure’ (Chanie 2007: 367). Such economic and policy centralization is nurtured and reinforced by the political hegemony of the Ethiopian Peoples’ Revolutionary Democratic Front (EPRDF) coalition, the member and affiliated parties of which control all governments at the federal and regional levels. Indeed, the EPRDF has demonstrated extreme intolerance to the development of any vibrant political opposition spaces at national and sub-national levels, and has instead established a Soviet-style authoritarian party state. In essence, while ethnic federalism has contributed significantly to the reconstruction and stabilization of the Ethiopian state, the federation still has a long way to go ‘in terms of enhancing the autonomy of the states, strengthening the institutions of democracy, and protecting minorities in the regional states’ (Fiseha and Habib 2010: 140).

Unlike Nigeria and Ethiopia, post-apartheid South Africa does not explicitly define itself as a federation. Indeed, federation is supported primarily by narrowly based minority parties – the white-dominant National Party and Democratic Alliance (DA), and the Zulu nationalist Inkatha Freedom Party – and remains unpopular with most black supporters of the dominant African National Congress (ANC), who favour a strong, centrist, and consolidated developmental state. Reflecting the inclusive, consociation-oriented nature of South Africa’s political transition to multi-racial democracy, however, the country’s Constitution recognized national, provincial, and local authorities as ‘distinctive, interdependent, and interrelated’ spheres of governance (Simeon and Murray 2004: 277).

Inspired by the German national cooperative model of federalism, the South African federal system is strongly centrist, ‘with the predominance of legislative power assigned to the center, with a centralized system of public finance, and with considerable central powers to oversee provincial performance’ (Simeon and Murray 2004: 297). Although they are constitutionally empowered to develop their own constitutions, all of South Africa’s current nine provinces lack real political and decision-making autonomy. Rather, their authority is subject to the overriding prerogatives of the centre to maintain the country’s security and unity, and their major role has been effectively limited to implementing and delivering the services mandated by the national government. Like their counterparts in Nigeria and Ethiopia, the South African provinces are very weak financially, depending on central revenue transfers for over 90 per cent of their resources. Unlike the more financially autonomous localities, which collect property taxes, the South African provinces do not have any significant base of own-source taxation or revenue.

A major centralizing force in South Africa remains the overwhelming electoral dominance of the centrist ANC. Unlike Ethiopia and Nigeria, however, South Africa is a credible electoral democracy, with genuine freedom for opposition parties to organize and mobilize politically at national and sub-national levels. The potential for this relative electoral integrity to eventually develop into a more viable political pluralism and federalism is reflected in the significant electoral and governance successes of the DA, first in Cape Town, and then in the Western Cape Province.

Despite their political and economic over-centralization, the federal systems in Nigeria, Ethiopia, and South Africa have emerged as largely durable and stabilizing features of government and politics in these countries. The Comoros, on the other hand, remains a very fragile and conflicted federation. The Comoros is a union of three small islands, one of which (Grande Comoros) is more populous than the other two (Anjouan and Moheli) combined. Reflecting this blatant structural imbalance, the federation has witnessed intense inter-island rivalries, separatist movements, numerous coups and attempted coups, and repeated constitutional overhauls and political reforms designed to create a less precarious federation.

A profoundly multi-racial, multi-ethnic, multi-religious, and multi-lingual state, Sudan is a prime example of the limited viability of federation when the system is introduced after inter-group conflicts have escalated and hardened to a disintegrative stage. The first Sudanese federalist autonomy arrangement (1973–83) was introduced under the 1972 Addis Ababa Peace Agreement after 17 years of violent rebellion mounted by the country’s southern, non-Muslim, black African minority, against political and cultural subjugation by the Arab-Islamic unitary government in Khartoum. When Khartoum unilaterally abrogated the Addis agreement in 1983, the South resumed its rebellion, which killed more than 2 million people, and displaced an additional 4 million, over a 20-year period (1983–2005).

Following protracted international mediation, federalism was reintroduced in Sudan as a cornerstone of the historic 2005 Comprehensive Peace Agreement (CPA). Reflecting the legacy of two bitter civil wars, however, the CPA was a classic partitionist blueprint, elaborately providing not only for power-sharing, oil revenue-sharing, and security arrangements between Northern and Southern Sudan, but also for a referendum on secession in the South. Predictably, in January 2011 Southern Sudan voted overwhelmingly to become an independent country in its own right. Yet, in addition and despite their internal heterogeneity and ethnic conflict, the two Sudanese successor states have clashed violently over boundary demarcations, oil revenues, and new citizenship arrangements. Consequently, federalism and decentralization will remain indispensable to any constitutional proposals for their stabilization.

Decentralized Unitary States

To reiterate, some of the more ambitious experiments in political decentralization in Africa since the 1990s have taken place in countries with a previous, often contentious, history of experimentation with federation. Uganda’s 1962 Independence Constitution, for instance, provided for the devolution of powers to local councils and four historic kingdoms, including the Buganda kingdom, which enjoyed a special, asymmetric federal status, and where enthusiasm for federo (federalism) remains strong to this day. The dismantling of this quasi-federal structure in 1966 under the authoritarian rule of Milton Obote marked the beginning of an extended process of state failure and implosion in Uganda. Under the charismatic leadership of Yoweri Museveni, the National Resistance Movement (NRM) progressively seized control of Uganda in the mid-1980s, establishing ‘resistance councils in the areas under their control in order to mobilize and politicize the masses’ (Steiner 2008: 42). Since its official legalization in 1987, the resistance council system has developed into an elaborate five-tier devolutionary structure, with elected councils and committees at the village, parish, sub-county, county, and district levels. The system involves the extensive transfer of planning, legislative, and executive powers to elected sub-national authorities on the basis of the subsidiarity principle. Coexisting with this comprehensive devolution scheme is a deconcentration element, including the office of the district commissioner, a presidential appointee who supervises the local implementation of functions (defence, security, and so on) that are not devolved.

Decentralization in Kenya also has been implemented against a background of initial experimentation with, and continuing debates about, federalism. The country came into independence in 1963 on the basis of a constitution that largely enshrined the demands of many of Kenya’s non-Kikuyu smaller ethnic communities for ‘majimboism, or regionalism’ (Ndegwa 1997: 605). The key federalist features of this constitution included a bicameral national parliament, the decentralization of powers and resources to relatively strong regional governments with their own legislatures and executives, and a requirement for the approval of any changes to regional powers by super-majorities in each house of the bicameral national parliament or in a national referendum. Within a year of Kenya’s independence, however, the ruling Kenya African National Union (KANU) dismantled the constitutional architecture of regional autonomy and transformed the country into a unitary, one-party state. Kenya’s transition to multi-party politics in the 1990s, however, revived the majimbo debate.

Following prolonged struggles and debates over political reform, a new Constitution for Kenya was approved in a referendum and promulgated in August 2010. Aside from enshrining a separation of powers between the executive and the legislature, the 2010 Kenya Constitution provided for the devolution of power to county (district) levels of government and reintroduced a bicameral federal parliament, with each county being represented by a Senator in the upper house of the parliament. Each of the country’s 47 counties will also have a county executive headed by a county governor and a county assembly made up of representatives of wards within the county. The counties will have powers over matters of local concern as well as enjoy guaranteed revenues based on proposals of a newly established Commission on Revenue Allocation. Although its decentralist provisions were less far-reaching than those of the 1963 Majimbo Constitution, which had devolved powers to the much larger regions/provinces, the 2010 Kenyan Constitution represents a remarkably ambitious and rigorous experiment in democratic decentralization within an otherwise unitary polity in Africa.

In the DRC, a formally decentralized system has emerged as a compromise solution to the political struggles between proponents of federalism and unitarism. The basic law that ushered the DRC into independence in 1960 had significant federal features, including the grant of autonomous legislative and executive powers to the country’s original six provinces. Changes to this basic law in the immediate post-independence period – including the expansion of the number of provinces to 21, in order to enhance the ethnic homogeneity of these units – consolidated the federal character of the Congolese state. Yet, this period was defined by catastrophic ethnic instability and political chaos, including the mutiny of the national army, the assassination of the country’s first prime minister, and the emergence of a violent ethnosecessionist movement in Katanga Province. This political disarray led to the takeover of power by army commander Joseph Mobutu, who imposed a centralized one-party dictatorship. The implosion of Mobutu’s predatory government in the wake of the introduction of multi-party competition in the early 1990s plunged the DRC into an extended period of violent disintegrative conflict, which officially ended with an internationally mediated peace and political transition process during 2002–2006.

A highlight of the Congolese transition was the making of the 2006 Constitution, the decentralizing features of which represent a ‘radical break with Congo’s long history of pseudo-centralized rule’ (Tull 2010: 653). The new Constitution, for instance, mandates the expansion of the number of provinces from the existing 11 to 26, provides for the division of powers and revenues between the provinces and central government, establishes a Senate and a Conference of Governors to ‘give voice to the provinces’, and creates a constitutional court to arbitrate centre-provincial conflicts. The revenue-sharing provisions of the Constitution are especially remarkable, providing for the retention by the provinces of 40 per cent of fiscal revenues generated from their territory, while establishing an equalization fund that would allocate up to 10 per cent of national finances to developmental projects in poorer provinces. Yet, reflecting the troubled state of democratization and post-conflict reconstruction in the DRC, the fiscal provisions of the 2006 Constitution and the envisaged territorial reconfiguration of the country into 26 provinces have not been implemented.

Procrastination and ambivalence toward political decentralization has been a characteristic political strategy in several other African states, including the previously federal country of Cameroon. Established in 1961 following the reunification of former British and French trusteeship territories in the old German colony of Kamerun, the Cameroon federation was riddled with multiple imperfections and contradictions from birth. These included: the existence of only two federal sub-units, one of which (Francophone East Cameroon) was vastly superior in population and influence to the other (Anglophone West Cameroon); failure to specify a clear division of powers and resources between the central and sub-national orders of government; the non-establishment of a federal legislative chamber, thereby denuding the federal arrangement of a key mechanism for representing regional interests in the national government; extensive federal intervention in the territorial administration of the federal sub-units through presidential appointees, like the federal inspector and the district officers; and the imposition of unitary, single-party rule in 1968. In essence, the formal abrogation of the federation following a referendum in 1972 merely put an end to an experiment that was more ‘shadow than reality’ (Stark 1976).

Cameroonian politics continues to be vexed by the Anglophone agitation for the territorial reform of the francophone-dominated centralized unitary state. While denouncing federation as a recipe for separatism, the Cameroonian central authorities have formally and explicitly embraced the transformation of the country into a decentralized unitary state. Constitutional reforms introduced in 1996, for instance, not only instituted the principle of political autonomy for two decentralized tiers of sub-national government (the commune councils and the regional council), but also introduced a 100-member Senate as a quasi-federalist second chamber of the national parliament, with 10 senators representing local governments in each of the country’s 10 regions. However, the Senate and the regional councils were not established.

Despite its discouraging record in countries such as the DRC and Cameroon, decentralization has been embraced in several African countries that have emerged from a centralist Francophone or Lusophone administrative colonial tradition, including Mali, Senegal, and Mozambique. Malian decentralization has, for instance, been described as the ‘heart’ of the country’s democracy. Originally initiated in 1992 to placate separatist Tuareg groups in the north of the country, decentralization in Mali has sought ambitiously to develop a ‘local-level democratic vitality’ through the establishment of elected and representative councils (with administrative and financial authority over basic health, education, and infrastructure) at the levels of commune, the circle, and the region, as well as the creation of a High Council of Local Government as a form of second national legislative house that represents commune governments at the national level (Pringle 2006). However, the resurgence of Tuareg separatism in 2012 (following an influx of arms from the 2011 Libyan conflict) and a subsequent military coup highlights the unstable regional context of multi-level governance in Africa, as well as the deficits of decentralization in Mali, where sub-national autonomy is seriously constrained by ‘a limited fiscal base and the practice of central state tutelage’ (Dickovick and Riedl 2010: 67).

In Senegal, decentralization has produced a complex, multi-level political structure. This includes regional governments with elected councils, and a centrally appointed administrative governor; elected urban and rural localities (communes and communautes rurales); and centrally appointed departments and arrondissements with oversight powers over local governments. Decentralization laws in 1996 introduced elections for the regions, established new revenue transfers for regional and local authorities, and expanded sub-national responsibility for the delivery of nine basic social services, including health and education (Dickovick 2005: 188). In Mozambique, the creation in the 1990s of politically autonomous municipalities (autarquias), with elected mayors and municipal assemblies, has significantly opened up an otherwise centralized dominant party state. The creation of the autarquias has produced substantial improvements in service delivery, encouraged the development of sub-national revenue autonomy and fiscal capacity, and enabled opposition control of power and challenges to the hegemony of the ruling Frente de Libertação de Moçambique (FRELIMO, Liberation Front of Mozambique) party that are not feasible at the national level (Linder 2009).

While acknowledging the ‘bourgeoning trend of decentralization in Africa’, however, the scholarly literature has documented several interrelated shortcomings of this process that echo many of the flaws that also afflict the continent’s fully fledged federations (Dickovick 2005: 184). These shortcomings include:

Overall, recent achievements in the establishment of new legal frameworks for devolution in Africa have not been matched by de facto advances in decentralization. Rather, sub-national fiscal autonomy, developmental capacity, public service provision, and democratic political accountability have lagged behind the formal institutionalization of decentralized legal authority (Dickovick and Riedl 2010: 1–6). As multifaceted and complicated as these shortcomings of decentralization may appear, they pale in comparison to the enormity of the impediments that afflict experiments at promoting inter-governmental relations at the international level in Africa (see Khadiagala, this volume).

Conclusion

Federalist governance principles have been extensively, if often grudgingly and unsuccessfully, invoked and implemented in Africa. These principles are a fundamental feature of the architecture of governance in some of the continent’s larger and more ethnically complex states, including Nigeria, South Africa, and Ethiopia. While they have abandoned or resisted fully fledged federal constitutions, most of the continent’s multi-ethnic states have used federal principles of self-rule and shared rule for the decentralization of their unitary governmental systems. These same principles have provided the basis for organizing the continent’s evolving supranational institutions that are designed to meet collective governance challenges at the international (regional and sub-regional) levels.

Yet, there can be little doubt that the record of federalism in Africa has been uninspiring. Indeed, a prime contradiction of African politics involves the obvious imperatives and relevance of federalism as a governance strategy on the continent on the one hand, and the absence of propitious conditions for the effective working of federal institutions and processes on the other. Most African states lack the fiscal resources, administrative capacities, and political cultures necessary to make federalism and decentralization flourish as instruments of democratic consolidation, political stability and unity, and economic development. Thus, Africa’s federations and decentralized unitary states have been too centralized to be authentic and successful experiments in self-rule and shared rule. The continent’s supranational or international integration schemes, on the other hand, have been too decentralized, fissiparous, or incoherent to make for the effective and constructive development and deployment of collective solutions to continental and sub-regional governance challenges.

All of this is not to understate the positive contributions of federalism to African governance and politics. The adoption or reconfiguration of federalist institutions has contributed to the relative stabilization of post-civil war Nigeria and Ethiopia, the consolidation of multi-racial democracy in South Africa, and the durability of the union of Zanzibar and mainland Tanzania. Decentralization has aided post-conflict reconstruction in Uganda, restrained political authoritarianism in Mozambique, and spurred governance reform in Kenya, among other countries. Furthermore, it is difficult to envisage any viable political alternative to democratic federalism and/or decentralization in conflicted states like the two Sudans and the DRC. Federal strategies and theories for designing intergovernmental relations will remain important for developing and improving Africa’s nascent institutions for supranational governance. For these reasons, federal ideas, principles, and options will continue to have tremendous appeal, relevance, and resonance in Africa.

Bibliography

Adamolekun, L. and Kincaid, J. (1991) ‘The Federal Solution: An Assessment and Prognosis for Nigeria and Africa’, Publius: The Journal of Federalism 21(4): 173–188.

Anderson, G. (2008) Federalism: An Introduction, Oxford: Oxford University Press.

Boone, C. (2003) ‘Decentralization as Political Strategy in West Africa’, Comparative Political Studies 36(4): 355–380.

Chanie, P. (2007) ‘Clientelism and Ethiopia’s post-1991 Decentralization’, Journal of Modern African Studies 45(3): 355–384.

Diamond, L. (1999) Developing Democracy: Toward Consolidation, Baltimore, CO, and London: Johns Hopkins University Press.

Dickovick, T. (2005) ‘The Measure and Mismeasure of Decentralization: Subnational Autonomy in Senegal and South Africa’, Journal of Modern African Studies 43(2): 183–210.

Dickovick, T. and Riedl, R. (2010) Comparative Assessment of Decentralization in Africa: Final Report and Summary of Findings, Washington, DC: USAID.

Fiseha, A. and Habib, M. (2010) ‘Ethiopia’, in Luis Moreno and Cesar Colino (eds) Diversity and Unity in Federal Countries, Montreal and Kingston: McGill-Queen’s University Press.

Horowitz, D. (1985) Ethnic Groups in Conflict, Berkeley, CA: University of California Press.

Kymlicka, W. (2006) ‘Emerging Western Models of Multination Federalism: Are They Relevant for Africa?’ in D. Turton (ed.) Ethnic Federalism: The Ethiopian Experience in Comparative Perspective, Oxford: James Currey.

Linder, W. (2009) ‘On the Merits of Decentralization in New Democracies’, Publius: The Journal of Federalism 40(1): 1–30.

Ndegwa, S. (1997) ‘Citizenship and Ethnicity: An Examination of Two Transition Moments in Kenyan Politics’, American Political Science Review 91(3): 599–616.

Pringle, R. (2006) Democratization in Mali: Putting History to Work, Washington, DC: United States Institute of Peace.

Simeon, R. and Murray, C. (2004) ‘Multi-Level Governance in South Africa’, in B. Berman, D. Eyoh and W. Kymlicka (eds) Ethnicity and Democracy in Africa, Oxford: James Currey.

Stark, F. (1976) ‘Federalism in Cameroon: The Shadow and the Reality’, Canadian Journal of African Studies 10(3): 423–442.

Steiner, S. (2008) ‘Constraints on the Implementation of Decentralization and Implications for Poverty Reduction: The Case of Uganda’, in G. Crawford and C. Hartmann (eds) Decentralization in Africa: A Pathway out of Poverty and Conflict? Amsterdam: Amsterdam University Press.

Stepan, A. (1999) ‘Federalism and Democracy: Beyond the US Model’, Journal of Democracy 10(4): 19–34.

Stren, R. and Eyoh, D. (2007) ‘Decentralization and Urban Development in West Africa’, in D. Eyoh and R. Stren (eds) Decentralization and the Politics of Urban Development in West Africa, Washington, DC: Woodrow Wilson International Center for Scholars.

Suberu, R. (2009) ‘Federalism in Africa: The Nigerian Experience in Comparative Perspective’, Ethnopolitics 8(1): 67–86.

Tull, D. (2010) ‘Troubled State-Building in the DR Congo: The Challenge from the Margins’, Journal of Modern African Studies 48(4): 643–661.

USAID (United States Agency for International Development) (2009) Democratic Decentralization Programming Handbook, Washington, DC: USAID.

Watts, R. (2008) Comparing Federal Systems, Montreal and Kingston: McGill-Queen’s University Press.