Introduction
Over the past three decades, the number and diversity of non-state providers (NSPs) of social welfare has expanded tremendously across most of Africa. From Benin to Zimbabwe, a range of non-state actors – including secular charitable non-governmental organisations (NGOs) such as CARE, faith-based NGOs such as World Vision, for-profit health care clinics and informal extended family networks, to name a few – provide access to and/or supply basic social services to ordinary people even more extensively than states.1 While the existence of non-state actors in social welfare provision is not new in Africa, their numbers have grown substantially in recent years. Moreover, the predominant patterns of non-state provision vary in puzzling ways.
This article seeks to answer two questions. First, why has the number of non-state providers increased so markedly during this contemporary time period? And, second, why do the predominant types of non-state providers vary in different African countries? I argue first that the neoliberal democratisation during the 1980s and 1990s created new opportunities and spaces for non-state providers. This explains the remarkable growth in the numbers of non-state providers across the continent, but not the cross-national variation in the patterns of non-state provision.
Hence, the second part of the argument is that colonial legacies mediate the numbers and types of non-state actors on the ground, and the article focuses on the political historical context of several cases in West Africa. Various colonial administrators historically conceptualised the respective roles of the state and non-state actors in the provision of social welfare in distinct ways. Thus, where the colonial state was more decentralised, for example in the British colonies in West Africa, the activities of a greater number and diversity of non-state providers in social welfare delivery were encouraged. In contrast, where the colonial state was more centralised, for example in the French colonies in West Africa and the Americo-Liberian colony of Liberia, efforts to replace precolonial non-state actors made less room for other faith-based or entrepreneurial non-state providers.2 These colonial-era differences were reproduced in the immediate post-independence era and in the contemporary era of neoliberal democratisation owing to respective differences in the normative orientation of post-colonial elites as well as relative capacities of the state and non-state actors established in the earlier part of the twentieth century.
This article’s analysis highlights that social provision in Africa is not simply an ahistorical policy diffusion of state spending and provision. Instead, colonial history mediates the important and expanding role of different non-state actors. Understanding the political history of these dynamics of non-state provision sheds light on its political consequences. While advocates highlight how non-state provision generates the possibility of new opportunities of access, fresh avenues of accountability for consumers and seamless complementarity with the state, I conclude that non-state provision produces new inequalities of access, complex barriers to accountability for citizens and the long-term undermining of state capacity.
In the next section, I review the existing theoretical explanations for the rise of different forms of non-state social protection in Africa as a whole. In the third section, I discuss the data and methods used to investigate these two questions. In section four, I demonstrate the increase and variation in the types of non-state actors over time on the continent. In the fifth section, I develop the arguments for why non-state provision has grown so rapidly in the past three decades, and why it varies across the continent in particular ways in West Africa. Finally, I conclude that this growth and heterogeneity of non-state providers has important consequences for citizens’ ability to obtain equitable access and accountability for services.
Theoretical explanations
Explaining the rise in the number of non-state providers
The dominant explanation for the rise in the number of NSPs in much of the economics literature is that non-state provision emerges from market failure in developing countries with endemic poverty and low levels of economic development (Alderman and Paxson 1992; Stiglitz 2005). According to this view, private sector firms have little profit motivation to provide social services such as education and health in areas where consumer demand and ability to pay are weak, and the supporting infrastructure of roads and electricity is very poor. The implication is that widespread market failure results in the emergence of not-for-profit non-state actors stepping in to deliver and/or finance needed social services.
But, market failure is not ubiquitous on the continent (Farzad 2007), or everywhere the same, and yet, NSPs are growing across Africa. For example, for-profit companies are enrolling growing numbers of students in private primary schools in both South Africa and Malawi (Rose 2006). Furthermore, even as some private sector markets have grown to meet the demands of an expanding middle class in such countries as Botswana, Kenya or Ghana, a host of other not-for-profit non-state providers persist and continue to grow.
An alternative perspective holds that the core problem is not the failure of markets, but rather the failure of the state (Acemoglu and Robinson 2012; Bates 2008; Gough and Wood 2004). While the media often portray Africa as a continent full of failed states, state failure is actually rather rare. Far more common is the persistence of state weakness (Jackson and Rosberg 1982). This article demonstrates that state weakness is a necessary condition for the tremendous growth in NSPs across Africa but does not help us understand the timing of this growth in the early 1990s.
Instead, I argue that the dominance of the neoliberal paradigm of development, especially hegemonic among the relatively impoverished and weak states in Africa from the 1980s, combined with a second factor, democratisation beginning in the 1990s, to stimulate the accelerated growth of non-state social protection over the past three decades. Beginning in the 1980s, donors pushed states to adopt neoliberal reforms to reduce what were viewed as excessive interventions in the market (World Bank 1981) and to delegate their roles to lower levels of administration (Grindle 2007). Even when only partially implemented (van de Walle 2001), neoliberal reforms in the 1980s and 1990s in the social welfare sector initially meant the elimination or drastic reduction of prior state subsidies for social services and the introduction of user fees to achieve partial cost recovery. This literature suggests that the overall reduction of the state’s role in the economy opened new spaces for non-state actors at the local level (Tripp 1997).
This first dynamic of neoliberal economic globalisation, initiated in the 1980s, was then reinforced when many African countries began to liberalise their authoritarian political regimes in the early 1990s (van de Walle 2001). Democratic openings not only created new popular pressures for the state to expand public programmes for social protection, but also created opportunities for non-state actors to emerge, operate and grow. By the early 2000s, many African states had softened their earlier structural adjustment programmes by making basic health and education either free or more affordable to all (MacLean 2012). As a result of neoliberal and democratic reforms, African states may now be expanding social services, but they are also competing with a tremendous number of NSPs to provide these services to their citizens.
Explaining the different types of NSPs
While the combination of these pro-market and democratic reforms explains the exceptional increases in the quantitative level of non-state provision in recent decades, it does not explain the qualitative variation across different countries in Africa. The analytic framework developed by Cammett and MacLean (2014) suggests that different levels of state capacity combine with various levels of NSP capacity to produce four different modes of NSP–state relations: co-production, state domination, substitution and appropriation. These four modes range from more positive and collaborative relationships to more antagonistic and competitive interactions.
Applying this framework to the analysis of variation within Africa has limited value owing to the overwhelming prevalence of weak states on the continent (see Table 1). Only a very few African countries, such as Botswana and South Africa, exhibit high state capacity for social welfare provision. In Botswana, where a strong state bureaucracy tightly controls the activities of a relatively weak sector of NGOs, community-based organisations (CBOs) and businesses, the dynamic is characterised as one of state domination. For example, the Botswana government has notably recentralised the authority from community-based organisations for natural resource management (Poteete and Ribot 2011). In certain provinces and municipalities of South Africa, state and NSP capacity are both high, and the resulting pattern spans from co-production to delegation. This mode of NSP–state relations is undoubtedly the most infrequent across Africa.
| State capacity to deliver and regulate social welfare | |||
|---|---|---|---|
| Low | High | ||
| NSP capacity to finance and deliver social welfare | Low | Appropriation (NSPs control access to state resources through brokerage, patronage, credit-claiming) Examples: Ghana; Côte d’Ivoire (until 2002); Angola (until 2002) | State domination (State control over financing and delivery of services) Example: Botswana |
| High | Substitution (NSPs take over when state does not perform or provide) Examples: Côte d’Ivoire (2010s–present); Liberia (2000s–present); Tanzania; Kenya; Angola (2002–present); Sierra Leone | Coproduction (Joint financing and delivery of services by state and NSPs) Delegation (Authority granted to NSPs to finance and/or deliver services) Example: South Africa |
The great majority of African states have weak capacity and thus NSP–state relations are characterised as either appropriation or substitution. When state capacity and NSP capacity are both low, a relationship of appropriation emerges, where NSPs control access to state resources by serving as brokers for state-provided benefits or by distributing public services as a form of patronage. Appropriation is probably the most frequent mode of NSP–state relations across Africa. For example, in Ghana, through informal social networks of reciprocity, an uncle may finance the tuition, fees and school supplies for his nephew to attend secondary school. Or, in Côte d’Ivoire, a hometown association of civil servants based in Abidjan organises and collects membership dues to construct a cafeteria for the public primary school in the natal village.
The fourth and final pattern of substitution is produced when state capacity is low, but NSP capacity is high. Here, NSPs are able and willing to take over social welfare functions from the state when it is not performing or to provide services that the state never provided in the first place. For example, in Kenya, NGOs often provide an increasing number of social welfare services that the state lacks the infrastructure to supply (Brass 2014, 2016). Substitution might also describe countries recovering from civil war, such as Angola or Liberia, where donors fund international NGOs to rebuild basic infrastructure.
This article’s comparison of Ghana, Côte d’Ivoire and Liberia in the region of West Africa thus makes an important contribution by attempting to disentangle the sources of variation in Africa’s weak-capacity states. This article explores whether important differences in colonial rule influence norms and shape the construction of state and non-state social welfare regimes. Where some scholars contend that colonial states had a shallow impact in the rural hinterlands in Africa (Herbst 2000), others argue that different patterns of colonial rule have long-term consequences for development on the continent (Lange 2009; Mamdani 1996).
The article breaks new ground in exploring how different experiences of democratic consolidation may interact with colonial legacies to shape the types of NSPs on the ground. Several scholars have shown that the wave of democratic transitions has sometimes faltered or been reversed in Africa (Bratton and van de Walle 1997). While much of the literature highlights the critical role of a vibrant civil society in promoting democratic consolidation (Gyimah-Boadi 1996), this article examines the reverse causal linkages, investigating how democratic consolidation may feed back, shaping the character of civil society and hence the different patterns of NSPs.
Research design, data and methodology
Very little cross-national quantitative data documents the extent of non-state providers across Africa, and the existing sources are limited; for example, restricted to only one social service, focused on one type of non-state provider or excluding theoretically important differences among African countries (e.g. Johns Hopkins data on civil society exclude francophone African countries). To conduct a more holistic analysis, I triangulate among a variety of quantitative and qualitative sources to uncover historical shifts and potential cross-national trends from the precolonial era to the present.
I then investigate the causal processes of the contemporary period from the early 1980s to the present in more depth with comparative case studies of three neighbouring countries in West Africa: Ghana, Côte d’Ivoire and Liberia. These countries share many similarities including: geography; weak state capacity; economic development; adoption of neoliberal economic reform in the early 1980s; and an attempt to transition to multiparty democracy in the 1990s. The three case studies also share similar ethnic and religious demographics (see Table 2). The analysis and interpretation of these data is deepened by previous fieldwork in Ghana, Côte d’Ivoire, Kenya and Uganda at various points from the late 1990s to 2016.
| Shared characteristics | Explanation for rise in numbers of NSPs | Explanation for variation in pattern of NSPs | |||||
|---|---|---|---|---|---|---|---|
| Geography | Level of economic development | State capacity | Neoliberal economic reform | Democratic transition to multiparty system | Colonial legacy | Democratic consolidation (Freedom House score in 2015) | |
| Ghana | Small territory on West African coast | Developing (lower middle income) | Weak | Partial implementation since early 1980s | 1992 | Decentralised British | Free |
| Côte d’Ivoire | Small territory on West African coast | Developing (lower middle income) | Weak | Partial implementation since early 1980s | 1990 | Centralised French | Partly free |
| Liberia | Small territory on West African coast | Developing (low income) | Very weak | Partial implementation since early 1980s | 1997 | Centralised American repatriated slaves | Partly free |
The expansion of different non-state providers of social welfare in Africa
This article conceptualises the expansion of non-state provision along two dimensions: (1) the quantitative increase in the number of non-state actors over time; and (2) the qualitative types of non-state providers. Where much of the existing literature on non-state actors focuses almost exclusively on NGOs (Mercer 1999), this article evaluates the prevalence of seven different types of NSPs including: for-profit firms, secular NGOs, ethnic and sectarian organisations, faith-based organisations, community-based organisations, informal brokers, and family and friendship networks.
Small numbers of locally organised welfare providers across precolonial Africa
Historically, in many parts of sub-Saharan Africa, what we conceptualise as non-state actors today have long supplied basic social welfare services to people at the local level. These locally organised welfare providers often have deep historical roots in the region and even predate the establishment of modern nation-states in Africa by many centuries. Like in Europe, families, village communities, religious groups and work associations self-organised to help each other in times of hardship and need (Berry 1993; Dei 1992). These precolonial providers were locally organised and grounded in the everyday life of the village, where politics, religion and communal values were frequently fused.
In contrast to the mythic image of timeless, homogeneous and completely autonomous reciprocity institutions, these local providers varied in organisational structure and were often constituted by the precolonial political systems. For example, in many decentralised, precolonial polities in Africa, sophisticated age-grade societies (Bernardi 1985), such as the Poro society in what is now Guinea, Liberia and Sierra Leone, and the Bwame in what is today the Democratic Republic of Congo, provided education and skills to generational cohorts of young men and sometimes women (Simiyu 1988). Meanwhile, in more centralised precolonial systems, such as the Ashanti kingdom in today’s Ghana, the village chief resolved and coordinated social welfare needs at the level of the village; in rare cases, issues might be referred upward in the hierarchy to the paramount chief or further to the Asantehene. The ‘Kafo’, found in the Mandinka areas of West Africa and dating back to the Ghana Empire from the late eighth century, are an example of a multi-village indigenous institution created to handle larger-scale social problems that exceeded the capacity of a single village chieftancy or community.
With the spread of Islam to Africa beginning in the seventh century, new types of welfare providers entered the scene; in particular, explicitly faith-based organisations. Koranic schools developed by the ninth and tenth centuries in East Africa, and the eleventh century in West Africa (Robinson 2004). As early as the 1500s, the international passage of people and goods along the trans-Saharan trading routes stimulated the development of three universities at the crossroads in Timbuktu, Mali. Islamic religious leaders or imams taught over 25,000 students from different parts of Africa, Europe and the Middle East at the Sankore Mosque or University of Timbuktu (Bleck 2011).
In sum, although precise and comparable quantitative figures are impossible to obtain for the precolonial era, we can tentatively conclude that because welfare provision in this period was more locally organised, the numbers of providers were smaller and growing relatively slowly over time. Furthermore, it appears the predominant types of welfare provision were community-based or faith-based organisations and varied according to the local political and religious structures.
Moderate numbers and increasing hetereogeneity of non-state providers with imperial and colonial rule
Centuries of exploitation by Europeans, beginning with the slave trade in the fifteenth century, and later colonial rule from the nineteenth century, disrupted indigenous African state formation and economic development. Central states were thus formed and democratised much later in Africa than in Europe, and the majority of African economies remain primarily agricultural.
The first sustained contact with Europeans took place with maritime exploration and the growth of the slave trade in the late fifteenth century. By the early 1480s, the Portuguese began building Elmina Castle along the coast of what is now Ghana. As early as the mid sixteenth century, European trading companies created ‘Castle schools’ to educate a small number of children of African merchants and local chiefs at the slave forts along the coasts in Ghana (George 1976). The company headquarters, based in Europe, provided only limited financing of these schools, so local fund-raising was often important. At this point, the non-state welfare provision by Europeans was so limited in number and so concentrated in just a few coastal settlements that there were no discernible distinctions in the types of non-state providers across African cases. The existing, precolonial community organisations and family networks remained dominant, with faith-based schooling gaining particular importance in parts of East Africa and the Sahel where Islam had gained ground.
Following this initial commercial opening, various European missionary groups began to found schools and health care clinics in conjunction with their evangelical work. But, the numbers, locations and operating styles of these missionary groups varied and were later incorporated in divergent ways by the respective colonial powers. In Ghana, Wesleyan (Methodist) and Basel (Presbyterian) missionaries arrived as early as 1828 and had built several schools by the mid nineteenth century. The missionary schools in Ghana were primarily supported by the missions and received only limited assistance later by the British colonial government.
In contrast, French Catholic missionaries were the first Europeans to arrive in Côte d’Ivoire, as early as 1637, but their presence was always much more limited than the missionary activity in their anglophone neighbours. Islamic schools also continued to grow in many parts of Africa. But French colonial officials tended to restrict the expansion of Islamic schools whereas British colonial officials were more open to these developments. For example, by 1907, nearly 3000 Koranic schools had been founded in present-day Guinea, and by 1931, approximately 30,000 Islamic schools enrolled between 200,000 and 360,000 students in Nigeria (Johnson 1975). Owing to the overall greater numbers, the faith-based schools and health clinics in the anglophone colonies began to reach further into the hinterland than in the francophone colonies, where non-state provision continued to be more concentrated in the largest urban areas and coastal capitals.
By the late nineteenth century, European industrialisation spurred new legislation to expand social welfare for European workers in the metropole while also stimulating the entrenchment of colonial administration and the drive to increase the extraction of necessary raw materials throughout the African territories (Austen 1987). In many of the African colonies in the 1880s, European powers slowly built central government-run schools and clinics to begin delivering social services directly. Nevertheless, the British and French constructed public provision in different ways. In general, the British colonial rulers more frequently continued to work with mission schools and clinics than the French, who more actively sought to replace them (MacLean 2003). For example, in Ghana, by 1931, only 19 of the total of 347 schools were completely government-run. In contrast, the French used direct taxation to provide full financial support for free services in public schools and clinics in Côte d’Ivoire (MacLean 2014). In Mali, the French, suspicious of the local Koranic schools, built their own franco-arabe schools and attempted to co-opt or eliminate their rivals (Bleck 2011).
By the end of the colonial era in the middle of the twentieth century, non-state providers were growing more quickly and delivered a significant share of the education and health services to African subjects. During this period of imperial and colonial rule, NSPs had become more diversified, with for-profit firms beginning to build schools and clinics, and internationally organised, with faith-based organisations often initiated by missionaries from overseas.
Contemporary explosion of different types of non-state social welfare providers
While non-state actors are not new in Africa, their numbers and diversity have exploded over the past three decades (see Table 3). Even though the growth in the numbers of NSPs seems to have occurred all over the continent, different patterns of the types of NSPs are highlighted below in various African countries.
| NSP type | Indicator | Ghana | Côte d’Ivoire | Liberia | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Year | 1990s | 2000s | 2010s | 1990s | 2000s | 2010s | 1990s | 2000s | 2010s | |
| INGOs | Number of international NGOs | 654 (2002) | 680 (2012) | 559 (2002) | 559 (2012) | 261 (2002) | 212 (2012) | |||
| NGOs | Number of NGOs | 80 (1980) | 5714 (2012) | 1347 (2012) | ||||||
| Number of development NGOs | 282 (2010) | 111 (2010) | 87 (2010) | |||||||
| For-profit firms | % of primary school enrolment in private schools/total primary enrolment | 10 | 18 | 22 | 10 | 11 | 12 | * | * | * |
| Extended family and friendship networks | Remittances received (US$) | 7.3 million | 43.5 million | 2 billion (2014) | 49.5 million (1992) | 120 million (2002) | 384 million (2013) | * | 58.4 million (2004) | 495 million (2014) |
Quantitative data from Africa on the number of NSPs over time are difficult to obtain, even for this more contemporary time period. Of all NSP types, international NGOs (INGOs) are most likely to be monitored on a global scale. The global aggregate data available show that the number of INGOs around the world has grown exponentially over time, declining only slightly during the Great Depression in the 1930s (Davies 2008). The aggregate data for Africa from the Union of International Associations (UIA 2013) show a remarkable 25% increase in the number of international NGOs in Africa.
Country-level data suggest that domestic NGOs were growing in number and spreading outside of capital cities in Africa. According to the Johns Hopkins Center for Civil Society Studies, a significant percentage of the economically active population in four sub-Saharan African countries have been working as paid staff or volunteers in the civil society sector between 1995 and 2000 (Salamon, Sokolowski, and Associates 2004). Brass similarly points to the rapid increase in the numbers and activities of local NGOs in Kenya (Brass 2014). Data from Ghana reveal that NGOs expanded from just 80 in 1980 to 4772 in 2010 (USAID 2010).
This growth in NGOs seems more pronounced in formerly anglophone colonies than francophone ones, however. Thus, when focusing on a subset of NGOs working in development, Côte d’Ivoire has only 111 organisations listed whereas Ghana had 282 organisations in 2010 (Wesselink 2014, see also UIA 2013). This trend is reinforced when comparing two countries considered to be regional hubs for development: Kenya, with 797 development NGOs, and Senegal, with 323 development NGOs.
For-profit firms as non-state providers of social welfare also appear to have increased over the past several decades around the world and in Africa. World Bank data indicate that private primary school enrolments have increased at a remarkably accelerated rate across all world regions (except in OECD member states) between 2000 and 2010. In sub-Saharan Africa, the percentage increase of private primary school enrolments during the 1990s exceeded 100%. The increased activity is not only from local for-profit firms, but also from large multinational corporations engaging in corporate social responsibility initiatives. For-profit providers are also mushrooming in health care. For example, Katusiimeh (2015) finds that increasing numbers of Ugandans obtain health care in for-profit clinics in and around Kampala.
But, again, important patterns emerge between anglophone and francophone countries. Where private primary school enrolment grew rapidly in Ghana from 10% in 1992 to 22% in 2002, it grew in Côte d’Ivoire only slightly, from 10% to 12% (see Table 3).
Although little quantitative data exists, the increasing role of ethnic and sectarian organisations as well as faith-based organisations has been heralded in Africa. Scholars working in Ghana, Nigeria and elsewhere around the continent have noted the expanding number and activities of ethnic hometown associations in enhancing community infrastructure for social welfare (Honey and Okafor 1998; Mercer, Page, and Evans 2008). Weiss (2002) has highlighted the important role in social service delivery of Muslim organisations, such as the Muslim Brotherhood in Egypt, while Bleck (2011) has examined the growing role of madrassa schools in Mali. Aremu finds a remarkable increase in the number of private Christian and Islamic universities in Nigeria since 1999, where private, mostly faith-based, universities grew rapidly from only 7.3% to over 39% of all universities in 2013 (Aremu 2014).
Current cross-national data on less formal types of NSPs are even more difficult to evaluate. International remittances are evidence of the growing importance of international family and friendship networks (Manuh 2006). World Bank data suggest that remittances received in Ghana increased from $500,000 in 1980 to over $2 billion in 2014 (World Bank 2010) (see Table 3). In Côte d’Ivoire, remittances increased from $49.5 million in 1992 to over $384 million in 2013. Similarly, remittances grew in Liberia from over $58.4 million in 2004 to $495 million in 2014. These dramatic rates of growth in Côte d’Ivoire and Liberia might suggest that remittance transfers were intensified by the recent experience of conflict.
The above conclusion that the role of non-state welfare providers is becoming increasingly important over time in Africa does not mean that the state is absent or declining. The reduction in public welfare provision over recent decades has been accompanied in some countries by a net increase in public social expenditures. Public health spending has increased as a percentage of GDP in Africa in the past decade (World Bank 2012). During the 1990s, public education spending in Africa first fell below 1980 levels with the austerity of structural adjustment programmes but then rebounded to a higher level by the late 2000s (Ibid.).
Yet, these data on state social expenditures provide a misleading view of the operation of African social welfare systems. These increases in state social expenditures have often occurred simultaneously with the donor call for non-state actors to take on more of the operational burden of supplying the basic social welfare services and infrastructure. Furthermore, central state money is not enough. Strong state regulatory capacity is essential to ensure that both state and non-state providers offer high quality services in an equitable and accountable way to Africans (Hill et al. 2012). In interviews in 2012, 2014 and 2016, parents in Ghana or Kenya complained that the expansion of free primary education by the state had resulted in such overcrowded classrooms that the quality of instruction had declined dramatically. With sometimes over 80 or 90 students crammed into a single classroom, even parents with little means exclaimed that they would do everything to avoid those public services, including paying significantly higher tuition and fees at a private school.
Explaining the extent and patterns of non-state provision within Africa: how colonial legacy mediates neoliberal democratisation
Non-state provision has increased remarkably all over Africa, particularly since the 1990s, but the patterns of non-state provision vary significantly across different African countries. The above descriptive analysis of the extent of NSPs in Africa over time illuminates the importance of a critical juncture in the early 1990s. Certainly, NSPs have been active players in Africa since the precolonial era, but their numbers were relatively small, and continued to grow moderately until the 1990s. All of the evidence suggests that African countries experienced a veritable explosion of NSPs beginning in the 1990s that had not been witnessed in earlier eras.
This data analysis highlights the potential explanatory power of factors that changed significantly across most African countries by the late 1980s and early 1990s. Whereas geography, state capacity and the level of economic development remained constant for most African cases during this time period, two hypotheses from the literature remain relevant and will be further explored in the case study comparison of three countries in West Africa: the adoption of neoliberal economic reforms and a democratic transition.
Notably, non-state provision has become more heterogeneous beginning in the colonial period, with distinct patterns emerging between decentralised (anglophone) and more centralised (francophone and Americo-Liberian case) states. Furthermore, even in the most contemporary post-colonial era, differences persist in the types of non-state providers that thrive in more decentralised versus more centralised country contexts.
These findings suggest that neoliberalism and democratisation are important for explaining the timing of the acceleration of growth of NSPs since the 1990s but less useful in theorising the patterns. Since neoliberalism and democratic transitions are experienced to some degree across most of the continent, I tease out how colonial legacies may mediate broader trends of neoliberal democratisation in Ghana, Côte d’Ivoire and Liberia in West Africa.
The role of colonial legacies during neoliberal democratisation in Ghana, Côte d’Ivoire and Liberia
The nature of non-state provision differs in important ways across these three country cases in West Africa. The mode of non-state provision in Liberia is substitution, where NSPs have greater capacity and frequently take over service delivery functions when the state does not provide. In contrast, the mode of non-state provision for both Ghana and Côte d’Ivoire is appropriation, where the capacity of both the state and NSPs is relatively low, and thus, NSPs frequently broker access to state welfare services but do not always provide and deliver them independently. But, even here, within the same analytic category, the number and types of non-state providers differ significantly between Ghana and Côte d’Ivoire.
To begin, Ghana has a greater number of active NSPs of all types overall in the post-colonial era, in comparison to either Côte d’Ivoire or Liberia (see Table 4). The numbers of INGOs and NGOs is higher, and the dominance of private provision of schooling is greater in Ghana than in Côte d’Ivoire or Liberia. Indeed, evidence from multiple sources suggests that the numbers of international NGOs are actually declining in Liberia as the Liberian government strengthens and donor-led projects are concluded (UIA 2013; USAID 2012, 80).
| Precolonial era | Missionary/colonial era | Post-colonial era | |
|---|---|---|---|
| Centralised states (francophone and Americo-Liberian colonies) | LOW NUMBER Rural and urban dispersion Village and multi-village associations Faith-based in Islamic areas | LOW NUMBER Urban and coastal concentration Village, regional and national organisations Restrictions on faith-based | MEDIUM NUMBER Urban concentration Regional and national organisation |
| Decentralised states (anglophone colonies) | LOW NUMBER Rural and urban dispersion Age-grades within villages Implicit faith-based centred on indigenous African religions | MEDIUM NUMBER Rural and urban dispersion Extended family and village-level organisations Expansion of faith-based | HIGH NUMBER Rural and urban dispersion Community-based organisations Expansion of faith-based and for-profit firms |
Moreover, a greater number of formal non-state providers are located across the country in both urban and rural communities in Ghana (Ibid., 61), whereas NSPs are more heavily concentrated in the national capital of Abidjan in Côte d’Ivoire and Monrovia in Liberia (USAID 2012). In Liberia, the lack of adequate road and energy infrastructure outside of the capital hinders the operation of INGOs, larger NGOs and businesses in many of the more remote counties (Ibid.). In Côte d’Ivoire , insecurity is still high in some regions, particularly in the north and west of the country, so many NSPs remain based in the national capital, where electricity, roads and the airport are now reconstructed and well maintained (Sturm 2013).
In addition to the greater number and spread of NSPs in Ghana, the range of NSP types is more diverse in Ghana than in Côte d’Ivoire or Liberia. In Ghana, NSPs are both large and small in size, more locally based and entrepreneurial than the larger international NGOs prevalent in Côte d’Ivoire and Liberia. In Ghana, the most prevalent NSPs are informal networks, community-based organisations, faith-based entities and for-profit businesses (MacLean 2010). Faith-based organisations seem to be much more active in Ghana than in Côte d’Ivoire overall (MacLean 1999).
What explains the growth and divergent character of non-state provision in these three cases in West Africa? Certainly, the pressure to adopt neoliberal reforms was experienced by all three countries in the 1980s, and all three implemented, at least partially, some degree of structural adjustment programmes (World Bank 2001). Likewise, all three cases attempted some democratic transition to a multiparty system in the 1990s. Thus, the overall acceleration of growth in all three cases can be largely explained by these major shifts in the economic and political regime.
A more nuanced explanation however digs into the timing and degree of implementation of neoliberal reform as well as democratic consolidation, and the way that these twin reform efforts have been mediated by the history of state formation and colonial legacies. For example, in Ghana, the gradual and more wholesale adoption of neoliberal economic reform has contributed to the reinvigoration of a historically lively small-scale business sector for Ghanaian men and women (Clark 1995; Opoku 2010). Furthermore, Ghana’s partial economic recovery since the stagnation and crisis of the 1970s and early 1980s (Aryeetey and Kanbur 2008), and the more recent discovery of off-shore oil reserves in 2007, have benefited a wealthy elite and a growing middle class. With a per capita GDP in 2016 of $4300, some Ghanaians can now choose to pay out-of-pocket for private welfare services, which are perceived to be of greater quality than public ones. Despite similar pressure to adopt neoliberal economic reform, the economies of Côte d’Ivoire and Liberia remain weaker, with Gross National Income per capita less than Ghana’s at $3400 in Côte d’Ivoire, and only $900 in Liberia.
The consolidation of democracy also plays a role in all three cases, particularly in Ghana. In Ghana, three alternations of power to the opposition since the advent of a multiparty system in 1992 have meant that civil and political liberties are increasingly protected. This democratisation has also contributed to the growth of for-profit firms, as well as CBOs, faith-based organisations (FBOs), and local and national NGOs (USAID 2012). In general, the Ghanaian state allows local businesses and non-profit organisations to register with ease and operate without harassment or intimidation (Ibid.).
The Liberian regime has also had two rounds of democratic elections since 2005, but no alternation among political parties as of yet (Sawyer 2008). The ruling party is also accorded expansive executive powers according to the current Liberian constitution, prompting initiatives from society and the state for constitutional reform. According to Freedom House, the current Liberian regime is only partly free. One implication of this assessment is that NSPs may be more tightly regulated by the government of Liberia. Historically, since 1977 (and amended through 2002), the government has regulated both local and international NGOs as well as community-based organisations through the Associations Law of Liberia with the stated purpose of aligning the activities according to national priorities (Griffiths 2010).
Democracy remains even more fragile and contested today in Côte d’Ivoire than in either Ghana or Liberia. After two rounds of multiparty elections where the dominant party continued to reign, democracy faltered in 1999 with a military coup d’état and two subsequent civil wars that divided the country. Conflict over candidate eligibility and boycotts by different opposition parties threatened the outcome and perceived legitimacy of the elections in 2000–01 and more recently in 2010–11. President Ouattara struggled to reunite the country and rebuild basic infrastructure, but was rewarded with a landslide victory in the October 2015 election. Despite improvements, general insecurity poses a serious challenge to NSPs in Côte d’Ivoire.3
None of the above variations in neoliberal democratisation can be properly understood, however, without examining the continued importance of colonial legacies on the number and type of non-state providers in Ghana as compared with Côte d’Ivoire and Liberia today. During colonial rule in Ghana (then called the Gold Coast), the British believed that the most effective social welfare system already existed in the customs of the African village. In one report from 1931, the British articulated their task as ‘(adapting) and (building) upon this communal machinery for social security … [thus] (supplementing) the tribal or village organization’ (GNA 1931). The British thus encouraged the continued role of indigenous and more locally based social welfare networks and organisations. Kay (1972) argues that the preservation of these local institutions was designed to entrench colonial power and facilitate the economic exploitation of the colony by the British. For example, they resisted direct funding of social welfare services in their colonies and instead relied on missionary organisations to provide health and education services. The British also called for revenue generation from the treasuries of the local chiefs, the creation of local advisory committees and the use of vernacular textbooks and native language instruction.
In contrast, the French took a more active, highly centralised and secular role in the provision of social welfare in Côte d’Ivoire. The presence of French Catholic missionaries had historically been much more limited than that of the Protestant missionaries in the neighbouring Gold Coast. Where the British worked to reinforce the indigenous extended family system, the French called for its dissolution. One French official wrote in the early 1950s that ‘the disaggregation of the traditional family corresponds to emancipation, a liberation of its members that one can’t complain about’ (FNA 1951–52). As a consequence, the French used revenues from direct taxation to provide a broader level of public social services in Côte d’Ivoire than in Ghana.
The colonial legacy in Liberia is unique on the continent, but nonetheless important for the nature of non-state provision today. Liberia is one of the few African nations that were never colonised by a European power. Instead, the country was colonised beginning in 1820 by African Americans, the majority of whom were freed slaves. These colonists became known as Americo-Liberians and never left; they remained the dominant minority in Liberia and ruled a highly centralised republic from 1847 until the first coup in 1980. The Americo-Liberians’ ideology of racial supremacy (Hanson Thomas McPherson 2010, reprinted from 1891) meant that the indigenous tribes from inland Liberia were officially excluded from citizenship until 1904. Even thereafter, with political power wielded from the centre by the political minority, economic development and social welfare were heavily concentrated in the capital of Monrovia to the benefit of Americo-Liberians.
In the immediate post-independence period, these different colonial legacies were reproduced and continued to shape the nature of non-state provision in Ghana, Côte d’Ivoire and Liberia. A diverse and numerous group of non-state social service providers continued to be relied upon more extensively by President Kwame Nkrumah in Ghana than by President Félix Houphouët-Boigny in Côte d’Ivoire. For example, churches played a larger role in providing primary and secondary school education and health care in Ghana than in Côte d’Ivoire (Fadayomi 1991). Even when the Ghanaian government began to increase public funding for secondary and technical schools under the Seven Year Plan (1963–70), they urged local voluntary associations and district-level governments to support primary and middle schools. Furthermore, throughout the 1970s, local communities were frequently asked to support social services by supplying communal labour, supplying local building materials and contributing money to special fundraising collections.
In contrast, in Côte d’Ivoire, Houphouët-Boigny emphasised the state’s responsibility for financing and delivery of social welfare services directly to the population; the central state left little room for non-state providers of any type. In general, this highly centralised social welfare system meant that public expenditures were concentrated on government hospitals and schools in Abidjan and a few other major urban cities in Côte d’Ivoire.
Meanwhile, in Liberia, the Americo-Liberian minority continued to discriminate against the indigenous tribes in Liberia. In the post-World War II period, the Liberian government led by President William Tubman turned to foreign investment, particularly from the US, to begin to modernise the country’s infrastructure. The Liberian government was not only highly centralised, like in Côte d’Ivoire, but had a tremendously powerful executive, so the smaller number of non-state providers remained closer to the capital city and had less autonomy than in Ghana.
Even in the contemporary period, under the homogenising pressures of neoliberal globalisation and democratisation, colonial legacies continued to influence the number and types of non-state providers active on the ground in different ways in the three cases. The causal mechanisms for the reproduction of these colonial legacies since the achievement of independence around 1960 are twofold: (1) the enduring normative frameworks of political elites; and (2) the persistent organisational capacities of state and non-state actors.
In Ghana, we see the explosion of a range of secular and faith-based, non-profit and for-profit non-state providers in urban and rural communities throughout the country. In interviews with contemporary political elites in Ghana, they articulated a normative orientation where non-state actors were expected and encouraged to play a prominent role in social welfare, to help the needy ‘indirectly’.4 Ghanaian NSPs – be they community-based organisations, local or international NGOs, or for-profit providers – are vibrant and capable of fulfilling the expected roles in urban and rural areas.
This more decentralised type of non-state provision in Ghana had important consequences for social inequality and accountability. The growth of a wide range of non-state providers drawing on community-based revenues and resources at the local level aggravated the inequalities between the wealthier and the poorer districts and regions in the country. Social provision was essentially delinked from the centre, and so were the mechanisms of accountability for citizens. The greater number and diversity of non-state providers appeared complex and highly fragmented to Ghanaian citizens, who often lacked basic knowledge of which actors were providing what type of service at what cost.
In contrast, in Côte d’Ivoire, the historic centralisation and secular orientation of the state since colonial rule have slowed the growth and diversity of non-state providers compared with Ghana. Ivoirian NSPs tend to be larger, secular international NGOs or larger, for-profit businesses based in Abidjan and, to a lesser extent, the regional capitals. The normative framework of Ivoirian political elites also clearly differed from that of their Ghanaian counterparts. Ivoirian government officials spoke of working with community-based organisations in terms of encadrement,5 a conceptualisation that called for a more top-down and interventionist role for the state compared with the Ghanaian mode of collaboration with existing local initiatives. The conflict in Côte d’Ivoire from the late 1990s through the 2000s further reinforced the history of a centralised approach to non-state provision, as the central government struggled to regain control over the state’s full territory and relied on outside international NGOs for assistance with service delivery.
This more centralised type of non-state provision in Côte d’Ivoire also had significant consequences for inequality and accountability. The inequalities between the rural towns and urban centres have worsened over time, intensified by the regional conflicts which made transportation between areas less secure. While the source of accountability for welfare provision was more readily knowable in a more centralised system, the geographic and cultural barriers were greater for Ivoirian citizens from small rural communities far from Abidjan.
In Liberia, colonial legacies of discrimination were arguably the root cause for the political instability and over two decades of civil war led by Samuel Doe and later Charles Taylor beginning in 1980. The post-conflict administration of President Ellen Johnson Sirleaf (2005–present) is attempting to reconstruct the social welfare system, depending heavily on the international support donors working through international NGOs. In an apparent effort to redress past colonial inequalities and create a more decentralised system of governance (Sawyer 2005), the Sirleaf government articulates a policy goal of achieving a more equitable distribution of social welfare services (GOL 2011), but the urban bias and international dependence on foreign donors are difficult to overcome. Part of the problem is a demographic one: the violence of two successive civil wars resulted in high mortality rates and internal displacement, creating an extremely youthful and urban population (Ibid.). This dependence was gravely exacerbated by the Ebola pandemic in Liberia in 2014. The role of non-state providers is growing slowly with the containment of Ebola, improved security and a rebounding economy, but the organisational sustainability (USAID 2012) and quality of Liberian NSPs is uneven.6
The patterns of growth and predominant types of non-state providers in Liberia vary from both of the other cases owing to the persistent challenges posed by its unique history. Ironically, the nature of non-state provision in Liberia differs most markedly from anglophone Ghana and resembles more closely francophone Côte d’Ivoire because of the heavily centralised nature of state formation there. Like in Côte d’Ivoire, this centralisation of non-state provision reinforced historic inequalities between the rural periphery and the urban capital, but in Liberia, these inequalities overlapped with perceptions of ethnic difference and marginalisation. Moreover, the routes to seek accountability were not only geographically and culturally remote, but this distance further weakened the average citizen’s trust in government institutions.
Conclusion
In conclusion, non-state actors are an essential component of social welfare regimes in Africa. Non-state provision is not new on the continent, but the numbers of these NSPs have increased remarkably over the past three decades owing to the contemporary projects of neoliberal democratisation. Moreover, the patterns of non-state provision diverge in significant ways across country cases in West Africa. The in-depth comparative analysis of Ghana, Côte d’Ivoire and Liberia revealed that colonial histories of state formation have stimulated the emergence of certain different systems of non-state provision during the colonial era of the nineteenth and early twentieth centuries. Subsequently, these colonial legacies have been reproduced over time, such that contemporary dynamics of neoliberal economic reform and democratisation have resulted in the more rapid growth of different types of non-state providers today.
The awareness of the growing role of non-state actors across Africa and in many other contexts, including even the highly industrialised and wealthy United States (Allard 2014), is important because these changes are not neutral, technocratic changes, but instead are politically contested with far-reaching consequences. The rise of non-state provision is creating new inequalities of access and complex mechanisms of accountability for African citizens. For example, some NSPs employ restrictive eligibility criteria, whether in their stated, formal rules or informal practices, ensuring that their beneficiaries do not include all needy people in their catchment areas. Paller (2015) highlights how only the well-connected individuals gain access to affordable housing in the urban settlements of Accra, Ghana. Few NSPs have either the resources or the mandate to provide broad, equitable access to services (Wood 1997). For example, MacLean and Brass (2015) find that NGOs and businesses struggle to reach the rural poor in Kenya and Uganda with affordable energy solutions. Since NSPs do not have mandates to serve all citizens universally, it may be harder for people, especially poor and marginalised groups, to identify providers or to demand recompense for insufficient or poorly rendered services. This is particularly difficult when NSPs are informal or when citizens face an increasingly fragmented scene of multiple non-state providers, such as the NGO landscape in post-conflict Liberia. Rose (2006) suggests that this fragmentation may be mitigated by the formation of umbrella groups that coordinate provision and accountability among non-state providers, but then acknowledges that these associations usually privilege more elite Africans based in urban areas. Currently, much of the literature focuses too narrowly on service provision by the public sector and the effects on accountability and participation (Harding and Stasavage 2014); scholars need to broaden the analytical lens to include the interactions between citizens and non-state providers.
Scholars examining social welfare in African countries must also recognise and pay attention to the interactions between the state and non-state actors. Many NSPs are indigenous organisations or informal institutions with long histories in particular African contexts. Even with the transmission of new international norms through transnational advocacy networks, African non-state actors, states and citizens work to construct new hybrid forms of organisation. We should therefore broaden our conceptualisation of international interdependency beyond the scope of the contemporary era of accelerated, neoliberal globalisation that may constrain a state’s social spending. We need to acknowledge how the relatively brief experience of different types of colonialism continues to shape the role of non-state actors in Africa today. We also must investigate how the waxing and waning of post-colonial rule by authoritarian dictators has combined with the dictates of the multilateral club of donors to open and close new spaces for non-state actors to provide social welfare over time. Despite the widespread legitimation of the expanding number of non-state actors by the advocates of neoliberal democratisation, neither non-profit altruism nor for-profit incentives will adequately provide social welfare for all Africans. ‘Rolling back the state’ has not enabled development for all. Persistent poverty in relatively remote and difficult-to-access communities demands the continued commitment and democratic responsiveness of the African state to its citizens.