Keywords

1 Introduction

The Sustainable Development Goals (SDGs) require significant levels of finance to be mobilized to meet development objectives . The United Nations Conference on Trade and Development (UNCTAD) puts the global investment needs to meet the SDGs at between US$ 5 trillion and US$ 7 trillion per year. For developing countries alone, the investment needs range from US$ 3.3 trillion to US$ 4.5 trillion. This includes finance for basic infrastructure such as roads, rail and ports, power stations, and water and sanitation . It further extends to investment required to address food security such as agriculture and rural development, climate change mitigation and adaptation, health, and education (UNCTAD 2014). The economic and financial shocks associated with COVID-19 have further increased the levels of finance required for the achievement of the SDGs by 2030, especially at the local government level (UN 2020; OECD 2020).

Most suggested measures to address the SDG finance gap focus on ways to mobilize or increase sources of funding, such as Official Development Assistance (ODA), Foreign Direct Investment (FDI), and other private funding and domestic revenues (Move Humanity 2018), even if these have been declining (UN 2020). However, governments have the mandate and provide the institutional apparatus for the SDGs to be implemented. Public finance should therefore be central to debates on SDG implementation. Moreover, considering the importance of local governments in achieving the SDGs in urban contexts, such debates should consider the importance of subnational government finance.

While global agendas recognize the importance of local governments, most overlook the issue of local finance. For example, Agenda 2030 only makes reference to finance as part of SDG 17.1, which focuses on the need to strengthen domestic resource mobilization, including through international support to developing countries, to improve domestic capacity for tax and other revenue collection. Meanwhile, SDG 17.2 and 17.3 link to ODS and FDI levels. Similarly, the Addis Ababa Action Agenda (AAAA), the most important international framework dedicated to development finance,Footnote 1 makes no mention of cities at all and only has one mention of subnational governments.

The New Urban Agenda (NUA) is the only global agenda that provides suggestions for urban local government financing; however, these provisions are very general and fail to provide robust guidance (UN 2017).Footnote 2 Recent UN reports have started to include more attention to subnational financing, especially when it comes to infrastructure finance in contexts of rapid urban growth (UN 2018). However, action areas that are identified for subnational financing are generic and do not take the specificity of existing fiscal systems into account.

In most African countries, the public finance apparatus—through which funding for the SDGs is meant to flow—remains a work in progress. Reforms implemented through the decolonization period, and later structural adjustment, have laid a patchy groundwork for fiscal governance. In many countries, reforms are still underway, with powers and functions shifting, at times with and at times without the resources needed. These reforms are overlaid on dynamic processes of urbanization, wherein metropolitan regions, cities, and towns are growing and placing new pressures on local states. In order to assess the prospects for SDG financing in African cities, we must have a better understanding of the workings and dynamics of the local fiscal space.

In this chapter, we therefore draw attention to the importance of local government finance and raise questions about the ways in which finance flows, who controls it, and how much of it there actually is, in order to highlight the importance of fiscal decentralization for SDG implementation in African cities.

2 Why Decentralization Requires Effective Multilevel Governance

Global development agreements generally represent the ambit of national governments, as the main signatories and entities responsible for implementation. However, many of the areas that development agendas address—such as water or health—have been devolved to subnational levels of government. Subnational government refers to all of the levels or tiers of government below the national. In many contexts, this includes regional governments, such as states or provinces, and local municipal governments. As such, subnational levels of government, despite their often-limited role in global forums, are central to the effective implementation of global agendas. An important part of localizing the SDGs , therefore, is establishing robust multilevel governance systems that can allow for different levels of government to play their role.

A key part of multilevel governance systems is determining what different levels of government can and cannot do, what they are and are not responsible for, and how the relationships between them will operate. As countries are internally differentiated, another important part of designing these systems is addressing the diverse needs and capabilities of different types of areas and by extension those authorities responsible for them. There are several interrelated concepts which together have produced a strong discourse around the most appropriate assignment of functions, political powers, and resources to subnational governments (Smoke 2001; Bird and Vaillancourt 2008). The first significant concept is the principle of subsidiarity (Moeti et al. 2007). Subsidiarity argues that public goods or service provisions should be assigned to the smallest or lowest level of government, granted that this is practical and that economies of scale are still possible (Boadway and Shah 2007). Subsidiarity is closely aligned with the neo-classical “decentralization theorem” which argues that assignment should be made to the lowest level at which provision is efficient and public good maximized (Oates 2008; Buchanan 1989). The classical economic case for fiscal decentralization is predicated on the assumption that welfare gains are maximized when decision-making is brought closer to constituents (Oates 2008).

The decentralization of functions, political power, and fiscal resources to subnational levels of government is key to this agenda. Decentralization refers to the process of shifting powers from central governments to subnational levels of governments. The decentralization of functions, political power, and resources is widely assumed to be a precondition for achieving good governance in developing countries (Faguet 2014). In theory, decentralization can achieve more accountable and efficient allocation of resources that respond to local needs and priorities, combatting top-down decision-making and corruption. As a development mantra, it has been pushed by multilateral institutions, think-tanks, and policy-oriented scholars (UN-Habitat 2009; Lincoln Institute and World Bank 2016). It features in key United Nations agendas including the AAAA, Agenda 2030, and NUA , all of which articulate the importance of empowering subnational governments to enable them to play more central roles in development processes (UN 2015, 2017).

3 Urban Africa’s Multilevel Government Experience

There is much to debate regarding the applicability of these neo-classical arguments and their applicability to the African context. As the wider debates on decentralization have progressed, many questions have been raised regarding the assumptions that underpin these theories. Regardless, decentralization has been aggressively pursued in Africa since the 1980s, including in both political and fiscal decentralization reforms (Ribot 2002; Smoke 2003; Tanzi 2016; Farvacque-Vitkovic and Godin 1998). Dovetailing with the Structural Adjustment of African economies, decentralization formed part of a suite of reforms that focused on “institution-building” and “good governance” (Becker et al. 1994; Clarke Annez et al. 2008). In some African countries, reforms were supported by central states in order to access debt relief. In other cases, decentralization allowed states to “neutralize regional ethnic tensions,” offering fiscal, administrative, or political power to regional elites and curbing opposition (e.g., in South Africa, Ethiopia, and more recently Kenya) (Smoke 2003, p. 12).

Decentralization reforms have reshaped African countries, shifting powers, functions, and resources away from central governments. However, these reforms, both political and fiscal, have been hotly contested. In many cases, reforms have been resisted by national governments who fear the loss of both fiscal and political control (Wunsch 2001). Africa’s capital cities have been a particularly important site for this resistance, as has been apparent in cities in metropolitan areas such as Gauteng and Dakar (Bekker and Therborn 2012). Private sector actors, such as construction companies and infrastructure lenders, have also resisted decentralization, preferring to do business with centralized technical agencies, rather than democratized local authorities. This is particularly true when it comes to large infrastructure projects that require coordination between several local governments and in cases where local governments are perceived to be heavily influenced by party politics. While great strides have been made in terms of democratic decentralization, the question of how money flows, and who controls and shapes budgets, is a vital part of understanding decentralization in practice . In other words, political decentralization and fiscal decentralization have not always gone hand in hand. Understanding the nature of the fiscal requires understanding more than just the legal and regulatory frameworks which underpin fiscal processes—it requires understanding how these operate in practice .

4 Method

Unpacking fiscal dynamics in Africa requires both understanding aggregates and comparisons across the continent, as well as the specificity of particular places. Both require contextualization of data, either among a family of cases or of the particular histories, arrangements, and development trajectories of countries and cities. For this purpose, this chapter consolidates and synthesizes material from several studies of subnational finance in Africa. The bulk of the insights are drawn from a study on subnational government in Africa, which collected data on 21 African countries.

The 21 countries, listed in Table 9.1, fall into different income groups and population sizes and were selected due to their availability of fiscal data (see Table 9.1, columns 2 and 3). The fiscal data was collected for 2016 as this was the year with the most complete data set in the year of collection. The authors for this chapter were integrally involved with the data collection process. The raw data was collected from enacted government budgets, central banks documents, public finance transparency forums, and organizations such as the International Budget Partnership (IBP), the Commonwealth Local Government Forum (CLGF), and United Cities and Local Governments of Africa (UCLGA) and using International Monetary Fund (IMF), UNICEF, and World Bank reports (when official government budgets were unavailable) (see Table 9.1, column 5). The collected data was standardized through a framework developed as part of a multi-year fiscal data observatory. For several countries, including South Africa, Nigeria, Ethiopia, and Namibia, more detailed fiscal breakdowns were available and have been used to provide nuance. The raw data is available online as part of the United Cities and Local Governments (UCLG) and Organization for Economic Cooperation and Development (OECD) as part of the OECD-UCLG World Observatory on Subnational Government Finance and Investment (OECD and UCLG 2019a).

Table 9.1 Overview of fiscal data for 21 African countries (Compiled by authors. Income categorization based on the World Bank Atlas method, published by the World Bank, using 2018 data. Population is based on the United Nations Department of Economic and Social Affairs (UNDESA ), World Population Prospects, 2016 data. Country data based on available Fiscal Year (FY) and exchange rates in Purchasing Power Parity (PPP))
Full size table

5 The Fiscal Multilevel Governance Context in Africa

The rest of this chapter unpacks the findings from this work. We have arranged these findings into three key insights. First, we argue that there is limited comparative data on African countries decentralization. Ambitious efforts to consolidate data on subnational finance have usefully demonstrated the incredible diversity and heterogeneity of Africa’s experience of fiscal decentralization. Second, these data exercises have foregrounded several challenges related to data and comparison, including limited fiscal transparency and different fiscal accounting and reporting styles. The implications are that such data sets need to be considered in context and triangulated with qualitative information. Finally, there is considerable conflation between local government and city government. This has serious implications for fiscal decentralization, which requires tailored approaches for rural, urban, and metropolitan areas.

5.1 Comparing Across Diversity

Ambitious efforts to consolidate data on subnational finance have usefully demonstrated the incredible diversity in African contexts. The data shows that African subnational governments today are highly heterogeneous (Paulais 2012; Clarke Annez et al. 2008). According to the UCLG data sets, there are over 15,000 recognized subnational governments across the 54 countries in Africa. Most African countries have emerged from decentralization reform processes with two levels of subnational government: the regional level and the municipal (local) level, which forms the lowest level of government (UCLG and Cities Alliance 2015). For example, Zimbabwe has two tiers, with 92 local councils at the local (municipal) level and 8 provinces and 2 metropolitan cities at regional level. There are also several cases that have three subnational levels. The level between the regional and the local is often referred to as the intermediate level. For example, Togo has a three-tier system of 5 regions, 39 prefectures at the intermediate level, and 116 municipalities at municipal level. Finally, some African countries only have one level of subnational government. In these cases, it is often hard to tell if this single level should be referred to as a local level or a regional level. For example, Malawi has only 35 councils, and Kenya has only 47 counties (OECD and UCLG 2019b). These differences in subnational architecture have significant impacts on the public finance systems and operations of countries, shaping how money flows.

The extent to which subnational governments control expenditure is an important indication of the level of fiscal decentralization. Table 9.2 outlines the subnational spending per capita, as a proportion of total government spend is divided between current and capital. Unfortunately, the UCLG data set did not differentiate between regional and local government spending for most countries. It is only possible to ascertain this for federal countries, such as South Africa, and countries with only one tier of subnational government (as it can thus be assumed that it is all local spend).

Table 9.2 Subnational expenditure data for selected countries in Africa (OECD and UCLG 2019c)
Full size table

When comparing subnational expenditure levels per inhabitant (see Table 9.2, column 2), the highest level of subnational expenditure per capita is in South Africa. South Africa, however, is a severe outlier, as the next highest country is Botswana, with less than a quarter of the per capital spending of South Africa. The lowest per capita expenditure is in Burundi and Malawi. These extremes show the diversity across the continent. Notably, while there is no clear trend in terms of regions, the upper-middle-income countries have higher levels of subnational spending per capita. In the cases of Botswana and Mauritius, two of the higher spending countries, this can be attributed to a combination of the countries’ economic wealth, their small population number (resulting in higher expenditure per capita), and the decentralization of responsibilities in the sectors of general public services, economic affairs and transportation, environmental protection, and culture and recreation. In South Africa, the high level of expenditure per inhabitant is a direct result of the country’s well-established and comprehensive legal fiscal framework at subnational levels and its overall wealth (although unequally distributed).

In terms of total spend, expenditures are highly centralized in Angola, Burundi, Senegal, Malawi, and Eswatini, where subnational expenditure share in the general government is equal to or below 5% (see Table 9.2, column 3). There are different explanations for this in the various countries. Angola and Eswatini experience highly centralized control of all government responsibilities and resources due to the current political regimes (although in Angola there appears to be a move towards allowing subnational governments greater control of their finances). In Malawi and Senegal, a lack of capacity to manage resources at the local level is largely the cause for their low levels of expenditure (OECD and UCLG 2019b). In the case of Burundi, the low levels of total subnational spending are mostly due to a lack of resource allocation from the central government, resulting in subnational governments not being able to fully implement and fulfill their devolved functions. According to a World Bank Public Expenditure Review, an adequate and transparent intergovernmental transfer system is yet to be defined and put in place by the Burundian government (World Bank 2014). In contrast to Nigeria, Ethiopia, and South Africa, subnational expenditure accounts for nearly half of all public expenditures (see Table 9.2, column 3). For Nigeria, the federal structure concentrates fiscal resources with the states. In South Africa, the provinces control big-ticket items, such as health, education, and housing subsidies.Footnote 3 The larger South African metropolitan areas also control large infrastructure grants for urban service delivery. In Ethiopia, the high proportion of subnational expenditure is a direct result of large social sector expenditure undertaken at the subnational level. For example, in 2016, about 60% of the national health on-budget expenditure and 47% of the national education expenditure were incurred by subnational governments (UNICEF 2017). Ethiopia stands out as the most decentralized country among the low-income countries, with 48% of public expenditure occurring at the subnational level. This is mainly due to Ethiopia’s public sector-led development strategy , which includes high levels of pro-poor expenditure at subnational level in the sectors of health, education, nutrition and food security, water, energy, electricity, and road construction.

Beyond the question of proportional control, it is important to understand the nature of subnational expenditure. In particular, it is important to understand how much control subnational governments have over capital expenditures and current expenditures. In Africa, subnational governments are important public employers, pushing up current expenditures (see Table 9.2, column 4). In some countries, current expenditure is more than 90% of subnational expenditure. In contrast, capital expenditure across the continent ranges between 7 and 58% of total subnational expenditure, with Benin and Tunisia leading (see Table 9.2, column 5). In the Tunisian case, local governments have few management responsibilities in areas requiring substantial current spending and therefore can play more of an investment role (OECD and UCLG 2019b). These cases, however, are exceptions, and capital investment on the continent generally accounts for a small share of total subnational expenditures (e.g., see Mali, Mauritania, Malawi, Uganda, and Zambia ).

Moving from expenditure to revenue, there are similar trends which can be seen (this is because governments can only spend what they have in revenue, except in cases of extreme over or under spending). Revenues include the revenues from own collection and grants. Similar to expenditure patterns, higher-income countries tend to have higher per capita incomes (e.g., South Africa, Botswana, and Mauritius). Most African countries fall in the range of US$ 10 to US$ 2500 per capita revenues (see Table 9.3, column 2). Similarly, the federal states of Nigeria, South Africa, and Ethiopia show the highest proportion of subnational revenue. In all three countries, regional (states, provinces) entities raise a larger proportion of subnational revenues, and local (municipalities) entities tend to rely significantly more on transfers from the federal government compared to regions. Table 9.3 shows the break down in subnational revenues between transfers and own-source revenues. Transfers dominate subnational revenue (see Table 9.3, column 5). Extreme cases include Uganda, Tanzania, and Rwanda, where transfers make up around 90% of subnational revenue. In Zimbabwe, by contrast, transfers are very low, less than 4%. This is largely due to severe political tensions between local and national government in Zimbabwe which result in limited transfers. For instance, Zimbabwe’s 2013 Constitution (Article 301), introduced the fiscal transfer to provinces and local authorities of 5% of national revenue raised, but this has not yet been implemented. This is not uncommon, as in many African countries central governments resist transferring money to subnational governments (OECD and UCLG 2019b).

Table 9.3 Subnational revenue data for selected African countries (OECD and UCLG 2019c)
Full size table

Overall, this data on revenue and expenditures shows that there is incredible diversity among African countries.

5.2 Need for Contextualized Understanding

Apart from the vast diversity in fiscal experiences across the continent, there are some serious gaps in the data which make it difficult to both understand what is going on, and to compare across contexts. Everything from effective forward planning to risk calculation relies on being able to understand what is happening, or could happen, in different geographies. Notably, we have significantly lower levels of insights into the fiscal data space in Africa compared to many other parts of the world. While this knowledge gap is as much political as it is technical, a well-rehearsed and important part of this story relates to the limited data which is collected in and on Africa. As the possibilities of the digital age emerge, there are many questions about what sort of data can and should be collected in Africa and who should be responsible for analyzing, sharing, and protecting such data. Limited subnational data collection is not particularly surprising considering the many challenges regarding access and availability at the city level (Borel-Saladin 2017). However, the gap in fiscal documentation is more difficult to comprehend as public finance (unlike other aspects, such as the informal economy, corruption, or other more subjective data) is generally part of the financial accounting frameworks of countries. Unfortunately, there are very few contemporary examples of publicly available and disaggregated fiscal data on Africa.

The recently published data set on subnational government finance, developed by UCLG with the support of the OECD, aims to fill some of the gaps in subnational fiscal data collection globally, with a considerable sample of African countries. The observatory/database aims to track data regularly over time, providing insights into trends and patterns regarding subnational finance and fiscal decentralization. As we show above, useful insights can be gained from this exercise. However, while the framework provided a robust method for data collection, the actual data collected on Africa remained limited. Some insights are possible, but there are also gaps which require additional research and contextualization.

Two important data issues are worth mentioning. First, there are many contexts where it is simply impossible to access reliable subnational data. For many countries, data is not published online, or if it is, the information is scattered. This issue is attributed to limitations in terms of capacity, accountability, and transparency (IBP 2017). African countries, except for South Africa, rank low on international budget transparency indexes for their limited budget accessibility and transparency to the public. Fiscal data on expenditure, revenue collection, and debt at subnational levels are largely on a cash basis and remain unknown and inaccessible for 33 countries on the continent, situated mostly in West, East, and Central Africa. Second, a particularly important and underexplored issue related to fiscal data is the incomparability and incommensurability of the data, owing to different systems of public finance accounting.

The differences among African countries stem from diverse fiscal legal regulations and accounting systems used in different countries. Most national governments in Africa have inherited and adopted—entirely or at least partially—the judicial and institutional framework of their respective colonial powers. For example, Anglophone African governments have adopted a mixed legal system of English common and customary law. The Francophone and Arab governments have adopted a mixed legal system of French civil law and Islamic law. The handful of Lusophone countries have inherited the legal system of Portuguese civil law and fiscal policies. When comparing British, French, and Portuguese colonial legacies, there is a greater uptake of accrual financial accounting methods and adoption of international accounting standards and reporting practices among Anglophone-speaking countries with legal systems based on the English common law. Overall, these challenges make comparing fiscal data difficult as the categorization of information is not uniform across the different contexts.

5.3 Local vs. City Government

In the context of African urban debates, there is considerable conflation between “local government” and “city government” as there are many different ways in which subnational governments are administratively defined. This has serious implications for fiscal decentralization, which requires tailored approaches for rural, urban, and metropolitan areas. For instance, local governments in Africa are increasingly characterized as either “urban” or “rural.” As shown in Table 9.4, examples of countries which categorize local government as urban or rural include Botswana, Burkina Faso, Burundi, Eswatini, Guinea, Malawi, Mauritius, Namibia, Niger, Senegal, Senegal, and Zimbabwe. Further differentiation may be made for larger cities, which can be granted “city ” or “metropolitan” status (examples of local governments like this include Mombasa, Johannesburg, Dakar, Kampala, and many others). The definition of “urban” used across African countries is also highly differentiated (Farvacque-Vitkovic et al. 2008). However, the growing attention to rural urban differentiation—in particular, when it comes to the structure of public funding—is a useful recognition of the complex processes of urbanization evident on the continent (Ribot 2002; Pieterse 2008).

Table 9.4 Data on urban differentiation in 25 African countries (Source: compiled by authors)
Full size table

While South Africa is the only country with official metropolitan classifications of local government, there are several cases where some level of metropolitan level coordination is being attempted in capital cities, for example, in Senegal (Dakar, established in 1996) and Ivory Coast (Abidjan, established in 2011), both of which have created bodies to coordinate the local government units which make up big cities. In other cases, where there are very large cities, these metropolitan areas more closely align with regional government boundaries, for example, in the case of Nairobi City County, Maputo City province, Addis Ababa [Chartered] City , and Lagos State. There are also cases wherein national bodies have been created, in parallel to the elected local government, to manage key cities. A good example of this is the Nairobi Metropolitan Services (NMS). These different arrangements reflect the efforts that African countries are making to grapple with their largest urban areas.

In the context of rapid, and highly uneven, urbanization in Africa, it is impossible to consider decentralization without accounting for urban dynamics. Rapidly growing urban and metropolitan areas on the continent have the potential to become the engines of inclusive and resilient growth. However, this requires a better understanding of these cities, in terms of how they operate and how they fit into the spatial and fiscal hierarchies of countries and regions. Overall, and particularly in comparison to Western counterparts, data on African cities is a challenge. This is particularly true for fiscal data which may be collected at the local level, but are rarely made public or aggregated for analysis. However, where data has been collected, there are interesting insights about inter-city fiscal differentiation which can be seen. Reflecting on data from 27 cities in Africa, we see several interesting factors related to city level finances (see Table 9.5).

Table 9.5 City level expenditure and revenue data for selected African cities (UCLGA 2017)
Full size table

As with the country data, revenues and expenditures track each other closely. Across African cities, per capita spending is wildly variate, ranging from less than US$ 1 in Dar es Salaam and Conakry to US$ 439 in Rabat (notably, no South African cities were included in this data set). This is a substantial level of variation which would have been further exacerbated by the inclusion of South Africa. Most African cities which were reviewed for this work spent below US$ 50 per inhabitant per year. The nature of this expenditure is also important. In most of the cities, most of the spending is on current expenditure, in other words staff costs and operations (UCLGA 2017). There is limited capital expenditure, ranging from US$ 15 to less than US$ 1 per inhabitant per year. There are exceptions to this, for example, Rabat has a total capital expenditure of US$ 210 per inhabitant per year, and Ezulwini has a total capital expenditure of US$ 126 per inhabitant per year. The extent to which cities are able to spend on capital and development has fundamental implications for the role they play in shaping the urban fabric. Together, Tables 9.4 and 9.5 point not only to the rising importance of urban areas and city finance but also the important role that metropolitan regions are now playing in the fiscal structures of African countries.

6 Conclusion

Public finance will be central to any chances of successfully implementing the SDGs . The question is not only about how much funding is needed but also who controls available funding. Given the limitations in fiscal data, it is very difficult to get a clear picture of the extent to which city governments in Africa control money. And by extension, it is difficult to assess the extent to which city governments are fiscally capacitated to support SDG implementation.

Despite this challenge, the chapter makes three important arguments. First, we show that there is incredible diversity in subnational government control over finance. Making broad-brush statements about the extent to which subnational governments are or are not fiscally capacitated thus requires more detailed interrogation. Second, the data is uneven and incomplete, therefore requiring a more holistic understanding of the complexity of the local government fiscal space. In addition to an overall lack of fiscal data, there is insufficient data on the divide in fiscal control between intermediate and local government. There are also serious challenges related to the commensurability of data. This is owing both to gaps in data collection and very different methods of fiscal and financial accounting.

The vastly differentiated multilevel governance arrangements and urban classifications make homing in on urban fiscal data a challenge. Significantly more work is therefore needed to get a clear understanding of the complex spatial and fiscal dynamics that are emerging through Africa’s urbanization process.