Photo: Jeff Attaway
Sub-Saharan Africa is currently the fastest urbanizing region of the world; by 2030, a majority of Africans will be living in urban areas.
As a result, providing adequate services in urban areas — such as sanitation, potable water, sustainable housing and electricity — is an increasingly important priority for African governments. Yet, due to the embracing of decentralization policies in Africa, a trend that has been enthusiastically supported by the international donor community, responsibility for providing these services has often been transferred to sub-national authorities. Where a situation of vertically divided government prevails — meaning that an opposition party is in control at the sub-national level — fulfilling these responsibilities can become more complex.
Today, vertically divided authority is a growing trend within the region, and a number of important African cities, ranging from Cape Town to Lagos to Nairobi, currently are in the hands of the opposition.
UNU-WIDER’s Decentralization and Urban Service Delivery in African Cities project examines the impact of politics on decentralization and urban service delivery by focusing on four countries: Kenya, Senegal, South Africa and Uganda. In all four countries, recent fieldwork was conducted with local government officials, decentralization ministries, civil society organizations and the donor community.
These four countries were chosen for a number of reasons. First, as can be seen in the table below, they all share a context of vertically divided authority in key cities (e.g., Nairobi, Dakar, Cape Town and Kampala). Second, major decentralization reforms were implemented in each country during the mid-1990s, and all four are considered to be among Africa’s most decentralized countries.
Third, there are key differences among them that can be leveraged to uncover when and why vertically divided authority is most problematic for service delivery. For instance, South Africa is a quasi-federal country with its decentralization laws clearly enshrined in its constitution. Moreover, inter-governmental transfers play a much larger role in Uganda and Kenya, while urban areas rely much more on their own tax revenues in Senegal and South Africa. Finally, while Senegal and Uganda have highly personalistic party regimes, South Africa possesses the most institutionalized party system, with Kenya occupying an interim position.
in control (major city)
|Orange Democratic Movement (Nairobi)||Socialist Party (Dakar)||Democratic Alliance (Cape Town)||Democratic Party (Kampala)|
|Decentralization legislation||Piecemeal since 1960s with key changes in 1990s and 2000s||Loi de la dé-centralisation, 1996||National Constitution, 1996||Constitution (1995) and Local Government Act (1997)|
|Vertical imbalance (intergovernmental transfers as share of local expenditures)||40.1%||17.2%||34.1%||97.6%|
|Expenditure of subnational governments as share of central government expenditure||3.9%||3.1%||19.4%||27.5%|
Sources: IMF Global Financial Statistics (2010), Global Financial Statistical Yearbook (2008), Municipal Development Agency of Senegal (2008, 2010), Guide des ratios Financiers
Our findings show that vertically divided authority augments the trade-offs between autonomy and accountability that are inherent in the decentralization process, with attendant impacts on service provision. Under circumstances of vertically divided authority, service delivery performance may provide one party an advantage, or disadvantage, among voters in subsequent elections.
As such, both central and local government actors will opt for a decentralization process that maximizes their autonomy, but they will prefer to be held accountable only when service delivery outcomes are favourable. The central government will, therefore, have a preference for employing “strategies of subversion”, or tactics to purposely reduce the autonomy of local government under conditions where the latter may be held accountable for good service delivery and increase autonomy when local government can be targeted for poor performance.
These tactics include removing elected officials within opposition-controlled municipalities and replacing them with government appointees, reducing inter-governmental transfers and local tax-raising authority, and increasing administrative ambiguity over who is responsible for delivering which services. Our research found that such tactics can negatively impact, for example, garbage collection in Dakar and housing in Nairobi.
Such strategies are easiest in contexts of low education, opaque information and complex institutional environments because citizens often may not understand whether local or national politicians are primarily accountable for the quality of service delivery. They are also more likely in personalized regimes where power is centralized in one major urban area, where fiscal decentralization is low, and where decentralization policies are ad-hoc rather than enshrined within key legal instruments. Consequently, we find that “strategies of subversion” are most explicit in the Ugandan case and least apparent in the South African one.
This research yields four main contributions.
First, it reconciles three literatures that often do not communicate with each other. Specifically, the decentralization literature rarely uses cities as an explicit unit of analysis. At the same time, the party politics literature predominantly focuses on the national level, while the urban studies literature rarely examines how macro political contexts impact dynamics at the city level.
Second, with regards to methodology, the research is not only comparative across countries and cities within Africa but also draws on insights from other regions of the developing world with similar experiences of vertically divided authority, especially Latin America.
Third, the cases provide a set of rich data on municipal finances, urban service delivery and local elections that typically are extremely difficult to obtain for African cities.
Finally, we believe our investigation holds important policy implications for the donor community. Many donors support decentralization as a means of improving service delivery in developing countries. In Africa, this support is reinforced by the fact that 25 countries stipulate that decentralization is a key objective in their country’s Poverty Reduction Strategy Papers. The World Bank, for example, allocated approximately US$10.6 billion to decentralization projects worldwide over the period spanning 1990-2006.
While most donors wish to remain apolitical in their interventions, this research illustrates that recognizing and addressing the ongoing contestation of power between local and national authorities in such divided contexts is crucial for ensuring the effectiveness of development assistance.