8 Quality and Composition of Public Spending in Africa: Suggestions for Increased Effectiveness

Laura Wallace
Published Date:
January 1999
  • ShareShare
Show Summary Details
Luca Barbone

Africa lags behind the rest of the world in human development. Recent data show that the infant mortality rate is 91 per 1,000 live births. The total primary school enrollment rate is 75 percent, with the figure for females lower at 67 percent. Only about 50 percent of the population has access to safe water, and there are about 11 main telephone lines per 1,000 people.

Tremendous work thus remains for improving the quality of life in Africa and of assisting Africans to improve their own lot. Donors, business, and most governments hold the view that market development is the key to poverty reduction. Yet, there remains a substantial role for government in ensuring that rules promote competition and bring out the best of what each country has to offer, and in providing basic services when markets fail or externalities are to be found.

Alongside the poor social conditions (I say alongside because causality can go both ways) Africa suffers from a crisis of statehood—that is, a crisis of capability, as the World Bank’s 1997 World Development Report points out. Public institutions are not always effective, the rule of law is not always dominant, and checks and balances are at times lacking, so power can be abused. This clearly inhibits governments from delivering quality services, be they specific social services in a narrower sense, or an institutional environment conducive to economic growth and poverty reduction in a broader sense. Moreover, government behavior is responsible for, or reinforces, the crisis in statehood.

For the development community, the crisis poses a problem in strategy. Providing development projects with imported materials and management is a relatively simple task.2 But how does one assist countries to attain development objectives while at the same time abstaining from direct management control in respect of a borrower’s sovereignty and in an attempt to build up the borrower’s capacity to manage?

This paper will explore these questions after looking at the crisis in budget management and its roots. The proposed solutions are based on ongoing work at the World Bank to monitor the effectiveness of public expenditures, and on popular reflections on the role of institutions in shaping policies and the delivery of public services. A key recognition stemming from this work is the importance of institutions and the difficulty of reforming them—or, as is the case in many African countries today, resurrecting them after their demise.

To borrow from Douglass North, the 1993 Nobel Laureate, economic institutions consist of formal rules, informal rules, and enforcement mechanisms for those rules. In the past, we have been preoccupied with introducing formal rules, but have accomplished little—as political reality and institutions shape each other continuously, with no guarantee of a desirable outcome. In the future, we should also focus on informal rules, in part because they shape political reality, which in turn shapes government activity. Thus, the challenge for reformers throughout Africa and the donor community is to find a way to stimulate incentives so that both formal and informal rules are aligned to permit an effective use of public resources.

Crisis in Budget Management

Just how poorly are budgets managed in Africa? A World Bank study of public expenditures and budgetary practices and outcomes in 22 African countries3 reveals some interesting, albeit probably not surprising, facts:

  • Budgets generally remain only an approximate guide to the actual allocation of resources, despite several years of technical assistance and budgetary reforms that were driven by Policy Framework Papers.4 As Figure 1 (top panel) shows, over one-third of the countries had a discrepancy greater than 30 percent between the allocation of resources decided at the time of budget formulation and the outcomes. Less than one-half registered what could be considered a moderate deviation of 10 percent or less. These figures should give us pause, as they imply that for a large majority of African countries, budgets are a figment of the imagination.
  • What is even worse is that the resulting unpredictability of resources results in wide variations—even within sectors—in the allocation of funding. Figure 1 (bottom panel) shows that in a large number of countries, the sectoral destination of resources within priority sectors has not been respected. Funds for schoolbooks go for ministerial limousines. Allocations for rural health centers get diverted to white-elephant heart centers.
  • All this happens because accountability, as evidenced by expost controls, remains low. Indeed, Figure 2 shows that only one in seven countries manage to produce audits of public expenditures within a year of completion of the fiscal year. By then, it is old news and the political debate is already centered on next year’s priorities. Thus, ministers do not really have to account for their actions—or inactions.

Figure 1Budgets Do Not Tell the Story …

Source: Christos Kostopoulos, Progress in Public Expenditure Management in Africa—Evidence from World Bank Surveys, Africa Region Working Paper Series, No. 1, (Washington: World Bank, 1999).

Figure 2How Timely Are Audits?

Source: Christos Kostopoulos, Progress in Public Expenditure Management in Africa—Evidence from World Bank Surveys, Africa Region Working Paper Series, No. 1, (Washington: World Bank, 1999).

Certainly, these are bleak pieces of evidence—all the more distressing because most of the countries surveyed have benefited from considerable assistance from the World Bank, the IMF, and a score of foreign donors. It should be emphasized, however, that the aggregate picture masks a lot of variation, and there are a number of encouraging examples, such as Botswana and Uganda. Nevertheless, it is not an exaggeration to argue that the failure of the state institutions is a major determinant of the limited progress in the improvement of human development indicators.

Roots of the Problem

To better understand the reasons for the crisis of the state in Africa—and the role of the donor community in this crisis—it is useful to borrow from a World Bank study coordinated by Malcolm Holmes5 that classifies official budget systems as operating at three levels.

  • Level 1 is the capacity of the budget to achieve the government’s fiscal objectives—the level with which the IMF is concerned (budgetary discipline).
  • Level 2 is the capacity to allocate resources according to the government’s strategic objectives (political choice), essentially a political process, but one disciplined by cost.
  • Level 3 is the capacity to deliver services efficiently and effectively.

All the levels are linked. For instance, effectiveness (level 3) is certainly influenced by political and program choice tradeoffs (level 2), along with budgetary limits and intersectoral discipline (level 1).

To characterize in a schematic way the sequence lived by many African countries over the last two decades, one would say that the initial mandate given to the state in defining its level 2 priorities was very ample. This reflected the ideologies of the time, which emphasized the direct intervention of the state not only in the provision of public goods and services but also in the setting of development priorities in the productive sectors. Institutions were created to fulfill these mandates, but over time a breakdown of fiscal discipline emerged (destruction of level 1)—the result of external shocks and low growth caused by the excessive mandate of the state—leading to unsustainable macroeconomic outcomes that required correction. These corrections, however, were carried out in an environment of both increasing resource constraints and shifting, but as yet not well redefined, paradigms. As a consequence, across-the-board cuts, rather than a refocusing of mission, were the rule. It is no wonder that under these circumstances, institutions crumbled and the ability of the state to deliver basic services was reduced (level 3), accompanied by rising corruption and a lack of accountability.

So how does the donor community fit in? It is not too much of an exaggeration to argue that donors, despite the large amounts of resources deployed, have often taken a narrow view of assistance to Africa. They have been guided by the principle of project implementation, measuring success by the degree to which implementation—and only later development—objectives have been met.

In doing so, the donor community has often failed to recognize that foreign aid plays into the three levels at which budgets operate and profoundly affects the system of incentives faced by policymakers and bureaucrats. The availability of foreign financing for projects—nominally determined jointly with governments, but in reality driven by donor considerations and affinities—becomes a substitute for an effective reform of level 2. In conjunction with the requirements of level 1, donor assistance can become a vehicle for undermining institutions and the effectiveness of the state (level 3) as resources are directed toward areas that receive funding to the detriment of a coherent institutional strengthening strategy. And although some rudimentary services may be delivered because of donor financing, the notion of statehood is undercut, as service delivery is not based on self-reliance and confidence building, but on handouts.

In short, capacity is not being built from the ground up; rather, already low capacity is just being supplemented with technical assistance. Foreign assistance becomes part of the political tradeoffs in an arena of institutions that are often new to the region. Democracy is one such institution; statehood is another. Public administrations in one sense are political outcomes. And national budgets, to which we attach so much importance, are part of that outcome.

A Strategy for Reform

With this in mind, what is the appropriate strategy for reform and ultimately for increased effectiveness of public expenditures? The key is institution rebuilding.

First, assuming that the task of overall budget stabilization is accomplished, concentrate on fixing level 2. A sharp refocusing of the state’s priorities is badly needed. Depending on the nature of the markets, measures should be taken to increase contestability (through proper regulation or market-making measures); divest the state (in its broader sense) from activities that do not form part of its mandate; and use alternative delivery mechanisms (witness the role of the private sector and the civil society in the Democratic Republic of the Congo) where it can be presumed that these will work better.

This has begun to occur in a number of cases, but much too often the reality is that the state’s capacity has withered away under the pressure of bringing budgets under control to meet fiscal objectives. An emphasis on retrenchment, necessary as it may be, often results in a wholesale demoralization of the civil service and increased incentives for corruption, as the choice between integrity and starvation is an easy one. Rather than reducing state involvement and concentrating on core capacities—while increasing the role of the private sector and civil society in providing public goods and services—the initial strategy (or lack thereof) results in an overall deterioration of capacity.

Unfortunately, implementing a strategy for reform is proving to be a challenge, even in countries that recognize the scope of the challenge. The problem is that everything depends on everything else, and adopting the right sequence of reforms is a tricky endeavor indeed. Technical solutions will not work by themselves. The decomposition of institutions is such that the introduction of, say, a new budgeting system or computerized large taxpayer units is likely to be rejected by the system, rather than contribute to its reform. Similarly, technical assistance aimed at civil service reform is a glaring example that generally meets with negative results.

So what can be done? We increasingly realize that ingredients for success must include:

  • top-level support and determination;
  • a buying-in of existing personnel; and
  • work on simultaneous fronts—redefining the mission of the state, revamping revenue capacity, reforming the civil service, allowing civil society and private sector participation, and modifying the systems and incentives at the institutional level.

Above all, it should be clear that reform of the state takes time to be effective. The practice of using adjustment lending or the formulation of Policy Framework Papers as the occasion to “negotiate” imaginary civil service reform timetables, for instance, is one of the elements that may lead to a formalistic buy-in of reform, and ultimately to its chances of success.

Role of the Donor Community

While the responsibility for successful institution rebuilding must rest primarily with the African leadership, the substantial amounts of resources still accruing to them imply that the donor community can also have a great impact. In particular, the project enclave mentality that underlies much donor assistance must be replaced by an approach that emphasizes institution building, resource predictability, and accountability of public officials—or the odds will be stacked against reform.

How can this be done? First, by gradually switching the emphasis from individual projects to medium-term expenditure frameworks. Second, by focusing on institution building—recognizing that institutions are inputs as well as outcomes—rather than simply capacity building. And third, by introducing transparency, accountability, focus, and vision, and designing projects so that their performance is monitorable.

Role of Medium-Term Expenditure Frameworks

An MTEF is aimed at moving toward an integrated program approach with “internal consistency” between political and development objectives, domestic and foreign aid resources, and management capacity. Most importantly, once the MTEF has been agreed upon by government, donors, and stakeholders, it should provide a stable basis for decision making in the management of expenditures. The World Bank and other donors have three roles to play: serving as honest brokers between political players to maintain the MTEF agreement; assuring program financing with budgetary aid, all the time staying flexible; and maintaining traditional project lending as part of an MTEF. Ideally, if there were complete agreement on an MTEF, covering the entirety of the budget, donor assistance would not need to remain project specific—support to the budget would be all that is needed. Toward this end, the World Bank is now working on a new class of lending instruments.

But how can we arrive at a budgetary process that incorporates MTEFs as integral parts of its architecture? Clearly, what is required is a quantum leap away from the old practice of compiling public investment plans. A workable MTEF needs to assess country needs in the short and medium term, take stock of current and projected financial and management resources, and adjust activities to achieve high priority outcomes given finite resources.

The government’s commitment, therefore, is essential for the MTEF approach to succeed. Resource availability must be assessed in a realistic way, so that the government’s own commitments under the MTEF can be fulfilled (sometimes this may involve simply maintaining teachers’ salaries). In addition, an assessment of the scope of the state’s mandate and managerial capacity should lead to partnerships with the private sector and nongovernmental organizations for services they are best suited to provide. Indeed, partnerships with the private sector are now commonplace in the transportation and telecommunications sectors, and have also gained prominence in the education sector—in some countries with the assistance of nongovernmental organizations.

Role of Monitoring

A crucial element of keeping a medium-term project on track is monitoring—in essence, the buzzword for obtaining management information.6 Of course, monitoring requires financial resources, and these need to be programmed up front as part of overall project costs. So which indicators should be monitored? I would like to suggest three steps.

The first step is to establish desired project outcomes—an approach that differs markedly from the past focus on implementation rather than development objectives. Examples of outcome indicators would include employment (and salaries) of graduates for a technical school project, and earnings of farmers based on shifts to new crops in an agricultural extension project. These indicators are difficult to obtain and sometimes materialize only after the project’s completion. Nevertheless, it is vital to establish them at the project design phase, and to seek out those indicators that allow for early tracking of project performance. As outcomes make several assumptions about factors unrelated to the project that still affect project performance (be it climate or GDP growth), it is important to single these out and identify potential risk factors.

The second step is to establish project or program outputs. These are tangible goods and services—deliverables—for which the project is responsible. They might refer to the number of graduates from a technical school, or the number of methods learned by new farmers. Monitoring outputs is a more doable task, and such indicators are often compiled by project managers, implementing agencies, or auditors. Similar to outcome indicators, output indicators require assumptions that need to be articulated. For instance, many projects rely on partner contributions (be they national or district governments, nongovernmental organizations, or other donors), and associated risks need to be spelled out.

The final step is to devise indicators for monitoring the progress in implementation of the project. These are the input indicators. They measure the various processing stages of project implementation and are among the easiest to gather. In the technical school example, inputs would be classrooms built and teachers trained; in the agricultural example, they would include the number of seeds delivered.


In summing up, Africa’s low level of human development is a consequence of the political and economic institutions that influence the provision of social services. So far, project assistance from donors has proven simply to be a Band-Aid approach—and one where the patient is constantly changing positions, so the Band-Aid eventually comes off. For there to be a consistent delivery of services, there needs to be both a stable flow of resources and sound management.

How can a stable flow of resources be assured? This will require a commitment on the part of the government. As budgetary allocations are often the result of political decisions, “de-politicizing” of the allocation process will help. Introducing donors as “budgetary partners” with clear developmental objectives and financing may also go a long way.

How can sound management be assured? For this, policymakers and donors need to think in terms of desired outcomes over the medium and longer term, breaking objectives into modules, and monitoring performance. Such an approach will improve the quality of public spending and contribute to the much needed capacity building in Africa.


The views presented here are the author’s alone. Many thanks, without implication, are due to Mike Stevens and Christos Kostopoulos.


In a similar fashion relating to the private sector, foreign direct investment is often successful because foreign technology and foreign management use local labor. But notice that foreign investors have control, and they are not interested in cultivating domestic managers until the operation is already running smoothly. By contrast, development assistance tries to provide project and management training concurrently, while respecting the sovereignty of the borrower.


See Christos Kostopoulos, Progress in Public Expenditure Management in Africa: Evidence from World Bank Surveys, Africa Region Working Paper Series, No. 1 (Washington: World Bank, 1999).


Policy Framework Papers are forward-looking documents drafted by the national authorities in collaboration with the staffs of the World Bank and the IMF.


World Bank, Public Expenditure Management Handbook, Poverty Reduction and Economic Management Network (Washington, 1998).


Roberto Mosse and Leigh Ellen Sontheimer, Performance Monitoring Indicators Handbook, World Bank Technical Paper No. 334 (Washington: World Bank, 1996).

    Other Resources Citing This Publication