Bursting the development target bubble? The reality of sectoral spending targets

It can’t be denied that progress in development needs to be accompanied by greater spending on sectors that contribute to development, such as education, health, social protection, etc. But exactly how much needs to be spent, on which sectors, and what is affordable for developing country governments?

Much of the debate on development financing is characterised by a sector-oriented ‘silo’ approach, where discussions on spending needs and affordability focus on a single development sector, rather than considering the broader picture.

The increasing number of sectoral spending targets put in place by developing country governments and donors reinforce this silo approach. When looking at sub-Saharan Africa, for instance, we can find concrete or implicit spending targets in six development sectors that suggest a certain level of government expenditure for each. Some are very specific: for example in the Abuja Declaration in 2001 governments aspired to spend 15% of government expenditure on health. They also agreed to spend 20% of government expenditure on education as part of the Education for All Initiative in 2000, 10% on agriculture according to the Maputo Agreement in 2003 and 0.5% on sanitation in the EThekwini Declaration in 2008. At an African Union Assembly in 2009 governments agreed to increase spending on infrastructure, and in 2008 agreed to aim towards providing a specific minimum social protection package.

In isolation these targets may seem reasonable, but what happens when you look beyond a single sector – are these targets jointly feasible? The short answer is no – countries have not met the targets and cannot meet all targets simultaneously.

Recent work by ODI looks at whether these six targets fit the fiscal realities in five sub-Saharan African countries, and shows that government spending on key development sectors falls short in most countries. Only Ethiopia and Malawi meet two targets each, Kenya and Mozambique meet one and Uganda meets none. But with the combined cost of these targets amounting to more than 100% of government expenditure on average, this is not surprising – it’s impossible to meet all six targets!

Complicating the story further is that much of development aid is being delivered ‘off-budget’, with aid being paid directly to a project account and not included on a government’s official budget. Off-budget aid can be very high, for example around 50% of aid in Uganda is delivered off-budget. So, would governments have met the targets if off-budget aid were included in the analysis? Our data for Malawi and Uganda show not. As these expenditures are not recorded on government budgets, this clearly undermines effective budgetary management.

International agreements, such as the Accra Agenda for Action and the International Aid Transparency Initiative of 2008, set out principles and practical actions towards better alignment of aid to recipient country requirements, but these agreements are still in their infancy and implementation remains imperfect.

As the full ODI reportshows, if governments want to meet the targets, they’d need to increase domestic expenditure by 110% on average, keeping spending in other sectors constant. Likewise, doubling aid funding could ensure that all targets can be met. Neither of these options is feasible in the aftermath of the global economic crisis and the resulting fiscal deficits within OECD economies, however.

Decisions on how much revenue is raised and spent, and how it is allocated across sectors, need to take fiscal sustainability and coherence across sectors into account.

Lobbying to achieve specific sector spending targets can result in silo-based spending decisions that challenge effective and efficient public financial management, and further undermine realistic budgeting. In effect, sector-based advocacy has lead to more and more targets being signed that are impossible to meet jointly, thus putting sectors in direct competition with each other.

Successful lobbying in one area could be to the detriment of performance in other key sectors with weaker lobbies, even if they’re equally important to development. For example, the social protection sector has entered the debate later than others, at a time when claims on government budgets are already well established. Likewise, Steer and Wathne argue that increases in social sector aid have favoured health over education, as the health sector has had more high-level political support and media coverage. They outline a lobbying strategy for education and suggest how advocates can do more to ‘capture’ the global stage and compete with health spending.

It is claimed by some that developing country governments lack the political will to allocate resources to certain sectors – Michael Cichon argued recently at a UN Commission for Social Development meeting that the ‘key [to increasing social protection expenditure] is political will’. Political economy analysis suggests there may be implicit financial and political incentives for governments to become signatories to international initiatives, even if they have little intention of working actively to meet them – as the work of Dijkstra on Poverty Reduction Strategy Papers shows. While political preferences largely shape expenditure decisions, however, there’s no escaping the fact that not all sectoral financing demands can be met simultaneously without huge increases in budgets.

It’s important to caution against the notion that any particular sectoral spending target is inherently ‘affordable’ in any objective sense.

Affordability is contingent on broader political preferences, and the realisation of one target is likely to be at the expense of another. Given the unavoidable overall financing shortfall, the key questions become prioritisation of existing resources, and opportunity cost to other sectors when allocating resources. This is clearly an area that needs ongoing dialogue between donors and governments.

24 March 2011
,
This publication is an output of the following project: Government resource allocations to the social protection sector