A wide range of public services and investments is crucial for the attainment of the PRSP targets which in the second generation PRSPs are commonly related to the MDGs. Therefore, the government budget is a key instrument for PRSP implementation. It is especially at the level of the budget composition that the relationship between budgets and the poverty-related targets is apparent (Adam and Bevan 2005). Yet, there is a missing conceptual link between the stated PRSP targets and interventions to provide public services on the one hand and the allocation of financial resources in the budget as inputs on the other hand. Generally, PRSPs hardly provide guidance on how to allocate public funds, especially in scenarios with limited resources. The medium term expenditure frameworks (MTEF), introduced in several PRSP countries as the link between PRSPs and the budget, are not based on a systematic costing and prioritization of policies to achieve PRSP targets. This gap is increasingly recognized. For
instance, the Joint Staff Advisory Note on the Tanzanian PRSP suggests to prepare a “medium-term quantitative framework underlying the MKUKUTA4 based costing and prioritization of the intervention strategies needed to reach the MKUKUTA targets” (IMF/IDA 2006).
Only recently, initiatives in some countries referred to as PRSP/MDG costings or needs assessments have started to put a price tag on PRSP interventions and to estimate the full financial cost of PRSP implementation (see Box 1). Costings can be seen as the first step to link the budget with the PRSPs. In many cases, they have demonstrated that there are large financing gaps and shortages of other inputs for public service provision (e.g. doctors). As donor funds and local resources remain limited, many interventions cannot be scaled-up to the extent necessary to achieve the targets. Therefore, government spending must be made more
effective in order to increase its impact on the targets. In particular, decisions regarding resource allocation across sectors of government activities (inter-sectoral allocation), resource allocation across different interventions (intra-sectoral allocation), resource allocation across time (sequencing) and resource allocation across geographical regions (spatial allocation) must be made on a less arbitrary basis as it is common in the budgetary decision making processes at present.
Box 1: Costing the second generation PRSP in Tanzania
Within the framework of the implementation of the Tanzanian National Strategy for Growth and the Reduction of Poverty, the Tanzanian Government conducts costings in several sectors including health, water, agriculture, education, roads and energy. Generally speaking, the costings are supposed to provide an estimate on what it takes in terms of human, institutional and financial resources to meet the PRSP as well the MDG targets by 2010 and 2015, respectively. They estimate the overall resource envelope as well as the specific resource requirements by interventions or public services. This allows the assessment of prospective financing levels. In case of a financing gap, the costings provide guidance about which interventions to prioritize. In addition, non-financial constraints are identified which potentially prevent the attainment of the targets in case additional measures are not taken even if sufficient financial resources are available.
The costing of the water sector showed that there is a financing gap based on current financing levels and that the cost of many interventions greatly differs. For instance, hygiene education requires expenditures of 45 million USD until 2015 whereas the construction of functioning water supply infrastructure requires 2 billion USD for the same period. However, at the same time, from a health perspective, hygiene education is as important as physical access to safe water. Therefore, the costing suggests that in scenarios with limited resources, hygiene education should be fully financed whereas the construction of physical infrastructure should be sequenced according to resource availability.
The health sector costing highlighted a dramatic shortage of trained doctors, specialists and nurses as a critical constraint for scaling-up interventions in the health sector. This nonfinancial constraint underlines complementarities with other sectors (education, civil service reform, etc.) and suggests that as a precondition to reach the health related PRS/MDG targets,
large investments in education as well as governance reforms are necessary.
By estimating the total resource requirements and by providing guidance for resource allocation in resource constrained scenarios, the costings are an essential input for mediumand long-term financial planning and thus for PRSP implementation. However, in Tanzania, an overall framework for budget allocation based on the costings, which mostly provide
sector-specific views, is still missing.
Making public spending more effective is inherently complex. First, targets can be multidimensional. If all targets have the same priority but if at the same time resource constraints prevent the achievement of all targets, trade-offs inevitably arise. Second, the impact of particular interventions on poverty-related outcomes is obscure. While a large range of public interventions likely contribute to poverty reduction, the challenge is rather to choose the most effective ones. Economic research has largely failed to keep pace with the challenges in development policy so that theoretical and empirical literature on public expenditure is of little use for allocation decisions in practice (Paternostro (2005)). Available technical tools provide valuable information for policy makers, but they must be complemented by additional analysis. Third, the budget process is highly influenced by political considerations, and policy makers are led by subjective or superficial criteria. On the one hand, allocations that are more effective from a technical point of view may conflict with political interests. On the other hand, the consciousness among policy makers that in resource constrained scenarios, every spending decision may potentially entail large opportunity costs in terms of foregone benefits of alternative allocations is only slowly emerging. Allocation decisions are still made based on what is assumed to beneficial, and not based on what is assumed to be most beneficial for poverty reduction.
Due to these difficulties, there are hardly mechanisms for strategic medium and long-term planning of public expenditure. The prevailing approach to make public spending more effective is to prioritize particular components of public spending (e.g. spending on primary education) each of which comprising a range of interventions considered as ‘pro-poor’. However, criteria used to identify these spending categories are typically based on naïve reasoning which does not take into account the complexities of this challenge5. Results-based public expenditure systems introduced in developing countries are essential, but they rely on a poor understanding of the links between inputs and poverty-related outcomes. In addition, the political will among governments and development partners to systematically review budget allocations is often lacking. Development partners often believe that attempts to influence the budget process using economic arguments are futile due to its political nature so that they do not support the development of new analytical tools.
The main purpose of this paper is to present an analytical framework referred to as pro-poor budgeting. It assumes that the overall target is poverty reduction through promoting economic development. It is based on the key principle that public spending should aim at increasing the returns on labour, which is the most important asset of the poor, by removing country-specific constraints on private investment6. The framework hence serves as a basis for medium- and long-term financial planning needed for PRSP implementation by identifying priorities to increase effectiveness. In contrast to prevailing approaches in development policy, this approach is more holistic and takes into account recent advances in development economics. Second, the paper shows how the framework can be implemented and considered in the budget process in practice given that political considerations dominate the budget process.
The remainder of the paper is organized as follows. The second section illuminates the economic background on public expenditure as an instrument for poverty reduction. The third section analyzes the prevailing approach in budgetary decision making to prioritize public spending components. The fourth section introduces a framework which provides guidance about how to make resource allocation more effective. The fifth section indicates how this framework can be implemented in practice. The sixth section concludes.
MKUKUTA is the acronym of “National Strategy for Growth and the Reduction of Poverty (NSGRP)” in Swahili language.
Spending programs for primary education are typically focused on hardware, leading to the construction of a
large number of new school buildings, typically neglecting the software and other complementary factors which are necessary to produce the desired education outcomes.
The World Bank plans to link research on public expenditure with ongoing work on the binding constraints to
growth (Development Committee 2006).