GRIPS Development Forum > Diversifying PRSP > ch2. Global Development Trend and PRSP

"Goals for Development: History, Prospects and Costs"
written by Shantayanan Devarajan, Margaret J. Miller, and Eric V. Swanson
World Bank Discussion Paper No. 2819, April 2002

Main Points:

This paper provides a preliminary estimate of the additional aid required if countries would work vigorously toward meeting the Millennium Development Goals (MDGs). The authors have adopted two approaches in estimating the resource gap.

--> paper "Goals for Development: History, Prospects and Costs" pdf file

Approach 1: estimating the additional resources necessary to increase economic growth so as to reduce income poverty

The following steps are taken to calculate additional aid requirement for halving poverty.

  • use a simple, "two-gap" growth model in which growth depends on the level of investment and the efficiency with which investment is turned into output (ICORs).
  • assume that for a given rate of growth per capita GDP, the rate of poverty reduction depends on the shape of the income distribution and the level of average income relative to the poverty line.
  • estimate the amount of additional investment needed (= ODA for the poorest countries), by calculating the average rate of growth required to reach the income poverty goal in 2015, based on the existing poverty level and income distribution.

Furthermore, to better reflect the factors for aid effectiveness, developing countries are divided into the following groups, and necessary adjustments are made in applying the above growth model.

  • Group I: 86 countries where additional aid has little impact on poverty reduction (aid has little effect on growth rates, for being "on track" with the current resources etc.);
  • Group II (primarily low-income countries): 65 countries which are unlikely to achieve the MDGs without aid. This group can be further divided into: (i) countries with good policies and (ii) countries with poor policies.

It is estimated that incremental costs of achieving this goal to be in the range of $54-$62 billion.

Approach 2: estimating the cost of meeting specific goals in education, health and environment

The authors acknowledge the difficulty in calculating the cost of attaining the social goals, due to the interdependent nature of these goals and a danger of double-counting. (For example, child mortality reduction is significantly achieved by increasing the number of mothers with primary education.) Thus, the paper has made use of existing estimates of some of the individual goals, some of which are based on country-level analysis and judged relatively reliable. Moreover, the following steps are taken.

  • In education, a unit cost approach has been used to estimate the cost of attaining universal primary education by 2015. Also, the cost of meeting the 2005 gender equality target in secondary education is calculated by assuming constant average costs for enrollment and then increasing the numbers of girls in school so that the ratio of girls to boys would be 1:1: by 2005.
  • In health, critical inputs for selected health interventions are identified and costed including: (i) rehydration therapy, vaccinations and promotion of breast feeding (for infant mortality reduction); (ii) prophylaxis treatments for pregnant women and mosquitoes (for malaria control); (iii) DOTS program (for tuberculosis control); (iv) the costs of prevention and treatment campaigns (for AIDS control).

These estimates yield a rough breakdown of the additional costs of achieving the social goals as follows:


$10-$30 billion

Health $20-$25 billion
Environment $5-$21 billion


$35-$76 billion

In conclusion, the paper states that because the estimate using the two approaches are in the same range, there is some reason to think that an increase in foreign aid of an amount equal to current foreign aid ($57 billion) is about the right order of magnitude for achieving the development goals.

Comments (by GRIPS Development Forum)

This paper was prepared at the request by Bank senior management to calculate the World Bank's estimates of costing the MDGs. We consider that while this is a timely effort, the paper needs more careful examination, particularly on the appropriateness of its underlying assumptions. At least, the following issues should be further analyzed.

  • The paper is based on two approaches, which are quite different in their assumptions of poverty reduction mechanisms as well as their policy implications. To reduce poverty, the first approach supports the policy measures to promote broad-based growth (investment-growth-poverty reduction), while the second approach supports the pro-poor targeting measures. The paper does not explain clearly the relationship between the two different approaches.
  • Neither is it clear why additional resource requirement must entirely come from ODA. Mobilization of private investment and domestic savings should also play an important role and be promoted.
  • The paper appears to be optimistic about the absorptive capacity of aid recipients, when aid is doubled.
  • This paper does not address alternative ways of allocating resources to attain the MDGs. Although the estimate assumes country-based poverty reduction efforts, it may be more efficient to geographically concentrate resources on those countries where the absolute number of the poor is significant (e.g., China, India).

Note: The comments are based on our interview with one of the authors in May 2002.