1. Introduction
The Millennium Development Goals (MDGs) are a set of internationally agreed targets that
poor countries aspire to attain by 2015. The eight broad MDGs (as well as the corresponding
targets and indicators) span poverty reduction, primary education, gender parity, child
mortality, maternal mortality, reversal of diseases, environmental sustainability, and
development cooperation.
The MDGs are fast becoming the touchstone for directing as well as assessing socio-
economic progress across the developing world. Increasingly, country-level policy
frameworks such as Poverty Reduction Strategies (PRSs) are linked to the MDG targets;
estimates are now being made of the aid flows required to achieve the MDGs. Crucially, the
MDGs can be a useful focal point for social and political dialogue and action in both poor and
rich nations towards improving the lot of the poor around the globe.
The material and symbolic importance of these targets make it vital to assess the analytical
coherence of the MDG “project”. In this spirit, this chapter highlights complexities and
difficulties of the MDG approach that policy makers should consider. Section two sets out an
analytical framework for analysing the MDGs. The following four sections cover issues of
measuring progress, and achieving and valuing outcomes; sustaining outcomes; devising
policies during structural transformation; and implementing policies in a decentralised policy
system. These discussions draw attention to limitations of current methods of analysing the
MDGs. A final section concludes and proposes a more sophisticated and tractable approach
to the MDGs.
2. Conceptualising the MDGs as a policy “problem”
The MDGs are a set of time-bound, multi-dimensional socio-economic goals almost all of
which can in principle be translated into quantifiable targets and indicators2. Development
policy practitioners commonly believe that there are policy levers which governments can
use to move countries towards achievement of these targets.
Ideally, the “solution” to a policy “problem” then consists of setting instruments’ values so
as to hit the targets, while allowing for institutional changes, uncertainty, and some learning.
In mathematical terms, Tinbergen’s rule states that well-defined solutions in a static
framework require that the number of instruments equals or exceeds the number of
targets3, (Tinbergen (1952)). But economies are dynamic not static systems; policy makers
therefore seek to control the evolving system at a point in time (point controllability) and to
influence its trajectory over time (path controllability). In this dynamic perspective,
Tinbergen’s rule no longer holds for point controllability because rates of change of
instruments as well as their absolute levels can affect objective variables.
2 There are eight overall goals some of which contain a number of specific targets. There are a total of 18
targets.
3 More precisely, the number of linearly independent instruments must be equal or greater than the number of
linearly independent targets (Petit (1990)).