​The Millennium Development Goals (MDGs) have become a key metric for measuring the performance of developing countries in addressing critical development challenges. The goals are expressed in concrete, time-bound targets set in either relative terms – for example reducing relative poverty by half – or in terms of ‘getting to zero’ – for example achieving universal primary education. While originally framed as global targets, the MDGs have been widely applied as national targets to benchmark progress.

This country-level application has been problematic in some cases, however, and promoted a misperception of real progress made. This is because applying the same targets to all countries suggests similar efforts will result in similar ‘gains’ across different countries. But our analysis – alongside that of others – shows progress is rarely linear: improvements in people’s lives across different dimensions occur at varying rates across countries. For some MDG targets, progress has been faster for countries further from a target; for others it has been slower.

To examine the true patterns of progress on the MDGs, this paper explores seven indicators – one representing each of the first seven MDGs. For all indicators but extreme poverty, we find that typically, progress is easiest to attain for countries that are relatively deprived (or further from the target), though there are important differences between indicators. Progress is notably non-linear for educational enrolment, maternal mortality and HIV/AIDS – areas where there is most reliance on government services. Stipulation of similar rates of progress or absolute goalposts means these non-linearities are overlooked, and targets are framed in an overly ambitious way for many countries.

In most assessments of MDG progress by major international institutions, countries are classified as either ‘on track’ or ‘off track’ by comparing required rates of progress to meet the MDG target with a country’s actual performance. While recognising data gaps, we show that up to 46% of countries – depending on the indicator – have registered better-than-expected progress on a number of MDG targets even though they may not be ‘on track’ to meet them.

How can targets be set in a non-linear world with countries at different stages of development? Setting targets using methods that recognise non-linearities in progress and initial starting points could help close the gap between normative goals and national realities. A global goal to ‘get to zero’ on maternal mortality, for example, could be calibrated to the speed of progress that might be expected given a country’s initial level. To mobilise efforts and encourage action, it is important to set challenging but attainable MDG targets. This paper argues that there is a need to find a middle ground: to maintain the power of a unified set of goals while bringing in greater sensitivity to national realities. This focus would help bridge the gap between expectations and achievements.

Key messages

• Improvements across different MDG dimensions are often non-linear: they occur at varying rates at different times in different countries. On many MDG targets, countries’ starting positions strongly condition progress.
• When applied at country level, MDG targets require overly ambitious performance for many countries because they stipulate similar rates of progress or the need to reach absolute goalposts.
• Many poor countries – up to 46%, depending on the indicator – have registered better-than-expected progress on some MDG targets, even though they are not ‘on track’ to meet them.
• In a post-2015 agreement, to provide more appropriate incentives for individual countries to ‘leave no one behind’, setting targets in a technical way could help bridge the gap between normative commitment and greater sensitivity to national realities.
  

Laura Rodriguez Takeuchi and Emma Samman, with Liesbet Steer