Eight goals with deadlines· 3 of 4
The baseline problem
Progress against MDG targets was measured from 1990, a year fixed a decade before the goals existed — which meant the starting line itself shaped every headline.
A decade already in the bank
The eight Millennium Development Goals, adopted in 2000, almost all used 1990 as the reference year for measuring change. The headline target on extreme poverty, for instance, called for halving the proportion of people living on less than a dollar a day between 1990 and 2015. That formulation handed signatories a ready-made head start: a full decade of change — the 1990s — counted as progress even though it preceded the agreement by ten years.
This was not simply a political convenience. Setting 1990 as the baseline gave statisticians a longer run of data and aligned the goals with earlier commitments made at 1990s UN summits. But it had a direct consequence: countries and regions that had moved fastest in the 1990s, East Asia above all, appeared to be dramatically on track by 2005 even when the rate of change after 2000 had slowed. The measured achievement was partly retrospective rather than attributable to any post-2000 programme.
The problem compounds when the 1990 figure itself is examined. For many low-income countries, household survey data from 1990 either did not exist or was thin enough that the baseline was an estimate interpolated from surveys conducted years before or after. The World Bank's PovcalNet (now the Poverty and Inequality Platform) has published successive revisions to these estimates as later data arrived, meaning that the starting line has, in several cases, moved after the race was already being run.
Rebasing of national accounts adds a further layer. When a statistical office revises its GDP series — as Ghana did in 2010 and Nigeria in 2014, in both cases shifting measured output sharply upward — welfare aggregates anchored to the old series can become inconsistent with the new ones. The 1990 baseline was not a fixed point; it was a figure that national statistical offices continued to revise throughout the MDG period.
None of this makes progress unreal. It does mean that headline statements about how much of an MDG target was achieved carry an uncertainty range that official scorecards rarely stated explicitly.
Read next