Zunia

How development got measured, and what the measuring changed.

Everything, in one list →

A household survey questionnaire on a clipboard held at a doorway

The number is made in somebody’s front room

The number is made in somebody’s front room

Behind a headline poverty figure is an enumerator at a door with a questionnaire. Every property of the number — its coverage, its comparability, its uncertainty — starts there.

A household survey questionnaire on a clipboard held at a doorway.

Three rooms, six years, and one instrument each

A question about paperwork →
A conference room after a session, chairs pushed back, papers on the table
After the sessionEvery instrument in this part was agreed in a room of this kind, and what it contains is procedure rather than money.
The courtyard of a ministry building in Paris
Paris, 2005The Paris Declaration on Aid Effectiveness: five principles, fifty-six partnership commitments, twelve indicators of progress.
Two stacks of differently formatted reports side by side
Accra, 2008The Accra Agenda for Action sharpened the framework and put untying and predictability into the text.
A spreadsheet printout with columns of country rows
Busan, 2011The Busan Partnership widened the table and moved monitoring to the Global Partnership for Effective Development Co-operation.
ParisAccraBusanRome
Four roomsThe high level forums and the precursor meeting at Rome. Each pin opens the piece about what was agreed there.

Where the texts were agreed

Rome set the agenda, Paris numbered it, Accra sharpened it, Busan replaced it

The Rome Declaration on Harmonisation of 2003 flagged duplication; Paris attached targets to it; Accra added transparency and predictability; Busan brought in providers outside the OECD Development Assistance Committee and revised the indicators, which is why the two monitoring series are not straightforwardly comparable.

  • Paris Aid effectiveness was a question about paperwork
  • Accra Who owns the plan, and who agreed to stop asking
  • Busan Paris, and the twelve indicators it set
  • Rome The five principles, as the text defines them
“The welfare of a nation can, therefore, scarcely be inferred from a measurement of national income.”

Simon Kuznets · National Income, 1929–1932 · report to the United States Senate, 1934

The number is made in somebody’s front room

A published figure does not say whether it was measured or estimated

Where a household survey exists, the number is counted. Where none exists, it is modelled from whatever is available and printed in the same column, in the same typeface, with the distinction surviving only in a footnote. A rebasing of national accounts can move a headline figure substantially without anything real having changed.

Counted from a surveyEstimated from a model

Behind every international estimate is a sampling frame, a questionnaire and an enumerator at a door. Every property of the number starts there.

All four pieces →

A printed data table with some rows footnoted
The number is made in somebody’s front room

Modelled against measured

Where no survey exists the figure is estimated, and published series rarely make that distinction visible.

Read →
Two editions of a statistical yearbook side by side
The number is made in somebody’s front room

The revision

Rebasing national accounts can move a country’s headline figures by a large fraction overnight, without anything real changing.

Read →
A statistics office interior with staff at desks and printouts
The number is made in somebody’s front room

The offices that publish it

National statistical offices, not international agencies, produce most of the underlying data — and their capacity varies enormously.

Read →
A printed goal framework poster on an office wall
Eight goals with deadlinesThe framework ran from 2000 to 2015: eight goals, twenty-one targets after the 2007 revision, and sixty indicators to file against them.

Start here

Eight goals with deadlines

The MDGs ran from 2000 to 2015 and were the first time the field agreed on a short numbered list with dates attached.

A target that is measured gets managed; one that is not, does not. That is the central criticism of the framework and also the central point of it.

Read the piece →

  1. 1Eight goals with deadlinesThe MDGs, and what a target does.4 pieces →
  2. 2A question about paperworkAlignment, harmonisation and who owns the plan.4 pieces →
  3. 3Cancelled against conditionsHIPC, MDRI and the arithmetic of relief.3 pieces →
  4. 4The number is made in somebody’s front roomWhere the figures come from.4 pieces →

Everything, in one list →