State capacity metrics: what the standard indices miss
Composite governance indices are widely cited but rarely interrogated. A closer look at three leading indices shows they agree less than most users assume.
Governance indices are routinely used to rank countries, justify aid conditions and benchmark reform. Three of the most cited composite indices claim to measure a similar underlying concept, described variously as "state capacity" or "government effectiveness". Our analysis compares their country rankings across the 2015–2024 period.
The correlation between index pairs is weaker than commonly assumed, particularly among middle-income states where capacity is neither clearly strong nor clearly weak. Two countries can trade places by ten or more ranks depending on which index a report cites, without any change in the underlying data.
Much of the divergence traces to the differing weight given to perception-based survey items versus administrative data such as tax collection or civil-service turnover. Indices that lean heavily on expert-perception surveys are more sensitive to recent news coverage than to underlying administrative performance.
None of this means composite indices are without value; it means they should be read as one input among several, with attention to what each index actually weights, rather than treated as a single settled number. Our data appendix sets out the item-level composition of each index for readers who want to check the sourcing themselves.