Loading data…
State capacity over time
Pick up to eight countries. Grey shows the median and interquartile range of all countries observed that year.
State capacity around the world,
Click a country to add it to the comparison. Hatched areas have no score for that year.
Rankings
Countries ranked by the selected index. Click a row to add the country to the comparison.
| Rank | Country | Score | Change in score | Change in rank |
|---|
What drives the index
The six V-Dem indicators behind the index for one country, against the world median. The weight is the indicator’s fixed loading in the selected index (Table 2 of the paper).
Download the data
Annual country-year panels with all three indices and the six V-Dem inputs. Data are released under CC BY-SA 4.0. Please cite the paper and V-Dem (see Method & citation).
Variables
| Variable | Description |
|---|
Earlier update files named the narrow indices c_base and c_fiscal; these are now c_nar and c_fisc, following the paper. The values are unchanged. Every number on this site is reproduced by the build pipeline, which checks itself against the published paper and earlier releases.
Method
State capacity is the ability of a government to achieve the goals it sets: raise revenue efficiently, control its territory, enforce the law impartially, and provide public goods that support markets. The index is the first principal component of six indicators from the Varieties of Democracy (V-Dem) dataset, each covering one of these dimensions.
Factor loadings from the principal component analysis on V-Dem v9 (Table 2 of the paper). Inputs are standardised before weighting, so a loading is the weight of a one-standard-deviation change in that input.
Three versions
- Narrow (
c_nar) uses rule of law, state authority over territory, impartial public administration and public goods spending. It has the longest coverage, back to 1789 for more than 20 countries. - Narrow with fiscal capacity (
c_fisc) adds the state’s main source of revenue. Fiscal data are missing for some important cases, for example the United States between 1920 and 2005. - Comprehensive (
c_comp) adds educational equality. It covers every dimension but starts in 1900.
Updates to newer V-Dem releases
The original series is the published one, built on V-Dem v9 and ending in 2018. Each update applies the same fixed v9 loadings to a newer V-Dem release. The inputs are standardised on that release’s own sample, so each vintage has mean zero. Because V-Dem re-estimates its whole history with each release, an update revises earlier years slightly as well as adding new ones. Correlations with the original series on overlapping years are about 0.99. The download files also include pca_* columns that re-estimate the principal component on the newer data, for robustness checks.
Summary statistics. Scores are principal-component scores and are not rescaled, so the standard deviation equals the square root of the first eigenvalue. The original rows reproduce Table 3 of the paper.
How to cite
If you use the index, please cite the paper:
O’Reilly, Colin, and Ryan H. Murphy. 2022. “An Index Measuring State Capacity, 1789–2018.” Economica 89 (355): 713–745. https://doi.org/10.1111/ecca.12411
@article{OReillyMurphy2022,
author = {O'Reilly, Colin and Murphy, Ryan H.},
title = {An Index Measuring State Capacity, 1789--2018},
journal = {Economica},
volume = {89},
number = {355},
pages = {713--745},
year = {2022},
doi = {10.1111/ecca.12411}
}
Data source and attribution
All six inputs come from the Varieties of Democracy (V-Dem) Project, which made this index possible. V-Dem asks users to cite both the dataset and its measurement model. Please cite the V-Dem release that matches the vintage you use:
Pemstein, Daniel, Kyle L. Marquardt, Eitan Tzelgov, Yi-ting Wang, Juraj Medzihorsky, Joshua Krusell, Farhad Miri, and Johannes von Römer. “The V-Dem Measurement Model: Latent Variable Analysis for Cross-National and Cross-Temporal Expert-Coded Data.” V-Dem Working Paper No. 21. University of Gothenburg: Varieties of Democracy Institute.
V-Dem data are licensed under Creative Commons Attribution-ShareAlike 4.0. The index is a derivative work and is released under the same licence: you may share and adapt it, including commercially, provided you give credit and share adaptations under the same terms. The indicator values here are V-Dem’s. Errors in constructing the index are the authors’ own; V-Dem does not endorse this index.
Map boundaries: Natural Earth, via the world-atlas project (public domain). Borders are modern and do not imply any position on disputed territory.