Transparency International’s Corruption Perceptions Index was updated this summer and before we dive into the lessons learned on the country-level, it is worth to see how the CPI got to be one of the most important public metrics of how corrupt countries are.
The Corruption Perceptions Index (CPI) is the most widely used global indicator of public sector corruption, published annually by Transparency International since 1995. It ranks countries and territories on a scale from 0 (highly corrupt) to 100 (very clean), based on expert and business executive assessments of corruption levels in the public sector. The 2025 edition covers 182 countries and territories, making it the most comprehensive anti-corruption ranking available.
History and Development
The CPI was first introduced in 1995 by Transparency International, a non-governmental organization founded in 1993 to combat corruption through collaboration between business, civil society, and government. Initially covering fewer countries, the index has expanded over three decades to include approximately 180 countries annually.
A significant methodological update occurred in 2012, when the scale was standardized to 0–100, enabling year-on-year score comparisons for the first time. Prior to 2012, longitudinal comparisons were not methodologically valid due to changes in data sources and calculation methods.
How CPI Scores Are Calculated
The CPI is a composite indicator constructed through a rigorous four-step process:
The index draws upon 13 distinct data sources from 12 different institutions, including the World Bank, World Economic Forum, African Development Bank, Bertelsmann Stiftung, Economist Intelligence Unit, and Varieties of Democracy Project. Each source must meet strict quality criteria.
All source data are standardized to a 0–100 scale, where 0 represents the highest perceived corruption and 100 the lowest. Three sources are reverse-scored to maintain directional consistency.
Each country’s CPI score is calculated as a simple average of all available rescaled scores for that country. A minimum of three data sources is required for a country to receive a score. Approximately 180 countries meet this threshold annually; absence from the index indicates insufficient data, not an absence of corruption.
Each CPI score is reported with a standard error and 90% confidence interval to reflect variance among source data. Only changes confirmed by a majority of underlying data sources are considered statistically significant.
What the CPI Measures
The CPI specifically assesses the following manifestations of public sector corruption:
- Bribery of public officials
- Diversion of public funds
- Officials using public office for private gain without consequences
- Government capacity to contain public sector corruption
- Excessive red tape creating corruption opportunities
- Nepotistic appointments in civil service
- Financial disclosure and conflict-of-interest laws for officials
- Legal protection for whistleblowers and corruption reporters
- State capture by narrow vested interests
- Public access to government information
The Caveats
The CPI measures perceptions rather than actual corruption levels. Critics argue that perception-based data can be subjective and influenced by media exposure, cultural biases, availability heuristics, and the halo effect. Experts and business elites may have conflicting understandings of corruption that do not align with local definitions.
The exact set of sources used for each country varies annually, and question wording differs across surveys. This makes precise estimation of net changes in corruption perception difficult.
The standardization process, particularly the reversal of certain source scores, can create inconsistencies. Countries with complex regional corruption dynamics may be inadequately captured by standardized scales.
Transparency International does not publish original scores from private data sources, limiting full transparency in CPI construction. While the organization verifies methodologies with source institutions, researchers cannot independently validate all underlying data.
It can be seen that although it has shortcomings, the Corruption Perceptions Index remains the most comprehensive and widely cited global measure of public sector corruption, despite acknowledged methodological limitations. By synthesizing multiple expert assessments into a single comparable indicator, it provides valuable insights into cross-national corruption patterns and trends. Users should interpret CPI scores alongside complementary data sources and recognize that the index measures perceptions rather than absolute corruption levels.