Worldwide Governance Indicators

Governance can be broadly defined as the set of traditions and institutions by which authority in a country is exercised. This includes (1) the process by which governments are selected, monitored and replaced, (2) the capacity of the government to effectively formulate and implement sound policies, and (3) the respect of citizens and the state for the institutions that govern economic and social interactions among them.

The WGI measure six broad dimensions of governance:

1. Voice and Accountability (VA) – capturing perceptions of the extent to which citizens can participate in selecting their government, including electoral integrity, and perceptions of accountability mechanisms for citizens—reflected in the ability to access information, governmental oversight bodies, and a robust traditional/digital media landscape.
2. Political Stability (PV) – capturing perceptions of the extent to which political power and governance are secure from destabilization, and of the likelihood that authority will be challenged or altered through violent, coercive, or unconstitutional means.
3. Government Effectiveness (GE) – capturing perceptions of the quality of public services, the civil service, policy formulation and implementation, and the credibility of a government’s decisions.
4. Regulatory Quality (RQ) – capturing perceptions of the government’s ability to design and implement policies and regulations that promote private sector development.
5. Rule of Law (RL) – capturing perceptions of the extent to which agents respect and follow the rules of society, including contract enforcement, property rights, the police, courts, and the likelihood of crime and violence.
6. Control of Corruption (CC) – capturing perceptions of the extent to which public power is used for private gain, including both petty and grand corruption, as well as capture of the state by elites and private interests.

The six composite WGI measures are helpful as an initial lens for broad cross-country comparisons and for evaluating broad trends over time. However, they are often too coarse to guide the design of specific governance reforms in particular country contexts. Such reforms, and evaluation of their progress, need to be informed by much more detailed and country-specific diagnostic data that can identify the relevant constraints on governance in particular country circumstances. The WGI is complementary to a large number of other efforts to construct more detailed measures of governance, often just for a single country. Users are encouraged to consult the disaggregated individual indicators underlying the estimated WGI scores to gain more insights into the particular areas of strengths and weaknesses identified by the underlying source data.

The Worldwide Governance Indicators were updated every two years between 1996 and 2002. After 2002, they are updated on a yearly basis. Annual updates are usually released in October of each year.

No, 1996 will remain the starting year. Extending the time series further back would require removing several data sources that only became available later. Doing so would reduce the precision of the estimates (i.e. increase the standard error) and complicate the interpretation of changes over time (since changes in a country’s scores could simply reflect the loss of underlying sources rather than real shifts in governance).

The 2025 WGI update introduced a set of methodological refinements. These include a stricter protocol for screening data sources, the addition of new qualified sources, closer alignment of indicators with an institutional-functions framework, targeted refinements to indicator mapping across the six dimensions, revisions to the aggregation model that allow global averages to vary over time, and the introduction of an absolute 0–100 scale anchored by fixed benchmark countries. As a result, the values in this year’s dataset may differ from versions downloaded in previous years. To ensure a fully consistent series, all historical estimates have been recalculated back to 1996 using the revised methodology.

The entire WGI dataset is publicly available and downloadable in Excel and Stata formats.

Yes.  A full reproducibility package for the WGI is available here.

The WGI draws on 35 data sources that capture the views and experiences of survey respondents and experts across multilateral organizations, non-governmental organizations, academic institutions, and commercial data providers. These data sources include: (a) household and firm surveys (e.g. the Gallup World Poll, the regional Barometers, the World Bank’s Enterprise Surveys, and the World Economic Forum’s Executive Opinion Survey), (b) non-governmental organizations (e.g. Freedom House and Reporters Without Borders), (c) commercial data providers (e.g. the Economist Intelligence Unit, The PRS Group, and S&P Global), and (d) multilateral organizations (e.g. the Country Policy and Institutional Assessments (CPIA) of the World Bank and regional development banks). For a complete list of sources used in the current update of the WGI, click here.

Yes, the raw source data—in its original form, before rescaling—is compiled across all countries and years from 35 sources into a single Excel file. The raw data are reported alongside the minimum and maximum obtainable values, along with the variable orientation, where 1 (0) indicates that higher (lower) values correspond to better governance. Note that for commercial data sources, the data are averaged at the dimension level. This webpage also contains the 35 individual Excel files of raw source data, one for each data source, with additional documentation on the sources.

The WGI draw on 35 data sources that fall into two broad categories: surveys of households and firms (14 data sources) and expert assessments (21 data sources). Household and firm surveys capture the direct perceptions of citizens and business managers regarding governance in the countries where they live and work. Expert assessments reflect the views of specialists engaged and/or convened by the organizations producing the data, many of whom are also based in the regions or countries they evaluate. In terms of geographic diversity, seven data sources are produced by organizations headquartered in the Global South (the African Development Bank CPIA, Africa Integrity Indicators, African Electoral Index, Afrobarometer, Arab Barometer, Asian Development Bank CPA, and Latinobarómetro). Another eight are surveys that collect responses directly from households and firms in developing countries (Americas Barometer, Gallup World Poll, IMD World Competitiveness Survey, Transparency International’s Global Corruption Barometer, World Bank and EBRD’s Business Environment and Enterprise Performance Surveys, World Bank’s Enterprise Surveys, the World Economic Forum’s Executive Opinion Survey, and the World Justice Project). Many data sources—whether survey-based or expert assessments—also maintain offices in developing regions or partner with local institutions. For more information on the WGI data sources, please consult the data files for the individual data sources posted here, as well as Annex 2 of the WGI methodology paper.

A description of the source variables contributing to each of the six WGI dimension scores can be found by clicking on each dimension name:

• Voice and Accountability
• Political Stability
• Government Effectiveness
• Regulatory Quality
• Rule of Law
• Control of Corruption

For a description of each data source and access to underlying data click here.

The WGI standardize and compile information from 35 existing data sources that report the perceptions of citizens, entrepreneurs, and experts from around the world, on various aspects of governance. The WGI data sources are selected based on six criteria:

1. Relevance– Sources must capture subjective perceptions relating to at least one of the six WGI governance dimensions.
2. Originality – Sources must be based on primary data collection from household/firm surveys or expert assessments. Compilations or secondary products derived from other governance indicators are excluded.
3. Cross-country coverage – Sources must cover multiple countries to permit meaningful international comparison, typically at regional, global, or major economic-grouping scales.
4. Regular updates – Sources must be updated on a recurring basis. Annual frequency is preferred, but lower-frequency updates (e.g., every three years) may be considered for high-quality sources with broad coverage, particularly large-scale surveys. 
5. Transparency – Sources must provide well-documented information on their methodology, including respondent selection, data collection processes, scoring methods, and quality assurance procedures.
6. Independence – Sources must be produced by recognized multilateral organizations, nongovernmental organizations, academic institutions, or commercial data providers. Indicators produced solely by national governments are excluded.

Suggestions for additional data sources that meet these criteria are welcome – please email suggestions to wgi@worldbank.org. 

Perception-based measures provide essential insight into how governance functions in practice, complementing formal and administrative indicators. First, people and firms make decisions based on how they experience governance—not just on what laws say. Expectations about whether police will respond, contracts will be enforced, or regulations will be applied fairly shape real-world behavior. Second, for some areas of governance, especially those without a clear “paper trail,” perception data may be the only feasible source of evidence. Corruption is a well-known example: formal audits or prosecutions often capture only part of the picture, whereas surveys and expert assessments provide broader signals of how institutions work in practice. Third, perceptions help reveal gaps between legal commitments and actual performance. A country may have strong laws on equal access to justice, but if citizens report intimidation in courts or businesses expect selective enforcement, these perceptions point to shortcomings not visible in administrative indicators.

At the same time, perception data have limitations. They can be influenced by expectations, ideology, or recent events. Objective information—such as administrative records or audit findings—offers an important complement. A balanced assessment of governance therefore requires drawing on multiple types of evidence.

Data sources based on expert assessments have advantages and disadvantages relative to surveys. An advantage is that they are well suited to cross-country comparisons, given their structured design and quality assurance processes. Expert assessments can also provide more specialized judgments on particular institutions that typical survey respondents may not be able to assess directly. They are also less affected by issues of respondent reticence, which can arise in household and firm surveys when addressing sensitive governance topics. On the other hand, a shortcoming of expert assessments is that they reflect the views of a narrower set of respondents than household or firm surveys, and in some cases their judgments may overlap with those of other expert assessments. To mitigate this, the WGI does not use expert assessments that explicitly draw on other existing data sources.

Individual questions from the underlying data sources are mapped to one (or maximum two) of the six governance dimensions. For example, an establishment-level survey question on the business regulatory environment would be assigned to Regulatory Quality, or a CPIA indicator on transparency, accountability, and corruption would be assigned to both Voice and Accountability and Control of Corruption. A full description of the individual variables used in the WGI and their mapping to the six  dimensions is available here. Each variable is then rescaled to range from 0 to 1, with higher values indicating better governance outcomes. When multiple variables from the same source map to the same dimension, they are averaged so that each data source contributes only one signal per dimension. 

Next, the averaged rescaled data from the individual sources are standardized and combined using a weighted average—estimated through a statistical technique known as an Unobserved Components Model (UCM)—to construct each of the six WGI dimension estimates. Rescaling and standardization adjust for the fact that different indicators use different explicit and implicit units to measure governance. The UCM weights reflect the pattern of correlations among the sources: sources that are more strongly correlated with each other receive more weight, while those that are less correlated receive less. This weighting improves the precision of the aggregate indicators, although a country's governance score remains similar even when unweighted averages are used.

For details, please refer to the WGI methodology paper.

All measures of governance and the investment climate are unavoidably imprecise. The WGI make this explicit by reporting margins of error that show the statistically likely range of a country’s governance score. These margins of error reflect the number of underlying sources and the level of agreement among them: when sources align closely, margins are smaller; when they diverge, margins are larger.

Margins of error also shrink as more sources are available. For example, a country score based on a single source would have a margin of error about twice as large as a country score based on five sources. In recent years, the typical country’s governance score has been based on about 9 separate data sources. These margins acknowledge the inherent difficulty of measuring governance—something that most other governance or investment climate indicators leave implicit. The WGI website reports 90 percent confidence intervals for all scores, and the margins of error have declined over time as the number of data sources has expanded. 

For details on the construction and interpretation of these margins of error, please refer to the WGI methodology paper.

The six WGI dimension scores are estimated using a statistical methodology known as an Unobserved Components Model (UCM). This methodology has two key advantages over other simpler approaches such as simply rescaling and averaging indicators together. First, it provides a way of assigning greater weight to indicators that are more informative about governance, resulting in a more precise overall average. Second, the methodology generates margins of error around the estimates that capture the unavoidable uncertainty in measuring governance, with a clear statistical interpretation.  These margins of error should be taken into account when comparing the WGI governance scores across countries and over time. Please refer to the WGI methodology paper for more details.

The weights assigned to each data source reflect the precision of that source, with more precise data sources receiving greater weight. Precision is estimated using a statistical technique known as an Unobserved Components Model (UCM); see Section 3 of the WGI methodology paper for details. Because countries rely on different sets of sources for each dimension and year, the resulting weights vary across countries, dimensions, and years. Average weights (across countries and years) for each data source and governance dimension are available on the WGI website in Excel and Stata format. Weights for any specific country–indicator–year combination can be viewed using the Excel tool that reproduces the WGI calculations also available on the WGI website.

The six WGI dimension scores reflect a series of decisions about the selection of data sources, the assignment of questions to dimensions, and the approach to aggregation. The definitions of the dimensions themselves also involve judgment, as alternative conceptualizations of governance might imply a different structure than the one used in the WGI. For this reason, the WGI website provides easy access to all source data. The raw underlying source data in a single Excel file is available; alternatively, the averaged rescaled data at the dimension x source level are also available  in Excel and Stata formats. The full reproducibility package replicates the WGI calculations, enabling users to explore alternative ways of organizing the source data and aggregating indicators to suit their specific needs.

When comparing two countries—or one country at two points in time—it is important to take margins of error into account. A useful rule of thumb is that if the two confidence intervals overlap, the difference is not statistically, or likely practically, significant. In other words, small differences or changes in aggregate governance scores should not be over-interpreted when they fall within the reported margins of error; in such cases the WGI are not informative about differences across countries or trends over time. For more details, click here.

As with cross-country comparisons, comparisons over time should consider margins of error. If confidence intervals in two periods overlap, the WGI should not be interpreted as indicating a meaningful change. Most year-to-year changes are too small relative to margins of error to be viewed as statistically or practically significant. Over longer horizons, however—such as a decade—significant trends are evident in a number of countries.

Changes in country scores reflect two factors: (1) movements in the underlying source data that capture changes in governance performance, and (2) the addition of new data sources that become available only in later years or the removal of sources that were discontinued. For large, statistically significant changes over longer periods, changes in the source data are usually the primary driver. For smaller, short-term movements, both factors typically play a role.

Full access to the raw underlying data sources, available here, can help users understand the drivers of changes in a country's score.

Yes, the updated WGI methodology allows the global average of governance to change over time, making it possible to observe a time trend. Please refer to the updated WGI methodology paper for full discussion of global trends across the six governance dimensions.  (Note that in the 2024 and earlier WGI editions, the global average of governance was fixed at zero each year, which made global trends mechanically unobservable.)

Yes. For approximately 60 percent of all pairwise country comparisons in the WGI, the 90 percent confidence intervals do not overlap in 2024, indicating statistically significant differences. Over longer periods such as two decades, about 14 percent of countries show a significant improvement or decline. However, many small changes over shorter periods—or small differences between similarly scored countries—are not significant and should not be over-interpreted.

The six dimensions are not combined into a single overall composite governance score for conceptual and statistical reasons. Conceptually, a single score would be extremely broad and difficult to interpret.  Statistically, many WGI data sources contribute to more than one dimension, creating correlations driven by closely related indicators from the same source, rather than true relationships across dimensions of governance. Some indicators also map to two dimensions, which mechanically introduces correlation across the six dimensions. These features complicate the construction of margins of error—an essential requirement for interpreting cross-country and over-time differences in any overall composite indicator.

The Political Stability (PV) dimension does not measure how long a government has been in power. Rather, it captures perceptions of the likelihood of politically motivated violence, including civil unrest and terrorism. For example, the United States experienced a decline in its PV score between 2000 and 2002. This decline reflected heightened perceptions of terrorism risks following the events of September 11, 2001, rather than changes in the U.S. political process itself. Similarly, countries that may otherwise appear politically stable but are experiencing episodes of civil unrest may also receive lower scores on this dimension.