The World Inequality Database (WID) provides open access to the most extensive database on the historical evolution of economic inequality. Updated annually by the World Inequality Lab in collaboration with a global network of researchers, the WID represents the most up-to-date and coordinated effort to measure and track inequality worldwide.
Each year, our regional coordinators work with this network and statistical agencies to improve our methodology and expand existing series. The WID now covers 216 countries, with data going back as far as 1820 to 2024 for some countries. Our latest update includes new data points and revisions of earlier estimates for pre-tax income data, gender inequality data, and wealth inequality data. Here are five highlights:
1. Updated regional income inequality ranking
We accessed new administrative data for Brazil and Chile revealing that inequality at the very top (within the top 1%) was previously underestimated. As a result, Latin America once again emerges as the most unequal region, overtaking MENA, when comparing the income shares of the top 10% with those of the bottom 50%. Europe remains the least unequal region in the world.
For more details about the income distributional updates by region, read the Technical Notes.
2. Gender gaps in labour income persist worldwide
This update substantially improves the measurement of Female Labor Income in China; for other countries, the incorporation of new data and methodological refinements mainly strengthens data quality and confirms previously observed regional trends in Female Labor Income Shares.
For China, we find that the decrease in Female Labor Income Shares over the last three decades, from 44.7% in 1988 to 36.2% in 2018, is linked to the country’s structural transformation. This pattern mirrors a fall in the Female Employment Share, which declined from 49% to 43.1%. The drop reflects both the rapid expansion of wage employment and a decrease in women’s participation in wage jobs. While men moved more quickly from agricultural self-employment into wage employment, women remained more concentrated in declining agricultural activities and, to a lesser extent, in non-farm self-employment. As a result, structural change translated into a declining female share of both employment and labor income.
For more details about the income distributional updates by region, read the Female Labor Income Share Technical Note.
3. Extended historical coverage of wealth inequality
We extended the wealth distribution series from 1980 worldwide, and back to 1820 for a limited set of countries. Our current wealth inequality estimates remain unsatisfactory due to limited access to country-level household wealth survey and tax data.
For more details about the methodology for wealth inequality, read the this Technical Note.
4. A revised methodological framework: DINA Guidelines 2025
The 2025 Distributional National Accounts (DINA) Guidelines are the go-to methodological document for researchers who want to understand the concepts, data sources, and methods used in the WID. Edited by Lucas Chancel, Ignacio Flores, Rowaida Moshrif, Gastón Nievas, and Thomas Piketty, the guidelines document how national income and wealth totals are consistently distributed across individuals and percentile groups, ensuring full compatibility with macroeconomic aggregates.
After their initial publication in 2016 and major revision in 2020, the database has expanded quickly, with new variables, improved estimates, and wider coverage. This third edition consolidates this progress by introducing:
- revised macroeconomic aggregates,
- updated income and wealth definitions,
- enhanced data-quality flags that capture changes over time and across series,
- new sections on wealth distribution and gender inequality.
To make the most of the WID, we recommend consulting the DINA guidelines and the other methodological resources listed here.
5. Inequality data remains scarce or inaccessible
Our objective is not to claim that we have perfect data series, but rather to make explicit what we know and what we do not know. We attempt to combine and reconcile in a systematic manner the different data sources at our disposal. The research papers upon which our series are based are available on-line and present our methods and assumptions in the most transparent manner. Raw data sources and computer codes are released so that our work can be extended and improved by others. If you want to contribute to a data series or have any technical questions, get in touch.
