World Bank Jobs Data

World Bank Jobs Data is how the whole planet fits onto one labor spreadsheet. The World Bank's jobs topic bundles hundreds of labor-market indicators drawn mainly from the World Development Indicators, with the ILO Modelled Estimates database (ILOSTAT/ILOEST) as the primary underlying source - so unemployment in Peru and participation in Poland arrive measured the same way. Featured series include Unemployment, total (SL.UEM.TOTL.ZS), labor force participation rate (SL.TLF.CACT.ZS) and wage and salaried workers as a share of total employment (SL.EMP.WORK.ZS), backed by employment by economic activity and status, vulnerable employment, working poverty, youth unemployment and child labor.

The extracts land wide: Country Name, Country Code (ISO3, e.g. GBR), Indicator Name, Indicator Code, then one column per observation year from 1960 to 2025, blank where missing. Each bundle travels with country classification metadata (Region, IncomeGroup, SpecialNotes, TableName) and indicator provenance notes naming the originating organization. Scale runs from a single indicator - 265 country rows by roughly 66 year columns, about 140 KB - to a full WDI extract of several GB. Nothing else in the human resource and employment services industry catalog reaches this geography: roughly 265 economies against the American series' one.

Datadory reshapes those extracts into typed country-year rows for the human resource and employment services industry that join against openings, payroll and vacancy feeds. Get a sample of this dataset and put 265 economies next to the two or three you chart today.

Rows captured during the August 2026 research pass, exactly as they land from Datadory - one row per economy-year, values in percent of the labor force:

country_name     country_code  indicator_code   year  value
United Kingdom   GBR           SL.UEM.TOTL.ZS   2023  4.025
India            IND           SL.UEM.TOTL.ZS   2023  4.172
Aruba            ABW           -                -     -

Three rows and the shape of the collection shows itself. The first two are the same indicator read in two very different labor markets - a 4.0 percent developed-economy unemployment print beside a 4.2 percent reading for an economy of 1.4 billion people, directly comparable because both carry the modeled-ILO-estimate methodology rather than two national definitions of who counts as unemployed. The third row looks empty and is not: Aruba arrives with no unemployment value on this pass but with its classification metadata intact - Latin America & Caribbean, high income - which is how small economies enter the panel long before any given indicator fills in for them.

That blank-value pattern is worth planning around rather than cleaning away. Roughly 265 economies times hundreds of indicators produces a matrix with real holes in it, concentrated in the earliest years and in smaller states; treating missingness as signal (which economies publish labor statistics at all) is often more informative than imputing over it.

Each economy-year observation carries an identity block (economy name, ISO3 code, indicator name and code), a value, and the classification metadata that rides along in every bundle. Definitions below are verified against the record card, not inferred.

Two structural choices matter more than the field count. First, the identifier discipline: indicator codes such as SL.UEM.TOTL.ZS pack topic, measure and unit into one string, so a filter on any dimension is a string operation rather than a lookup across tables. Second, the wide-versus-tidy question: extracts ship one column per year, while Datadory can deliver the same observations as tidy country-year rows - pick whichever your warehouse prefers, because the underlying observations are identical.

Additional fields on request. Beyond the core row shape, the wider collection spans indicator provenance notes naming the originating organization, the SpecialNotes and TableName identifiers attached to each economy, and sex and age-group disaggregations on selected series. Each gets pinned down against your sample before you commit, so the dictionary you buy is the dictionary you tested.

• Geo: roughly 265 World Bank member economies and aggregates worldwide - from the largest labor markets to island states - each carrying region and income-group classification metadata.
• Temporal: annual observations from 1960 through 2025, with most labor series concentrated between 1991 and 2025; the early decades thin out fast outside large economies.
• Granularity: one value per economy-year per indicator, with selected series disaggregated by sex and age group - youth unemployment being the canonical cut.

Breadth is the product here, and the aggregates are half of why. Regions, income groups and lending categories appear as entries in their own right, so a Europe-and-Central-Asia rollup costs no aggregation work - but the flip side is that a careless global average counts Germany twice, once alone and once inside Europe. Filter aggregates out of country-level panels and lean on the WDI aggregate rows distinction when the rollup is the deliverable.

API, files, or your warehouse. Daily, weekly, or hourly.

Pick the indicators, pick the cadence, pick the landing zone - the same rows arrive whichever way you take them. The annual rhythm of the underlying series is a property of macroeconomic statistics, not of your pipeline: once the panel lands in your warehouse on your schedule, every downstream model reads the same table. The sample comes first, so the indicator mix and the cadence are settled facts before any commitment.

• Market researchers and consultants put client markets side by side in one comparable frame - participation rates, wage-worker shares, sector mixes across dozens of economies in a single deck slide; the market researchers use cases page shows where the panel slots into broader work.
• Investors and quant researchers screen countries on unemployment trajectories and participation trends for allocation and expansion signals; the investors quants use cases page covers the workflow.
• Data scientists and ML engineers build the country-year labor panel behind cross-country features and backtests, keyed on ISO3 code and indicator code; the data scientists use cases page maps the modeling plays.
• Sales and growth teams size labor pools and wage-earning populations when picking expansion countries for staffing and hiring clients; the sales growth teams use cases page walks the territory-selection play.
• Competitive intelligence teams contextualize a competitor's new markets with the labor-force fundamentals beneath them; the competitive intel product teams use cases page covers the setup.
• Journalists and academics cite World Bank labor indicators with ILO-sourced methodology in papers and explainers, quotable across any border; the journalists academics use cases page has the citation patterns.

Teams wiring the feed into dashboards will find delivery-side detail in the developers builders use cases page.

• BLS Job Openings and Labor Turnover Survey (JOLTS) - monthly openings, hires and separations for the United States; the World Bank panel supplies the international frame around the American cycle.
• BLS QCEW - Employment Services NAICS 5613 - county-level establishments, employment and wages for the staffing industry itself, the supply side beneath any global labor narrative.
• BLS Occupational Employment and Wage Statistics (OEWS) - occupational wage percentiles across about 530 areas, turning country comparisons into compensation benchmarks.
• California EDD WARN Notices - named-employer layoff filings at event grain; there is a direct head-to-head in the World Bank Jobs Data vs California EDD WARN Notices comparison if you want breadth against grain.
• **O*NET Database (downloadable)** - occupation requirements and task ratings to attach skills context to any labor-pool reading.
• World Employment Confederation Statistics - agency-work market size and penetration rates across roughly 40 countries, the private-employment-services layer on top of the macro panel.

Two vocabulary notes sharpen the setup before you commit: how a country-year panel structures repeated observations, and what ISO country code fields guarantee when joining economies across sources. The wider shelf sits on the best human resource & employment services datasets ranking.