World Bank Renewable Electricity Output (% of Total) Data

It is the renewable share of the grid, one economy and one year at a time. Series EG.ELC.RNEW.ZS, published by the World Bank inside the World Development Indicators, reports renewable electricity output as a percentage of total electricity output for each economy - hydro plus all other renewables in the numerator, everything generated in the denominator. Annual observations run 1990 through 2021 in the current vintage, across about 266 entities: every World Bank economy plus regional and income aggregates such as the Arab World, Africa Eastern and Southern, and Central Europe and the Baltics.

Three numbers frame the file's scale: roughly 10,000-12,000 country-year observations, eight fields per row, and a bulk footprint under 1 MB. In Datadory's renewable-electricity slice it is one of just four datasets that publish an explicit percent-of-electricity figure, and it carries a quality score of 9 out of 10.

The denominator is the detail most analyses get wrong. This is a share of electricity output - generation - not of installed capacity and not of total energy consumption, so a hydropower-heavy economy can score high here while staying oil-dependent elsewhere in its energy mix.

Exactly as they land after normalization - three real observations from 2021, chosen because they show the range the indicator produces:

country=Africa Eastern and Southern  iso3=AFE  year=2021
  value=33.1591703379323

country=Central Europe and the Baltics  iso3=CEB  year=2021
  value=24.8023669805393

country=Arab World  iso3=ARB  year=2021
  value=4.79067848099356

The spread is the point. Africa Eastern and Southern draws close to a third of its electricity from renewable plants, the Central-European-and-Baltic aggregate about a quarter, the Arab World under five percent - one row set returns the whole distribution instead of a map screenshot.

Two things to notice before the first join. These leading rows are aggregates, not countries: the entity list mixes sovereign economies with World Bank groupings, each carrying its own three-letter code (AFE, ARB, CEB), so filter on the ISO3 column before ranking anything. And value arrives as null where nothing was reported - a hole in the underlying statistics stays null, never zero.

Eight verified fields, written against the published row layout rather than guessed:

• Indicator and Country arrive as nested objects carrying a stable id plus a readable label - the indicator object pins EG.ELC.RNEW.ZS to its full definition, the country object pairs a two-letter World Bank id with the display name.
• ISO3 Code is the join-friendly three-letter key, and the single column that separates countries from aggregates when you filter.
• Date is the observation year as a four-character string; Value holds the percentage itself, null wherever no observation exists.
• Unit, Obs Status and Decimal are reporting metadata - the unit rides inside the indicator label for this series, status flags sit empty in normal conditions, and Decimal records the precision each value is published to.

Everything beyond the eight core columns - companion World Development Indicators metadata such as topic classification and the underlying-source note - is confirmed and appended at sample time rather than promised blind.

Geography: worldwide by construction. Every World Bank economy is in the file wherever it sits on the income ladder, and the regional and income aggregates ride along as ordinary rows - useful for continental context, poisonous for rankings unless filtered.

Temporal: 1990 through 2021, annual. The current vintage was compiled from the IEA Energy Statistics Data Browser with an access date of 2025-03-25, and it currently stops at 2021 even though the underlying IEA statistics extend further - treat 2021 as the series' real edge and pair with Our World in Data — Renewable Energy & Electricity Mix when analysis needs later years.

Granularity: one renewable-share figure per economy per year. There is no subnational breakdown and no technology split - separating hydro from wind from solar takes a companion series, and the comparison with the IEA Renewables Data Explorer walks through exactly that trade-off.

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

Pick the channel your stack already speaks. The rows arrive identical either way - one record per economy and year with the indicator identifier, ISO-3 code, value and reporting metadata - extracted so nobody on your team re-stitches a development-indicator database by hand.

Because sibling World Development Indicator series share the same row conventions, this feed snaps into cross-country panels without remapping keys - join on iso3 and year and pull in population, GDP or any neighboring series next.

Every delivery ships with the field dictionary above plus these sample rows for validation, so your first benchmark lands against evidence, not hope.

• Market researchers and consultants benchmark national grids in minutes - 266 economies ranked by renewable share turns 'is this market electrifying cleanly' from a deck anecdote into a sortable column; worked flows continue on our market sizing page.
• Investors and quant researchers screen country exposure before capital commits: the Arab World reading under five percent and the African-aggregate reading near thirty-three price very different transition risks, and feed straight into ESG and emissions analysis.
• Data scientists and ML engineers get a tidy panel - iso3 by year, typed numerics, preserved nulls - that joins to any neighboring World Development Indicator series on identical conventions; see ML model training.
• Economists, academics and policy teams cite a stable, long-run series - up to 32 annual observations per economy back to 1990 - that has anchored cross-country studies for decades; see citation-grade research.
• Developers and builders read the aggregate rows as demand signals - markets where the renewable share climbs fastest are the ones adding interconnection, storage and balancing work next.

Market researchers and consultants get the benchmarking layer - one percentage per economy per year, comparable across every continent, with no currency or unit conversions to defend. Investors and quant researchers get a country-risk factor that predates the ESG label by decades, with nulls preserved so absent data stays absent instead of masquerading as zero. Data scientists and ML engineers get the cleanest panel shape in the catalog - eight typed columns, sub-1 MB, ready for a groupby before the coffee cools. Journalists, academics and students get numbers that resolve to a definition rather than a press release: EG.ELC.RNEW.ZS names exactly what was measured, for which economy, in which year.