Electricity access more than doubled in Sub-Saharan Africa, and 73 million more people now live without it
Between 2000 and 2024 the share of people in Sub-Saharan Africa with electricity rose from 25.7% to 55.1%, more than doubling. Over the same years the number there with none rose from 505.8 million to 579.0 million, because the population grew faster than the connections. That headcount peaked at 623.0 million in 2015 and has fallen since, so the region is catching up — it is simply still 73 million behind where it started. Worldwide both measures agree: access reached 91.9% and the number without fell by 684 million.
Sub-Saharan Africa more than doubled the share of its people with electricity between 2000 and 2024, and ended the period with 73 million more people living without it than when it started.
Is this happening worldwide?
No — worldwide the two measures agree. Global access rose from 78.2% to 91.9% and the number of people without electricity fell from 1.34 billion to 657.3 million, a decline of 684.4 million. Sub-Saharan Africa is the exception rather than the pattern, which is exactly why quoting only the global percentage, or only the global headcount, would leave a reader with a different impression of the same data.
What accounts for most of the global fall?
India, arithmetically. India went from 60.3% access in 2000 to 99.9% in 2024, and from 420.0 million people without electricity to 1.5 million. That fall of 418.5 million is about 61% of the world’s 684.4 million decline. India is part of the world total, so this is an accounting share of a sum, not a claim that India caused the global change — but no other single country in this dataset moves the world number remotely as much.
Is the number still rising?
No, and this matters enough that the chart would contradict the headline without it. Sub-Saharan Africa’s unserved population peaked at 623.0 million in 2015 and has fallen in most years since, reaching 579.0 million in 2024 — down 44.0 million from the peak. Connections finally overtook population growth around the middle of the 2010s. The 73 million in the title is the gap against 2000, not a claim that things are still getting worse. Quoting the rise without the turn would be the same selective framing this page is about.
Which number is the honest one to quote?
Both, together. The share is the right measure of whether a country is electrifying its population; the headcount is the right measure of how many people the problem still has in it. A tool that picks one for you is picking your conclusion. This is the reason the chart on this page plots the headcount while the text keeps the share beside it, and why the downloadable CSV carries the figures rather than the framing.
What can’t this data tell you?
Whether the power is any good. Access here means a connection exists; it says nothing about outages, hours of supply, cost, or whether the connection carries enough current to run anything beyond lighting. It also cannot tell you why any line moves — grid investment, urbanisation, off-grid solar, birth rates and conflict all sit behind these numbers and none of them are separated out here. And these are modelled estimates built from surveys taken in different years in different countries, so single-year moves should be read as estimates rather than measurements.
How this was measured
- Indicator
- — World Bank — access to electricity, % of population (EG.ELC.ACCS.ZS) and total population (SP.POP.TOTL)
- What it means
- — Access to electricity is the percentage of the population with an electricity connection, compiled for the World Bank from household surveys and national data by the SDG 7 custodian agencies. It counts connections, not reliability, hours of supply or affordability — a household with four hours of power a day is counted the same as one with twenty-four. The people-without figures charted here are not published by anyone: they are population × (1 − access share), computed at the World Bank’s own annual estimates for both inputs.
- Period covered
- — 2000 to 2024
- Published
- — 27 July 2026
- Figures last verified
- — 27 July 2026. Every value in the CSV below was compared against the live source on that date; the page is not edited and this date does not move unless they still match. The source can revise its own figures at any time, and Furlen does not independently audit it.
- Rights
- — World Bank open data (CC BY 4.0) — cited with attribution
Common questions
How many people have no electricity in 2024?
On World Bank figures, about 657.3 million worldwide — down from roughly 1.34 billion in 2000. Sub-Saharan Africa accounts for 579.0 million of the 2024 total, or about 88% of it, having accounted for 38% of the 2000 total. The figure is computed as population × (1 − access share) from the World Bank’s own estimates for both, because no source publishes the headcount directly.
Did Sub-Saharan Africa get worse at electrification?
No, and the data says the opposite. Access more than doubled, from 25.7% to 55.1%, which is the fastest sustained improvement in the series for any region charted here. The unserved population grew anyway because the total population roughly doubled over the same period — the World Bank publishes it as SP.POP.TOTL if you want to check. Improving and finishing behind are not contradictory when the target is moving.
Where does this data come from and can I check it?
The World Bank’s access to electricity series (EG.ELC.ACCS.ZS) and total population series (SP.POP.TOTL), both pulled 2026-07-27, for the World and Sub-Saharan Africa regional aggregates and for India. Using one publisher for both inputs is deliberate: multiplying one body’s population estimate by another body’s access share would make the result partly an artefact of two methods. The exact rows behind the chart are downloadable as CSV from this page, and every cell is re-verified against the live World Bank API on a schedule — including the multiplication, which is redone rather than trusted.
Make one from your own data
This chart was built in Furlen from the CSV above, and every figure in the text is recomputed from those rows before export. You can do the same with a spreadsheet, or ask for a public dataset by name.
More on how claims are checked on the Truth Engine page.