Do 449 million people still have working off-grid solar?
Energy-access funding draws on a solar reach estimate built from sales, not checks of working devices in homes.
The question the lab is investigating
How many of the roughly 449 million people that the official SDG7 tracking report says were served by off-grid solar in 2024 still have a working solar device, judged against public household surveys instead of company sales records and an assumed product lifespan?
What's going on
- An official report estimates off-grid solar served 449 million people in 2024.
- The count relies on company sales, assumed market share and assumed device lifespans.
Why it's stuck
- Surveys often record the main lighting source, not which solar devices still work.
- Even the best available survey cannot reveal device survival on its own.
What we'd try
- The lab would rebuild the sales-based count and compare it with household surveys.
- It would test whether revised assumptions better predict surveys left out of the fitting.
Where we stand
- On hold: the count's owners are revising it, and no user for this audit was found.
- Sparse surveys may leave only a few countries to compare, with wide uncertainty.
- Reopening needs usable device histories, plus a willing user or published revised counting rules.
Written by the lab's AI (openai/gpt-6.1-sol) and checked against the lab's record by a model from another company (google/gemini-4-argon). The record has changed since. A new summary is on its way.
- Good it would do
- 3 / 5
- Can AI help
- 3 / 5
- How neglected
- 4 / 5
- Can we check it
- 4 / 5
Read the lab's full wording
Why it's on hold
Parked: nobody has checked the 449 million sales-based count against households, but its owners are revising it now (MTF calibration dropped; GOGLA lifespan rule to be revised within a year, likely upward), surveys mostly record main lighting source rather than working Tier 1+ devices, and the best survey (Kenya MTF) cannot identify device survival on its own. No decision user found.
Why it's stuck
The headline number is built from sales, not from looking at homes, and it sits between the solar industry, which produces it, and survey statisticians, who rarely check it. The 2026 Tracking SDG7 chapter on electricity access says its estimate of 449 million people served by Tier 1 and Tier 2 off-grid solar in 2024 is 'based on sales data reported by GOGLA affiliates', that affiliates are assumed to hold a 29 percent market share, and that 'Product lifespans are assumed to be 1.5x warranty periods ... typically three years for solar home systems'.
What we'd try
Rebuild the sales-based estimate country by country (the published GOGLA country sales series and the stated lifespan, market-share and discount assumptions), then compare it with what nationally representative household surveys found in the same country and year: the World Bank's Multi-Tier Framework surveys (public-use microdata, for example Kenya 2016-2018, which records solar-based devices), and the solar questions in LSMS and the DHS rounds that include them. Fit the device lifespan and non-affiliate share that best reconcile the two, with uncertainty, and report which countries' official numbers look too high or too low.
We'd look again if
Both: (1) a reachable household survey or panel is shown to record past, failed or discarded solar devices or purchase histories well enough to estimate device survival in at least one country without assuming market share (check newer MTF rounds, LSMS-ISA panels, the Malawi panel); and (2) a named user (GOGLA Impact Metrics consultation or the Tracking SDG7 team) says it would use household-based survival evidence, or the revised Impact Metrics are published so the audit targets the rule actually in use.
Think we got it wrong?
Ruled-out problems can come back. Tell us what we missed.
Members can read the full record for this problem, check its sources and comment. Open the member view → Not a member? Here is how to join.