Send money straight to a family in extreme poverty.
No program and no intermediary purchase: a direct mobile-money transfer to a household in extreme poverty, to spend as they judge best.
Choose your ripple
Every.org passes your gift to GiveDirectly and sends the receipt; it suggests a tip you can set to zero. Or give on givedirectly.org.
Hundreds of millions of people live below the World Bank's line for extreme poverty, most of them in sub-Saharan Africa. Cash given directly is spent on whatever the household judges most urgent: a roof, school fees, livestock, a small business. Decades of studies have found it is not squandered, and that the gains last.
The figures
What GiveDirectly does
GiveDirectly identifies households in extreme poverty, enrolls them, and sends money to their phones. Recipients decide what to do with it: a roof, school fees, a business, food, medicine.
How GiveWell gets there
Direct cash is the most-studied intervention on this site. Multiple randomised controlled trials have tracked what recipients do with unconditional transfers and what happens afterwards to consumption, assets, earnings, and psychological wellbeing. GiveWell uses cash as the benchmark unit: every other charity's cost-effectiveness is expressed as a multiple of what the same money would do as a direct transfer.
What could be wrong
- Cash is the benchmark, not the maximum. GiveWell's top health charities are estimated to do considerably more good per dollar than cash, which is exactly why cash is the yardstick rather than the headline pick.
- The ~90% figure is a program-level efficiency ratio, not a guarantee about your specific gift, and it varies by program and country.
- The strongest long-run evidence covers a subset of programs and geographies. Effects measured years out are smaller and noisier than the effects measured immediately.
Every figure is an average program cost, not a receipt. How we phrase things.