●●● high
The struck-through letters are the unreachable upper bound (ceiling). A confirmed − (below) sets the ceiling; independently verified + decide the position within it.= non-additive meter
GiveDirectly: Direct, unconditional cash to people in extreme poverty. What happens if aid hands poor people cash itself—rather than in-kind goods or programs—directly? GiveDirectly has answered that question with some of the most rigorous evidence available. Founded in 2009 by Harvard/MIT economists (Michael Faye, Paul Niehaus, Jeremy Shapiro and others), it gives people living in extreme poverty cash with no strings attached, delivered mainly via mobile money (such as M-Pesa). It has reached Kenya, Uganda, Rwanda, Malawi, the Democratic Republic of the Congo, Liberia, Morocco, and even disaster survivors in the United States, delivering more than US$1 billion in cumulative cash. It was the first nonprofit dedicated to cash transfers and was named a GiveWell Top Charity (2012–2022). The evidence is unusually strong: the randomized controlled trial by Haushofer & Shapiro (2016, Quarterly Journal of Economics) found that recipient households in Kenya improved their assets, consumption, food security, and psychological well-being (less stress), with no rise in alcohol or tobacco. A large basic-income experiment in Kenya (by Banerjee, Suri and others, across thousands of people over a decade) likewise shows that lump-sum and long-term transfers raise enterprise and income. It is an organization that has used data to overturn the conventional belief that “poor people waste cash.” The letter is A; certainty is high. In 2022, in the DRC (South Kivu), staff colluded with an outside vendor and a former employee to divert cash transfers meant for about 1,900 households (about US$1.2 million by the final investigation) (category i—an effect on extremely poor households, a protected stakeholder—and category iii). GiveDirectly self-disclosed and publicly apologized in 2023, responding with external audits (Grant Thornton / Humentum) and resumption of the original payments. The loss equaled about 0.8% of that year’s global disbursements. Responsibility is recorded as a control failure, but given the self-detection, public disclosure, and resumption of payments to recipients, the ceiling is left in place (A retained—though B is tracked as a reasonable alternative reading). sets the assessment’s upper bound. Unconfirmed concerns are placed under “Watching.” (As of 2026-Q2; estimate based on public information.)
Main narrative
What happens if aid hands poor people cash itself—rather than in-kind goods or programs—directly? GiveDirectly has answered that question with some of the most rigorous evidence available. Founded in 2009 by Harvard/MIT economists (Michael Faye, Paul Niehaus, Jeremy Shapiro and others), it gives people living in extreme poverty cash with no strings attached, delivered mainly via mobile money (such as M-Pesa).
It has reached Kenya, Uganda, Rwanda, Malawi, the Democratic Republic of the Congo, Liberia, Morocco, and even disaster survivors in the United States, delivering more than US$1 billion in cumulative cash. It was the first nonprofit dedicated to cash transfers and was named a GiveWell Top Charity (2012–2022). The evidence is unusually strong: the randomized controlled trial by Haushofer & Shapiro (2016, Quarterly Journal of Economics) found that recipient households in Kenya improved their assets, consumption, food security, and psychological well-being (less stress), with no rise in alcohol or tobacco. A large basic-income experiment in Kenya (by Banerjee, Suri and others, across thousands of people over a decade) likewise shows that lump-sum and long-term transfers raise enterprise and income. It is an organization that has used data to overturn the conventional belief that “poor people waste cash.”
One person’s story (N1)
+ before → after
A household living in extreme poverty in rural Kenya or Uganda. Until now, a thatched roof that had to be re-laid every rainy season drained their money; with no savings and no collateral, a little bad luck could undo their lives. When unconditional cash (on the order of several hundred to a thousand dollars) arrived on their phone from GiveDirectly, the household first switched to a rust-free metal roof, invested in livestock and a small business, and paid school fees. Rather than having someone else decide how the money is spent, they put it toward the needs they know best—and the RCT showed that, in doing so, their assets, their table, and their peace of mind all grew, while alcohol and tobacco did not.
Source nature: Haushofer & Shapiro / Quarterly Journal of Economics / P1 peer-reviewed academic study (QJE). Positive effects are not used to offset negatives.
Positive / negative effects
+ effects
- GiveDirectly was the first nonprofit dedicated to cash transfers, a GiveWell Top Charity (2012–2022), and has delivered more than US$1 billion in cumulative cash directly to people in extreme poverty (Kenya, Uganda, Rwanda, Malawi, the DRC, Liberia, Morocco, US disasters, and more). The randomized controlled trial by Haushofer & Shapiro (2016, QJE) found that recipient households improved their assets, consumption, food security, and psychological well-being, with no rise in alcohol or tobacco. A large basic-income RCT in Kenya (Banerjee, Faye, Krueger, Niehaus, Suri and others) is also testing the effects of transfers.P1 independent evaluation (GiveWell) / peer-reviewed RCT / GiveWell
− effects (confirmed)
- 2022In 2022, in South Kivu, DRC, GiveDirectly staff colluded with an outside mobile-money vendor and a former employee, abusing recipients’ SIM registrations to divert cash transfers meant for about 1,900 households (about US$1.2 million by the final investigation). The affected households (people in extreme poverty) did not receive their scheduled payments, and some incurred debt or other losses. GiveDirectly detected the fraud in January 2023, suspended payments, apologized publicly, underwent external audits (Grant Thornton / Humentum), and responded by resuming the original payments. It equaled about 0.8% of that year’s global disbursements.Confidence: high (org self-disclosure + independent reporting) / GiveDirectly / The New Humanitarian
- Tracking of long-term effects
- Final results of the UBI experiment
- Local prices / general-equilibrium effects
- Accuracy of targeting
- Cash vs. programs
- Efficiency at scale
A second look
Cash is powerful but not a cure-all: it does not by itself change structural factors (missing markets, infrastructure, and public services), some effects fade over time, and the impact of transfers on local prices is debated. The question of what to benchmark cash against (cash benchmarking) also continues. It is a one-off transfer rather than the building of productive capacity, and it depends on targeting and on mobile-money penetration. The 2022 fraud in the DRC (staff-enabled SIM fraud that diverted payments meant for about 1,900 households) showed that even in a cash model, local operational controls are essential—GiveDirectly responded with self-disclosure, external audits, and resumption of the original payments.
Sources
How to read this assessment
- Reachable upper bound (ceiling): a confirmed − sets the ceiling, and independently verified + decide the position within it. + do not cancel out −.
- The weight of evidence is not symmetric: only confirmed − are counted; the volume of disputes or allegations goes under “Watching.” + are counted from independent evidence, while an organization’s own PR is treated as “reference.”
- Size is not value: scale is not used in the assessment. Matters that stay within money or competition—investors, shareholders, sanctions, trade secrets—are also excluded.
- The letter (assessment) and certainty (how reliable the information is) are separate axes.
This is a translation; the Japanese version is authoritative. The assessments here are generated automatically by AI based on published criteria. The operator does not alter individual results. Because they are AI-generated they may contain errors, and they are opinion and commentary, not statements of fact. Where evidence is insufficient, the entry is marked “On hold.” Requests for correction are accepted via the form.
Terms: Narrative Value = an assessment (A–G) of the distance between the story an organization tells and its reality / Ceiling meter = a visualization of the reachable upper bound / Watching = unconfirmed matters not counted / Protected stakeholders = people, animals, nature, and future generations. | Generated by: AI | As of: 2026-Q2 | Back to top