Development aid assumed poor people couldn't be trusted with cash — a charity tested that assumption directly and found it was the aid model itself wasting money

the problem

An entire industry's design assumes the people it serves need decisions made for them, and that assumption is rarely tested against the alternative

background

what everyone would do

The standard aid model was in-kind or curated programs — livestock, seeds, job training, school supplies — selected on behalf of recipients by an NGO or agency, built on the assumption that unrestricted cash would be misused on drink or short-term wants rather than what an outside expert judged people actually needed. That assumption shaped nearly the entire industry's program design without ever being tested directly against its alternative.

what they saw

GiveDirectly's founders saw that the paternalistic assumption underneath the whole aid industry had simply never been tested with the same rigor applied to conventional programs — nobody had run unconditional cash transfers as a randomized comparison to see whether recipients, given full discretion, actually did worse. Treating 'just give people the money' as a measurable hypothesis rather than an ideological position turned a philosophical debate into an experiment.

the move

GiveDirectly sends one-time or recurring unconditional cash transfers directly to recipients' mobile-money accounts, with no restrictions on how the money is spent and no accompanying training or program — recipients decide entirely for themselves, and GiveDirectly measures the outcomes with the same experimental rigor development economists use to evaluate conventional aid programs.

why it works

A recipient who knows their own local prices, urgent needs, and immediate opportunities can direct cash exactly where it matters most, while a distant program design fixed in advance can only guess — funding livestock when the real bottleneck is medical debt, or seeds when the real bottleneck is transport. Removing the curation and logistics layer that an in-kind program requires (procurement, distribution, administration of a physical good) also cuts overhead, so a larger share of each donated dollar reaches the recipient rather than the delivery apparatus around it. Mobile-money transfer makes the transaction itself auditable without requiring control over how it's spent, satisfying donors' need for accountability while still leaving the allocation decision with the person who has the most relevant local information — and the resulting randomized studies then produced hard, comparable evidence (multiplier effects, mortality reductions) that no curated program had ever been measured against in the same way.

the payoff

A large randomized study across 653 villages and over 10,500 households in rural Kenya (transfers averaging roughly $1,000, an injection worth more than 15% of local GDP) found a local economic multiplier of about 2.5–2.7 — every dollar transferred generated $2.50–2.70 in local economic activity, with large positive spillovers even to non-recipient households and minimal price inflation. A separate large-scale follow-up study found unconditional cash transfers were associated with 48% fewer infant deaths before age one and 45% fewer child deaths before age five in recipient communities.

where it breaks

The mechanism depends on functioning local markets where cash can actually buy what's needed — food, medicine, tools, services — and on a payment rail that reaches recipients reliably; it does less where a critical good or service simply isn't available for purchase locally, such as a vaccine that must be administered by a health worker rather than bought. It also assumes the named recipient genuinely controls the money, which breaks down under household power imbalances where a transfer gets seized or redirected by someone else in the home, and it cannot substitute for public goods like roads or water infrastructure that no individual household's cash can produce, only their private share of. And GiveDirectly's own 2023 finding of roughly $900,000 in employee fraud in its DRC program is a reminder that removing paternalistic control over recipients doesn't remove the separate need for financial oversight inside the organization delivering the cash.

what came after

GiveDirectly held GiveWell's 'top-rated' charity status from 2012 to 2020 and its research helped establish unconditional cash transfers as a credible, rigorously evidenced category within development economics rather than a fringe idea — though the organization has not been without its own operational failures, including a 2023 investigation that uncovered roughly $900,000 in employee fraud in its Democratic Republic of Congo program, a reminder that removing paternalistic program design doesn't eliminate the need for organizational oversight elsewhere.

references

1. [1]General Equilibrium Effects of Cash Transfers: Experimental Evidence from KenyaNational Bureau of Economic Research, 2019nber.org
2. [2]Cash TransfersGiveWell (independent charity evaluator), 2024givewell.org
3. [3]Can Cash Transfers Save Lives? Evidence from a Large-Scale Experiment in KenyaNational Bureau of Economic Research, 2025nber.org

keep it

same kind of clever

• 2010 · American ExpressAmex didn't advertise its card on the biggest shopping weekend of the year — it invented a whole new day inside it
• 1962 · AvisAvis told customers outright it was only #2, and the confession made everything else it said believable
• 2025 · AXA France (with Publicis Conseil)AXA covered domestic violence like a house fire by changing three words in a contract millions of people already had