Executive summary: The author argues that “cash benchmarking”—comparing any aid program to simply giving recipients equivalent cash—should be the default standard for doing good, because many well-designed programs underperform cash transfers unless they can clearly justify why they outperform them.
Key points:
A program can outperform a no-treatment control in a randomized controlled trial yet still be worse than giving recipients the equivalent amount of cash, which is what “cash benchmarking” tests.
Evidence from programs like Huguka Dukore shows that while such interventions can improve outcomes relative to controls, equivalent cash transfers can produce larger gains in hours worked, assets, savings, income, consumption, wealth, and subjective well-being.
Programs often underperform cash because recipients have better information about their own needs and stronger incentives to allocate resources efficiently than external donors do.
Some interventions plausibly outperform cash when they address systematic biases (such as undervaluing preventative healthcare), generate positive externalities, secure bulk discounts, fund public goods, enable political change, or benefit beings who cannot transact in markets.
The author proposes that the default for altruistic spending should be to give money to very poor people, and any deviation from this should come with a clear explanation of why it is better than cash.
Some effective altruist funders operationalize this logic by setting explicit benchmarks relative to cash, such as GiveWell’s benchmark based on doubling consumption at the $2.15 poverty line and Coefficient Giving’s “Coefficient Giving dollar.”
This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, andcontact us if you have feedback.
Executive summary: The author argues that “cash benchmarking”—comparing any aid program to simply giving recipients equivalent cash—should be the default standard for doing good, because many well-designed programs underperform cash transfers unless they can clearly justify why they outperform them.
Key points:
A program can outperform a no-treatment control in a randomized controlled trial yet still be worse than giving recipients the equivalent amount of cash, which is what “cash benchmarking” tests.
Evidence from programs like Huguka Dukore shows that while such interventions can improve outcomes relative to controls, equivalent cash transfers can produce larger gains in hours worked, assets, savings, income, consumption, wealth, and subjective well-being.
Programs often underperform cash because recipients have better information about their own needs and stronger incentives to allocate resources efficiently than external donors do.
Some interventions plausibly outperform cash when they address systematic biases (such as undervaluing preventative healthcare), generate positive externalities, secure bulk discounts, fund public goods, enable political change, or benefit beings who cannot transact in markets.
The author proposes that the default for altruistic spending should be to give money to very poor people, and any deviation from this should come with a clear explanation of why it is better than cash.
Some effective altruist funders operationalize this logic by setting explicit benchmarks relative to cash, such as GiveWell’s benchmark based on doubling consumption at the $2.15 poverty line and Coefficient Giving’s “Coefficient Giving dollar.”
This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.