@IPRatNU posted long-term RCT results for the Graduation program in Ethiopia, providing very poor households a productive asset (e.g. livestock) & training in its use. Quick take: High-quality RCT finds a sizable 18% gain in household consumption at the 3-year mark, fading somewhat to 9% at 7 years.


  • Per the study report, program participants received a one-off transfer of either sheep & goats, oxen, bees, or inventory for petty trade, plus technical training and coaching, and were given access to local bank accounts. The program's cost was approximately $4011/household in US$.

Study Design:

  • The study sample comprised 925 households, randomly assigned via lottery to treatment (T) vs control (C). Based on careful review, this was a high-quality RCT (e.g., good baseline balance, negligible sample attrition).


  • 18% gain in per-capita consumption in year 3 after study entry, fading somewhat to a 9% gain in year 7 (both impacts were statistically significant). The impact on food security ("enough food every day?") was 72% T vs 64% C at 3 years, fading to 85% T vs 83% C (& no longer statistically significant) at 7 years.

  • These impacts are meaningful (as evidenced by the 3y gain in food security), but the T & C groups remained very poor - e.g., at the 3y mark, annual per-capita consumption was just $666 T vs $563 C in US dollars. It's unknown whether total program benefits will exceed costs over time (it depends on how long the impacts endure).


  • Various RCTs of the Graduation program have found large impacts in some places (e.g., West Bengal, India; Afghanistan), modest in others (e.g., Ethiopia, described above), & null in Andhra Pradesh, India.

  • Overall, I think Graduation is a very promising program model, but more research is needed to identify the conditions/settings where it's most effective. Here's the Ethiopia study report.

We use cookies to improve your experience and to help us understand how you use our site. Please refer to our cookie notice and privacy statement for more information regarding cookies and other third-party tracking that may be enabled.

Intuit Mailchimp logo