Back in 2010, after the terrible earthquake that hit Port-au-Prince killed 200,000 people, Haiti was for a time a substantial recipient of foriegn aid. But it's now seven years later. Other disasters have happened. Canada has at times been the single most important foreign aid donor to Haiti, and Haiti was the top recipient of Canadian foreign aid in 2010. But Haiti now by ranks 16th among countries receiving aid from Canada. Thus, the Canadian government decided to fund a study to set priorities for its foreign aid. It provided funding to the Copenhagen Consensus group for the Haiti Priorise project, which carried out benefit-cost studies on a number of possible aid priorities.
It's worth noting that the specific numbers in benefit-cost calculations are always a little shaky: if you are comparing two policies with benefit-cost ratios of, say, 4.2 and 3.5, it's probably fair to conclude that for practical purposes they are about the same. But a lack of perfect precision doesn't mean a lack of any insight. If the benefit-cost calculations find that fortifying wheat flour with micronutrients has a benefit-cost ratio of 24, while building pit latrines in urban areas has a benefit-cost ratio of 0.9, then it's sensible to decide (at least provisionally, barring some dramatic change in the available information) that one option is strongly preferable to the other.
If you want to look over the research behind Haiti Priorise recommendations, this page offer the links to each of the presentations of evidence; in turn, each page has a link to an underlying background paper.
Here's a list of the policy options where the benefit-to-cost ratio is 11 or higher, with the benefit-to-cost ratio listed first and the policy option afterward. Thus, wheat flour micronutrient fortification has a benefit-to-cost ratio of 24. If you want more information, the presentation is here and the underlying paper, written by four professors at UC-Davis, is here.
- 24 Wheat flour micronutrient fortification
- 22 Reform electricity utility
- 18 Access to contraception
- 17 Early childhood stimulation
- 16 Train first responders
- 14 Shelters for women and children
- 13 Immunization for children 0-1
- 12 Expand mobile broadband
- 12 Domestic violence hotline
- 11 Electronic registration of birth certificate
- 11 Maternal and newborn health
On the other end of the scale, here are the policy options with a benefit-to-cost ratio of 1 or less, again with the benefit-to-cost ratio first, and the policy option next.
- 1.0 Solar photovoltaic power
- 1.0 Increase domestic worker wage
- 1.0 Increase public service pay 10%
- 1.0 Urban container-based sanitation
- 0.9 Pit latrines in urban areas
- 0.8 20% rice tariff for 10 years
- 0.8 Improved and intensified rice production
- 0.8 Prevent teen dating violence
- 0.8 Vaccinate girls against cervical cancer
- 0.5 Solar reflective power
The value of this approach comes in several forms. In some cases, results are non-intuitive (at least for me). For example, I probably wouldn't have guessed both that a domestic violence hotline and shelters for women and children would be near the top of the list, while also guessing that seeking to prevent teen dating violence would be near the bottom of the list. I'm not surprised that vaccinations for newborns and maternal-child health is high on the list, but I am surprised that the gains from vaccinating girls against cervical cancer ranks so low.
The more broad-based advantage is the potential for reallocating resources. Shifting aid funds from programs with the lowest benefit-to-cost ratios to those with the highest benefit-to-cost ratios will not change expenditures, but would increases benefits tenfold and more.
Finally, if you disagree with a benefit-cost analysis and wish to dispute it, the reasons for the number out in the open. Have at it.