RCTs stand for Randomized Controlled Trials and in the international development world, they are a golden ticket to legitimacy. Mention an aid program and sure you are likely to get some support for investing in a good cause. BUT, mention that your program has been tested in RCTs and results show a positive impact — Boom. Your cause is automatically considered a “legitimate” one.
RCTs are used to evaluate the impact of an intervention against a control group, or a group that remains unchanged. This method is common in medical experiments to reduce bias and has become a popular method in development economics since the mid 90’s.
The idea is that, after finding eligible people to participate, you randomly assign people to one of two (or more) groups. The only difference between the groups is the treatment received. There will always be one control group that receives no treatment.
For a simple, hypothetical example of a project that wants to give sheep to poor families, an RCT might look like this:
Eligible families are picked and randomly assigned to group one or two
Group one receives two sheep per family
Group two, the control group, receives nothing
Researchers collect data for a given amount of time to assess how differently the sheep have impacted those that received them than families that received nothing.
Evidence generally plays a small role in the international charitable sector so the mainstreaming of RCTs is promising. RCTs have become synonymous with effectiveness; however, many economists claim that using control groups to compare development interventions is setting the bar too low. Instead, they argue for measuring by using cash benchmarking.
In the cash benchmarking method, the effectiveness of a project is compared to a group that receives cash equivalent to the cost of the project as well as a control group.
If a sheep in the example above costs $10 (considering all overhead expenses as well) then you could benchmark the project by introducing a third group of people who receive $20 per family instead of two sheep.
Paul Niehaus, associate professor of economics at UC San Diego, mentions a common sentiment expressed by proponents of cash benchmarking: if we spend ten million dollars on a project, we should make sure that the project is more effective in doing good than handing out ten million dollars in cash to people.
Non-profits have already started using cash benchmarking. The United States Agency for International Development (USAID) recently made news for using cash benchmarking to measure the effectiveness of some of their programs. The results have come back for the first study suggesting that their Gikuriro program was not more effective than just handing families cash.
Other projects are more effective than cash which is why it should be a benchmark and not treated as a magic bullet. In a study comparing unconditional cash transfers to Building Resources Across Communities (BRAC's) Transfers to the Ultra-Poor (TUP) graduation program in South Sudan, consumption increased from both interventions in the short-run but there was a persistent wealth effect in the long-run for those receiving the graduation program.
It's time we start thinking of giving as critically as we think about investing. It's easy to think about donating as just handing over money that you would never see again. This approach might lead to not really caring about where it goes or where it will have the best impact.
On the contrary, when we invest, we carefully research where our money will have the most return, not just if it will have a return. We should approach donating just as we do investing, with meticulous research.
Kinder is constantly evaluating organisations to recommend those that are performing at the top of their game. You can explore stay on top of these insights by following us on Facebook.