Regarding Baseline consumption, GiveWell addressed it here (in their report of GD, the section discussing selection of recipients).
Endline results from this study in July 2018 found that mean consumption per capita per day in the full population of the control villages was $0.79.43 We would guess that this population is fairly similar to current beneficiaries of GiveDirectly’s program in Kenya. [...]
Given our estimate that households in places where GiveDirectly previously used the targeting model had a mean per capita consumption level of approximately $0.78 per day,45 the results from the general equilibrium study suggest that the income levels of recipients under the targeting and saturation models are very similar. Prior to seeing the new data, we expected recipients under the village saturation model to be somewhat wealthier, on average, than the average recipient under the household targeting model. This result may be an indication that the targeting model was not particularly effective at identifying the poorest households in a village and/or that GiveDirectly is now working in areas that are poorer overall than where it worked previously.
Now that I’ve realised this, I will remove the entire baseline consumption consideration. As projecting forward I assume GiveDirectly will just get better at selecting poor households to counteract the fact that they should be richer. Thanks for pointing this out!
A good time to mention that I thought this is really great work, and nice spotting of gaps in the original analysis (even if specifically this seem to have been addressed, it wasn’t at all clear from the CEA spreadsheet) :)
Regarding Baseline consumption, GiveWell addressed it here (in their report of GD, the section discussing selection of recipients).
Oh no, I’ve missed this consideration! I’ll definitely fix this as soon as possible.
Now that I’ve realised this, I will remove the entire baseline consumption consideration. As projecting forward I assume GiveDirectly will just get better at selecting poor households to counteract the fact that they should be richer. Thanks for pointing this out!
👍
A good time to mention that I thought this is really great work, and nice spotting of gaps in the original analysis (even if specifically this seem to have been addressed, it wasn’t at all clear from the CEA spreadsheet) :)