As part of researching the most effective charity options, Charity Entrepreneurship (CE) has previously conducted research on Conditional Cash Transfer (CCT) programs (see here). Since then a few things have changed that we feel are important enough to merit an update (available here).
In brief, this new evidence base points towards a large effect of CCTs on increasing clinic visits and institutional deliveries, and also suggests a larger effect of CCT programs targeted at healthcare workers, rather than healthcare patients, for certain outcomes such as antenatal care visits, clinic visits, and institutional deliveries. Why it was worth updating the evidence base
More Evidence There are a large number of studies (70+) that were not included in the previous write-up and which provide a much larger sample for estimating average treatment effects. Crowdedness An initial reason for CE’s interest in CCTs was that the field is relatively uncrowded. The largest non-profit we know of that implements CCTs is New Incentives, which currently operates a program that offers CCTs to incentivize infant vaccination in Nigeria, after moving away from encouraging the prevention of mother-to-child transmission (PMTCT) of HIV by incentivizing birth in clinics, antiretroviral adherence, and HIV screenings for newborns. After this additional round of research we did not find any new large non-profits working on CCTs that we were not already aware of, although, we have been able to gather much more evidence from the more than a dozen government-run programs. Wider Array of Conditions Given the identification of these additional studies and the changes in some organisations’ (e.g. New Incentives) target outcomes, we believed there was likely to be evidence on a larger set of conditions than we considered before. Goal We are looking for more specific options in the field of conditional cash transfers. Our initial report did not make a strong specific recommendation, instead listing a few possible options. Improving the evidence base is a necessary step in the process. We plan on publishing a fully updated CCT report with specific details on updated options by the summer of 2019. What process was used to find the studies
Consult previous CE research notes.
Read what GiveWell has written, including updates on New Incentives.
Search for reviews from Cochrane reviews, Campbell Collaboration etc.
How to read the spreadsheet The evidence is grouped into “Directly applicable evidence”, “Related studies”, and “Organizations currently doing C/UCT”, in addition to some rough Average Effect estimates. Author, Year, Study Effect: Includes the percentage point increase or decrease in the target outcome mentioned in the study. Where possible confidence intervals or p-values were included. Sample size (Control + Experiment): The number of individuals with the targeted outcome (including both those in treatment and control groups). In some instances individual-level numbers were unavailable and rougher units such as number of households are provided. Target Outcome: These include Antenatal Care (ANC), Immunizations, Institutional delivery, etc. Location & Region: The country where the program being studied was conducted. The region (Asia, Africa, Latin America) was also included for comparative analysis. Experiment Type: Lists whether the study was an RCT, observational data study, longitudinal or cross-sectional or quasi-experimental RDD or DinD. Intervention: Whether the program was a CCT for Education, Healthcare Utilization, or Healthcare Worker Performance (which could be targeted at education or healthcare utilization target outcomes). Size of Cash Transfer (USD): Many of the studies provide the dollar per month equivalent of the cash transferred to recipients. Where the program distributed cash at some other time interval the monthly size has been calculated. Where no US dollar amount was listed the USD equivalent was used when it was easy to do so. Link & Alternative Link: A URL to the original source in online pdf or through a journal access website. Notes: Any other points of interest.