Shikilia does not target specific genders because our targeting mechanism has largely been based the level of need. From our operations so far however, 2/3 of recipients have been women and 1/3 men. Additionally, 96% of sampled surveys have been individuals with more than one child and 60% were single, including widows or separated individuals.

Although it is hard to track how recipients use the money that we send them, research has shown that most money received from cash transfers is actually used to purchase food and household items. Additionally, as most families that we will target will not have access to adequate funds to purchase food and pay bills, we anticipate that their priorities will be to cover necessities before spending any money on vices.

Yes, Shikilia will post updates to our websites where all our funders / supporters can view program progress, results and updates.

The two main corruption risks that typically arise in transfer programs involve (a) manipulation of the list of eligible recipients and (b) diversion of transfers sent to eligible recipients. We address the first through due diligence on list providers, surveying individuals and checking data against any available secondary data sources, and post-transfer audit steps (currently being conducted remotely). We address the second through identity-matching between our records and those of our payment providers, as well as follow-up surveys after enrollment to ensure money is reaching the intended recipients.

In order to be as fair to all as possible, we are selecting recipients using a systematic and proactive set of criteria, and do not accept inbound applications. Unfortunately, we expect that we will not be able to provide cash transfer to everyone in need.

All transfers target individual adults and not households, given the risks of door-to-door enrollment during COVID, the need to act quickly, and our desire to avoid any incentive for misrepresentation on household size. We have sized the transfers per adult (given that the cash is generally supporting individuals now out of work), with the aim that as we scale, we would ideally support all vulnerable adults in all areas, including multiple in each household, comparable to a basic income program.

Individual organizations collaborating within Shikilia are responsible for data protection. Currently, all recipient data is currently being managed by GiveDirectly, which captures and stores all recipient data on leading cloud-based platforms (including Salesforce and Google Suite), and as an organizational policy forbids storage of recipient data on local or personal devices.  Data hosted by GD will only be accessible to a limited number of approved GD staff and third-parties who require the information to monitor the program, determine eligibility and process payments. GiveDirectly also leverages the latest security methods, including password-management software and 2FA for organizational systems, to monitor and prevent any data breaches.

All individuals are surveyed prior to enrollment or transfers, with explicit consent required (via SMS) for inclusion. Where CBOs or businesses do not have permission to share list data or it is not clear, consent is being sought prior to any outreach by GiveDirectly.

We first identify a location of interest then find trusted CBOs in that region that have close community ties to work with us. To pick CBOs to work with, we look at:

1. Risk (robustness of list from an anti-fraud perspective)
2. Reputation (political / community reputation in terms of trust for the CBO)
3. Need of audience (whether the recipients are more / less needy than others we can reach)
4. Reliability (how/where it was gathered)
5. Quality (data completeness, including poverty related data, and recency)
6. Potential for scale (orgs that can help us reach >1500 people)

We are already working with 4 CBOS in different regions: SHOFCO, Access Afya, Ghetto Foundation and Carolina for Kibera that have given us access to 98, 000 individuals from different communities. We are also currently evaluating ~40 CBOs.

Recipient targeting involves the following steps:

• Geographic poverty analysis to identify high need areas (currently focused on urban slums), with government engagement prior to launching enrollment in an area
• Sourcing of lists of potential enrollees (vulnerable individuals) through CBOs and businesses operating in the priority areas, with CBOs and lists screened and prioritized based on factors like organization risk, reputation and capacity, data quality, and level of vulnerability, and MOUs signed. This process is being conducted by GiveDirectly, with oversight on the process from other Shikilia collaborators.
• Within large lists, prioritization based on vulnerability factors.

Currently, more than 100,000 potential individuals have been identified, and list / CBO screening in priority areas is ongoing. Current enrollment areas include Kibera, Mathare, Korogocho, Mukuru and Kawangware in Nairobi, with priority areas in Mombasa and Nairobi being actively considered. Current list providers include SHOFCO and Ghetto Foundation (prior GiveDirectly cash program partners), Access Afya (a low cost health clinic operator), and a growing list of other CBOs.

In addition to this list-based approach, Shikilia organizations are exploring the potential for geographic targeting using telco geodata, which Dalberg and GiveDirectly have previously partnered to pilot and are now working to scale for emergency response outside Kenya. Shikilia is also considering needs in rural areas and how best to support them.