Promoting Livelihoods and Productive Inclusion: Direct Impacts and Local Spill-overs in Chad

The Chad cash transfer program provides small monthly transfers to 6,200 poor households. Bene ficiary households receive a transfer of CFA 45,000 every three months for a period of two years. Payments are usually made to the wife of the household head. The fi rst payments started in December 2017.

To overcome various constraints preventing productive inclusion and resilience of poor Chadians, a subset of cash transfers bene ficiaries was assigned a package of productive accompanying measures, starting in the second year of the cash transfer program (December 2018). The package includes a lump-sum cash grant as part of a set of interventions that were developed to address a range of constraints to productivity identifi ed in the Sahel region. The package is a lighter version of the "capital package" tested in government-led safety net programs in other Sahel countries (Bossuroy et al. 2018, https://www.socialscienceregistry.org/trials/2544) and is similar to traditional graduation programs.

The productive inclusion intervention includes the following components:
- Component 1: Facilitation of savings groups for cash transfer bene ficiaries. This component facilitates the creation of community savings groups and provides technical support to their management, using the village savings and loan association model.
- Component 2: Micro-entrepreneurship training. A week-long group-based training covers basic business skills. The training focuses on
cross-cutting micro-entrepreneurship skills, including basic accounting and management principles, market research, planning and scheduling, saving, and investing. In addition, the training helps beneficiaries chose a livelihood activity.
- Component 3: A lump-sum cash-grant (around USD 200) aims to promote investments in income generating activities. The cash grant is unconditional, and provided after components 2 and 3.
- Component 4: Coaching: A coach supports bene ficiaries throughout the program, including to elaborate a simple plan for the use of the cash grant, as well as to provide facilitation of market access.

2018-12-01

2020-06-30

- Consumption: per capita measures of food and non-food consumption
- Food security: standardized measures of food security (FIES, and FCS)

- Income-generating activities and Diversifi cation: new business, expansion or diversifi cation of existing business
- Income and Revenues: income from business, wage employment; livestock and agricultural revenues; number of types of income sources.
- Assets: value of household durable, productive assets, and livestock holding.
- Finance: borrowing, lending, and savings
- Proxies for psychological and social well-being
- Women's empowerment: decision-making power over her own earnings and income-generating activities, control over other resources in the household.

Key variables will be constructed consistently with Bossuroy et al., 2018. (https://www.socialscienceregistry.org/trials/2544)

To estimate the impact of the productive inclusion measures, we use a multi-arm clustered randomized design. The study is implemented in villages targeted for the pilot cash transfer program in the regions of Bahr-el-Ghazal and Logone Occidental. In these two regions, 116 villages participate in the cash transfer program. We do not include villages with less than 10 cash transfer bene ficiaries in the sample. In addition, in villages with fewer than 20 CT beneficiaries, all CT beneficiaries are allocated to receive the productive measures intervention. The remaining 88 eligible villages were be randomly assigned to treatment (CT+ productive package, 57 villages) and control (CT only, 31 villages). Further, within treatment villages, a subset of CT beneficiary households (1798 out of 4005) was randomly selected to receive the productive inclusion measures.
This design generates the following categories of households:
1. Treatment villages:
a) CT + Productive Package (1,798)
b) CT (2,207)
c) Non-CT (10,554)
2. Control villages:
a) CT (1,861)
b) Non-CT (4,666)

Randomizing villages and households allows to measure the impact of the productive measures as well as spillovers by taking single differences in post intervention outcomes between the relevant treatment, control and spillover groups.

Speci fically, the impact of the productive inclusion package is estimated by comparing group 1a and 2a. The local spill-overs of the productive measures are estimated by comparing groups 1b and 1c to 2a and 2b. We will measure spill-over of cash transfers beneficiaries and non-beneficiaries to maximize power. We will also estimate separately spillovers on CT bene ficiaries (by comparing 1b and 2a) and spillovers on non-CT bene ficiaries (by comparing 1c and 2b) (exploratory analysis).

The follow-up survey will be collected one year after completion of the intervention.

in office by a computer

village, then household (see description above).

Yes

88 villages

4000 households

(Planned sample break-down, subject to adjustments)
1. Treatment Village:
1a CT + Productive: 1000 (average cluster size: 18)
1b CT Only: 680 (average cluster size: 12)
1c Non-CT: 680 (average cluster size: 12)
Total in treatment villages: 2360 (average cluster size: 42)
2 Control Villages:
2a. CT: 1000 (average cluster size: 32)
2b. Non-CT: 640 (average cluster size: 21)
Total in control villages: 1640 (average cluster size: 53)

see design documents for power by indicator and comparison