Benchmarking a livelihood support program for refugees in Nairobi

How to cite

Zeitlin, Andrew; Khan, Sana. 2020. "Benchmarking a livelihood support program for refugees in Nairobi." AEA RCT Registry. AEARCTR-0005621. https://doi.org/10.1257/rct.5621-2.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Andrew Zeitlin* , Sana Khan

* Primary investigator

Registry ID
AEARCTR-0005621
DOI
https://doi.org/10.1257/rct.5621-2.0
First registered
Last updated
Status
On going
Trial dates
2019-08-01 – 2021-12-31
Countries
Kenya ()
Registry record
http://www.socialscienceregistry.org/trials/5621

Abstract

The increasing movement of refugees crossing geographic and political boundaries to escape conflict and to search for economic opportunity presents a challenge not only to wealthy countries, but also to developing-country hosts. With limited opportunities for wage employment in firms, the best opportunity for economic integration often lies in microenterprise. But little is known about the best way to support the growth of small businesses among such vulnerable populations: as Jayachandran (2020) notes, the empirical literature to date is ambiguous about the returns to relaxing human capital and liquidity constraints for entrepreneurs in general. This academic and policy uncertainty is even more pervasive in relation to at-risk populations such as refugees, with a high population of female entrepreneurs among them.

To address these questions, this study experimentally evaluates the impact of an International Rescue Committee (IRC) BILLY program designed to increase employment likelihood and income of young urban refugees and vulnerable populations in Nairobi. We pursue an individually randomized controlled trial to understand how a package of bundled services compares to an experimental control group and two different treatment arms that receive a combination of services and cash grants of equivalent cost to the IRC. Moreover, the study incorporates an incentive-compatible 'choice' arm that allows us to assess the extent to which refugees have access to information about what mix of training and cash will best allow their business to succeed.

This project will evaluate the impacts of these programs on the employment status, time use, productive assets, earnings and consumption of the target population. In addition, we will examine impacts on a range of mechanisms of impact and alternative measures of well-being, ranging from mental health to perceived self-efficacy and cognitive bandwidth. By evaluating these impacts in parallel for populations of refugees and comparably situated native residents, our study will speak to the generalizability of the existing microentrepreneurship literature onto the crucial problem of putting refugees skills to productive employment.

Intervention

The study uses an individually randomized, controlled trial to evaluate the impacts of an IRC program that provides microentrepreneurship training to a vulnerable population consisting primarily of refugees in Nairobi, Kenya. Specifically, we evaluate the microentrepreneurship track offered in the IRC's Building Income and Leveraging Livelihoods for Youth (BILLY) program. This program will be 'benchmarked' against the impact of a cash transfer equivalent in costs, and a third active treatment arm unbundles the training to examine responses to a mixture of cash and training -- again, holding total costs constant. Finally, to understand how the target population's own information may be harnessed to learn not only about their willingness to pay for specific alternatives, but also about their idiosyncratic returns to human capital versus liquidity, we implement an incentivized 'choice' exercise.

Experimental design

We enrolled individuals who meet IRC's eligibility criteria and express interest in starting or growing a business on a rolling basis, with a new cohort every two weeks spread across two phases.
In the first phase (Fall 2019), one cohort consists of three treatment groups:
1. Full microenterprise program, which includes business skills training (called L2E), business grants, required participation in urban savings and loans association, and coaching and mentoring.
2. Business grant only, including the business grant and an additional cash amount corresponding to the cash value of services for clients.
3. Business grant and business skills training (L2E), including an additional cash amount for the other services of the program.
The second phase of allocation began in February 2020. During this phase, the beneficiaries are assigned to either one of the treatment groups described, or to a control group that will not receive any service until after endline.

The incentivized 'choice' exercise is conducted with participants who were randomly assigned to group 2 and group 3. Group 2 participants are asked their willingness to pay to receive the 4-day program of business skills training (L2E), as well as their willingness to pay to receive the full suite of services offered by IRC. Group 3 participants already receive L2E, so they are only asked about their willingness to pay for the full suite of services. Using a public lottery system, some participants may purchase the services with the additional cash received and effectively switch treatment arms.

Design details

Randomization method
The allocation of individuals to each experimental group is undertaken on a randomized basis for each cohort in an office by a computer and blocking for female and refugee status.
Randomization unit
The unit of randomization is the individual refugee/applicant.
Clusters
Study participants are clients registered for the microenterprise track of the BILLY program between August 2019 and April 2020. The assignment protocol is conducted on a rolling basis, with a new cohort every two weeks spread across two phases. In the first phase (Fall 2019), a total of 535 individuals were assigned to one of the treatment arms: 171 to full microenterprise program, 185 to business grant only, and 179 to business grant and business skills training. For phase 2 (Spring 2020), we will assign approximately 765 individuals to complete a sample size of 1300 people, including 1000 receiving active treatment.
Observations
735 individuals.
Arms
Full microenterprise program: 504 Business grant only: 518 Business grant and business skills training: 512 Control: 300
Primary outcomes
Primary outcomes examined in this study relate to the development of microenterprises under the BILLY program and its variants. These include employment status and hours worked, business assets and profits, and household consumption. These outcomes will be measured at endline, two years after the enrollment of the first clients, and high-frequency surveys will provide descriptive evidence on the evolving time path of these outcomes. In addition, we examine impacts on a range of mechanisms and alternative measures of well-being, including measures of participants' mental health, sense of empowerment, and satisfaction, and measures of their cognitive performance designed to understand how programs that alleviate (financial) bandwidth constraints may improve decision-making.
Secondary outcomes
Please see above and pre-analysis plan. Secondary outcomes will be finalized upon completion of baseline data collection.

Cited by

No citation data available for this registration yet.

Keywords