Apply for funding

Ideas42 & World Bank — Impact Brief

Introduction

ideas42 are a non-profit organization using behavioural science to address complex social problems. One very promising of focus of their work is the design and implementation of innovative solutions rooted in behavioural science to increase the effectiveness and efficiency of cash transfer programs targeting the poorest households.

Over the past twenty years, and further accelerated by the pandemic, the number and coverage of social protection programmes with a cash transfer component in developing countries has rapidly expanded. These programmes provide safety nets for millions of people and uplift them out of poverty. ideas42 use a problem driven process to understand drivers of behaviour of cash transfers recipients and draw from the behaviour science literature to design cost-effective nudges, or minor adjustments to such social protection programmes, which may enable people to make better use of the transfers they receive. If these nudges are effective, they could be easily scaled and adapted for programmes across Sub-Saharan Africa.

Use of GIF Funds

In 2017, GIF awarded a $2.22 million grant to ideas42 to partner with the World Bank’s Social Protection and Jobs Global Practice and the social protection agencies of Kenya, Tanzania, Ghana, and other countries to develop and test behavioural enhancements to their cash transfer programmes. A follow-on grant of $2.4 million was awarded in 2020.

Objectives

Enable ideas42 to partner with social protection agencies within governments in sub-Saharan Africa and the World Bank’s Social Protection and Jobs Global Practice to deploy a problem-driven approach to implement behavioural design interventions in social protection cash transfer programmes in at least five countries. Rigorous evaluation will test the impact of behavioral additions on productive inclusion and human development outcomes, and ideas42 will support the capacity of the World Bank and local governments to apply these methods and scale proven approaches.

  1. Implement the innovation in ongoing social protection cash transfer programmes focused on productive inclusion or human development outcomes in Kenya, Tanzania and five other countries by using a lean design process to customize and field test behavioural interventions.
  2. Generate and disseminate evidence on the innovation across a wide spectrum of programmatic contexts in Kenya, Tanzania and four other cash transfer programmes.
  3. Build capacity of the staff of the World Bank Social Protection and Jobs Global Practice and the national Implementing Agencies, and produce key resources to equip them to initiate and scale the innovation.
  4. Facilitate the adoption of the innovation and its integration in processes within the World Bank’s ongoing and new social protection programmes.

Impact to Date

Ideas42 have refined and finalised behavioural nudge designs for cash transfers through co-creation with the World Bank and government implementation bodies. The designs encompass a self-affirmation activity, social norms poster, partitioning pouch, and goal-setting and plan-making exercise. In Kenya and Tanzania, ‘lab in the field’ evaluations, randomized at the individual level (and therefore not immune from spillover effects) for 900 beneficiaries have been completed, with the control group receiving an informational intervention about ‘the importance of saving’ as an alternative to the nudge. Key results one month after the intervention include:

  • Intention to invest: Intention to invest is measured in terms of having a productive goal. The reported productive goal increased in both Kenya (by 7pp or 9%) and Tanzania (3pp or 3%)
  • Saving: Savings is measured in two ways –one, incidence of savings and two, actual amount of the transfer saved. In Tanzania, the incidence of savings increased (8pp or 13%). In Kenya, the results showed no significant increase in the incidence of savings but a marginal (but high magnitude) significant increase in the amount saved (5pp or 41%).
  • Productive investments: In Tanzania, the results show an increase in incidence of productive investment (5 pp or 8.9%). While results are suggestive of positive effects, there are no other conclusive outcomes.
  • Household vulnerability: In Kenya, the results show a significant increase in complete repayment of debt within a month (14pp or 60-70%).

Overall, based on the one month follow up results of the trial we observe that in both Kenya and Tanzania beneficiaries are setting productive goals and increasing their savings (in anticipation of the productive goal planning). Short term results are encouraging and demonstrate the intermediate behaviour change necessary for long term outcomes were facilitated by the nudges. A cluster randomized control trial is now underway in Tanzania which will look at effects after 6 months.

This builds on evaluation undertaken in Madagascar’s Human Development Cash Transfer Programme (not funded by GIF). This multilevel, multi arm randomized control trial with a sample of more than 6,000 households was conducted 12months after the intervention, with positive impacts observed across some combinations for income, likelihood of repaying loans and food security, as well as expenditure on education and adoption of positive parenting behaviours.

In 2020, ideas42 completed a baseline for a more rigorous cluster randomised trial in Kenya, but subsequent intervention roll-out was disrupted mid-way by COVID-19. ideas42 pivoted to conduct Computer Assisted Telephone Interviews with some significant results for those who received the intervention on goal setting, savings, productive investment and income received from other sources.

ideas42 have closely engaged with partners in Kenya, Ghana, and Tanzania to strengthen their ability to apply behavioural principles in their work, especially in pandemic response. All processes from design, user-testing, to randomised trials were co-designed and co-delivered. ideas42 also conducted workshops for government agencies, for example, as part of the World Bank Financial Literacy Workshop in Kenya.

To facilitate easy application of behaviour science to cash transfer programmes, ideas42 have generated several public goods including a checklist tool for practitioners to design behaviourally informed programs, a policy paper on learning and results to date, a guidance note and infographic on applying behavioural insights to social protection cash transfers in emergency responses, and presentations to document processes. ideas42 further developed key operational tools such as training and facilitation manuals, and testing procedure documentation.

To date, ideas42 has reached 12,170 recipients with behavioural nudges integrated in cash transfer programmes, and, as a result of the work under this grant,13 new social protection programs have expressed interest in adopting behavioral insights.