Since the onset of the COVID-19 pandemic, cash transfer programs across the globe have been trying to respond to the increased demand for assistance from those who need it most. In many countries, cash transfers are a vital component of the livelihoods of those living with extreme poverty. Timely and effective delivery of these programs is critical, and insights from behavioral science can aid cash transfers in having a greater impact on the lives of those in need. 

For the last six years, ideas42 has been working with governments in sub-Saharan Africa to incorporate behavioral designs into cash transfer programs in order to enhance their impact. Though our in-person work in the field came to a halt at the beginning of the global COVID-19 pandemic, we still worked to generate insights about our designs in this new context. For example, in Kenya, we worked with the Kenyan Ministry of Labor and Social Protection to determine if designs created pre-COVID were still effective during the pandemic.

However, the pandemic persisted, and there was a greater push for innovative solutions to provide timely assistance to support those directly impacted by the growing economic crisis. Practitioners and governments agree that the COVID-19 pandemic created a turning point for the delivery of social protection programs. This, in turn, presented us with a challenge: how to adapt interventions designed to be delivered in-person in a context that for all intents and purposes no longer existed, and maintain their effectiveness for cash transfer recipients.

We worked with the World Bank and the Fonds Sociale de la République Démocratique du Congo (FSRDC) to add a behavioral lens to their COVID-19 response cash transfer program. The goal of this program was to assist people experiencing poverty in urban Kinshasa in managing their cash payments, enabling them to spend cash in a way that would build resilience and reflect their own goals. Potential recipients were identified through satellite mapping, after which they registered over text message or through a telecommunications company on the ground. Once identified, they received the equivalent of $25 per month for six months, delivered through mobile money. Given the digital nature of this program, we aimed to assess whether a behavioral intervention, delivered fully through an SMS platform, would have a positive impact on recipients’ spending in accordance with their own goals.    

To help recipients manage their money and spend in line with their priorities, we worked to develop a pair of behaviorally informed text messages to be sent to recipients around the time of their cash transfer payment. These messages sought to help recipients identify a priority goal for spending their cash, and then spend their cash accordingly. The first message, sent the day before the transfer, prompted recipients to identify their priorities and to write out a plan for how they wished to spend their money. A day after receiving the mobile money transfer, recipients received a second message encouraging them to recall their plan and follow through with it. Both messages centered the recipient’s agency and ability to change their household stability through the program. 

To assess if the messages were effective, we rolled out a randomized controlled trial (RCT) in which half of the recipients were randomly selected to receive the messages and the other half did not. We then sent out a survey using interactive voice response (IVR) three and a half weeks after participants received both of our messages and their cash. We aimed to learn whether the messages had an impact on recipients identifying a goal for their cash and if it was a future-oriented goal (such as spending on a business or education, paying off debt, or saving), and whether they created a plan to follow through on this goal. We also wanted to know if they spent their cash on their identified goals. 

Overall, the results showed a positive, and significant, impact for those who received the messages:

• 2.5% increase in recipients reporting having identified a priority.
• 4.1% increase in recipients reporting having written down a plan for spending their cash. 
• 3.6% increase in recipients reporting having spent their cash in line with the identified priority.  
• 6% increase in recipients who reported having identified priorities that were future-oriented. This could be in part due to the framing of the text messages aiming to empower them to use the cash to improve their family’s livelihoods. 

The positive results from this work are similar to what we’ve found previously with in-person interventions, for example, in Kenya and Tanzania where we tested behavioral designs. Though these designs were delivered differently, they utilized the same behavioral principles and tools of goal-setting, plan-making, self-affirmation, and timely reminders. We’re excited to see that the results from this fully text-based intervention are in line with our previous work, as a digital approach could be more cost-effective and scalable than in-person delivery methods. 

COVID-19, and the economic crises that followed, created a turning point in the social protection landscape. Many in the field are seeking new ways to innovate as a result of the challenges posed by the pandemic, and behavioral science can offer insights and context-driven tools to design more effective transfers that can reach large numbers of new transfer recipients. Over the coming months, we plan to work with the FSRDC in the Congo to scale our messages in order to reach thousands more cash transfer recipients and help practitioners and governments alike make the most of the resources they have available.