Rose, a mother in a rural village in Kenya, takes care of five children. Two are her own, and three are the children of her sister, who passed away two years ago. Rose wants to send them all to school, but it has been difficult—last year, one child got sick, and the cost of medical care meant that she could only afford school fees for three of the five children. She hopes to open a small bakery and sell mandazi, a popular fried bread, but start-up costs are high. Given all her other responsibilities, she has struggled to save enough to buy bulk ingredients and rent a location. “It is very difficult to save when there is a sick person who needs medicines,” she said.
Rose, whose name we’ve changed for privacy reasons, is not alone. Millions of households across low- and middle-income countries face similar challenges in covering essential costs, which make it much harder to plan for the future. For instance, in Kinshasa, the capital of the Democratic Republic of the Congo, many people work in small, unregulated firms and struggle to meet their needs. Shocks to the economy, like the recent pandemic, are particularly damaging without the government protection that is more often afforded to workers in regulated firms. In South Sudan, rapid depreciation of the South Sudanese pound makes it impractical for households to save even if they have extra cash—any money they save loses value almost immediately.
What many of these households have in common is that they participate in cash transfer programs. Such programs, through which governments or philanthropic organizations provide recurring payments to households, are an evidence-informed way to reduce poverty and improve the lives and livelihoods of people around the world.
Cash transfers mean people have to allocate a large influx of cash across multiple needs and wants, a challenge in any circumstance. We all have finite mental bandwidth, so we can only process and act on a limited amount of information. People who experience poverty, a state of chronic scarcity, must use much of that mental energy on urgent needs like feeding their families, leaving little to devote toward figuring out and following through on the most financially efficient way to do so.
Budgeting the cash from these transfers isn’t easy—especially under conditions of scarcity, whether brought on by chronic poverty or exacerbated by pandemic-related shocks.
People who receive cash transfers often wish to put their money toward long-term goals, such as saving up to pay for their children’s school fees, opening small businesses, or investing in livestock. However, programs are not generally designed to account for common decision-making tendencies, such as present bias and limited attention, and don’t facilitate the complex budgeting decisions required to balance basic needs with future goals.
My team and I at ideas42, in partnership with the World Bank, have spent the past seven years developing behavioral interventions for cash transfer programs to support participants in making and following through on those decisions. Results from randomized controlled trials (RCTs) in Kenya, Tanzania, South Sudan, and the Democratic Republic of Congo show that these interventions, when customized to suit the socioeconomic context in which they are delivered, can support recipients saving or spending in line with their priorities and making investments that can help them generate income in the future—cost-effectively enhancing the impact of cash transfer programs.
Our first step was to speak with cash transfer recipients to better understand their goals and priorities, and the decisions they must make to use their cash accordingly. We learned that many participants had long-term ambitions that required a large amount of capital—more than they received in any one transfer. One participant in Tanzania told us, “I want to open a shop, but I will never have the capital to do it… I need 200,000 shillings to buy ingredients to start, but I only got 20,000 [from this transfer].”
Through our conversations, we also learned that many recipients thought of each transfer as one-off. The framing and structure of programs didn’t set recipients up to consider what they might achieve by saving cash from multiple transfers. Beyond that, participants had immediate consumption needs, such as buying food for their families. Visiting the market immediately post-transfer can be overwhelming—participants reported finding it challenging to determine on the fly how much cash to put toward those urgent needs versus how much to save in support of their longer-term goals.
Understanding the context was crucial—if we had simply copied and pasted designs that promoted saving, what was helpful in one country likely would have been ineffective or even harmful in another.
In collaboration with local communities, we developed simple, low-cost tools that recipients can use at this critical juncture. Goal-setting and plan-making tools prompted participants to identify an achievable goal and plan exactly how much of each transfer they wished to put toward that goal. Participants were then provided with a physical tool, such as a pouch or envelopes, that they could use to immediately separate their long-term from their short-term cash.
The interventions were delivered as close as possible to the actual transfer to take advantage of the “cognitive plenty” recipients experience when they receive their cash. In this moment, the mental constraints of scarcity are temporarily eased because recipients have the resources they need to meet their basic needs—an optimal time plan for the future.
Take, for example, a household that wants to start a basket-weaving business in Kenya or Tanzania. They might set a goal to purchase materials such as dye and reeds. Using the plan-making tool, they could decide to put aside one fourth of each of their next few transfers for those purchases. Then, once they receive their cash, they could put their specified amount aside in their pouch to keep it separate from the money they’ll take to the market.
Of course, these designs were not one-size-fits-all. In fact, helping recipients save their cash is not always good advice. In South Sudan, encouraging cash transfer recipients to save would have been impractical or even harmful—rapid depreciation of the South Sudanese pound means that it loses value almost immediately. In this context, we had to dig deeper to learn how South Sudanese people were “saving without saving.” The solution was a plan-making tool that prompted participants to identify what items they could purchase immediately that would maintain their value, such as dried food that could be sold later or tools necessary to build their small business.
In South Sudan, encouraging cash transfer recipients to save would have been impractical or even harmful—rapid depreciation of the South Sudanese pound means that it loses value almost immediately.
In other cases, it was important to match the types of tools with the medium of delivery. In the Democratic Republic of Congo, cash transfers were completely digital: participants could register via text and receive their cash electronically. So we adapted the interventions to be digital as well. One day before the transfer, participants received goal-setting and plan-making prompts over text. One day after their transfer, they received a text reminding them to separate their money according to their plan.
The extra effort to understand the context was crucial in designing interventions that worked—if we had simply copied and pasted designs that promoted saving, what was helpful in one country likely would have been ineffective or even harmful in another.
Between 2017 and 2022, we conducted five RCTs to assess these customized designs and found encouraging results. The structure of the tests varied—from smaller, individually randomized pilot tests where outcomes were tracked one to two months after the interventions were delivered to much larger cluster-randomized trials.
A large trial in Tanzania in 64 villages with over 2,600 participants showed that six months after the intervention, participants were 65 percent more likely to report saving and 22 percent more likely to report that they made a productive investment. Participants planned to use these investments—such as livestock, farm materials, or items for their small businesses—to generate income in the future, a step toward improving their livelihoods and escaping poverty.
In shorter, smaller trials in Kenya, South Sudan, and the Democratic Republic of Congo, we found evidence that the interventions can support recipients in overcoming barriers they face and support them in achieving their short- and long-term goals. In Kenya, interventions increased the amount participants saved from their transfer by five percent. In the Democratic Republic of the Congo and South Sudan, participants were four percent and seven percent more likely to report spending on their priorities. We currently have two additional six-month tests in progress in Ghana and Ethiopia.
By incorporating these insights, cash transfer programs can make their limited resources go further, helping them reach even more households.
Finally, we calculated a cost-effectiveness multiplier as an alternative measure of the impact of our approach. It takes into account the per-person cost of the intervention, the effect of the intervention, and the change in behavior we would expect if people received the cost of the intervention in cash (rather than the intervention itself). This helps us estimate how much additional cash participants would have to receive to match the positive impact of our behavioral interventions. We estimate that when they are delivered in-person with physical materials, cash transfers are 1.2 times more cost effective than providing extra cash. And we found that they were up to eight times more cost-effective when sent via text message. By incorporating these insights, cash transfer programs can make their limited resources go further, helping them reach even more households.
These behavioral designs have reached almost 200,000 recipients so far. While our steps toward scaling are in early stages, we’ve seen some successes—the government of Tanzania has incorporated money into their program budget to scale the interventions to half a million households. However, this is just a small portion of cash transfer recipients worldwide. Cash transfers have grown in number and reach during the pandemic: since March 2020, 17 percent of the world’s population has received at least one COVID-related cash payment.
Budgeting the cash from these transfers isn’t easy—especially under conditions of scarcity, whether brought on by chronic poverty or exacerbated by pandemic-related shocks. But we can make it easier. Building in decision-making tools grounded in an understanding of human behavior is a natural and essential step toward more effectively reducing extreme poverty around the world.
Disclosure: Kate MacLeod is an employee at ideas42, which provides financial support to Behavioral Scientist as a Founding Partner. Founding Partners do not play a role in the editorial decisions of the magazine.
