eMBeDding behavioral insights in development projects – an update


This page in:

Also available in: Español, Français中文

People think fast and often automatically, respond strongly to social incentives, and use mental models or specific worldviews to interpret information and perceptions. So, shouldn’t we be taking into account their thinking and behaviors while designing policies? 

The answer to that came from the World Bank’s 2015 World Development Report on Mind, Society, and Behavior, which demonstrated that a more realistic understanding of choice and behavior can make development interventions much more effective. Taking this as a cue, we launched two parallel initiatives that focused on finding ways to encourage and support project teams at the World Bank in incorporating behavioral insights into their projects. The models had a similar approach: find interested teams in the institution, embed as team members, diagnose a behavioral problem they face, come up with potential solutions, and test them rigorously.

Today, we collectively have more than 80 behaviorally informed projects underway across 50 countries in a wide range of sectors, from health, gender, education, environment, and water and sanitation to financial inclusion, jobs, and taxation. We are tackling problems as simple as increasing tax compliance in Central America and as complex as reducing sex preferences at birth in Georgia, where parents’ preference for sons is distorting the sex ratio at birth.

Positive results

We are seeing positive and lasting results in our projects. Take the example of Peru, where there are wide gaps in test scores between students from high- and low-income households. The usual approach to address this problem would be to invest more on teacher training and learning materials. Instead, the team designed a growth mindset intervention to change beliefs and mental models on the part of educators and students alike. An impact evaluation showed that this simple, one-time 90-minute session increased test scores by 0.2 standard deviations, roughly the size of having a parent with three more years of education. The cost of the intervention was $0.20 cents per student, which allowed the pilot to reach 50,000 students in the first year. Now the intervention is being implemented in Indonesia and South Africa reaching more than 500,000 students.

You may be wondering why such interventions aren’t applied to more projects. In fact, when we engaged with project teams, we found a huge appetite for this type of work, but very little time for teams to see it through. In our lingo, teams had low cognitive bandwidth, juggling too many tasks and hence tended to focus on immediate goals and deadlines. So while more often than not, teams knew exactly what they needed to do (behaviorally speaking) – they just needed a little push from us to help them do it. And so we did.


Going forward, such support is getting more streamlined. The promising success of the two parallel teams has led to the establishment of a new unit, eMBeD (Mind, Behavior, and Development), to mainstream and scale up the use of behavioral science in World Bank Group operations.  The new unit will help diagnose a wide set of psychological, social and economic factors that influence decision making and offer low-cost and quick solutions that can dramatically increase project impacts.

Our aspirations

As we eMBeD ourselves in projects, we also hope to build capacity within the World Bank to use behavioral insights across the institution in an organic, cost-effective, and hands-on way. And we will continue to grapple with how behavioral science can help tackle complex problems, such as the challenges around the refugee crisis and gender-based violence.

We also want to build capacity in our government clients, like we did with the Ministry of Education in Peru where we supported them to establish MineduLab, an in-house cost-effective innovation lab. Today, MineduLab is running more than 15 interventions to reduce teacher biases, improve teacher and student motivation or increase parents’ engagement.

Ultimately, we want to create a one-stop-shop where everyone can come to find behavioral solutions that can be adapted to their needs, which at the end will be the way to scale up this work.

We would love to hear about the problems that you are trying to solve. Visit our website or sign up for our newsletter to stay connected.


Renos Vakis

Lead Economist, Poverty and Equity Global Practice, World Bank

Varun Gauri

Senior Economist, Development Research Group, World Bank

Join the Conversation

Leonard Haggai Oduori
April 27, 2017

For the first time, I can see World Bank becoming innovative by adopting a behavioral approach to development. Local experts need to buy into the idea of helping communities and individuals to learn and change behavior by evaluating positive and negative cultural norms and values which influence their decisions and choices in development activities. There are too many technologies and practices which communities have either ignored or simply talk about but do not adopt to change their situation. Eating habits and farming practices (Monoculture) are a case in point. These behaviors and practices are contributing to malnutrition, food price volatility, thin markets for traditional foods, poor land use and worsening of climate change effects. Whole communities and influential individuals need to change their mindsets, behavior and role in development investments.

May 03, 2017

Dear Leonard
Thank you for your comment. Could not agree more, mental models are indeed key for how we interpet the world and therefore how we make decisions. You correctly point to the need to do better diagnostics in the context we work to better understand what those belief systems are so we can come up with solutions that take them into consideration.