Mirror, mirror on the wall: Applying a behavioral science lens to decision-making

What do a minister in the Philippines, a mayor in Nigeria, a public officer in Macedonia, and you all have in common? More than you might think.  

In a recent paper, we discuss how decision-making in the public policy sphere is influenced by three overlapping contexts: institutional, group, and individual.

• The institutional context refers to how systems are established in each setting or organization; how rules and processes are designed, implemented, and monitored.  

• The group context refers to the role of social dimensions, such as the individual’s group identity, social norms or rules, and standards regarding behaviors in a society.  

• Finally, the individual context encompasses dimensions pertaining to one’s motivation to perform, interpretation of information, and abilities. It covers multiple aspects, such as the role of monetary and non-monetary rewards, beliefs about how the world works, cognitive biases, and non-cognitive skills. 

These three contexts are interdependent. For instance, beliefs, motivation, and social identities can influence the behavior of those working in policymaking toward beneficiaries of projects. Likewise, organizational culture and bureaucratic norms can help shape decisions and actions of civil servants. Ultimately, the way these contexts materialize in each setting will influence policy professionals’ behaviors and decisions and, ultimately, the success of policies and programs. 

Not me, you are thinking. Yet a recent experiment done with staff from the World Bank and the United Kingdom’s Foreign, Commonwealth & Development Office (FCDO) found that confirmation bias — when we interpret data by looking for patterns that confirm our prior beliefs — inhibited staff from answering an objective question correctly. Meanwhile, sunk-cost bias — when it’s hard to let go after we invested a lot of effort or resources in something — resulted in few people admitting failure and stopping a project. 

So, just like us, anyone — from presidents, ministers, judges, policymakers at large to frontline workers like teachers and nurses – who interacts directly with people decides behaviorally. Understanding this and exploring ways to address it can therefore play a critical role in service delivery and development outcomes. 

Looking in the mirror — and evolving 

By highlighting examples of recent work, we illustrate how the use of behavioral thinking can help identify behavioral bottlenecks among policy professionals and potentially address them for better policy outcomes.   

For example, recent projections suggest that targets set by the Philippines for child stunting prevalence for 2025 will need another 80 years to be met despite four decades of state-sponsored programs to improve child feeding and nutrition practices. These policies have typically targeted pregnant women, lactating mothers, and new fathers.  

Surprisingly, a large-scale national study that we conducted found that what is influencing directly nutrition outcomes is actually the beliefs of frontline health workers. While many of these workers held the belief that nutrition and prenatal care are the key cause of stunting — issues that could be addressed — there was a large share of health workers that believed that genetics and race are the cause of stunting. In other words, they believed that stunting was unavoidable. Even more telling, the analysis revealed that this second group of frontline workers was more likely to work in facilities with significantly worse maternal and child nutrition services and outcomes (Sen et al., 2020). Could their beliefs affect their own behaviors and effort? 

In Nigeria, we designed an intervention that asked people to display a “certificate of excellence” prominently in health facilities, distinguishing workers in the best performing facility. This led to improved workers’ performance. The key to the success of this inexpensive four-week intervention was the improved sense of accountability (institutional context) by providing social recognition (group context), which increased workers’ motivation to perform the needed record-keeping tasks. 

Finally, looking at the current challenges around the COVID-19 vaccine roll-out, ongoing work indicates that health workers’ own beliefs on vaccine efficacy and safety are likely to affect their motivation as well as communication to patients around vaccine uptake. In the 19 countries we have conducted experiments so far, we also have found that hesitancy among health workers is as widespread as the levels we find among the general population. This is a crucial (and worrisome) insight since the analysis also revealed that health workers are the main source of credible health-related information for most people, including on vaccine safety and efficacy. Do we expect a nurse or doctor to help a patient get the vaccine if they do not believe that themselves? Using these results, we are now designing communication and training approaches that can reduce misinformation among health frontline workers and help them make informed, science-based decisions around vaccination. 

So what about that mirror? A first step in solving any problem requires acknowledging it and reflecting. Behavioral science can provide a useful lens to understanding how all of us working in public policy make decisions (just like every human on earth) and how to try to overcome the challenges in policy design and implementation, by addressing bottlenecks linked to individual’s biases, beliefs, motivations, group dynamics, and organizational processes. Becoming behavioral science professionals could be what it takes to unlock developing impacts at scale around the world.