Samantha Power, Administrator USAID
This post is adapted by the BPP Blog from USAID Administrator Samantha Power’s keynote remarks to launch United Nations Behavioural Science Week on June 21
I should declare that I am not a behavioral scientist, nor do I play one in the media. I am somebody who has the privilege of running the United States Agency for International Development (USAID), and I am committed to working with my colleagues to ensure that behaviorally-informed public policy becomes even more central to our programs.
When we talk about behavioral public policy, much of it can sound intuitive and intrinsically appealing: low-cost interventions that have great outcomes. What could be wrong with that?
But please hear this at the outset: None of us are yet getting the most out of behavioral insights. And we will not get the most out of them — we will not optimize in our programming and in our leadership — if we do not depart in critical ways from “business as usual”.
On the one hand, everyone of course knows that human behavior matters. This is not a novel concept. We have all insisted for a long time, for example, on people-centered design, on paying attention to culture, and on emphasizing local conditions. I hope that all of this is now baked into our DNA as policymakers. Even if we are not always perfect in our adherence to these ideas, they are very familiar to us, they are broad and helpful, and they have become an important part of the work of the UN and USAID.
Broad principles, however, are not really what behavioral scientists in London, New South Wales, Buenos Aires, New Delhi, or Cape Town mean when they talk about making modern behavioral science central to public policy.
And what we’re talking about today is how we can achieve a “delta” — a real measurable difference — at places like USAID, and in development work more broadly between where we are now and the full use of what is now available.
Behavioral science is much more specific and more concrete than simple broad principles about meeting people where they are, about respecting local culture, about taking note of human behavior and adherence.
Behavioral science is new. It is not old. It is not actually all that familiar. It is not what we have been doing all along.
It’s really important to bear in mind that standard economic theory, which informed public policy across the board, stipulated that human beings are rational — that we are good planners; that we respond well to incentives; that we consider the long term as well as the short term; that when we are given information, we process it ably; that we are neither too optimistic nor too pessimistic; and that we have a decent ability to figure out whether risks are small or large.
These were the going-in assumptions, and they were the predicates on which public policy was shaped and implemented for a very long time.
Modern behavioral science has questioned all of these standard assumptions. The work of Daniel Kahneman and Amos Tversky challenged conventional wisdom about human behavior, showing not that people are “irrational” (the simplistic way in which economists can talk about this), but that we depart from the concept of “perfect rationality” in specific and predictable ways. We often do so instantaneously and unconsciously.
Over recent decades, extensive research has shown, for example, that human beings focus on the short-term and often ignore the long term.
Research has shown that people can be unrealistically optimistic. We exaggerate some risks and ignore others altogether.
Even in circumstances in which we know that making a change is in our interest, we can have an overwhelming attachment to the status quo. Inertia is an immensely powerful force over our daily lives.
Research has shown that we care about what is fair and what is perceived to be fair. This means that we will be inclined to sacrifice our material interests in order to comply with social expectations. These “social norms” are an incredible weapon in our collective toolbox for public policy.
Much of this sounds rather obvious now. But public policy has not started to be predicated on these findings until very recently — and these findings have specific and concrete implications for our work in development. They suggest that our long-held intuitions about behavior might not just be a little wrong, but might be completely wrong. And if they are completely wrong, that is going to be harmful to our long-term development goals.
If you take these findings to their natural conclusion, it would suggest that what we are now doing in a whole range of areas might not work. But behavioral science research also suggests very promising paths forward. I’ll briefly offer two cases that demonstrate how promising these paths are.
“Speak Up!” Tackling Public Transport Safety
Kenya has an expansive public transportation network, but also experiences frequent fatal bus crashes. In April 2021, an accident in southeastern Kenya left 15 people dead. This is an all-too-common occurrence in a country where fatalities from road accidents exceed 3,000 annually. These numbers represent people who rely on public transit for their livelihoods.
When a pair of Georgetown University researchers, in partnership with USAID, set out to learn more about what was behind these accidents, they found passengers were not comfortable calling out drivers for speeding or aggressive driving. The perceived social cost of being the first to complain was too high. People did not know if others would join them, especially given the desire to arrive quickly to one’s destination. But the researchers found that specific reckless behaviors could be influenced by changing social norms.
They devised and tested a simple, inexpensive solution built on behavioral insights: stickers. By putting stickers on buses with motivational messages to “Speak up!” about reckless driving, people were made to feel more comfortable telling bus drivers to slow down. The results (as detailed in the researchers’ studies from 2010 and 2015) were significant compared to rates for buses without the “evocative messages”. People felt empowered to speak out. This straightforward and cost-effective intervention led to fewer deaths, and road accident insurance claims dropped by one-third.
Today, some 4 million people ride on buses with these stickers. That is the impact that behaviorally informed initiatives can have on people throughout the world.
Tackling Stigma; Complexity in HIV Prevention
Women make up more than half of people living with HIV around the world, yet research shows that the stigma of taking pills and visiting clinics, as well as the complexity of adhering to a specific regimen, causes many young people—and many young women in particular — to avoid taking a once-a-day HIV prevention pill known as PrEP. “If it’s medical, you’re sick,” was the common refrain heard by development professionals in the field.
Remember, the old development models would have baked in the assumption that people benefiting from life-saving drugs would pursue their own interest in obtaining them. But it turned out that there was way more standing in the way than had been previously understood, and simply funding more awareness campaigns couldn’t address those hurdles.
After extensively engaging young women, USAID helped develop a way of delivering PrEP that fits into a routine and encourages women to generate a positive association with the treatment. Rebranded as “V,” PrEP is offered in a pill container that mimics lip gloss and would look normal in any makeup case. Taking PrEP is presented like any other daily self-care routine.
What looks like lip gloss actually contains a full week’s worth of medication. Text messages nudge participants to keep focusing on making themselves feel good by remembering to take the day’s pill, and rather than asking women to embrace new rituals, staying safe from HIV is framed as being part of the already routine behavior of putting on makeup and looking good for friends and family.
Behavioral Insights and Public Policy
The central premise of behavioral science is simple. To make progress, we have to understand human behavior, not on the basis of intuitions, but using new findings and concrete data. We must learn how the people we hope to serve act (or do not act) in response to everyday challenges. And rather than making assumptions or applying what works in one culture to another, we need to gather evidence and data from the specific communities in which we serve.
From the pages of academic journals into the building blocks of actual policies, behavioral science insights enhance the potency of our efforts. They allow us to use precommitment strategies to help workers save more for retirement as they get pay raises later in their careers—an approach that can more than double average saving rates. Behavioral science helps us get more people vaccinated against COVID-19 by recognizing that the accessibility of vaccination sites is half the battle. Behavioral science helps us develop more effective approaches to reducing greenhouse gas emissions by changing utility companies’ default settings, allowing people to opt out rather than requiring them to opt in for using green energy.
USAID works in more than one hundred countries and is one of the world’s leading funders of research and programs that incorporate behavioral findings. Our work has evolved to include insights from behavioral economics, human centered design, behavior change, communications, and the use of “nudges”. The evidence shows that all of this can save lives and improve livelihoods.
More examples abound of how behavioral insights are changing policy:
People’s choices are heavily influenced by inertia, starting points and loss aversion. We can use strategies such as automatic enrolment (as mentioned above) to make sure important programs are reaching more people.
We know that complexity, jargon, and small print (things that the UN may lead the world in, and USAID certainly can hold its own in that competition) can defeat the whole enterprise of increasing people’s capacity for agency. We also know that pervasive features of life, including poverty, hunger, and loneliness, impose significant cognitive costs that make it even more difficult to solve problems. Simplification, reducing barriers to access information, and disclosures are more important tools than previously recognized.
Even when people do want to change, we have limited attention and that stands in our way — so consistent and well-designed reminders that point to a desired destination can make a major difference in helping people remain focused on what they are trying to achieve.
Graphic warnings can address our unrealistic optimism and engage the instantaneous affective responses that drive so many of our decisions. Think of those on cigarette boxes, for example.
The choice architecture determining how options are presented greatly influences our decisions. What people see first, whether it’s a name on an election ballot or certain kinds of food in a cafeteria, makes a big impression on our ultimate choices.
By invoking social norms and particularly emerging norms, like “more and more people are switching to green energy”, we can marshal the power of informational cascades as well as our innate desire to be on the right side of history.
I have not yet mentioned directly the behavioral analysis around “sludge”. This comprises the needless paperwork burdens and administrative requirements that get in the way of beneficiaries actually coming to programs. There is often so much paperwork that stands in the way. I know from my time at the UN that paperwork is a not insignificant administrative burden there as well. At USAID, part of our future efforts will include cutting down on the sludge that deters potential partners from working with us, and slows us down when contemporary challenges require agility and speed.
An exciting future
What we are doing with behavioral science and insights at USAID is only the beginning. I could not be more excited about the new research being conducted around the world every day, research that we want to incorporate in our work wherever we have programs. It is also thrilling to have seen the World Bank adopt behavioral insights six years ago, and to now see the UN Secretary General adopting this new guidance on behavioral science — thrilling because in order to be faithful to the people that we are attempting to serve, in order to contemplate advances of the scale we know we need in fulfilling the Sustainable Development Goals, and in order to fulfil national plans and development objectives, we need to optimize. We need to lose the reliance on perfectly rational individuals. We need to inject this learning into our programming. It could not be a more exciting new frontier for us to work on together.
We have to be realistic about the obstacles, and addressing them forthrightly will be necessary for the Secretary-General’s guidance to have the impact it envisions. The same is true for the behavioral agenda I hope to see USAID accomplish in the coming years. But we know we can accelerate our own progress toward the global community’s goals, and we look forward to seeing our UN partners help lead the way.
When I became USAID Administrator, my message to the Agency was that for all the lofty pronouncements we can make about America’s global leadership, it is our country’s actions — our ambition and our ability to get big things done — that truly moves minds and changes futures. Behavioral insights have so much to offer in this urgent work, by putting individuals’ most basic needs at the forefront of policy decisions and generating the pathways of hope that flow when we invest in a person’s agency and self-reliance.
I am immensely honored to be associated with the Secretary-General’s guidance and its launch. This will have been a really important new set of innovations at the United Nations that will make a profound difference for people around the world.
Samantha Power serves in the Biden-Harris Administration as Administrator of the United States Agency for International Development. She previously served in the Obama-Biden Administration as the U.S. Ambassador to the United Nations.