The World Bank is now using behavioural insights to tackle wicked problems.
The World Bank’s Global Insights Initiative (GINI) was formed to put the key findings of your World Development Report into practice. Please summarize those findings.
The main message that came out of that report was that development policy is due for a redesign, based on a more realistic understanding of how people think, behave and make decisions.
In contrast to the rational model — whereby people have unlimited computational power to make rational decisions on the basis of consistent and self-interested preferences and explicit information — realistic thinking is very different. It relies on three principles: We usually think automatically and fast; we are highly influenced by our environment and by social norms; and explicit and implicit information derived from our worldviews (i.e. mental models) affect how we interpret information and make decisions. In the report, we emphasized that everyone thinks this way — not only end users and consumers, but policy makers, heads of companies, and even heads of state.
Put simply, decision making is highly contextual. You might typically assume that if you raise the price of
something by 10 per cent, people will demand less of it and supply will increase; but when you make something 10 per cent more salient in an interaction, depending on the environment, a wide variety of things can happen. That’s why it is so important to understand the decision-making context.
Why are insights from behavioural science so well suited to the policy realm?
Because every policy relies on explicit and implicit assumptions about how people make decisions and why they do what they do. As indicated, most of those assumptions rely on the rational, idealized model of how people think. By embracing behavioural insights, we can redesign public policies using new tools, including choice architecture; social rewards and recognition; the way information is presented; and activating new social norms.
Your goal is to incorporate behavioural insights into some of the World Bank’s work. How are you going about this?
We have three main criteria for selecting projects. The first is, do they address an important developmental challenge? Areas that address basic human rights are given high priority. Second, is the initiative ‘owned’ by the World Bank, or a development partner? That is really important for sustainability and ensuring a long-term impact. Then finally, can the intervention be evaluated? At this stage of our work, it’s really important to show that we’re having an impact, so we look for projects where we can provide some form of evaluation.
So far, our main areas of work are in tax compliance, digital finance and savings, education, gender norms and organizational effectiveness.
Why is this ‘extra work’ necessary?
Because time and again, we find that people have a very hard time following through on their commitments. We might want to save more, lose weight, or stop smoking, but we just can’t follow through because we overvalue the present and the willpower exertion is too large. Governments often offer programs — from conditional cash transfers to health and education programs — and are shocked when people don’t take them up on it. This happens a lot, particularly in the poorest environments. The question is, why? One reason is that availing yourself of such programs requires effort and takes time. Even though in the long run, it makes sense to take advantage of the program, in the short run, people may not have the bandwidth to do it.
As a result, all people — rich and poor alike — sometimes make choices that don’t promote their own well-being. People may also get stuck in habits, succumb to inertia, and repeatedly procrastinate despite intentions to do otherwise.
There was a study in Morocco in which the government was providing a credit subsidy for a water hookup. Water availability is a big problem there: people spend seven to nine hours each week collecting water, and it is often the women doing the work. Yet, people weren’t signing up for the program. The researchers found that the problem was the ‘hassle factor’ — making time to sign up for the program. So, they decided to go directly to peoples’ homes and get the necessary documentation, and they brought it to the municipality. The result: the sign up rate went up from a baseline of 10 per cent to 69 per cent! The implication is that making things easier for people can have a huge impact.
Should governments be getting involved in shaping individual choices?
I believe they should, for three reasons. First, doing so helps people obtain their own goals. Reminders to save money or take medicine help people achieve objectives that they themselves have set. Research shows that commitment contracts, which markets under-provide, can reinforce decisions to adopt beneficial behaviours. Matching the
Development policy is due for a redesign, based on a more realistic understanding of how people think.
timing of social transfers to the timing of charges for school enrollment, or making it easier to buy fertilizer at harvest time when cash is at hand, helps to overcome intention-to-action divides for people who may be forgetful or possess insufficient willpower (that is to say, most of us). Many development policies that operate at the boundary of Economics and Psychology can be understood in these terms.
Second, because decision making is often based on only the most accessible and salient information — and is also influenced by subtle social pressures and existing mental models — peoples’ preferences and immediate aims do not always advance their own interests. Individuals might choose differently, in ways more consistent with their highest aspirations, if they had more time and scope for reflection. The assumption that we always make choices that promote our own interests — often a fundamental benchmark for policy analysis—is misguided. But if decision makers do at times require assistance, what guidelines are to be used for the policy interventions aimed at shaping choice? We believe that governments should focus on the most important freedoms, and in the development context, those include freedom from poverty, disease, and oppression.
Third, socially-reinforced practices can block choices that promote well-being and prevent individuals from even conceiving of certain courses of action. Our Report argues that social interdependence and shared mental models affect choices, sometimes creating traps for communities and individuals, including low trust, ethnic prejudice, and gender discrimination.
You believe that behaviourally-informed approaches and traditional economic interventions can be complementary— and that some of the biggest payoffs emerge from adapting traditional interventions. Please walk us through this.
Traditional interventions usually take two forms: incentives and education/information. Incentives usually take the form of changing prices in some way, or changing penalties associated with rules. The key idea is that people don’t respond to incentives plainly: they respond to incentives as they are represented in their own minds. Sometimes, they forget about the consequences of a certain behaviour and pay little attention to it; or they might miscalculate the probabilities associated with a penalty. This provides scope for changing interventions that highlight penalties. For instance, in the tax compliance domain, people know that there’s a penalty for non-payment, but if you highlight this fact, it might change how they behave. If you think about it, virtually all information can be made more salient, more timely and easier to understand.
How can leaders know if behavioural insights apply to a particular project?
They would need to think through a few things. First, are individuals aware of what they need to do, but unable to accomplish it, or does a desired behaviour need to be activated? Second, are individuals motivated enough to nudge themselves? Third, will more deliberation lead people to undertake the desired behaviour, or are they already suffering cognitive overload? And fourth, is the desired action not being accomplished because of a competing action, or due to inertia? Consequently, should you aim to discourage a competing action or encourage a target action?
GINI has developed a five-step process to using behavioural insights (see page 116). Which step have you found to be the most challenging?
The most challenging is probably learning from piloting an intervention (the fifth and last step), because time is always short for policymakers and they tend to want things done very fast. Carving out time for a pilot, especially one that has learning embedded into it, requires a mindset change. We have found it challenging to make the case to policymakers about the importance of this step.
People don’t respond to incentives plainly: We respond to incentives as they are represented in our own minds.
The problem is, policymakers are often looking for offthe-shelf solutions, in the way that economists might advise them to ‘raise the price of something by five per cent’. Done. They don’t quite realize that if you are going to walk down this road of believing that decision making is highly contextual, you need to take the time to make sure you get your intervention right. Maybe someday we’ll be able to give a quick prescription like, ‘Make this information 10 per cent more salient’; but we’re not there yet.
Also, the target behaviour (step 2) needs to be very precise. It should not be decomposable into other constituent behaviours. For example, ‘increasing retirement savings of a household’ is general and could be decomposed into precise behaviours such as ‘opening a new retirement account’, ‘setting aside more money each month’, ‘reducing spending on discretionary items’, etc. The behaviour should also be easily observable.
Describe some of the complex issues you are currently tackling.
We are finding that a richer understanding of why people save, use preventive health care, work hard, learn and conserve energy can provide a basis for innovative and often very cost-effective policies and development interventions.
At the moment, we have a few projects on increasing women’s participation in economic and social life. In Jordan, we’re doing a very long pilot. Rather than going right to an intervention, we’re trying to understand how social norms influence women’s decisions about whether to join the labour force, participate in public life, and open a bank account.
We started out with a survey that asks households — men and women — about how they perceive social norms in this area, what they think other people do and what they think other people think they should do. We’re also looking at implicit biases, to see whether these affect decision making on the part of women as they decide whether or not to join the workforce. We’re taking a long time to do this, because we want to really understand whether our interventions should target social norms, the stories people tell about themselves or implicit biases.
Looking ahead, how do you see the GINI initiative evolving?
Our aim is to start using behavioural insights to address some of the world’s biggest problems. What we need to do, therefore, is remain relevant to policymakers by working with them to enable ‘quick wins’. For example, we can help them collect more revenues through taxes relatively quickly, or reduce energy consumption with changes in messaging. We are truly understanding more about influencing peoples’ behaviour every day.
Varun Gauri is Head of the Global Insights Initiative (GINI) and Senior Economist in the Development Economics Vice Presidency of the World Bank. He serves on the editorial boards of the journals Behavioural
Public Policy and Health and Human Rights, the World Economic Forum Council on Behaviour and the Advisory Board of Academics Stand Against Poverty.
Nina Mažar is an Associate Professor of Marketing at the Rotman School of Management and Co-director of Behavioural Economics in Action @ Rotman (BEAR). She is on leave from the School until November of 2017 to serve as a Senior Behavioural Scientist for the Global Insights Initiative (GINI) at the World Bank.