Behavioural insights to guide Covid-19 containment strategies

The main purpose of this article is to discuss how research findings can inform policies/strategies for Covid-19 containment in the context of a developing country like Bangladesh. Primarily, insights will be shared from previous research in the field of behavioural economics, done with my co-authors, over the past decade(s) in order to shed light of those general findings in the current context to promote voluntary compliance by ordinary citizens with respect to the expert's guidance as well as government's directives related to Covid-19 health protocol. In doing so, the article will also refer to other research findings with greater relevance to the issue at hand.
There is a large amount of research suggesting why people might be in disagreement with experts regarding the judgement of risk. The primary reason is that experts and policymakers possess more information about the magnitude of the risk at issue than the ordinary citizen; the ordinary citizen lacks such information. In evaluating risks experts mostly focus on numberslives or life years at stake, while ordinary people's risk judgement might be affected by many other qualitative factors (and not just the number of lives at risk), including a rough sense of experts' concern, such as the understanding of risk, its seriousness, susceptibility to risk, voluntariness of risk, controllability of risk, time of effect, whether children are at special risk, and not least, trust in applicable institutions. Often people fail to take into account the externality involved in their actions and consider large risks as part of life and treat small (statistical) risks more seriously, resulting in failure of overall risk management. Intrinsic motivation such as trust, altruism and cooperation can potentially amend failure in risk management.
Based on research findings related to risk perception and risk valuation by people in Bangladesh, which are consistent with research findings in psychology elsewhere, the following main points warrant attention: people's risk perception is biased implying that they systematically underestimate the large risks and overestimate the small risks and that people attach low value to risk reduction(s). For example, when informed about age-specific mortality risks in Bangladesh, it was found on average that young individuals overestimate their own mortality risks and that older individuals underestimate their own mortality risks. In separate studies, when asked about their willingness to pay for mortality risk (associated with contagious disease) reduction through participating in vaccination programme and mortality risk reduction from air pollution through paying for air pollution improvement programme, average valuation is found very low compared to similar studies in other countries, while also suggesting the role of affect in their risk valuation. In another study, whereby individuals were asked about their willingness to pay for safety device that would reduce mortality risk from road accident, it was found the average valuation of mortality risk reduction in the magnitude compared to studies elsewhere in the world. This divergence in valuation of risk reduction might be attributed to the nature of the good in question, public vs private good, ceteris paribus; in the first two cases, there are externalities involved, although vaccine is regarded as private good. In other studies eliciting people's ethical preferences over life-saving programmes, whereby people were asked to choose between alternative life-saving programs, they prefer to save more lives and life years; people particularly prefer to save younger lives. Also, in a societal context, people prefer to save lives from the risks of water contamination and air pollution (involuntary risks), more than from the risks of road accidents (voluntary risks). The expense of controlling risks may affect valuation as people may find it is expensive to control risks; controllability is slightly problematic if individuals suffer from asthma. Also, people may think it's expensive for them to avoid risks. It was also found that people in disaster affected areas become more risk loving ex-post natural disaster than people in unaffected areas implying that big negative events shape individuals' preferences towards risk.
With respect to policy, people's value judgement (not mistakes) may be incorporated to the extent that can survive a process of reflection. In addressing the divergence of risk judgment between experts and ordinary citizens and the implication of it, the most important is to inform people of real facts; communicate the benefits of risk reduction; spend more resources where the cost of controlling risk by individuals seems higher. It was found that explaining risk information and the concept of risk reduction improves individuals' risk assessment and valuation of risk reduction. Research also shows that risk attitude or risky behaviour can be addressed using appropriate incentives and nudge.
Insights from behavioural sciences suggest that human behaviour is greatly influenced by the context or environment within which they make decisions and choices and that there is evidence that health interventions (e.g. handwashing, water treatment) focusing on information provision often show limited effectiveness in influencing behaviour as they fail to influence the way people process information.. This shows the importance of nudge in changing behaviour. In the context of Covid-19 health behaviour, nudging people in a way so as to change their belief about the risk to health that they themselves and others are posing, and that others are going to cooperate or reciprocate with desired behaviour, remains extremely important. Because everyone is better off with an outcome when cooperation is increased, whereby everyone is wearing masks (safe use) and washing hands frequently. If it is regarded as good, individual and pro-social behaviour, people would do that for various reasons -for its expected benefit, to be good to others- other-regarding behaviour and solidarity or for benefit in afterlife. It was found that people tend to help others for benefit in the after-life. It was also found that people express more solidarity ex-post disaster as substantial risk sharing was observed among people in affected areas ex-post disaster, where people are more likely to form risk sharing groups with many others than people in unaffected areas.How to trigger desired behaviour in the context of Covid-19 risk containment? Most importantly, communicating the benefit of risk reduction to the community vis a vis individual risk reduction. For this, people need to be asked unambiguously of maintaining expected behaviour with pro-social messaging, for example, don't spread the virus, others are at risk from your behaviour and they expect you to comply. It is also important who communicates the risk information. Trust in applicable individuals and institutions is crucial in this context. Belief about cooperation increases trusting behaviour as observed in many research. For example, religious and or patriotic priming might be useful along with pro-social messaging. One way to trigger people's desire to be seen as a good member of the society is attributing the credit of good social outcome to individuals' behaviour. Importantly framing behaviour change messages in a way that the act of free riding (or believing that I don't have to comply as others are complying already) is discouraged. Now if there is reason to believe that others are not willing to cooperate- I may give up! For greater social benefit, all need to cooperate, whereby the role of leadership/coordination is to align everyone's belief on cooperation behaviour; increasing belief that others are willing to cooperate is important. There was strong association between interpersonal trust and trust in institutions and that trust in local institutions is relatively higher than trust in, say, higher order institutions. Local level coordination can ensure that everyone's belief towards cooperation, maintaining health guidelines, is aligned. It is therefore crucial to incentivise the local level trustworthy leaders/institutions who will align everyone's belief in cooperation. The starting point is thus well defined as our research in the context of Bangladesh suggest that people have biased risk perception, their risk judgment depends on many factors, their behaviour is influenced by the context or environment they make choices- preferences are malleable. Moreover, they exhibit other-regarding behaviour, they are conditional cooperators- willing to give benefit of doubt, trust local institutions more. Taking into account of these insights would make Covid-19 containment strategies smart.
To end this article, it is worth mentioning the World Health Organisation's principle regarding the Covid-19 management, which states that the communities should have a voice, they need to be informed and engaged and participate in the process of risk containment.

Dr Minhaj Mahmud is a Senior Research Fellow of Bangladesh Institute of Development Studies (BIDS).
[email protected]