Covid and the Limits of ‘Nudge’

In recent years the idea that governments can use ‘nudges’ to influence how people behave has become very fashionable. A ‘nudge’ can be defined as a means of influencing individual and social behaviour, for example through the provision of information, that does not involve official coercion. A ‘nudge’ is contrasted with a law or regulation that must be obeyed, or a tax that must be paid, and that involve penalties if they are disregarded. This blog contends that the experience with behaviour in the COVID pandemic suggests that the claims in favour of ‘nudges’ have been greatly exaggerated.

The cognitive background
 
In 1901, an academic at Stanford University, Edward Ross, published a book, ’Social Control: A Survey of the Foundations of Order’, that marked the start of modern discussion of regulation and regulatory techniques. Ross described the many ways in which societies can coordinate social relationships and, in particular, how societies can coordinate relationships without government or the law. He influenced much subsequent writing on regulatory techniques and on the ways in which regulators, and those they regulate, respond to each other, often in informal ways.
 
In the post war world, the motivation to look further into the techniques of persuasion received an enormous boost from cognitive research into the way people actually reason. In economics, Herbert Simon argued that people take shortcuts in the way they reason, are rational in bounded ways that takes into account the time and effort in gathering information. They often resort to simplifying assumptions to frame and make sense of the world around them. Amos Tversky and Daniel Kahneman launched a program of cognitive research into ‘heuristics and biases’ that has had widespread influence, popularised in such distinctions as ‘fast’ and ‘slow’ reasoning and thinking.
 
Both economics and political science have had to incorporate the findings coming out of this research program. In economics it goes to the heart of how people respond to market incentives, and how far their behaviour ‘maximises’ the pursuit of a self-interested objective. In politics it goes to the heart of how far we can treat the average citizen or voter as open to reason and deliberation in elections, ‘Town Hall’ gatherings and ‘Citizen Assemblies’. It identifies what styles and processes of political communication might lead people to change their mind or to confirm their prior beliefs.
 
Nudges arose out of this programme because they seemed pertinent both to ‘fast’ reasoning where people choose based on prior positions, or other short cuts, as well as ‘slow’ reasoning where people are ready to absorb and digest new and pertinent information about the implications of their choices. It offers techniques that feed into slow forms of reasoning by, for example, providing information on the externalities and social effects of behaviour such as information on the harmful effects of plastics in oceans. It offers techniques that make use of fast reasoning by, for example, pointing to the choices made by others who are similarly situated. It also offers techniques that rely on automatic behaviour and inertia through framing default positions for choices that yield the socially desired result. We encounter these default techniques every day on the internet when we tick the ‘agree’ box without reading the small print.
 
The normative appeal of nudge
 
Quite apart from the background to nudges arising from research into cognition there has also been a powerful normative appeal to the idea that people will respond to signals or forms of guidance that do not involve commands or control. The normative appeal consists of two main elements. One reflects the view that reliance on the evolution of informal social norms to achieve social coordination is normatively ‘superior’ to other systems of social coordination because it is consistent with a small state and minimum government. In addition, there is a long tradition of viewing the regulatory world as generally ‘inferior’ to other forms of social coordination, not only to informal social norms but also to politics and the law. Adam Smith relegated regulation to the ‘police’ functions of government and Jeremy Bentham simply recognised the role of what he termed ‘indirect’ legislation.
 
The other strand of the normative appeal of nudges emphasises the normative superiority of decision taking by individuals who can internalise what is the right thing to do, who can act according to their own reasoning, and don’t need to be ordered what to do by the ‘nanny’ state. Adam Smith referred to the importance of ‘conscience’ that, in his assumptions about behaviour, linked the emotional drives of individuals to perceptions about how others view their actions.
 
Both classes of fear have been brought together in discussion of the alleged ‘crowding out’ effects caused by regulation. ‘Crowding out’ can apply either to displacing other forms of social coordination or in respect of diminishing personal responsibility. Nudges avoid any such crowding out effects because they leave the individual in charge of their own decision making and leave politics and the law out of the picture. A larger space is left for informal social norms.
 
These different elements underlying the normative appeal of nudges are sometimes combined under the label of ‘liberal paternalism’. Nudges are paternalist because they guide behaviour towards what is socially responsible as seen by those who hold authority in a particular context and setting. At the same time, they belong to the liberal tradition because they appear consistent with a small state.
 
The weaknesses of nudge
 
Empirical studies on the effectiveness of nudges have so far yielded contradictory findings. Experience with COVID suggests that the normative appeal of ‘nudges’ has outrun the evidence of when they are, and when they are not, effective as a guide to decision taking by individuals.
 
Simplification and licensing
 
First, in the COVID pandemic, regulation performed its traditional and essential role in the licensing process for vaccines. Despite the fears of anti-vaxxers, most people accepted the assurance of regulators that the benefits of the vaccines outweighed the risks. The traditional role of regulation in simplifying the environment for actors by providing assurances about health and safety was fulfilled. Even with licensing, unfounded claims were made for unauthorised products. It is not difficult to imagine the much greater confusion that would have arisen in the absence of a formal licensing process.
 
Disassociation and compartmentalisation
 
Secondly, despite abundant evidence that the COVID virus was highly infectious, mandatory lockdowns, closures, distancing and quarantining became necessary to limit the spread of the virus. People did not appear willing to take the risks of spreading infection into account in their own behaviour towards social contact with others. There was abundant evidence and information to encourage considered personal responses to infection rates, but, nudging in the form of disseminating information about infection rates did not seem to have been sufficient to influence behaviour in ways that would have limited the spread of the virus.
 
One way of characterising the gulf that emerged between voluntary personal behaviour and socially responsible behaviour is in terms of ‘disassociation’. Disassociation takes a variety of forms. At its most general it refers to when people are so pre-occupied with running their own lives that they become detached from their responsibilities towards others outside their immediate circle. From this perspective, mandatory instruments were needed to counteract the compartmentalisation of lives and behaviour in modern urban societies.
 
Dissensus and the disaggregated
 
Thirdly, in order to achieve sufficient uptake to contain the effects of the virus, officially certified proof of vaccination and COVID-free health has had to be required in a multitude of social settings such as travel. Despite individuals having strong personal incentives to pay attention to the advice of their political leaders and leading health sector professionals, people did not seem prepared to respond to their advice and get vaccinated. (Leaving out of the picture the erratic example set by President Trump).
 
The underlying issue is about the effectiveness of different tools of government in the face of social dissensus. Politics relies on the aggregation of opinion and sentiment. In contemporary pluralist and diversified societies, the socially aggregative role of politics seems to have weakened and the role of political leaders in speaking to ‘the people’ seems to have diminished. The advice of health professionals also does not seem to have been fully heeded, even if trusted more than that of politicians. In these circumstances, problem solving seems to rely on disaggregated techniques involving, in this case, programs of individually targeted vaccinations to different age groups according to their vulnerability.
 
Thus, in the case of COVID, nudges did not work either to strengthen slow reasoning or fast reasoning. The advice of political leaders or experts should have triggered short cut or ‘fast’ reasoning. Disseminating information about the infectiousness of COVID should have stimulated slow reasoning. Neither worked. In addition, informal social norms about what to do lacked consistency and influence.
 
Conclusions
 
It remains the case that governments and regulators can and should draw on a wide range of techniques for influencing behaviour and that soft forms of persuasion may often be as or more effective than coercive measures. But the circumstances when nudges can overcome the habits of people who lead compartmentalised lives or can overcome the difficulties of achieving an aggregated social response in pluralist societies, seems to have been greatly exaggerated. Putting this conclusion in another way, nudges can work where choices are already simplified, in contexts where people don’t have other significant distractions, and where they can be highly targeted. In addition, there has to be an absence of contradiction from informal social norms and predispositions. These pre-conditions limit the role of nudges. They also suggest when default positioning can take advantage of inertia.
 
We should also be careful about viewing the regulatory space as offering an ‘inferior’ form of social coordination. Each form of social coordination has its own strengths and weaknesses. In this case politics was not trusted, the law lacked relevant expertise and informal social norms gave weak and ambiguous signals. A regulatory response was needed, short of those draconian measures deployed in China and possible only in an autocratic state.