The power to nudge: can we democratise choice environments?
|Date:||22 February 2018|
Behavioural science shows how we systematically diverge from the ideal of rational choice familiar from economics textbooks. What does this imply for public policy? One conclusion is that public policy should shed unrealistic models of human agency. While information campaigns, for example, might work well for the idealised homo economicus, often enough they won’t budge real-life people. Proponents of nudge theories suggest a different approach to use people’s cognitive biases to their advantage. Public policy nudges do not change the options themselves nor the costs and benefits associated with them. Instead, by tweaking choice environments, they seek to nudge agents towards options that are good for them and/or good for society.
Nudging has proven popular with governments around the world, with the UK in many ways being a trailblazer. But alongside nudge enthusiasts, critics abound as well. The first chamber in the Netherlands voted in February 2018 to make organ donation the default. Unless you opt out, Dutch citizens are automatically signed up to be organ donors. Default switching is arguably the most famous of all nudges. Nonetheless, Dutch politicians were far from united, voting with just 38 to 36 votes in its favour.
So, why do people object to nudging? In my article The Power to Nudge in the American Political Science Review, I discuss one prominent objection. Various commentators worry that systematic nudging might allow the government to exercise problematic social control over its citizens. Such critics sometimes draw on neo-republican views of freedom: a central goal in institutional design should be to prevent power relations that allow one agent – individual or collective – to impose their will on others. And you can impose your will on others if you hold power over them that they in turn do not adequately control (or that is not adequately controlled in their name). Nudges, so the argument goes, furnish governments and policy-makers with such power. Nudges are insidious ways to influence citizens. Given their lack of transparency, they are hard to control individually and democratically.
In my article, I argue that worries around dominating power provide no principled objection to the nudge approach, at least not in sufficiently democratic contexts. What is more, caring about problematic power relations can sometimes also support an argument for more rather than less public policy nudging. Take the transparency argument first. Critics sometimes argue that the nudge approach faces a dilemma: not making nudges transparent seems objectionable; but make nudges transparent and they might just stop working. However, whether transparent nudges would work or not is an empirical question. The evidence we have at this stage, patchy as it might be, suggests that transparency in no way stops nudges from working. And, in any case, if transparency does make nudges slightly less effective, we might happily pay this price for greater transparency and democratic control.
Critics also worry that nudges inherently fall into a different category to traditional interventions. Compare nudges with a law that penalises driving without a seatbelt, for example. In the latter case, people understand what is going on and the government is open about its intent and method. Nudges in contrast seem more insidious and harder to track. However, notice how the situation changes if we move away from simple examples like seatbelt laws. Can we really say the same about complex pharmaceutical regulations, car safety regulations or the Common Agricultural Policy? Such regulations, and others like them, influence our choices every day. Yet they are hardly more transparent, let alone easier to understand, than nudges. The Dutch organ donor case is a good example of how policy-makers can meet worries about transparency. Not only is the law easy to understand, it also includes a provision that every permanent resident receives two letters informing them of the law and allowing them to opt out easily. Moreover, they receive a notification once they are officially signed up and retain the option to opt out any time thereafter. (Further provisions allow relatives to overrule organ donation, if they have reason to believe the deceased did not want to be an organ donor or had simply not read the letters.)
More generally, democratising the policy process for nudging does not strike me as any harder than doing so for non-behavioural regulations and policies. Arguably, the conditions seem particularly conducive. Nudging has been debated intensely, with widespread attention in politics, academia and the media. Moreover, its proponents typically favour an evidence-based approach to policy-making, regularly assessing how effective behavioural interventions are. Such assessments should also make it easier to control nudges democratically.
Consider now a stronger claim in defence of nudging. Whilst eminently fashionable, nudging is not entirely new. Marketing companies and PR departments have been pursuing similar strategies for quite some time. In their book Phishing for Phools, George Akerlof and Robert Shiller show how companies rely on our cognitive biases to elicit purchase decisions that are good for their bottom line but bad for us. For instance, most people have a preference for giving away less private data than they actually do, yet the default privacy settings on Google, Facebook and many apps on your phone do not reflect such preferences. Or think of long-term gym contracts that exploit our optimism bias. Recovering the day after New Year’s Eve, we might sanguinely sign a year-long contract, optimistic in the belief we will hit the gym three times a week. Yet only a few weeks after, we find ourselves locked into a disadvantageous contract as our resolve wanes (no surprise that such contracts are heavily advertised just after New Year). Environments within which we make decisions about what to eat and drink are rife with many more such examples.
‘Disadvantageous private nudges’ not only leave us worse off, they are also instances of problematic power relations. Companies often lack the incentives to make private nudging transparent – think of your privacy default setting – nor is their power to use such nudges systematically controlled by consumers. This is where public policy might help. For example, we might try to ‘counter-nudge’ disadvantageous private nudges or regulate companies in ways that prevent such nudges. We could, for example, impose stricter conditions on what kinds of privacy settings internet firms can use as defaults. Or we might adjust food environments in ways that will help people stick to healthier diets. Adjusting choice environments through public policy, with nudging being an important tool among others, could furnish us with more democratic control over our choice environments.
We typically think of democracy as a value for political institutions. But many of the reasons we care about democracy apply to other social relations too, as recent work on workplace democracy emphasises, for example. But if my arguments are correct, consumer markets should be arenas for democratisation as well, and we should welcome public policy nudging as one avenue for increased democratic control.