Elizabeth Robinson, Magenta Consulting
In recent years, behavioural nudges have been used to support college students to manage their loan debt, encourage consumers to adopt renewable energy sources, strengthen community policing (Social and Behavioral Sciences Team, 2016), and to promote many other goals deemed worthy and acceptable by society. Studies show that nudges in Western countries are largely supported by the public—which has implicitly been cited as a justification for their use. But what happens when nudges are used to support women’s access to gender-based violence (GBV) services in Afghanistan, or to promote women’s employment in the security sector in Jordan? In such cases, local cultural norms may conflict with Western concepts of welfare, dignity, and autonomy, and the public may not agree with the goals the nudges seek to achieve. Should the lack of public support make policymakers pause? Or can nudges still ethically be used as a tool for needed social change?
In an August 2020 article published in Behavioural Public Policy, Olivier Drouin summarizes evidence from a recent study showing that behavioural nudges targeting parents and children to improve children’s health outcomes are generally accepted by adults. Drouin’s findings add to a growing body of literature demonstrating that behavioural nudges are largely accepted by the public as tools of policy (Drouin, 2020).
Indeed, nudges are widely supported in Western countries, including by individuals from across the political spectrum and different demographic groups in the US (Sunstein, 2016), and generally in Europe. Approval of nudges has been shown to be high in the UK, Italy, Germany, and France, though with lower approval in Hungary and Denmark (Reisch and Sunstein, 2016). A recent study with participants from Australia, Brazil, Canada, China, Japan, Russia, South Africa, and South Korea also showed that nudges garner wide approval in these nations, as long as they support legitimate goals and align with the interests and values of the country. Japan was an outliner in this study, consistently showing lower approval rates for nudges than most other countries (Sunstein et. al., 2018).
This research into the acceptance of nudges is critical, given their increasingly wide-spread use in government policy. Whether the public supports the policies used by the state is key to the legitimacy of the state. That nudges are a relatively new tool, and may seek to influence the unconscious biases of the target audience, is further cause to explore public opinion.
However, there is far less evidence on public support of nudges in developing countries, where nudges and other social and behaviour change interventions are gaining traction as tools of not only government policy, but also for international development and social change.
The World Bank’s Mind, Behavior, and Development Unit focuses on “using the behavioral sciences to fight global poverty and reduce inequality,” with a portfolio spanning 65 countries that applies behavioural interventions to diverse issues such as women’s labour force participation in Jordan (Gauri et. al., 2019), tax compliance in Latvia (World Bank, 2019), and open defecation (Neal et. al., 2016). The UN has also begun to include behavioural insights and social and behavioural change communication (SBCC) in its programming, even setting up a Behavioural Insights Group through the UN Innovation Network. The intersection of behavioural science and international development has also sparked the emergence of specialised consulting firms, including MAGENTA, where this author works.
While research and case studies on the design and impact of nudges and other behavioural interventions are now quite widespread, there is far less evidence on the public acceptability of nudges in developing countries.
One example of this needed research was a study conducted by the international polling firm IPSOS in 2010, which surveyed 18,500 adults across 24 countries—including lesser-studied and developing nations such as India, Indonesia, Mexico, Saudi Arabia, and Turkey. The research found that public acceptability of nudges was generally high across all countries, with more support for government intervention among less wealthy nations—including India, Indonesia, and China—and more support for prohibitive legislation against certain behaviours among countries with a higher inequality of power, including Russia, Mexico, China, Indonesia and India (Branson et. al., 2012).
The IPSOS study appears somewhat unique; further—and more recent—research is urgently needed. In particular, there are two aspects of public acceptability of nudges that future research in developing countries ought to explore:
First, different nations likely have different expectations as to how much influence the government (or other official entity) should have on individual or family behaviour. In the US and many European countries, the public generally accepts that the government can and should have some role to play in the lives of citizens, in exchange for maintaining social order and providing public goods. Governments may also have a mandate to influence behaviours given that, as the World Bank points out:
“Governmental inaction does not necessarily leave space for individual freedom; rather, government inaction may amount to an indifference to the loss of freedom”.(World Bank, 2015).
However, in some countries—for example nations where tribal loyalties provide the main framework for local decision making, not formal state mechanisms—families and communities may recoil at the idea of politicians trying to influence their behaviour.
In addition, in many developing countries the state has far less legitimacy and may be less trusted, possibly creating inherent scepticism among citizens targeted by nudges. Indeed, the above-mentioned study that found lower levels of approval of nudges in Japan—and also citing diminished approval in Denmark and Hungry—suggested lower levels of trust in government as a potential explanation. As the authors of the study point out, in some countries, “if the government plans to do it, it is probably a bad idea” (Sunstein et. al., 2018). Nudges implemented in a country where citizens have less trust in the government may be less effective (though not necessarily any less effective than other policy tools used by the same government).
The second aspect of public acceptability of nudges that warrants further research and ethical consideration—and on which the rest of this post will focus—is whether it’s acceptable to implement a nudge that is not supported by citizens because the goals promoted by the nudge conflict with local cultural norms.
The existing literature on the public acceptability of nudges is fundamentally based on the premise that it’s important for us to know whether nudges are accepted by the public, because this may affect the decision whether or not to use them. Implementing nudges that are not accepted by a majority of the public may lead to (further) loss of trust in the state, or deterioration of legitimacy.
Perhaps more importantly, there is also a concern around whether it is ethically acceptable to pursue a goal that the majority of the public doesn’t support, and/or to do so without the public being explicitly aware that the state (or other entity) is aiming to change their behaviour (given that in some cases, nudges are designed to encourage us to make different choices automatically).
Such ethical issues have been addressed extensively in the nudge literature. Cass Sunstein argues that nudges are ethical if they promote “welfare, autonomy, and dignity” (Sunstein, 2015), and provides three principles to ensure nudges are designed ethically:
- Nudges should be transparent, and not misleading
- It should be as easy as possible to opt out of a nudge
- There should be evidence that the behaviour promoted by the nudge will improve the welfare of those being nudged (Sunstein and Thaler, 2009).
Scholars from University College Dublin have developed the FORGOOD framework, which includes the following seven ethical dimensions of nudges that should be considered (Figure 1).
To some degree, ethical concerns are also mitigated by the design and definition of nudges to begin with: by changing choice architecture or helping us overcoming cognitive biases, nudges make it easier for us to do what we want to do—to do the right thing—in the first place. Indeed, existing literature on the acceptability of nudges typically examines nudges for behaviours that we generally want to do, but often find it difficult to follow through on, such as donating to the Red Cross, off-setting carbon emissions (Sunstein et. al., 2018), or in the case of Drouin’s research, improving children’s health outcomes (Drouin, 2020).
Applying these ethical frameworks and guidelines in different cultures—that is, cultures different from the one in which such frameworks were developed—is not straightforward. Sunstein’s principles of welfare, autonomy and dignity may be defined differently by different cultures, and some societies may prioritise different values altogether—such as considering family honour to be more important than individual autonomy. If we try to determine if a behavioural policy ‘serves good and legitimate goals,’ as per the FORGOOD framework, from whose perspective do we answer this question—ours, as Western “experts,” or from the perspective of the communities where the policy is being implemented? And how does the principle of delegation apply if the entity implementing behavioural interventions are not policy-makers to begin with, but rather international development agencies?
Whether or not a nudge is ‘ethical’ may depend on which set of values is prioritised—which also raises the question of who has the right to make that decision. Such lines of inquiry may ultimately also lead us to question whether it’s ‘right’ to apply ethical frameworks developed in one culture to another—under the assumption that there is, in fact, a fundamentally shared set of human values—or whether localised versions are needed.
To take an example: a nudge to make it easier for women in Afghanistan to access gender based violence (GBV) services may certainly improve the welfare, autonomy and dignity of women, and would by all accounts promote women’s rights. However, such an intervention could easily be seen (including by some women) as compromising family honour by bringing what’s considered private family issues into the public sphere. Many Afghan men may disagree that such a nudge would ‘serve good and legitimate goals,’ and see the nudge as threatening their own freedom to choose what they see as best for their family.
To be clear, all human beings deserve the same human rights, and promoting human rights—including through access to GBV services in Afghanistan or other contexts—should not be conditional on local interpretations of who is entitled to which rights. That being said, nudges that conflict with local norms and lack public acceptance runs the risk of, at a minimum, being ineffective, and in the worst case, actively causing harm.
At the risk of stating the obvious, nudges need to be contextualised to local communities, and tailored to meet societies where they are in the process of social change, rather than imposing foreign ideals and values. But public acceptance can still be tricky.
If, for example, a nudge could dramatically reduce the likelihood of a child contracting malaria or other deadly disease, should questions of ethics and local acceptance still carry the same weight? Applying the implicit Western standard that nudges must be publicly acceptable may, in this case, come across as naïve and overly cautious.
It is tempting to justify the contrasting standards that may exist for public acceptance of nudges (and other aspects of development programming) between Western countries and developing countries in light of the increased urgency to find solutions in the latter, or other very real contextual factors. This is a slippery slope that, in the past, has led to harmful interventions in the Global South—such as young, untrained missionaries starting up medical NGOs—that would never have been permitted in the Global North.
As nudges and other behavioural tools are used more and more widely outside of Western contexts, there is a need to critically examine not only whether ethical guidelines, including around the need for public acceptance, developed in the West ought to be applied to other cultures, but also the justification in cases when they are not.
While public acceptance of nudges may seem like a challenging litmus test in some countries, the lack of such support is a good indication that behavioural and other interventions are trying to change too much, too fast, and need to be better contextualised in order to be effective. Creating demand for behaviour change—the want to want to change—is a first step we shouldn’t skip, in any country.