Nudging behaviour is ineffective, naïve and unethical

Interventions to ‘nudge’ people into desirable behaviours have become popular with policy-makers internationally.  In the UK, the Behavioural Insights Team – established under former premier David Cameron – are prolific nudgers, designing (amongst many others) inputs to reduce meat consumption, protect police officers from phishing, and reduce the gender pay gap.

Unlike earlier economic models of behaviour based on a theory that humans are ‘rational actors’, nudges are based on the contrary idea that much of the time people make irrational choices.  A nudge simply enhances the attractiveness of one option over another.  It could be a financial incentive – such as a discount for early payment of a fine, or a simple ploy such encouraging users of a works canteen to buy fruit by placing it at eye-level.

Gurus of the nudge – behavioural economists Richard Thaler and Cass Sunstein – see nudging as ‘libertarian paternalism’.  Rather than policy makers forcing particular decisions on the public, instead a gentle nudge can make the ‘right’ option easier to choose.  Sunstein argues that such approaches “do not force anyone to do anything and … maintain freedom of choice, but … have the potential to make people healthier, wealthier and happier”, and simply offer a little help to steer them in that direction.

But are nudges and other behavioural interventions as benign and as cost-effective in improving the public’s health and well-being as their supporters suggest? Our recent research casts a critical eye over two very different behaviour-shaping initiatives in the UK (cash for breastfeeding) and Australia (a debit card to encourage responsible spending).  For each, we asked the question: what does this intervention actually do?  In addition to its efficacy in achieving its stated aim, what are its broader physical, psychological, emotional and social effects in terms of capacities and limitations?

Breastfeeding during infancy has been linked to many positive health benefits during both childhood and adult life.  These include physical and cognitive development, protection from infections and lower incidence of obesity and diabetes in adults who were breastfed.  However, rates of breastfeeding vary widely, dependent upon social class, ethnicity and other cultural factors, and geographical location.  It’s lower among working class mothers, young mothers, less educated mothers and those living in economically-developed countries.

Public health efforts to address some of these disparities have included nudging mothers into breastfeeding their new-borns by offering cash or shopping vouchers as incentives.  Previous research has suggested that encouraging new mums to start and continue breastfeeding needs frequent contact with health professionals.  A shopping voucher nudge could prove far more cost-effective.

A recent controlled trial by Clare Relton and colleagues tested the efficacy of such attempts to nudges mothers into breastfeeding. Over 5000 new mothers from localities with low rates of breastfeeding were offered £40 shopping vouchers on five occasions between two days and six months after birth of a child if they had breastfed since the previous contact.  A control group received standard advice from midwives and health visitors.  Offering vouchers led to a modest increase in the mums breastfeeding their infants at least once (up from 32 to 38 per cent).  But they failed to encourage significantly more new mothers to breastfeed their baby for the first time, or to breastfeed exclusively – with no use of formula milk.

Our analysis suggests reasons for these findings. We looked at a range of research evidence, to map the broad cultural and material contexts surrounding new mothers. Mums are caught up in complex assemblages (networks of human and non-human interactions) comprising family, friends, relationships, health professionals and wider cultures, while living in circumstances that may be materially and financially challenging.  When a new born infant arrives, this adds new demands on time, priorities and family finances.

Not least among these is the physical requirement to provide frequent and dependable nourishment: literally a matter of life or death.  However, many factors can affect whether breastfeeding is possible.  Mothers may experience difficulty or pain, or have demands on their time from family or other responsibilities.  They may not know the benefits of breastfeeding, or their culture and family norms may not encourage breastfeeding or disapprove of breastfeeding in public.  Reaching for the formula milk may be the simplest solution.

None of the latter affects are countered by the offer of vouchers.  Vouchers cannot solve health problems, cultural issues or the pressures of family life.  The trial findings suggest that a financial bonus can nudge a few mothers (perhaps those without serious impediments) towards breastfeeding; but no effect on getting others to initiate breastfeeding or abandon bottle-feeding altogether.

We would suggest that offering a financial nudge may indeed have some unintended and potentially negative consequences for new mums.  Such incentives act most powerfully upon those struggling financially, while not addressing the societal and cultural forces which are already making life challenging for this group.  In addition, this benefit may be quickly withdrawn if painful breastfeeding, time demands or family/peer pressure prevent continued breastfeeding.  The intervention also acts to define gender roles: infant feeding becomes the sole duty of a female parent, potentially producing stress in interpersonal relations between male and female carers.

The other behavioural intervention we analysed was a cashless debit card (CDC) used in parts of rural Australia to pay welfare benefits.  Its aims were to promote socially-responsible behaviour, by restricting cash purchases of alcohol, illegal drugs and gambling.  The card trialled in the two rural areas in South and Western Australia, where many of the people drawing benefits were Indigenous Australians.  For recipients of disability, parenting, carer, unemployment or youth allowance benefits, 80 per cent of their weekly income was paid via the CDC, with the remainder in cash.

An Australian Government evaluation in 2018 found that 52 per cent of people ran out of money to buy food, 45 per cent had no money to buy schoolbooks or other goods for their children, and 35 per cent could not pay their bills.  Research by Klein and Razi revealed that users suffered financial hardship and increasing levels of domestic violence, and added to social divisions within these communities.

Again we analysed the broad contexts within which users interact with the CDC.  While the aim of the Card is to change individual behaviour, its use impacted on a far broader social and material assemblage that links welfare recipients, welfare payments, the goods and services that these payments purchase, governmental welfare agencies, and the political interactions between government, people on welfare and Indigenous communities.

The addition of the CDC into this complex mix augments and alters these interactions, to produce a novel range of (in)capacities in its users.  The card limits welfare recipients choices, both in terms what they can buy and where they can shop.  Purchasing via the card marks out and may stigmatise users as welfare recipients, while technical failures and users’ limited understanding of the card’s operation can produce both material hardship and embarrassment.  The CDC also interferes with customary/barter economies operating in these communities, further constraining access to food and other tradeable goods.

The research by Klein and Razi also revealed impacts of the CDC upon already fraught relations between settler and Indigenous communities in Australia.  It added to racist sentiments by non-indigenous residents who were not subjects of the trial, and engendered conflict between Indigenous leaders who accepted the trial and Indigenous community members who resented the imposition of the CDC.

Our analysis of these two interventions gives us great cause for concern about this behavioural approach to policy implementation. Offering financial incentives for breastfeeding seems to be founded on the naïve assumption that mothers who use formula milk are irresponsible, uneducated or lazy.  Analysing the broad contexts of breastfeeding suggests far more complexity lies behind the difficult choices facing new mothers. Many may not be able to breastfeed for health reasons, while for others family demands and cultural constraints may overwhelm any desire to breastfeed.

Similarly, the policy underpinning the CDC is founded upon the view that people receiving a welfare payment from the government in the trial sites have behavioural problems with alcohol, drugs and gambling. Yet most people using the CDC do not have such issues, and receive welfare payments because of unemployment, disability, parenting or caring responsibilities.  Many struggle with poverty, compounded by the difficulties of living in challenging outback terrains, in rural and remote locations where there are simply not enough jobs for all citizens.

Neither the offer of vouchers for breastfeeding nor the use of the cashless debit card to manage income fits the foundational premise of behavioural economics that ‘nudges’ encourage people to make choices that are in their best interests.  Both interventions can add to the distress or deprivation of already vulnerable people.  The CDC adds to users’ financial hardship, limits their capacity to engage in the cash economy or even to feed their children.  Women offered vouchers for breast feeding may find themselves coerced into difficult or painful breastfeeding or unsettling family relationships.

Our conclusions?  Based on these two case studies, the use of behavioural approaches to achieve government policy objectives is highly questionable ethically within democratic societies that hold freedom of choice as a principle.  They are sociologically naïve about the complex reasons behind why people do what they do.  And the kicker: evaluations indicate they are ineffective in achieving the policy objectives set by their authors.

About the authors: Nick J Fox (@socnewmat) is professor of sociology at the University of Huddersfield, UK.  His most recent book (with Pam Alldred) is Sociology and the New Materialism (Sage, 2017).  Elise Klein (@elisejklein)is lecturer in development studies in the School of Social and Political Science, University of Melbourne, Australia.