Welcome to this edition of Human Capital Matters (HCM)—the digest for leaders and practitioners with an interest in human capital and organisational capability. Human Capital Matters seeks to provide Australian Public Service leaders and practitioners with easy access to the issues of contemporary importance in public and private sector human capital and organisational capability. It has been designed to provide interested readers with a guide to the national and international ideas that are shaping human capital thinking and practice. The inclusion of articles is aimed at stimulating creative and innovative thinking and does not in any way imply that the Australian Public Service Commission endorses service providers or policies. It is intended that the articles are accessible for the general reader, do not require subscriptions to specific sites and, where possible and appropriate, editions of HCM have been reviewed by topic specialists to provide range and currency on topical issues.
Additional hyperlinks/references for those with librarian support or access to specific, user-pays sites may also be provided.
Thank you to those who took the time to provide feedback on earlier editions of Human Capital Matters. Comments, suggestions or questions regarding this publication are always welcome and should be addressed to: humancapitalmatters [at] apsc.gov.au. Readers can also subscribe to the mailing list through this email address.
This edition looks at 'nudging' decision-making to affect behavioural changes and the role it might play in developing public policy.
A traditional approach to behaviour change is education. We have long thought that providing timely, accurate and helpful information will help people make better decisions. While information can lead to behaviour change, it is only one way to influence decision-making. It works well with some of the people, some of the time.
Human decision-making is not based entirely on rational thinking. Overt and unconscious biases, variations in cognitive abilities and multiple sources of information and motivation (eg social norms, altruism, fear of loss, incentives and sanctions) also influence our decisions and consequently our behaviour.
Behaviour is often automatic and influenced by variables beyond conscious recognition. Our reactions and consequently our behaviour are influenced by context, social cues and our own predisposition.
It is from this understanding of human behaviour that the concepts of nudging and nudges emerged. Thaler and Sunstein defined nudges as:
... any aspect of the choice architecture that alters people's behaviour in a predictable way without forbidding any options or significantly changing their economic incentives. To count as a ... nudge, the intervention must be easy and cheap to avoid. Nudges are not mandates. Putting fruit at eye level counts as a nudge. Banning junk food does not1.
The principles of nudging recognise the human tendency to take shortcuts, to act automatically. Hence if an option is easy, attractive, social and timely2 people will tend to favour that choice. This is what nudging is about: making it easier and more attractive to make better choices.
Nudging is not about coercion. Incentives and sanctions will influence choices people make however. Nudging recognises people react to options and are more likely to choose options that are easy, attractive, social and timely (EAST). Nudging works with these insights into behaviour by using the EAST design principles to influence people's choices and consequent decisions.
The potential for ethical dilemmas about the use of nudge techniques has not gone unremarked3, . Regardless, humans can and will be influenced and there is potential to influence them in ways that serve the greater good 'while also respecting freedom of choice' (Thaler and Sunstein, 2003 as reported in Vlaev, 2016). Transparency about what and why something is being done (public defensibility) and rigorous testing of the intended nudge before it is embedded in policy are critical to addressing ethical concerns about nudge techniques.
Influencing people's choices is thought to have potential for the development of policy. Vlaev (2016) suggests '(p)olicies that change the context or "nudge" people in particular directions have captured the imagination of policy makers at the same time that the limitations of traditional approaches have become apparent ...' (p.551)
- The first article provides public administration practitioners with an accessible summary of behavioural economics and its potential application to policy.
- The second article is from the Public Sector Innovation Toolkit. Although originally written in 2014 it provides a summary and links to useful and current sites. Of particular interest is the link to the NSW Behavioural Insights Team (BIT) and community of practice which provides a link (hyperlinked separately) to the BIT Annual Report for 2015-16, published in September 2016. This report provides examples of how nudges have been used effectively across the public sector.
- The third article is a media summary of PM&C's behavioural economics team.
- The fourth article is more detail about the Behavioural Economics Team of the Australian Government (BETA) and provides links and references for further reading.
- The fifth article is another one from the media providing a summary and important links to a recent publication providing nudge guidance from the US Government's Nudge Unit.
Vlaev, I., King, D., Dolan, P. & Darzi, A. The Theory and Practice of "Nudging": Changing Health Behaviours. Public Administration Review, Vol 76 (4), July/August 2016
Vlaev notes that not everyone shared the enthusiasm for integrating nudging into public policy. This reluctance was based on a lack of evidence and a framework to facilitate policy making. These gaps were addressed by the UK Government's Behavioural Insight Team and resulted in the 'Mindspace' framework. The team had been tasked to deliver innovative ways of encouraging, enabling and supporting people to make better choices.
Vlaev's article describes and builds on existing academic scrutiny of Mindspace to discuss behaviour change interventions, specifically in the domain of health. He starts with the Mindspace explanation of how most people behave or act across various situations.
People's brains have two distinct systems for cognitive processing. One is automatic, effortless and unconscious. This system is affective, driven by emotions and habit and is evident in one's immediate reaction. The second system is that part of cognitive processing that is more effortful, controlled, conscious and rational. This system underlies considered, thoughtful responses. There is neurological evidence that both systems exist.
Traditional education and training approaches to behaviour change use this two system model of cognitive functioning. So too does the nudge model of behaviour change. The nudge approach proposes however that the first system, automatic reactions, can be utilised to create behaviour change rather than intervening only through the second system.
Further detail about the theoretical underpinnings of the Mindspace approach to behaviour change is given in the paper. In summary, Mindspace is an acronym for the framework that attaches behavioural responses to both brain and psychological processes; to give an example for 'M' and 'I:
|Mindspace technique||Behaviour||Brain system||Psychological process|
|Messenger||we are heavily influenced by who communicates information to us||Impulsive||Attraction, trusting|
|Incentives||Our responses to incentives are shaped by predictable mental shortcuts such as strongly avoiding losses||Impulsive||Greed, fear|
Vlaev's discussion would resonate with most policy designers:
Traditional ways of changing behaviour, such as legislation, regulation, and incentives, can be very effective. Behavioural economics does not attempt to replace these methods. Rather, it extends and enhances them, adding new dimensions that reflect fundamental but often neglected influences on behaviour. (p.556)
He concludes that 'much is unknown about the ... nudge approach in general. There remains uncertainty over how long the nudge effects last and how well they work in different segments of the population'. It is likely that different aspects of Mindspace work better for different interventions: some have short term effects and some induce long-standing change. He recommends that nudge techniques require rigorous evaluation before claims of effectiveness can inform evidence-based policy.
Public Sector Innovation (13 June, 2014) Nudging policy makers—an overview of Behavioural Exchange 2014. Australian Government, Canberra.
This summary of proceedings from the first international behavioural insights conference held in Sydney in 2014 provides a brief description of nudges, reasons for nudging, how to nudge and limits on nudging. It also provides a list of further readings.
Of interest from this article are the links to other sites. For example, the NSW Government's Department of Premier and Cabinet maintains a website on behavioural insights for a community of practice.
In August 2016 the department released a 12 page 'Behavioural Approaches to Increasing Workforce Diversity' document, complete with references, which refers to the effect of unconscious bias in attracting, selecting and promoting employees. It referred particularly to Indigenous peoples and women. The document provides possible behavioural interventions to address bias, adding the caveat 'BIU (Behavioural Insights Unit) recommends that where there is a lack of empirical evidence or consensus on the effectiveness of an intervention, the approach should be trialled before rolling it out—ideally using a rigorous methodology such as a randomised controlled trial' (p.6)
An example of a nudge-style intervention to reduce possible gender bias is given as follows:
|Potential intervention||Rationale/background||Biases targeted||Stage of recruitment process|
|Assess more than one CV at a time and, if possible, evaluate each section of an application separately||Joint evaluation of job candidates (looking at more than one CV at a time, side by side) has been shown to decrease gender biases.5 Evaluation of CVs should be conducted in sections so that positive/negative effects from one section do not spill over into the evaluation of another section||Stereotyping
(each of these are explained, with examples, more fully in the document)
The BIT annual report for 2015-2016 is also available on line. The report provides summaries of work over 12 months in areas such as health and wellbeing, education and skills, and reducing fraud, error and debt across projects in the UK, Sydney, New York and Singapore.
Listed among the achievements for the BIT are:
- Health. Informing doctors that they are prescribing more antibiotics relative to 80 percent of their peers reduced the number of unnecessary prescriptions by 3.3 per cent (more than 73,000 prescriptions). This was reported as helping to address what the Chief Medical Officer identified as the greatest medical threat of our age.
- Education. The two year 'Alert' trial showed that regular text messages designed to encourage learners to keep going increased pass rates by 12 percent.
- Taxation. In a group of trials to assess the impact of different letters to help prevent tax debts and fines it was found that SMS messages to those who had been late in previous tax years increased payment rates by 50 percent.
Stephen Easton. (23 November, 2015) BETA testing: new behavioural economics team in PM&C. The Mandarin. [Accessed 15 September, 2016].
The departments involved in the Behavioural Economics Team of the Australian Government (BETA), 'nudge' unit within PM&C include the ATO, PM&C, and DHS. The unit is led by Professor Michael Hiscox (on leave from Clarence Dillon Professor of International Affairs at Harvard University and member of the Behavioural Insights Group at Harvard's Centre for Public Leadership).
Senator Scott Ryan introduced BETA in late 2015 with advice that the ATO had already been using behavioural insights. It used tailored text messages to taxpayers' phones to assist them in avoiding debt. The messages provided advice based on social norms (typical ways in which most people behave). This led to an increase in timely tax payments.
Ryan went on to say that such tools, whilst available, 'do not, on their own, automatically provide the political or public justification ... to change policy or to seek to alter human behaviour.' Tools based on behavioural insights do not negate the need to develop policy 'based on consultation, communication and research, and planning for effective implementation'.
Behavioural Economics Team. Behavioural Economics. Department of Prime Minister and Cabinet [accessed 15 September, 2016].
The Behavioural Economics Team (BETA) started work in PM&C in February 2016. Its goal is to provide simpler, clearer and faster public services. BETA's mission is 'to build behavioural economics capability across the public service and drive its use in policy design by testing what works, where and in what context.' All this is aimed at helping people make better decisions.
BETA defines behavioural economics as the application of psychological insights to explain and predict how people make decisions:
Experience has shown that inexpensive improvements based on a better understanding of human behaviour can increase efficiency within the public service and help people put their good intentions into action. Initiatives like plain packaging of cigarettes, mysuper and pre-filled tax forms were designed with real human behaviour in mind
To address the issue of '... testing what works, where and in what context' BETA advocates the use of randomised control trials (RCTs).
BETA's guide to developing behavioural interventions for randomised controlled trials was published on 2 September, 2016. It provides nine guiding questions in two parts: Initial discovery questions and diagnosis questions.
The four discovery questions are: what is the outcome of interest; can we accurately, directly measure the outcome using existing data; can we deliver standardised interventions to a reasonably large randomised population; and, is an intervention in this space feasible?
The five diagnosis questions are: How can we better understand the behaviour; what behaviour is leading to the outcome; what is our theory, step-by-step, of the current behaviour; what interventions might influence the behaviour; and, what is our theory, step-by-step, of how and why that intervention will change the behaviour?
There is a detailed guide for each question.
David Donaldson (20 September, 2016). Four public sector uses for behavioural insights from Obama's nudge unit. The Mandarin [accessed 21 September, 2016]
This article is based on the 2015 implementation guide from the US government nudge unit. The guide identifies that behavioural insights can be used by government agencies to drive better outcomes for citizens. It is organised around four aspects of policy 'where research and practice show that behavioural factors play an especially strong role in program outcomes': determining access to programs, presenting information to the public, structuring choices within programs and designing incentives.
The guide recommends how agencies might make changes around each of these four aspects of policy implementation:
- Determining access to programs. Based on the insight that 'small barriers to program access can have large impacts on participation and outcomes' suggested implementation techniques to improve access include: simplifying forms, autofilling forms and automatically enrolling eligible individuals.
- Presenting information to the public. This is based on the insight that 'how individuals understand and respond to information depends on its presentation'. The guide recommends that information should be presented in a meaningful manner for the intended audience. This coincides with BETA's advice to make choice architecture 'attractive' and easy to understand. Bolding important information, using colour and images and personalised messages are all thought to be effective in influencing decisions.
- Structuring choices. The central insight at work here is that 'complex or difficult choices in programs can lead individuals to choose inconsistently'. Research demonstrates that individuals can have difficulty choosing, and choosing consistently, when choices involve numerous alternatives, vary along multiple or complex dimensions, involve assessments of probability or risk, or have a substantial time dimension (i.e., consequences long into the future). Examples of possible interventions include reducing the complexity of choices and setting defaults to the option likely to be the best for most people.
- Designing incentives. People are more likely to respond to an incentive that is framed in terms of potential loss rather than gain. They also respond to non-financial incentives. Immediate incentives are likely to be more effective than ones in which any reward is delayed. A pertinent quote from the guide is:
In cases where the goal of an incentive is to encourage a particular behavior, agencies should ensure the incentive is salient to individuals. Incentives may be more salient if they are provided in isolation, rather than as part of a larger payment such as an income tax refund. Incentives may be less salient if they are embedded in otherwise complicated programs or schedules, such as the tax code. Simple reminders can be an effective way to keep incentives salient.
As well as additional recommendations based on behavioural insights, the guide provides an exhaustive reference list for review.
Additional links to interesting sites:
From the Corporate Research Council: Behavioural Insights Applied to Policy: European Report 2016 [accessed 30 Sep 16]
This report focuses on developments across Europe and provides a state of the art view of the contribution of behavioural insights to policy making, while also putting forward an analysis of institutional developments. It showcases examples of behavioural interventions in a range of policy areas, such as employment, consumer policy, health, taxation, environment or transport. The report provides reviews for 200 behavioural policy initiatives across Europe.
Selinger, E., & Whyte, K. Is There a Right Way to Nudge? The Practice and Ethics of Choice Architecture. Sociology Compass 5/10 (2011): 923-935 [accessed 22 September, 2016]
Despite the fact this is 'old' in the world of academic publications, this is an informative and easily read paper. Selinger and Whyte outline the ethical and political objections as well as the conceptual problems and practical difficulties of nudging.
4 Selinger, E., & Whyte, K. Is There a Right Way to Nudge? The Practice and Ethics of Choice Architecture. Sociology Compass 5/10 (2011): 923-935 [accessed 22 September, 2016]
5 Bohnet, I., Van Geen, A. & Bazerman, M.H. (2012). When performance trumps gender bias: joint versus separate evaluation [online]. Working Paper No 12-083. Boston, MA: Harvard Business School. Available at www.hbs.edu/ faculty/Publication%20Files/12-083.pdf [Accessed 02 February 2016].