Governments around the world are talking about citizen-centric services. Public sector CIOs are attempting to understand user needs and behaviour, so that they can design, develop and deliver services citizens will use and benefit from. But how exactly do you understand user citizens’ needs and wants, how do you factor them in your technology design, how do you design an interface which users will adopt, use and benefit from.

Behavioural science could provide some of the answers. In a 2008 book, ‘Nudge’, University of Chicago economist Richard H. Thaler and Harvard Law School Professor Cass R. Sunstein talked about nudges or interventions that steer people in a particular direction while preserving freedom of choice. They were drawing on ideas first defined by Nobel prize winner, Daniel Kahneman and Amos Tversky. Daniel Kahneman summarised his life’s work in the book, ‘Thinking, fast and slow’. He proposed two systems of thinking: "System 1" is fast, instinctive and emotional; "System 2" is slower, more deliberative, and more logical.

What is behavioural science?

Behavioural science is fundamentally about cognitive biases associated with each type of thinking. From framing choices to people's tendency to replace a difficult question with one which is easy to answer, they proposed and verified through studies over decades that too much confidence in rational human judgement.

The way a question or choice is framed or presented can influence people’s decisions and answers in numerous subtle ways. In the public sector, this could mean that governments can gently nudging citizens toward certain choices.

Examples of the sort of critical insights which behavioural science provides are:

Availability/recency heuristic:  People rely on the most immediate examples that come to a given mind when evaluating a specific topic, concept, method or decision. This is because of the notion that if something can be recalled, it must be important, or at least more important than alternative solutions which are not as readily recalled.

Sunk cost fallacy: People tend to persist on a course of action once they have invested some effort into it to avoid ‘wasting’ resource that is unrecoverable.

Loss aversion: People prefer avoiding losses to making gain. Loss aversion implies that one who loses $100 will lose more satisfaction than another person will gain satisfaction from a $100 windfall.

Rising use in government

Starting from 2014, behavioural insight teams have been created within several governments to inform policy decisions and apply specific practical guidance to drive real desired outcomes. The first such unit to be created within government was the Behavioural Insights Team (BIT) in the British government. It was originally located within the Cabinet Office but was spun off in 2014 as a social purpose company, partly owned by the Cabinet Office, employees and Nesta. The Social and Behavioral Sciences Team (SBST) was established by President Barack Obama in 2015.

Meanwhile, the Australian government set up the Behavioural Economics Team of the Australian Government (BETA), a joint initiative of 17 agencies, hosted by the Department of Prime Minister and Cabinet. The European Nudging Network (TEN) was created in 2014 to ensure a scientifically and ethically responsible dissemination of applied behavioural insights throughout Europe and beyond. The Network is managed by The Center for Science, Society and Policy (ISSP) in a collaboration between ISSP, the Organisation for Economic Cooperation and Development (OECD) and HEC Paris.

Technology and behavioural science

Behavioural science and technology could be related in three ways, sometimes one leading to another, forming a virtuous cycle.

• Implementing Behavioural Science insights using technology
• Designing, implementing or increasing uptake of technology using behavioural insights
• Using technology to gather better, real-time and more accurate data for improved behavioural science insights

The Organisation for Economic Cooperation and Development (OECD) published a report called ‘Behavioural Insights and Public Policy: Lessons from Around the World‘ this year in March. It said that behavioural insights till now have primarily been used for policy implementation. However, there is great scope for using them for policy design.

The report emphasises that good or reliable data is key to applying behavioural insights. Also, that data does not equal evidence and statistical significance must be taken into account. The short and long term effects have to be continuously monitored, the work published for transparency and accountability and the costs to government measured.

We pick several examples from the OECD report of behavioural insights being applied in governments from Canada to Colombia, from Kenya to UK and add information from other sources. These examples were selected because of the involvement of technology in one or more of the three ways described above.

Application examples

Encouraging adults to stick with literacy and numeracy programmes(Country- UK; Area- Education)

In September 2014, the UK’s Department for Business, Innovation and Skills (BIS) established the Behavioural Insights Research Centre for Adult Skills and Knowledge (ASK), in collaboration with BIT. ASK was established with the aim of conducting research, running trials to produce evidence-based policy practical tools for employers, training providers and adult learners.

In one of its first trials, ASK tried to find if behavioural science insights could encourage adults with low maths and English skills to stick with literacy and numeracy programmes. Two thousand students were sent multiple text messages and prompts on behalf of the college, telling them that what the students are learning is important, that they can succeed, that practice matters and that they belong in college. They were also sent messages helping them plan attendance, and revise course materials. Learners in control classes received no text messages. Average attendance increased by 7% compared to control group, while dropout rate decreased by 36%. Several colleges in the UK are considering implementing a similar text message regime.

Roll-out of smart meters (Country- UK; Area- Energy)

The UK government has been using behavioural science insights in Smart Energy GB, the national campaign for smart meter rollout in every home and small business across Scotland, England and Wales by 2020.The government requires energy companies to install smart meters for their customers, and has set out rules around data access and privacy, security technical standards for the smart metering equipment and meeting the needs of vulnerable consumers. However, consumers are free to choose whether to install one.

A study published in July 2016, titled ‘A smart route to change’ summarised the lessons derived from theory, surveys and consumer focus groups to find the most effective interventions which would build consumer confidence in smart meters and motivate them to install one. Insights from the qualitative research from blogs by users and non-users included points like perceived ‘hassle’ is a major barrier to engagement and installation. Examples of findings from the focus groups included positive response to interventio
ns which brought greater clarity to energy usage and the impact of changes in behaviour. It was found that ideally interventions should tap into existing behaviours and avoid creating additional effort and that more personalised feedback and advice would help to maintain behaviour change.

Encouraging healthy behaviour among citizens (Country- Canada, Singapore; Area- Healthcare)

Nearly 1 in Canadian children and youth and 6 among 10 adults are overweight or obese. Public Health Agency of Canada (PHAC) wanted to see if it could use behavioural science to tackle this complex problem.

Canadians are amongst the highest users of loyalty points in the world. Using nudge theories, PHAC launched a mobile app called ‘Carrot rewards’ which awards users loyalty points (Aeroplan® Miles, Petro-Points™ and more) a for their chosen programme for learning about and adopting healthy behaviours. They are awarded for downloading the app, for referring friends and family, for taking healthy actions, such as taking a healthy heart quiz, learning about healthy eating choices, participating in physical activity or visiting a local YMCA.

It is an ongoing project. Partners include British Columbia Ministry of Health, Newfoundland and Labrador Government, the Heart and Stroke Foundation, Diabetes Canada, and YMCA Canada.

It was launched in British Columbia in 2016 and will be extended to interested provinces and territories over 5 years, till 2020. Success will be measured by data on acquisition of new users, levels of engagement, attrition/retention rates, demonstration of improved knowledge against a baseline and information from wearable devices (the programme has been designed so that it can be linked to wearable devices). Analytics is used to drive modifications and tailored offers.

In Singapore, the Health Promotion Board (HPB) is leveraging behavioural insights and wearable technology to influence lifestyle changes and nudge Singaporeans to integrate more physical activity into their daily lives. Consulting with the UK’s BIT and the Ministry of Manpower’s Behavioural Insights and Design Unit.

         For example, HPB used the concept of gamification where points are earned based on the number of steps taken in the National Steps Challenge™, integrating the use of wearable technology with a steps tracker and HPB’s Healthy 365 mobile application to incentivise participants to sustain behaviour change.
         
         
         

Preventing patients from stopping tuberculosis treatment using virtual observation(Country-Moldova; Area- Healthcare)  

Moldova is facing an increased incidence of multi-drug resistant tuberculosis. One of the major reasons is the low drug adherence rate, people stopping treatment after leaving the hospital. Patients are required to take the pill in the presence of a medical professional. This Directly Observed Treatment (DOT) is to ensure that infection is completely eliminated. However, the travelling to-and-fro from the hospital, to swallow a pill in front of a doctor, proved too high a cost.

The United Nations Development Programme (UNDP), in partnership with a local NGO, Act for Involvement and the UK BIS is testing the replacement of DOT with VOT or Virtually Observed Treatment. It would eliminate the need for the patient to travel to the hospital He/ she can do it in front of a computer and send a video message using an app. Four hundred randomly chosen patients have been placed under VOT, with a control group of 200 patients receiving DOT to judge whether VOT leads to better adherence rates and health outcomes.

Reduction in missed follow-up doctor appointments(Country- UK; Area- Healthcare)

In 2014-15 around 5.6 million (9% of the total) NHS outpatient appointments were missed in England. Missed first outpatient appointments cost the NHS up to £225 million in 2012 to 2013.

         Patients may miss their appointment because they forget. Tackling this by telephone or text message reminders substantially reduce missed appointments.
         
         
         

Research was conducted to test how different reminder messages affect missed appointment rates. It compared the existing standard reminder message in use at Barts NHS Trust with 3 new messages designed by the research team. The message stating specific costs to the NHS for each appointment missed was found to be the most effective, more than an appeal to empathy or social norms or a statement regarding costs in general without stating a number.

Increasing uptake of online renewal of vehicle license stickers(Country- Canada; Area- Government digital services)

The government of Ontario, the second largest province of Canada tried to use behavioural science to boost uptake on online renewal of vehicle license stickers. The license is renewed annually.

A randomly selected sample of over 620,000 vehicle owners were sent modified renewal notices. The concepts of ‘salience’ and ‘gain-loss framing’ were tested.

The black text of the subject line regarding renewal was embedded in a blue background to make it stand out on the exterior of an otherwise standard black-and-white ServiceOntario envelope. Secondly, to increase the salience of benefits from online renewal, in the subject line on the envelope, the wording “Instant and easy renewal online” (highlighting the ease) rather than “Renew online and receive a 10-day extension (not all consumers require a 10-day extension.)

Research has demonstrated that when people think about an immediate action, they often primarily focus on hassles or the mental effort required to perform the action instead of the benefits. To address this, the original messaging about the online renewal option printed on the back of the renewal form found inside the pressure-sealed envelope was changed to prompt consideration of the benefits before a person had time to think about any hassles or costs associated with undertaking an unfamiliar process. In the second intervention, people’s attention was directed to gains or positives associated with the online renewal. In the third intervention, the negatives, particularly the time cost, of not choosing to renew online.
         
         
         

The salience-gain and salience-loss letters had significant positive effects on online renewals and only the salience-gain letter had any effect on on-time renewals. The salience-gain letter prompted the best response. The increased use of online renewal service during the eight-week study saved the government approximately C$28,000 in transaction fees by reducing the number of in-person transactions in ServiceOntario centers. Annual savings of C$612,000 were projected if the best-performing salience-gain treatment condition is adopted permanently in Ontario, and this gain would be achieved at virtually zero cost to the government.

Encouraging businesses to regularly update central registry data (Country- Denmark; Area- Government digital services)

The Danish Business Authority maintains a register, where Danish businesses have to register their data and keep it up to date. The data in the register is publicly available and used by businesses in their dealings with each other.

It is very difficult to estimate the accuracy of the data because of the nature of the register. So, DBA created a nudge intervention to help businesses keep an eye on their own data. There were two separate behavioural barriers to overcome, a lack of attention (for updating information such as change in contact person or address) and a lack of understanding of the procedure for updating. The proposed solution was a pop-up on a homepage by DBA, virk.dk, where businesses go to submit forms for getting subsidies, filing VAT returns and many other activities. The page is visited by most businesses several times a year.

As soon, as businesses logged in on the page, the pop-up prompted them to accept or change the current information about their company in the database. During a 20-day testing period, the pop-up was shown 14,377 times. 52.5% of users confirmed their current company data, 41.6% pressed ‘Correction’ and only around 6% chose the option, ‘I am not responsible for business data’.  42% of people who picked ‘Correction’ did not make corrections. The reason for that was found in a later user survey to be anti-intuitive characteristics of the flow for changing. However, the pop-up itself was effective in getting the attention of busy businessmen who might not consider updating registry information as a priority task.

Increasing citizen engagement with government by sharing work updates on mobile app (Country- USA; Area- Government digital services)

The Boston Mayor’s Office of New Urban Mechanics (MONUM) wanted to find ways to use technology for improving civic engagement. Citizens were using an app called Citizens Connect to submit public service requests to the city government. Using the app, MONUM provided the citizens with images of work being done of requests they had made, for example, filling potholes, cleaning graffiti and fixing streetlights.

In the following months, number of public service requests submitted by citizens jumped by 19.6% and the requests were made in 9.3% more categories. Operational transparency had a positive impact on Boston residents’ trust in the city government, which increased engagement.

Tackling behavioural problems in digital cash transfer for food aid (Country- Kenya; Area- Social welfare)

Cash for Assets (CFA) is a joint World Food Programme (WFP)/Government of Kenya conditional cash transfer scheme that reaches food insecure households in seven arid and semi-arid counties in eastern and coastal Kenya, where recipients work on community assets to build resilience against drought.

When an electronic payment system was launched in 2010, recipients faced issues of overcharging, battery problems, unfair treatment, service unavailability due to network outages, unreliable service hours and forgetting PINs.

CGAP (the Consultative Group to Assist the Poor) supported WFP Kenya to identify measures to deal with these problems based on insights from behavioural science. CGAP and WFP conducted two mystery shopping experiments in August 2014 and August 2015 in Isiolo county in Kenya. It sought answers to the questions around merchant and user behaviour. For instance, in busy shops with long lines, will merchants lose patience with beneficiaries trying to use their cards, mishandling their personal information and refraining from explaining how the system works? Or would there be a “lottery effect” with beneficiaries, where they treat the electronic payments less cost-consciously than “earned” money?

The study showed indications of positive effects on shoppers’ price sensitivity by mandating itemized receipts from POS devices. Merchants charged varying prices for the same item across customers. However, most consumers were not negotiating on prices, contrary to usual shopping behaviour in Kenyan markets. This could be due to the “lottery effect”. To address this, the introduction of mandatory itemized receipts was suggested to strengthen beneficiaries’ capacity to protect themselves from additional charges and put merchants on alert that their pricing is being monitored.
         
         
         

Several merchants did not fully protect PIN information of beneficiaries, entering the information themselves or asking beneficiaries to leave their PIN and/or card with them. A possible remedy could be to integrate mystery shopping on a permanent basis into the programme. An award system could be created for merchants that comply with these good practices where mystery shoppers provide them with “good practice” stickers or certificates when they refuse to sell non-permitted items or make consumers enter their PIN in the POS device on their own.

Improving understanding of telecom service conditions (Country- Colombia; Area- Telecom services)

Comisión de Regulación de Comunicaciones (CRC) or the Commission for Regulation of Communications in Colombia found that users of mobile telephony and Internet services had very poor information about consumer rights and obligations. They lacked even superficial legal knowledge while signing post-paid contracts.

Working with Konrad Lorentz University, CRC conducted a number of experiments on a sample of 851 mobile users, across different socio-economic strata, of both genders, between the ages of 18 and 65 with pre-paid and post-paid plans from different operators.

Principles of bounded rationality (Rationality of individuals is limited by the information they have, the cognitive limitations of their minds, and the finite amount of time they have to make a decision) were applied. To simplify the contracts, all information relevant to the user was presented in a clear and organised manner, such that it could be read in 12 minutes.

Researchers studying the effect of the simplified contract among sample users found that reading time is related to perceived risk and invested money, oral communication was often preferred and using introductory titles allowed for easy contextualisation of topics. Level of education was not a significant factor in understanding. Consistent users expressed confidence in the operator based on past experience. And when users believed they had no choice but to accept the conditions laid down by the operators, the ‘helplessness’ reduced their motivation to read.
         
         
         

CRC concluded that the success of simplifying contracts would depend on a combination of consumer psychology, behavioural science and visual communication techniques.  

Context, ethics

The OECD report stresses that results must be validated through replication to ensure that observed results are correct in the same context and setting and also in different ones, where governments might be thinking of applying them. Context is crucial.

Government actions will be necessarily targeted at a group and conclusions which might be statistically valid for a population might be difficult to apply for every single individual within the same population group. Technology (AI applied through mobile applications based on data from wearables for example) can help in making the interventions more targeted and personalised.

There could be ethical concerns in attempting to ‘modify’ behaviour. Thaler and Sunstein dealt with this by talking about ‘Choice architecture’ which describes the way in which decisions are influenced by how the choices are presented. People can be "nudged" by arranging the choice architecture in a certain way without taking away the individual's freedom of choice.

Transparency about the interventions could also go a long way in maintaining trust. Ultimately the aim is not to intercede and decide on behalf of citizens but to enable them to take the ‘best’ possible decisions, ‘best’ being decided by themselves.

Technology can help implement the interventions or the interventions can increase adoption of technology. These examples demonstrate repeatedly how simple and often cost-free interventions, derived from behavioural science research can be applied to help people make better decisions, without restricting freedom. It could be as simple as a well-placed pop-up or a change the background colour in which a message is embedded.