The governments of the Gulf Cooperation Council1 (GCC) countries are implementing ambitious national transformation programs that can succeed only if they reach key social, economic, and environmental objectives. Traditional policymaking involving incentives, rules, and regulations alone will not be enough to achieve many of these objectives. This is because people have “cognitive biases,” natural tendencies to resist change, even against their best interests. To counter such biases and improve the efficiency of new policies, many countries are integrating behavioral insights into the policymaking process.
Integrating behavioral insights into policy design complements conventional policy levers by providing a more realistic understanding of human behavior. Behavioral science combines insights from economics, neurology, psychology, and sociology to create a range of tools such as framed communication messages, peer comparisons, modified default options, role models, games, competitions, etc. Together, these encourage people to choose desired actions. Such behavioral interventions have proven highly effective to achieve public-sector priorities, particularly environmental sustainability, healthy lifestyles, and civic engagement. They are relatively inexpensive and easy to adjust for optimal results. However, they should be conducted carefully because they are not automatically replicable in different contexts. Policymakers should choose the right tools and apply them transparently, focusing on people’s welfare without restricting their freedom, in order to guarantee the desired outcomes.
To implement this approach, each GCC government needs to initially establish a centralized behavioral insights unit to gather data and insights, choose the right behavioral tools to apply in behavioral interventions, and optimize them. This will facilitate the transfer of knowledge, ensure better coordination between departments, and save costs. The unit will require a clear mandate and political backing. It should begin developing capabilities by leveraging skills from leading international institutions and seconded experts, then establish credibility by scoring quick wins. Finally, it should build internal and external partnerships to scale up and commission broader interventions.
Behavioral science can provide policymakers with a better understanding of why people behave in these ways and help them design tools to change these behaviors, thereby enriching the economic models in place.
Behavioral science can help GCC governments reach many of the social, economic, and environmental objectives outlined in their various national transformation plans, such as Abu Dhabi Economic Vision 2030, New Kuwait 2035 Vision, Saudi Vision 2030, and UAE Vision 2021. Policymakers have the task of introducing new policies to achieve these ambitious objectives, but are often frustrated by people’s inability or refusal to buy in. Behavioral science can provide policymakers with a better understanding of why people behave in these ways and help them design tools to change these behaviors, thereby enriching the economic models in place.
Conventional policy levers used by governments have not always succeeded in altering behaviors. This is because the design of these levers is often derived from standard economic models and design policies that use educated guesses about how people respond. This approach assumes that individuals make rational choices leading to the best possible outcome, irrespective of the context. In reality, however, people often behave irrationally: They resist saving money, they indulge in unhealthy habits like smoking or eating fast food, and they procrastinate when it comes to meeting deadlines — even when they are aware of the negative consequences of their behavior. These deviations from rational behavior are due to “cognitive biases.” For example, individuals have a tendency, known as “present bias,” to perceive things as less valuable if they are farther away in time and will therefore seek immediate gratification instead. This explains why someone might incur credit card debt to buy something immediately rather than save up money to make the purchase. “Negativity bias” is a tendency to let negative experiences influence new decisions and prevent change; for example, remembering how their previous diet didn’t work could hold people back from starting a new one. Interviewed for this study, Dr. Susan Mayer, co-founder of the Behavioral Insights and Parenting Lab at the University of Chicago, explained that “Cognitive biases exist everywhere but they can vary in degree. They are distinct from problems related to lack of access to information, lack of means, personal beliefs, or structural enablers.”
Cognitive biases impair the effectiveness of new policies, prevent change, and exacerbate existing problems, such as the prevalence of non-communicable diseases or high energy costs. Conventional policymaking does not automatically account for individuals’ cognitive biases, which results in undesired behaviors. For example, simply conducting awareness programs to reduce the incidence of diabetes might not be enough to reduce a country’s obesity rate. This is partly because educating individuals about the outcomes of diabetes only, without following up with behavioral interventions that prevent people from relapsing into unhealthy eating habits, will not effectively trigger the desired change.
The behavioral science approach diagnoses cognitive biases through incorporating a broad range of insights from economics, psychology, and sociology. Behavioral science then uses these personal insights (beliefs, attitudes, emotional behaviors) to design interventions that shift people’s behaviors. Behavioral interventions may take the form of “nudges,” tools that influence people’s decisions without imposing restrictions or altering their incentives, thus preserving their freedom of choice. We interviewed Dr. Michelle Harrison, CEO of the WPP Government and Public Sector Practice, who explained that: “It’s about understanding their beliefs, their attitudes, and their emotional behaviors in order to create a story and message that shifts behavior in a way that feels natural to them. These insights are the foundation for every part of the execution plan, including measurement and optimization.”
Governments have traditionally been anchored to the scientific, academic approach when designing policy, but they are gradually learning from successful private-sector marketing techniques, according to Dr. Harrison. She noted that adoption over the past 10 years has progressed slowly and steadily, but the pace has been accelerating recently as more policymakers have begun recognizing the effectiveness of the solutions. In 2010, the U.K. government established the Behavioural Insights Team (BIT), which was the first entity to have successfully applied behavioral insights to policymaking.2 Since then, other countries including Australia, China, France, Kenya, Singapore, and the U.S. have established similar entities. What seems to be clear is that many countries are looking for global leading practices that can be implemented locally.
Behavioral tools should be transparent and not misleading, easy to opt out of, and should aim to improve people’s welfare. The most common examples include:
Behavioral science insights complement standard policymaking by delivering a more realistic view of human behavior and allowing for fine-tuning and adaptation throughout the process. They then serve to design behavioral interventions by using different behavioral tools to achieve the desired outcome. Such experimentation is not possible with standard policies alone. Dr. Mayer confirmed this, stating that “behavioral intervention tools cannot substitute conventional policy levers entirely. For example, encouraging parents to read to their children will fail if the appropriate structures are not in place, that is, if parents cannot afford to buy books or access a public library.”
Therefore, if governments want behavioral interventions to succeed, they should put in place the supporting infrastructure and adequate standard policy levers, in addition to implementing the right behavioral tool or combination of tools (see Exhibit 1). For instance, a policy to increase recycling rates will not succeed if municipalities do not provide the right waste management infrastructure.
Communication is a key lever of behavioral policy delivery. The role of communication is about more than education and raising awareness, it is a motivator of change. Communication can address people’s emotions, values, or perceptions of social norms to make the desired change acceptable. It can alter their motivations, biases, and habits.
For communication to achieve this, the choice of message, messenger, and channels involved should be incorporated with saliency and efficiency. People rarely behave the way they actually believe, or say, they do. As such, central questions around why people behave the way they do need to be uncovered. In addition, the scale and speed at which people are exposed to multiple forms of content on a daily basis are increasing rapidly. The source of information or the messenger — and people’s reactions to it — is also crucial in influencing people to change their behavior. For example, people are more likely to act on information if an expert or a friendly person delivers it.3 The right form of data collection via the relevant research methods will help determine the appropriate communication tools for reaching the right people at the right time in the right way, tracking progress, and evaluating the overall outcome.
Selecting the right combination of tools is key to maximizing the impact of behavioral interventions. Change4Life, a Public Health England campaign aimed at helping prevent childhood obesity in the U.K., effectively illustrates this. Launched in 2009, the campaign leveraged communication tools, gamification features, and partnerships to reach its objective. Change4Life introduced 10-minute “shake ups” — activities featuring Disney cartoon characters, which increased its appeal to the target audience. A mass media campaign created awareness about the initiative. Locally, participants received personalized activity plans and alerts about activities near them. As a result, the campaign increased the number of “non-sporty” children choosing to play sports on a weekly basis in schools by 166 percent, and encouraged schools to introduce more sports activities.4 The campaign is still ongoing and providing updated content to four million participating families at the time of publishing this report.
Kantar TNS, a WPP company, recently conducted a survey in Saudi Arabia and the United Arab Emirates (UAE) to illustrate how the “decoy effect,” a tool used in behavioral interventions, can nudge people to change their preference between two existing options. This was achieved by introducing a third option that is “asymmetrically dominated,” meaning it is similar but clearly inferior to one of the two existing options. The decoy thereby encourages more people to pick an existing option. Respondents in Saudi Arabia and the UAE were divided into two groups. The first group was given the choice between an expensive premium gym membership with full-day access to the facility, including all classes and the swimming pool, and a pool. More than half the respondents in the first group selected the cheaper option.
The purpose of the second group was to test whether respondents would shift their preference away from the cheaper option when presented with the third asymmetrically dominated option, the decoy. This decoy option was in between the two initial options — it was slightly less expensive than the premium option although with fewer benefits, but it was more expensive than the cheap one. The experiment with the decoy succeeded in getting more people to select the premium gym option (see Exhibit 2).
However, looking at the results of men and women separately revealed a variation in the magnitude and relative difference of choices in the two countries. In Saudi Arabia, men were much more likely to choose the cheaper option when the decoy option was not offered. However, approximately the same proportion of men and women chose the cheap membership when the decoy option was included. By contrast, in the UAE, there was a marked difference between the genders, with the decoy having a strong effect on women’s preference (see Exhibit 3), more than halving the number who selected the cheap membership. Kantar TNS recommends further research to gain more meaningful insight into this specific anomaly and determine how it may affect using such data in a full-blown behavioral intervention. Behavioral insights need to be analyzed to understand such anomalies if they are to be applied in policymaking.
By applying behavioral interventions in the right places, GCC countries can effectively achieve the key objectives outlined in their national transformation plans. Strategy& analyzed transformation programs in GCC states and found that behavioral interventions could possibly be applied to 12 objectives in these plans. Then, Strategy& prioritized these objectives by weighing the proportion of behavioral interventions against structural levers used to achieve them, as well as the success of behavioral interventions implemented globally to achieve similar objectives (see Exhibit 4).
To illustrate the correct use of behavioral interventions and their advantages, this report will focus on the three national objectives with highest potential for behavioral intervention: Achieve environmental sustainability, Maintain a healthy lifestyle, and Increase civic engagement. These objectives are illustrated through four initiatives: optimizing water and electricity consumption, encouraging recycling, reducing obesity, and enforcing tax compliance.
Although structural levers are important to reduce water and electricity consumption in GCC countries, they should be complemented by behavioral intervention tools that address excessive household usage. Structural levers include reevaluating water and energy bills and increasing access to smart meters. What is vital is to change behaviors because the way in which people consume water and electricity exacerbates factors such as intensive irrigation and energy subsidies. When educated about the risks of water shortage or high energy costs and told to reduce their consumption, people may still not change their behaviors. The hidden reason behind this is the “status quo bias” (or “system justification”) whereby people prefer existing conditions (consuming as much water or electricity as they like) to alternatives (saving water or electricity), even if this affects their individual or collective well-being. They could also be unconsciously choosing to ignore the obvious negative effects of their behavior (the risk of power outages at peak consumption times and the detrimental environmental effects of desalinating water or burning fossil fuels) by what is known as “ostrich effect bias.”
The populations of GCC countries consume more water than the world average, even though the region has a water shortage.
The populations of GCC countries consume more water than the world average, even though the region has a water shortage. Bathrooms are the number one source of water consumption in households due to wasteful behaviors like taking long showers, leaving the water on when brushing one’s teeth, or using high-capacity toilet bowls and flushing repeatedly. According to a Strategy& report on achieving water sustainability in the GCC, Saudi Arabia and the UAE — the region’s largest consumers of water — consume between 10 and 39 times more than the amount of renewable water available to them.5 At the same time, the World Resources Institute projects that all six GCC countries will be among the top 10 most water-stressed countries in the world by 2040.6
On the energy side, electricity consumption has been rapidly increasing over time in most GCC countries. The increase in per capita electricity consumption outpaced per capita GDP growth in recent years, suggesting that the former was driven by underlying behaviors. According to the Saudi Standards, Metrology and Quality Organization (SASO), for example, energy consumption in Saudi Arabia has increased by 5.7 percent annually from 2005 (700,000 barrels of oil per day equivalent) to 2016 (1,400,000 barrels of oil per day equivalent), while the number of housing units increased by only 2.5 percent in that period. This spike in consumption, then, is mainly due to wasteful household consumption. Many homes in GCC countries run washing machines and dishwashers with partial loads. Too many consumers leave lights, air-conditioning units, and electronic appliances on when they are not being used. In 2017, more than 70 percent of houses still had inadequate insulation, which accounted for some 25 million air conditioning units with low energy efficiency in operation.7 As a result, GCC countries incur large costs from electricity provision.
GCC countries can modify the behaviors linked to excessive consumption of water and electricity through behavioral interventions and structural levers. This would involve tactical interventions and awareness campaigns on the behavioral side, followed by legislation to provide the structural levers. They could redesign energy bills to showcase the subsidy level, introduce bill comparisons between households in the same neighborhood, make “green plans” the default option, or introduce energy labels. In 2015, for example, the Saudi Energy Efficiency Center (SEEC) introduced certificates of conformity for all imported and locally manufactured air-conditioning units that match its new energy efficiency standards. Such initiatives promote the use of energy-efficient appliances such as washing machines and refrigerators with digital inverters or low-flow showers and toilet bowls. They could also draw comparisons between water and electricity consumption in households and their counterparts in countries with similar income levels. In parallel, public awareness campaigns could also educate people about the environmental impact of their excessive consumption.
Egypt provides a relevant example of the use of communication campaigns to make consumers aware of energy usage. In 2014, the demands on Egypt’s electricity supply were exceeding generating capacity, leading to daily power cuts. Household consumption was a large part of the problem, accounting for 42 percent of the total, compared to industry and public facilities that were 32 and 26 percent, respectively. Egypt’s National Initiative for Energy Conservation campaign exploited traditional and digital media channels to send messages about consumption rates and nudges in the form of energy efficiency tips. For example, the campaign told consumers to set air conditioning units to the right temperature, use the right fuel for cars, and turn off electrical kitchen appliances when not in use. The campaign discussed the issue on talk shows. TV stations displayed a power meter on their screens showing real-time peaks in national electricity consumption. The campaign achieved a 3.7 percent reduction in electricity demand in two months.8
GCC countries do not recycle much. In 2016, for example, Kuwait, Oman, and the UAE each recycled only 11 percent of their municipal solid waste, according to the Gulf Petrochemicals and Chemicals Association, while Saudi Arabia and Qatar recycled only 1 percent each.9 By comparison, European Environment Agency figures show that the recycling rate of municipal solid waste in EU member states (excluding Croatia) has been steadily improving year-on-year, reaching 44 percent in 2014, with the leading recyclers being Germany (64 percent), Austria (56 percent), and Belgium (55 percent).10 Part of this problem is the limited consumer awareness of the value of recyclable goods for the economy and environment, or of the need to avoid plastics litter, sort materials for collection for recycling, or reduce the consumption of non-recyclables. This is coupled with an insufficient recycling infrastructure and a lack of related policies at both the municipal and national levels. However, cognitive biases also interfere in the way people recycle (or don’t). For example, people tend to sort intact papers, cans, bottles, and the like, for recycling, but prefer to throw away objects in the trash when they appear too damaged, even if they are recyclable, something known as the “distortion bias.”11
In 2016, the China Environment Protection Agency (CEPA) incorporated behavioral insights into a communication campaign to improve trash disposal in the correct recycling bins. Prior to launching this campaign, China had tried unsuccessfully for 20 years to educate the public about sorting recyclable waste. The campaign triggered Chinese citizens’ sense of peer pressure and attention to social standing to improve their sorting performance. The CEPA did this by rebranding standard recycling bins on a busy food street in Shanghai and linking them to IQ levels — the more sophisticated the bin’s waste segregation, the higher the person’s IQ. It became a matter of pride for citizens to project a public image of themselves as having high IQs, and they made a conscious effort to use the more segregated bins. In one day, more than 89 percent of the total waste collected in these bins had been placed in the correctly designated slot (see Exhibit 5).12
As GCC residents move to a more sedentary lifestyle, rates of obesity and other chronic ailments are rising, with the population becoming more prone to excessive consumption of unhealthy fast food and limited physical activity. According to data from the World Health Organization (WHO), obesity rates across the GCC are among the highest in the world (see Exhibit 6). WHO data also show that the lack of physical activity has reached alarming levels in the GCC. For instance, in 2016, it had reach 53.6 percent in Kuwait, 33.3 percent in Qatar, 58.5 percent in Saudi Arabia, and 30.2 percent in the UAE.13 In addition, high temperatures contribute to the lack of exercise by confining people indoors.
Once again, cognitive biases can lead people to make unhealthy eating choices or abandon their weight loss goals. “Unit bias” compels people to finish a certain portion or unit of food because they perceive it as appropriate. Customers at fast food chains may feel compelled to finish their combo meals whether or not they are full, even if they purchased “plus-size” options (e.g., large fries or soda). They may also unconsciously experience “loss aversion,” the sense that eating only part of their meal or not taking full advantage of their “free refill” option somehow reduces their enjoyment of their meal. Another cognitive bias that sabotages people’s attempts to make healthier choices is “optimism bias,” the unrealistic belief that everything will be alright — until it is too late. This bias can prevent people from wanting to change in the first place, even if the health risks of overeating are explained to them, or it might make them think it is okay to cheat on their diet and eat food that is too sweet or too fat “once in a while,” until this becomes more frequent, with detrimental effects.
The rising incidence of chronic diseases caused by obesity and overeating imposes steep financial costs on GCC healthcare systems. For example, the total direct and indirect cost of non-communicable diseases, including diabetes, was close to US$36 billion in 2013, according to a Strategy& study, “The $68 billion challenge: Quantifying and tackling the burden of chronic diseases in the GCC.”14
GCC countries could decrease obesity rates and the associated healthcare costs if they instill healthy lifestyle habits in their citizens.
GCC countries could decrease obesity rates and the associated healthcare costs if they instill healthy lifestyle habits in their citizens. A behavioral intervention to encourage more physical activity is an important component in reducing the prevalence of obesity. For instance, Singapore’s Health Promotion Board started the National Steps Challenge campaign in December 2015. The intervention was designed to encourage working professionals who have limited spare time to lead a more active lifestyle by walking 10,000 steps a day, a number that was perceived as daunting. The campaign, which is ongoing, allows participants to download a step-counting application and enter a draw every time they accumulate enough points from walking a certain number of steps. It became influential by exploiting Singaporeans’ love of social media — giving participants the chance to collect points by participating and accumulate more points by sharing the application, photos, and videos through social media. The campaign quickly became a social trend, with 126,000 individuals signing up for the initiative, around 8.8 percent of the inactive adult population (see Exhibit 7).15
GCC countries can also lower obesity rates by instilling healthy eating habits through information and nudges. A team led by Professor Janet Schwartz from Duke University found that providing people with the right information had little impact on their eating, but a nudge was effective. Schwartz and her team conducted three experiments in a Chinese restaurant on the university campus. Customers were asked if they wanted to reduce the size of their side dishes. Some customers were offered a 25 percent discount and others were not. What was interesting was that just being asked resulted in 14 to 33 percent of customers accepting to downsize their portions. These customers did not compensate by ordering more main courses or other dishes, and so consumed 200 fewer calories. Interestingly, more customers opted to downsize when the restaurant did not include calorie labels on the menu (21 percent) than when it did (14 percent).16
The Dubai authorities have designed certain initiatives that included behavioral insights to address obesity. Between July and September 2014, for example, the Dubai Municipality launched the “Your Child in Gold” program to encourage parents to raise healthy children. For every kilogram lost over 2 kilograms, overweight families were awarded 2 grams of gold, and individuals 1 gram.17 Although the program had around 7,000 participants, it suffered from two major limitations: first, it was applied for only a short period of time; second, it was financially unsustainable in the long run.
GCC countries are introducing a value added tax (VAT) of 5 percent starting in January 2018 or January 2019, depending on the country. Although it is still early for GCC countries to assess the impact of introducing new taxes, it is safe to assume that implementation could be difficult. Limited awareness of the concept of VAT and its benefits could lead to non-compliance. Traditionally the GCC has been a low- or no-tax environment in which individuals and businesses have a relatively light tax compliance burden compared with their peers in countries where tax systems are already in place. This could provide the grounds for “status quo bias,” wherein companies and individuals prefer things the way they are, as opposed to change. Research from other countries as well suggests that a large percentage of individuals are skeptical about the concept of taxation and misinterpret how taxes are actually being spent by the government.18 This is due to the “isolation effect bias,” people’s tendency to focus on an immediate perceived problem (i.e., having to pay taxes) rather than the bigger picture in terms of the benefits it can bring (e.g., funding for social programs). The key to altering such wrong perceptions is through effective communication campaigns and behavioral interventions.
Kenya has faced similar problems with tax compliance, with citizens apprehensive about the online tax filing process in particular. People continued to file on paper and in person rather than through the more efficient and convenient iTax online system. By investigating the reasons why so few people used iTax, behavioral scientists uncovered a range of barriers including that: citizens saw the process as complex, had low confidence in their accounting skills, and the social norm was to wait until the last minute. The Kenya Revenue Authority therefore designed an integrated awareness campaign in 2014 that used colloquial language to communicate with taxpayers through different channels and encourage them to reconsider their attitudes. In the month preceding the deadline to submit tax forms, the campaign initiated a countdown to convey a sense of urgency. The result was a 312 percent increase in users registered on iTax, $9 billion in taxes collected for the fiscal year 2014–2015, and much shorter queues to file tax returns in person (two million fewer people).19
In the U.K., the BIT conducted effective trials aimed at increasing tax compliance and reducing evasion. First, it introduced social norm messages in control letters to some taxpayers, such as pointing out that most people in the recipients’ local area had already paid their taxes. This increased taxpayers’ response rate by up to 5 percent. It also worked with the Driver and Vehicle Licensing Agency (DVLA) to reduce the number of unlicensed vehicles on the road. Despite DVLA letters to owners of these vehicles, there were an estimated 249,000 unlicensed vehicles in the U.K. in 2011, representing £40 million (around $53 million) in lost revenue. By including photos of their vehicles in the letters to non-payers, the BIT was able to increase payment rates from 40 to 49 percent.20
GCC governments will need to plan ahead carefully to maximize the efficiency of integrating behavioral science into the policymaking process and to manage its challenges. The novelty of this approach means taking deliberate steps in terms of process and thinking through the challenges.
Behavioral insights are mostly integrated at the policy formulation and implementation stages, but should also be taken into account across the policymaking process to ensure the soundness of the policy design, reduce redundancy costs, and allow for effective fine-tuning of interventions (see Exhibit 8).
The traditional policymaking process consists of five stages, according to the Organisation for Economic Co-operation and Development (OECD). These are agenda setting, policy formulation, policy implementation, monitoring, and enforcement.21
Behavioral interventions start with up-front research to determine whether cognitive biases are at the roots of a problem and, accordingly, prepare the right behavioral tools to influence target individuals’ behavior. This is the longest part of the process and can take up to six or eight months. The actual period of application of behavioral tools is shorter and usually spans three to four months. Finally, once the tools are removed, it is important to monitor the target individuals and verify that the change in their behavior is lasting.
At the agenda setting stage, policymakers should determine whether a problem is caused by cognitive biases and set the targeted objectives accordingly. Wrong judgments at this stage can restrict the spectrum of available policy options and jeopardize the success of the eventual policy.
At the policy implementation stage, a behavioral intervention should first be tested in a randomized control trial.
At the policy formulation stage, policymakers define the problem, target population, and attainable objectives. They analyze behavioral drivers affecting relevant outcomes to determine the appropriate behavioral intervention. They also write an implementation protocol to serve as a manual for conducting the intervention and account for possible ethical issues that could undermine the intervention. Data, surveys, experiments, and lessons on what works or does not work in the implementation of policy and regulation should all feed into the early design of policies. They can outline the various cognitive biases behind certain behaviors and highlight any unique insights. These steps can narrow the research question and allow for a targeted intervention with a higher probability of success.
The importance of taking time to understand the implications of the evidence was underlined by Dr. Harrison of WPP, who told us that “It takes a significant amount of time to derive quality empirical evidence, and too often governments make the mistake of presuming that this first phase of a successful policy implementation can be rushed. One can’t make up these insights. Only time and highly skilled research professionals can yield the results that will be the cornerstone of success.”
At the policy implementation stage, a behavioral intervention should first be tested in a randomized control trial (RCT). This consists of a small-scale study in which people are chosen at random to receive one of several interventions. One control group of people receives a standard intervention or no intervention at all, and serves as a basis for comparison with the test groups that received different interventions. This allows checking that the mechanics of intervention are working properly. After learning from the RCT, researchers can improve the concepts, messages, and content. The results have to be analyzed using statistical methods to accept or reject one of the initial hypotheses and confirm the effectiveness of the intervention. Once the intervention achieves its objectives, it is rolled out at a larger scale.
At the monitoring stage, policymakers analyze the intervention and obtain data and statistics that allow them to understand the uptake of their interventions.
At the enforcement stage, policymakers fine-tune and adapt their policies, and ensure compliance with the interventions.
Applying behavioral interventions poses three main challenges to policymakers related to standardization difficulties, ethical concerns, and undesired outcomes.
GCC countries must each proceed to mainstream the use of behavioral insights in policymaking.
Globally, governments have adopted various operating models to institutionalize behavioral insights, with differing institutional setups and degrees of behavioral intervention. Each GCC government should start by centralizing behavioral insights expertise, and assess alternative models as it builds know-how and gains credibility.
Behavioral insights units have been incorporated into public institutions in one of three structures: a centralized model, a decentralized model, or a hybrid one (see Exhibit 9). The choice of model depends on the institutional culture of the country and the intensity of behavioral interventions. Models can co-exist and can evolve over time. Additionally, these units could be either permanent in-house teams or governmental task forces assembled on a project basis.
A centralized model consists of a single behavioral insights unit that is fully owned by the national government and usually derives from a high-level entity such as the prime minister’s office. Such a unit has the mandate of integrating and using behavioral insights in policymaking and supporting the use of behavioral interventions across government bodies. Australia’s BETA, established in 2016, is one example of a centralized model. Originally, 17 different departments within the Australian government had started experimenting independently with applying behavioral insights in public policy. As behavioral insights became more mainstream, policymakers recognized the benefits and economies of scale they could achieve by combining their resources and sharing their experience. This led them to merge as a single central unit under the Department of the Prime Minister and Cabinet. Another example of a centralized unit is France’s Secretariat-General for Government Modernization (SGMAP), which is part of the office of the Prime Minister. The SGMAP aims to encourage more public institutions to use behavioral insights in policymaking and supports the digitization of government.
A decentralized model involves different government entities heading their own separate behavioral insights units for their specific policymaking purposes. The governments of Denmark and the Netherlands, for example, do not have a specialized unit for the application of behavioral insights. Instead, different public institutions have formed or are currently forming their own behavioral insights teams to complement policies related to their respective mandates. These include the Danish ministries of employment, children, and taxation, and the Dutch councils of societal development and infrastructure and the environment.24
A hybrid model is the result of the combined features and functionalities of both centralized and decentralized models. This is the case with the BIT in the United Kingdom. In addition to projects with the central government related to taxation, energy, and health, among others, the BIT works directly with the Greater Manchester Combined Authority and other local and metropolitan authorities across England and Wales. Today, behavior insights capabilities are hosted across the BIT and different departments of the U.K. government.
The BIT was originally established in 2010 as a fully government-owned agency to support policymaking in the U.K. Cabinet Office. It was set up at a time of austerity when departments were facing major cuts and being encouraged to avoid costly initiatives. The initial team had seven members, a modest budget, and limited office space.
The BIT’s main objectives included transforming the approach of at least two major government departments by championing a new understanding of human behavior. As a condition for its continuation, the BIT team had the objective to produce results and deliver at least a tenfold return on costs within two years.
Since then, the BIT has conducted numerous behavioral interventions that have led to significant social, economic, and environmental improvements in the U.K. and abroad. It draws on behavioral science literature and runs RCTs to find ways of making public services more cost-effective and easier for citizens to use. In the U.K., it is responsible for recruiting 100,000 more organ donors annually, bringing in an additional £200 million ($260 million) in income taxes, and persuading 20 percent more households to switch to a “green” energy provider.
The success of its early interventions contributed to making behavioral insights more mainstream in the U.K. civil service and in several European countries, with more governments worldwide expressing an interest in the field. To answer this growing demand, the BIT spun out of the government in 2014. Today, the BIT has over 60 employees and has become a for-profit social purpose company, jointly owned by its employees, the Cabinet Office, and the Nesta foundation. Its mission is to provide behavioral insights capabilities to different public-service organizations in the U.K. and worldwide (the Government of New South Wales in Australia and the Ministry of Manpower in Singapore are customers).25
Internationally, it worked on increasing the corporate tax compliance rate in Costa Rica and Poland, and also launched schemes to improve medical programs in Mexico and Moldova. Additionally, it is tackling broader projects within the World Bank and United Nations Development Programme. The BIT also organizes Behavioral Exchange, the annual international conference for behavioral scientists.
GCC governments must each start with a centralized behavioral insights unit.
Initially, each GCC government will need a specialized, centralized unit to start launching behavioral intervention programs. This unit consists of a small core team of five to eight behavioral intervention experts supported by additional researchers. According to Dr. Harrison of WPP, the most efficient team likely consists of both public and private talent working together. Private teams to conduct the research and uncover the behavioral insights and public expertise to translate the research findings into effective public policy.
Such a unit is necessary at this early stage because most departments and civil servants found in each of the separate GCC governments are unfamiliar with the concept of behavioral science in policymaking.
The unit should start by commissioning leading institutions and experts outside the government to lead behavioral interventions, and gradually own the process as it builds internal expertise. Such a centralized model will allow an individual government to:
For the unit to operate at maximum efficiency, the government should put in place a clear governance process, build up the right capabilities, expand them, and strike key partnerships.
For a centralized behavioral insights unit to operate at maximum efficiency, it is essential that it have a clear level of engagement in the policymaking process, strong support from political leaders, and a well-defined mandate regarding the sectors it should be covering. Such a unit should also support the sharing of tools and processes for conducting interventions across government entities.
The behavioral insights unit should start by conducting a simple intervention that yields a quick win to build up its credibility. At an early stage, it may not have the sufficient resources and experience to conduct this intervention with its in-house team, and should commission leading behavioral institutions and scientists outside the government to lead it. Leveraging skills from such experts will help the team build their own capabilities and communicate the usefulness of behavioral insights to different policymakers. For example, after the New South Wales Government in Australia partnered with the U.K.’s BIT in 2012, it was able to establish its own Behavioral Insights Unit to support its policymaking.
The unit should also consider secondment to reduce the risk of data infiltration and information leakage. In the U.K., Her Majesty’s Revenues and Customs (HMRC) arranged on several instances for someone from the BIT team to be directly seconded by a member of their department. This gave HMRC oversight of the persons or teams running RCTs and reassured it when it came to guaranteeing data, without the need for statutory barriers.
Government entities and departments that have been exposed to behavioral interventions, and have acquired know-how in the topic, should ensure that their expertise is distributed across other government bodies. “What matters most is that the skill set is world class and able to read across multiple types of evidence to uncover the insights that will lead to lasting behavior change,” said Dr. Harrison. “This quality of t
For instance, Danish policymakers from different government entities engaged in their own behavioral insights programs are taking the initiative to teach their colleagues in other public functions how behavioral insights can inform policymaking or how to spot policy areas that could benefit from such insights. They are working to promote the use of behavioral insights in policymaking and establish common standards for interventions.
To be able to commission more sophisticated and effective interventions, a behavioral insights unit should build partnerships with different stakeholders, including governmental organizations, academia, international behavioral institutions, and the private sector. Government entities facilitate the unit’s access to resources and data that are not publicly available — while taking measures to prevent the leaking of sensitive information or breaches of privacy. Researchers and academics can support the design of future interventions by sharing their expertise, methods, success stories, and the challenges they faced when conducting interventions in different contexts. Private-sector stakeholders that share common objectives with government-led behavioral interventions can also provide valuable support. For example, a sportswear brand can sponsor events and competitions or co-brand its marketing messages to reinforce a governmental campaign to increase citizens’ physical activity.
With these four elements in place, the centralized unit can grow or evolve into different models as needed. In fact, as the use of behavioral interventions progresses, the need for a central model to provide cross-support, advice, and coordination diminishes. At that point, different government departments will have acquired sufficient experience to apply behavioral tools in policymaking where necessary and drive change.
Behavioral science is a new policymaking discipline in which there has recently been growing interest by governments around the world. For GCC governments, it holds considerable promise if used appropriately. Applying interventions based on behavioral science insights could help them achieve important socioeconomic objectives where standard policy levers alone may not suffice. Most prominently, behavioral interventions can help make water and electricity consumption more efficient, encourage more recycling by GCC residents, promote healthier lifestyles, and create a culture of compliance with new taxes.
To use behavioral interventions, each GCC government should set up a behavioral insights team. The team should be dedicated to designing interventions to support the enforcement of policies and to continuously gathering behavioral science insights to optimize the effectiveness of these interventions. By adopting this approach, each GCC government will be able to complement its large-scale national transformation plan.