How is nudging applied as a behavioural tactic to
decrease smoking behaviour?
Laura Admiraal – 10800697 Bachelor Thesis
Universiteit van Amsterdam Edward Huizenga
June 27th 2017
Word count abstract: 156 Word count thesis: 12,523
2 Abstract
This research studies how Thaler and Sunstein’s nudging is applied as a behavioural tactic to decrease smoking behaviour. Therefore, one campaign and two policies are
analysed, namely the American Truth campaign, the Australian plain packaging policy and the snus legislation in Sweden. This study developed a tool to assess to what extent a campaign or policy is a nudge: the Nudginess Checklist. This is a checklist with six criteria based on existing literature. All three approaches are analysed with this checklist. The results show that the American Truth campaign is a mere nudge, in contrast to the Australian and Swedish policies, which scored 4 and 5 out of 6 respectively. None of the three approaches were explicitly meant as a nudge. Thus, even though an approach is not intended as a nudge, it possesses many principles of nudging. Further research should investigate a possible relationship between the nudginess of a campaign or policy and its effectivity.
3 Contents 1. Introduction 4 2. Theoretical Framework 5 2.1. Behavioural Science 5 2.2. Nudging 9 2.3. Smoking 11 2.4. Politics 13 2.5. Ethics 13 2.6. Conceptual Framework 14 3. Method 15 3.1. Material 15 3.2. Checklist 16 4. Results 18
4.1. United States of America 19
4.2. Australia 20
4.3. Sweden 22
5. Discussion 25
4 1. Introduction
Consumer behaviour theories are implemented in designing frameworks to change consumer behaviour. In order to change behaviour, an understanding of it is needed first. Several theories try to provide this understanding. On one side, there are the theories that are based on the idea that human decision-making is a rational, conscious and elaborate process. This is called type 2 thinking (Kahneman, 2011). Two examples of this theory are the consumption and production theory. The consumption theory states that when making a decision, a person’s goal is to optimize his or her utility. So people give all the features of the different options a relative weight, depending on how important a specific feature is to them. Then, every feature gets scored based on how good it is for that specific product. This is done for all the alternatives. The option with the highest total score is the one that has the highest utility. Therefore, this option should be chosen, according to the consumption theory. The production theory argues that it are the profits that are maximized, not the utility. So options are also compared to each other, but the criterion now is how profitable the options are. On the contrary side there is the theory that human decision-making is an irrational,
unconscious and automatic process. This is called type 1 thinking, which is the fast, automated and unconscious thought process (Kahneman, 2011). Previous theories like the consumption and production theory reason from the type 2 thinking, but new insights have caused the focus to shift to type 1 thinking. Kahneman and Tversky (1979) developed the type-1 prospect theory as a critique to the consumption theory. The prospect theory states that when under certain circumstances, people do not make rational decisions. Exemplary circumstances are time pressure or when risks are involved. When people are under time pressure they become more averse to the risk of losing, then when that pressure is absent (Kocher, Pahlke & Trautmann, 2013). This means rational decision making theories are not applicable to all situations. Clearly, under certain circumstances people use other thought processes to come to a decision. Therefore the need for an alternative theory arises to explain this different thought process. A theory that takes human’s irrational nature into account. Nudging is such a theory, based on type 1 thinking. Nudging is an applicable theory that strives to change destructive consumer behaviour. Consumers do not always behave in a manner that is good for themselves or the environment. They tend to behave in irrational ways (Feher da Silva, Baldo & Chialvo, 2012). One very clear example of this is that everyone knows that you should eat healthy and that snacks and sweets are bad for you. Another principle that can be considered common knowledge is that one should engage in some form of physical exercise. Despite this knowledge, in the Netherlands 43% of the population is overweight (CBS, 2016). If people made rational decisions, this would not be the case. But people do not always make the optimal decision, so they eat unhealthy food and do not get enough physical exercise. This manner of irrational behaviour is quite easy to predict (Ariely, 2008). Nudging is concerned with changing the behaviour of individual consumers. It embraces the system 1 theory that states people use fast, unconscious thinking for most of their decision making, instead of elaborate thinking. Nudging is part of the political movement of libertarian paternalism, which momentarily mainly takes place in the United Kingdom (Thaler & Sunstein, 2008). Libertarian paternalism is best explained by separately explaining both words. First, the libertarian part is about freedom. It says
5 consumers must be free to make their own decisions. So campaigns and policies should be constructed whilst aiming to increase or at least maintain consumer’s freedom of choice. Second, the paternalism part is about the idea that the nudge should benefit the chooser. So a campaign should try to make consumers and their environment healthier. The biggest example of behaviour that should be nudged is probably smoking. Although the potentially catastrophic consequences of smoking are widely known, still a substantial part of the world population smokes. This causes problems for the public health, an area that is in need of a successful strategy of changing wrong consumer decisions. Smoking raises the chances of lethal diseases such as lung cancer and respiratory diseases. So smoking has a negative effect on public health. This is a problem because the life expectancy of people will decline and with that also their quality of life. Also, large health issues bear large health costs, which are partially being paid indirectly by the rest of the inhabitants of the country via government taxes. Numerous of campaigns have been implemented all over the world, but still almost one out of four people smokes. So there is a need for a more effective approach.
Therefore this study investigates the following research question: how is nudging applied as a behavioural tactic to decrease smoking behaviour? In order to answer this question, tobacco control measures of three countries, U.S.A., Australia and Sweden, will be assessed. First a theoretical framework will be given with information about behavioural science, which constitutes the theory on which nudging is based. Then, an overview of previous research on nudging is given. Thereafter follows a summation of tactics that have been used in the past to decrease smoking behaviour. This follows the World Health
Organization’s MPOWER model, a model that lists six ways to counter smoking behaviour. Because nudging is meant to help in the making of policies there is a section about politics included. Also, nudging is quite contentious because of some ethical issues, therefore the theoretical framework ends with a section that discusses these issues. Then the method follows, where the ‘Nudginess Checklist’ is developed based on the existing literature. This tool is designed to determine to what degree an approach can be judged as a nudge, so that multiple approaches can be compared on their so-called nudginess. Thereafter, the policies and approaches of the three countries are discussed in the result section. This article will end with a discussion, followed by a conclusion and recommendations for further research.
2. Theoretical Framework 2.1. Behavioural Science
Behavioural science is an interdisciplinary science, originated from a combination of economist and psychologist. One big category is behavioural economics, which is concerned with the decision process that people go through when considering a purchase. This falls out into two broad categories: the classic theory of rational man and the theory of irrational man. It has become clear that certain modifications are needed to the theory of rational behaviour, because in many experiments it has been proven that the classical theory is not sufficient (Simon, 1959). Even processes that appear to be rational can sometimes be performed irrational (Langer, Blank & Chanowitz, 1978). In three experiments results suggested that a
6 lot of actions are unexpectedly performed mindlessly. An example is one experiment where the experimenter tries to skip ahead in the line of the copying machine. When he gave a reason as to why he should be let first, the people in the queue where more likely to let him go first. The surprising finding was that it did not matter whether that reason was
informational or not. The experimenter gave the reasons ‘because I need to make copies’ and ‘because I am in a rush’. The first one consists of placebic information, since everyone in line needs to make copies, and the second of real information. But in both cases the experimenter was let in front of the line. This shows that people in everyday situations do not process information rationally, although it seems like it at face value (Langer et al., 1998).
Another demonstration of human’s irrational thinking was given through research of framing effects. Framing effects are the consequences of how a questions or choice is framed on the decision maker’s decision. Tversky and Kahneman (1981) researched these effects. They found that when faced with a monetary decision people became more aversive of perceived loss and more attracted towards perceived gains. The same tendency was found for making decisions about human life and death. This is illustrated by the experiments executed in the research. Two groups of respondents had to choose between two options to save as many of the 600 people that were threatened by the outbreak of a deadly Asian epidemic. The first group had to choose from the following two options: option A: “200 people will be saved” and option B: “There is 1/3 probability that 600 people will be saved, and 2/3 probability that no people will be saved”. The second group had the options: option C: “400 people will die” and option D: “There is 1/3 probability that nobody will die and 2/3 probability that 600 people will die”. Important to notice is that option A and C and option B and D are the same, although they are framed differently. A and B are framed as a gain and C and D are framed as a loss. In the first group 72% choose option A and in the second group 78% choose option D. So when presented in a scenario framed as a loss, people take more risks and when presented as a gain people take less risks. These framing effects grow under time pressure, so these effects originate from the intuitive system (Guo, Trueblood &
Diederich, 2017).
Dan Ariely is a behavioural economist that has written a lot about the irrational behaviour of people. He has supported the theory of irrational behaviour with numerous surprising and exciting experiments. In his book Predictably Irrational (Ariely, 2008) he makes an interesting statement. Apart from the theory that people behave irrationally in making decisions, he also stated that the pattern of these irrational decision is systematic and therefore predictable. One of the given examples is that Americans do not save enough for their retirement. This cannot be explained by the theory of rational behaviour, but when applying the irrational behavioural theory it is explainable and even predictable. Namely, the underlying and systematic mechanism is that people procrastinate. This personal trait is very common and explains a lot of the observed irrational behaviour. People do not think enough about the future, but instead procrastinate most of the work as long as possible, including saving up for retirement. Also, people simply do not always understand the future costs associated with not saving now and the future benefits of depositing a fixed amount every month in a savings account. Many more explanations of this sort are possible.
7 He advocates that economists should take this into account by focussing more on how
people actually think instead of on how they should think. In this one instance a solution was offered that does take irrational behaviour into account, called “Save More Tomorrow” or the “SMarT program” (Thaler & Benartzi, 2004). It builds on the fact that people are not prone to put money aside from their current salary for future benefits. But, people find it less hard to allocate a proportion of their future salary increase to their retirement. When this option was offered to employees 78% joined the plan and 80% of them was still part of the plan during the fourth pay rise. This resulted in a substantial increase of employees’ saving rates. The point Ariely (2008) demonstrates here is that when reasoning from the idea that people behave irrational better solutions can be created to counteract this irrationality, than when reasoning as if people are rational.
Besides procrastination and weak understanding of the options to choose from, there are more often used thinking mechanisms. A mechanisms that floats in the middle of
irrational and rational behaviour is the use of heuristics (Albar & Jetter, 2009). These are rules of thumb that are used when due to human limitations and circumstances information
cannot be processed rationally. The biggest of the those limitations are the limited cognitive resources of humans and the limited working memory in particular. In the working memory one can hold a number of seven items plus minus two items. If more items are to be
processed heuristics are often used. Three of the most commonly used economical heuristics to estimate probabilities and forecast values will be explained next (Tversky & Kahneman, 1974). First, the representativeness heuristics is used to answer questions regarding the probability that a certain object belongs to a certain class. When that object is more
representative of the class, the probability that it belongs to that class is estimated higher. For example robins are more representative for the category birds than penguins are (Gilhooly, Lyddy & Pollick, 2014). So people will estimate the probability that robins belong to the class birds higher than that of penguins. In many instances, this heuristic is quite useful and effective and leads to the right decision. However, there are some serious errors that can arise from using this heuristic. One is the conjunction fallacy (Tversky & Kahneman, 1983). In one experiment people first read a description of Linda which stated here as a feminist. Then statements were presented among which “Linda is active in the feminist movement”, “Linda is a bank teller” and “Linda is a bank teller and active in the feminist movement”. The participants had to rate statements from most likely to least likely. The most chosen sequence from most likely to least likely was: feminist, bank teller plus feminist, bank teller. Also, nearly all participants rated the conjunction of bank teller and feminist higher than bank teller or feminist alone. This is irrational, since the likelihood of Linda being one of the two options is always higher than the likelihood of her being both options. These irrational rankings can be attributed to the representativeness heuristic, since people find the
conjunction more probable than bank teller alone, because Linda’s description made her more representative for feminists than for bank tellers. This violates the extension law of probability and therefore demonstrates how the representativeness heuristic can lead to errors (Tversky & Kahneman, 1983).
8 The second heuristic is the availability heuristic. When judging the probability of a certain event to happen, people often depend on how easy information of the occurrence of such an event comes to mind (Tversky & Kahneman, 1974). For example, after a recent plane crash travellers estimate the probability of another plane crash higher. This effect is even bigger when the crash has been covered in the news extensively, because people can easier bring the event to mind after all the commotion (Gilhooly et al., 2014). As is clear from the example, the availability heuristic often leads to fallacious thinking and decisions. The Word problem is an experimental instrument designed to demonstrate this (Tversky & Kahneman, 1983). Here participants had to determine whether a word containing three or more letters was more likely to be found in a random English text when it started with an r (e.g. road) or when it had an r in the third position (e.g. car). Most people said that the chance to encounter a word starting with an r was bigger. They make this judgements even for consonants that are more present in the third position than in the first position, for example the r and k. Simply because it is easier to bring words to mind when given the starting letter, than when given a letter in the middle or end of the word. Another example is that death causes that receive more elaborate publicity (e.g. murder) are considered to be more likely than death causes that receive less publicity (e.g. suicide), (Lichtenstein, Slovic, Fischhoff, Layman & Combes, 1978). Illustrated by these two examples, the availability heuristics leads to errors in the form of availability bias.
Third, the anchoring and adjustment heuristic. People often estimate the unknown value of objects and events by adjusting from an known initial value (Tversky & Kahneman, 1974). This initial value can be given through formulation of the problem or other sources of information and is called the anchoring point. Important is that the anchoring point – relevant or not – influences the final estimated value. In an experiment that demonstrates this effect, participants had to estimate the percentage of African countries in the United Nations. Prior to making that estimation a wheel was spun to determine the anchoring point. Clearly, this anchoring point is non-informative. First participants had to indicate if the percentage of African Countries was higher or lower than the anchoring point and after that they had to state the estimated value. The participants that got 10% as an anchoring point, estimated a final value of 25% and participants with a anchoring point of 65% estimated a value of 45%. This shows that the initial value that is given has a clear effect on the final estimated value, even though the initial value carries no information. So when the initial value is relatively low, the final estimated value will also be relatively low. The anchoring and adjustment heuristic is also used when conducting a partial computation. For example, while 1 x 2 x 3 x 4 x 5 x 6 x 7 x 8 and 8 x 7 x 6 x 5 x 4 x 3 x 2 x 1 have exactly the same
outcome, the estimations of the outcome differ greatly. The first computation is estimated to have a value of 512, while the second computation is estimated at an outcome of 2250. These different estimations of the right answer, which is 40320, can be explained by the anchoring and adjustment heuristic. When estimating the outcome, the first part of the computation is calculated to form an anchoring point. Then, based on that partial calculation the number is adjusted towards the estimated outcome. With the first computation the anchoring point is lower, because computing the first part of the computation leads to a lower value than computing the first part of the second computation. The lower anchoring point of the first
9 computation directly influences the lower final estimated value, and vice versa for the
second computation. These experiments illustrate how the anchoring and adjustment heuristic leads to erroneous decision making (Tversky & Kahneman, 1974).
For this knowledge about irrational thinking and behaviour to be useful, it should be forged into a practical tool. Thaler and Sunstein (2008) did this by using the knowledge on irrationality to guide people to the right decisions, since people apparently do not always make decisions that are good for them. They provoke behaviour that is good for the people involved and for the environment, without the people noticing that they are being
influenced. This is called nudging. The essence of nudging is best described through a citation. “A nudge, as we will use the term, is any aspect of the choice architecture that alters people’s behavior in a predictable way without forbidding any options or significantly changing their economic incentives. To count as a mere nudge, the intervention must be easy and cheap to avoid.” (Thaler & Sunstein, 2008, p. 6)
2.2. Nudging
A book on nudging was first written in 2008, so nudging is a relatively young research area (Thaler & Sunstein, 2008). Nudging originated at the toilets of Amsterdam Airport, where economist Kieboom decided to implement a new approach (Lawton, 2013). The men’s toilets were dirty because a lot of pee spilt from the urinal. This led to unhygienic sanitary and high cleaning bills. Kieboom came with a simple solution: he placed a sticker of a fly in the urinal, which changed colour when it came in contact with urine. Men aimed at the fly and substantially less urine was spilt. Men did not think about this, it was an
instinctive reaction (Vicente, 2006). This resulted in cleaner bathrooms and lower cleaning costs. Kieboom’s idea and its success gave rise to the nudging approach and is often cited as the first application of nudging, even though the term nudging was not present yet.
However, this first initiative was merely a business policy, not a scientific research. So no results can be drawn from this one example. But it inspired a lot of research that is scientific.
Since nudging is such a young object of study most of the research is exploratory. Lots of different subjects have been studied, but all to a limited extent. The research focuses mostly on two broad topics: public health and the environment. This study is concerned with the role of nudging in smoking behaviour and will therefore focus on the topic of public health. Public health is the health of a certain population as a whole, this population is usually the inhabitants of a country. A good public health in this research consists primarily of high life expectancy and a low patient/non-patient rate. Many public health subtopics are minimally researched and not a lot of topics are studied by more than two studies. Because of this, the theoretical framework consists of many different topics, that are all related to the role of nudging in public health.
A big burden on public health is breast cancer, since its prevalence in the Netherlands increases every year since 2013 (CBS, 2017). In 2008 it was the biggest cause of death among women between 35 and 50 years old. When it is discovered early the chance of curing from the disease is substantially bigger (Ploug, Holm & Broderson, 2012). Therefore screening programmes are developed, to identify and diagnose breast cancer in the early stages. The
10 public health will thus benefit from more people enrolling in this program. This is where the irrational behaviour comes in again: women know that participating in this program only has benefits. Namely, either reassurance of their health or early discovery of breast cancer. Nevertheless, a big proportion of the women does not enrol in the program. As a reaction the Danish government has applied several aspects of nudging in their invitation to
participate in the screening program. First, a default options was used. Women get a pre-set appointment in their invitation and they have to cancel in case they do not wish to attend that appointment. Second, the invitation includes a strong recommendation from the Danish National Board of Health to participate in the program. Hence, a choice to not participate in the program is associated with a bigger health risk and opposing the national healthcare institution. Whether this approach is effective does not become clear in the article. It is argued that the screening programme causes more harm for the women than it benefits them. So this nudge might not lead to the optimal health choice. Thus the take away from this is that with examining every nudge it should be analysed whether the choice people are being stimulated towards is purely health-enhancing, since this is not always the case.
Another big and worldwide growing threat to public health, as mentioned before, is obesity. Individual or small group interventions have proven to be successful, but to decrease obesity substantially and on a greater scale, changes in the food environment are necessary (Thorndike, Riis, Sonnenberg & Levy, 2014). The role of nudging in this
intervention is twofold: target automatic process and preferences for convenience. In the cafeteria of a 907-bed teaching hospital in the U.S.A. a traffic light coding system was applied to the products. Healthy products were labelled green, products that contained equally as much healthy as unhealthy ingredients were labelled yellow and last, the unhealthy products were labelled red. After the implementation of the coding system, choice
architecture was applied. Green labelled products were made more visible by for example putting them at eye level and placing them in baskets throughout the cafeteria. In the same manner the red labelled products were made less visible. The proportion of green items increased from 41% to 45% after 12 months and to 46% after 24 months. Complementary, the red items decreased from 24% to 21% after 12 months and remained 21% after 24 months. Concluding, this study found long-term effectiveness of nudging in making healthier nutrition choices. A problem with this study is that it investigates the effects of the traffic-light labelling and the choice architecture simultaneously. However, both interventions are nudges. So it has proven that nudging is effective in combatting obesity, one way or another (Thorndike et al., 2014)
Also, a big public health struggle in many countries is the shortage on organ donors. This results in long waitlists, which are unfortunately even too long for many of the patients. Thaler and Sunstein (2008) give approaches to organize organ donor registration ranging from explicit consent to what they call “routine removal”, which means that the state ones the rights to deceased peoples body parts. A nudging approach is the presumed consent, where people are presumed to be an organ donor, unless they indicate differently. Changing their status to ‘not an organ donor’ should be easy and costless, to count as a pure nudge. In the United Kingdom a research has been conducted to see if nudging can be effective in
11 increasing the number of organ donors (Morgan, Deedat & Kenten, 2015). Many countries, including the United Kingdom use an opt-in system for organ donor registration, which falls under the explicit consent approach. So to become an organ donor one has to register
themselves online or check the corresponding box when applying for governmental
documents. The default option with this policy is not becoming an organ donor. In this study they found that for populations with many ethnic or religious minority groups a lot of factors contribute to the decision to become an organ donor, for example the absence of an organ donor system in their home country and the trust in donation and transportation process. Therefore, a nudging strategy regarding organ donation might be less effective in these populations. This effect decreases as the minority group becomes more heterogenous (Morgan et al., 2015).
2.3. Smoking
Worldwide, a lot has already been undertaken to decrease smoking or to prevent it all together. This action has mostly been undertaken by the government or other public organizations. The World Health Organization set up the WHO Framework of Tobacco Control in 2005 as a first step to create a treaty between countries to decrease the supply and demand of tobacco (World Health Organization, 2015b). It has now expanded over more than 150 countries. A tool presented by the Framework is the MPOWER package, which lists six effective policies to counteract smoking. MPOWER is an acronym for the six policies: monitor, protect, offer, warn, enforce and raise (World Health Organization, 2008.
Primary to all other five polices, monitoring is necessary in order to understand the consequences of tobacco. Without this, the following five interventions cannot be employed effectively and its success cannot be measured. It must among other monitor the prevalence of tobacco use, the impact of policies and interventions and the marketing of the tobacco industry (World Health Organization, 2008).
First, the most applied tobacco preventing policy is smoke-free legislation to protect non-smokers from negative health consequences (World Health Organization, 2013). Exposure to tobacco smoke, by second-hand smoking, is harmful to a person’s health. This can be counteracted by smoke-free laws, that prohibit smoking in public places and
workplaces. In Scotland the air quality in bars, restaurants and cafes was showed an 86% reduction in small airborne particles, which is an air marker for second-hand smoke, two months after implementation (Semple, Creely, Naji, Miller & Ayres, 2007). By enforcing strong smoke-free legislation public health will benefit quickly and substantially, primarily through reduction in heart attacks (Lightwood & Glantz, 2009). Worldwide 1.1 billion people, 16% of the world’s population, are protected by strong smoke-free laws (World Health Organization, 2013).
A second effective measure is offering help in the cessation of tobacco (World Health Organization, 2013). There are three effective ways to offer this help. One is cessation advice in primary health care, which is advice from doctors and other healthcare employees.
Another are the quit lines, which offers counselling via a free telephone line. There is a direct positive relation between the educational level of the caller and the length of the call and
12 pharmacological therapy. The advice is to also offer services more focussed on lower
educated smokers (Willemsen, Van Der Meer, & Schippers, 2010). Last is the provision of pharmacological therapy, which is most common in the form nicotine replacement therapy like nicotine gum. However, only 1 billion people live in the 21 countries that offer help to stop smoking and little progress has yet been made in getting this number up (World Health Organization, 2013).
Third, despite the common idea that smokers are aware of the harmful consequences of their behaviour, a big part of them considers it as a bad habit instead of behaviour that involves serious health risks (World Health Organization, 2008). Most people believe that a small consumption of tobacco is not that harmful. Additionally, apart from the increased risk of lung cancer, many smokers cannot name more risks associated with tobacco use. One clear implementation of this policy are the mandatory health risk warnings on tobacco packaging. This appears in the form of textual and graphical warnings, and often in combination. An Australian study suggested that graphical warnings might be more effective than textual warnings (Borland et al., 2009). Graphical warnings are especially more effective in less developed countries with high illiteracy rates and for young people, including the children of smokers (World Health Organization, 2008). Another form of warning are governmental initiated anti-smoking counteracting campaigns, to warn people about the health risks associated with direct and second-hand smoking (World Health Organization, 2008).
Fourth, efforts can been made to reduce the promotion of smoking and other tobacco products by enforcing bans on tobacco advertising. Since tobacco advertising increases tobacco consumption, a logical step in decreasing tobacco use is banning the advertisement of it all together (Saffer & Chaloupka, 2000). This ban is only effective when it is a
comprehensive ban, where all media advertising on tobacco is prohibited. Many prosperous countries have taken this logical step. Currently 24 countries have enforced a complete ban on all sorts of tobacco advertisement, promotion and sponsorships, a so called TAPS. As a result 10% of the world’s population is now protected by a TAPS. Only two out of these 24 countries are European, namely Spain and Turkey (World Health Organization, 2013). 24 countries might nog seem like a lot, but another 27 countries are only one criterion away from the status of complete TAPS. Point-of-sale advertising is the last remainder of tobacco advertising in most of those countries, followed by sponsorships.
Last, the most effective measure to counteract tobacco consumption is the raise of taxes (World Health Organization, 2013). By raising the price of tobacco with 70%, one-fourth of the deaths related to smoking can be prevented (Alleyne et al., 2006). Also, a study in South-Africa showed that with every 10% price increase, the consumption of tobacco will drop between the 5% and 7% (Van Walbeek, 2003). This method is largely accepted and has been applied for centuries in countries worldwide. Acceptation comes from the fact that tobacco is not an essential good and is harmful for people (World Health Organization, 2008). Also, the popularity of this intervention increases when the money from the taxes is allocated to further tobacco prevention programmes. This intervention has the most effect on young and/or poor sociodemographic groups, because they are the most sensitive to the
13 price of consumer goods. Additionally, lower educated people are more sensitive to tax increase (Willemsen, Van Der Meer & Schippers, 2010).
2.4. Politics
There is a strong connection between nudging and politics, because nudging is meant as a tool for public and private organizations to design policies that affect people’s choices by influencing the outcome of it in the interest of the chooser’s well-being (Thaler & Sunstein, 2008). The political movement that carries out these principles is called libertarian
paternalism. It does not belong to the right or left wing, but is a compromise between the two streams. The use of libertarian and paternalism in one term seems paradoxical, since the first is associated with freedom and the second with regulations and restraints (Ménard, 2010). However, nudging combines the two, by applying paternalism in steering people in the right direction and libertarianism by preserving people’s freedom of choice (Thaler & Sunstein, 2008).
Libertarian paternalism is most popular in the United Kingdom and the U.S.A. In the U.K. the government is particularly interested in behavioural science. The United Kingdom’s Cabinet Office has even set up a special unit to monitor and apply the development in this area. This unit is called the Behavioural Insights Team (COBIT) or the nudge unit (Quigley, 2013). Its priority is to improve public health by applying behavioural knowledge
throughout all departments, not just health care. For example, within the transport department it could stimulate the development of public rent-a-bike schemes (Quigley, 2013). Apart from the aimed effect of limiting the use of other polluting manners of transport, it also contributes to public health by getting people to engage in more physical exercise. The COBIT mainly focuses on organ donation, smoking, diet, alcohol consumption, physical activity and decreasing errors in the National Health Service (Quigley, 2013). These subjects are of importance because they are associated with high national costs, like the costs of obesity-, alcohol- and smoking-related diseases.
Politics should be concerned with decreasing smoking, because it harms not only smokers, but also non-smokers through second-hand smoking. Hence, governmental intervention is beyond just legitimized, it is crucial (Alemmano, 2012). The argument that people should be able to live their live the way they want, without intrusion from the state or others, does not hold its strength against the deadly consequences of smoking and inhaling second-hand smoke. Moreover, the World Health Organization urges countries to intervene in the unacceptable smoking behaviour ‘in the name of public health’ (Alemmano, 2012). According to Alemmano (2012) several tobacco control interventions count as a nudge: graphic warnings on the packages, plain packaging (see Australia) and a visual display ban. However, according to the checklist designed and used later in this article, not all three interventions will count as a mere nudge, but they do possess certain criterion.
2.5. Ethics
A lot of articles have been written on the ethical side of nudging and often academic articles also include a section about ethics. Therefore it is striking that the creators of the
14 principle of nudging, Thaler and Sunstein (2008) do not mention any ethical issues regarding their behavioural approach, even though they mention seven other objections against
nudging. Ethical issues around nudging foremost concern the trade-off between stimulating better health-related choices on one side and hiding facts and intentions from the people involved on the other. Moreover, to what extent can a government or other organization apply nudging for it to remain ethical (Alemanno, 2012).
One main concern is the impairment on the freedom of choice. Even though Thaler and Sunstein (2008) claim to retain the freedom of choice, ethical issues are risen because of the influence that is exerted on people’s choice, without them being aware of it (Selinger & Whyte, 2011). True, people are not impaired in their freedom of choice, however they are unaware of the tactics that possibly steer their choice in another direction than it would have been without the interference. An additional issue is that people habituate to the slight impairments on their freedom of choice. Whilst it begins with a simple, relatively free nudge, it will incrementally lead to further paternalization. This is referred to as the ‘slippery slope’ (Rizzo & Whitman, 2009). In the long term nudging will step by step lead to greater
impairments on the freedom of choice, by for example forbidding products that are harmful for one’s health altogether.
Another issue is that applying nudging abundantly may lead to moral laziness. Because people will get used to being nudged into making the right choices, without participating actively in the decision making of such choices, they might unconsciously assume a sort like mechanism in other situations as well. They expect other members of society to also take responsibility in their health-related choices. This moral laziness is hazardous in situations where no mechanisms help to make the right choice and can therefore eventually lead to people making the wrong choice (Selinger & Whyte, 2011).
2.6. Conceptual Framework
As shown in Figure 1 below, nudging is one of the many facets of behavioural science and it can be deployed to decrease smoking behaviour. Decreased smoking behaviour will in turn lead to better public health, because of all the health issues that are related to smoking. This framework can be used in further research by filling in the boxes with the issues that are studied. When conducting the researches new variables, mediators and moderators might be found. These can be added to the framework afterwards, to create a more cohesive
15 Figure 1. Conceptual Framework
3. Method 3.1. Material
This study completely relies on the availability of public data, particularly
information on different campaigns and policies. Information on campaigns that are initiated by non-governmental organizations is retrieved through its website and other sources, like social media. Every campaign has its own website and often it is active on multiple social media platforms. Facebook is often the most informative social media platform, since it enables people to post long texts, clips, pictures and many other features. Data about policies implemented by governments is obtained from official national sources, e.g. the website of the government and publications. Additionally, the World Health Organization publishes a lot on tobacco control and gives elaborate analyses of countries’ national policy.
First of all the countries that are being researched have to be selected. A criterion is that all countries involved have a Western culture, for this is necessary for comparison. Nudging will probably not have the same effects in every country. This is the case because between countries exist big cultural differences, that cause people to react and behave
differently (Morgan et al., 2015). Nudging has been designed by Western countries to benefit their own countries and are therefore concerned with and aimed at changing situations in countries with Western cultures. So studying the effects of nudging across different cultures is a whole other area of research, that is not within the scope or aim of this study. Another reason why all countries are Western is because nudging originated from here, mainly from the United Kingdom and Australia. Therefore, most research on nudging has been
conducted in Western countries. Another criterion is that all countries that are studied belong to the same income group, in order to keep the circumstances as equal as possible. Since smoking is quite expensive, lower income populations might react differently to certain policies than higher income populations. For most research is conducted in Western
16 countries and Western countries often belong to the higher income group, this study focuses on high-income countries.
The countries that will be investigated in this research are the United States of
America (U.S.A.), Australia and Sweden. The U.S.A. have been selected because they are the biggest country with the largest population of all Western countries. Also, the U.S.A. has the biggest public health budget. Therefore their campaigns will be the least hindered by
monetary constraints. Australia has been selected because it is a has a large expenditure on public health. The latest available information comes from 2014. In this year Australia spent 155 084 345 AUD (104 363 624 EUR, converted by the currency rate on December 31st 2014) on tobacco control. Aside from that it is one of the pioneers in nudging and has produced a lot of research. A third reason is that Australia has implemented a unique strategy that is now evaluated and considered by many other countries, and is currently under heavy debate in the United Kingdom. Sweden has been selected because it is has an extraordinary low percentage of smokers, namely 5%. It is interesting to investigate whether nudging has contributed to this low number. Naturally all three countries have a Western culture and have a high-income population.
First, data has to be collected regarding the campaigns. In the United States one of the biggest campaigns is Truth. This campaign has its own website through which all necessary information will be collected. Moreover, the different social media platforms on which Truth is active will be analysed and searched for additional information. In Australia the policy of plain packaging will be analysed. This policy has been implemented through national legislation. Therefore this information will be retrieved from the website of The Department of Health of the Australian Government. Also, additional academic studies on the effectivity of plain packaging are cited. Sweden’s legalisation of oral tobacco is also part of national legislation. Since Sweden is part of the European Union and its allowance of oral tobacco forms an exception on the Union’s general tobacco control policy, much information is retrieved from publications from reports of the European Commission. Furthermore
academic articles are used to determine the harmfulness of oral tobacco in comparison with smoked tobacco.
3.2. Nudginess Checklist
The different campaigns will be judged on how much they apply nudging in their
approach. For this a checklist will be used, with criteria that determine whether something is considered to be a nudge or not. An intervention is hardly ever a 100% nudge, many will check for a part of the criteria. Therefore instead of comparing an intervention that is not a nudge with one that is, this checklists is meant to compare the ‘nudginess’ of interventions. The more criteria an intervention checks, the more it resembles a nudge, thus the higher the nudgines’. To make this checklist the literature from the theoretical framework will be used, mainly the book by Thaler and Sunstein (2008), since they invented the term nudging. In particular the following, already mentioned, quote is used to derive the checklist from: “A nudge, as we will use the term, is any aspect of the choice architecture that alters people’s behavior in a predictable way without forbidding any options or significantly changing their
17 economic incentives. To count as a mere nudge, the intervention must be easy and cheap to avoid.” (Thaler & Sunstein, 2008, p. 6). Although not explicitly mentioned in Thaler and Sunstein’s definition of a nudge, a nudge should be deployed with the purpose of improving the chooser’s well-being (Bovens, 2012). Based on this literature, a nudge should meet the six criteria of Figure 2.
Figure 2. Nudginess Checklist
The first criterion is thus that the change in behaviour should be predictable. This is the case because nudging is designed to benefit the chooser’s welfare (Thaler & Sunstein, 2008). If someone is being nudged, but the outcome of that nudge is not predictable, it is not
necessarily a beneficial outcome. Therefore, it should be clear what the outcome of the nudge will be. This automatically leads to the second criterion. A side from it being predictable, that predicted behaviour should make the chooser better off by reducing the chance of the
chooser engaging in risky behaviour (Bovens, 2012). This falls under the paternalistic principle that nudging prescribes, which entails that it is justified for policy makers to influence people’s decisions and behaviour when it makes people’s life longer, healthier and better. (Thaler & Sunstein, 2008). Nudging is not meant as a tool to manipulate people into doing things that is better for others, but as a tool to improve the chooser’s own well-being. A condition for the chooser’s well-being to be improved, is that the choosers agrees that his or her well-being is improved (Thaler & Sunstein, 2008). Third, the intervention must not forbid any choices. This means that for example legislation that prohibits advertisement of tobacco products, which is part of the MPOWER model, does not count as a nudge. This is because nudging relies on the libertarian principle and therefore respects the freedom of choice. When options are forbidden or coerced the freedom of choice is damaged, and therefore does not count as a nudge. The fourth criterion states that no monetary incentives are changed. Nudging is a form of choice architecture, by formulating a choice differently whilst keeping the same options one’s decision can alter. So only the environment in which the choice is made is changed, not the options itself (Bovens, 2012). Also, nudging is cheaper, because it requires less public funding than approaches that do change monetary incentives (Bovens, 2012). Fifth is the principle that a nudge should be easy to avoid. This means that an
Nudginess Checklist
1. the change in behaviour is predictable
2. the change in behaviour improves the chooser’s own well-being
3. no options are forbidden 4. no economic incentives
are changed 5. it is easy to avoid 6. it is cheap to avoid
18 approach that the consumer will be confronted with no matter what, is not a nudge. This could also be linked to the principle of libertarianism. Namely, an intervention that confronts the consumer unwillingly can be considered as intrusive on people’s freedom. By giving people the chance to avoid it the principle of freedom of choice is respected, because any person can choose for him- or herself to avoid the approach or not. Finally the last principle argues that a nudge must be cheap to avoid. This means that people should not need to pay in order to avoid an intervention. For example, when the production of a good is forbidden in a certain country, people have to pay extra to import a similar product from another country. Just like the previous principle, this limits the freedom of choice, because certain choices are made more expensive than others. So all six principles on the checklist somehow descend from the two main principles of nudging: libertarianism and paternalism.
4. Results
Figure 3 summarizes information on multiple facets per country. It states the policy or campaign that is studied, the period it ran and its goal. Background information per country is also provided, like the income group, prevalence of smoking and information concerning the price of cigarettes. Lastly the annual expense on tobacco control in 2014 is given, since no more recent information is available. All information is retrieved from the World Health Organization (2015b).
Country U.S.A. Australia Sweden
Campaign/policy Truth Plain packaging Sweden
Income group High-income High-income High-income
Period 2000 – present 2012 – present Allways
Goal Let the current
generation of teenagers and adolescents end smoking
Decrease smoking No stated goal
Prevalence of daily smoking 14% (2013) 14% (2013) 5% (March 2017) Price of package of 20 cigarettes in international dollars 6.23 10.99 6.5 Total taxes on cigarettes 42.54% 56.76% 68.84 National expenditure on tobacco control in 2014 481.200.000 USD 144.624.111 USD (155.084.345 AUD) 5.946.564 USD (41.000.000 SEK)
19 4.1 United States of America
In the U.S.A. the prevalence of smoking in 2013 was 14% (World Health Organization, 2013). Many measures to tackle smoking behaviour have already been tried. In the period from 1990 to 2014 the cigarette prices were raised by almost 350%. However this was not
completely successful, because the prices of roll-your-own tobacco and pipe tobacco did not increase as much as the cigarette prices. The U.S.A. have signed the Framework Convention on Tobacco Control in 2004 but still has not ratified it signature. This means that they formally have not stated their participation in tobacco control yet (World Health
Organization, 2015b). This might explain why the reports on tobacco control in the U.S.A are not complete yet (World Health Organization, 2015a).
In the USA one of the biggest anti-smoking campaigns is Truth, which has been running since 2000. The aim of the campaign is to let the current generation of teenagers and adolescents not only quit smoking themselves, but to end smoking all together. Their
method is to provide hard facts about the consequences of smoking to teens and adolescents. In this they work together with governmental organizations, among others with the U.S. Food and Drug Administration and Centers for Disease Control and Prevention.
Truth is mainly active in fulfilling their goals by touring across the U.S.A. in orange truth trucks. Every summer these trucks cross the countries to stop at events where teenagers come together, such as music festivals and sport events. Here, the Truth crew engages in one-on-one conversations with teens about the use and consequences of tobacco. They distribute Truth merchandise as well, such as T-shirts and bandanas. During an average summer, the trucks stop at around fifty cities and twenty-five states. Truth also entered a partnership with High School Nation, which is an organization that organizes dance parties for high schools. Through this collaboration the truth truck has a bigger scope of events to attend.
Truth is active on various social media platforms: Facebook, Twitter, YouTube and Instagram. The reach on Facebook is large, over two million people have liked the Truth-Facebook page. If someone likes your page on Truth-Facebook, they will automatically get status updates and other information from your page in their news feed. The posts on Facebook are diverse. Truth develops short commercials that end with a fact and a statement. The
commercials are meant to communicate one specific fact, for example “There are more tobacco ads in black neighbourhoods than in other neighbourhoods”. And it calls for action, for example “Stop profiling”. These clips are posted on the Truth Facebook page. On Twitter Truth has been active since 2009 and here it posts the same clips as on Facebook. However, the reach with Twitter is substantially smaller, because they have little over 100,000
followers. On Instagram the same clips are posted, with a reach of around 45,000 followers. Last, Truth has a YouTube-channel. On this clips of approximately one minute are posted. YouTube has a function to see how many views a clip got, one clip got over nine million views. So through YouTube a lot of people are reached. Thus Facebook and YouTube are the social media platforms with the biggest scope. Apart from social media Truth also has a website on which people can find a good overview of what Truth does. All the events that
20 Truth attends and organizes are on there as well. Via the website it is possible to sign up for exclusive texts from Truth. This are around four texts per month. It is characteristic for the campaign that is has chosen texts instead of a newsletters to fit the target audience better. Likewise, when browsing through the website and social media of Truth, it is apparent that they put effort into addressing the teenagers in an informal and playful manner.
Abbreviations are being used, like “merch”, short for merchandise. Also, words that are usually only used by teenagers are present in the clips and texts.
Truth is funded by Truth Initiative Foundation and Affiliate, previously known as the American Legacy Foundation. This Foundation is directed by state governors, legislators, attorneys general and experts in the medical, education and public health fields. Their goal is twofold: discourage smoking among youngsters and educate smokers about the diseases related to the consumption of tobacco. This is realised through national advertisement and education programs. Truth Initiative allocated 74,005,000 USD to Truth in 2016, which was substantially more than the 66,132,000 USD received in 2015.
Nudginess – How nudgy is Truth?
To determine to what extent Truth can be considered a nudge it will be compared with the previously composed Nudginess Checklist. First of all the behaviour is predictable. With this campaign people will learn the facts about tobacco use. These facts are merely negative. Thus, when smokers encounter these facts their smoking habits will either stay unchanged or decrease. This automatically answers the second criterion, namely that the behaviour should change to the benefit of the chooser. Because people will either keep smoking, smoke less or quit smoking, Truth has only a potential benefit for the chooser’s health. Third, Truth does not forbit smokers any options. They even clearly state that they do not hold anything against smokers and that some of the crewmembers are smokers as well. Truth leaves it to the facts and conversations, further they let people make their own choices. The next criterion demands that no economic incentives are changed. Truth does not offer a monetary rewards on decreased smoking whatsoever. Also, it does not influence cigarette prices. Therefore, no economic incentives are changed. Fifth, it is easy to avoid. The truth campaign mainly consists out of trucks on certain events and social media messages. The trucks are in about fifty cities during the summer, so if you even encounter a truck, one can still avoid it by not interacting with the crewmembers. Last, it is cheap to avoid. Truth does not charge anything for anyone and does not forge itself on smokers. So no costs will be incurred when avoiding this campaign.
4.2 Australia
Australia is the first country in the world to apply plain packaging as a strategy to tackle smoking and therefore a pioneer in tobacco prevention. In 2003 Australia signed the World Health Organization Framework Convention on Tobacco Control and in 2004 this
commitment was ratified (World Health Organization, 2013). The Tobacco Plain Packaging Act 2011 and the Tobacco Plain Packaging Regulations 2011 were accepted in 2011 and implemented in December 2012. The act forbids tobacco packaging from showing logos, brand imagery, and promotional text on the products and its packaging. Also it concludes
21 strict rules about the appearance of the packaging: colour, size, format and materials.
Additionally all packages should bear extensive health warnings accompanied by shocking and repulsive images to illustrate the consequences of smoking. The brand name and/or variant is still featured on the package, on the bottom in white plain fond. So not in the original fond of the brand. All packages are olive green and in the right bottom corner of the package the number of cigarettes is named.
In addition to this policy, cigarettes in Australia are highly priced in comparison with other countries. Starting in March 2014 Australia decided on four 12.5% tax increases on cigarette prices, that were implemented every first of March and September until 2016 (World Health Organization, 2015b). A package of Marlboro now costs 26 AUD (17 EUR). This same package costs 6 EUR (10 AUD) in Amsterdam. With this price, Australia is worldwide the most expensive country when it comes to cigarettes, all compared on the price for a package of Marlboro. These high prices have proven to be effective in decreasing smoking behaviour in Australia (Dunlop, Perez, & Cotter, 2011). Instead of buying cheaper cigarettes or using other sources, around 20% of the smokers tried to quit. A majority of the smokers reported positive changes in their thinking and behaviour towards smoking as well. When checking for the effectivity of plain packaging on decreasing smoking behaviour, it is hard to determine which effects are the results of which intervention, because both
interventions run simultaneously. In 2013 14% of the Australian population were daily smokers (World Health Organization, 2015b).
Instinctively, one can imagine that plain packaging would decrease smoking behaviour in a way. But the exact underlying mechanisms that should cause this decrease require research. Four main reasons have been found to explain the potential effectiveness of plain packaging (Cunningham & Kyle, 1995). The first reason is that the packaging itself is a powerful form of advertising. It communicates the brand and everything it stands for and it functions as a ‘badge’ for the consumer by adhering to a certain image. The role of
advertisement through packaging has increased, since tobacco advertisement is prohibited in many countries. So by eliminating this, the appeal of smoking as part of an image or life style will decrease. Another part of the advertisement explanation concerns the power of previous advertisement. Namely, despite the fact that advertisement is currently prohibited, past advertisement still has an effect, especially on the people that have been smoking for a longer period of time. Plain packaging breaks the connection between the leftover effects of old advertisement in a radical way. Second, the efforts of public education through health warnings on the packaging is more effective on a plain package. Mainly because the noise, like promotional texts or salient features of the packaging, can distract the reader from the message of the warnings. Also, it strengthens the message that tobacco is not any other consumer product and that the risks of consuming it should not be taken lightly. Thirdly, people’s perception of the quality and taste of food, drinks and other consumable goods is influenced strongly by the appearance of those goods. Generic products are often evaluated as being inferior to branded products, or minimally branded products in this case. So, even though the product inside remains the same, people can have a less good experience because of the plain packaging. The last reason is that the number of tobacco brands will decline as a
22 consequence of the limited components to compete on. It will no longer be feasible to have the hundreds of brand variations that the market offers. When people cannot find their favourite product they might consider stopping or at least smoke less. Considering the fact that applying plain packaging will lower the tobacco consumption in combination with the low costs for the government to implement this, it has potential to be a great policy. Another side benefit is that because the government dictates the material of the package, it can opt for a biodegradable material and removal of the plastic foil (Cunningham & Kyle, 1995).
Nudginess – How nudgy is plain packaging?
Plain packaging satisfies the first criterion of the Nudginess Checklist that states that the change in behaviour is predictable, as this policy can only decrease the tobacco
consumption or have no effect at all. Smokers will not smoke more after implementing this policy, since there are no new benefits to smoking. This policy only makes smoking less attractive. The second criterion to serve as a nudge states that the policy should benefit the chooser’s well-being. Because the only possible change in behaviour is less smoking, this criterion is also satisfied. Since less smoking will contribute to better health. The third
criterion, that no options are forbidden, is met as well. Plain packaging does not prohibit any tobacco companies to sell their product in Australia, nor does it prohibit smokers from buying them. Fourth is the requirement that no economic incentives are changed by the policy. Even though Australia has the highest tobacco prices in the world, this is disjointed from the plain packaging policy. Thus, the fourth criterion is also met. However, effects from these two interventions are hard to separate. The fifth and sixth criterion are not satisfied, because the policy is neither easy, nor cheap to avoid. It has been effectuated throughout whole Australia, so avoiding it will require to leave Australia or smuggle cigarettes from other countries, which is not easy. It is not cheap either, because leaving Australia as well as smuggling cigarettes is not cheap, let alone free.
4.3 Sweden
Sweden also signed the WHO Framework Convention on Tobacco Control on June 16th 2003, which was ratified on July 7th 2005 (World Health Organization, 2015b). In 2014, 21% of the Swedish adult population smoked tobacco, and 11% of the population smoked daily. But in march 2017 the number of people of the Swedish population that identified themselves as ‘daily smokers’ had dropped to 5%, which is the lowest percentage of all Member States of the European Union (European Commission, 2017). When looking at the report of the European Commission (2017) Sweden often forms the exception on statistics. For example, it is the country with the biggest proportion of people that have successfully quit smoking, namely 41% (European Commission 2017). Another striking fact is that only a little over halve of the smokers smokes daily in Sweden, which is the lowest number of daily smokers of all EU countries. Also, this number used to be 72% but decreased 20 percentage points, which in turn is the largest decrease of daily smokers in the entire EU. Also, whilst in other EU countries there were no big changes in the proportion of people that smoked boxed cigarettes, in Sweden this number dropped 20 percentage points. In terms of tobacco control efforts, Sweden is again not like other countries. The smoke-free legislation, that has been
23 broadly adopted by numerous countries, is not yet applied in Sweden. This means that smoking is still allowed in restaurants, public transport and even health care facilities (World Health Organization, 2015b). Warnings are mandatory on tobacco products and must cover a minimum of 35% of the rear and front of the packaging combined, these warnings do not include a photograph or other graphic. Advertisement of tobacco is largely prohibited, except for advertising at the point of sale, which signals where tobacco can be purchased. However, sponsorships are not forbidden in Sweden. So in the categories of the World Health Organization’s MPOWER model, there is still a lot of room for progress.
The policy of Sweden that will be evaluated is the legalisation of oral tobacco, called snus. In all Member States of the European Union the use of oral tobacco is prohibited, except in Sweden (European Commission, 2017). When entering the European Union Sweden listed this as a criterion. Therefore this deliberate choice can be seen as policy on smoking behaviour. In 2017 50% of the Swedish population had tried snus, chewing tobacco or nasal tobacco, contrasted to 5% of the inhabitants of all Member States. Daily users of snus in Sweden comprise 20% of the population. For 16% of the Swedish snus was the first
tobacco product that they used. Contrary, there is also a large proportion, 19%, that has used snus, chewing or nasal tobacco in order to quit smoking or in attempt to do so (European Commission, 2017). Snus is relatively cheaper than cigarettes. Daily users of cigarettes smoke on average 13 cigarettes per day and daily users of smoke consume 15 grams of snus a day on average. The relative price of the daily use of cigarettes in relation to snus is 6.65, which means that smoking is 6.65 times more expensive than the use of snus (Bask & Melkersson, 2003).
The remaining question is whether snus is less harmful than smoking. Snus emits a higher nicotine concentration than cigarettes, whilst emitting lower concentrations of other harmful chemicals (Foulds, Ramstrom, Burke & Fagerström, 2003). Meaning that users can get their nicotine kick with less damage to their health. Contrary to smoked tobacco, like cigarettes and cigars, snus does not cause elevated levels of risks of cancer or respiratory disease. It does increase the risk of cardiovascular disease and is harmful to foetuses during the pregnancy (Foulds et al., 2003). However, all combined these risks are lower and less harmful than the ones associated with smoking. Overall, smokeless tobacco is at least 90% less harmful than smoking tobacco (Bates et al., 2003). The risk of smokeless tobacco can be further decreased by legislation on the dose of harmful chemicals in the product (Bates et al. 2003). Evidence that proves that smokeless tobacco is indeed less hazardous to human health is provided by multiple studies. For one thing, there was a decrease in smoking behaviour over the period of 1976 to 2002, with the biggest decrease for the male population. Of the male ex-smokers 30% was now using snus. In this same period the incidence of lung cancer and myocardial infections had decreased more rapidly among men than among women. This suggest a possible positive relationship between snus, quitting smoking and better health (Foulds et al., 2003). Another study shows that the consumption of snus is lowest in the North of Sweden, e.g. Stockholm, Gothenburg and Malmö. In these areas the incidence of head and neck cancer is the highest (Lewin et al., 1998). So there may be a potential negative relationship between the use of snus and the incidence of forms of cancer.
24 Importantly, Swedish snus meets two major conditions to service as a tool for tobacco control. First, it is less hazardous to human health than smoking. Second, it is a possible substitute for smoking and can increase smoking cessation (Bates et al., 2003). An important side note is that other forms of oral tobacco are associated with oral cancer, but Swedish snus is not. With all this evidence provided, one may question why the EU has put a ban on oral tobacco. First of all, the main reason for the ban is that oral tobacco is harmful for people’s health. But since it is proven to be less harmful than smoking and that it is helpful in the cessation of smoking, the question arises whether it is ethical to deny this product to
consumers (Bates et al., 2003). Another reason is labelled as the ‘gateway effect’, which is the concern that people that use snus would otherwise not have smoked if snus was illegal. However, it is impossible to test for such an effect, because one cannot know what the users of smokeless tobacco would have done in the scenario that smokeless tobacco was
prohibited. Also, Sweden has a tobacco prevalence similar to their neighbouring countries Denmark and Norway. This suggests that the legalisation of oral tobacco does not increase the tobacco prevalence, it just influences the type of tobacco that is consumed (Bates et al. 2003). Thirdly, the effects of smokeless tobacco are not yet entirely know. Therefore, some argue for application of the precautionary principle of banning it until all effects are
completely known. A counter argument to this is the evidence provided by Sweden, where the decades long use of snus has not led to increased levels of any diseases. The European Commission needs to revise its decision in the light of scientific development (Bates et al., 2003).
Nudginess – How nudgy is snus legalisation?
First of all the criterion whether the change in behaviour is predictable. This criterion is met, because countries with higher levels of snus consumption have lower levels of
smoking. Also, only a small proportion uses snus complementary to smoking. So legalisation of snus will lead to the predictable behaviour that a proportion of the current smokers will use snus instead of smoking or as a means to quit smoking. The second criterion asks whether the change in behaviour improves the chooser’s own well-being. This criterion is also satisfied, because the consumption of snus is less harmful than smoking. Smoked tobacco is associated with higher risks of cancer and respiratory diseases, but no such relationship has been found with snus. Thus if people use snus instead of smoking it
improves their well-being. Thirdly, a nudge does not forbid any options. For snus this is the case, because it actually adds an option, without removing any other options, since smoking is still legal. Thereafter follows the criterion that no economic incentives are changed. This criterion is not entirely met. Although the price of smoked tobacco remains the same, snus is relatively cheaper in daily use than cigarettes. When thinking about replacing cigarettes for snus, economic considerations could play a role, since snus can be seen as a cheaper
substitute to get the desired nicotine. The last two criteria are that a nudge is easy and cheap to avoid. This is true for snus, because no extra efforts have to be made and no extra money has to be spent to avoid using snus.
25 Criterion U.S.A. - Truth Australia – plain
packaging Sweden - snus 1. the change in behaviour is predictable Yes - if the behaviour changes, people will smoke less or quit
Yes
- if the behaviour changes, people will smoke less or quit
Yes
- higher levels of snus use are related to lower levels of smoking
2. the change in behaviour improves the chooser’s own well-being
Yes
- smoking less and/or quitting is good for one’s health
Yes
- smoking less and/or quitting is good for one’s health
Yes
- snus is less harmful than smoking 3. no options are forbidden Yes - no ban on tobacco products or companies Yes - no ban on tobacco products or companies Yes - no ban on tobacco products or companies 4. no economic incentives are changed Yes - the price of smoking does not change
- no rewards are offered for smoking less or quitting
Yes
- high cigarette prices are disjoint from the policy
No
- snus is relatively cheaper for daily users than smoking
5. it is easy to avoid Yes
- by not interacting with the
crewmembers or avoiding the events Truth attends No - avoidance will require smuggling or illegal import Yes
- people that do not want to use snus are not confronted with it
- unlike smoking, snus is not harmful for bystanders 6. it is cheap to avoid Yes - no costs are associated with avoidance No - smuggling or illegal import is not cheap
Yes
- not using snus does not raise any costs
Score 6/6 4/6 5/6
Figure 4. Overview of scores on Nudginess Checklist for the U.S.A., Australia and Sweden
5. Discussion
This research investigated how nudging is applied as a behavioural tactic to decrease smoking behaviour. The literature showed that nudging has proven to be effective in many public health domains. Tobacco control measure that possess many criteria for a nudge are graphic warnings, plain packaging and visual display bans (Alemanno, 2012). But how nudging are real-life running campaigns and policies? For this purpose the Truth campaign
