LITERATURE REVIEW (12-16 pages) This essay is basically synthesizing 10 sources to support the new research im proposing that needs to be done. The sources needs to be categorized into 3/4 subheadings

Ehrenberg’s ‘‘weak’’ theory suggests advertising will support unhealthy eating beha- viors, while the wide range of sales promotions employed will prompt trial and reward continued consumption. This article presents an alternative analysis of how marketing contributes to obesity and uses behavior modification theory to analyse the ‘‘fast-food’’ industry’s promotions. We also review the New Zealand government’s response to obesity and suggest policy interventions that would foster healthier eating behaviors. Although obesity always has been associated with serious health conditions such as diabetes and heart disease, its increased incidence, particularly among young people, has led health lobbyists to call for a reexamination of the factors that influence young people’s eating patterns (Diabetes New Zealand Inc. and Fight the Obesity Epidemic Inc., 2003). Unprecedented numbers of children now feature in obesity statistics (Cristol, 2002; Ebeling, Pawlak, & Ludwig, 2002; Fight the Obesity Epidemic [FOE], 2004), leading commentators to describe the growth in obesity as an epidemic.

Health lobby groups have responded to the increase in obesity levels by calling on governments to take urgent steps to prevent further escalations in the proportion of obese individuals (FOE, 2004; Hawkes, 2004). In New Zealand, health groups such as Fight the Obesity Epidemic (FOE) and the Obesity Action Coalition have lobbied strongly for a ban on advertising of all ‘‘fast-food’’ items during children’s television programs, among other measures. Analyses of television advertising direc- ted at children have led to concern over the number of ‘‘fast-food’’ advertisements many children see. Zuppa, Morton and Mehta’s (2003) content analyses of Austra- lian advertisements shown during children’s programs revealed that promotions Address correspondence to Professor Janet Hoek, PhD, Department of Marketing, College of Business, Massey University, Private Bag 11-222, Palmerston North, New Zealand.

E-mail: [email protected] Journal of Health Communication, 11:409–423, 2006 Copyright#Taylor & Francis Group, LLC ISSN: 1081-0730 print=1087-0415 online DOI: 10.1080/10810730600671888 409 featuring high-fat, high-sugar, and high-salt items accounted for more than half their sample (see also Hill & Radimer, 1997; Kotz & Story, 1994). These findings have focused attention on the effect food advertisements may have on young people’s behavior.

Not surprisingly, the advertising industry has reacted strongly to suggestions that governments restrict or ban advertising screened during children’s programs and there has been intense media debate over advertisers’ rights and responsibilities.

This debate has raised several issues, including the role played by parents in moderating their children’s behavior, the effects advertising has on behavior, and the multiple factors alleged to contribute to obesity (Ambler, 2004).

Advertising and industry spokespeople argue that children’s food consumption is a matter for parents to determine (Sykora, 2003). Parents, advertisers assert, should guide their children’s media use, help them interpret advertising, explain why certain food groups should be consumed in moderation, and resist requests that would exceed a moderate consumption level. Ellis (2003) summarized this view: ‘‘We shouldn’t get too far away from the key to all this, which is parental responsibility.

Kids might watch the food advertisements, but it’s the parents who do the shopping and who control the environment’’ (p. 2). Parents’ role in guiding their children’s media and good consumption is undeniable. Even the most cautious parents cannot monitor their children’s viewing behavior constantly, however, and the argument that advertisers have the right to communicate with children, while parents must take responsibility for the consequences of these communications, is clearly unbalanced.

Advertisers also claim that food advertising aims to change brand preference and suggest it has little or no effect on the aggregate demand for a product category (Young, 2003). According to this argument, advertising does not create a desire for ‘‘fast food,’’ but reorders the hierarchy of brands from which consumers choose.

Recent research from Hastings, Stead, and McDermott (2004), however, challenges this view and highlights the need to consider the broader effects of food advertising on consumption.

Because the causes of obesity are complex, advertisers have argued that their products and advertising make a negligible contribution to the problem (Divehall, 2003). Instead, they note that people now exercise less, have more sedentary lifestyles, and make greater use of convenience foods (Lvovich, 2003). While dispro- portionate consumption of ‘‘fast foods’’ may contribute to obesity, advertisers argue that the specific role played by these types of food remains impossible to define. As a result, they conclude that measures targeting one component of the problem are both unfair and unlikely to succeed (Young, 2003).

The competing views of health and industry spokespeople appear irreconcilable.

Advertisers demand evidence of a causal relationship between food advertising and obesity before they will accept that regulatory intervention may be warranted. Social science research can never establish causality with certainty, however, and instead provides estimates of the probable relationship between two variables. Because many inter-related factors, including poor nutritional habits, overeating, and reduced levels of physical activity, all contribute to obesity, it is difficult to quantify the pre- cise role played by each of these variables. Deferring action until a specific causal relationship between advertising and obesity levels has been established would over- look actions that could be taken using knowledge of how advertising supports and maintains behavior. 410 J. Hoek and P. Gendall In this article, we thus examine an alternative means of framing the debate and explore the role advertising plays in reinforcing behavior patterns. We suggest that food advertising is not innocuous and that, in depicting frequent consumption and consumption of larger portion sizes (‘‘up-sizing’’) as normal behaviors, it contributes to the rise in obesity that nations around the world have experienced. We begin by examining the arguments presented by health lobbyists and the advertising and fast- food industries before analysing different perspectives on advertising and how it affects consumers’ behavior. We conclude by discussing a public policy response to obesity, the New Zealand Healthy Eating, Healthy Action Plan, and suggesting additional measures more likely to shape desirable behavior. The Advertising–Obesity Debate Parental Control Though advertisers and health lobbyists agree that childhood obesity figures warrant serious attention, they disagree over the action that ought to be taken to address the problem. Advertisers have argued strongly that individuals need to take greater responsibility for their own behavior and the behavior of those for whom they are responsible, and have claimed public opinion supports their stance. For example, Spungin (2004) noted that 80% of parents surveyed said their children had asked them to purchase a specific brand. Less than half of these, however, would examine the nutritional value of the product before deciding whether to buy it, although only 14% would allow their children to try a new product without vetting it in some way.

Despite these rather discrepant findings, Spungin concluded that parents willingly act as gatekeepers who review the appropriateness of their children’s requests. Even putting to one side the equivocal status of these results, Spungin’s evidence applies only to new products and does not offer insights into how parents respond to adver- tising for products they or their children have previously consumed.

Spungin (2004) also found that only 12% of parents believed advertising to children should be banned, while 45% believed advertisements were a commercial reality. Because Spungin did not disclose either the actual statement wording or the full response distribution, however, it is difficult to tell what parents responded to in his survey. That is, it is not clear whether respondents disagreed with a ban on advertising directed at children, or whether they were simply ambivalent about this prospect. Surveys based on simple attitude statements do not adequately test the range of regulatory interventions possible, and are not a sufficient basis for con- cluding that no further action is necessary. Furthermore, because it is difficult for parents to reject responsibility for controlling young children’s access to fast-food items, social desirability error will inflate the proportion of respondents who agree with statements that parents are primarily responsible for controlling what their children eat.

While it is reasonable to expect parents to exert control over their children’s con- sumption behaviors and the with which media they engage, the advertising industry appears reluctant to accept that fast-food advertising is likely to make parents’ tasks more complicated than would otherwise be the case. Although the New Zealand Bill of Rights affords some protection to commercial speech, it is important to balance advertisers’ rights to promote products against parents’ rights to raise their children in an environment that is conducive to healthy eating behaviors.Advertising and Obesity 411 Confounding Factors Few contributors to the debate would dispute that obesity is a complex problem, created through the interaction of a range of factors, of which food advertising is only one. Changed lifestyle patterns involving a reduction in physical exercise are also likely to contribute to obesity, particularly when coupled with an increase in energy intake. Although researchers agree that obesity represents a complex pro- blem, however, the individual factors that contribute to it may still be addressed through voluntary industry initiatives or policy changes. The debate over the effects of tobacco promotions and the relationship between bans on these and overall con- sumption highlights the difficulty of defining the precise contribution made by each factor. More seriously, efforts to quantify the specific role played by each factor risk deflecting attention away from the more important task of identifying and evaluat- ing options that could reduce the current levels of obesity.

Advertising—Caused or Causal?

Advertisers, and those who oppose a ban on advertising of fast foods, have argued that there is no causal link between advertising and the consumption of fast food (Young, 2003). For example, Young and Webley (1996) argued claims of causality are mistaken, because research into the effects of fast-food advertising is rarely longitudinal or experimental. Advocates of tobacco advertising bans, however, have argued establishing causality is an impossible goal. Hastings, Aitken, and Mackintosh (1994) suggested it required that: ‘‘We would not pronounce any link between smoking and lung cancer untilallsmokers died of the disease and allits victims were smokers’’ (p. 197).

Because they believe there is no causal link between advertising and aggregate consumption, opponents of advertising restrictions argue these would have little effect. They claim that advertising changes brand preferences within mature markets and does little to generate primary demand for a product category (Ambler, 2004).

Put simply, advertising persuades consumers to buy one brand rather than another, thereby increasing that brand’s market share, but does not increase overall consump- tion of a product category. As one commentator has noted, advertising shampoo is unlikely to make people take more showers to wash their hair more often, though it may prompt them to use Brand A instead of Brand B (Calfee, 2003).

These arguments parallel the logic relied upon by the tobacco industry. Cigarette manufacturers fought a rear-guard action to stave off advertising restrictions by arguing that their promotions had no effect on nonsmokers and sought only to change brand preferences among existing smokers. Like fast-food manufacturers, tobacco companies claimed that new recruits were attracted to smoking not by advertising, but by the behavior of their peers and other social groups (Smith, 1990). The role tobacco promotions play in smoking initiation remains disputed.

Evidence relating to food advertising, however, suggests it may affect primary demand. Hastings and colleagues’ (2003) comprehensive and systematic review of research examining food advertising found only weak evidence of brand switching and much stronger evidence of category switching (p. 19).

Despite the growing evidence to the contrary, the advertising industry conti- nues to argue that advertising does not affect overall market size. This view leads logically to the conclusion that additional regulation of food advertising is overly maternalistic and unnecessary. Some researchers have drawn parallels with tobacco 412 J. Hoek and P. Gendall regulation; they suggest that the alleged failure of tobacco advertising bans to reduce tobacco consumption means food advertising restrictions would be ineffective (Ambler, 1996; Boddewyn, 1989; 1994).

This argument overlooks the highly addictive nature of smoking however, and erroneously assumes the removal of a weak reinforcer such as advertising can pro- duce immediate short-term results (Hoek, 1999). High fat and sugar foods do not create a physiological addiction in the same way as nicotine does; thus, removal of advertising could produce more immediate results. At the very least, it would eliminate a reinforcer that supports continued consumption of fast foods. Restric- tions on food advertising also would reduce the social approval that advertising confers on a product category.

In summary, arguments that advertising of ‘‘fast foods’’ affects only brand pre- ference, and that restrictions are therefore unwarranted, rely on flawed logic because they assume that advertising works only through persuasion. Although advertising can prompt behavior, its main role is to reinforce and maintain existing behavior patterns. The view of advertising as a reinforcer rather than as a cause of behavior aligns strongly with operant conditioning and a more behavioral theory of advertis- ing. Ehrenberg (1974) first applied behavioral learning theory to advertising and developed what has become known as the ‘‘weak’’ theory of advertising; we examine this theory in the following section before using it to explore the advertising–obesity debate.

Ehrenberg’s ‘‘Weak’’ Theory of Advertising For more than three decades, Ehrenberg and his colleagues have questioned the view that advertising works by persuading consumers. In a much-cited comment, he argued:

‘A minor actress pretending to drink a cup of Nescafe in a TV commercial may marginally encourage someone who usually buys another brand to do so again. But it is unlikely to make someone who usually buys another brand to switch to Nescafe. It is even less likely to persuade a non-coffee drinker to start drinking coffee.’ (Barwise & Ehrenberg, 1988, 167) Ehrenberg (1974) challenged hierarchy-of-effects models of advertising and their emphasis on persuading consumers to adopt new behaviors. Instead, he argued that because advertising is noticed primarily by users of the promoted brand, it reinforces existing behavior patterns and maintains a brand’s salience among its users, most of whom also regularly purchase competing brands. His model originally included three components: awareness, trial, and reinforcement; advertising’s primary role occurred in the reinforcement stage. (Barnard and Ehrenberg, 1997, extended the model by conceding that advertising could ‘‘Nudge’’ consumers’ behavior.) This model differs sharply from the traditional view of advertising as a persuasive force capable of instilling new behavior patterns in consumers, and led Ehrenberg to describe his approach as the ‘‘weak’’ theory. Ehrenberg’s model suggests advertising is a form of operant conditioning, and his views fit logically with behavior modification theory (Nord & Peter, 1980).

At first glance, Ehrenberg’s view of advertising appears to support the advertis- ing industry’s contention that its promotions are unlikely to persuade people toAdvertising and Obesity 413 perform new behaviors, since Ehrenberg, like industry proponents, rejects the notion that advertising persuades. However, Ehrenberg’s argument that advertising maintains behavior should not be interpreted as minimising its potential influence on obesity. By supporting the continuation of unhealthy behavior patterns, advertis- ing reduces the likelihood that individuals will either recognise the behaviors as unhealthy or seek to change these.

Interestingly, industry attempts to minimise the effects of ‘‘fast food’’ advertising have highlighted its potential to reinforce behavior. For example, Fraser (2003) stated: ‘‘So if television commercials promote Pepsi, fruit loops, burger rings and snickers bars they’re not leading kids into new and wicked temptations. They’ve always bought this kind of stuff.’’ Fraser’s comments imply that reinforcement is innocuous and support his contention that attempts to restrict it are unwarranted.

By reminding users to consume ‘‘fast foods’’ and by providing extra incentives to do so, however, advertising may contribute to excessive consumption in a variety of ways. For example, bundling promotions combine several menu items and foster beliefs that ‘‘up-sizing’’ is both sensible (it represents better value) and normal (why else would super value combos exist?). Overall, advertising of fast foods maintains the impression that consumption of these products is consistent with a healthy diet, but offers little or no guidance about what is required to achieve a balanced and moderate food intake.

Acceptance of Ehrenberg’s arguments raises the obvious question of how beha- vior develops, if not through advertising. As with the initiation of smoking, it is likely that unhealthy eating behaviors initially develop in an environment where that behavior is accepted, viewed as normal, or even promoted as desirable. Just as chil- dren whose family and peers smoke are more likely to develop a smoking habit than those from smoke-free environments, so children who regularly eat high-fat, high- salt, and high-sugar foods will come to view these as part of a ‘‘normal’’ diet.

Although Ehrenberg’s model of advertising emphasises repeat purchase, which he views as critical to the long-term success of a brand, the ‘‘Trial’’ component recog- nises the potential for sales promotion activities to attract new users. Prominent sales promotions, such as price discounts, product bundling, and loyalty programs, stimu- late and reward purchase, and may encourage more frequent and heavier consump- tion. Similarly, invitations to ‘‘upsize’’ for a trivial increase in price, delivered at the point of purchase, actively promote larger portion sizes and higher energy intake.

This aspect of Ehrenberg’s model fits logically with Nord and Peter’s (1980) descrip- tion of respondent conditioning, which occurs when a stimulus precedes a particular behavior and increases the likelihood of that behavior occurring.

The relationship between the environment and behavior highlights the need for adults to shape the development of healthy eating habits in children. Parents’ task is complicated however, by the on-going presence of promotions that support quite different eating habits. Removal of fast-food advertising and sales promotions would eliminate environmental factors that support food consumption habits that, in excess, lead to obesity. Yet, just as removal of tobacco promotions has not eliminated smok- ing, removal of fast-food advertising will not eliminate obesity. To counter obesity, those responsible for health strategies need to develop programs that change indivi- duals’ environment so that healthier lifestyles become easier to follow.

Both health lobbyists and advertising and food industry representatives recognise the need to change unhealthy eating behaviors, and both have developed initiatives to address this goal. For example, the advertising industry has proposed 414 J. Hoek and P. Gendall developing an education program and already has launched a campaign featuring ‘‘Willie Munchright,’’ a cartoon character owned by McDonald’s but used without branding to promote healthier eating habits. The industry’s reliance on advertising to promote healthy eating behavior, however, seems somewhat at odds with their argument that advertising does not cause behavior. While they assert that advertising does not persuade nonconsumers of fast food to begin consuming this type of food, they appear to believe it may persuade consumers to moderate their intake and begin a regular exercise program.

According to Ehrenberg’s theory, such a campaign is highly unlikely to motivate people who do not eat moderately to change their behavior. Instead, it will reassure those who do maintain a careful diet that they have made sensible and healthy life- style choices. Similarly, advertising alone is unlikely to prompt individuals to adopt a regular exercise program, although it will remind those who do exercise of the benefits they will receive by continuing with this behavior. Although maintenance of beneficial behaviors is important, however, different interventions are required to change unhealthy eating and lifestyle habits. We examine these issues in the following section, where we analyse the New Zealand government’s Healthy Eating, Healthy Action Plan, and outline an alternative intervention agenda based on behavior modification theory (BMT).

A New Intervention Agenda In discussing social marketing, Rothschild (1999) noted that interventions may occur at the individual, community, or social levels, and could be stimulated or reinforced by education, marketing, or the law. Like Ehrenberg, Rothschild’s approach also cor- responds to BMT, which suggests many marketing decisions could be simplified by analysing how environmental factors shape behavior (see also Nord & Peter, 1980).

This perspective stresses the need to pay more attention to stimuli within consu- mers’ choice context. At present, less attention is paid to consumers’ environment than is devoted to their thought processes. For example, many education and social marketing campaigns provide information in the belief that this will increase aware- ness of an issue, improve attitudes toward the desired behavior, and eventually lead to the behavior itself. Such an approach relies on the view that changing consumers’ beliefs and attitudes will promote changes in their behavior. By contrast, behavior- ists eschew the need to understand consumers’ cognitive processes and focus instead on the environmental stimuli that prompt and reinforce their behavior patterns (Foxall & Greenley, 2000).

Adopting a behavioral perspective implies that health-related interventions may occur at several levels. In the first instance, behavioral interventions may occur at an individual level; for example, offering an incentive, such as free nicotine patches or condoms, increases the proportion of individuals who decide to quit smoking or practice safe sex. The availability of specific stimuli such as these triggers behavior by introducing new factors to an environment. Once behavior has occurred, it can be maintained by applying reinforcers. These may be direct, such as the continued availability of the initial stimulus, or indirect, such as advertising that promotes the desired behavior.

As Rothschild noted, interventions may also occur at a community level. Exam- ples of these include participation by schools in ‘‘healthier food’’ programs, a decision that may be influenced by policy changes that direct schools to establish specificAdvertising and Obesity 415 canteen guidelines. By limiting the availability of less healthy food choices, schools can directly influence what children who purchase food from school canteens may consume. Other initiatives include ‘‘walking bus’’ schemes designed to provide super- vised care as children walk to and from school. Although these interventions provide new opportunities for individuals to engage in healthier behaviors, however, they do not directly reinforce those individuals by rewarding them for making healthier choices.

Behavior changes also may occur in response to policy developments, the third area of intervention Rothschild identified. Laws and policy initiatives constrain the behaviors a society may exhibit by controlling the ease with which these may be performed, the rewards that apply to them, and the disincentives used to discourage unwanted behaviors. For example, the decreasing tolerance of driving while under the influence of alcohol has arguably resulted from policy changes that have reduced the legal blood–alcohol limit, increased the level of enforcement and thus the likeli- hood of detection, and raised the fines applied to those convicted of drink driving offences. Similarly, both the penetration of smoking and the level of tobacco con- sumption appear likely to decrease following legislation that prohibits smokers from smoking in bars and restaurants (Longo, Johnson, Kruse, Brownson, & Hewett, 2001). Marketing campaigns that support the resulting new behaviors reinforce these legal interventions, maintain their salience, and support compliance.

Implementation of policy creates the environment within which behavior occurs.

As a result, it is logical to consider social and legal interventions, since these deter- mine many of the reinforcers that shape the behaviors of interest. Eggers and Swinburn (1997) also argued that initiatives designed to address obesity need to shift away from viewing obesity as an individual disorder to considering it as a ‘‘normal response to an abnormal environment’’ (p. 477). That is, they suggested researchers focus more attention on the range of advertising and promotional stimuli present within individuals’ consumption environment and that foreground less healthy eating behaviors. Although opponents of advertising restrictions have argued strongly that individuals should take more responsibility for their own behavior, Eggers and Swinburn point out that individuals’ ability to make healthy choices decreases when their choice environment actively promotes less health options.

Yet despite the growing calls for greater attention to be paid to environmental factors, most behavior change programs begin with what is loosely described as social marketing. This type of marketing usually attempts to educate individuals, in the apparent belief that greater knowledge or awareness of an issue will lead to behavior change. Programs such as ‘‘Push Play’’ (which encourages people to exer- cise for at least 30 minutes each day) and ‘‘Five Plus’’ (which promotes consumption of five or more servings of fruit and vegetables each day) rely on individuals recog- nising the future benefits associated with these behaviors. A recent Cancer Society survey, however, found nearly half of New Zealand adults do not consume recom- mended quantities of fruit and vegetables, even though more than two thirds recog- nize the importance of these items in a healthy diet (Sullivan et al., 2004).

These findings are consistent with Ehrenberg’s (1974) argument that advertising alone is insufficient to instill new behavior patterns and that, at best, it will only maintain the salience of existing behaviors. The New Zealand Cancer Society research also suggests that awareness and knowledge do not necessarily lead to behavior. Overall, these findings do not augur well for programs that rely heavily on educating people about the desirability of alternative behaviors. 416 J. Hoek and P. Gendall Obesity Reduction Plans Analysis of the goals set out in the New Zealand Ministry of Health’s Healthy Eating: Healthy Action Implementation Plan (HEHA Plan) suggests these empha- size education, awareness, and knowledge, while placing considerably less emphasis on the environmental factors that affect eating behavior (Ministry of Health, 2004).

Reviews of other obesity management plans indicate that these also have education as a dominant goal, although many suggest regulatory changes that might be imple- mented (Bureau of Chronic Disease Prevention, 2001; CDC, 2005).

While the HEHA plan represents an important statement about the New Zealand government’s recognition of obesity as a critical health issue, we suggest plans designed to reduce obesity will be more successful if they have an explicit behavioral orientation. This implies that policy interventions designed to create an environment where unhealthy eating behaviors become more difficult to perform and maintain logically precede education and awareness raising campaigns. In Table 1, we summarize the goals and objective of the HEHA Plan before suggesting how these might be better aligned with social marketing and advertising theories.

Although the HEHA goals acknowledge the importance of environmental changes, the focus of these remains very much on individuals and on how they might be persuaded to adopt healthier behavior patterns. For example, the HEHA plan highlights the development of communication plans, training programs, and other measures designed to increase knowledge and disseminate information. While a more informed populace and more knowledgeable influencers may help increase awareness of the importance of healthy eating and exercise behaviors, this informa- tion alone is unlikely to stimulate behavior change that could have a material effect on obesity levels. As Ehrenberg (1974) suggested, advertising is not powerful enough to induce new behavior patterns, especially (as Rothschild, 1999, noted) when it offers no direct benefit to trigger or reinforce the new behavior.

Reducing the level of obesity requires interventions that begin at a societal level to create an environment in which social marketing efforts are more likely to suc- ceed. Ehrenberg has argued that changes in awareness or attitudes are not necessa- rily predictive of behavior change and may even follow (rather than precede) behavior (Ehrenberg, 1974; Ehrenberg, Barnard, & Scriven, 1997). This reasoning implies that marketing will support existing behaviors, yet not be powerful enough to alter these or to promote the adoption of entirely new behaviors. To bring about more fundamental changes in behavior, it is thus necessary to consider the broader environmental factors that shape behavior. In particular, consumers’ environment should foreground healthy behaviors and reduce the ease of performing behaviors associated with less healthy outcomes. For these reasons, we propose inverting the traditional approach andbeginningwith changes to regulations and public policy, rather than with campaigns directed at individuals.

Policy and Regulatory Interventions Given that advertising theory suggests sales promotions trigger and reinforce behavior, and social marketing theory proposes individuals exhibit higher levels of behavior when offered additional benefits, we suggest regulatory attention turn first to these activities. Specific initiatives could include regulating the development of merchandise based on children’s television programs or movies as gifts or compo- nents of continuity programs. Currently, many fast-food retailers pair movie-basedAdvertising and Obesity 417 Table 1.Healthy eating healthy action implementation plan objectives HEHA objectives Actions Build healthy public policy Adoption of WHO Global strategy; database of stakeholder groups; establish steering group across relevant sectors.

Exploration of policy options; encouragement of sector links; review of nutrition and physical activity policies; Development of joint planning and policy processes Create supportive environmentsInvestigate advertising and social marketing options; Fruit in Schools pilot study; expand breakfast programs; encourage use of healthier cooking products among the food industry; monitor availability of healthy food choices; develop walking and cycling strategy; develop programs for older people and those with disabilities.

Strengthen community actionSupport development of diverse community programs; ensure information shared.

Develop personal skills Extend education and training programs and opportunities; enhance teachers’ ability to deliver curriculum; enhance knowledge of food industry and physical activity industry workers.

Reorient health services Develop list of evidence-based weight loss programs; identify and fill gaps in these; develop training guidelines for obesity service providers; monitor and extend Green Prescriptions; integrate HEHA strategy into district planning; increase promotion of nutrition in primary care settings; ensure meaningful Maori participation.

Monitor, research, and evaluateDevelop research agenda and link with relevant funding bodies; initiate research; monitor information on health status, nutrition, and obesity; improve evaluation of nutrition and physical activity programs.

Communication Develop communication plan relating to nutrition and media strategy; evaluate plan; develop social marketing to facilitate behavior change; track effectiveness of campaign; develop HEHA brand.

Workforce Assess needs of workforce, particularly Maori and Pacific workforce; develop training programs and strategies. Include nutrition and physical activity in relevant curricula. Increase capacity of Maori and Pacific health professionals. 418 J. Hoek and P. Gendall giveaways with specific products; this provides an incentive to purchase the meal item linked to the gift, and to continue purchasing until a complete collection has been acquired. Because each range is available for a limited time only, the promo- tions prompt frequent consumption as this ensures access to the full range of items (Schlosser, 2001). Although it is normally very difficult to change the rate at which consumers purchase a product (Ehrenberg, 1988), these promotions may shorten consumption cycles by applying time constraints that do not affect other purchase behaviors.

Retailers bundle promotions, which means they combine together items on a menu and offer these at a reduced price. This may promote multiple purchases of an item where purchase of several items becomes better value than purchase of an individual item (e.g., $5 for 2 items instead of $3.95 each). Bundling may also cross-sell related products by offering a price discount when the items are purchased together (e.g., $5.95 for a hamburger, chips, and soft drink instead of a summed unit price of up to $2 more). Further encouragement to purchase multiple items occurs at the point of sale, where staff invite consumers to take advantage of the deals currently on offer.

Use of sporting role models by fast-food manufacturers strengthens the associa- tion between healthy activities, such as exercise, and the consumption of foods high in fat, salt, and sugar. Because young people may identify strongly with role models, associations that pair ‘‘fast-food’’ items with individuals they esteem poses a risk that health messages relating to these foods will be discounted (Hoek, Gendall, & Stockdale, 1993). The differences between most young people’s energy requirements and those of high-performance athletes, such as Olympic Gold medalists Sarah Ulmer and Hamish Carter (both of whom have featured in McDonalds’ promotions) are never mentioned in promotions. Sponsorship’s pairing of role models’ attributes with brands may create the misleading impression that consumption of high-fat, high-sugar, and high-salt items is consistent with peak sporting performances.

The frequency with which these sales promotions and sponsorships are used encourages regular purchases of ‘‘fast-food’’ items, provides on-going reinforcement of these purchases, and relies on associations that may mislead young people about the composition of a healthy diet. Regulating the range and types of sales promo- tions that can be used to foster sales of fast foods could remove powerful purchase incentives and rewards, thereby enabling other food options to assume a higher place in consumers’ choice hierarchy.

Other policy changes could include regulation of the food options available through school canteens and the sponsorship–supply arrangements schools can enter into to supplement their operations grants. As noted, constraining food choice within schools modifies the behavior children can exhibit. At present, however, schools with insufficient funds enter into supply contracts with, for example, soft- drink manufacturers, who install vending machines and donate a proportion of sales revenue to host schools. These arrangements provide guaranteed access to a captive market and, because scholars authorize placement of these machines, they convey a clear message that schools approve of the food items sold. Regulations that limit both the type of food schools may sell and the sponsorship contracts they may enter into would also create an environment that fostered healthier food choices.

In summary, the emphasis on regulatory and environmental change we propose would begin by exploring restrictions on the promotions used to support ‘‘fast-food’’ sales. More specific regulations could reduce the salience of these products and limitAdvertising and Obesity 419 the rewards currently provided to prompt or reinforce purchase. According to Ehrenberg and Rothschild’s reasoning, advertising restrictions would remove rein- forcers that promote and reward regular consumption of up-sized product options, and make this behavior appear consistent with a healthy diet. Stricter regulation of school supply arrangements would reduce opportunities for consumption and so make it easier for healthy eating habits to develop. Required modifications to man- ufacturing processes would ensure healthier food choices, irrespective of consumers’ willingness to seek these out. Changes such as these would promote healthier eating patterns, create an environment where community-level activities will be more likely to succeed, and ensure individuals would be more receptive to social marketing messages about healthy nutrition and exercise programs.

Community-Level Interventions The HEHA Plan, and others like it, outline some excellent interventions operating at the community level, such as the ‘‘Fruit in Schools’’ program run successfully in late 2004. The key advantage offered by this program was its intrusiveness and explicit location within schools themselves. Children’s food choices expanded to include a new, low-risk and interesting option that many took the opportunity to try. This direct intervention provided a key opportunity to stimulate new behaviors and, through repetition, to encourage the establishment of a behavior pattern. The targetting of this particular program also recognised existing discrepancies in the availability of healthier food options, and thus contributed to the overall equity of the implementation program.

The ‘‘Fruit in Schools’’ and other initiatives could succeed by offering new and different food choices, and also by reinforcing these through associations with posi- tive role models, prizes, and other reinforcers. The tactics used so successfully by fast-food manufacturers could increase the adoption and maintenance of healthier behaviors. Community-level interventions need to provide alternative choices and associate clear benefits or reinforcers with these. Simplifying the ease with which new behaviors can be performed, and ensuring these are more rewarding than alter- native behaviors, will help establish new patterns that advertising campaigns directed at individuals can reinforce.

Individual-Level Interventions Policy environments that support community-level social marketing programs will also create the conditions where individuals are more likely to be responsive to education programs such as ‘‘Five Plus’’ or ‘‘Push Play.’’ In Table 2, we invert the traditional social marketing approach and outline an alternative program that begins with regulatory change and concludes with education campaigns aimed at individuals.

Conclusions In addressing obesity, policymakers have directed their efforts primarily at creating more informed individuals, without first having created a regulatory environment that will support social marketing and education programs. Behavior modification theory and advertising theory suggest that social marketing and education 420 J. Hoek and P. Gendall campaigns will achieve limited results since they are rarely powerful enough to alter established behavior patterns. Both Ehrenberg’s ‘‘weak’’ theory of advertising and BMT suggest the key to behavior change lies in consumers’ environment. Using policy to decrease the visibility of less healthy foods, while also making these more expensive or more difficult to access, reduces the salience of these foods and dis- courages consumption. Overall, policy initiatives seem very likely to increase the probability that behavior change will occur.

Although logic and theory support the introduction of additional regulatory measures, policymakers typically have two concerns about proposals designed to restrict marketing activities. First, they are sensitive to accusations of state matern- alism and, second, they are concerned that policy changes may not reduce unhealthy behaviors. Future research could explore both concerns. Analyses of public opinion could estimate the likely support for stricter policy measures. Recent studies suggest a public mandate does exist (King’s Fund 2004) although each jurisdiction contem- plating policy changes would wish to ascertain its own constituents’ opinions.

Experimental work, such as stated choice preference experiments, could explore the attractiveness of specific promotional activities, such as bundling and the effect of celebrity endorsement, on choice behavior. Where these promotions significantly alter the overall utility of particular food items, voluntary restraint or further regula- tion would appear warranted.

Such research could clarify both the support for regulatory intervention and the types of interventions most likely to change behavior, and would provide clear guidance to policymakers. Empirical evidence would complement the theoretical rationale for stricter regulation, which in turn would support social marketing and education initia- tives and increase the probability that these will support healthier eating behaviors.

Table 2.Obesity prevention efforts Level Strategies Regulatory structure Regulate school sponsorship–supply arrangements Develop school food sale regulations (U.S. National School Lunchtime Program) Institute stricter food preparation standards Regulate the use of loyalty and bundling promotions Introduce more specific rules relating to youth role models Ensure full and meaningful disclosure of nutritional details Community level ‘‘Fruit in Schools’’ ‘‘Walking bus’’ ‘‘Good employer’’ Workplace exercise and fitness programs Heart Foundation ‘‘Best Fish and Chip’’ Competition Use of fruit rather than sweets as reinforcers in schools Support safe cycling and walking routes ‘‘5-2-1-GO’’ (U.S. SCHOOL-BASED food, television, and exercise program) Individual level ‘‘Push Play’’ ‘‘Five Plus a Day’’ ‘‘VERB’’—It’s what you doAdvertising and Obesity 421 References Ambler, T. (1996). Can alcohol misuse be reduced by banning advertising?International Journal of Advertising ,15 (2), 167–174.

Ambler, T. (2004). Do we really want to be ruled by fatheads? International Journal of Adver- tising and Marketing to Children ,5(2), 25–28.

Barnard, N. & Ehrenberg, A. (1997). Advertising: Strongly persuasive or nudging? Journal of Advertising Research ,37 (1), 21–32.

Barwise, P. & Ehrenberg, A. (1988). Television and its audience. London: Sage Publications.

Boddewyn, J. (1989). There is no convincing evidence for a relationship between cigarette advertising and consumption. British Journal of Addiction,84 , 1255–1261.

Boddewyn, J. (1994). Cigarette advertising bans and smoking: The flawed policy mechanism. International Journal of Advertising ,13 , 311–332.

Bureau of Chronic Disease Prevention. (2001). The obesity epidemic in Florida. 2001 Report.

Retrieved June 20, 2004. from www.doh.state.fy.us Calfee, J. (June 20, 2004). Can advertising make you fat? American Enterprise Institute Presen- tation.

Cristol, H. (2002). Trends in global obesity. The Futurist,36 (3), 10.

Diabetes New Zealand, Inc. and Fight the Obesity Epidemic, Inc. (2003). The POD Report. Prevention of obesity and type 2 diabetes in New Zealand children. The facts . Retrieved June 20, 2004, from http://foe.org.nz/archives/pod_report.doc Divehall, D. (2003, September 2). Fast food industry perspective. Childhood Obesity Symposium, Massey University at Albany. Retrieved June 20, 2004, from http://childhoodobesity.

massey.ac.nz/presentations/RestaurantBrands.pdf Ebeling, C., Pawlak, D., & Ludwig, D. (2002). Childhood obesity: Public health crisis, com- mon sense cure. Lancet,360 , 473.

Eggers, G. & Swinburn, B. (1997). An ‘‘ecological’’ approach to the obesity pandemic. British Medical Journal ,315 , 477–480.

Ehrenberg, A. (1974). Repetitive advertising and the consumer. Journal of Advertising Research ,14 , 25–34.

Ehrenberg, A. S. C. (1988). Repeat–Buying: Facts, theory and applications . London: Charles Griffin & Company Limited.

Ehrenberg, A., Barnard, N., & Scriven, J. (1997). Differentiation or salience. Journal of Adver- tising Research ,37 (6), 7–15.

Ellis, G. (2003, September 2). Speaking Notes. Childhood Obesity Symposium, Massey University.

Fight the Obesity Epidemic. (2003). The POD Report. Retrieved June 20, 2004, from http:// www.foe.org.nz/archives/pod_report.doc Fight the Obesity Epidemic (FOE). (2004). Our children are among the fattest in the world .

Retrieved June 20, 2004, from http://www.foe.org.nz/facts2.html Foxall, G. & Greenley, G. (2000). Predicting and explaining responses to consumer environ- ments: An empirical test and theoretical extension of the Behavioral Perspective Model.

Services Industry Journal ,20 (2), 39–54.

Fraser, I. (2003, March 5). Food for thought—The broadcasters’ response . Advertising Standards Authority Food Symposium. Retrieved March 20, 2003, from http://www.

asa.co.nz/food_symposium/ian_fraser.htm Hastings, G., Aitken, P. P., & Mackintosh, A. M. (1994). Children, smoking and advertising: The evidence is there for those who wish to see it (A reply). International Journal of Advertising ,13 , 209–215.

Hastings, G., Stead, M., & McDermott, L. (2003). A review of research on the effects of food promotion to children . Report prepared for the Food Standards Authority by the Centre for Social Marketing, Strathclyde University. Retrieved September 30, 2003, from http://www.ism.stir.ac.uk/pdf_docs/final_report_19_9.pdf Hawkes, C. (2004). Marketing food to children: The global regulatory environment . World Health Organisation Report, Switzerland.

Hill, J. & Radimer, K. (1997). A content analysis of food advertisements in television for Australian children. Australian Journal of Nutrition and Dietetics ,54 , 174–81. 422 J. Hoek and P. Gendall Hoek, J. (1999). Effects of tobacco advertising restrictions: Weak responses to strong mea-sures? International Journal of Advertising ,18 (1), 23–40.

Hoek, J., Gendall, P., & Stockdale, M. (1993). Some effects of tobacco sponsorship advertise- ments on young males. International Journal of Advertising ,12 (1), 25–37.

King’s Fund (2004). Public attitudes to public health policy . King’s Fund Publication.

Retrieved 15 August, 2004, from http://www.kingsfund.org.uk/resources/publica- tions/public_attitudes.html Kotz, K. & Story, M. (1994). Food advertisements during children’s Saturday morning television programming: Are they consistent with dietary recommendations? Journal of the American Dietetics Association ,94 , 1296–1300.

Longo, D., Johnson, J., Kruse, R., Brownson, R., & Hewett, J. (2001). A prospective investi- gation of the impact of smoking bans on tobacco cessation and relapse. Tobacco Control, 10 , 267–272.

Lvovich, S. (2003). Advertising and obesity: The research evidence. International Journal of Advertising and Marketing to Children ,4(3), 35–40.

Ministry of Health. (2004). Healthy eating, healthy action implementation plan . Retrieved June 25, 2004, from http://www.moh.govt.nz Nord, W. & Peter, J. P. (1980). A behaviour modification perspective on marketing. Journal of Marketing ,44 , 36–47.

Rothschild, M. (1999). Carrots, sticks, and promises: A conceptual framework for the man- agement of public health and social issue behaviors. Journal of Marketing,63 (4), 24–37.

Schlosser, E. (2001). Fast food nation. London: The Penguin Press.

Smith, G. (1990). The effect of advertising on juvenile smoking behavior. International Journal of Advertising ,9(1), 57–80.

Spungin, P. (2004). Parent power, not pester power. International Journal of Advertising and Marketing to Children ,5(3), 37–40.

Sullivan, C., Oakden, J., Young, J., Lau, M., & Lawson, R. (2004). Pulp fiction – The facts harvested. A study of New Zealanders’ physical activity and nutrition . Retrieved 20 June, 2004, from http://www.cancernz.org.nz/Uploads/Pulp_Fiction_Nutrition_Analysis_ Report_Final.pdf Sykora, A. (2003, 2 September). Our approach. Presentation on behalf of Coca Cola New Zealand to Childhood Obesity Symposium, Massey University.

Young, B. (2003). Does food advertising make children obese? International Journal of Adver- tising and Marketing to Children ,4(3), 19–26.

Young, B. & Webley, P. (1996). The role of television advertising in children’s food choice .

London: Ministry of Agriculture, Fisheries and Food.

Zuppa, J., Morton, H., & Mehta, K. (2003). Television food advertising: Counter-productive to children’s health? A content analysis using the Australian guide to healthy eating.

Nutrition and Dietetics ,60 (2), 78–84. Advertising and Obesity 423