Obesity: Can Behavioral Economics Help? | SpringerLink

ann. behav. med. (2009) 38 (Suppl 1):S47–S55 DOI 10.1007/s12160-009-9119-2

ORIGINAL ARTICLE

Obesity: Can Behavioral Economics Help? David R. Just, Ph.D. & Collin R. Payne, Ph.D.

Published online: 17 September 2009 # The Society of Behavioral Medicine 2009

Abstract Background Consumers regularly and predictably behave in ways that contradict standard assumptions of economic analysis such that they make decisions that prevent them from reaching rationally intended goals. These contradictions play a significant role with respect to consumers’ food decisions and the effect these decisions have on their health. Discussion Food decisions that are rationally derived include those that trade short-term gains of sensory pleasure (hedonic) for longer term gains of health and wellness (utilitarian). However, extra-rational food decisions are much more common. They can occur because of the contexts in which they are made—such as being distracted or pressed for time. In these contexts, heuristics (or rules of thumb) are used. Because food decisions are made with little cognitive involvement, food policies designed to appeal to highly cognitive thought (e.g., fat taxes, detailed information labels) are likely to have little impact. Furthermore, food marketing environments influence not only what foods consumers buy but also how much. As a general principle, when individuals do not behave in their own interest, markets will feed perverse and sub-optimal behaviors.

D. R. Just (*) Applied Economics and Management, Cornell University, 254 Warren Hall, Ithaca, NY 14853, USA e-mail: [email protected] C. R. Payne Marketing Department, New Mexico State University, MSC 5280, PO Box 30001, Las Cruces, NM 88003-5280, USA

Conclusion Given the limited ability of individuals to retain and use accurate health information coupled with varying levels of self control, profit motivations of marketers can become predatory—though not necessarily malicious. Alternative policy options that do not restrict choice are outlined, which enable consumers to make better decisions. These options allow for profit motivations of marketers to align with the long-term well being of the consumer. Keywords Behavioral economics . Food policy . Food psychology . Obesity . Consumption decisions

Introduction Once thought to be a problem only in high-income countries, overweight and obesity are now problems in every region of the world. The World Health Organization projects that by 2015, 2.3 billion people will be overweight and 700 million will be obese [1]. Differences in education, race, age, location, sex, and income have been implicated in varied incidence rates of overweight and obesity. Despite these varied incidence rates of overweight and obesity, at every level of each of these variables, substantial increases have occurred during the past 25 years [2]. Besides increasing the risk of hypertension, osteoarthritis, dyslipidemia, type 2 diabetes, coronary heart disease, stroke, gallbladder disease, sleep apnea and respiratory problems, cancers (i.e., endometrial, breast, and colon), and death, the economic consequences of this disease to the nations’ health care system is immense [1, 3]. In response to overweight and obesity, usual recommendations of diet and exercise apply.

S48

However, these recommendations have been based on rational or cognitive approaches to modeling human behavior resulting in very modest “leveling-off” of overweight and obesity rates [4, 5]. Significant improvements in reducing obesity and overweight may come from approaches that account not only for rational consumers but also consumers whose decision making is based on heuristics. Rational decision making supposes that the consumer has a set of consistent preferences and systematically calculates to make decisions that maximize their well-being [6]. Heuristic-based decision making assumes that people rely on “rules of thumb” (i.e., status quo or habits) to guide their decision making. The heuristic component of food decision making suggests that behavioral economics can inform not only the obesity epidemic but also food policy created to fight it. Previous research regarding causes of the obesity epidemic suggests an increase in consumption of processed calorie dense (from fat and sugar) food and decreases in physical activity [7–9]. In the first case, the argument has been made that processed calorie dense foods are generally considered cheaper than produce [10, 11]. This is the reason why higher rates of obesity are suggested to occur in socioeconomic groups/areas that are marked by poverty [12, 13]. In contrast, it may also be that these same socioeconomic groups simply do not have the same access to healthier foods as more affluent socioeconomic groups. Mere availably of these foods may play a more important role than prices in shaping food choices that lead to overweight and obesity [14, 15]. In the second case, decreases in physical activity have come about because of busier lifestyles that include longer working hours, commute times, and sedentary leisure activities such as television and computer viewing. Both work and play now emphasize convenience [13–15]. The consequence of convenience may not only be less exercise but also a state of being that discounts cognitive effort in favor of heuristic-driven consumption decisions that can be easily be manipulated by marketing environments. This work addresses standard economic approaches to reducing overweight and obesity but suggests that the reason that they have largely failed is because they do not mirror reality. That is, standard approaches assume that people’s food-related behavior is a function of rational decision making. We suggest that, in limited cases, this may be true, but in most cases, marketing environments (i.e., packaging, price, and promotion) leverage peoples’ heuristic consumption decisions that are anomalous to rational decision making. Heuristic driven consumption behavior can be easily steered by marketing environments to meet profit interests. The

ann. behav. med. (2009) 38 (Suppl 1):S47–S55

more distracted people are, the more they will rely on the environment to inform their consumption decisions. If food policies fail to account for heuristic-based food decisions, they are destined to fail simply because they do not mirror consumer reality.

Standard Assumptions of Consumption Decisions: Rational Models Standard economic theory assumes that individuals maximize their own utility—often interpreted as wellbeing—subject to a set of constraints. For example, an individual may be assumed to derive utility from eating, consuming other goods, and from future health. Constraints on their choices may involve limits on income, the time available to prepare food, or on the health information they may have available at the time of purchase. It is assumed that their utility is determined by preferences and habits, which are generally taken as primitive and immovable. These assumptions are used to derive and estimate the relationships between food purchases and other factors. These standard assumptions pose two primary challenges to analyzing food behavior. First, utility maximization does not consider the possibility of systematic mistakes in decision making. In other words, the model assumes that individuals are making the best choices available to the decision maker in equilibrium. Under this assumption, the only chance for improving the welfare of any individual is by reducing market failure. A market failure occurs when (1) the market fails to provide the necessary information for an individual to make a decision or (2) the consequences of an individual decision will impact others not directly involved in the decision. Because the individual is assumed to make the best decisions given their constraints, there are only two potential improvements the researcher can possibly identify. First, individuals can be made better off by loosening the constraints on choice (e.g., by lowering the prices of food or providing health information). Second, overall welfare can improve if the policymaker can cause the individual to respond to how their decisions will affect others (e.g., by taxing the foods that will increase future costs to others or subsidizing foods that will decrease these costs). With regards to the individual, standard economics relies heavily on the libertarian principle—that expanding choice sets improves individual well-being. Two primary market failures have been proposed to justify policy intervention. First, individuals may be eating poorly because they do not have enough health information. This is the rationale for the various public information campaigns regarding food and nutrition. Secondly, the


health costs of obesity are often borne by taxpayers and thus may represent a way in which individuals do not pay for the full consequences of their actions. This has led many to suggest the use of fat taxes or even outright bans on various foods. The Impact of Price The field of agricultural economics has long examined the relationship between food prices and food demand. They characterize this relationship with elasticity of demand parameters. The elasticity of demand is defined as the percentage change in demand resulting from a 1% change in price. The literature finds generally that food price elasticities are very low. For example, Regmi et al. [16] find that the price elasticity of demand for food in developed country consumers range from −0.0 to −0.5. Thus, it would take at least a doubling of prices for a typical food to reduce consumption by 50%. Of course, the context will also play a role. For example, raising the price of a Hershey bar may induce individuals to switch to substitutes (like a Dove chocolate bar). However, raising the price of all chocolate bars is not likely to have as big of an impact on overall consumption of chocolate bars. Bonnet et al. [17] find that starchy foods have extremely low price elasticities. Kuchler et al. [18] find that a 20% increase in the price of potato chips would only decrease bodyweight by around a quarter pound per year. Furthermore, altering the price of some food nutrients may have unintended consequences through demand interaction effects. For example, decreasing the price of fruits and vegetables may actually increase the demand for some fats [19]. Drewnowski and Specter [11] link price and calorie density. More correctly, they show that dollars per calorie decreases with calorie density. Lipsky [20] convincingly argues that this relationship is a mathematical artifact of the data structure and says nothing of the relationship between price and calorie density. She shows that when price, calorie, and mass data are generated independently and randomly, one obtains the same apparent relationship between price per calorie and calorie density due to the appearance of calories in the numerator of the dependent variable and the denominator of the independent variable. In fact, with similar data, she finds that prices for lower calorie dense options are in fact lower priced. Importantly, Drewnowski and Specter’s study fails to take account of the market nature of food pricing. Haavelmo [21] has shown that, when estimating pricing relationships, one must take account of the fact that prices are determined endogenously based on both consumer desires and suppliers costs. Thus, if we were to find that lower priced foods were necessarily less healthy, we would still need to determine if this was the case because either lower income individuals preferred

S49

less healthy foods or because such foods were cheaper to produce. Interestingly, there is substantial evidence that prices are not the primary driver of many food decisions. Many purchases are separated from the consumption either by time or individuals making it difficult for price to play a direct role [22, 23]. For example, Dickson and Sawyer [24] find that about half of individuals cannot recall the price of items they have placed in their shopping carts only seconds after the fact. If individuals do not recall prices at the time of purchase, it is hard to imagine price playing much of a role in consumption decisions once the item is in their home. Multiple grocery items purchased together results in consumers dissociating the item from the purchase price in a process referred to as payment decoupling [25]. Furthermore, many households will have one primary shopper and several consumers who may never observe prices. While these individuals’ requests for foods may drive purchases, their requests cannot respond to prices. Certainly, the nutritional gatekeeper in the home will play a role in determining how prices influence purchases, but the complex nature of this relationship may diminish the role of prices in consumption decisions. Regardless, there may be consumer segments that are likely to make rational food decisions that maximize their well-being. Inflating prices to target one group of consumers must consider those consumers that already consume healthy quantities of foods so as not to adversely affect their already healthy consumption patterns. The Impact of Health Information A small literature in economics has examined the impact of health information on food purchasing decisions. For example, Brown and Schrader [26] use an index created by searching the popular press for references to cholesterol to measure health information. They then use this as a factor in estimating the demand for high cholesterol foods, finding a substantial impact. However, such an approach is highly subject to spurious correlation. The food marketing literature has used a much more individual-based approach, employing survey techniques. This approach produces rather disappointing results. First, individuals tend to have very little knowledge of how eating impacts their health [27]. Thus, if they were to act on the health information they possessed, it is not certain that their health would be positively impacted. Secondly, individuals tend to cite convenience and taste as the primary drivers of food decisions—with health playing only a very minor role [28, 29]. Thirdly, marketers need to provide short, simple, and often repeated health messages to have any sustained effect on purchasing behavior [30]. In fact, health information is probably more likely to be misused in a market setting. For example, food items will

S50

often display in bold letters the best health attribute of the item while not mentioning any negative attributes. Thus, consumers may respond to “no transfats” labels and ignore calorie density, overall fat content, or other important attributes. Moreover, brands that obtain a reputation for healthier choices may lead consumers to believe that all foods carrying a particular brand name are healthy choices. For example, Chandon and Wansink [31] find that those eating at Subway tend to underestimate their calorie consumption by far greater margins than those eating at McDonalds. For this reason, it is important when modeling food decisions to recognize that consumers may not only lack health information, but they may also seriously misperceive or misuse the health information that they have.

Nonstandard Assumptions of Consumption Decisions: Heuristic Models Standard assumptions of consumption decisions based on rational economic models negate one of the primary motivations for policy. In many cases, people may make poor decisions for lack of will power or through some sort of systematic mistake in calculation. If people make systematic mistakes, then policy makers may have another justification to help the decision maker in improving their decisions. This is a primary strength of using an approach based on behavioral economics. Behavioral economic models combine heuristic decision rules with economic decision making. In some contexts, we can interpret the models as providing both a normative measure of behavior as well as a positive description. Thus, some behavioral economic models seek to tell us how individual decisions deviate from the decisions that might make the individual better off. In the context of food, we may be able to model and identify when individuals make decisions to consume that disregard or underweight health information—or, perhaps, when individuals distort the amounts of food they think they are consuming. A key question in deciding whether to employ a rational or heuristic model should be whether the individuals are able to freely choose the outcome they want. In this case, it is difficult to argue that obese individuals have chosen deliberatively to become so. Rather, individuals invest 40 billion dollars annually in attempts to restrict their own eating behavior in the form of diet plans [32]. Many of these programs prominently feature regular weigh-ins, counseling visits, and other mechanism that increase the social cost of overeating. If this is the case, the standard model relies too heavily on dubious rationality assumptions. For example, standard economic models assume that individuals can easily discard unwanted goods. However, if


individuals of necessity pay money to restrict their own choices, clearly, this assumption does not characterize their decision context. Thus, we conclude that individuals face some costs to enforcing their desired actions with regard to eating. Resisting tempting food is not costless in terms of mental or other exertions. We cannot observe these decision costs because they largely occur within the mind of the individual. However, we can use experimental methods to document the behavioral regularities that result from these decision costs. This is perhaps an imperfect substitute when engaging in normative policy, but at present, it is the only option available. Work by Kozegi and Rabin [33], Bernheim and Rangel [34], and the recent book Nudge by Thaler and Sunstein [35] point out that heuristic models provide an opportunity to alter behavior without constricting the choice set. The libertarian principle, that expanding choice sets improves individual well-being, fails to apply generally under behavioral economic models. Thus, providing individuals with inferior choices that are in some way tempting may make them worse off. In this case, it may be possible to use psychology to remove the temptation without removing choice. In other words, by eliminating some immediate affective response to the stimulus, the individual can be enabled to make a more thoughtful decision based on their true well-being. Or, failing that, the choice can be designed so that the superior choice will be more tempting than the inferior choices. This has been called libertarian paternalism by Thaler and Sunstein [36]. Notably, libertarian paternalism requires a moral judgment by the policy maker as to what the desired choices are. Then, a choice structure is designed that suggests a behavioral norm without restricting what choices can be taken. For example, if workers are by default enrolled in an optional retirement plan, they are much more likely to save for retirement than if the default option is for individuals to not enroll [37]. In this case, the policy maker must make the moral judgment that retirement savings are desirable and then institute a choice structure encouraging savings. Importantly, this structure would allow those who, after thought and deliberation, believe they do not need retirement savings to opt out. Thus, the incidence of the policy on those who do not share the values of the policy maker should be minimal. In a similar way, behavioral economics may lead us to policies that encourage better eating without limiting the choices available. Thus, those who already have healthy food habits or those who consume too few calories may be impacted in a minor way. To understand better, however, how behavioral economics may inform food policy to help combat the obesity epidemic, it is important to first describe variables known to control consumption.


Variables Known to Control Consumption Examples of variables thought to influence individual consumption decisions are given in Fig. 1. These variables are of three types: those that we take as primitive (factors determined prior to the decision-making process), those controlled by the manufacturer (marketing), and those controlled by the individual. Factors such as personal preferences and habits are taken as primitive and fixed. Over time, it may be possible (and in fact desirable) to influence these factors. Yet, over the course of a single decision, these factors are not under the control of any decision maker. Manufacturers control the economic factors that govern decisions—prices and product information. They also control many of the attributes of the food item (packaging, content, etc.) that can predictably impact consumer decisions regarding what to purchase and how much to consume. In a restaurant setting, the marketer may control a greater set of parameters including the restaurant design and atmosphere. Finally, the individual may control some of the variables as well, such as whom they choose to eat with and under what circumstances. Ultimately, the individual chooses what to purchase and how much to consume based on the interaction of these other competing variables. Variables that have received the least amount of attention in the policy context, but may be the most important, are environmental drivers of food consumption. Generally speaking, two types of environments play a critical role in food consumption decisions. The first includes environments in which food is chosen, served, and consumed. The second includes environments in which

Fig. 1 Variables known to control consumption

S51

the food itself is arranged, labeled, packaged, placed, and becomes salient [38]. Consider the first type of environment. Distracting stimuli (e.g., shopping and eating with others, media viewing, bright or subdued lighting, fast or slow tempo music) may influence consumption by limiting the ability to deliberate on choice and intake. An inability to deliberate leads to the reliance instead on heuristics that suggest what is appropriate, normal, or conventional to buy and how much to consume [39–41]. Heuristic-based food decisions may result in selecting and consuming types of food that have short-term hedonic benefits but long-term health detriments [42]. In the reinforcement history of a person, foods are generally known to lie somewhere on a subjective continuum of pleasure [43]. As a consequence, when people are distracted they make snap judgments of what is “good” and/or “bad” food. However, this judgment is not based on whether the food is under a certain threshold of calories or whether a person is able to afford it (i.e., deliberative thinking) but rather whether the marketing environment “suggests” that the interaction with the food will be pleasurable (i.e., heuristic thinking). The second type of environment includes attributes surrounding the presentation and serving of food. These attributes can suggest a consumption norm that dictates what and how much to eat. Consider the size of food containers. Given the same amount and type of food in two containers, the larger container will result in people not only serving themselves more but also consuming more. This phenomenon has been reported for both meal and

S52

snack-related foods [44]. Interestingly, even when given poor tasting food, people still consume more when eating from large than small containers [45]. The size of food packages may suggest a consumption norm of what is appropriate to serve to oneself and consume [46]. Environments can have a powerful yet subtle influence on consumption decisions. Even in the best case social welfare scenario, a given environment may only slightly increase consumption. Considering, however, that the rise of obesity in the USA can be attributed to only 15 additional calories a day [47], it is alarming to consider the real impact of multiple environments acting simultaneously on the consumer. It is not likely that consumption decisions will become less distracted and impossible that the manufacturing/marketing environment will stop the creation of environments built to incur purchasing (and consumption). What then can policy makers do to make progress in an epidemic that has slowed, but not stopped? A possible solution is to cede control to the environment— environments that are built to profit not only the manufacturer or marketer but also the consumer. To do this, it is necessary to first understand how creators of food environments and consumers interact. How Creators of Food Environments Interact with Consumers Given the great number of high impact factors marketers may control, their decisions regarding food marketing can have a big impact on the food decisions individuals make. In the tradition of von Neumann and Morgenstern, this creates a game in which the marketers select the product attributes, price, and environmental factors so as to maximize their profits. Alternatively, the individual controls their environmental factors, purchasing and eating decisions to maximize their utility. The decisions of each will impact the outcome of the other. Generally, behavior in such a game is described by the Nash equilibrium principle. Nash equilibrium occurs where each individual is maximizing their own perceived well being given the strategy of each other player. However, within heuristic models, while the marketer (through experimentation etc.) will behave as if they know how individuals will respond to changes in the environment, our evidence suggests that the individual is unaware of how the environment affects their consumption decisions. This creates an information asymmetry where the marketer holds private information about the consumer who is not only unaware of how the environment affects their behavior, but of how this information is leveraged by the marketer. In economics, this sort of game is referred to as a mechanism design problem. In such problems, there is a loss of efficiency due to the asymmetric information. In


other words, it would be possible to make at least one of the participants in the game better off without making the other one worse off if the information asymmetry could be eliminated. This asymmetry of information about how the individual acts will persist no matter what the prices or health information provided. Thus, simply taxing foods or providing more health information is not likely to eliminate this inefficiency. In this context, the mechanism design problem will lead the marketer to find a way to exploit the individual’s environmental responses to induce greater profits at the individual’s expense. Thus, the environment that is causing individuals to consume too much food is endogenously determined by the individuals own behavioral responses to food. Policy measures that do not account for the mechanism that creates the food and eating environments may only have temporary effects as marketers find ways to adjust the environmental factors to maximize profits—likely subverting any policy goals. Policy could instead be designed to help individuals overcome their behavioral response to environmental factors. In many cases, not knowing how he or she responds to the environment is likely to lead the individual to over-consume—giving the marketer a tool to derive greater profits. However, policies could be designed to allow food marketers to charge extra for environments that encourage under-consumption. For example, a restaurant could charge for a private room in which to eat with fewer distractions or charge proportionally more for smaller portion sizes. The following examples are derived from field studies that suggest the potential for immediate and long-lasting change.

Field Studies that Suggest Change We provide here several field study illustrations of how food marketers may subtly impact eating decisions in trying to maximize profits. We then suggest easy ways that may change the eating environment that could still maximize profit but in more a more responsible way. These win–win strategies are of the type that may be beneficial in the long term in curbing overconsumption. All You Can Eat Thaler [48] first noted the impact of the sunk cost fallacy on food consumption. In an all-you-can-eat restaurant, an individual pays a fixed price for a meal and then decides how much to consume. Given that there is not an additional cost for each trip to the buffet, an individual should simply decide to eat until the next slice of pizza provides no more enjoyment—disregarding the price paid. Instead, Just and Wansink [49] find that those paying a higher price for the


same buffet will eat as much as 25% more in an attempt to get their money’s worth. There may be pressure on restaurateurs to increase prices if the additional food consumed is relatively cheap inadvertently increasing individual consumption levels. Alternatively, all-you-can-eat restaurants may instead use small plates, small utensils, and arrange seating so that patrons are farther away from and are not facing the buffet. Reducing consumption this way would allow restaurants to be more profitable by charging the same price but not having to purchase as much food to prepare. In contrast to fixed price all-you-can eat restaurants, work and school cafeterias provide a context to understand how to nudge consumers to purchase healthier options. Cafeterias Previous research suggests that altering payment methods by which people can purchase foods may subtly lead to better or worse choices in food type and volume [50]. Regarding different payment methods, mental accounting can lead people to view different types of money allocations in fundamentally different ways [25]. Using a debit type card, for instance, might endogenously lead consumers to view the available money more liberally [51]. In contrast, Prelec and Lowenstein [52] find that cash payments are more closely associated with the pain of payment than other payment methods, such as the use of a credit or debit card. Paying with cash may even increase one’s involvement in the decision because trade-offs are more visible [53]. Thus, individuals may seek more hedonic goods, such as relatively unhealthy good tasting food, when using a debit like card in contrast to cash. Students who regularly ate school lunch were recruited to participate in a “lunch” study. In some cases, students were given $20 cash; in other cases, they were given a debit card worth $20, and still in other cases, they were given a debit card wherein $10 could buy anything they wish and $10 could only buy “green dot” items. The “green dot” items were somewhat healthier than the non-green dot items. Interestingly, those using the debit card were more likely to purchase and consumed less healthy foods and foods with more caffeine and sugar versus cash or restricted debit card [54]. While school cafeterias may institute cashless payment systems because of their convenience in transactions and the degree of anonymity provided to free or reduced lunch students, they may be inadvertently contributing to childhood overweight and obesity. Alternatively, school lunch and other cafeterias may want to consider restricted debit cards. By offering consumers a choice between either unrestricted debit cards that allow patrons to purchase whatever foods they wish or restricted debit cards that place some limits on card

S53

purchases, the cafeteria does not arbitrarily limit the ultimate food choices of the consumer. Patrons may even be willing to pay a premium for restricted cards much like the popular 100-calorie packs have become popular for their portion control features. Policy Implications If food decisions are characterized by the game we suggest takes place between consumers and marketers, food policy must take into account the response of both consumers and marketers. If policies simply proscribe some behaviors by consumers, raise the relative prices of some foods, or provide more health information, this will not fundamentally alter the structure of the game. Thus, these policies may simply induce responses by marketers that wash out the effects of the policies. Alternatively, policies that address lower order behaviors of the consumer may be effective in creating both greater profits for the marketer and better health outcomes for the consumer. For example, current US law prohibits marketers from claiming that a certain type of packaging could discourage overeating. Such a policy removes the possibility of marketers profiting from selling food packaged to reduce consumption. Absent this potential strategy, food marketers are left only with strategies that charge for the opportunity to over-consume. Alternatively, a government agency could provide a third party certification for portion size norms much like they currently do for organic foods. Thus, a company that wishes to label their foods as healthy portioned would submit their packaged foods to a government process to determine if the packaging encourages healthy levels of consumption or greater ability to monitor consumption. The certification would be accompanied by an information campaign educating consumers about what the certification means and how to use the certification information. Again, food companies may want to design their packages to meet the standard in order to increase sales or charge more for the same foods. Individuals would seek out the labels in order to help control their own behavior within the eating episode, thus overcoming some portion of the information asymmetry problem. By working more closely with marketers and acknowledging the behavioral nature of food decisions, policies can lead to more healthy eating decisions and higher food marketer profits without proscriptively reducing the food choices available to consumers. Alternatively, disregarding marketer motives may lead to perverse affects. For example, taxing fats may lead all-you-can-eat restaurants to increase prices, leading to increased consumption in a particular episode.

S54

Discussion We conclude that many of the proposed solutions to the obesity problem are unlikely to have much of an impact. In particular, policies that adjust prices or increase health information are likely to be ignored by exactly those individuals who have a tendency toward obesity. Rather, individuals have become obese or overweight by making food decisions that ignore cognitive factors, relying more heavily on heuristic decision mechanisms. In this case, cognitive factors such as price and health information might be less effective than policies that employ sophisticated marketing techniques to combat obesity. These techniques could help individuals by creating contexts that encourage healthier decisions. For examples, policies could encourage a marketplace where individuals are allowed to make binding decisions prior to being exposed to the sensual temptation. Behavioral economic principals suggest that these policies could encourage healthier eating without abridging the choices available to the individual or reducing the profit opportunity of the food manufacturer. While these behavioral economic-based policies have been tested in the field, they have not been tested in contexts that allow for competition between firms in multiple environments. The question remains: Can firms secure long-term profitability in the face of behavioral economic-based food policies? Ultimately, this question is one of consumer satisfaction; that is, will consumers be satisfied with consuming less resulting in repatronage of a firms’ food product? Within a reasonable range of consumption, we believe that the answer is yes (i.e., 100 calorie packs). As previously stated, because people do not believe that their consumption decisions are under their complete volitional control, the decisions they do make are justified as occurring because of inherent preferences [35]. Nevertheless, until these policies are tested in complex environments, any conclusions about long-term profitability will need to be held in abeyance. Behavioral economics informed by food psychology provides a key resource to study how individual food choices are influenced by environmental factors. Understanding the interaction of these environmental factors with food choices has provided food marketers with a tremendous set of tools to increase their profits. On the other hand, policy makers can use these same tools to design a market system that can benefit both marketers and consumers. Economics has been a powerful tool generally to help design policies to resolve problems where individual choices may affect the welfare of others. Additionally, by explicitly taking account of the psychological relationship, behavioral economics provides a new tool to design policies that can resolve problems where an individual’s own decisions do not fully account for their well being.


Policy makers must be careful when applying these tools to preserve individuals’ free will. This can be accomplished by making subtle changes in decision contexts that individuals themselves may not recognize as having an impact. We document several such potential changes in this paper.

References 1. World Health Organization. Obesity and overweight. Available at: http://www.who.int/mediacentre/factsheets/fs311/en/index.html/. Accessibility verified November 12, 2008. 2. International Obesity Taskforce. International Obesity Taskforce prevalence data. Available at: http://www.iotf.org/database/index. asp/. Accessibility verified November 12, 2008. 3. Wang Y, Beydoun MA, Liang L, Caballero B, Kumanyika SK. Will all Americans become overweight or obese? Estimating the progression and cost of the US obesity epidemic. Obesity. 2008; 16: 2323-2330. 4. McCarthy M. The economics of obesity. Lancet. 2004; 364: 21692170. 5. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States, 1999–2004. J Am Med Assoc. 2006; 13: 1549-1555. 6. Hartman RS, Doane MJ, Woo CK. Consumer rationality and the status quo. Q J Econ. 1991; 106: 141-162. 7. Pan SY, DesMeules M, Morrison H, Wen SW. Obesity, high energy intake, lack of physical activity, and the risk of kidney cancer. Cancer Epidemiol Biomark Prev. 2006; 15: 2453-2460. 8. Wansink B, Payne CR. The joy of cooking too much: 70 years of calorie increase in classic recipies. Ann Intern Med. 2009; 150: 291-292. 9. Wang Y, Beydoun BA. The obesity epidemic in the United States —Gender, age, socioeconomic, racial/ethnic, and geographic characteristics: A systematic review and meta-regression analysis. Epidemiol Rev. 2007; 29: 6-28. 10. Drewnowski A. The real contribution of added sugars and fats to obesity. Epidemiol Rev. 2007; 29: 160-171. 11. Drewnowski A, Specter SE. Poverty and obesity: The role of energy density and energy costs. Am J Clin Nutr. 2004; 79: 6-16. 12. Baker EA, Schootman M, Barnidge E, Kelly C. The role of race and poverty in access to foods that enable individuals to adhere to dietary guidelines. Preventing Chronic Disease [serial online] 2006. Available at: http://www.cdc.gov/pcd/issues/2006/jul/ 05_0217.htm. 13. Frank LD, Andresen MA, Schmid TL. Obesity relationships with community design, physical activity, and time spent in cars. Am J Prev Med. 2004; 27: 87-96. 14. Hill JO, Wyatt HR, Reed GW, Peters JC. Obesity and the environment: Where do we go from here? Science. 2003; 299: 853. 15. Wansink B. Mindless Eating. New York: Bantam; 2007. 16. Regmi A, Deepak MS, Seale JL, Bernstein J. Cross-country analysis of food consumption patterns. In Regmi A. (ed.) Changing Structure of Global Food Consumption and Trade. Agriculture and Trade Report, WRS-01-1. US Department of Agriculture, Washington, May 2001, pp. 14–22. 17. Bonnet CP, DuBois, Orozco V. Food Consumption and Obesity in France. Working Paper, Toulouse School of Economics, March 2008. 18. Kuchler F, Tegene A, Harris M. Taxing snack foods: Manipulation of diet quality or financing information programs. Rev Agric Econ. 2005; 27: 4-20.

ann. behav. med. (2009) 38 (Suppl 1):S47–S55 19. Kinsey J, Bowland B. How can the US food system deliver food products consistent with dietary guidelines? Food marketing and retailing an economist’s view. Food Policy. 1999; 24: 237-253. 20. Lipsky L. The Economics of Obesity: Dietary Energy Density and Energy Cost. Paper presented at the Seminar in Nutritional Sciences, Cornell University, November 4, 2008. 21. Haavelmo T. The statistical implications of a system of simultaneous equations. Econometrica. 1943; 11: 1-12. 22. Gourville JT, Soman D. Payment depreciation: The behavioral effects of temporarily separating payments from consumption. J Consum Res. 1998; 25: 160-174. 23. Thaler RH. Mental accounting and consumer choice. Mark Sci. 1985; 4: 199-214. 24. Dickson PR, Sawyer AG. The price knowledge and search of supermarket shoppers. J Mark. 1990; 54: 42-53. 25. Thaler RH. Mental accounting matters. In: Camerer CF, Loewenstein G, Rabin M, eds. Advances in Behavioral Economics. Princeton: Princeton University Press; 2004: 75-103. 26. Brown DJ, Schrader LF. Cholesterol information and shell egg consumption. A J Agric Econ. 1990; 72: 548-555. 27. Fulmer S, Geiger CJ, Parent CRM. Consumers’ knowledge, understanding, and attitudes toward health claims on food labels. J Am Diet Assoc. 1991; 91: 166-171. 28. Asp EH. Factors affecting food decisions made by individual consumers. Food Policy. 1999; 24: 287-294. 29. Nestle M. Food politics. Keynote Address, American Agricultural Economics Association Annual Meetings, Montreal, Quebec, Canada, August 2003. 30. Wansink B. How do front and back package labels influence beliefs about health claims? J Consum Aff. 2003; 37: 305-316. 31. Chandon P, Wansink B. The biasing health halos of fast-food restaurant health claims: Lower calorie estimates and higher sidedish consumption intentions. J Consum Res. 2007; 34: 301-314. 32. Klein S. Does paying for obesity therapy make cents? Gastroenterology. 2005; 128: 530. 33. Kozegi B, Rabin M. Revealed mistakes and revealed preferences. Working Paper Department of Economics. University of California, Berkeley; 2007 34. Bernheim BD, Rangel A. Beyond Revealed Preference: Choice theoretic foundations for behavioral welfare economics. Working Paper 13737, National Bureau of Economic Research, Cambridge MA; 2008 35. Thaler RH, Sunstein CA. Nudge: improving decisions about health, wealth, and happiness. New Haven: Yale University Press; 2008. 36. Thaler RH, Sunstein CA. Libertarian Paternalism. Am Econ Rev. 2003; 93: 175-179.

S55 37. Choi JJ, Laibson D, Madarian BC. Metrick A. For better or for worse: Default effects and 401(k) savings behavior. In: Wise DA, ed. Perspectives on the Economics of Aging. Chicago: University of Chicago Press; 2004: 81-126. 38. Wansink B. Environmental factors that increase the food intake and consumption volume of unknowing consumers. Annu Rev Nutr. 2004; 24: 455-479. 39. Wansink B, Payne CR, Chandon P. Internal and external cues of meal cessation: The French Paradox redux? Obesity. 2007; 15: 1957-1960. 40. Simonson I. The effect of purchase quantity and timing on variety-seeking behavior. J Mark Res. 1990; 27: 150-162. 41. de Castro JM. Eating behavior: Lessons from the real world of humans. Nutrition. 2000; 16: 800-813. 42. Cabanac M, Bonniot-Cabanac M-C. Decision making: Rational or hedonic? Behav Brain Funct. 2007; 3: 45-52. 43. Wise RA. Dopamine, learning and motivation. Nat Rev Neurosci. 2004; 5: 1-12. 44. Wansink B. Can package size accelerate usage volume? J Mark. 1996; 60: 1-14. 45. Wansink B, Kim JK. Bad popcorn in big buckets: Portion size can influence intake as much as taste. J Nutr Educ Behav. 2005; 37: 242-245. 46. Wansink B, van Ittersum K, Painter JE. Ice cream illusions: Bowls, spoons, and self-served portion sizes. Am J Prev Med. 2006; 31: 240-243. 47. Hill JO, Wyatt HR, Reed GW, Peters JC. Obesity and the environment: Where do we go from here? Science. 2003; 299: 853-855. 48. Thaler RH. Toward a positive theory of consumer choice. J Econ Behav Organ. 1980; 1: 39-60. 49. Just DR, Wansink, B. The fixed price paradox: Conflicting effects of “all-you-can-eat” pricing. (Working Manuscript). 50. Just DR, Mancino L, Wansink B. Could behavioral economics help improve diet quality for nutrition assistance program participants? Report # 43, US Department of Agriculture, Economic Research Service; 1999. 51. Hirschman EC. Differences in Consumer Purchase Behavior by Credit Card Payment System. J Consum Res. 1979; 6: 58-66. 52. Prelec D, Loewenstein GL. The red and the black: Mental accounting of savings and debt. Mark Sci. 1998; 17: 4-28. 53. McClure SM, Laibson D, Loewenstein GL, Cohen JD. Separate neural systems value immediate and delayed monetary rewards. Science. 2004; 306: 503-507. 54. Wansink B, Just DR, Payne CR. Constrained volition and healthier school lunches. (Working Manuscript).