Abstract
The overconsumption of sugar-sweetened beverages (SSBs) is associated with noncommunicable diseases such as obesity, type 2 diabetes, and cancer. Voluntary industry codes have largely failed to curb the overconsumption of sugar, and governments globally are increasingly willing to impose taxes on SSBs. However, the effectiveness of SSB taxation varies significantly. Drawing on a systematic review of the most recent literature (N = 79), the authors find that heterogeneity in outcomes is likely to result from idiosyncratic conditions within marketing systems. Building on marketing systems theory, they identify and critically evaluate (omitted) factors within the marketing environment that have an impact on the effectiveness of SSB taxation. Findings reveal that most studies to date focus on demand-side issues, often omitting supply-side responses such as reformulation or pass-through rates. Furthermore, studies largely disregard evidence from marketing and behavioral sciences, which show that taxation works through psychological mechanisms other than price. Finally, the authors find that few studies have systematically evaluated the complementary effects of SSB taxation and other health-promoting policies. By highlighting these blind spots in the current SSB taxation knowledge, the authors provide fruitful avenues for future research at the nexus of marketing and public policy.
As far back as 1776, Adam Smith observed that “sugar, rum and tobacco are commodities which are nowhere necessaries of life…which are…objects of almost universal consumption, and which are therefore extremely proper subjects of taxation” (Smith 1776, p. 445). The current motivation for a tax on sugar is very different from that of the 1700s: rather than a luxury, sugar is now a food consumed daily by most people and to such excess that it has led to high levels of overweight and obesity globally. Sugar-sweetened beverages (SSBs) have become the main target for such “sin taxes.” Because SSBs largely fail to activate satiating signals in the body, they are a key source of excess calories in American diets (Han and Powell 2013). Thus, SSBs are a key contributor to the obesity crisis, which is the third leading cause of preventable deaths in the United States (Danaei et al. 2009).
Although evidence suggests that taxation curbs overall SSB consumption significantly, research also finds great variance in outcomes across and within populations (e.g., Nakhimovsky et al. 2016; Park and Yu 2019). As a result, the effectiveness of such taxes in curbing the (over)consumption of sugar and improving health outcomes is still contested by the academy, the political establishment, and the media (e.g., Brownell et al. 2009; Falbe et al. 2016). One explanation for the varying success of SSB taxes involves forces within the specific marketing systems in which the taxes are implemented. Marketing-related factors, however, have received little attention from academics and policy makers. This provides a rationale for further investigations and is where this study makes an important contribution to the literature.
We critically evaluate the empirical evidence (N = 79) on SSB taxation between 2010 and 2020. We review the literature through a marketing systems lens (Mittelstaedt, Duke, and Mittelstaedt 2009) because a perennial problem facing our current knowledge base is that existing studies have operationalized SSB taxation predominantly through elasticities of demand, often not accounting for the motivations, actions, and relationships of actors within complex and adaptive systems in which taxes are implemented (e.g., Alvarado et al. 2019).
This review has four objectives. First, we integrate and synthesize core insights from the disparate empirical research on SSB taxation. We organize existing studies according to focal outcomes, methods, and key empirical findings. Second, we highlight congruent findings by collating the evidence from previous meta-reviews and thus shed further light on the effectiveness of SSB taxation. The findings also highlight some significant variations between and within studies, providing the rationale for analyzing the existing evidence through a broader lens that considers various factors within marketing environments. Third, we critically review the current evidence through a marketing systems lens. We highlight important inconsistencies and gaps in the current body of research, many of which lie in the broader domain of marketing theory, and we identify and discuss potential explanations. Finally, drawing on the insights generated from this systematic review, we derive clear directions for future research at the nexus of marketing and public policy.
The Rise of Sugar Taxation as a Policy Instrument
This study is motivated by the shift in the burden of disease globally over the past 40 years. Although the prevalence of infectious diseases has decreased, there has been an increase in the prevalence of noncommunicable diseases (NCDs), including obesity, cardiovascular disease, type 2 diabetes, cancer, and diseases of the mouth (Kassebaum et al. 2017; Lim et al. 2012; Moodie et al. 2013; Naghavi et al. 2017).
In an effort to address the health and economic burden posed by NCDs, health experts have called for the implementation of prevention measures delivered at the population level (Fidler et al. 2017; Lobstein and Brinsden 2014; Shill et al. 2012). As the quality of nutrition, dietary patterns, and the food environment all contribute to the development and progression of these diseases, improving these areas has become a key target for action (Contreary et al. 2015; Wakefield et al. 2006).
However, a fault line is discernible in the consensus that the problem of obesity needs to be addressed. Some stakeholders have advocated for self-regulation within the food and beverage industries, claiming that the design and monitoring of regulation is costly and time-consuming (Karnani 2011) and that paternalistic policies can trigger reactance that could exacerbate unhealthy behaviors (e.g., Dumitrescu, Shaw Hughner, and Shultz 2016). Others have stated that self-regulation in the food industry can lead to regulatory capture, the reframing of evidence bases to suit corporate agendas (Bes-Rastrollo et al. 2013), and a lack of transparency and accountability (Freudenberg 2014; Nestle 2010; Norum 2005).
Some governments have developed and implemented sugar taxes, specifically targeting SSBs like soda and soft drinks. Indeed, the call for fiscal policies has become louder in light of increasingly “seductive” market environments in which consumers are faced with aggressive marketing and the convenience, salience, and availability of calorie-dense, nutritionally poor foods—factors that are now widely recognized as important determinants of the (over)consumption of such products (e.g., Murray et al. 2012).
Taxation has long been used as an instrument of social policy, including in the reduction of tobacco and alcohol consumption. The theoretical justification for a sugar tax is based on the Pigouvian principle (Pigou 1920) that a tax is imposed on goods or services that are undesirable because they generate a social cost (e.g., health care provision) that is not met by the individual. An increase in the relative price is thought to incentivize utility-maximizing consumers to reduce consumption of sugary foods and beverages to more socially acceptable and less harmful levels. The rationale for a tax intervention is often countered by an argument that consumer food literacy and empowerment are the foundation of the war against obesity: the problem is information deficit, and educating consumers will enlighten and empower their choices, driving the market to offer healthier alternatives. Thus, pertinent questions are whether welfare gains caused by SSB taxation are significant enough to justify its use as a policy instrument and what factors are likely to determine the effectiveness of such taxation.
Literature Review
To identify factors in the marketing environment that are likely to influence the effectiveness of SSB taxation as well as to highlight any gaps in the literature, we conducted a systematic literature review. We used the following online databases in our search: ScienceDirect, EBSCO Host, Scopus, and ProQuest. All publications between January 2010 and January 2020 were considered. Results were limited to full-text, peer-reviewed publications written in the English language.
For our first step, we searched the term “sugar tax*.” We included in our review all articles that directly referred to the taxation of sugar in the title, abstract, or keywords. The asterisk (*) was used for multiple character searching of the original word stem, allowing the database to return results that included words like “taxation” and “tax.” This first step led to a broad range of results beyond the scope of our endeavor. For example, studies included taxation of product categories such as confectionary or other sweet foods, or historical studies related to the sugar, rum, and tobacco trade of the 18th century. To refine the search, we connected the initial search term “sugar tax*” with “soft drink” OR “sugary drink” OR “sugar sweetened beverage” OR “SSB” by using an “AND” condition.
Next, we scanned each article to identify studies (including meta-reviews) that empirically investigated the relationship between sugar taxation and consumption, obesity, or other distal health outcomes. Studies without original empirical research (e.g., editorials or essays) were then excluded. In a final step, overlaps throughout the different databases were eliminated, resulting in a final set of N = 79 articles that we deemed relevant. Table 1 provides an overview of the search results found in each database.
Literature Review.
Note: Six meta-analyses are included in the total number of articles.
Finally, we analyzed and categorized each article. We synthesize and discuss important findings in the body of this article, and the Web Appendix provides a detailed overview of all empirical studies (N = 73), excluding the six meta-reviews. In the columns of the Web Appendix, each article is classified according to (1) study type (e.g., modeling study), (2) geographic location (e.g., United States), and (3) SSB tax type (e.g., penny-per-ounce). These descriptive findings are summarized and discussed in detail in the next section. Furthermore, the Web Appendix provides information regarding the (4) focal dependent variable(s) (e.g., change in population health) and (5) key measures (e.g., prevalence of obesity among adults), as well as (6) a summary of the main empirical findings (e.g., reduction in the number of obese adults in the population). We also identified and listed (7) key control variables for each study. We next discuss these results in detail, highlighting significant congruencies as well as variations between and within studies.
Mapping the Field: Research on SSB Taxation
The number of empirical studies on SSB taxation has grown exponentially over the past ten years. Over 80% of the articles were published in 2015 or later. Furthermore, all studies are published in science, medical, and health journals. No studies have been published in marketing or consumer research journals. Indeed, almost 40% of articles are published in four leading publications: PLoS ONE (N = 9), PLoS Medicine (N = 8), BMC Public Health (N = 7), and American Journal of Public Health (N = 6). Other publications include the International Journal of Behavioral Nutrition and Physical Activity (N = 4), Public Health Nutrition (N = 4), Appetite (N = 3), BMJ Online (N = 3), and Health Affairs (N = 3). Together, these journals account for almost 60% of empirical studies on SSB taxation (Table 2).
Studies by Journal.
Most studies were conducted in North American (55%) or European (18%) contexts. A significant number are based on data from Latin America (11%). Studies conducted in New Zealand and Australia (7%), Africa (4%), or Asia (4%) account for a relatively small proportion of articles.
Methodologically, most sugar tax research is based on modeling studies (54%). In the absence of “real-life” intervention data, these studies use available demographic information as well as economic data (e.g., price elasticities) to estimate the impact of price changes on consumption choices and/or health outcomes in a population. However, with more countries opting for SSB taxation, recent studies are often observational (28%). Many of these studies are based on natural experiments, which evaluate relevant outcome variables before and after the introduction of SSB taxes and/or differences within differences (e.g., in product categories). Other studies compare differences in outcomes between comparable jurisdictions. Only a small number of studies have used randomized controlled experiments (10%), and even fewer use surveys (7%) or qualitative methods (1%).
In terms of outcomes, 6% of studies evaluate the impact of SSB taxation on prices, effectively calculating tax pass-through rates. Most studies estimate the impact of SSB taxes on sales and/or purchases of SSBs (33%). Regarding sales, studies often compare the hypothetical or actual (total, average, or relative) number of SSBs that individuals, households, or populations purchased before and after the introduction of the tax. In addition, 19% of studies evaluate the effect of SSB taxation on consumption, which is often measured in terms of change in SSB intake or the associated intake of energy (kcal). Many of these studies also account for substitution effects in other product categories such as water, milk, or diet drinks. Other studies evaluate the impact of SSB taxation on weight-related factors (18%) or health outcomes (15%). Health indicators capture a wide variety of outcomes, including prevalence of NCDs such as type 2 diabetes, cardiovascular disease, or stroke. Studies also measure the impact of SSB taxation on health-adjusted life years or the reduced number of premature deaths associated with a reduction in SSB consumption. Two studies explicitly focus on the cost effectiveness of SSB taxation, and only one study explicitly assesses different industry responses, such as reformulation. However, a small proportion of studies control for a limited number of industry responses to taxation by accounting for and/or conducting sensitivity analysis around possible pass-through rates and/or substitution effects. The Web Appendix provides a more detailed overview of the focal outcomes and measures.
Because our aim is to identify specifically the factors within the marketing environment that are likely to influence SSB tax effectiveness, we do not provide a more detailed analysis of the empirical findings of these studies. Instead, we briefly summarize evidence from the existing meta-reviews and focus on the role of factors in the marketing environment, many of which have received little attention from academics and policy makers.
Is an SSB Tax Effective? Evidence from Meta-Reviews
Our analysis included six meta-reviews, which primarily investigated the impact of an SSB tax on consumption and distal health outcomes. They differ from the analysis presented in this study, in which we are more interested in identifying the underresearched factors in the marketing environment that are likely to influence the relationship between SSB taxation and consumption.
For example, in a meta-review of nine SSB taxation studies in the United States, Mexico, Brazil, and France, Escobar et al. (2013) calculated an average price elasticity of −1.299 (95% CI: [−1.089, −1.509]), suggesting that higher prices result in lower SSB consumption as well as in potential reductions in body mass index (BMI) and prevalence of overweight and obesity. Furthermore, the authors find evidence for substitution effects, showing that higher SSB prices are associated with greater demand for alternative drinks such as juice or milk. The authors call for further research into how context-specific factors influence price elasticities and, specifically, the cost-effectiveness of SSB taxation.
Niebylski et al. (2015) cast a wider net and review a total of 78 studies that estimate the effects of taxation of SSBs and other foods as well as healthy food and beverage subsidies. They conclude that taxation influences dietary behaviors and is likely to be most effective when increasing prices by at least 10%–15% and when used in tandem with subsidies. The authors also acknowledge that the implementation of SSB taxes requires policy makers to engage with myriad stakeholders within a marketing ecosystem, which “poses a complex dilemma, especially considering [stakeholders’] averseness to increased ‘taxation’ and influenc[es on] their freedom to food choices” (Niebylski et al. 2015, p. 789).
Backholer et al. (2016, p. 3070) investigated the differential impact of SSB taxation on “beverage purchase and consumption, weight outcomes and the amount paid in SSB taxes according to SEP [socioeconomic position].” Their meta-review of the literature up to 2015 incorporated 11 studies. These studies investigated the relationship between taxation and SSB demand (N = 1), consumption (N = 3), and hypothetical modeling studies, which use established price elasticities to estimate potential effects on consumption, calorie intake, and body weight (N = 7). The findings indicate that an SSB tax is likely to be regressive, but the differences between low- and high-income households are likely to be very small (i.e., .10%–1.0% and .03%–.60% of annual household income paid in SSB tax for low- and high-income households, respectively), suggesting that SSB taxation will deliver similar benefits across different socioeconomic groups.
The meta-review undertaken by Nakhimovsky et al. (2016) focuses on the effects of SSB taxation in middle-income countries. Drawing on nine studies conducted in Brazil, Ecuador, India, Mexico, Peru, and South Africa, the authors find that price elasticities range between .6 and −1.2. Their estimates suggest that a 10% increase in SSB prices would result in a reduction in average caloric consumption of 5 to 39 kilojoules per person per day. The authors also find evidence for possible substitution effects, with some studies estimating a negative relationship between SSB prices and obesity outcomes due to consumers’ switching to other (high-calorie) products. Furthermore, the meta-review points out that most studies assume that taxes are fully passed on to consumers and are not absorbed by retailers and manufacturers. The authors indicate that the impact of manufacturer and retailer responses is not well understood.
In a more recent meta-review, Park and Yu (2019) identified 17 studies that measure the impact of soda taxes on consumption as well as obesity and other distal health outcomes. Their descriptive analysis reveals that SSB taxation is consistently associated with reductions in consumption, average BMI, and obesity, as well as cardiovascular disease. However, 14 of the 17 studies selected for analysis are modeling studies, and the authors highlight that these are based on several assumptions, which are “limitations because they often cannot consider real-world dynamics” including a static industry response (Park and Yu 2019, p. 94). The authors thus propose that future studies pay closer attention to the dynamic factors in “the specific context of each region and [that] controlling confounders are necessary for establishing the internal and external validity” (p. 94).
Finally, Teng et al. (2019) conducted a systematic review of “real-world” SSB taxation (i.e., observational studies) and examine by meta-analysis the impact on beverage purchases and dietary intake. Their analysis includes only studies that have a temporal (pre- vs. post-tax) or geographic (taxed vs. untaxed jurisdiction) basis for comparison. The findings reveal that “the equivalent of a 10% SSB tax was associated with an average decline in beverage purchases and dietary intake of 10% (95% CI: [−5.0%, −14.7%], n = 17 studies, 6 jurisdictions),” and associated substitution effects averaged a 1.9% increase in total untaxed beverage consumption but were not statistically significant (Teng et al. 2019, p. 1187). Although these results provide further evidence that SSB taxation appears to consistently reduce SSB purchase and dietary intake, the analysis also reveals considerable heterogeneity between jurisdictions (i.e., local context) and tax design and somewhat lower heterogeneity within jurisdictions. This suggests that the effectiveness of SSB taxation is contingent on critical factors within the respective market environments. For example, the authors contend that some of the differences between jurisdictions stem from non-price-related factors such as health signaling pathways, which depend on the public’s health consciousness and general awareness of the tax. Other factors may concern supply-side factors, such as tax pass-through rates (e.g., Vecino-Ortiz and Arroyo-Ariza 2018) or reformulation (e.g., Acton and Hammond 2018).
To summarize, most studies evaluate the effectiveness of price signals to influence SSB purchases and consumption. Meta-analyses report that taxation influences consumption choices of SSBs in the desired manner (Escobar et al. 2013; Lal et al. 2017; Niebylski et al. 2015). Consistent conclusions from meta-analyses include that larger taxes are associated with more significant changes in consumption of the taxed product and that a tax needs to be nontrivial (e.g., 10%–15%) to garner sufficient changes in consumption (e.g., Backholer et al. 2016; Nakhimovsky et al. 2016). Furthermore, most studies express preferences for a multifaceted approach to targeting sugar consumption (e.g., Niebylski et al. 2015). Importantly, results show that the beneficial effects of SSB taxation are small to moderate and vary greatly across and within studies. This heterogeneity provides a rationale for further research into the factors that are likely to mediate or moderate the impact of SSB taxation on consumption and health outcomes (e.g., Alvarado et al. 2017).
Broadening the Lens: SSB Taxation and Marketing Systems
Marketers have long emphasized the importance of marketing systems in designing policy (see Hunt 1981). In his seminal work, Layton (2007, p. 230) defined marketing systems as “network(s) of individuals, groups, and/or entities linked directly or indirectly through sequential or shared participation in economic exchange that creates, assembles, transforms, and makes available assortments of products, both tangible and intangible, provided in response to customer demand.” He explains that marketing systems are “environments” (Layton 2007, p. 237), reminding the reader that such systems exhibit qualities that are difficult to comprehensively capture using cause–effect models because (1) events happen dynamically and continuously; (2) relevant variables interact in complex ways, making each difficult to isolate; and (3) the actors and institutions can have simultaneously contradictory roles.
In our context, taxing SSBs aims to correct the negative externalities that stem from overconsumption of sugar by changing the decisions of both producers (e.g., reformulation) and consumers (e.g., reduction in consumption). However, the evidence discussed previously highlights the need to analyze the tax policy process in the context of the unique marketing system in which it is implemented (Mittelstaedt, Duke, and Mittelstaedt 2009). Alvarado et al. (2019, p. 9) concluded that “currently most evaluations test the implicit hypothesis that price change drives change in SSB consumption,” often failing to adequately acknowledge that “SSB taxes exist in complex and adaptive systems, and it is likely that they (like tobacco taxes) may operate in more complex ways.” Many of these influences reside within the marketing environment in which SSB taxes are implemented.
Specifically, we argue that the welfare gains associated with such a tax (i.e., reduced levels of obesity and noncommunicable diseases) are largely determined by the decisions and interactions of key actors within the marketing system, as well as the unique characteristics of the marketing environment. Layton (2015, p. 306) argues that “in thinking about a marketing system it is essential to identify with precision the individuals, groups, and entities that comprise the system or systems, and the institutions, infrastructure, artefacts and devices that are part of the system setting and environment.” Consistent with a marketing systems perspective, the tax policy process is characterized by a complex network of interactions between the government, suppliers (i.e., manufacturers and retailers), and consumers, as well as moderating influences that arise from the macro environment. The flow diagram in Figure 1 illustrates how the interrelated behaviors of key stakeholders are likely to influence the relationship between SSB taxation, consumption, and obesity/NCDs. Figure 1 also highlights potential feedback loops as well as boundary conditions that reside within the marketing environment. Figure 1 thus serves as a useful starting point to visualize and better understand how decisions from government (e.g., tax rates and bands) as well as responses from industry (e.g., pass-through rates) and consumers (e.g., substitution) are likely to mediate the influence of tax-induced price changes on SSB consumption. In the following section, we review the empirical evidence regarding the role of these stakeholders and boundary conditions within the sugar tax ecosystem in more detail. Our review follows the flow of the SSB taxation process as shown in Figure 1.
Mapping the SSB taxation process within marketing systems.
SSB Tax Effectiveness and Tax Design
SSB taxes vary greatly in their design and scope. To correct market failure and to maximize net welfare gains, governments in various jurisdictions have opted for different tax designs and have applied them at different rates (Figure 1). However, few studies to date have explicitly investigated the impact of different tax types, sizes, bands, or the complementary role of other policies.
Tax rates
Findings suggest that larger taxes are associated with more significant changes in consumption (Thow et al. 2010). Researchers have stressed that a tax needs to be nontrivial, for example 15%–20%, to gain sufficient changes in consumption, and there is consistent evidence that the higher the tax, the more effective it is (Acton and Hammond 2018; Bes-Rastrollo et al. 2016; Nakhimovsky et al. 2016; Shemilt et al. 2013; Waterlander, Ni Mhurchu, and Steenhuis 2014; World Health Organization 2015). Furthermore, there is emerging evidence in these studies that taxation has a greater positive net health gain in lower socioeconomic groups (e.g., Paraje 2016; Peñalvo et al. 2017). Such results indicate that larger taxes not only have a disproportionate impact on consumption and health but also a greater potential to reduce disparities in health outcomes between lower and higher socioeconomic groups.
Tax types and bands
Other studies show that different types of tax are likely to affect purchase decisions differently. Generally, SSB taxes can be applied per calorie value or grams of sugar (i.e., specific tax) or per value of sales unit (i.e., ad valorem tax), and they can be implemented as an excise tax (i.e., before sales) or as a sales tax (i.e., at point of sale). Zheng, Huang, and Ross (2019) find that excise taxes (when compared with sales taxes) drive much steeper reduction in demand, mainly because soda firms pass the excise tax on to consumers, who will ultimately face a noticeably higher price when purchasing soda.
Briggs et al. (2013a) modeled different industry responses to the United Kingdom’s proposed two-tiered levy on SSBs in which the tax amount was determined by grams of sugar. Their study compared three scenarios in which industry would (1) engage in reformulation to reduce sugar concentration, (2) increase product prices, or (3) change the market share of high-sugar, mid-sugar, and low-sugar drinks. The analysis suggests that the reformulation scenario is likely to result in greater reductions in obesity prevalence as well as type 2 diabetes and tooth decay compared with price or market share changes. The study thus raises important questions regarding tax design and the incentives it creates for industry. For example, ad valorem taxes are likely to result in price changes but might be less effective at incentivizing manufacturers to reformulate their products.
Importantly, the tax type and rate are likely to influence the salience of the tax. People’s awareness of SSB taxes is likely to influence the effectiveness of the tax, with studies showing that the salience of a tax at the point of purchase has a significant impact on people’s consumption choices. For example, in a field experiment, Chetty, Looney, and Kroft (2009) found that posting tax-inclusive price tags reduced demand by 8% more than in the control group, in which the tax was not made salient. These findings suggest that explicit labeling of an added-sugar tax might see a significantly larger reduction in SSB consumption. There is also evidence to suggest that when cohorts are mostly unaware of an SSB tax (Ortega-Avila, Papadaki, and Jago 2018), or if the channel is low-involvement/high-convenience such as a vending machine, the tax is less effective (Seah et al. 2018). These findings also mirror research in domains such as food labeling that suggests front-of-package nutrition labeling has a significant influence on product perceptions and purchase intentions (e.g., Andrews, Burton, and Kees 2011; Howlett et al. 2012; Newman, Howlett, and Burton 2014).
Complementary policies
Emerging evidence suggests that taxation of SSBs results in greater benefits when launched in conjunction with subsidies or other policy tools. In a field experiment, Breeze et al. (2018) evaluated the impact of a local SSB tax of 20 pence that was introduced alongside a health promotion policy in leisure center venues in Sheffield, UK. They found that the combination of pricing and promotion policies resulted in a 31% reduction in SSB units sold per attendance in the year since the policy was introduced—a significantly larger change in behavior than has been observed in other studies (e.g., Colchero et al. 2017). Although evidence is also emerging that plain packaging and warning labels on SSBs have significant effects on purchase intention, no interaction between taxation and the labeling or plain packaging policies have been reported (Bollard et al. 2016; Seah et al. 2018).
In a comparative policy study, Pearson-Stuttard et al. (2017) modeled the impact of various policies on the prevalence of cardiovascular disease mortality in the United States and found that a combination of SSB taxes, educational campaigns, and subsidies for fruit and vegetables is more effective than stand-alone policies in reducing cardiovascular disease burdens. Likewise, Caro et al. (2018) modeled the impact of various policies on consumption of different nutrient groups in Chile and found that a 10% ad valorem SSB tax would likely result in a 13% reduction of SSB purchases.
SSB Tax Effectiveness: Marketing Decisions
Because taxation is a process within a social matrix rather than a one-off event, the relationships between taxation, sugar consumption, and health outcomes are complicated by the (marketing) decisions and actions of SSB manufacturers. During the implementation stage (Figure 1), manufacturers not only determine the tax pass-through rates but also make decisions about the nutritional content of products, distribution strategies, and communication related to SSBs (e.g., Griffith et al. 2016).
Price changes
Prior studies have mainly focused on demand-side factors, whereas supply is often assumed to be perfectly elastic (i.e., 100% pass-through of tax). In a theoretical study, Dharmasena, Davis, and Capps (2014) modeled both demand-side (i.e., consumer) and supply-side (i.e., industry) responses. Their findings show that assumptions about the supply side are often more important and that ignoring supply-side effects severely overestimates quantity and calorie effects and slightly underestimates revenue effects. More recent studies have therefore controlled for and explicitly modeled the impact of industry pass-through rates (e.g., Briggs et al. 2013b; Wilde et al. 2019).
Observational studies from various jurisdictions, however, suggest that SSB manufacturers might be strategically adjusting pass-through rates to benefit from variations in price elasticities. For example, Stacey et al. (2019) observed that pass-through rates in South Africa varied by product size. Their findings show that average pass-through rates for carbonates was approximately 68%, with 100% pass-through on small containers and 50% pass-through on larger containers. Likewise, Colchero et al. (2015) found evidence that pass-through rates in Mexico varied by package size and product type and that incomplete pass-through rates could be observed in some regions. However, more research is needed to better understand how industry’s pricing decisions influence the effectiveness of SSB taxation. For example, Vecino-Ortiz and Arroyo-Ariza (2018, p. 111) modeled the impact of various SSB tax scenarios on obesity prevalence in Colombia. Their study shows that in “the most conservative scenario (40% pass-through), a tax rate of at least 75 cents of a Colombian peso per milliliter (24% of the average price) is needed to have statistically significant effects on both overweight and obesity prevalence among lower SES households.”
Formulation changes
SSB manufacturers might also change their products in response to taxes. Indeed, taxes that are levied by grams of sugar per liter provide strong incentives for firms to reformulate their products to bring sugar content below a set threshold. Research in response to the introduction of a levied sugar tax in the United Kingdom modeled three potential industry responses to a sugar tax and found that SSB reformulation (vs. price increases and change in market share) produced the best health outcomes (Briggs et al. 2013a). Results suggest that supply-side-led reformulation may prove more effective in reducing the prevalence of obesity and NCDs than demand-side responses to changes in price. In an observational study, Hashem, He, and MacGregor (2019) compared the change in sugar and energy content on SSB labels before (2014) and after (2018) the introduction of a levied SSB tax in Great Britain and Northern Ireland. Their findings show that the average sugar content of 83 products decreased by 42%. Although the tax seems effective, the authors also conclude that sugar content still varies considerably and that the levy thresholds could be reduced and the tax increased to drive further reformulation of soft drinks.
Likewise, Stacey et al. (2019) found that a levied tax in South Africa resulted in a significant reformulation response on the part of manufacturers. Furthermore, products that were reformulated to be below the levy experienced a similarly large price increase despite the zero tax liability. One possible explanation is that brands’ sugar content and prices were strategically adjusted, as they were all owned and marketed by a single company. Conditionally on demand price elasticities, it is possible that firms compensate for profits lost on higher-sugar and therefore higher-tax products by increasing margins on lower-sugar products (particularly if demand for these products is less price sensitive). More research is needed to better understand product reformulation decisions and their impact on demand across a variety of beverages.
Changes in channel and promotional activities
Purchase decisions and price elasticities are likely to differ by distribution channel and intermediaries. In one of the few studies that observed differences in SSB tax by store type, Silver et al. (2017) found that pass-through rates in Berkeley, California, were complete in large chain supermarkets, small chain supermarkets, and chain gas stations; partial in pharmacies; and negative in independent corner stores and independent gas stations. The authors attribute these differences to possible delays in tax implementation for self-distribution stores, but more work is needed to systematically investigate SSB tax pass-through rates and demand elasticities across various channels of distribution over time.
Industry’s promotional efforts in response to SSB taxation are one of the least studied aspects of supply-side factors. Indeed, none of the 79 studies analyzed in this meta-review accounted for potential changes in advertising, public relations, and promotional strategies. However, some studies focused on the promotional aspects of the proposed policy changes, showing that SSB taxation can trigger behavioral changes through “signaling effects,” which subsequently influence purchase behaviors, for instance though changes in attitudes or social norms (e.g., Álvarez-Sánchez et al. 2018; Ortega-Avila et al. 2018; Taylor et al. 2019). For example, Taylor et al. (2019) investigated the impact of a promotional campaign that preceded an election and subsequent introduction of the SSB tax (Measure D) in Berkeley, California. Measure D was passed with 75% of voters in favor and was preceded by an aggressive campaign dubbed “Berkeley vs. Big Soda.” The empirical results showed that soda purchases dropped significantly in the immediate aftermath of the election, months before the tax was implemented, leading the authors to conclude that the effects of media coverage and the election campaign on consumer behavior might have been larger than the tax itself. Likewise, Álvarez-Sánchez et al. (2018) investigated the potential signaling effect of a Mexican tax by analyzing the association between awareness and self-reported changes in consumption of taxed SSBs. Adults who were aware of the tax were more likely to report a decrease in SSB consumption, suggesting that highly visible campaigns may amplify the impact of taxes. These findings show that consumers are receptive to governments’ promotional campaigns, but less is understood about how manufacturers’ promotional activities have changed in response to SSB taxes.
SSB Tax Effectiveness: Consumption Choices
Consumers are expected to respond to price increases by curbing their consumption of sugary drinks and/or switching to healthier substitute products (Figure 1). However, their consumption and substitution choices are influenced by the decisions and actions of the government and the producers, which codetermine the effectiveness of the tax. Furthermore, elasticities of demand can mask psychological mechanisms such as changes in health awareness or social norms, which are widely unaccounted for.
Own-price elasticities
Most studies to date have utilized own-price elasticities to estimate the potential reduction in demand associated with SSB taxation. Research shows that price elasticities are typically in the range of −.8 to −1.3. For example, modeling from Brazil suggests that a 1% increase in the price of SSBs would result in a .85% reduction of calories consumed from SSBs (Claro et al. 2012). Escobar et al. (2013) report an own-price elasticity of −1.29, and Eyles et al. (2012) report a mean own-price elasticity of −.93 for taxes on carbonated soft drinks. Manyema et al. (2014) use an own-price elasticity of −1.29 in their study in South Africa. Other estimates include −1.37 for soft drinks in Chile (Guerrero-López, Unar-Munguía, and Colchero 2017) and −1.16 for SSBs in Mexico (Colchero et al. 2015). Also, elasticities seem higher for SSBs than for non-SSBs (e.g., Lal et al. 2017; Langellier et al. 2017).
Substitution
An important consideration is the potential for substitution to result from the introduction of a sugar tax. To improve health outcomes, any substitution must be to healthier alternatives. The availability of substitute products varies significantly across markets, and evidence shows that “substitution effects are highly dependent on the products considered and specification of the empirical model” (Levy and Friend 2013, p. 303). Most studies evaluated substitution of SSBs to low-sugar—and therefore untaxed—products such as bottled water or milk. Colchero et al. (2016), for example, found a 7.3% sales reduction of SSBs in the two-year post-tax period in Mexico and a 5.2% increase in plain water sales (manufactured by the nonalcoholic beverage industry). This is in line with Guerrero-López, Unar-Munguía, and Colchero (2017), who found a cross-price elasticity for plain water of .63, suggesting that a 10% increase in the price of soft drinks could lead to an increase of 6.3% in the price of water. Falbe et al. (2016) and Lee et al. (2019) found evidence for a significant increase in water consumption after the SSB introduction in Berkeley.
Regarding other beverages, Breeze et al. (2018) found that a 31% reduction in units of SSB sold per attendance in the post-tax period was associated with a 27% increase in sales of other artificially sweetened non-SSB products. No substitution to fruit juice or water was found. Cornelsen et al. (2017) used point-of-sale data from 37 restaurants in the United Kingdom. The intervention was a 10 pence per drink levy on SSBs. Sales per customer of SSBs that were subject to the levy were measured at 12 weeks and six months after introduction of the levy, and effects on nonlevied beverages were also assessed to consider substitution effects. The study reports declines of 11.7% in the SSBs at 12 weeks and 9.8% at six months, with increased purchase of nonlevied beverages, particularly fruit juices.
Others have argued that consumers possibly compensate for the reduction in sugar from SSB by switching to other sugary foods or drinks. Finkelstein et al. (2013), for example, evaluated substitution effects by utilizing nationally representative panel data to investigate whether an increase in SSB prices results in substitution effects within 12 major food categories among U.S. consumers. The findings show that a 20% price increase of SSBs would result in a 4.7% reduction in energy (kcal) purchases per day per household member. Finkelstein et al. found no evidence for substitution to other sugary foods. Quimbach et al. (2018) used a randomized controlled trial in a virtual supermarket to test for potential substitution effects. Although people in the tax condition purchased significantly fewer SSBs, no significant effects on purchases in other food or beverage categories were observed. Likewise, Waterlander et al. (2014) could find no evidence for substitution effects in an experiment with Dutch consumers.
Consumer differences
Although individual differences such as personality traits or values are important determinants of consumption choices, these factors are generally less actionable for policy makers. Most studies to date have thus focused on observable differences in consumer demographics. For example, sugar taxes are often criticized as regressive, affecting low-income earners disproportionately. There is solid evidence of socioeconomic differences in food consumption patterns, with lower socioeconomic groups purchasing higher proportions of energy from less healthy food and beverage categories compared with those in higher socioeconomic groups (Pechey et al. 2013). Furthermore, people in lower socioeconomic groups appear to be more price sensitive (Paraje 2016), which results in greater behavioral changes (Colchero et al. 2016). As a result, consumers in lower socioeconomic groups tend to experience greater health gains from SSB taxation. For example, modeling in Australia shows that 49.5% of the total health gains from introducing a 20% SSB tax would occur in the two lowest quintiles (Lal et al. 2017). Although those in the lowest quintiles are likely to have the greatest health gain, they are also likely to incur the greatest cost. Other studies show that the effects on health outcomes from a sugar tax are likely to be most evident in low-income individuals (Claro et al. 2012; Sharma et al. 2014).
Growing evidence also suggests that SSB taxation might be more effective at triggering behavioral changes in adults than in adolescents or children, which has important implications for policies that aim to reduce childhood obesity. Cawley et al. (2019) studied the effect of a 1.5¢ tax per ounce (including diet sodas) on SSB purchase probabilities in Philadelphia and found an 11.1% reduction in the probability to regularly consume soda among adults but not among children. Langellier et al. (2017) also modeled the effect on children of a 1.5¢ tax per ounce in Philadelphia. Results show a small reduction in consumption of approximately 1.3 drinks/week, and the effects were slightly larger among children below the poverty level. Studies in the United States by Fletcher, Frisvold, and Tefft (2010) and Sturm et al. (2010) also report no significant effects on BMI in schoolchildren or consumption and weight gain in children, respectively. However, existing research on children and adolescents has predominantly studied the impact of small taxes that are unlikely to have visible effects.
Theoretical Implications and Directions for Future Research
In our literature review, we identified key variables within the marketing environment that are likely to determine the relationship between price, SSB consumption, and distal health outcomes. The focus of the review has been to critically evaluate the empirical evidence regarding these variables in determining SSB tax effectiveness and to highlight potential gaps in the knowledge base.
We next highlight four substantive areas within the marketing environment that have received comparatively little attention and that provide potentially fruitful avenues for future research. Specifically, we discuss how the following factors are likely to impact SSB tax effectiveness: (1) tax design and tax salience, (2) multiplier effects of complementary anti-obesity policies, (3) strategic industry responses, and (4) psychological mechanisms and individual differences in consumer responses to SSB taxes. We examine these challenges to SSB tax effectiveness through a marketing systems lens and outline key theoretical propositions and directions for future research.
Tax Design and Tax Salience
Studies show that tax design, and specifically tax salience, influences the effectiveness of SSB taxation. Taxes need to be of a certain magnitude, generally 10%–20%, to be noticed by consumers. Although higher prices are clearly important, sin taxes also “signal” to consumers that products like SSBs are socially undesirable, which might increase awareness of potentially adverse health outcomes or even change social norms that have been associated with behavioral change in adjacent consumption domains (e.g., Dowd and Burke 2013).
Research indicates that warning labels have an information function but also signal to consumers threat and danger frames that can be quickly processed and counteract health halos caused by branding (Arrúa et al. 2017). These could be transposed to sugar tax designs. Indeed, research on SSB taxation could utilize and build on the rich literature and empirical evidence on nutrition labeling and policy related to food packaging (e.g., Cecchini and Warin 2016, a meta-review). Future research could more systematically evaluate how different SSB tax designs, including tax labeling, affect awareness of price increases, perceptions of SSBs, and subsequent consumption choices.
Researchers could control for cohort differences in consumers. For example, research suggests that adults might be more receptive to sugar taxation than children or adolescents (e.g., Cawley et al. 2019). Research also shows that taxation, like other paternalistic interventions, is perceived by some as a constraint of their personal freedom, which can result in reactance and potential boomerang effects (e.g., Elliott 2017). Controlling for cohort differences is thus critical to better understand how tax designs might affect reactions and consumption among different population groups.
The effectiveness of SSB taxation is likely to differ depending on the purchase context. While most studies to date have not differentiated between purchase decisions in different environments and contexts, emerging evidence suggests that price elasticities and attention vary significantly with purchase occasion and channel. For example, in a randomized controlled field experiment, Seah et al. (2018) found no empirical evidence for the impact of tax salience on SSB purchases from 21 vending machines on a Singaporean university campus. The authors tested whether a 10% tax in combination with a tax-related or subsidy-related message would result in differences in weekly SSB purchases. The results show no significant differences in weekly units of high-sugar beverages sold in the tax message and subsidy message conditions compared to the control condition. Indeed, the authors contend that vending machine purchases constitute impulse/low-involvement purchases and that people might therefore be less price sensitive or receptive to health messages. Future research might compare changes in purchases in retail environments with highly convenient, low-involvement, or impulse-stimulating channels such as vending machines, airports, or night clubs.
Multiplier Effects of Complementary Anti-Obesity Policies
Research typically reports that of all the potential policy instruments, a tax appears to provide the best single outcome (Gortmaker et al. 2011). However, an optimal outcome will be achieved by a suite of policy approaches rather than a tax in isolation. This is in line with recommendations by the Lancet Commission (2019), which proposes that global health problems such as climate change, malnutrition, and obesity have common drivers. Following this guidance leads to multiple interventions at the systems level, which, when combined, result in double-duty or triple-duty actions to tackle myriad issues. As previously noted, emerging evidence also seems to suggest that the effectiveness of SSB taxation increases disproportionately when combined with other policies, such as labeling or awareness campaigns (e.g., Breeze et al. 2018). Empirical evidence on the magnitude of these multiplier effects and the mechanisms through which SSB taxation and other policies trigger behavioral change is scarce. However, combinations of SSB taxes and subsidies for fresh fruits and vegetables have been shown to have greater effects on obesity and distal health outcomes (Pearson-Stuttard et al. 2017), as they result in more significant changes in relative prices.
Our discussion also shows that taxation does not solely work through the price mechanism. Studying the effects of SSB taxation in combination with food labeling (e.g., color codes), advertising restrictions (e.g., plain packaging), or educational campaigns (e.g., in schools) would be potentially fruitful avenues for future research. Specifically, future studies could measure the interaction effects of SSB taxation with other interventions at the population or individual level, including but not limited to fiscal policies such as subsidies (e.g., for fresh fruit and vegetables); changes in the “obesogenic” environment (e.g., removal or restocking of vending machines or fountains); bans on advertising (e.g., plain packaging); school-level interventions (e.g., health literacy); and/or policies encouraging physical exercise. Ideally, such studies should be conducted in more controlled settings, for example, using experimental or quasi-experimental approaches.
Industry Responses
Our review has shown that supply-side factors are often unaccounted for or are assumed to be static (e.g., 100% pass-through of tax). However, these assumptions often do not hold, and in disregarding supply-side effects, researchers risk overestimating quantity and calorie effects and underestimating tax revenue (Dharmasena et al. 2014). While some studies have controlled for industry pass-through rates (e.g., Briggs et al. 2013b; Wilde et al. 2019), fewer have accounted for manufacturers’ strategic changes in product or channel strategies, and no studies to date have evaluated changes in promotion as a response to SSB taxes.
Regarding product changes, research has shown that manufacturers strategically adapt to SSB taxation by reformulating products or adjusting pass-through rates to product size (e.g., Stacey et al. 2019). Research shows that pass-through rates might be significantly lower for larger containers than for smaller ones, which might incentivize (price-sensitive) consumers to purchase larger quantities or to buy in bulk, potentially undermining tax effectiveness. Likewise, manufacturers might be inclined to account for differences in price elasticities between different channels and adjust pass-through rates.
Another blind spot in our current knowledge concerns manufacturers’ promotional and lobbying responses to SSB taxation. Although manufacturers pose as part of the solution by reformulating high-sugar products and espousing in-house codes of responsibility, they actively resist regulation through lobbying government, funding counterstudies on obesity, and sponsoring professional organizations (e.g., Iivonen 2018; McKee, Steele, and Stuckler 2019). Similar practices have been observed in the cigarette and alcohol industries (Moodie et al. 2013). Thus, a more holistic account of the soft drinks industry response would measure how the SSB industry shapes taxation in contradictory ways. For example, Taylor et al. (2019, p. 1483) point out that “since 2009, the soda industry has spent more than $117 million to stop soda tax initiatives in the United States.” In California, this has resulted in the California Sugar-Sweetened Beverages Tax Initiative (#18-0004) no longer being on the ballot as a combined initiated constitutional amendment and state statue for November 2020, and other states have passed laws barring local government from enacting SSB taxes (e.g., Arizona, Michigan, Washington). Despite considerable efforts by the soda industry, Philadelphia was the first city in the United States to approve an SSB tax, which was upheld by a Supreme Court ruling in 2018 concluding that the tax was on distribution and not on sales.
Food studies funded by industry are seven times more likely to report results that are favorable to the industry donor’s business operations (Nestle 2018). Congruent with a marketing systems perspective, studies that measure tax effectiveness longitudinally may also capture changing behaviors in the population, as well as pressures to enact “sunset” clauses in tax regulation, which involve reviewing the tax’s effectiveness on health outcomes and amending the tax legislation if it is perceived to be ineffective (Stewart 2019).
Another important industry response not investigated within sugar taxation research involves promotional efforts to stimulate or shift SSB consumption in the post-tax period. Such efforts could include promotional tactics such as “special deals” (buy-one-get-one-free), discounts, and increases in advertising to reclaim consumer attitudes and boost sales. Indeed, weak and strong theories of advertising have long predicted and empirically shown that advertising spend correlates with consumers’ attitudes and habitual behaviors, as well as food and beverage sales, particularly among children (e.g., Beales and Kulick 2013; Desrochers and Holt 2007; Harris, Bargh, and Brownell 2009).
Although beverage manufacturers might shift promotional efforts to diet or no-sugar products in their portfolio, evidence from the alcohol industry shows that zero-alcohol advertising and public messages on responsible drinking drive brand awareness over and above the actual message or product detail (Jones, Hall, and Kypri 2017). Similar effects may be found in SSB industry responses to taxation.
Consumer Behavior
Research on SSB taxation has predominantly focused on own- and cross-price elasticities and has only partially accounted for differences in socioeconomic groups and age. No study to date has explicitly investigated psychological differences in people’s responses to SSB taxation.
Research from other consumption domains shows that consumer traits (Wood 2012), beliefs (Leary, Minton, and Mittelstaedt 2016), or values (Yakobovitch and Grinstein 2016), to name a few, are important determinants of food-purchasing behavior. Specifically, health awareness and social norms have been identified as mechanisms through which SSB taxation might trigger behavioral change (e.g., Taylor et al. 2019). However, few studies have drawn from the vast consumer behavior and consumption literatures, and therefore consumer responses to SSB taxes have not been evaluated more accurately. Although evaluating these literatures in detail is beyond the scope of this study, we identified several areas on the basis of our review that provide promising avenues for research.
Emerging evidence suggests that habits are likely to play an important role in SSB consumption and people’s responses to penalties like SSB taxation (Verplanken and Wood 2006). For example, Debnam (2017) accounted for differences in elasticities of demand between low- and high-consuming households in response to the penny-per-ounce tax on soda that was introduced in Berkeley. Findings show that high-consuming households are less responsive to price changes following the tax, suggesting that SSB taxes might be less effective in breaking consumption habits of those who would potentially benefit the most. Furthermore, ignoring individual differences and applying elasticities homogeneously across consumers may obfuscate the impact of a tax on the at-risk population.
Debnam’s (2017) finding implies that SSB taxation may not be the most effective policy for breaking the habitual behaviors of high SSB consumers. This is echoed by research in adjacent disciplines, which suggests that breaking habits in the realms of food consumption, transportation choice, energy use, or recycling can often be achieved more effectively through positive behavior change strategies such as prompts, incentives, or feedback (e.g., White, Habib, and Hardisty 2019). The role of habits and different responses in SSB consumption warrants further investigation.
Research also suggests that penalties, like taxation, can even trigger negative affect and defensive responses in some individuals, which may lead to rebound effects at the population level. For example, findings from social psychology provide evidence for reactance behaviors such as an increased desire to engage in a threatened or restricted behavior or increased preference for threatened or restricted goods (e.g., Brehm and Brehm 1966; Rosenberg and Siegel 2018). If this holds true in the context of SSB taxation, reactance might be problematic for regulators. If (high-consumption) individuals respond to restrictions and penalties with stronger preferences for the targeted items, the price incentives created by taxes might generate countervailing effects. For example, Dillard, Kim, and Li (2018) found that anti-SSB messages could trigger reactance, which was heightened by prior message exposure, conservative political orientation, and prior consumption of SSBs. The role of reactance in responses to SSB taxation thus provides avenues for future research.
Finally, few studies have accounted for various food-marketing environments, which affect the consumer in ways that have a well-documented impact on overserving and overeating because the salience, availability, hyper-palatability, and low price of sugary and highly calorific food create obesogenic cultures (e.g., Chandon and Wansink 2007; Swinburn, Egger, and Raza 1999). Consumer research is tooled with theoretical frames to assess reactions to sugar taxes in such dynamic environments, most of which currently do not feature in the empirical research on SSB taxation.
Conclusions
This study highlights the importance of sugar taxes in tackling the problem of overconsumption of sugar, which leads to obesity and associated diseases. Overwhelming evidence suggests that sugar taxes are an effective policy in reducing overall demand for SSBs and thus decreasing obesity rates and associated illnesses. However, SSB tax effectiveness varies significantly across and within studies. Our aim was to evaluate the empirical evidence through a marketing systems lens (Layton 2011) to identify key factors within complex and adaptive environments that affect the relationships between taxation, consumption, and health outcomes and might help explain heterogeneity in outcomes (Alvarado et al. 2017).
Our critical analysis of N = 79 articles has shown that most studies operationalize the relationships between taxation, consumption, and health outcomes primarily through elasticities of demand, often treating supply-side factors as static or ignoring individual differences and important psychological mechanisms in consumer responses. The contribution of this study thus lies in the systematic analysis of the evidence base and the identification of current blind spots. In this way, we highlight numerous avenues for further research, which will ultimately help scholars to better understand the factors and contingencies that have an impact on the effectiveness of food-related taxation in tackling the global obesity epidemic. Policy makers, social scientists, and medical researchers can benefit from adopting a marketing systems perspective to unravel the interconnected decisions and reactions of multiple stakeholders and their potential impact on the associations between sugar taxes, obesity, and health outcomes.
Supplemental Material
Supplemental Material, WebAppendix_R4_Final_PDF - Are Sugar-Sweetened Beverage Taxes Effective? Reviewing the Evidence Through a Marketing Systems Lens
Supplemental Material, WebAppendix_R4_Final_PDF for Are Sugar-Sweetened Beverage Taxes Effective? Reviewing the Evidence Through a Marketing Systems Lens by Marius Claudy, Gerardine Doyle, Lisa Marriott, Norah Campbell and Grace O’Malley in Journal of Public Policy & Marketing
Footnotes
Associate Editor
Clifford Shultz
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
