Principal investigator: Bruno Benevit
Original title: How Well Targeted Are Soda Taxes?
Authors: Pierre Dubois, Rachel Griffith and Martin O'Connell
Location of the Intervention: United Kingdom
Sample Size: 2.449 individuals
Sector: Tax Savings
Primary Variable of Interest: Demand for soft drinks
Type of Intervention: Taxation
Methodology: OLS, Logit
Summary
Excessive sugar consumption is associated with the development of several serious diseases, posing health challenges for governments. To change the population's consumption habits, several countries apply higher taxes on the consumption of foods with high sugar levels. In this sense, this study conducted a series of analyses to verify the impact of taxes applied to soft drinks on their demand, considering different consumer profiles. The evidence found demonstrated that taxes on soft drinks are relatively effective in reducing sugar intake among young people, but less effective in reducing intake among those with high total dietary sugar intake. The results suggest that the incidence of these taxes is unlikely to be regressive, especially for consumers who benefit from the avoided internalities.
- Policy Problem
Sugar consumption represents one of the greatest health challenges in developed countries. Excessive consumption is linked to the development of several serious diseases, such as diabetes, cancer, and heart disease (DUBOIS; GRIFFITH; O'CONNELL, 2020). The incidence of these diseases constitutes the economic phenomenon defined as internalities, where the consumption of a particular good or service implies long-term consequences for the consumer.
According to the CDC (Centers for Disease Control and Prevention(2016), non-alcoholic beverages are the biggest contributors to the high rates of sugar consumption in the diet today, with soft drinks being the most prominent category. The groups most susceptible to excessive consumption of soft drinks include individuals with diets high in sugar and the young population, especially those with low incomes.
Faced with this problem, governments in several countries have adopted the application of taxes on soft drink consumption with the aim of inducing a reduction in its consumption. In December 2019, taxes targeting soft drinks were being implemented by 43 countries. Among these countries, the United Kingdom implemented the Soft Drinks Industry Levy in 2018, imposing a tax of £0,24 per liter on variants of Coca-Cola and Pepsi from April 1st of that year (DUBOIS; GRIFFITH; O'CONNELL, 2020). In this sense, measuring the consequences of applying this type of tax on the well-being of the population involves understanding how the demand for these products reacts to groups of consumers with different consumption patterns and price sensitivities.
- Policy Implementation Context
The study examined the non-alcoholic beverage market, considering consumer choice between alternative beverages and substitution with snacks/appetizers. Non-alcoholic beverages include soft drinks (with and without sugar), energy drinks, fruit juices, sweetened milk-based drinks, and other sweetened non-alcoholic beverages, as well as bottled water. "Soft drink taxes" are typically applied to soft drinks containing sugar, including diet drinks. On the other hand, beverages such as pure fruit juices without added sugar and beverages predominantly composed of milk are exempt from this type of tax.
The immediate consumption of these beverages is relevant to understanding their demand. Immediate consumption represents a significant portion of the market and a substantial source of sugar, especially for younger people. Approximately half of the sugar intake from sweetened soft drinks comes from purchases for immediate consumption. Furthermore, identifying immediate consumption behavior allows us to distinguish the consumption preferences of households from the consumption of individuals within different age groups.
Soft drinks available for immediate consumption in the UK non-alcoholic beverage market comprise products from brands belonging to Coca-Cola Enterprises, PepsiCo, GlaxoSmithKline (GSK) and Barrs, as well as a large number of smaller brands (a group of brands representing 16% of the market). Additionally, the study also considered fruit juices, flavored milks and flavored waters, which together represent just under 10% of the market, and bottled water, which represents a further 11%.
- Evaluation Details
The data used in this study comes from the immediate food consumption survey of Kantar Worldpanel The survey, conducted by the market research company Kantar, tracks food and beverage purchases made by individuals for immediate consumption outside the home and for consumption at home. Individuals in the immediate consumption survey are randomly selected from households registered in the database. This allowed for the identification of information about the products purchased, the transaction price, the store where the purchase was made, household attributes, and consumer characteristics.
Based on these transaction-level prices, and for each product, it was possible to calculate the average monthly price in each type of store, allowing for an estimation of product demand. The average monthly price was calculated in two ways. For national chains in the United Kingdom, with nationally defined prices, and for vending machines, their respective national prices were considered. For independent stores, the prices were calculated regionally.
The data comprises a panel of 2.449 individuals between June 2009 and December 2014. The study observations were made at the "choice occasion" level, defined as a day when individuals purchase a non-alcoholic beverage or a snack. The final sample consisted of 616.544 choice occasions, where 95% and 60% of consumers reported observations on more than 25 and 100 choice occasions, respectively.
- Method
The study considers a demand model that takes into account various characteristics of products, transactions, consumers, and periods. Regarding product characteristics, the study considered product type (beverage or snack), brand, whether or not sugar was added, and price. To control for transaction characteristics, the reseller firm, location, and type of advertising were observed, considering different types of target groups. Regarding consumer characteristics, the fixed effects of each individual were considered, and groups of individuals were distinguished in terms of gender and age range.
The authors estimated the extent to which the tax on soft drinks is passed on to market prices. To do this, an event study method was adopted for the prices (per liter) of the two main brands, Coca-Cola and Pepsi. The period analyzed included the year before and after the date of implementation of the tax, considering the characteristics of the products, transactions, and consumers.
Consumption preferences for beverages were identified in order to capture the heterogeneity of individuals' tastes. Thus, a multinomial logit method was used, allowing the estimation of the probability of individuals purchasing a given product on each choice occasion. This approach allowed the evaluation of the tax impact for consumer profiles with different age ranges and sugar preferences.
Through the estimated demand model, the impacts of the tax on soft drinks on the amount of sugar ingested during immediate beverage consumption were verified. Additionally, the effectiveness of targeting taxation to priority groups was verified: young people, individuals from low-income families, and individuals with high annual sugar intake. Finally, the study analyzes the impact of the policy on consumer welfare and whether the tax is regressive.
- Main results
Estimates revealed that beverage prices affected by the soft drink tax increased by $0,28, but varied unevenly according to the size of the beverage packaging. Soft drinks with smaller packaging saw an increase of £0,241 per liter, while larger packaging experienced an increase of £0,294 per liter. Considering that the Soft Drinks Industry Levy after implementing a tax of £0,24 per liter for the sugary variants of Coca-Cola and Pepsi, these results indicated a strong pass-through of taxes to the price paid by consumers.
Preferences for price and beverages varied little among the different age groups, with the exception of older individuals preferring beverages over snacks. However, sugar preferences varied considerably with age. Individuals under 30 were more likely to prefer varieties of sugary products, while older individuals were less prone to this. For those with moderate sugar preferences, younger individuals had a significantly higher average sugar preference than those aged 22 to 30, who, in turn, preferred more sugar than those over 30.
Estimates related to consumer profiles indicated that consumers who buy soft drinks reduced, on average, the amount of sugar consumed immediately by about 245g per year, a reduction of 21%. Part of this reduction is offset by substitution with non-taxable sugar drinks and alternative snacks. Even considering behavioral responses, the compensatory variation remains high for these groups. Considering a can of Coca-Cola as a reference, an internality of at least £0,62 per can would be necessary for these consumers to experience an increase in well-being, assuming no benefit from the tax revenue.
Young individuals and those from poor families obtain more sugar from soft drinks consumed immediately, making them more affected by the soft drink tax. Estimates showed that consumers aged 13 to 21 reduce their sugar intake by an average of 280g and have a compensatory variation of £4,94. Young people and individuals in low-income households are more impacted by the tax and show greater reductions in sugar consumption. Individuals with high dietary sugar intake showed smaller responses, both in absolute and percentage terms.
- Lessons in Public Policy
This article investigated the impact of the tax on sugary drinks. Soft Drinks Industry Levy From the United Kingdom regarding sugar consumption in situations of immediate consumption. To this end, an analysis was conducted using data on individuals' daily immediate consumption patterns, considering consumers' reaction to price changes caused by the tax.
The evidence from this study highlighted that consumers are moderately sensitive to price increases due to the tax. The implementation of the tax resulted in a significant reduction in sugar consumption, especially among young people and low-income individuals, emphasizing the effectiveness of the tax in targeting these specific groups. Furthermore, the study highlighted the importance of consumption preferences and substitution with non-taxed beverages in determining consumer response. The design of this policy reveals how the price structure can induce healthy consumption changes, contributing to reduced sugar intake and improving long-term public health.
References
DUBOIS, P.; GRIFFITH, R.; O'CONNELL, M. How Well Targeted Are Soda Taxes? American Economic Review, v. 110, no. 11, p. 3661–3704, 1 Nov. 2020.
