The information in this brief is intended to provide educational information on the cost-effectiveness of SSB taxes.
Implementation of a city excise tax of $0.015/ounce of sugar-sweetened beverages (SSBs) and diet drinks, administered by the city department of revenue and based on proposals considered by federal, state, and local governments.1-4
SSBs include all beverages with added caloric sweeteners; diet drinks include any beverages with artificial sweeteners. The modeled excise tax does not apply to baby formula, medical foods, or drinks with 50% or less fruit juice or 50% milk products.5 Although SSB consumption has declined in recent years, children and adults in the U.S. consume twice as many calories from SSBs compared to 30 years ago.6-8 Randomized trials and longitudinal studies have linked SSB consumption to excess weight gain, diabetes, and cardiovascular disease. Consumption of SSBs increases the risk of chronic diseases through its impact on BMI and other mechanisms.9,10 The Dietary Guidelines for Americans, 201511recommends that individuals reduce SSB intake to manage their body weight. Drawing on the success of tobacco taxation and decades of economic research, public health experts have called for higher taxes on SSBs and documented their likely impact.12-15 In 2009, the IOM recommended that local governments implement tax strategies to reduce consumption of “calorie-dense, nutrient-poor foods,” emphasizing SSBs as an apt target for taxation.16
Increased city excise tax linked to change in BMI through change in SSB price and consumption.
Impact of Tax on Price of SSBs to Consumers
We assume 100% pass through of the tax over the ten years. Empirical studies of SSB excise taxes in Mexico and France indicate that approximately the full amount of the excise tax is passed on to consumers.17 Short term studies for the local tax in Berkeley indicate less than complete pass-through.3,18,19 The expected percent increase in SSB price was estimated based on the average $0.059/ounce reported in a review of beverage demand elasticity (inflated to $0.0612 in 2014 dollars).20 The price per ounce in this study was based on a weighted average across stores, restaurants and other sources proportional to the source of consumed SSBs in NHANES 2009-2010. The price per ounce of SSBs purchased in stores was calculated using weighted averages of two-liter bottles, 12-can cases, and single-serve bottles or cans based on the distribution of package sizes estimated from 2010 Nielsen Homescan data. The $0.015/ounce excise tax would result in a 24.5% price increase. We assumed that the tax rate would be adjusted annually for inflation to maintain the 24.5% price increase throughout the ten-year modeling time frame.
SSB Consumption and Price Elasticity of Demand
We used regionally-adjusted estimates of total SSB consumption in 2015 published in the UCONN Rudd Center Revenue Calculator for Sugar-Sweetened Beverage Taxes to adjust national age, sex, and race/ethnicity-specific consumption data from NHANES 2005-2010 to estimate current SSB consumption levels in Philadelphia.21 Powell et al reviewed studies published 2007-2012 and estimated a mean own-price elasticity of demand for soft drinks (including regular and diet) of -0.86, ranging from -1.86 to -0.41.22 Recent research concerning the Berkeley tax indicates a 21% reduction in SSB intake among low income populations. 18
Direct effect of change in SSB consumption on change in BMI
We conducted evidence reviews for impact of change in SSB intake on BMI, taking into account any dietary compensation.15 Four large longitudinal studies in adults23-26 of sufficient duration were identified. The relationship was modeled using a uniform distribution based on the range of the estimates of the effect of a one serving reduction on BMI (from 0.21 to 0.57). Among youth, a double-blind randomized controlled trial conducted over 18 months found that an additional 8 oz serving of SSBs led to a 1 kg greater weight gain.27 We have no randomized controlled trials documenting impact of a reduction in diet drink intake. Diet drinks are non-caloric, so the model assumes no effect of reduction in diet beverage intake on change in BMI.
The intervention reaches all youth and adults ages 2 years and older in Philadelphia.
The policy change will involve start up and ongoing labor costs for municipal tax department administrators. To implement the intervention, the municipal government will need to process tax statements and conduct audits. Businesses will also need to prepare tax statements and participate in audits, which will require labor from private tax accountants. Cost information was drawn from states with planned or implemented excise taxes on soft drinks.15 The cost and benefit estimates do not include expected tax revenue.
The CHOICES microsimulation model for Philadelphia was used to calculate the costs and effectiveness over ten years (2015–25). This is a stochastic, discrete-time, individual-level microsimulation model designed to simulate the experience of the Philadelphia population from 2015 to 2025. Cases of obesity prevented were calculated at the end of the model in 2025. The model uses data from: US Census, American Community Survey, Behavioral Risk Factor Surveillance System28, NHANES, National Survey of Children’s Health29, the Medical Expenditure Panel Survey, and multiple national longitudinal studies. We calculated uncertainty intervals using Monte Carlo simulations programmed in Java over one thousand iterations of the model for a population of one million simulated individuals scaled to the city population size.15
We estimated the impact of the tax-induced reduction in SSB intake on diabetes incidence for adults ages 18-79 years using a published meta-analysis of the relative risk of developing diabetes due to a one-serving change in SSB consumption30 as well as local estimates of diabetes. On average, each 8.5 oz serving per day of SSBs increases the risk of diabetes by 18%. In Philadelphia, we estimated that the proposed SSB excise tax would lead to a reduction in diabetes incidence of 8% and an estimated 968 cases of diabetes prevented over a one-year period once the tax reaches its full effect.
According to the city of Philadelphia, a 1.5 cent per ounce excise tax on SSBs and diet drinks could raise approximately $91 million anually.31
|Cost per Year with Obesity Prevented||Cost-saving|
|Cost per Quality Adjusted Life Year (QALY) Gained||Cost-saving|
|Cost per Case of Obesity Prevented||Cost-saving|
|QALYs Gained||4,060 (1,190; 8,480)|
|First Year Population Reach*||1.54 million|
|Decrease in 12-oz Servings of SSBs per Person in the First Year*||102 (56; 159)|
|Cases of Obesity Prevented*||14,300 (4,340; 30,400)|
|Years with Obesity Prevented||102,300 (31,010; 214,000)|
|Life Years Gained||1,190 (337; 2,490)|
|Deaths Averted*||349 (102; 736)|
|Annual Intervention Cost||$222,000 ($147,000; $298,000)|
|Net Cost (negative means savings)||-$76.8 million (-164 mill; -21.0 mill)|
|Health Care Cost Savings per $1 Invested||$35.60 ($9.96; $86.70)|
All metrics reported for the population over a 10-year period and discounted at 3% per year, unless otherwise noted.
Concerns have been raised regarding the impact of the tax on households with low incomes. Because of the elasticity of -0.86, our analyses indicate that households will spend less on SSBs (about 2% less) after the tax goes into effect, providing disposable income for other purchases. In addition, we project that greater health benefits will accrue to low-income consumers who on average consume more SSBs than higher income consumers; the same is true for a number of racial and ethnic groups. Disparities in obesity outcomes should thus decrease following implementation of the proposed tax. By taxing diet beverages, the policy change makes these non-caloric beverages more expensive, and hence should reduce the attractiveness of this option. In addition, revenue raised from an SSB tax can be reinvested in low income communities; for instance, in Berkeley, CA SSB tax revenue has been allocated for spending on school and community programs, several with a focus on low income or minority populations, to promote healthy eating, diabetes and obesity prevention.32,33
There is opposition from the beverage industry, which spends over $4 billion/year nationwide on marketing.34 Public support for such taxes generally increases with earmarking for prevention activities.35 Relatively small beverage excise taxes are currently applied across many states. The proposed tax is likely to be sustainable if implemented based on the history of tobacco excise taxes. There is potential for a shift in social norms of SSB consumption based on evidence from tobacco control tax and regulatory efforts.36
We project that the proposed SSB excise tax policy will prevent thousands of cases of childhood and adult obesity, prevent new cases of diabetes, increase healthy life years and save more in future health care costs than it costs to implement. Revenue from the tax can be used for education and health promotion efforts. Implementing the tax could also serve as a powerful social signal to reduce sugar consumption.
Results prepared by the CHOICES project at the Harvard T.H. Chan School of Public Health: Gortmaker SL, Long MW, Ward ZJ, Giles CM, Barrett JL, Resch SC, Cradock AL. Funded by The JPB Foundation and Healthy Food America. Results are those of the authors and not the funders. For further information, contact firstname.lastname@example.org.