OuncesofPrevention-ThePublicPolicyCaseforTaxesonSuggaredBeverages.pdf

Perspective

The NEW ENGLAND JOURNAL of MEDICINE

april 30, 2009

1805n engl j med 360;18 nejm.org april 30, 2009

The obesity epidemic has in-spired calls for public health

measures to prevent diet-related diseases. One controversial idea is now the subject of public debate: food taxes.

Forty states already have small taxes on sugared beverages and snack foods, but in the past year, Maine and New York have pro-posed large taxes on sugared bev-erages, and similar discussions have begun in other states. The size of the taxes, their potential for generating revenue and reduc-ing consumption, and vigorous opposition by the beverage indus-try have resulted in substantial

controversy. Because excess con-sumption of unhealthful foods underlies many leading causes of death, food taxes at local, state, and national levels are likely to remain part of political and pub-lic health discourse.

Sugar-sweetened beverages (soda sweetened with sugar, corn syrup, or other caloric sweeteners and other carbonated and uncar-bonated drinks, such as sports and energy drinks) may be the single largest driver of the obe-sity epidemic. A recent meta-analysis found that the intake of sugared beverages is associated with increased body weight, poor

nutrition, and displacement of more healthful beverages; in-creasing consumption increases risk for obesity and diabetes; the strongest effects are seen in stud-ies with the best methods (e.g., longitudinal and interventional vs. correlational studies); and in-terventional studies show that re-duced intake of soft drinks im-proves health.1 Studies that do not support a relationship between consumption of sugared bever-ages and health outcomes tend to be conducted by authors support-ed by the beverage industry.2

Sugared beverages are market-ed extensively to children and adolescents, and in the mid-1990s, children’s intake of sugared bev-erages surpassed that of milk. In the past decade, per capita intake of calories from sugar-sweetened beverages has increased by nearly 30% (see bar graph)3; beverages now account for 10 to 15% of the

Ounces of Prevention — The Public Policy Case for Taxes on Sugared BeveragesKelly D. Brownell, Ph.D., and Thomas R. Frieden, M.D., M.P.H.

Sugar, rum, and tobacco are commodities which are nowhere necessaries of life, which are become objects of almost universal consumption, and which are therefore extremely proper subjects of taxation.

Adam Smith, The Wealth of Nations, 1776

PERSPECTIVE

1806 n engl j med 360;18 nejm.org april 30, 2009

calories consumed by children and adolescents. For each extra can or glass of sugared beverage consumed per day, the likelihood of a child’s becoming obese in-creases by 60%.4

Taxes on tobacco products have been highly effective in reducing consumption, and data indi-cate that higher prices also reduce soda consumption. A review conducted by Yale University’s Rudd Center for Food Policy and Obesity sug-gested that for every 10% in-crease in price, consumption decreases by 7.8%. An indus-try trade publication report-ed even larg er reductions: as prices of carbonated soft drinks increased by 6.8%, sales dropped by 7.8%, and as Coca-Cola prices increased by 12%, sales dropped by 14.6%.5 Such studies — and the econom-ic principles that support their findings — suggest that a tax on sugared beverages would encour-age consumers to switch to more healthful beverages, which would lead to reduced caloric in-take and less weight gain.

The increasing affordability of soda — and the decreasing af-fordability of fresh fruits and veg-etables (see line graph) — proba-bly contributes to the rise in obesity in the United States. In 2008, a group of child and health care advocates in New York pro-posed a one-penny-per-ounce ex-cise tax on sugared beverages, which would be expected to re-duce consumption by 13% — about two servings per week per person. Even if one quarter of the calories consumed from sugared beverages are replaced by other food, the decrease in consump-tion would lead to an estimated reduction of 8000 calories per

person per year — slightly more than 2 lb each year for the aver-age person. Such a reduction in calorie consumption would be ex-pected to substantially reduce the risk of obesity and diabetes and may also reduce the risk of heart disease and other conditions.

Some argue that government should not interfere in the mar-ket and that products and prices will change as consumers demand more healthful food, but several considerations support govern-ment action. The first is exter-nality — costs to parties not di-rectly involved in a transaction. The contribution of unhealthful diets to health care costs is al-ready high and is increasing — an estimated $79 billion is spent annually for overweight and obe-sity alone — and approximately half of these costs are paid by Medicare and Medicaid, at taxpay-ers’ expense. Diet-related diseas-es also cost society in terms of decreased work productivity, in-creased absenteeism, poorer school performance, and reduced fitness on the part of military recruits, among other negative effects.

The second consideration is in-formation asymmetry between the parties to a transaction. In

the case of sugared beverages, marketers commonly make health claims (e.g., that such beverages provide energy or vitamins) and use techniques that exploit the cognitive vulnerabilities of young children, who often cannot dis-tinguish a television program

from an advertisement.A third consideration is

revenue generation, which can further increase the societal benefits of a tax on soft drinks. A penny-per-ounce ex-cise tax would raise an esti-mated $1.2 billion in New York State alone. In times of economic hardship, taxes that both generate this much rev-enue and promote health are better options than revenue initiatives that may have ad-verse effects.

Objections have certainly been raised: that such a tax would be regressive, that food taxes are not comparable to to-bacco or alcohol taxes because people must eat to survive, that it is unfair to single out one type of food for taxation, and that the tax will not solve the obesity problem. But the poor are dis-proportionately affected by diet-related diseases and would derive the greatest benefit from reduced consumption; sugared beverages are not necessary for survival; Americans consume about 250 to 300 more calories daily today than they did several decades ago, and nearly half this increase is accounted for by consumption of sugared beverages; and though no single intervention will solve the obesity problem, that is hard-ly a reason to take no action.

The full impact of public poli-cies becomes apparent only after they take effect. We can estimate changes in sugared-drink con-

Ounces of Prevention — The Public Policy Case for Taxes on Sugared Beverages

1x col

AUTHOR:

FIGURE

JOB: ISSUE:

4-CH/T

RETAKE 1st2nd

SIZE

ICM

CASE

EMail LineH/TCombo

Revised

AUTHOR, PLEASE NOTE:Figure has been redrawn and type has been reset.

Please check carefully.

REG F3rd

EnonARTIST:

Brownell

1 of 2

04-30-09

36018

Dai

ly C

alor

ic In

take

from

Sug

ar-

Swee

tene

d D

rink

s in

the

U.S

. 225

100

125

150

175

200

141

190

1977–78 1994–96 1990–2000

Daily Caloric Intake from Sugar-Sweetened Drinks in the United States.

Data are from Nielsen and Popkin.3

PERSPECTIVE

1807n engl j med 360;18 nejm.org april 30, 2009

sumption that would be prompt-ed by a tax, but accompanying changes in the consumption of other foods or beverages are more difficult to predict. One question is whether the proportions of calories consumed in liquid and solid foods would change. And shifts among beverages would have different effects depending on whether consumers substi-tuted water, milk, diet drinks, or equivalent generic brands of sug-ared drinks.

Effects will also vary depend-ing on whether the tax is de-signed to reduce consumption, generate revenue, or both; the size of the tax; whether the revenue is earmarked for programs relat-ed to nutrition and health; and where in the production and dis-tribution chain the tax is applied. Given the heavy consumption of sugared beverages, even small taxes will generate substantial revenue, but only heftier taxes will significantly reduce consumption.

Sales taxes are the most com-mon form of food tax, but be-cause they are levied as a per-centage of the retail price, they encourage the purchase of less-expensive brands or larger con-tainers. Excise taxes structured as a fixed cost per ounce provide an incentive to buy less and hence would be much more effective in reducing consumption and im-proving health. In addition, man-ufacturers generally pass the cost of an excise tax along to their customers, including it in the price consumers see when they are making their selection, where-as sales taxes are seen only at the cash register.

Although a tax on sugared beverages would have health ben-efits regardless of how the reve-nue was used, the popularity of such a proposal increases great-ly if revenues are used for pro-grams to prevent childhood obe-sity, such as media campaigns, facilities and programs for phys-

ical activity, and healthier food in schools. Poll results show that support of a tax on sugared bev-erages ranges from 37 to 72%; a poll of New York residents found that 52% supported a “soda tax,” but the number rose to 72% when respondents were told that the revenue would be used for obe-sity prevention. Perhaps the most defensible approach is to use rev-enue to subsidize the purchase of healthful foods. The public would then see a relationship be-tween tax and benefit, and any regressive effects would be coun-teracted by the reduced costs of healthful food.

A penny-per-ounce excise tax could reduce consumption of sug-ared beverages by more than 10%. It is difficult to imagine produc-ing behavior change of this mag-nitude through education alone, even if government devoted mas-sive resources to the task. In con-trast, a sales tax on sugared drinks would generate considerable rev-

Ounces of Prevention — The Public Policy Case for Taxes on Sugared Beverages

3x col

AUTHOR:

FIGURE

JOB: ISSUE:

4-CH/T

RETAKE 1st

2nd

SIZE

ICM

CASE

EMail LineH/TCombo

Revised

AUTHOR, PLEASE NOTE:Figure has been redrawn and type has been reset.

Please check carefully.

REG F3rd

EnonARTIST:

Brownell

2 of 2

04-30-09

36018

Pric

e In

dex

(198

2–19

84=

100)

350

100

150

200

250

300

Fresh fruits andvegetables

Consumer-price index

Sugar and sweets

Carbonated drinks

1979

1981

1983

1985

1989

1991

1993

1995

1999

2001

2003

2005

2007

2009

1997

1987

Relative Price Changes for Fresh Fruits and Vegetables, Sugars and Sweets, and Carbonated Drinks, 1978–2009.

Data are from the Bureau of Labor Statistics and represent the U.S. city averages for all urban consumers in January of each year.

PERSPECTIVE

1808 n engl j med 360;18 nejm.org april 30, 2009

enue, and as with the tax on to-bacco, it could become a key tool in efforts to improve health.

No potential conflict of interest relevant to this article was reported.

Dr. Brownell is a professor and director of the Rudd Center for Food Policy and Obe-sity, Yale University, New Haven, CT. Dr. Frieden is the health commissioner for the City of New York.

This article (10.1056/NEJMp0902392) was published at NEJM.org on April 8, 2009.

Vartanian LR, Schwartz MB, Brownell 1. KD. Effects of soft drink consumption on nutrition and health: a systematic review and meta-analysis. Am J Public Health 2007;97:667-75.

Forshee RA, Anderson PA, Storey ML. 2. Sugar-sweetened beverages and body mass index in children and adolescents: a meta-analysis. Am J Clin Nutr 2008:87:1662-71.

Nielsen SJ, Popkin BM. Changes in bever-3. age intake between 1977 and 2001. Am J Prev Med 2004;27:205-10. [Erratum, Am J Prev Med 2005;28:413.]

Ludwig DS, Peterson KE, Gortmaker SL. 4. Relation between consumption of sugar-sweetened drinks and childhood obesity: a prospective, observational analysis. Lancet 2001;357:505-8.

Ounces of Prevention — The Public Policy Case for Taxes on Sugared Beverages

global health

Rationing Antiretroviral Therapy in Africa — Treating Too Few, Too LateNathan Ford, D.H.A., Edward Mills, Ph.D., and Alexandra Calmy, M.D.

Related article, p. 1815

The past 6 years have seen striking advances in access

to antiretroviral therapy in Africa. From 2002 onward, the interna-tional drive to scale up antiret-roviral treatment gained consid-erable momentum, most notably with the establishment of the Global Fund to Fight AIDS, Tu-berculosis, and Malaria, the “3 by 5” Initiative of the World Health Organization (WHO), and the U.S. President’s Emergency Plan for AIDS Relief (PEPFAR). Today, an estimated 3 million people in the developing world are receiv-ing antiretroviral therapy.

The momentum has now be-gun to wane, with various groups arguing that the focus on AIDS has had its day and that health care funding should now be re-directed to other areas, such as maternal and child health and primary care. But before the in-ternational community gives up on prioritizing care for patients with HIV infection, we believe that on-the-ground discussions must address not only whether enough has been done to scale up treatment but also whether

the treatment that patients are receiving is good enough.

The standard approach to HIV treatment in Africa is to wait un-til people are visibly sick, treat them with effective but poorly tolerated drugs, and then wait until they are sick again before switching regimens. There are sev-eral problems with this approach.

The first is that too few peo-ple are receiving treatment. The 3 million people receiving anti-retroviral therapy are usually said to account for about 30% of the need for such treatment, but even this rate reflects the use of stringent eligibility criteria that have been abandoned in wealth-ier countries.

Second, we are waiting until people are symptomatic before they are treated. In most African countries, patients begin receiv-ing treatment when the CD4+ count falls below 200 cells per cubic millimeter, at which point most patients already have symptomatic and severe (WHO stage 3 or 4) infection. In the United States and Europe, treat-ment is initiated earlier — as

soon as the CD4+ count reaches 350 cells per cubic millimeter — and increasingly, experts are ar-guing that even that is too late.

In many patients in Africa, the CD4+ count takes only about a year to decline from the cutoff for such early initiation to that for the later initiation now prac-ticed in developing countries.1 Although delaying therapy may mean saving money on drugs during this period, the long-term cost of such delays is in-creased substantially by the need for more intensive clinical care, decreased income, and likely regimen switches. Cost is thus no longer a tenable justification for delaying therapy. More im-portant, recent observational data presented by Kitahata et al. in this issue of the Journal (pages 1815–1826) show that the risk of death increases by 69% when the initiation of therapy is delayed until the CD4+ count drops be-low 350 cells per cubic millime-ter. Patients’ immunologic nadir — how low their CD4+ count is allowed to drop — is predictive of the degree of benefit they will