writing summary

Legal and Administrative Feasibility of a Federal Junk Food and Sugar-Sweetened Beverage Tax to Improve Diet

Jennifer L. Pomeranz, JD, MPH, Parke Wilde, PhD, Yue Huang, MS, Renata Micha, RD, PhD, and Dariush Mozaffarian, MD, DrPH

Objectives. To evaluate legal and administrative feasibility of a federal “junk” food

(including sugar-sweetened beverages [SSBs]) tax to improve diet.

Methods. To assess food definitions and administration models, we systematically

searched (1) PubMed (through May 15, 2017) for articles defining foods subject to taxes,

and legal and legislative databases as well as online for (2) US federal, state, and tribal

junk food tax bills and laws (January 1, 2012–February 28, 2017); SSB taxes (January 1,

2014–February 28, 2017); and international junk food tax laws (as of February 28, 2017);

and (3) federal taxing mechanisms and administrative methods (as of February 28, 2017).

Results. Articles recommend taxing foods by product category, broad nutrient criteria,

specific nutrients or calories, or a combination. US junk food tax bills (n = 6) and laws

(n = 3), international junk food laws (n = 2), and US SSB taxes (n = 10) support taxing foods

using category-based (n = 8), nutrient-based (n = 1), or combination (n = 12) approaches.

Federal taxing mechanisms (particularly manufacturer excise taxes on alcohol) and

administrative methods provide informative models.

Conclusions. From legal and administrative perspectives, a federal junk food tax ap-

pears feasible based on product categories or combination category-plus-nutrient

approaches, using a manufacturer excise tax, with additional support for sugar and

graduated tax strategies. (Am J Public Health. 2018;108:203–209. doi:10.2105/

AJPH.2017.304159)

Poor diet is a leading risk factor for car-diovascular disease, type 2 diabetes, and obesity.1 One evidence-based strategy to improve diet is the enactment of policies to alter the price of food and beverages: disin- centives (taxes) for unhealthy food and in- centives (subsidies) for healthier food.2,3

Whereas local taxes on sugar-sweetened beverages (SSBs) are rapidly growing in pop- ularity and political acceptance, these do not have the broad effects that a national tax might have. In addition, substantial disease burden results from other dietary factors, including processed meatsandultra-processed foodshigh in starch, added sugars, salt, and trans fat.4

However, federal SSB or unhealthy “junk” food taxes have not yet gained substantial political traction in the United States. Besides concerns over regressivity and healthy food access, barriers include key unanswered policy questions, such as the complexity involved

in defining the food to be taxed and the appropriate taxing mechanisms for imple- mentation. Previous research has examined the typology of beverage tax options,3,5 global foodtaxpolicyarchitecture,6 andcompiledUS state definitions of food,7 but no study to our knowledge has evaluated the legal and ad- ministrative mechanisms to define and im- plement a US federal junk food tax to improve diet quality.

To address these gaps in knowledge, we evaluated the legal and administrative

feasibility of federal taxes on junk food based on the scientific literature, contemporary legislative experiences from the United States and other countries on defining targeted foods, US federal taxing and administrative mechanisms, and potential differences by tax purpose and type.

METHODS We systematically reviewed the scientific

literature, US and international food tax bills and laws, and federal taxing mechanisms to identify contemporary evidence on potential approaches for a federal junk food (including SSBs) tax. Previous literature identified potential methods to classify foods for a junk food tax that guided our investigation: by product category,8,9 nutrient content,3 and processing.2,10,11 We considered these def- initions, which type of tax would be most effective to reach various public health goals, and which federal taxes and mechanisms can serve as models for a junk food tax. Criteria to judge legal and administrative feasibility included achievability of implementation as demonstrated by proposed and successful past execution of similar taxes and com- patibility with other federal taxing mecha- nisms and laws. For all parts of the study, 1 author (J. L. P.) reviewed and extracted data and the remaining authors checked for consistency.

ABOUT THE AUTHORS Jennifer L. Pomeranz is with the College of Global Public Health, New York University, New York, NY. Parke Wilde, Yue Huang, Renata Micha, and Dariush Mozaffarian are with the Friedman School of Nutrition Science and Policy, Tufts University, Boston, MA.

Correspondence should be sent to Jennifer L. Pomeranz, Department of Public Health Policy and Management, College of Global Public Health, New York University, 715 Broadway, 10th Floor, New York, NY 10003 (e-mail: [email protected]). Reprints can be ordered at http://www.ajph.org by clicking the “Reprints” link.

This article was accepted September 25, 2017. doi: 10.2105/AJPH.2017.304159

February 2018, Vol 108, No. 2 AJPH Pomeranz et al. Peer Reviewed Public Health Law 203

AJPH LAW & ETHICS

Scientific Literature Review We conducted a systematic literature

search of PubMed through May 15, 2017, to identify peer-reviewed journal articles that provided in-depth discussions and sugges- tions for defining targeted products for junk food taxes. Keywords included the follow- ing: tax, unhealthy, snack, process, nutrient, product, and energy dense (Figure A, available as a supplement to the online version of this article at http://www.ajph. org). We excluded articles that focused solely on fast food taxes or beverage taxes or that evaluated 1 country’s or jurisdiction’s tax, and commentaries reporting on another publication, study protocols, and non- English–language articles. One author (J. L. P.) reviewed all abstracts and the full text of articles deemed potentially relevant. Data extracted included each consideration and recommendation related to how to define the product to be taxed. Review was supplemented with gray literature searches.

Selection of Policies To inform US federal tax policy, we

researched US and international taxes. First, using Congress.gov and the UConn Rudd Center’s Legislative Database, we researched US federal and state bills and laws to tax junk food that were proposed and passed from January 1, 2012, through February 28, 2017. Second, when a bill was proposed to explicitly amend existing law, we retrieved the law current as of February 28, 2017, using LexisNexis. Third, for in- ternational junk food taxes in effect as of February 28, 2017, we focused on Orga- nisation for Economic Co-Operation and Development (OECD) countries, consulted a 2016 World Health Organization (WHO) report,3 and conducted Internet searches to retrieve information for review. For all food tax bills and laws, we excluded legis- lation that taxed all food; solely restaurant, “prepared,” or vending-machine food; and laws conferring differentiated tax status on specific products, which is when the law excludes specific products (typically only SSBs and candy) from the definition of untaxed food. The latter category has different policy considerations, and such laws are common among US states—unlike

broad exclusions for, or taxes on, junk food.7

Fourth, because so few junk food taxes exist, we supplemented this research with US SSB tax laws to inform food tax definitions and mechanisms more broadly. We thus gathered the SSB taxes enacted between January 1, 2014, and February 28, 2017, and all federal bills proposed on this topic.

The data extracted for the tax policies included bill or law year, definition of how to identify the taxed product, exclusions from the tax, tax type (e.g., sales tax), tax rate, and, when relevant, existent law related to a bill.

Selection of Taxing Mechanisms To assess taxing mechanisms to implement

a federal junk food tax, we researched the US federal tax code using the US Master Excise Tax Guide,12 LexisNexis, and the Depart- ment of Treasury’s Web sites. First, we researched the federal taxation authorities and existent federal taxes to determine which type of tax—sales or excise—would be the most administratively feasible. Excise taxes are generally levied on the manufacture, distri- bution, or sale of commodities; revenue can be earmarked, or dedicated, to a specific purpose, and the taxed entity determines the extent to which it will pass on the tax to consumers. Sales taxes are paid by consumers and collected by sellers to remit to the taxing authority, which deposits the revenues into the general treasury.

Second, we analyzed existing federal taxes to determine if the federal government has previously defined products to be taxed similar to the category, nutrient, and pro- cessing approaches of interest here. Although Congress can amend the tax code to create new taxes, we considered compatibility with existent federal taxes a measure of adminis- trative feasibility.

RESULTS Our systematic review of the scientific

literature, US and international food tax bills and laws, and federal taxing mechanisms provided data and models to assess the legal and administrative feasibility of a federal junk food tax.

Scientific Literature Review Of 63 unique articles identified, 13 met

inclusion criteria (Figure A). Of these, 4 recommended a tax based on product cate- gory6,8,13,14; 4 on broad nutrition criteria (i.e., using more than 1 nutrient cutoff)15–18; 2 on combined category and broad19 or specific nutrient (saturated fat) criteria20; 1 on the specific nutrient sugar21; 1 on calories22; and 1 on both broad nutrition criteria and specific nutrient (sugar) approaches.23 Three publi- cations evaluated junk food taxes based on salt, sugar, and fat, and all 3 favored targeting sugar15,21,23 (Table A, available as a supple- ment to the online version of this article at http://www.ajph.org).

Similarapproacheswerealsoreflectedinthe gray literature. For example, the Streamlined Sales and Use Tax Agreement, adopted in 24 states to simplify sales tax collection, modeled taxing food by product categories (e.g., soft drinks, candy).24 WHO favors a broad nutri- tioncriteriatax,3whereastheUrban–Brookings Tax Policy Center favors a specific nutrient approach that targets sugar.25

The scientific literature also recommended a graduated taxation strategy, by which the tax increases as the nutritional quality of the food decreases. Specifically, Epstein et al. recommended that the price of food should be graduated to reflect its nutritional con- tent,16 and the Urban–Brookings Tax Policy Center recommended that the tax increase as the sugar content increases.25

In summary, the scientific literature rec- ommended 4 main methods to classify foods to be taxed: by (1) product category, (2) broad nutrition criteria, (3) specific nutrients or calories, and (4) a combination of these ap- proaches. None of the literature discussed taxing foods according to level of processing, although many articles assumed or posited that taxed foods would be processed.17,26,27

Additionally, some literature recommended a graduated taxation strategy.

Bills and Laws Legislative data provided complementary

empirical evidence on US and international junkfoodtaxesandUSSSBtaxes.IntheUnited States,8statesand1tribalgovernmentproposed or enacted junk food tax bills (n=6) and laws (n=3) during the study period; there were no federal junk food tax bills or laws (Table 1).

AJPH LAW & ETHICS

204 Public Health Law Peer Reviewed Pomeranz et al. AJPH February 2018, Vol 108, No. 2

Three bills, Texas’s tax, and Maine’s law taxed products according to category alone, with Maine’s law additionally using a pro- cessing component to identify 1 category of taxable foods (i.e., processed nuts and seeds). These 5 pieces of legislation were sales taxes

or removed tax exemptions for revenue production purposes. Examples of taxable products included SSBs, candy, processed meat snacks (e.g., jerky), and snack foods, which most often included chips, pretzels, frozen desserts, baked goods, and cereal or

granola bars (Table B, available as a supple- ment to the online version of this article at http://www.ajph.org).

By comparison, 3 bills and 1 law targeted health by incorporating nutrition consider- ations into the text and did not use sales tax

TABLE 1—Junk Food Tax Bills and Laws

Location and Bill or Law Taxed item(s) By Product Category

By Processing

By Nutrition Tax

US food tax bills and laws proposed or

enacted January 1, 2012, to February

28, 2017

Hawaii S.B. 2239 (2010)a,b Snack foods Yes Yes Yes Business privilege tax for selling snack food

(% not specified)

Maine 36 M.R.S. 1811; 36 M.R.S.

1752 (2015)c Not grocery staples Yes Yes for nuts

and seeds

No Sales tax (5.5%)

Maryland H.B. 1214 (2013)b Snack food Yes No No Eliminate sales and use tax exemption (% not

specified)

Minnesota H.F. No. 1249 (2013)b;

companion: MN SB 1205 (2013)b Snack food Yes No No Sales and use tax (% not specified)

Navajo Nation Health Diné Act of

2014a,c Minimal-to-no nutritional value food Yes No Yes Business privilege tax on gross receipts (2%)

New Mexico S.B. 416 (2017)b Minimal-to-no nutritional value food Yes No Yes Remove tax exemption for minimal-to-no

nutritional value food purchased from

a retail food store (% not specified)

New York H.B. 843 (2011)b Food and drink listed as sweets or snacks

in the US Department of Agriculture’s

national nutrient database for standard

reference

Yes No No Sales receipts (1%)

Texas Tax Code § 151.314(h) (2013)c Individual-sized portion snack items Yes No No Remove exemption from sales and use tax as

food (% not specified)

Vermont H.B. 477 (2017)b Packaged food products containing sugar No No Yes Excise tax on every distributor of food sold in

the state ($0.0005/gram of sugar)

International food tax laws implemented

as of February 28, 2017

Mexico law (2014)a,c Nonbasic foods (English documents use

the words “junk food”)

Yes No Yes Special tax over products and services (8%)

(in addition to VAT [16%], which applies to

all goods and services not exempt)

Hungary law (2011)a,c Public health product tax Yes No Yes For products produced in Hungary, the tax is

collected as part of general tax revenues

from the manufacturer (or first domestic

seller for imported foods). Hungary’s VAT

rate on most products is 27%. These taxes

are in addition to a VAT rate on most foods

of 25%. Some foods have lower rates (e.g.,

18% for dairy-, grain-, flour-, and starch-

based products, and 5% for medicines and

medical herbs).

Note. VAT = value added tax. aThe legislation expressly noted that 1 of its purposes was addressing nutrition or health. bBill. cLaw.

AJPH LAW & ETHICS

February 2018, Vol 108, No. 2 AJPH Pomeranz et al. Peer Reviewed Public Health Law 205

mechanisms. For example, Hawaii’s excise tax bill taxed snack food using a mix of category and nutritional considerations while taking processing into account. The Navajo Nation’s Health Diné Act taxed “minimal- to-no nutritional value” food using a gross- receipts tax. Although reported as a sales tax,28

a gross-receipts tax is imposed on the gross volume of business, and is thus a tax on the business activity of the seller, not on the buyer’s purchase.29 Vermont’s bill targeted certain food products containing sugar by assessing an excise tax per gram of sugar.

Internationally, Hungary and Mexico had junk food excise taxes; both used a combi- nation category and nutrient or calorie ap- proach (Table 1). Hungary’s law set forth detailed nutrient cutpoints primarily based on sugar and salt, to distinguish taxable foods within the following specific product cate- gories: soft drinks, candy, salty snacks, con- diments, and fruit jams. Mexico’s law taxed specific categories of foods—snacks, sweets, nut butters, cereal-based prepared products— when they additionally met a calorie threshold (Table B). Mexico’s combination category–calorie definition is similar to that of several US SSB taxes (Table 2).

In terms of SSB taxes, there was 1 federal bill proposed by Representative Rosa DeLauro in 2014 and 2015; it ultimately had 4 cosponsors and was popular among public health advocates but opposed by the SSB industry and died in committee both years. The bill targeted sugar in SSBs directly, by levying a tax per gram of caloric sweetener. There were 9 SSB tax laws (1 state, 7 mu- nicipal, and the Navajo Nation Act) that used a category approach (n = 3), set forth calorie or sugar thresholds totrigger the tax (n = 4), or were based on specific nutrients: caloric or artificial sugar (n = 2; Table C, available as a supplement to the online version of this article at http://www.ajph.org). The federal bill and all municipal SSB tax laws used an excise tax mechanism; except for Philadelphia’s tax, these were proposed or enacted explicitly for health or nutrition ra- tionales. The Navajo Nation and Maine SSB taxes are 1 element of their food taxes de- scribed in this section. The only SSB sales tax identified was Maine’s law, which was enacted for revenue purposes.

Two bills, the Vermont food tax and the federal SSB tax bills, used a graduated tax

strategy, by which the tax increases as the sugar content increases.

In summary, to define the taxed products, the identified bills and laws used (1) category (n = 8), (2) specific nutrient (n = 1), and (3) combination (n = 12) approaches. The most frequently targeted categories were SSBs, candy, processed meat products, and sweet and salty snacks, and the most frequently targeted nutrients were sugar, calories, and salt. No legislation used a broad nutrition criteria approach alone, targeted single nu- trients other than sugar alone, or taxed foods by level of processing.

Taxing Mechanisms Federal administrative mechanisms and tax

types can provide models for the adminis- tration and specification of a federal junk food tax. We identified no federal sales taxes. The Department of Treasury has 2 tax-collecting bureaus that administer excise taxes. The Internal Revenue Service collects excise taxes on multiple products that vary in their status as necessities or luxuries, including, for example, vaccines, fuel, and sports fishing equipment. These are revenue producing and not intended to be regulatory. Conversely, the Alcohol and Tobacco Tax and Trade Bureau (TTB) administers excise taxes on tobacco, alcohol, and firearms. The TTB also engages in regulatory activities for tobacco and alcohol (e.g., labeling) that are similar to the Food and Drug Administration (FDA) and US De- partment of Agriculture’s regulation of food.

Excise taxes may be levied at different points in the supply chain. Our research uncovered few excise taxes levied at the point of sale; a primary exception is the federal indoor tanning tax. Similarly, we found no federal taxes levied specifically on the acts of distribution or wholesale, although some wholesalers may be subject to a tax depending on other conditions. By contrast, we iden- tified excise taxes on manufacturers as a common tool for the federal government. Federal manufacturer excise taxes are applied when the manufacturer (which includes producers and importers) sells the product to the purchaser,12 who in the case of food would be distributors, wholesalers, retailers, restaurants, and food service operators.

We identified several existent federal ex- cise taxes that could provide a potential model

for a federal junk food tax. Most federal manufacturer excise taxes are category based (e.g., tires, medical devices) and enacted for revenue production purposes.12 The primary exception is federal alcoholic beverage taxes, which are revenue producing and regulatory. Certain alcohol taxes take into consideration ingredient levels (alcohol) and processing (carbonation). Specifically, wine taxes are graduated by alcohol level: wine lower in alcohol (< 14% alcohol by volume) is taxed at lower levels, $1.07 per gallon; wine that is moderate in alcohol (14%–21%) is taxed at $1.57 per gallon; and wine higher in alcohol (21%–24%) is taxed at $3.15 a gallon.30

Wine with an alcohol content higher than 24% it is taxed like distilled spirits ($13.50 per proof gallon). These alcohol percentages are also required to be disclosed on the products’ labels. Related to processing, artificially carbonated wines are taxed less ($3.30 per gallon) than sparkling wines ($3.40 per gallon) with the primary difference based on how producers create the carbon- ation.31 Both are taxed at a substantially higher rate than most still wine, even though they may have lower alcohol content.

In sum, the federal government regularly imposes federal manufacturer excise taxes. Most use a category method to identify taxable products; major exceptions are certain alcohol beverage taxes, which are based on ingredient levels and processing, and a grad- uated tax strategy on still wine, whereby the tax increases along withthe amount of alcohol in the product.

DISCUSSION This investigation provides new evidence

on the legal and administrative feasibility of imposing a federal tax on junk food. The scientific literature, empirical evidence from US and international food tax bills and laws, and existing federal taxes provide concordant evidence supporting a definition of the taxed products through (1) product- specific categories or (2) a combination of product-specific categories and multiple or individual nutrients. We identified additional support for a single nutrient (sugar) and graduated tax approach. Contemporaneous federal taxation methods include excise taxes but not sales taxes, and the food tax bills

AJPH LAW & ETHICS

206 Public Health Law Peer Reviewed Pomeranz et al. AJPH February 2018, Vol 108, No. 2

TABLE 2—US Sugar-Sweetened Beverage (SSB) Taxes Enacted 2014–2016 by State, Municipal, and Tribal Governments and Proposed or Enacted by Federal Government

Location (Bill or Law) Taxed Item(s) By Product Category By Processing By Nutrition, Calorie, or Ingredient Tax

Albany, CA lawa Sugar-sweetened beverage

product

Yes No Yes: 1 or more added caloric

sweeteners has been added and

that contains at least 2 cal/fl oz.

One cent per fluid ounce “on the

privilege of distributing” SSB products

in the city

Berkeley, CA lawa Sugar-sweetened beverage

product

Yes No No One cent per fluid ounce “on the

privilege of distributing” SSB products

in the city

Boulder, CO lawa Sugar-sweetened beverage

product

Yes No Yes: beverage thatcontains at least 5 g

of caloric sweetener/12 fl oz.

Two cents per fluid ounce on the

distribution of SSB products in the city

Cook County, IL lawa Sweetened beverages Yes No No $0.01 per ounce imposed on the retail

sale in the county of all sweetened

beverages and per ounce of sweetened

beverage produced from syrup or

powder

Federal bill H.R. 1687

(2015); H.R. 5279 (2014)

SWEET ACTa

Sugar-sweetened beverage

product

SSBs

generally

No Yes: by gram of caloric sweetener One cent per 4.2 grams of caloric

sweetener contained in the SSB

product; tax on the sale or transfer of

the SSB product “by the manufacturer,

producer, or importer thereof”

Maine law Soft drinks and powdered

and liquid drink mixes

Yes No No Sales tax (5.5%)

Oakland, CA, lawa Sugar-sweetened beverage

product

Yes No Yes: 1 or more caloric sweeteners are

added and contains ‡ 25 cal/12 fl oz of beverage

One cent general excise tax per fluid

ounce on the “privilege of

distributing” SSB products in the city

Navajo Nation Health

Diné Act of 2014 lawa Sweetened beverages Yes No Yes: high in sugar Business privilege tax on gross receipts

(2%)

Philadelphia, PA, law Sugar-sweetened beverages Yes No Yes: any form of caloric sugar or

artificial sugar substitute

A tax is imposed on each of the

following: the supply of any SSB to

a dealer; the acquisition of any SSB by

a dealer; the delivery to a dealer in the

city of any SSB; and the transport of

any SSB into the city by a dealer

Tax rates: (a) for SSBs, 1½ cents ($0.015)

per fluid ounce; (b) for syrups and

other concentrates, the rate per ounce

of syrup or other concentrate that

yields 1½ cents ($0.015) per fluid

ounce on the resulting beverage,

prepared to the manufacturer’s

specifications

San Francisco, CA, lawa Bottled sugar-sweetened

beverage, syrup, or powder

Yes No Yes: contains added caloric sweetener

and contains > 25 cal/12 fl oz of beverage

One cent general excise tax per fluid

ounce of a bottled SSB and SSB that

could be produced from syrup or powder

upon the initial distribution of the

bottled SSB or syrup or powder

aThe legislation expressly noted that 1 of its purposes was addressing nutrition or health.

AJPH LAW & ETHICS

February 2018, Vol 108, No. 2 AJPH Pomeranz et al. Peer Reviewed Public Health Law 207

and laws with a nutritional purpose over- whelmingly use an excise tax mechanism. Among excise tax options, federal taxes and taxing mechanisms support the legal and administrative feasibility of a manufacturer excise tax for junk food.

Several considerations might explain the identified metrics for defining taxed food items. Taxing products by category allows specific focus on the most problematic products in the American diet, based on their levels of consumption and their links to poor health. Such a tax may also be the easiest administratively because it does not require the additional evaluation of the nutrition composition of the taxable products and aligns with the majority of federal excise taxes, which are category based.8,9 Potential dis- advantages of a pure category approach are that it would not encourage reformulation nor facilitate a graduated tax strategy, and it could capture healthier products within taxable categories.

By comparison, a combination of category and nutrition criteria retains many of the advantages of a category tax yet at least partly addresses these limitations. For example, adding nutrient criteria allows for a graduated tax or exceptions for healthier choices within categories, which also promotes industry reformulations.25 Such a tax could be more administratively complex, in that the gov- ernment and manufacturers would be re- quired to consult nutrition-related data points to confirm and track compliance initially and over time as products change. Nonetheless, taxes with a nutrition or health purpose overwhelmingly used this method, suggesting that the strengths of this type of product definition are outweighed by the limitations.

We identified a relative discordance be- tween the frequency of expert recommen- dations to focus on sugar15,21,23,25,32 and the few proposed food tax bills or laws that fo- cused on this single ingredient. This novel finding highlights the need to investigate and understand the reasons behind this discor- dance. The reasons could include, for ex- ample, the national focus on taxing SSBs, which is a major source of added sugar in the diet; the relative newness of public focus on sugar in foods; the fact that sugar is often added to products deemed healthy or nec- essary (e.g., cereals, yogurts, breads); and the complexities of defining and determining

whether to tax total versus “free” versus added sugars.

Interestingly, the 2 bills that taxed sugar directly were also the only 2 that included a graduated strategy, by which higher sugar products would be taxed more than lower sugar alternatives. Graduated taxes may be a promising method to guide consumers toward healthier products and encourage industry reformulation, and have also been suggested by economists in other health contexts.33 In addition, a graduated junk food tax seems administratively feasible on the basis of identified evidence on existing federal graduated alcohol beverage taxes.

Only 1 article and no tax bill or law focused on calories alone. This is consistent with scientific evidence that foods rich in healthy fats (e.g., nuts) have more calories per gram than much less healthful foods high in starch and sugar, and that calorie content alone is a poor predictor of which foods are linked to long-term weight gain.4 Thus, a tax on “high calorie” foods would not necessarily align with improving diet quality or averting diet-based diseases. Similarly, we did not identify a fat tax except for the history of the failed implementation of Denmark’s satu- rated fat tax; such a tax raises political concerns about taxing staple foods like milk and cheese and scientific concerns about defining satu- rated fat as inherently unhealthy regardless of its amount or food source.34 We also did not identify examples of taxes based on sodium alone or trans fats alone or within nutrition criteria. The FDA banned partially hydro- genated oils, the major source of artificial trans fats, effective 2018; given the health impli- cations of this additive, implementation should be monitored and potential trans fat taxes considered accordingly.

We did not identify legislative precedent for a tax defined by multiple nutrient crite- ria in the absence of food categories. This is likely because such a tax would apply to all products in the food supply, dramatically increasingly complexity and administrative burden,15 while also raising challenges for scientific validity given the increased likeli- hood of misclassifying foods (for tax purposes) as more or less healthy based on nutrients alone. We also identified discordance be- tween previous research recommendations to target “processed” foods and actual empirical tax bills and laws, perhaps for similar reasons.

In terms of administration, manufacturer excise taxes on food are most in line with existing federal taxes. The federal govern- ment does not have sales taxes, but it does administer dozens of manufacturer excise taxes. One advantage of such administration is that the composition of the taxable prod- ucts can be clearly specified to manufacturers, who then have the opportunity to refor- mulate if nutrition criteria are incorporated into the tax.6,25 Additionally, manufacturers of taxed products are subject to the tax re- gardless of whether the food is sold to retail, wholesale, school food program, or restaurant buyers.11

There is little federal precedent for a tax at the point of sale, which additionally impli- cates policy concerns for the Supplemental Nutrition Assistance Program,35 or for dis- tribution or wholesale taxes, even though municipal SSB excise taxes were pre- dominantly imposed on distribution. Federal distribution and wholesale taxes might en- courage manufacturers to sell directly to re- tailers to avoid such taxes.

We evaluated the legal and administrative feasibility of a federal junk food tax, not the political feasibility. Political feasibility in the current political climate is uncertain and seems unlikely. However, compared with legal or administrative feasibility, political considerations can evolve quickly depending on current events, public attention, election results, and shifting alliances.

Strengths and Limitations The strengths of this investigation are that

it evaluated diverse sources of evidence, in- cluding the scientific literature, US bills and laws, international laws, and federal taxing mechanisms. To our knowledge, this is the first study to analyze the legal and adminis- trative feasibility of defining a federal junk food tax on the basis of the concordance of these complementary types of evidence.

Limitations should be considered. We might not have captured all relevant bills or laws. Related to taxes targeting sugar, this investigation did not analyze the feasibility of assessing a tax directly on sugar processers and refiners.15 Additional potential issues outside the scope of the present investigation should be considered in future studies, in- cluding the regressivity of such taxes, access to

AJPH LAW & ETHICS

208 Public Health Law Peer Reviewed Pomeranz et al. AJPH February 2018, Vol 108, No. 2

healthy food, complements and substitutes, the need for concurrent incentives or sub- sidies for healthier food, the optimal levels of taxation, and the best use of tax revenues.

Public Health Implications A fundamental premise of public health

is that economic and social environments can influence food choice in beneficial and harmful directions. Our finding that a federal manufacturer excise tax on junk food— definedthroughproductcategoryorcombined category–nutrient approaches—appears to be legally and administratively feasible has strong implications for nutrition policy. Junk food taxes have the potential to substantially reduce the disease burden that results from unhealthy food and beverage consumption in the United States.

CONTRIBUTORS J. L. Pomeranz, R. Micha, and D. Mozaffarian concep- tualized the article. J. L. Pomeranz conducted the legal research and literature review, extracted the data, and wrote the first draft, and all authors constructively ana- lyzed and edited the draft and contributed to the content. R. Micha and D. Mozaffarian made equal contributions as senior authors.

ACKNOWLEDGMENTS Funding for this research was provided by the National Institutes of Health, National Heart, Lung, and Blood Institute (R01HL130735; R. M., principal investigator).

WethankUditP.Modi,KeyaM.Patel,andEvaWilson for their thorough research assistance.

HUMAN PARTICIPANT PROTECTION Institutional review board approval was not needed be- cause there were no human participants.

REFERENCES 1. Micha R, Peñalvo JL, Cudhea F, Imamura F, Rehm CD, Mozaffarian D. Association between dietary factors and mortality from heart disease, stroke, and type 2 di- abetes in the United States. JAMA. 2017;317(9):912–924.

2. Mozaffarian D, Rogoff KS, Ludwig DS. The real cost of food: can taxes and subsidies improve public health? JAMA. 2014;312(9):889–890.

3. World Health Organization. Fiscal policies for diet and the prevention of noncommunicable diseases. Technical Meeting Report. Geneva, Switzerland, May 5–6, 2015. Available at: http://apps.who.int/iris/bitstream/10665/ 250131/1/9789241511247-eng.pdf?ua=1. Accessed August 14, 2017.

4. Mozaffarian D. Dietary and policy priorities for cardiovascular disease, diabetes, and obesity: a compre- hensive review. Circulation. 2016;133(2):187–225.

5. Chriqui JF, Chaloupka FJ, Powell LM, Eidson SS. A typology of beverage taxation: multiple approaches for obesity prevention and obesity prevention-related revenue generation. J Public Health Policy. 2013;34(3): 403–423.

6. Caraher M, Cowburn G. Taxing food: implications for public health nutrition. Public Health Nutr. 2005;8(8): 1242–1249.

7. Bridging the Gap. Compilation of state definitions of food and food products; January 1, 2014. University of Illinois at Chicago, 2014. Available at: http://www. bridgingthegapresearch.org/_asset/ls7nqv/Food- Definitions-January-1-2014.pdf. Accessed November 16, 2017.

8. Franck C, Grandi SM, Eisenberg MJ. Taxing junk food to counter obesity. Am J Public Health. 2013;103(11): 1949–1953.

9. Jacobson MF, Brownell KD. Small taxes on soft drinks and snack foods to promote health. Am J Public Health. 2000;90(6):854–857.

10. Martínez Steele E, Popkin BM, Swinburn B, Monteiro CA. The share of ultra-processed foods and the overall nutritional quality of diets in the US: evidence from a nationally representative cross-sectional study. Popul Health Metr. 2017;15(1):6.

11. Pomeranz JL. Taxing food and beverage products: a public health perspective and a new strategy for pre- vention. Univ Mich J Law Reform. 2013;46(3):999–1027.

12. US Master Excise Tax Guide. 8th ed. Chicago, IL: Commerce Clearing House; 2012.

13. Powell LM, Chaloupka FJ. Food prices and obesity: evidence and policy implications for taxes and subsidies. Milbank Q. 2009;87(1):229–257.

14. Mytton O, Gray A, Rayner M, Rutter H. Could targeted food taxes improve health? J Epidemiol Com- munity Health. 2007;61(8):689–694.

15. Thow AM, Downs S, Jan S. A systematic review of the effectiveness of food taxes and subsidies to improve diets: understanding the recent evidence. Nutr Rev. 2014;72(9): 551–565.

16. Epstein LH, Finkelstein E, Raynor H, et al. Exper- imental analysis of the effect of taxes and subsides on calories purchased in an on-line supermarket. Appetite. 2015;95:245–251.

17. Bishai D. Generalized nutrient taxes can increase consumer welfare. Health Econ. 2015;24(11):1517–1522.

18. Clark JS, Dittrich LO, Xu Q. Empirical evidence of the efficiency and efficacy of fat taxes and thin subsidies. Cent Eur J Public Health. 2014;22(3):201–206.

19. Bond ME, Williams MJ, Crammond B, Loff B. Taxing junk food: applying the logic of the Henry tax review to food. Med J Aust. 2010;193(8):472–473.

20. Madden D. The poverty effects of a “fat-tax” in Ireland. Health Econ. 2015;24(1):104–121.

21. Cobiac LJ, Tam K, Veerman L, Blakely T. Taxes and subsidies for improving diet and population health in Australia: a cost-effectiveness modelling study. PLoS Med. 2017;14(2):e1002232.

22. Nederkoorn C, Havermans RC, Giesen JC, Jansen A. High tax on high energy dense foods and its effects on the purchase of calories in a supermarket: an experiment. Appetite. 2011;56(3):760–765.

23. Harding M, Lovenheim M. The effect of prices on nutrition: comparing the impact of product- and nutrient-specific taxes. J Health Econ. 2017;53:53–71.

24. Streamlined Sales Tax Governing Board. Streamlined Sales and Use Tax Agreement. December 16, 2016. Available at: http://www.streamlinedsalestax.org/ uploads/downloads/Archive/SSUTA/SSUTA%20As% 20Amended%202016-12-16.pdf. Accessed November 16, 2017.

25. Marron DB, Gearing ME, Iselin J. Should we tax unhealthy foods and drinks? Urban–Brookings Tax Policy Center, 2015. Available at: https://www.urban. org/sites/default/files/alfresco/publication-pdfs/ 2000553-Should-We-Tax-Unhealthy-Foods-and- Drinks.pdf. Accessed November 16, 2017.

26. Niebylski ML, Redburn KA, Duhaney T, Campbell NR. Healthy food subsidies and unhealthy food taxation: a systematic review of the evidence. Nutrition. 2015;31(6): 787–795.

27. Wiggins S, Keats S, Han E, Shimokawa S, Hernández J, Claro R. The rising cost of a healthy diet: changing relative prices of foods in high-income and emerging economies. Overseas Development Institute Report May. Executive Summary, May 2015. Available at: https://www.odi.org/sites/odi.org.uk/files/odi-assets/ publications-opinion-files/9579.pdf. Accessed Novem- ber 16, 2017.

28. Clark L. The Navajo Nation will soon have the country’s first-ever junk-food tax. Mother Jones. March 25, 2015. Available at: http://www.motherjones.com/ environment/2015/03/navajo-nation-junk-food-tax. Accessed August 14, 2017.

29. Amdur JA. Constitutionality, construction, and ap- plication of state and local public-utility-gross-receipts- tax statutes–modern cases. ALR 5th; 58(187):1–55.

30. 26 USC §5041.

31. 26 CFR 24.10.

32. Nnoaham KE, Sacks G, Rayner M, Mytton O, Gray A. Modelling income group differences in the health and economic impacts of targeted food taxes and subsidies. Int J Epidemiol. 2009;38(5):1324–1333.

33. Chaloupka FJ, Sweanor D, Warner KE. Differential taxes for differential risks—toward reduced harm from nicotine-yielding products. N Engl J Med. 2015;373(7): 594–597.

34. Micha R, Mozaffarian D. Saturated fat and cardio- metabolic risk factors, coronary heart disease, stroke, and diabetes: a fresh look at the evidence. Lipids. 2010;45(10): 893–905.

35. Pomeranz JL. Implications of the Supplemental Nutrition Assistance Program tax exemption on sugar- sweetened beverage taxes. Am J Public Health. 2015; 105(11):2191–2193.

AJPH LAW & ETHICS

February 2018, Vol 108, No. 2 AJPH Pomeranz et al. Peer Reviewed Public Health Law 209

Reproduced with permission of copyright owner. Further reproduction prohibited without permission.