## htnea2021008

## Source details

**Canonical URL:** [htnea2021008](https://www.imf.org/-/media/files/publications/howtonotes/2021/english/htnea2021008.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/howtonotes/2021/english/htnea2021008.pdf.md)
- [Structured JSON version](/-/media/files/publications/howtonotes/2021/english/htnea2021008.pdf.json)

---

### The Rise of Obesity-Related Diseases
- Shift from communicable to non-communicable diseases driven by tobacco, alcohol, physical inactivity, and unhealthy diets (Jamison and others 2013).
- Encouraging physical activity and curbing tobacco consumption have clear health benefits (WHO 2019a).
- Obesity and overweight directly responsible for at least 2.8 million deaths annually (WHO 2019b).
- Obesity represented $147 billion (1 percent of GDP) in the United States in 2008 (Finkelstein 2009).
- In the USA, 30.3 million people had diabetes in 2015, and 84.1 million prediabetes (CDC 2017).
- Obesity-related costs represented 9.1 percent of health care expenditures in the USA (Finkelstein 2009).

### Case for Excise Taxes on Unhealthy Foods and Beverages
- Objectives beyond revenue: address negative externalities, internalities, imperfect information; revenue-raising; public health.
- Excises valued for administrative simplicity, equity, and efficiency trade-offs.
- Key constraints: complex tax bases (wide variety of products and sugar sources), administrative feasibility in low-capacity environments, legal and jurisdictional limits.

### Economic Rationales: Externalities and Internalities
- Externality channel: health insurance pooling of risk; magnitude depends on institutional factors.
- Bhattacharya and Sood (2005) estimate per capita externality costs of $149 in 1998.
- Finkelstein (2009) estimates burden of obesity for 2008 at $488 per person, or $147 billion in the United States.
- Allcott and others (2019a) propose a tax of 0.41 cents per ounce of SSBs to cover third-party externality costs (roughly 10 to 20 percent of the sale price).
- Marginal cost (Pigouvian) tax level has not been assessed.
- Internality-based estimates: Allcott and others (2019a) estimate an internality-based optimal tax of roughly 1.1 cents per ounce for SSBs.
- O’Donoghue and Rabin (2006) show optimal taxes can be large even with few individuals with low self-control.

### Price Elasticities and Revenue Implications (numeric estimates preserved)
- Andreyeva and others (2010) review of 160 US studies: soft drinks average elasticity –0.79 (confidence interval –0.33 to –1.24; individual estimates range –0.13 to –3.18); juice –0.76; sweets/sugars –0.34.
- Cornelsen and others (2014) review of 78 studies across 38 countries: comparable results; higher own-price effects in low-income countries.
- Powell and others (2013): own-price elasticities of SSB and fast-food are –1.21 and –0.52, respectively.
- Nakhimovsky and others (2016): own-price SSB elasticities ranging from –0.6 to –1.2 in 9 middle-income countries.
- Allcott and others (2019a): estimate an elasticity of about –1.4 for SSBs.
- Conclusion: the case for stable revenue-raising via taxing obesity-related food is weak.

### Health Impacts, Substitution Effects, and Evidence Limits
- Medical research links obesity to small sustained dietary imbalances (a 5- to 10-percent increase in daily caloric intake; e.g., one or two cans of soft drinks).
- Key contributors to weight gain: soft drinks, chips and potatoes, unprocessed red meat, processed meat.
- SSBs may account for up to half of dietary imbalances in the United States.
- Empirical findings on taxes and health:
  - Taxes can decrease consumption of taxed goods, but ultimate impact on obesity and health outcomes is uncertain and likely small.
  - No studies yet document a link from an excise tax to a health outcome (Shekar and Popkin 2020).
- Substitution evidence (heterogeneous):
  - Edward (2011): simulations that include substitution effects show much lower impacts on Body-Mass Index.
  - Cornelsen and others (2014): cross-price elasticities tend to be significantly lower than own-price elasticities.
  - Harding and Lovenheim (2017): a 20 percent tax on SSBs reduces purchases by 10.35 percent, but related caloric intake by only 4.84 percent (considering substitution).
  - Dharmasena and Capps (2012): taxes on SSBs impact sales of juices, low-fat milk, coffee, and tea.
  - Mexico’s experience shows uncertain substitution patterns and health impacts despite significant drops in sales of taxed products.

### Policy Design Conditions for Effectiveness
- Conditions for impact:
  - Link obesity to one or a few clearly identifiable food items driving excessive caloric intake (added sugar, especially in SSBs).
  - Induce substitution away from high-calorie foods; better documentation of substitution patterns needed.
  - Combine taxes with complementary non-tax measures and multisectoral engagement.
  - Awareness campaigns can amplify tax effects.
- Design trends:
  - Recent health-motivated tax proposals often involve moderate hikes (for example, one cent per ounce of soft drinks, i.e., 15 to 20 percent of the value) on a few well-selected high-calorie products.
- Policy alignment:
  - Address inconsistencies between VAT and excise treatment (for example, reduced VAT on SSBs undermines excise objectives).
  - Consider tiered sugar-based specific excises to incentivize reformulation.

### Selected Country Cases and Empirical Findings
- Mexico (policy adopted 31 October 2013):
  - Excise tax of 1 Mexican peso per liter of SSB (equivalent on average to about 10 percent of the sales price); covers energy drinks and non-alcoholic beverages prepared by dissolving sugars in water.
  - Additional 8 percent tax on 9 categories of non-staple high-calorie processed foods with more than 275 calories per 100 grams.
  - Embedded regulatory adjustment related to cumulative inflation.
  - Observed impacts versus baseline:
    - Taxed products (soda and non-soda SSBs) decline in sales volume: 5.5 percent in 2014 and 9.7 percent in 2015.
    - Non-soda taxed products: –16.2 percent in 2014 and –29.4 percent in 2015.
    - Soda taxed products: –0.8 percent in 2014 and –0.3 percent in 2015.
    - Sales of non-taxed products increased by 2.1 percent on average over the study period.
    - Untaxed dairy products declined 14.4 percent on average over 2014 and 2015; taxed dairy products fell 2.5 percent on average (shorter data series for untaxed dairy).
    - Effect of the tax significantly higher among lower socio-economic classes.
    - In 2014, tax revenue from sweet beverages were 0.1 percent of GDP.
- Denmark (saturated fat tax; repealed after 15 months):
  - Specific tax: 16 kroner (EUR 2.15) per kilogram of saturated fat on food with fat content above 2.3 percent.
  - Reported 10 to 15 percent decline in consumption of targeted items; observed shift for some oils and fats from high-price supermarkets to low-price discount stores.
  - Repeal reasons: high administrative costs for producers, alleged cross-border shopping, disputed effects on consumption habits.
- Hungary (Public Health Product Tax, PHPT; introduced September 2011):
  - Example rates as of January 1, 2019: HUF 240 (85 US cents) per liter for soft drink syrups with total sugar content exceeding 8 g per 100 ml; HUF 600 (USD 2.15) per kg for fruit jams containing total sugar in excess of 35 percent.
  - Revenue: HUF 40.6 billion in 2018 (close to USD 150 million), or 0.1 percent of GDP.
  - Market impacts: price increases of 10 to 30 percent for taxed products; consumption decreases of 10 to 15 percent.
  - Survey: 26–35 percent of consumers reported consuming less of taxed products than one year earlier; among those, 80 percent attributed decreased consumption to higher prices; one in five reported increased awareness that the foods were unhealthy.
  - Some manufacturer reformulation observed; no analysis of actual public health impact reported to date.
- France (soft drink tax since January 2012):
  - Specific rate in 2015: €7.50 per hectoliter of product.
  - Revenue: in 2013 the tax generated €375 million against an initial objective of €280 million.
  - VAT treatment: reduced VAT rate of 5.5 percent applied to SSB instead of the standard 20 percent, raising coherence questions.

### Cross-Country Patterns, Revenue, and Administrative Observations
- WHO inventory: 75 countries with taxes on SSBs.
- Revenue performance examples (% of GDP):
  - Algeria 0.002; Belgium 0.01; Cambodia 0.16; Chile 0.07; Croatia 0.07; Congo DR 0.03; Finland 0.02; France 0.02; Latvia 0.06; Mali 0.05; Mexico 0.1; Rwanda 0.16; Senegal 0.01; South Africa 0.06; Togo 0.001; Uganda 0.14.
- General observations:
  - In low-income countries, excise revenue from unhealthy foods can represent a larger share of GDP and total revenue.
  - Health objectives favor specific taxes (quantity- or sugar-content-based); revenue favors ad valorem.
  - Tiered specific excises based on sugar content can encourage reformulation (United Kingdom, Portugal, South Africa examples).
  - Measurement and enforcement challenges arise from sugar content variability, especially for imports.

### Policy Recommendations and Considerations
- Design principles:
  - Choose tax type according to primary objective: specific (quantity- or sugar-content-based) for health goals; ad valorem for revenue.
  - Consider tiered sugar-based specific excises to incentivize reformulation.
  - Use manufacturer-level excises when administratively feasible; otherwise consider distribution- or retail-level approaches.
- Fiscal and legal coherence:
  - Address inconsistencies between VAT and excise policies (remove VAT exemptions/reduced rates on unhealthy food items).
  - “Soft earmarking” of revenues (commitment to fund health initiatives through the standard budget process) preferred over “hard earmarking.”
- Complementary measures:
  - Combine excise taxes with non-tax regulatory measures and awareness campaigns for stronger health impacts.
  - Recognize non-tax measures (for example, bans on trans-fats) can sometimes be more efficient.

### Risks, Limitations, and Research Gaps
- Potential for substitution to other unhealthy foods may reduce health impact.
- Administrative costs, legal constraints, and added tax-system complexity may offset benefits.
- Evidence on long-term public health impacts is limited and geographically concentrated; more empirical study is needed.
- No documented study yet linking excise taxes directly to health outcomes at time of review.

### Attribution and Authorship
- Prepared by Patrick Petit, Mario Mansour, and Philippe Wingender. Acknowledged comments from Ruud de Mooij, Michael Keen, Victoria Perry, and Christophe Waerzeggers, and external collaborators from WHO and the World Bank, and the Hungarian authorities.

*Source: htnea2021008 — How-to Note (December 2021).*

### Conclusion �������������������������������������������������������������������������������������������������������������

### Conclusion

### The Rise of Obesity-Related Diseases
- As income per capita increases, many countries experience a gradual shift from communicable to non-communicable diseases driven by four main risk factors: tobacco, alcohol, physical inactivity, and unhealthy diets (Jamison and others 2013).
- Encouraging physical activity has an unambiguous positive health impact and strong support; curbing tobacco consumption has been a core national and international objective (WHO 2019a).
- Alcohol consumption has a somewhat less straightforward health impact but has followed some positive trends in many countries.
- Concerns over poor diet increase with income growth and lead to obesity-related health conditions.
- It is estimated that obesity and overweight are directly responsible for at least 2.8 million deaths annually (WHO 2019b).
- The share of obesity-related diseases in total deaths is rapidly increasing.
- In the United States, obesity represented $147 billion (1 percent of GDP) in 2008.
- In the USA, 30.3 million people had diabetes in 2015, and 84.1 million prediabetes (CDC 2017).
- Obesity-related costs represented 9.1 percent of health care expenditures in the USA (Finkelstein 2009).

### Economic and Policy Challenges
- The economic burden of obesity is significant and has motivated policy interest.
- Fighting the obesity epidemic has proven difficult given the diversity of natural and (text cut off in source).

### Attribution and Authorship
- Prepared by Patrick Petit, Mario Mansour, and Philippe Wingender. The authors acknowledge comments from colleagues including Ruud de Mooij, Michael Keen, Victoria Perry, and Christophe Waerzeggers, and external collaborators from WHO and the World Bank, and the Hungarian authorities. Remaining errors and omissions are the sole responsibility of the authors.

*Source: htnea2021008 - Conclusion — How-to Note (December 2021).*

### 315.8 billion for 2010. Shekar and Popkin (2020) provide estimates

### HOW TO APPLY EXCISE TAXES TO FIGHT OBESITY

### The case for excise taxes
- Modern tax systems rely on broad-based taxes; narrow-based excises are justified for objectives beyond revenue, including: (1) addressing negative externalities, internalities, and imperfect information; (2) revenue-raising; and (3) other objectives such as public health.
- Excises remain important worldwide for reasons including administrative simplicity, equity, and efficiency trade-offs.

### Externalities, internalities, and measurement challenges
- Main economic justifications: externalities, internalities, and imperfect information.
- Conceptual and empirical challenges in identifying the externality component of obesity costs:
  - Externality largely channeled through health insurance pooling of risk; size depends on institutional factors (for example, pure out-of-pocket systems would entail no externality).
  - Bhattacharya and Sood (2005) estimate per capita externality costs of $149 in 1998.
  - Finkelstein (2009) estimates burden of obesity for 2008 at $488 per person, or $147 billion in the United States.
  - Allcott and others (2019a) propose a tax of 0.41 cents per ounce of sugar-sweetened beverages (SSBs) to cover third-party externality costs (roughly 10 to 20 percent of the sale price).
  - Assessing the marginal cost of the externality (the Pigouvian tax level) has never been done.
- Internalities (time inconsistency, low self-control) are conceptually and empirically challenging:
  - Allcott and others (2019a) estimate an internality-based optimal tax of roughly 1.1 cents per ounce for SSBs.
  - O’Donoghue and Rabin (2006) show optimal taxes can be large even with a small number of people with low self-control.
  - Internalities could justify higher corrective taxes than externalities.

### Revenue-raising, administrative feasibility, and equity
- Well-designed excises on clearly identified bases can yield convenient, stable revenue with minimal administrative/compliance costs.
- Challenges for food excises:
  - Complex tax base: wide variety of products, inputs, and production/distribution chains.
  - Sugar tax base complexity: sugar occurs naturally or is added; sources include sugarcane, beetroot, corn (syrup), fruits; final products have widely varying sugar contents, potentially requiring differentiated taxation.
  - Administrative feasibility of a comprehensive excise on high-calorie foods is a key concern, especially in low-capacity environments; may suggest narrower focus on carefully chosen goods.
- Equity considerations:
  - No evidence that high-calorie food is consumed more by higher-income individuals in general; cheap, highly-processed, high-calorie food often consumed more by low-income individuals.
  - The stronger long-term negative impact on low-income individuals and reduction in internalities for those persons may offset regressivity to some extent.
  - Increasing equity is not a prima facie strong rationale for such taxes.

### Price elasticity and expected revenue effects
- Own-price elasticities vary by product and context; preserving numeric estimates exactly:
  - Andreyeva and others (2010) review of 160 US studies: soft drinks average elasticity –0.79 (confidence interval –0.33 to –1.24; individual estimates range –0.13 to –3.18); juice –0.76; sweets/sugars –0.34.
  - Cornelsen and others (2014) review of 78 studies across 38 countries: found comparable results; higher own-price effects in low-income countries.
  - Powell and others (2013): own-price elasticities of SSB and fast-food are –1.21 and –0.52, respectively.
  - Nakhimovsky and others (2016): own-price SSB elasticities ranging from –0.6 to –1.2 in 9 middle-income countries.
  - Allcott and others (2019a): estimate an elasticity of about –1.4 for SSBs.
- Conclusion: the case for stable revenue-raising via taxing obesity-related food is weak.

### Health impacts, substitution effects, and evidence limits
- Medical research (largely US-based) links obesity to small sustained dietary imbalances (a 5- to 10-percent increase in daily caloric intake, e.g., one or two cans of soft drinks).
- Key food items responsible: soft drinks, chips and potatoes, unprocessed red meat, processed meat.
- SSBs have been linked to obesity and may account for up to half of dietary imbalances in the United States.
- Empirical findings on taxes and health:
  - Literature suggests taxes can decrease consumption of taxed goods, but ultimate impact on obesity and health outcomes is uncertain and likely small.
  - No studies yet document a link from an excise tax to a health outcome, given most health-motivated initiatives are recent (Shekar and Popkin 2020).
- Substitution effects are critical and heterogeneous:
  - Edward (2011) finds simulations that include substitution effects show much lower impacts on Body-Mass Index.
  - Cornelsen and others (2014) note cross-price elasticities tend to be significantly lower than own-price elasticities; policy relevance is the combined effect across many products.
  - Harding and Lovenheim (2017): a 20 percent tax on SSBs reduces purchases by 10.35 percent, but related caloric intake by only 4.84 percent (considering substitution).
  - Dubois and others (2019): significant reduction in caloric intake documented but mitigated by increased consumption of non-taxed sugary drinks and food.
  - Dharmasena and Capps (2012): taxes on SSBs impact sales of juices, low-fat milk, coffee, and tea, weakening overall impact.
  - Mexico’s experience shows uncertain substitution patterns and health impacts despite significant drops in sales of taxed products.

### Policy guidance and conditions for effectiveness
- Conditions under which health-motivated excise taxes could have impact:
  - Link obesity to one or a few clearly identifiable food items leading to excessive caloric intake (added sugar, especially in SSBs, is a core problem in many countries).
  - Substitution away from high-calorie food is essential for long-term policy efficiency; better documentation of substitution patterns is needed.
  - Taxes interact with other non-tax policy measures; inclusive approaches involving multiple ministries and nongovernmental organizations increase effectiveness.
  - Awareness campaigns and complementary policies can enhance the impact of taxes (for example, Mexico’s awareness linked to lower consumption).
- Policy design trends:
  - Recent health-motivated tax increases/proposals often involve moderate hikes (for example, one cent per ounce of soft drinks, i.e., 15 to 20 percent of the value) on a few well-selected high-calorie products.

### Mexico case (selected findings)
- Context: obesity prevalence 71 percent and diabetes mellitus prevalence 10.8 percent of adult population; soft drink consumption 163 liters per person per year.
- Policy adopted 31 October 2013:
  - Excise tax of 1 Mexican peso per liter of SSB (equivalent on average to about 10 percent of the sales price); covers a wide range of drinks including energy drinks and non-alcoholic beverages prepared by dissolving sugars in water.
  - Accompanied by a large-scale media campaign on healthy habits.
  - Additional 8 percent tax on 9 categories of non-staple high-calorie processed foods with more than 275 calories per 100 grams.
  - The specific tax includes an embedded regulatory adjustment process related to cumulative inflation.
- Observed impacts compared to baseline consumption:
  - Taxed products (soda and non-soda SSBs) saw decline in sales volume of 5.5 percent in 2014 and 9.7 percent in 2015.
  - Non-soda taxed products: –16.2 percent in 2014 and –29.4 percent in 2015.
  - Soda taxed products: –0.8 percent in 2014 and –0.3 percent in 2015.
  - Sales of non-taxed products increased by 2.1 percent on average over the study period.
  - Untaxed dairy products declined 14.4 percent on average over 2014 and 2015; taxed dairy products fell 2.5 percent on average (shorter data series for untaxed dairy).
  - Effect of the tax was significantly higher among lower socio-economic classes.
  - In 2014, tax revenue from sweet beverages were 0.1 percent of GDP.

*Source: HOW TO APPLY EXCISE TAXES TO FIGHT OBESITY, Fiscal Affairs Department, International Monetary Fund | December 2021*

### 0.08 percent of GDP for the non-staple high-calorie

### htnea2021008 - 0.08 percent of GDP for the non-staple high-calorie processed foods (PAHO 2015)

### Overview
- The note examines the use of excise taxes on unhealthy foods and beverages (notably sugar-sweetened beverages, SSBs) as a policy tool to fight obesity, drawing on recent country experiences and empirical evidence.
- Key trade-offs highlighted include health objectives versus administrative complexity, revenue potential, legal constraints, and substitution effects.

### Country case examples and empirical findings
- Denmark (Box 2: Tax on Saturated Fat)
  - Specific tax: 16 kroner (EUR 2.15) per kilogram of saturated fat on food with fat content above 2.3 percent.
  - Threshold effectively excluded drinking milk but included meat, cheese, butter, edible oils, spreads, snacks, etc.
  - Specific rate roughly equivalent to a 30 percent ad valorem rate for butter, 8 percent for a bag of chips, and 7 percent for a liter of olive oil.
  - Tax duration: repealed after 15 months.
  - Reported results: 10 to 15 percent decline in consumption of targeted food items; observed shifts for some oils and fats from high price supermarkets to low-price discount stores.
  - Repeal reasons noted: high administrative costs for producers, alleged cross-border shopping, disputed effects on consumption habits.

- Hungary (Box 3: Public Health Product Tax, PHPT)
  - PHPT introduced September 2011 on selected manufactured foods with high sugar, salt or caffeine content; excludes basic foodstuffs.
  - As of January 1, 2019, example rates: HUF 240 (85 US cents) per liter for soft drink syrups with total sugar content exceeding 8 g per 100 ml; HUF 600 (USD 2.15) per kg for fruit jams containing total sugar in excess of 35 percent.
  - Tax payable upon first sale in Hungary; exports exempt.
  - Revenue: HUF 40.6 billion in 2018 (close to USD 150 million), or 0.1 percent of GDP.
  - Reported market impacts: price increases of 10 to 30 percent for taxed products; consumption decreases of 10 to 15 percent.
  - Survey findings: 26–35 percent of consumers reported consuming less of taxed products than one year earlier; among those, 80 percent attributed decreased consumption to higher prices; one in five respondents indicated increased awareness that the foods were unhealthy.
  - Manufacturer responses: some reformulation by reducing or removing targeted ingredients.
  - No analysis of actual public health impact reported to date.

- France (Box 4: Soft Drink Tax)
  - Since January 2012, excise tax on all beverages with added sugar or artificial sweetener; taxable items must meet four criteria including customs tariff codes NC 2009 and NC 2202, any amount of added sugar or artificial sweetener, retail packaging, and alcohol level less than 1.2 percent.
  - Specific rate in 2015: €7.50 per hectoliter of product.
  - Tax remitted by suppliers; exports exempt.
  - VAT treatment: reduced VAT rate of 5.5 percent applied to SSB instead of the standard 20 percent, raising questions on tax policy coherence.
  - Revenue: in 2013 the tax generated €375 million against an initial objective of €280 million.
  - Legislative strategy: broader revenue-raising objective added to health objective allowed Constitutional Council to uphold the tax on revenue grounds.

- Mexico (Box 1 referenced)
  - Table entry: rate of 1 peso per liter (±10 percent of retail price); reported revenue share 0.1 percent of GDP.

### Cross-country patterns and administrative considerations
- Dozens of countries tax SSBs; WHO inventory notes 75 countries with taxes on SSBs.
- In many low-income countries SSBs are produced in easily controllable facilities and have been treated as luxury items, facilitating opportunistic and equitable revenue raising through ad valorem taxes.
- Observations:
  - Low-income country context: excise revenue from unhealthy foods can represent a larger share of GDP and total revenue compared to high-income countries.
  - Tax design:
    - Health objectives tend to favor specific taxes (quantity- or sugar-content-based).
    - Revenue objectives often lead to ad valorem taxes.
    - Administrative simplicity may favor ad valorem or volume-based specific taxes when sugar content is hard to measure.
    - Tiered specific excises based on sugar content can incentivize reformulation (examples: United Kingdom, Portugal, South Africa).
  - Measurement and enforcement challenges: sugar content variability, especially for imports; domestically produced consumer products with observable retail prices facilitate controls.
  - Legal constraints can restrict the feasible tax base and level of government that can impose taxes (Box 5 examples).

### Findings on effects, revenue, and behavior
- Evidence on consumption:
  - Denmark: 10 to 15 percent decline in targeted food consumption while tax was in place.
  - Hungary: 10 to 15 percent decrease in consumption of taxed goods; 26–35 percent of consumers reported consuming less a year after introduction.
- Price pass-through and behavior:
  - Hungary: taxed product prices rose by 10 to 30 percent; 80 percent of those consuming less attributed it to higher prices; some consumers reported increased awareness of unhealthy nature of products.
  - Denmark: substitution toward low-price discount stores for some fats and oils.
- Revenue performance:
  - Hungary: HUF 40.6 billion in 2018, or 0.1 percent of GDP.
  - France: €375 million in 2013 against an objective of €280 million.
  - Table 3 selected revenue (% GDP) entries: Algeria 0.002; Belgium 0.01; Cambodia 0.16; Chile 0.07; Croatia 0.07; Congo DR 0.03; Finland 0.02; France 0.02; Latvia 0.06; Mali 0.05; Mexico 0.1; Rwanda 0.16; Senegal 0.01; South Africa 0.06; Togo 0.001; Uganda 0.14.
- Overall revenue from excise taxes on unhealthy food remains low compared to other excise taxes (including alcohol).

### Policy recommendations and considerations
- Policy coherence:
  - Address inconsistencies between VAT and excise policies (e.g., reduced VAT on SSBs can contradict excise health objectives).
  - Removing VAT exemptions, reduced rates, and zero-rates on unhealthy food items is a recommended first step toward better health-coherent taxation.
  - Zero-rating of healthy foods is possible but controversial and increases tax system complexity.
- Design advice:
  - Choose tax type according to primary objective: specific (quantity- or sugar-content-based) for health goals; ad valorem for revenue.
  - Consider tiered sugar-based specific excises to encourage reformulation.
  - Use manufacturer-level excises where administratively feasible; where legal or jurisdictional constraints exist, consider distribution- or retail-level approaches.
- Use of revenues:
  - “Soft earmarking” (commitment to increase taxes and fund health-related initiatives through the standard budget process) is recommended over “hard earmarking” to dedicated external funds, for flexibility, efficiency, transparency, and political durability.
- Complementary measures:
  - Excise taxes are more likely to have larger health impacts when combined with complementary policies and a broader package.
  - Non-tax regulatory measures (for example, bans on trans-fats) can sometimes be more efficient than taxation and should be considered.

### Risks and limitations
- Potential for substitution to other unhealthy foods may reduce health impact.
- Administrative costs, legal constraints, and added tax-system complexity may offset benefits.
- Evidence on long-term public health impacts remains geographically limited and incomplete; more time and empirical study are needed to assess health outcomes fully.

### Conclusion
- A cautious case can be made for excise taxes on unhealthy foods under specific circumstances, particularly where:
  - administrative feasibility is reasonable (for example, SSBs in low-income countries),
  - the marginal value of government spending is high,
  - and policy coherence (including VAT alignment and complementary measures) is pursued.
- Inducing desirable substitution (toward healthier options) and ensuring coherent tax policy are key determinants of whether such taxes will persist and deliver health and fiscal benefits.

*Source: htnea2021008 - 0.08 percent of GDP for the non-staple high-calorie processed foods (PAHO 2015).*

### Appendix 1. Excises on Unhealthy Food: A Brief Overview of the Health Literature

### Appendix 1. Excises on Unhealthy Food: A Brief Overview of the Health Literature

### Cause of obesity and definition of the tax base
- Vartanian and others (2007): Meta-analysis based on 88 studies. Strong statistical relationship between soft-drink consumption, body weight and diabetes, as well as a negative impact on milk consumption.
- Johnson and others (2009): American Heart Association Scientific Statement. Small daily dietary imbalances (±125 calories, or 4 to 5 percent of daily caloric intake) cause long term effects on weight. Imbalances are brought by a few final consumption products with high concentration of specific base ingredient, most notably sugar (high-fructose corn syrup) in soft drinks. “Over the past 30 years, total calorie intake has increased by an average of 150 to 300 calories per day, and approximately 50% of this increase comes from liquid calories (primarily sugar-sweetened beverages)”
- McGranahan and others (2011): SSBs are a significant contributor to weight gain and discretionary caloric intake, they represented on average only 5.6 percent of total caloric intake among all Americans in 2007/08 (albeit with significant differences between population subgroups).
- Mozaffarian and others (2011): Cohort study of 120,877 non-obese (baseline) women and men from the USA. Weight gain was on average 3.35 lbs. per person per 4-year interval, and that weight gain was statistically associated with (in descending order): potato chips (1.69 lbs.), potatoes (1.28 lbs.), SSBs (1.00 lbs.), unprocessed red meat (0.95 lbs.), and processed meat (0.93 lbs.). Vegetables, whole grain, fruits, nuts and yogurt were associated with weight losses, as well as physical activity (21.76 lbs.).
- Woodward-Lopez and others (2011): Review of the literature. One-fifth of the weight increase in the US population between 1977 and 2007 is due to sweetened beverages.
- Chriqui and others (2013): Review of tax practices and policy options. The inclusion/exclusion of various products from the tax base is most likely going to require very granular understanding of market dynamic for a range of products, such as aspartame and saccharin-based drinks, less than 100% fruit juices, sports and energy drinks, lightly sweetened fizzling water, etc.
- Bes-Rastrollo and others (2016): Review of systematic reviews on: (1) SSBs and obesity; (2) metabolic adverse effects of fructose and SSB, and (3) the impact of SSB taxation on weight control. SSBs represent half of the added sugar intake in America (fruit drinks for children and sodas for adolescents and adults) and most studies conclude that they are a risk factor for obesity. The effect of taxes on weight gain is unclear, notably because of product substitution. Taxes on all SSBs could have a role as part of a multi-component strategy.
- Drewnowski and Rehn (2014): Cross-sectional study among 31,035 children, adolescents, and adults based on US surveys (2003–2004, 2005–2006, 2007–2008, and 2009–2010) to identify the source and form of consumed added sugar. Purchases in stores account for 65 to 76 percent of added sugar. Soda, energy and sports drinks account for 34.4 percent of added sugar, and fruit juices for 8 percent.

### Impact of current taxes
- Smed and others (2007): Estimation of a demand system for food in Denmark, using household-level panel data. Price sensitivity of various unhealthy food items changes with age and socio-economic conditions.
- Powell and Chaloupka (2009): Review of the literature on the impact of food and restaurant prices on weight outcomes. Small taxes and subsidies are unlikely to significantly affect Body Mass Index or prevalence of obesity. Higher taxes could have an impact on specific subgroups, such as children, adolescents, individuals with low socio-economic status, as well as for those at risk of obesity and overweight.
- Powell and others (2009): Individual-level panel data regression analysis. SSB taxes have no effect on adolescent Body Mass Index.
- Fletcher and others (2010a): Individual-level regression analysis on the effect of taxes on Body Mass Index. There is a significant but negligible effect of current soft drink taxes on adult weight. The effect is stronger for low-income individuals and Hispanics.
- Fletcher and others (2010b): Individual-level regression analysis on the effect of taxes on Body Mass Index and soft drink consumption. The effect of existing soft drink taxes on consumption is very small for children and adolescent and is fully offset by substitution towards other products (notably milk), hence no impact on obesity. Nevertheless, a 16 percent ad valorem tax on soft drinks could reduce consumption by 100 calories.
- Sturm and others (2010): Individual-level regression analysis on the effect of taxes on Body Mass Index and soft drink consumption. Taxing SSBs at current levels has very little impact, but sub-groups of at-risk children (overweight, low-income, African-American) are more sensitive.
- Tiffin and Arnoult (2011): Estimation of a demand system based on United Kingdom household food survey data, and simulation of the impact of a subsidy to fruits and vegetables combined with a tax on saturated fat. The impact of the tax on health is negligible.
- Jou and Techakehakij (2012): Review of obesity prevalence, tax rates and SSB consumptions in various countries. Various factors that could determine the impact of new taxes on weight, including the obesity rate itself (that is, how acute the problem is), the level of SSB (and other unhealthy food) consumption (that is, the scope for dietary improvement) and baseline tax rates (that is, potential to raise the current tax level).
- Powell and others (2013): Systematic review of the effects of price on consumption and weight outcomes. The studies that link soda taxes to weight outcomes show very little impacts on weight.
- Chriqui and others (2014): Compilation of taxes on unhealthy food in the USA, as of January 2014. Taxes were on average 5.2 percent among the 35 states that applied a sales tax on soft drinks (whether general sales tax or excise).
- Cornelsen and others (2014): Review of literature and meta-analysis on own-and cross-price elasticity in low-, middle-, and high-income countries. Own-price elasticities are larger in low-income countries. Cross-price elasticities vary widely and are highest in high income countries (see also Table 2).
- Finkelstein and others (2014): Review of the literature. Little evidence that targeted food taxes or subsidies alone could have a major effect on individual weight or obesity prevalence.
- Batis and others (2016): Use of consumer-level purchasing data to assess the impact of Mexico’s 8% tax on high-energy non-essential foods and 1 peso-per-liter tax on SSBs (that is, ±10% ad valorem) after one year of implementation. The impact of the tax on consumption was significant, but marginal (25 g per month per capita, or 70 to 110 calories), and substitution patterns and weight impact remain unclear.
- Arantxa-Colchero and others (2017): Use of consumer-level purchasing data to assess the impact of Mexico’s 1 peso-per-liter tax on SSBs (that is, ±10% ad valorem) after two years of implementation. Sales decreased by 8.2 percent for taxed beverages and increased by 2.1% for untaxed beverages. The impact was strongest among low socio-economic groups. Substitution patterns, impact on health and effects of other factors (concurrent information campaign) remain unclear.
- Teng and others (2019): Systematic review and meta-analysis of real-world sugar-sweetened beverage tax evaluations and related impact on purchases and dietary intake, based on 17 study outcomes. A hypothetical 10 percent tax on SSBs is associated with an equivalent average decrease in consumption and dietary intake, but there are considerable variations, which likely depend on the local context. The impact of the tax on non-taxed drink is not statistically significant.

### Simulations for tax policy proposals
- Mytton and others (2007): Use of elasticity, expenditure and consumption data from the United Kingdom’s 2000 National Food Survey and of meta-analysis-based biometric estimates to estimate the effect of extending the 17.5% United Kingdom VAT to various food categories. Applying the tax to three ever wider subsets of unhealthy food, authors concluded that a tax on a narrow subset of principal sources of saturated fat would decrease the intake of such fats, but that an increase in salt intake would offset the health benefits. Wider tax bases would bring the modest but tangible benefits of saving 2100 to 3200 deaths per year.
- CBO (2008): Use of existing price elasticity estimates to calculate the impact of a 0.25 cents tax per ounce on SSBs. Annual revenue would amount to USD 4 to 5 billion.
- Brownell and others (2009): Use of existing price elasticity and consumption estimates to calculate the impact of a 1 cent per ounce tax on SSBs (equivalent to 15 to 20 percent of the price of a 20-ounce bottle, or 12 cents on a standard size soft drink can). Calorie consumption would be reduced by 10 percent and estimated tax revenue increase by USD 14.9 billion.
- Allais and others (2010): Use of scanner data to estimate the demand for food in France and derive price- and “nutrient “elasticities”. A tax-induced price increase of 10 percent for a wide range of unhealthy food would result in a small decrease in caloric intake (for example, less than 20 calories per day per individual for cheese, butter and cream). However, the cumulative effect of small changes in calorie intake could amount to significant long-term gains, and a 10 percent tax could amount to a weight reduction of 2 to 3 kilograms after a decade.
- Smith and others (2010): Estimation of a demand system with individual-level and industry data. The decrease in SSBs induced by a 20 percent tax would be partly compensated by increased intake of juice, milk (among children), and mostly bottled water. The resulting weight loss would be on average 3.8 pounds after a year among adults, and 4.5 pounds among children.
- Andreyeva and others (2011): Use of existing price elasticity and consumption estimates to calculate the impact of a 1 cent per ounce tax on SSBs. The tax would generate USD 13.5 billion and reduce consumption by 24 percent (roughly 50 calories per day), hence a reduction of 5 pounds per year in body mass.
- Edwards (2011): Selective review of the literature to emphasize the importance of considering cross-price elasticities. The simulated impact on Body-Mass Index of a given tax using price elasticity estimates found in the literature is much lower for studies that take substitution pattern into account.
- Dharmasena (2012): Use of scanner data to estimate a demand system for various beverages. There are significant substitution patterns from SSBs towards fruit juices, low-fat milk, coffee and tea. Nevertheless, a 20 percent tax on SSB would reduce bodyweight by 1.54 to 2.55 pounds over a year.
- Jensen and Smed (2013): Use of household level data to assess the impact of the Danish tax on saturated fat on the level and place of purchase of taxed items. Tax-induced price changes in the range of 18–22 percent for butter, butter blends and margarine were associated with a 10 to 20 percent decline in consumption. Consumers appear to have shifted to discount stores.
- Wang and others (2012): Use of various existing estimates to calculate the impact of a 1 cent per ounce tax on SSBs. The tax would decrease health care cost by an average of USD 1.7 billion annually, in addition to raising USD 13 billion in revenue.
- Briggs and others (2013): Estimation of a demand system for sweetened drinks in the United Kingdom using individual data and simulation of the impact of a 20 percent tax. The tax would decrease prevalence of obesity in the United Kingdom by 1.3%. The effect would be only half of that in the USA, and substitution patterns appear to be widely different: whereas diet drinks and SSBs are complements in the USA, they are substitute in the United Kingdom.
- Finkelstein and others (2013): Use of scanner data to estimate a demand system for various food items. Increasing the price of SSBs by 20 percent would have little impact on consumed quantities of other products, except for fruit juice and to a lesser extent fat intake (for example, ice cream). The related average weight loss would be 1.6 pounds in the first year and 2.9 pounds over 10 years.
- Dubois and others (2019): Use of individual-level survey and household data to model “on-the-go” purchases of sugary and non-sugary drinks. A United Kingdom-style 25 p. per liter tax on sugary soft drinks would lower “on-the-go” soft-drink-related sugar intake by 195 g per person per year. The impact is stronger among youth and low-income individuals, but individuals with strong sugar preferences are relatively price insensitive. The reduction in caloric intake is mitigated by an increased consumption of non-taxed sugary drinks (125 g) and by food sources of sugar of a similar amount.

*International Monetary Fund | Appendix 1. Excises on Unhealthy Food: A Brief Overview of the Health Literature*

### 2001. The Modern VAT. Washington, DC: International

### htnea2021008 - 2001. The Modern VAT. Washington, DC: International Monetary Fund.

### Major thematic areas covered in the referenced literature
- Nutrition-related taxation and public health interventions
  - Focus on sugar-sweetened beverage (SSB) taxes, fat/saturated fat taxes, soda taxes, and targeted food taxes (examples: Finkelstein et al. 2013; Jensen and Smed 2013; Mytton et al. 2007; Smed 2012; Tiffin and Arnoult 2011).
  - Reviews and meta-analyses of impacts on purchases, dietary intake, and obesity (examples: Nakhimovsky et al. 2016; Teng et al. 2019; Vartanian et al. 2007).

- Empirical estimates of health and economic burden related to diet and obesity
  - Annual medical spending attributable to obesity (Finkelstein et al. 2009).
  - Studies linking diet, lifestyle, and long-term weight gain (Mozaffarian et al. 2011).
  - Studies estimating effects of taxes on health outcomes, diabetes burden, and cost savings (Wang et al. 2012).

- Behavioral and economic theory relevant to corrective taxation
  - Internalities, addiction, and optimal sin taxes (Gruber and Koszegi 2001; O’Donoghue and Rabin 2006; Gruber 2003; Pigou 1918).
  - Product- vs nutrient-specific taxes and substitution patterns (Harding and Lovenheim 2017; Finkelstein et al. 2013).

- Fiscal policy, competitiveness, and administrative/legal considerations
  - Fiscal fitness and limits of taxes to induce healthier behavior (Petit, Mansour, and Wingender 2014).
  - Competitiveness impacts in agri-food sector and local excise taxation jurisprudence (ECSIP 2014; Shoked 2018).
  - Constitutional and interjurisdictional constraints on state and local taxes (Kearns, Todorova, and Stone 2019).

### Representative study foci and methods (as indicated by titles)
- Systematic reviews and meta-analyses of SSB/food taxes and health outcomes (Nakhimovsky et al. 2016; Teng et al. 2019; Powell et al. 2013).
- Modeling studies projecting health and cost impacts of specific tax levels (Wang et al. 2012; Smith, Biing-Hwan, and Lee 2010).
- Natural experiments and short-run empirical evaluations of national/local taxes (Jensen and Smed 2013; PAHO 2015; NIHD 2013).
- Theoretical and normative analyses of corrective taxation and internalities (Gruber and Koszegi 2001; Pigou 1918; Fleischer 2015).

### Key cited numeric and bibliographic details (preserved exactly as in source)
- Finkelstein, E.A., J.G. Trogdon, J.W. Cohen, and W. Dietz. 2009. “Annual Medical Spending Attributable to Obesity: Payer-and Service-Specific Estimates.” Health Affairs 28(5):822–831.
- Finkelstein, E.A., C. Zhen, M. Bilger, J. Nonnemaker, A.M. Farooqui, and J.E. Todd. 2013. “Implications of a sugar-sweetened beverage (SSB) tax when substitutions to non-beverage items are considered.” Journal of Health Economics 32:219–239.
- Gruber, J., and B. Koszegi. 2001. “Is Addiction Rational? Theory and Evidence.” Quarterly Journal of Economics 116(4):1261–1303.
- Jensen, J.D., and S. Smed. 2013. “The Danish tax on saturated fat—Short run effects on consumption, substitution patterns and consumer prices of fats.” Food Policy 42:18–31.
- Nakhimovsky, S.S., A.B. Feigl, C. Avila, G. O’Sullivan, E. Macgregor-Skinner, and M. Spranca. 2016. “Taxes on Sugar-Sweetened Beverages to Reduce Overweight and Obesity in Middle-Income Countries: A Systematic Review.” PLoS One 11(9) e0163358. doi: 10.1371/journal.pone.0163358.
- Wang, Y.C., P. Coxson, Y-M Shen, L. Goldman, and K. Bibbins-Domingo. 2012. “A penny-per-ounce tax on sugar-sweetened beverages would cut health and cost burden of diabetes.” Health Affairs 31(1): 199–207.
- WHO. 2019a. “WHO report on the global tobacco epidemic 2019.” World Health Organization, Geneva.
- World Bank. 2020. “Taxes on Sugar-Sweetened Beverages: Summary of International Evidence and Experiences.” World Bank, Washington, DC.

### Policy implications and recurring recommendations in the literature (as suggested by titles and publication foci)
- Consider taxes on sugar-sweetened beverages and other targeted food items as part of public health strategy (PAHO 2015; WHO 2019b; World Bank 2020).
- Account for substitution effects and cross-price responses when designing beverage- or nutrient-specific taxes (Finkelstein et al. 2013; Harding and Lovenheim 2017).
- Balance corrective (Pigovian) tax objectives with competitiveness and administrative feasibility (ECSIP 2014; Petit, Mansour, and Wingender 2014).
- Use evidence from systematic reviews, modeling studies, and natural experiments to calibrate tax rates and anticipate distributional and socioeconomic effects (Nakhimovsky et al. 2016; Powell et al. 2013; Smed, Jensen, and Denver 2007).

*References excerpted from the source content.*

---


_Source: https://www.imf.org/-/media/files/publications/howtonotes/2021/english/htnea2021008.pdf_