Skip to main content
NCBI home page
The BMJ logoLink to The BMJ
. 2000 May 27;320(7247):1469.

VAT and fat

Will sales tax influence consumption?

Alan O'Rourke 1
PMCID: PMC1127653  PMID: 10877558

Editor—Marshall has produced an original but flawed article, opening debate on public and preventive health by arguing that serum cholesterol concentrations are closely related to diet and to ischaemic heart disease; that dairy produce, pastries, and puddings account for more than 40% of the saturated fat in the British diet; and that introducing value added tax (VAT) at 17.5% on these foodstuffs, but leaving foods such as semi-skimmed milk and low fat margarine VAT free, could, because of “price elasticity” in demand, save 900 lives a year.1

This raises the question, why do we have sales taxes? Assuming that your country isn't engaged in protectionism, there are two reasons: to raise revenue and to influence demand towards more desirable commodities—for example, lower taxes on unleaded petrol. Taxes on tobacco and alcohol have been largely in the first category (the puritanical element in British society no doubt delighted in the added expense of such sinful pleasures). But they are moving, especially taxes on tobacco, towards the second category. VAT is not paid on most foodstuffs, although it is paid on luxury items. But Marshall proposes making foods that damage health liable to VAT as well.

Three issues arise:

  • Equity: although Marshall acknowledges that such taxes are regressive—that is, they fall most heavily on poor people—he shoots himself in the foot by proposing to compensate poorer people by diverting the extra VAT into higher benefits. Such “hypothecated taxation” is not popular, and with their spending power restored poor people will buy exactly the same foodstuffs. Also, although this system compensates those people on welfare, it does little for those people who are employed on low wages.

  • Practicality: the wealthier, and healthier, have already ceased smoking, changed their diet, and joined gyms. These people with high living standards have clear interests in maintaining health to enjoy their status, and they also feel that their own actions influence their lives. In comparison, the poor often feel disempowered, and they may be so stressed about making ends meet that their future health is not on the list of things to worry about. For various reasons, those foods rich in saturated fat offer compensations in a bleak world. Such financial manipulations, especially if “compensated,” seem unlikely to alter consumption. Despite rising tobacco taxes, smoking is more prevalent among those who are less affluent.

  • Effect on mortality and morbidity: ischaemic heart disease is multifactorial in its aetiology, and it seems dangerous to concentrate on single factors.

Acknowledgments

Competing interests: None declared.

References

  • 1.Marshall T. Exploring a fiscal food policy: the case of diet and ischaemic heart disease. BMJ. 2000;320:301–304. doi: 10.1136/bmj.320.7230.301. . (29 January.) [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2000 May 27;320(7247):1469.

Taxing single nutrient is dangerous

John C Stanley 1

Editor—The relationship between diet and blood cholesterol concentration is much more complex than Marshall indicates in his recent article.1-1 This complexity has major implications for a fiscal food policy aimed at extending VAT to the major sources of saturated fat in the diet.

Stearic acid is the second most abundant saturated fatty acid and the fourth most abundant fatty acid in the British diet. It has been repeatedly shown that stearic acid is neutral with respect to its effects on the blood cholesterol concentration. This was recognised by Keys et al who, in the paper cited by Marshall, modified the term for saturated fat in his equation to exclude stearic acid.1-2 Marshall used the original equation, not the modified one. Clarke et al (cited by Marshall) recently concluded that stearic acid “did not seem to be significantly related to blood cholesterol concentration.”1-3

Other factors complicate the relationship between dietary fat and blood cholesterol concentration. Saturated fatty acids with fewer than 12 carbon atoms also have no effect on the blood cholesterol level. Other classes of fatty acid, notably trans unsaturated fatty acids, raise the blood cholesterol level. What Marshall's approach needs, therefore, is a measure of the fatty acids in a food that raise cholesterol, which is not synonymous with the food's saturated fatty acid content. His approach probably also needs to take into consideration the nature of the carbohydrate and protein in a food, which can also influence the blood cholesterol concentration. These considerations are far from being purely academic. This is well illustrated by the example of whole milk. Marshall identified whole milk as one of the main sources of saturated fat in the diet. The difficulty with his argument is that up to half the saturated fatty acids in milk are neutral with respect to the blood cholesterol concentration. This helps to explain why whole milk lowers the blood cholesterol concentration by 5%.1-4

Foods are complex mixtures of nutrients. Whole milk contains several lipid soluble anticancer agents that are not present in skimmed milk.1-5 In seeking to tax a food to reduce the risk of one disease Marshall may increase the risk of developing another. This illustrates the dangers of taxing a food on the basis of its content of a single nutrient.

Acknowledgments

Competing interests: Dr Stanley has received fees for work carried out for the National Dairy Council.

References

  • 1-1.Marshall T. Exploring a fiscal food policy: the case of diet and ischaemic heart disease. BMJ. 2000;320:301–304. doi: 10.1136/bmj.320.7230.301. . (29 January.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 1-2.Keys A, Anderson JT, Grande F. Serum cholesterol responses to changes in the diet. IV. Particular saturated fats in the diet. Metabolism. 1965;14:776–786. doi: 10.1016/0026-0495(65)90004-1. [DOI] [PubMed] [Google Scholar]
  • 1-3.Clarke R, Frost C, Collins R, Appleby P, Peto R. Dietary lipids and blood cholesterol: quantitative meta-analysis of metabolic ward studies. BMJ. 1997;314:112–117. doi: 10.1136/bmj.314.7074.112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 1-4.Howard NA, Marks J. Hypocholesterolaemic effect of milk. Lancet. 1977;ii:255–256. doi: 10.1016/s0140-6736(77)92880-x. [DOI] [PubMed] [Google Scholar]
  • 1-5.Parodi PW. Conjugated linoleic acid and other anticarcinogenic agents of bovine milk fat. J Dairy Sci. 1999;82:1339–1349. doi: 10.3168/jds.S0022-0302(99)75358-0. [DOI] [PubMed] [Google Scholar]
BMJ. 2000 May 27;320(7247):1469.

Evidence is contradictory

Uffe Ravnskov 1

Editor—Marshall's proposal to treat ischaemic heart disease by fiscal measures is based on the assumption that diet determines cholesterol concentrations, which again determine the prevalence of ischaemic heart disease.2-1 These assumptions have been iterated for decades in spite of many contradictory observations and experiments. Let it suffice to mention some of the most salient facts concerning the alleged link between dietary fats and ischaemic heart disease.2-2,2-3

Secular trends of national fat consumption and mortality caused by ischaemic heart disease contradict each other almost as often as they coincide.2-2 Among 21 cohort studies of ischaemic heart disease including 28 cohorts and more than 150 000 individuals, people who developed ischaemic heart disease ate considerably more saturated fats than those without ischaemic heart disease in three cohorts, and considerably less in one cohort; in 22 cohorts no difference was found. In three cohorts those people who developed ischaemic heart disease had eaten more polyunsaturated fats; in 24 cohorts no difference was recorded. (In a few studies the consumption of saturated or polyunsaturated fats was not recorded.)2-2

Three of four cohort studies analysed the correlation between the consumption of saturated and polyunsaturated fats and the degree of atherosclerosis at necropsy; none of them found any. All four analysed the total fat intake: in one study it was correlated with degree of atherosclerosis; in another study no correlation was found; and in two studies the correlation was inverse.2-2

In six case-control studies of patients with ischaemic heart disease and of controls matched for sex and age, no important differences were noted between their intakes of saturated or polyunsaturated fats.2-2

In a meta-analysis of nine controlled, randomised, unifactorial, dietary trials with reductions of saturated fats and additions of polyunsaturated fats that were more radical than any of the official recommendations, neither total mortality (odds ratio 0.99) nor ischaemic heart disease mortality (odds ratio 0.94, confidence interval 0.80 to 1.10) was lowered; the total number of deaths in the treatment and control groups was identical.2-2

In the only unifactorial, dietary trial that lowered fatal and non-fatal ischaemic heart disease considerably which was included in the mentioned meta-analysis, the cholesterol concentrations in the diet group and the control group were almost identical.2-4

The edible oil industry may applaud the idea of fiscal medicine, but those of us who prefer evidence based medicine will not.

Acknowledgments

Dr Ravnskov was paid by Dairy Farmers of Canada in 1999 for two identical lectures summarising reference 2.

References

  • 2-1.Marshall T. Exploring a fiscal food policy: the case of diet and ischaemic heart disease. BMJ. 2000;320:301–304. doi: 10.1136/bmj.320.7230.301. . (29 January.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2-2.Ravnskov U. The questionable role of saturated and polyunsaturated fatty acids in cardiovascular disease. J Clin Epidemiol. 1998;51:443–460. doi: 10.1016/s0895-4356(98)00018-3. [DOI] [PubMed] [Google Scholar]
  • 2-3.Ravnskov U. The cholesterol myths. http://home2. swipnet.se/∼w-25775/ (accessed 26 Apr 2000).
  • 2-4.De Lorgeril M, Salen P, Martin JL, Monjaud I, Delaye J, Mamelle N. Mediterranean diet, traditional risk factors, and the rate of cardiovascular complications after myocardial infarction. Final report of the Lyon Diet Heart Study. Circulation. 1999;99:779–785. doi: 10.1161/01.cir.99.6.779. [DOI] [PubMed] [Google Scholar]
BMJ. 2000 May 27;320(7247):1469.

Author's reply

Tom Marshall 1

Editor—My article on the use of taxation to influence diet has stimulated considerable discussion.3-1 O'Rourke says that taxation may be intended to raise revenue or to influence demand. Irrespective of intention, all taxation will do both. The magnitude of the price elasticity of demand determines which effect predominates. He also implies that the poor can imagine no better use for additional income than to buy taxed foodstuffs. Economics (household production theory) and common sense indicate otherwise. Housing, clothing, transport, entertainment, and other goods will also make claims on extra money. He suggests that focusing on a single risk factor may be dangerous—an argument that might equally be advanced against quitting smoking.

I agree that a fiscal food policy should be based on a detailed analysis of the net effects of foodstuffs on cholesterol levels.

Ravnskov's views on the irrelevance of cholesterol concentrations to heart disease are at variance with much evidence, including the demonstrated effectiveness of statins. Nevertheless, to avoid the accusation that either of us could be citing evidence selectively, there is a place for a further systematic review to resolve this issue.

Policies to influence food prices are not new. To reduce undernutrition, many developing countries subsidise staple foodstuffs. In my own view there are three arguments against a fiscal food policy: it would not result in dietary change, it would be inequitable, and it would erode our personal freedom. The first argument can be resolved only by the further research that I advocate. Regarding the second: data are available on average weekly expenditures on different foodstuffs.3-2 From these it is possible to estimate the equity effects of a fiscal food policy. On average, individuals could expect to spend an additional £0.30 a week in tax. The additional £15.2m a week raised by this would be sufficient to increase benefits to 10.4 million people on low incomes by £1.30 a week. They would therefore be better off. All families with children under two years old (for whom full-cream milk is recommended) could also be provided with full-cream milk vouchers worth £1 a week. This would affect 1.7 million people, and would represent an extra expenditure to the government of £1.7m. Would such a policy erode our personal freedom? Yes, by £0.30 a week. Would we each be prepared to forgo £0.30 a week so that 900 fewer of us die prematurely each year? In the end this is a political choice. But it is only by further investigation that we will be clear what alternatives there really are to choose between.

Acknowledgments

Competing interests: None declared.

References

  • 3-1.Marshall T. Exploring a fiscal food policy: the case of diet and ischaemic heart disease. BMJ. 2000;320:301–304. doi: 10.1136/bmj.320.7230.301. . (29 January.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3-2.Ministry of Agriculture, Fisheries, and Food. National food survey 1998. London: Stationery Office; 1999. [Google Scholar]

Articles from BMJ : British Medical Journal are provided here courtesy of BMJ Publishing Group

RESOURCES