In recent years there have been a number of high profile calls for government intervention in the form of taxes or bans to discourage unhealthy diets. These range from Mayor Bloomberg’s proposed soda ban in New York City to Hungary’s Hamburger tax. The Australian Medical Association has even weighed in on the debate and voiced support for similar measures in Australia.
This issue can raise various emotive and ideological arguments pertaining to government and individual rights. However, this article will analyse the proposition of a ‘fat tax’ from an economic perspective.
The logic behind a fat tax is simple. Unhealthy diets lead to obesity. The health costs associated with obesity are an economic cost to society. This is a negative externality. Therefore we should introduce a pigovian tax to correct the over-consumption of unhealthy food.
Some of the assumptions made in this logic are questionable.
Firstly, it assumes that unhealthy foods are easy to define. The truth is that they are not. Even the national media spokesperson for Nutrition Australia, Aloysa Hourigan, has conceded that, “the difficulty is defining what junk food is.”
Do you define unhealthy foods as being high in sugar? High in fat? High in saturated fat? High in cholesterol? Low in vitamins? A combination of some or all of these factors? There is no one definitive answer. Even if we could agree on what constitutes unhealthy foods, narrowing down each of these factors to the level of specificity necessary for tax legislation would be very complicated. Moreover any definition is likely to have unintended consequences. Some foods that are generally considered to be ‘healthy’ would also be subject to the tax. For example, if high fat content was a definitional factor then some fish that are high in healthy fatty acids might also be captured. The significant definitional challenges presented by a tax on unhealthy food can be contrasted to other items that are taxed for negative externalities which can be more easily defined, such as cigarettes and alcohol.
Secondly, the logic assumes that the government can successfully estimate the social cost of obesity and tax up to that amount to correct the market failure. Estimating the social cost of obesity is quite difficult and this could lead to inefficient over-taxation leaving society worse off after the tax than before it.
A third problem with the logic behind a fat tax is that even if we could successfully define an unhealthy diet and impose a tax on it there would be a high administrative cost to producers and outlets. This was one of the reasons cited by the Danish Tax Ministry for their decision to scrap the world’s first fat tax. The Danish tax used saturated fat content to define unhealthy food (processed foods containing more than 2.3% saturated fat were subjected to the tax). This meant producers and outlets had to tinker with their products and install new systems to calculate the tax. This costs time and money. Eventually the cost was passed on to the consumer. The only winners here were accountants and tax law experts.
While the intention of the Danish government was for foods with a high saturated fat content to become more expensive, the reality was across the board price rises in supermarkets. This is because supermarkets responded by spreading the cost increase for fatty foods out across all their products. Therefore the desired effect of less fatty food consumption was not attained, as prices of foods low in saturated fat increased by the same amount as those high in saturated fat. Furthermore, ordinary people had to spend more at the supermarket to purchase food. In this light the tax can be viewed as regressive as poorer people spend a higher percentage of their income on food and will therefore be the most negatively impacted.
Other outlets that were less diversified than supermarkets lost a large amount of customers. Specialised food retailers such as bakeries and butchers were unable to spread the price increase out as they specialise in products high in saturated fat. This meant the price gap between a cupcake at the bakery and a cupcake at the supermarket became larger than ever before, as supermarkets could keep the price down. Large supermarkets became winners from the tax and specialist food retailers such as bakers and butchers, the losers.
The fourth problem is that unhealthy food is highly substitutable. For example, if there was a tax on foods containing high amounts of fat, consumers might substitute sugary snacks instead and vice versa. The demand for a certain type of unhealthy food is highly elastic because consumers can easily respond to a price increase by substituting a different unhealthy food. The supply of these goods, in the short term, is inelastic as a donut manufacturer, for example, cannot quickly or easily change to a different product and the workers in the donut factory would not eagerly change careers in response to a change in market conditions. Thus the tax burden falls on producers. This would lead to a negative outcome for producers and their workers whilst raising little revenue as consumers substituted another unhealthy food free from tax liability. We have seen this before – see the early 1990s U.S luxury tax. Meanwhile the whole exercise will have failed to achieve its ultimate aim: healthier eating choices leading to lower obesity.
All of the above goes to show, as co-author of the best-selling ‘Freakonomics’ Steven Levitt put it:
“Politicians have all sorts of reasons to pass all sorts of laws that, as well-meaning as they may be, fail to account for the way real people respond to real-world incentives.”
The failed Danish fat-tax clearly demonstrates the truth of this statement.
Instead of adopting a punitive approach to incentivise lower consumption of unhealthy foods, governments should attempt to positively incentivise higher consumption of healthier food such as fruits and vegetables. This would undoubtedly be less regressive and punitive to poorer people than a fat tax and could escape many of the problems outlined in this article.
The views expressed within this article are those of the author and do not represent the views of the ESSA Committee or the Society's sponsors. Use of any content from this article should clearly attribute the work to the author and not to ESSA or its sponsors.