13.8 What is the 'right' level of tobacco taxation

The most significant issue that governments must face in considering taxation as a tobacco-control policy is in deciding on an appropriate level for excise and customs duty on tobacco products. There are several possible approaches.

13.8.1 Tobacco taxes to recoup social costs

Many economists would argue that the 'correct' level of tobacco taxation is that which ensures that revenue from tobacco taxes is at least equal to the costs likely to be imposed on society resulting from tobacco use (that is, those costs not borne voluntarily by smokers themselves).1 While theoretically this seems like an attractive proposition, in practice governments attempting to set tax levels on this basis would face several difficulties.

Quantification of costs have been established for Australia, as for elsewhere, but the most comprehensive of these are estimates of current costs based on past tobacco consumption rather than estimates of future costs based on consumption by current smokers.i It doesn't seem fair to expect current smokers to bear costs associated with smoking by previous generations of smokers over whose smoking they had no control. If taxes were set to cover the costs associated with past consumption, then remaining smokers would have to contribute the same total amount of revenue, regardless of the total quantity of tobacco smoked, and regardless of how many smokers managed to quit.

Some studies have attempted to project future costs associated with smoking by current smokers; however, because most health effects occur only after many years of use, and because medical treatments change rapidly, it is extremely difficult to quantify such costs with any accuracy. Gene therapies, for instance, could revolutionise the treatment of many cancers and cardiovascular problems over the next 10 to 15 years.

Another challenge would be to ensure that all the relevant costs were being included. For many of the health problems caused by smoking, the extent of mortality and morbidity has yet to be quantified. Also, there are many health conditions that scientists suspect may be caused or worsened by smoking, but insufficient research has been conducted to ascertain causality. Further, current costs are based on average treatment costs. The cost of treating diseases in people who smoke are likely to be higher than average due to problems such as slower healing of fractures and a greater incidence of post-operative complications.2 Authors of economic studies acknowledge that current estimates are likely to significantly underestimate the true costs for all these sort of reasons.3 If tax rates were to be set on the basis of likely future healthcare and other social costs associated with smoking, then a mechanism would need to be established for more comprehensively and regularly updating estimates of the health burdens associated with tobacco use.

13.8.2 Tobacco taxes as a corrective force to combat failures of self-control

While the costs borne by business and costs borne by government due to smoking are considerable, the costs borne non-voluntarily by smokers themselves and by their immediate families are a third, very large component of Collins and Lapsley's estimate of total tangible costs: $9871.2 million or close to 50% of the total 19 609.3 million estimated tangible costs attributable to smoking for Australia in 2004–05 (see Chapter 17, Section 17.3). In addition to almost $10 billion in tangible costs, smokers and their families bear 100% of the intangible costs attributable to tobacco use: $19 459.7 million in 2004–05. Collins and Lapsley's estimate of intangible costs included an estimate of the value to individuals of a year of life lost in 2004–05, but they did not attempt to quantify the pain and suffering of those who grieve for smokers who die early. And as well as bearing these economic costs, addicted smokers spent more than $10 billion in 2004–05 on tobacco products, which included more than $6 billion in excise and customs duty and goods and services tax (GST) on sale of tobacco products.ii

While the financial and health consequences of smoking might be considered disincentive enough, they largely occur in the future and are more easily dismissed or discounted in a cognitive fashion by smokers. By contrast, the difficulty of quitting (in particular of withstanding cravings) is something that must be faced in the present with little prospect of immediate benefit. Increases in the cost of tobacco products helps to tip the balance, so that the continuing costs of smoking are greater and the benefits of quitting more immediate. As described by Gruber and Koszegi, governments can provide 'a self-control device that will allow the consumer to avoid making sub-optimal consumption decisions' (p13).4 ... or making the healthy choice the easier choice.

As part of its report to the government on recommended reforms to Australia's tax system, a review panel, chaired by the then Treasury secretary Ken Henry, undertook an analysis of Australian tobacco tax rates5 based on the Gruber and Koszegi model.4 While acknowledging the difficulties of quantifying time preferences and warning that considerable uncertainties surrounding their calculations, the panel concluded that tobacco excise rates could be increased substantially. As stated in its report:

'The ideal tax rate depends on a number of parameters. It depends on how strongly smokers privilege current costs and benefits over future costs and benefits. The stronger the current period preference, the harder smokers find it to carry out plans to stop smoking and the higher the tax needs to be to help them overcome their self-control problems. The more harm is done by current smoking, the higher the tax needs to be to limit the damage suffered by smokers with self-control problems. The ideal tax rate also depends on whether the smoker understands their self-control problem – that is, whether they are 'naïve' or 'sophisticated'. A naïve smoker does not realise they have a self-control problem.

...

In [Table E6-1] , the value representing the strength of smokers' current period preference can range from 0 to 1. A value of 0 would mean that current period preference is absolute, so that a smoker places no weight at all on what happens in future. A value of 1 would mean that there is no current period preference, so that a smoker recognises the full costs of smoking, and can make a rational decision about their current smoking on that basis. The values tested in the table cover the mid-range of empirical estimates.'

Table 13.8.1
Range of optimal tobacco tax rates for Australia ($ per stick)

 

Strength of current period preference

 

0.60

0.65

0.70

0.75

0.80

Optimal tax rate for 'sophisticated' smokers

$0.31

$0.27

$0.22

$0.18

$0.14

Optimal tax rate for 'naïve' smokers

$0.47

$0.41

$0.35

$0.29

$0.23

Source: Table E6–1 from K Henry et al 20105

This model suggests that excise duty on tobacco might need to increase to almost 50 cents per stick before smokers who highly discounted the future and who were unaware that they had problems controlling their smoking would seriously think about quitting. At 50 cents per stick, the recommended retail price of a pack of Winfield 25s would increase by about 26%, from $17.15 (February 2012) to $21.60 per pack.

While this is an attractive model, the review panel points out a number of limitations:

  • 'It is very difficult to measure the strength of an individual's current period preference. iii
  • Different groups may vary in the strength of their current period preference. If so, a uniform tax will overtax some and under-tax others. In particular, young people underestimate their chance of becoming addicted and their smoking responds more to price changes, so estimates for sophisticated smokers may underestimate the appropriate overall rate.
  • It is difficult to estimate the value of life-years lost through smoking and of the other harms that smoking causes.
  • Not all smokers are the same but they all pay the same tax. High tobacco taxes impose costs on the small proportion of occasional smokers who are at little risk of harm.' 5

13.8.3 Pricing products out of the reach of young people

Another approach to establishing the optimal level of tobacco taxes is to attempt to price tobacco products out of the reach of young people.

The recommended retail price of a typical pack of 25 cigarettes in Australia in February 2012 was $17.15. This was less than going to a movie (adult ticket $18.50, Hoyts April 2012), less than the cost of purchasing a new release DVD ($19.95 for a single disc, Video Ezy April 2012) and less than half the amount the average teenager received in weekly spending money in 2011 ($45.50 per week for young people aged 15 years (according to unpublished data from the Australian Secondary Student Survey on Alcohol and other Drugs provided to the author by V White December 2012). Depending on what other goods and services young people would not be prepared to forgo, students on average pocket money would currently be able to afford to purchase up to two packs of cigarettes per week. In 2011, secondary school students who smoked at least weekly smoked an average of 17 cigarettes per week.

A third approach to taxing tobacco products would be to set tax levels on cigarettes so that a week's supply would cost more than the average week's disposable spending money. This might require the price of cigarettes in Australia to more than double. Theoretically governments could go further and tax products so as to price cigarettes out of the reach of all but the most 'cashed-up' young person, at higher than the pocket money received by, say, 90% of young people aged 15 years. Because some young people receive very high levels of money they are able to 'spend on themselves' (exceeding $100 a week), setting taxes that result in these sorts of prices would impose considerable hardship on addicted smokers unable to dramatically reduce their consumption. Such a dramatic doubling or tripling in price may also result in the development of a black market to supply remaining users.

While it may not be feasible to price young people out of the tobacco market altogether, making cigarettes less affordable to young people is likely to remain an important consideration in setting tax levels.

13.8.4 Taxing to take tobacco products to smokers' 'price point' for quitting

A less drastic and more pragmatic approach to establishing tax levels on tobacco products might be to base them on what smokers tell us about how expensive cigarettes would need to be before they would think about quitting. The average increase recorded in responses from smokers in Victoria who were asked this question in November 2010 was $6.20, an increase over prices current at that time of almost 40% (Hayes, L. Unpublished data derived from the 2009 and 2010 Victorian Smoking and Health Surveys, Centre for Behavioural Research in Cancer, 2011).

This equates to about 85 cents per stick, or $21.25 for a pack of 25 cigarettes. A pack of 25 cigarettes would cost $21 if excise duty was about 47 cents per stick. So this result is at the high end of estimates by the Australia's Future Tax System panel,5 based on the Gruber and Koszegi model.5

13.8.5 Frequent price rises to discourage tobacco use

Even if an ideal tax on tobacco products could be established, the question remains as to how quickly it should move to such a level. Very little information is available about the comparative effects on consumption of one very large increase compared with several smaller increases.

It is known that quit attempts increase significantly following tax increases,6–9 so it makes sense to ensure that increases are frequent and large enough to help to maintain the frequency of serious quitting attempts by remaining smokers. Complementing such tax increases with other elements of a comprehensive strategy such as media campaigns and access to services and products for smokers to assist cessation would maximise the potential benefits.

Relevant news and research

For recent news items and research on this topic, click here (Last updated January 2024)

 

i Current costs would be higher than future costs to the extent that numbers of smokers were declining. Current costs would be lower than future costs to the extent that the costs of medical, hospital and particularly pharmaceutical treatments were increasing.

ii This spending is not counted as a 'cost' to individuals, because, had they not been smokers, they would have spent this money on other goods and services. The cost (net of government taxes) is counted as a cost to business outside the tobacco industry.

iii Henry et al explain that the values 0.60 to 0.80 are notional, not based on actual data from Australian smokers.

References

1. Warner KE, Chaloupka FJ, Cook PJ, Manning WG, Newhouse JP, Novotny TE, et al. Criteria for determining an optimal cigarette tax: the economist's perspective. Tobacco Control 1995;4:380-6. Available from: http://tobaccocontrol.bmj.com/cgi/reprint/4/4/380

2. US Department of Health and Human Services. Women and smoking. A report of the Surgeon General. US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. Atlanta, Georgia, 2001. Available from: http://www.cdc.gov/tobacco/data_statistics/sgr/index.htm

3. Collins D and Lapsley H. The costs of tobacco, alcohol and illicit drug abuse to Australian society in 2004/05. P3 2625. Canberra: Department of Health and Ageing, 2008. Available from: http://www.nationaldrugstrategy.gov.au/internet/drugstrategy/publishing.nsf/Content/mono64/$File/mono64.pdf

4. Gruber J and Koszegi B. A modern economic view of tobacco taxation. Paris: International Union Against Tuberculosis and Lung Disease, 2008. Available from: http://www.tobaccofreeunion.org/content/en/217/

5. Review panel Australia's Future Tax System. Final Report . Part 2: Detailed analysis; Chapter E: Enhancing social and market outcomes; E6: Tobacco taxation; E6–3: Excise rates should be higher. Canberra: Treasury, 2010. Available from: http://www.taxreview.treasury.gov.au/content/FinalReport.aspx?doc=html/publications/Papers/Final_Report_Part_2/chapter_e6-3.htm

6. Kengganpanich M, Termsirikulchai L and Benjakul S. The impact of cigarette tax increase on smoking behavior of daily smokers. Journal of the Medical Association of Thailand 2009;92(suppl. 7):S46–63. Available from: http://www.mat.or.th/journal/index.php?command=vol&selvol=92

7. Gallet C and List J. Cigarette demand: a meta-analysis of elasticities. Health Economics 2003;12(10):821–35. Available from: http://www3.interscience.wiley.com/journal/101520325/abstract?CRETRY=1&SRETRY=0

8. Dunlop SM, Perez D and Cotter T. Australian smokers' and recent quitters' responses to the increasing price of cigarettes in the context of a tobacco tax increase. Addiction 2011;106(9):1687-95. Available from: http://dx.doi.org/10.1111/j.1360-0443.2011.03492.x

9. Chaloupka FJ, Straif K and Leon ME. Effectiveness of tax and price policies in tobacco control. Tobacco Control 2011;20(3):235-8. Available from: http://tobaccocontrol.bmj.com/content/20/3/235.abstract