2.7 Per capita consumption in Australia compared with other countries

Show / hide chapter menu

Per capita tobacco (or cigarette) consumption—the amount of tobacco (or number of cigarettes) per person per year sold for domestic consumption—is a useful index with which to try to compare different levels of tobacco use in different countries. However, as outlined in the section above, these indexes should be interpreted with great care. Figures calculated by different researchers may be based on varying ways of estimating either or both the tobacco numerator and the population denominator.

Estimates of the numbers of cigarettes sold are sometimes based on trade data, sometimes on customs and excise data, sometimes on industry sales figures and sometimes on a mixture of all three depending on availability. Often sources are incompletely or inconsistently reported or not reported at all.

Even for studies that exclusively compare the same type of consumption data, inconsistencies may still remain.

Different countries levy excise and customs duty on tobacco in different ways. Some levy it per cigarette stick sold. Others levy it on the basis of the weight of tobacco sold. Others still levy it on the value of tobacco sales. Among countries where tobacco is levied by weight, some include the weight of the filter and some don't. In some countries the weight excised is net of moisture content; in some countries the moisture is included in the excisable weight. To enable comparison with countries that tax tobacco on the basis of weight, figures from countries that levy tax per cigarettes are converted to weight, mostly assuming that one cigarette is equal to one gram—see, for instance, the OECD's compilations of tobacco consumption in various nations.78 However one gram is probably an overestimate of the weight of cigarettes in most countries, and actual average weight must have varied considerably between countries and over time.

It is also necessary to be careful where numbers of cigarettes are used in the numerator. Some estimates include only cigarettes and some convert cigars, pipe, roll-your-own and other tobacco to 'cigarette equivalents'. If only cigarettes are included, cigarettes per capita will underestimate tobacco consumption depending on the extent of use of these other products.

Record-keeping systems in national tax offices are designed not to provide valid estimates of tobacco consumption, but, rather, merely, to accurately record revenue. In some countries, for instance, it is not easy to distinguish imports of manufactured smoking tobacco from imports of manufactured tobacco sheet or even raw tobacco that are destined not for retail sale but for factories that produce tobacco products that will be subject to excise duty. Inclusion of such items in estimates of imported tobacco products results in double counting and an overestimate of the amount of tobacco consumed.

And, of course, official tax receipts also underestimate consumption in that smuggled cigarettes are not included; consumption of illicit products is a much more significant proportion of total consumption in some countries than in others.

Population figures used to calculate such indexes of tobacco consumption are also not straightforward. Sometimes these are based on the total population. Other times the figure includes only adults, sometimes adults 18 and over, sometimes adults 16 and over, and sometimes people 15 and over. In countries with very low female smoking prevalence, overall consumption figures will be lower than in countries with higher female smoking prevalence, even though cigarette consumption among males may be very high. In some analyses, researchers have calculated per capita consumption for some products just for males. Some researchers have included and some have excluded armed services personnel in population estimates.79

Finally, of course, consumption in various countries in one particular year says nothing about the relative reductions in consumption in such countries over time.

Table 2.24 compares per capita cigarette consumption among the 29 countries defined by the World Bank as 'high income'80 for which tobacco production or trade data was available in (or close to) the financial year 2004. The countries include both high-income countries in the Organization of Economic Cooperation and Development (OECD) and a number of high-income countries that are not members of the OECD.[28] The figures are based mainly on production and trade data, and were collected for an atlas published by the American Cancer Society in conjunction with the International Union Against Cancer (UICC) that compared various indicators of tobacco use and tobacco control across all countries.81 These figures update those compiled a few years earlier for a compendium of country-by-country profiles published by the American Cancer Society in conjunction with the World Health Organization and the UICC82, and researchers have gone to considerable effort to ensure consistency in the calculation of estimates for each country.

As can be seen from Table 2.24, among the 29 high-income (OECD and non-OECD) countries for which tobacco production and trade data was available for 2004, Australia had the eighth lowest tobacco consumption. It ranked sixth lowest of the 22 high-income countries with more than minimal female smoking rates. Among high-income English speaking countries, Australia ranked third lowest of five, lower than Canada and the United States, but higher than New Zealand and the UK.

Among the full 112 countries for which data is available however, Australia appears to do less well, ranking only 70th (from lowest to highest). Among countries where female smoking rates are at least 50% of male rates, Australia ranked 17th lowest out of the 38 countries for which data is available. It ranked 10th lowest out of G20 nations (excluding the European Union).

The OECD has also compiled estimates of tobacco consumption in member countries based on weight of tobacco products on which duties are levied.78 This provides a similar ranking of countries, but very different figures—refer Table 2.6 of the Australian Institute of Health and Welfare's Statistics on Drug Use in Australia 2006 which tabulates OECD per capita consumption estimates for 15 countries over the past 30 years.50