Price increases have long been seen as an effective strategy for reducing cigarette consumption.1^–3 Previous research has demonstrated that price is an especially effective instrument in preventing youth smoking.4^–7 Tobacco tax is widely used as the major tool to increase the price of tobacco products. Several researchers have focused on the affordability of tobacco products. This article reviews the previous literature on cigarette affordability and compares cigarette affordability in 60 cities using a modified calculation method.

AFFORDABILITY OF CIGARETTES

Lal and Scollo used the Big Mac index, published annually by the Economist as a purchasing power parity (PPP), to calculate the price of cigarettes in 30 countries.8 The advantage of this method is the ready availability of data. However, the use of the price of a single commodity as the PPP commonly produces misleading results. Table 1 presents the prices of selected commodities from five cities. If the price of fuel instead of a Big Mac is used as the PPP, purchasing power in Singapore and Hong Kong is significantly reduced. Moreover, since in Lal and Scollo’s paper there is no consideration given to the level of income, their contribution is limited to the comparison of cigarette prices in different countries but not affordability. More expensive does not mean less affordable.

Guindon et al noticed the distortion effect of the Big Mac index. In addition to using the Big Mac index, they also used the price of a 1 kg loaf of bread and rice as references.9 This approach minimised the flaws inherent in using the Big Mac index. Their most significant contribution was to use income (defined in minutes of labour) to calculate the affordability of cigarettes. The average wage of 12 occupations published by the Union Bank of Switzerland (UBS) was used to calculate cigarette affordability (minutes of labour required to purchase one pack of cigarettes) in 56 cities. Given that these 12 occupations in the UBS’s survey cover different ranks, the application of the average wage does not address the fact that income disparity was ignored.

Blecher et al recognised that the UBS’s survey on income was “a more direct way of estimating people’s income”; however, they also highlighted the limitations of the UBS survey being only conducted every three years and being based on fewer samples than the survey undertaken by the Economist. Instead, they used commonly available data: gross domestic product (GDP) per capita. Affordability was therefore defined as “the percentage of per capita GDP required to buy 100 packs of cigarettes.”10 Clearly, Blecher et al found an easier way to calculate cigarette affordability. Despite being aware of the issue of income distribution, they considered that GDP per capita was a good indicator of living standard and, therefore, no remedy was proposed.10

Previous research has demonstrated that different income groups respond differently to a price measure. The price measure for cigarettes is more effective in low income groups than those with high income.11^–13 Moreover, smoking is more prevalent in lower income groups.14 15 Smoking exacerbates poverty.16 Low income groups rely heavily on public medical services. Expenditure on treating smoking related diseases in the public medical sector could be greatly diminished if the smoking rate was reduced in low income groups. In other words, when we assess the affordability of cigarettes, more attention should be given to lower income groups. Furthermore, in those countries with a high level of income disparity, the use of average income as an indicator may produce a biased result. For example, the gross national product (GNP) per capita for Hong Kong in 2005 was $HK199 49817; however, the median monthly household income for 2005 was only $HK16 000. Taking into account the average household size (3.1), the median annual individual income was only $HK61 936.18 The difference between the two figures is more than 300%, revealing that the use of GNP per capita to calculate cigarette affordability in Hong Kong results in an overestimation of the affordability level.

Sources and methods

In this paper, the affordability of cigarettes is defined as the ratio of the price of one pack of cigarettes to daily income (cigarette price-daily income ratio; CPDIR). A low CPDIR means high affordability, and vice versa. In assessing the income level, I used the mean of the lowest seven daily wage occupations of the 2006 UBS survey, which covered 14 occupations with a range of ranks.19 This approach minimises the distortion effect of income disparity. The household income of a particular percentile (for example, 20th or 30th percentile) may be more accurately reflected as the income of a low income group; however, periodic worldwide data are commonly unavailable. The World Development Indicator, published annually by the World Bank, seems to be the best available option; however, its usefulness is limited by the fact that the survey is undertaken in different years (from 1983 to 2003).20

The world cost of living, as published by the Economist Intelligence Unit (EIU), periodically provides the prices of more than 160 items found in over 130 cities worldwide, including cigarette prices in supermarkets and mid-priced stores (Marlboro and local brands).21 This paper used the lowest priced cigarette in 2006 published in this survey to calculate the cigarette affordability.

The selection of cities is solely based on the availability of data. From the data sets of the EIU and UBS, only 60 cities appeared in both samples; all 60 are therefore included in this paper.

RESULTS

Table 2 shows cigarette affordability levels in 60 cities, representing the cigarette price as a percentage of the daily income (CPDIR). The affordability levels range from 0.03 (Taipei) to 0.30 (Kiev). The cigarette affordability level of 34 of the 60 cities (56.7%) is equivalent to or below 0.10, and only four cities (6.7%) returned values higher than 0.20. These results show that most cities still have high cigarette affordability levels. Table 3 lists the cigarette affordability levels for 59 cities in terms of their respective economies’ income level classified by the World Bank. There is a tendency for cities with higher income economies to also have higher cigarette affordability levels. The cigarette affordability for 68.6% (n = 24) of high income cities was at a high level (CPDIR⩽0.10), whereas only 2.9% (n = 1) of high income cities had low levels of affordability (>0.20). In the upper middle income group, the percentage of cities with a high level of cigarette affordability declines to 45.5% (n = 5), with none at the lower levels. Most of the sampled cities in this income group had mid-level cigarette affordability (54.5%, n = 6). The cigarette affordability of cities with lower middle income was evenly spread: no single level of cigarette affordability represented 50% of the cities in this group. An interesting finding is that all low income cities have a mid-level cigarette affordability. The reason behind this may come from its small sample size (table 3).

Table 4 lists cigarette affordability in terms of geographical location. Most of the cities in Western Europe and South America have a high level of cigarette affordability, whereas 66.7% of their counterparts in Eastern Europe have a medium level. In North America, the relatively low cigarette price of the four sampled US cities changed the landscape for cigarette affordability in this area. The average cigarette prices for these four cities is only 75% of that in the sampled cities in Canada. In addition, the average daily income in the US cities is 1.2 times higher than that in the Canadian cities. In Asia, all cities with high cigarette affordability (40%) belong to upper middle to high income economies, with the exception of the Philippines. Three out of every four cities with low cigarette affordability are located in Asia. Given China’s relatively low income levels, it is no surprise that Shanghai and Beijing are included. An interesting point is the low cigarette affordability of Singapore, the only high income economy with a high CPDIR. The African cities of Johannesburg and Nairobi have a high and medium cigarette affordability level, respectively. This difference in affordability between the two cities can be attributed to the wide income gap between them (over three times).

DISCUSSION

These results show that the cigarette affordability for most of the sampled cities, especially those with high income economies, is high. Article 6 of the World Health Organization’s Framework Convention on Tobacco Control (FCTC) concerning price and tax measures does not clearly set a standard for cigarette and tax measures. Instead, it requests that signatories submit periodic reports of rates of tobacco taxation and tobacco consumption trends. This requirement motivates the signatories to take effective action. Reviewing cigarette affordability will surely become a core task of the tobacco control agencies of each signatory country. It is hoped that periodic monitoring and a reduction in cigarette affordability will reduce both the smoking population and cigarette consumption.

There is a risk that the rate of increase in cigarette prices within newly emerging economies is still much lower than their high rate of economic growth. When data from 2003 and 2005 are compared (data are available upon request), the CPDIR of 20 cities from a total of 57 (35.1%) had decreased, with 70% of these cities situated in Asia, Eastern Europe and South America. In Eastern Europe and South America, 57.1% and 83.3% of the compared cities in these areas recorded a decrease in their CPDIR, respectively. This study found that most cities within high income economies have a high cigarette affordability level. As the newly emerging economies approach high income levels, immediate measures should be taken to avoid a duplication of the experiences of their predecessors.