In 1945 72 per cent of Australian men aged 16 and above and 26 per cent of women smoked. By 1976 male smoking had dropped to 43 per cent of malesbut female smoking had increased to 33 per cent of females. Male smoking has declined ever since and in 2007 (the latest year for which data is available) only 21 per cent of Australian men smoked. Female smoking has fallen to 18 per cent. This latter data is for adults over 18 years of age but an Australian Bureau of Statistics survey in 1977 found that 43 per cent of men aged 16 and above smoked while 29% of women smoked (here) so the slight inconsistency in categorization does not make a significant difference.
Using the ABS historical population statistics one can estimate the population of males and females over 18 in 1945 as 2.481m and 2.504m respectively whilst figures for 1976 are 4.448m and 4.537m and figures for 2007 are 7.616m and 7.885m (here).
Combining these two sets of data we have the number of male/female smokers in 1945 as 1.786m males and 0.651m females or 2.437m persons, in 1976 as 1.913m males and 1.497m females or 3.410m persons and in 2007 1.599m males and 1.419m females or 3.108m persons.
Smoking rates fell dramatically over the 31 years from 1945 to 1976 but absolute numbers of smokers increased due to population growth by about 40 per cent. Over the 31 years 1976-2007 smoking rates continued to fall but now absolute numbers of smokers fell by about 9 per cent despite continuing population growth.
There are various estimates of cigarette consumption per head of the overall population and per head of the smoking population (here). Our interest is mainly in the latter. The data is imperfect and the various ways of estimating consumption give different measures of it. To summarise what is known from the various approaches:
- Total consumption in 1976 was 30,431m cigarettes while in 2000 it was 23,850m.
- Total kilograms of tobacco consumption on which duty was paid (this should be most tobacco consumed) was 10.4m in 1945, 32.56 in 1976, and 18.56m in 2006.
- Total cigarettes consumed per adult smoker increased from about 17 per day in 1964 to about 27 per day in 1992 suggesting significant increase.
- Total cigarettes consumed annually per smoker varied from 2,117 in 1998 to 1,398 in 2006.
- Estimated cigarettes consumed per week by those 14 years and over increased from 140 in 1998-99 to 124 in 2001-02 to 134 in 2004-05.
- Self reported cigarettes smoked per smoker per day fell from 15.4 to 13.0 over the period 1997-2006.
Other measures are possible and discussed here. But there are interesting features of the measures already presented. Using measures 1. and 2. the prevalence of smoking fell less quickly than did cigarettes smoked per smoker fell from 1976 to 2006 – this trend is even clearer for the latter part of this period from estimates 4. This contrasts with the earlier (partly overlapping) period 1964-1992 when 3. suggests smoking per remaining smoker increased
The self reported results 6. and estimates 5. show a much smaller decline per capita consumption among smokers and, indeed, estimates 5. show an inexplicable non-monotonicity. Estimates 5. and 6. show much smaller effects than these.
It would be worthwhile knowing what is going on here and suggestions are welcome. Consumption per head rose at first during the post-war period – perhaps partly because of the introduction of low tar, low-nicotine cigarettes so total numbers smoked went up even though even though the prevalence of smoking dropped markedly. This reflects what tobacco researchers call the ‘compensation’ phenomenon – smokers smoke more when nicotine levels in cigarettes decrease. After about 1976, prevalence decreased but so too, according to estimates 1., 2., 4. did consumption per remaining smoker which fell dramatically.
This would suggest that the measures taken in recent years to reduce smoking (tax hikes, abolition of cigarette advertising, negative advertising) have reduced the numbers of smokers but have also reduced consumption among those who continue to smoke. That is interesting if you take the view that smoking is an addictive activity that eventually leads to an equilibrium level of dependence that is largely independent of price. According to this view the only way people can offset the effects of such things as tax hikes is to quit smoking. It is impossible to ‘scale back’ the level of smoking for a nicotine-addicted smoker. The data presented suggests this view is incorrect since people do seem to be able to also reduce the daily amount they smoke.
One way to revive the equilibrium addiction viewpoint is to suppose that, in response to higher taxes, smokers smoke more intensively – they puff harder and smoke right down to the filter tip. This of course delivers a higher load of nicotine that satisfies the addiction (and of carcinogenic tars). But given that the changes in per capita consumption are strikingly large and, in the absence of any evidence one can only guess, that this explanation seems far-fetched.
Another resolution might be that it is the heaviest smokers who either have quit smoking in recent years or who have died. This gloomy implication of the health consequences of smoking is again nothing more than a plausible guess but is presumably at least part of the story. but given the abruptness of the change it can only be part of the story.