The Economic Regulation Of Tobacco Consumption in New Zealand

The Economics of Tobacco control: Towards an Optimal Mix edited by I. Abedian, R. van der Merwe, Nick Wilkins, P. Jha ( Applied Fiscal Research: University of Captetown, 1998), the proceedings of a conference of the Economics of Control Project, School of Economics, the University of Cape Town, 18-20 February 1998, Cape Town.

Keywords Health.

Introduction: A Brief History of Tobacco in New Zealand (1)

Tobacco was introduced in New Zealand by the Europeans about 200 years ago. Smoking became very popular with the Maori: surgeon Henry Weekes wrote in the early 1840s that it was “universal among New Zealanders [the Maori] of both sexes.” (Rutherford & Skinner 1940) It was also used as a currency and a commodity of exchange in the middle of the nineteenth century. By 1860 attempts were made to grow tobacco leaf, although this was not a successful industry until the 1930s. Cigars were being manufactured by 1884.

About 80 percent of adult males and 35 percent of adult females born at the turn of the century were regular smokers at some stage in their lives. The Maori rates were about the same for males and 65 percent or more for females. (Easton 1995)

Initially, opposition to smoking encompassed a purity argument and a health argument. The understanding of the health effects of tobacco was primitive: the first New Zealand Medical Journal reported in 1887-8 that tobacco was associated with insanity. An Anti-Nicotine Society formed in Christchurch in 1883, and in 1889 the Women’s Christian Temperance Union (also influential in the women’s electoral franchise of 1893) denounced tobacco.

Almost all the discussion was about male smoking, with few references to women. A special concern was juvenile smoking, it being believed that smoking was damaging to growing youth (but not to adults, as a number of smokers in the parliamentary debates at the turn of the century were wont to mention). Three factors reinforced concerns in the late 19th century: the introduction of cigarettes which made secret smoking harder to monitor, a general – concern about juvenile misbehaviour, and similar concerns overseas (the most influential countries being Australia, Britain, and the United States). Moreover, per capita tobacco consumption levels rose from 1.9lbs a year in 1887-1894 and 2.6lbs a year in 1904.

In 1882 legislation was introduced into the New Zealand parliament to ban juvenile smoking, although by the time the Juvenile Smoking Suppression Act was passed in 1903, a number of other legislatures had passed similar provisions. Its seven clauses prohibited supplying tobacco products to those under the age of 15 and smoking in a public place. (2) The Act has long since been repealed, but the prohibition of selling to youth (currently defined as under the age of 18) remains in New Zealand law. Although there were some prosecutions, the Act proved little deterrent to the rise of tobacco consumption.

Table 1: Annual Consumption

Year(s) Kgs per person over 15
1887-94 1.48
1904 1.73
1910-14 1.79
1921 1.86
1926 2.04
1931 1.91
1936 2.09
1941 2.50
1946 2.81
1951 3.29
1956 3.14
1961 3.30
1966 3.32
1971 3.11
1976 3.15
1981 2.91
1986 2.30
1991 1.79
1995 1.40

Source: Various official sources.
(1000 cigarettes = 1 kg)

Table 1 derived from various official sources show per capita consumption rising sharply after the mid 1930s, to a peak in the 1950s to 1960s, at just over 3.3kgs per adult, or double the 1904 level, declining thereafter. By 1995 consumption levels were back at those of the late 19th century. While 80 percent of men born before 1925 had ever-smoked (regularly) the proportion fell to below 45 percent of men born in 1970. The crucial factor may have been the ending from 1945 of widespread military service, where tobacco was cheap and social pressures favourable. About half of women from the cohort born about 1910 have been ever-smokers, a ratio which remained through to 1970.

Table 2: ADULT (Over 15) SMOKING RATES BY ETHNICITY (1996)
(ETHNIC GROUP) (ADULT POPULATION) (Male Prevalence) (Male Ever Smoking) (Female Prevalence) (Female Ever Smoking)
( ) % % % % % %
NZ European-Pakeha 69.2 22.9 48.4 21.2 41.7

NZ Maori 11.7 39.7 57.4b 47.4 65.0b
English 8.7 21.2 61.6a 18.5 42.8a
Scottish 3.2 25.5 55.3a 23.7 46.6
Samoan 2.2 33.3 42.6b 23.4 31.1b
Chinese 2.2 16.8 26.8b 5.9 9.7b
Irish 2.1 26.5 54.7 25.7 48.2

Australian 1.5 27.4 53.7 22.3 45.7

Dutch 1.3 22.6 58.1a 18.5 43.5a
Cook Island Maori 1.0 37.1 48.7b 37.9 49.2b
Indian 1.1 9.6 15.1b 5.9 9.4b
Tongan .7 39.5 50.0b 19.3 26.9b
Niuean .4 34.4 46.2b 30.9 40.8b
German .4 25.0 51.8 22.3 44.8
Korean .3 35.6 52.8b 6.3 9.5b
Welsh .3 23.1 55.4a 20.0 43.6

Filipino .2 20.0 32.8b 5.5 9.2b
American (US) .2 17.9 43.4a 15.5 38.2a
South Slav .2 24.4 49.1 22.6 39.4b
Fijian (ex Fijian Indians) .3 25.7 41.1b 22.1 35.7b
Japanese .2 37.4 54.1b 14.3 25.8b
South African .2 13.1 38.8a 12.9 32.9a
ALL 100.0 24.8 49.7 22.8 42.1

a = quit rate above 55%
b = quit rate below 45%
Source: 1996 Population Census

Table 2 shows current and ever (regular) smoking rates by ethnicity, as self categorised in the 1996 Population Census.(3) Today average smoking prevalence is at similar levels for men and women at just over one fifth of the adult population. Many previous smokers having given up, with a quit rate of near 50 percent for the adult population as a whole.(4)

The ethnic data needs to be interpreted with care since different groups have different age structures and educational levels, which also affect smoking behaviour.(5) Because the data has just been published there is no systematic commentary on it, but it confirms that the (New Zealand) Maori smoking prevalence rates are substantially higher than average and their quit rates lower.(6) Pacific Island smoking rates also tend to be higher than average.

Smoking is thought to reduce the average life expectancy of New Zealand smokers by about 7 years (Peto et al 1994) Some tentative calculations suggest Maori non-smokers live as long as non-Maori non-smokers, Maori shorter life expectation being almost entirely attributable to the higher incidence of smoking. Easton (1995) This may exaggerate the mortality impact of smoking, but it may well be the right order of magnitude.

Following the growing evidence linking tobacco consumption to various medical conditions (convincingly from the 1950s) a handful of New Zealand physicians led by Dr David Hay, medical director of the National Heart Foundation, campaigned actively against smoking. Initially this was individual medical advice rather than national policy, but a plethora of anti-smoking organizations developed. A Ministerial Advisory Committee on Smoking and Health was instituted in 1977.(7) Another official initiative which had considerable influence, was questions about ever-smoking and now-smoking in the 1976, 1981 and 1996 Population Censuses.

Among other influences on public policy on top of the growing awareness of smoking’s health effects, was that the removal of import controls led to the elimination of tobacco growing and a substantial reduction in cigarette manufacture with increasing offshore supply. Thus the size of the production lobby was significantly reduced. Tobacco products are generally sold in general outlets, rather than tobacconist shops, again blunting the supplier lobby. Another key change was the identification of passive smoking, which created a far larger reaction than the statistics warranted, compared to deaths from smoking. But it gave anti-smokers a political lever to repress smoking in public, since it was no longer a matter of the tobacco induced disease affecting only smokers.

I calculated the social costs of tobacco use came to a 1.7 percent loss of material GDP with a 3.2 percent loss of the total value of life (covering mortality and morbidity costs) in 1990. (Easton 1997) A summary of the conclusions appears in Easton (1997).

Fiscal Policy and Tobacco (8)

Excise duties on tobacco were initially for fiscal purposes. In 1839, the British colonial Secretary, Lord Normanby, confidently advised putative governor Hobson “[d]uties on the import of tobacco, spirits, wine and sugar will probably supersede the necessity for other taxation …” (McLintock 1958:90) There was a brief period in 1844 and 1845 when the duties imposed in 1841 were repealed, but since then excise duties on imports on tobacco, and later on domestic manufactures, have been an integral part of New Zealand’s fiscal revenue.

Normanby’s forecasts of excise duties being sufficient to fund government proved wrong, and in the 1997 fiscal year excise duty on tobacco products amounted to $658m, or 2.0 percent of total taxation and .7 percent of GDP. In addition tobacco, like almost all other products, has the uniform 12½ percent GST (Goods and Services Tax, a VAT) levied on it.

A 1994 study divided the cost of a pack of 20 (Rothmans) cigarettes into manufacture 22.4%, excise tax 55.1%, wholesale margin 2.8%, retail margin 8.5%, GST 11.1%. However since the GST (a uniform comprehensive sales tax) is imposed upon the excise duty, the totality of excise duty plus the resulting GST made up 62 percent of the price.(9) James (1995) estimated that the average smoker paid $763 in 1993 in excise duty, which might be compared with the average weekly wage of about $556.

Historically the justification for excise duty on tobacco has been a combination of fiscal pragmatism with the a view that consumption of tobacco was indulgent if not downright sinful. Because tobacco is price inelastic it is a good substance to tax without efficiency losses (whatever the health and equity considerations), while that the consumption activity is considered problematic by a significant and vocal proportion of the population makes the technical recommendation politically feasible. For instance during the debate following the excise duty hike on tobacco of the infamous 1958 “Black” Budget, there is no mention of health effects. The 1967 Taxation Review Committee does not discuss health aspects of any tobacco excise duty, either.(10)

But the 1970 budget stated that “it is clear that cigarettes and tobacco can be subjected to additional tax without harming in any way the general welfare of the community. In fact it is increasingly argued that discouraging the consumption of these commodities is likely to make a positive contribution to our general health.” (Muldoon 1970) By the 1977 budget the view had moved to “the adverse effects on health of smoking and drinking have been well publicised.” (Muldoon 1977) A specific sales tax was imposed, the proceeds of which were used for community health services. The levy did not last long, but was consolidated into the overall excise duty. The New Zealand Treasury is strongly opposed to tagged or earmarked taxes. Another complication was excise duties were not then inflation proof. Today they are indexed to the consumer price index.

The notion that excise duty on tobacco consumption for health purposes is now an integral part of the justification for the duty. It might be that were there no such case, the excise duty would be eliminated, since fiscal policy in general has attempted to eliminate special taxes in favour of a uniform GST. However that does not determine the level of the duty.

The narrowest view is that the duty should be sufficient to cover the cost to the state of tobacco induced diseases and the like. In 1990 public health treatment costs came to about $180m, while excise duties were $560m. Even if the reduction of taxation from mortality and morbidity were included, the revenue would still include direct costs to the state of tobacco consumption of $160m. (Easton 1997)

But even ignoring the matters covered in the next paragraph, the rule that special tobacco duties should equal the costs to the government is clearly wrong. Optimal decisions involve marginal conditions, but this is an average rule, although it could be argued that the marginal social cost of the consumption of tobacco is the average cost (which is obviously not true for the consumption of alcohol). Even so, social costs induced by tobacco use mainly occur a considerable time (even decades) after the act of smoking.

However the costs to the state are not the entirety of social costs. For comparison, and using the official value of a quality life year prolonged of $200,000, the total social costs came to $22.4b against the $560m revenue in 1990. (Table 2) Again for the reasons mentioned in the previous paragraph, there is no logical necessity that revenue from excise duty on tobacco should cover all the social costs of tobacco consumption.

Moreover, some of these social costs are incurred by the smoker (although there is some allowance for this in the calculations). Except for the addictive element of tobacco consumption, the treatment of such costs would be straight forward. If there was no addiction the social costs incurred by the smoker would offset their private costs by the gains to them of smoking.(11) However where there is addiction the rationality assumption is harder to apply. Moreover the way in which the New Zealand official value of life is calculated implies that part of the value of the individual’s life to other people so a rational person smoking her or himself to death (or trying to commit suicide) still generates social costs to others.

Whether through health and social costs arguments, or through modern variations of the purity argument (presenting tobacco as a demerit good), the anti-smoking pressure groups have lobbied vigorously for higher taxation on tobacco, with marked success in the late 1980s and early 1990s. A government under fiscal stress found it convenient to respond positively to the lobbyists’ demand.

The 1991 tax hike resulted in little extra excise revenue. This might suggest that the excise duty rate had reached the no additional revenue level, but it is generally thought that other (income reducing) measures taken at the same time, which were particularly harsh on the poor (who are the heaviest smokers) resulted in the cutting back their smoking to save expenditure. However there has been no real increase in tobacco excise duties since 1991. In any case we see here the potential of a conflict between the fiscal purpose of raising revenue and the health purpose of reducing consumption.

In evaluating the impact of excise duty, we need to think of at least three groups of smokers with differing behaviour, although little is known about their proportions, other than anecdotally.

(1) The addicted, typically older, smoker who has already resisted a range of economic and non-economic incentives to give up smoking. For most a hike in excise duty reduces their real income, rather than their tobacco consumption.

(2) Occasional, light, and non-addictive (adult) smokers whose consumption may be price elastic/sensitive to price increases (although occasional smokers may not be). However this group may not be a high policy priority except to discourage their joining the first group.

(3) New and potential smokers, who in New Zealand are almost entirely teenagers, since few begin smoking after the age of 18, typically having highly price elastic demands. One estimate put the youth price elasticity at 1.1, although the sample population includes 20 plus year olds so the true elasticity for teenagers is likely to be even higher. (Laugesen & Meads 1990) Excise duty hikes may not appear to reduce their consumption in the short run, because it is already low. In the long run, because they dont join the first group their potential consumption may be cut drastically.

A tax hike will impact on these different groups in different ways. In particular it may not be particularly effective on the first group, but may be effective on the long term smoking behaviour of the third.

A complication is the magnitude of the price (and hence excise duty) change may affect the behavioral response (in contrast to standard economic analysis which assumes that the aggregate elasticity from a number of small changes will be the same as the elasticity for a one of increase of the same aggregate magnitude). The addicted may be inured to small changes and it is only when there is a large price hike that they take the short term pain to reduce their consumption and save revenue. Fiscally then, optimal revenue is to be gained by regular small increments, whereas reductions of consumption requires occasional large hikes. The strategy of the 1990s of indexing excise duty to the consumer price index belongs to the first strategy, while the occasional major hikes in the 1970s and 1980s to the second.

Note there is a leakage in the existence of duty free entitlements for travellers (all the more ironically on no-smoking flights). New Zealanders are big international travellers – averaging about one overseas trip every three years. The elimination of duty frees between European Union countries in the near future is likely to lead to pressures for a similar elimination between Australia and New Zealand where there is high economic integration. (There may also be pressure for elimination between the two and the independent South Pacific states.) Whatever the economic case for duty frees generally, the case for duty free tobacco imports is thin.

There have been a number of New Zealand econometric estimates of the price and income elasticities for tobacco products, and some for advertising (and even news items). These are summarised in James (1995:23). Not all the estimates are significant, consistent with one another, or with a priori theory. Sometimes the data period is far too short to provide quality long run estimates. For instance, if we believe higher prices discourage teenagers taking up smoking, the full effect of an excise duty hike will not be for decades.

Non-fiscal Regulation

As reported in the historical section, legal prohibition were imposed in 1903 on juveniles from being supplied with tobacco products, or smoking in public places. In principle such restrictions can be converted to fiscal measures (a prohibitively high tax rate on the activities), although practically the fiscal alternative may not be very operational. Other prohibitions began to be imposed in the late 1980s. The main ones can be summarised as follows:

(1) Restrictions on commercial access to tobacco by the young (currently under 18).

(2) Prohibitions on advertising, sponsorship, (with a few exceptions, typically concerned with international events), display, and compulsory labelling including health warnings. (Ten packs have just been removed from circulation, with the hope that this will reduce purchases by the young and poor. Plain packs are a major item on the reformer’s agenda.)

(3) Creation and extension of smoke free environments. (The rules have been supported by many businesses – most evidently airlines – happy to comply in their own interests while attributing the restrictions to the government.)

It is not my task to detail these measures. In economic terms the first two might be primarily thought as being concerned with limiting access and improving understanding of adolescents who are judged unable to make quality rational decisions. There is a complementary anti-smoking education campaign.

Non-smoking areas reflect another economic principle – the allocation of property rights. It was not so long ago that smokers had an informal social right to pollute other people’s air with their smoke (even if the polite asked for permission, the expectation was that it would be given). Today that position is reversed and it is the non-smoker who usually has the (legal) right in public places to determined the air quality (in regard to tobacco smoke).(13)

For completeness it is mentioned that there are also private restrictions on smoking. For instance the Maori, having taken over responsibility for their (typically government funded) own anti-smoking campaigns, have applied smoking prohibitions to their marae (meeting areas and halls). (On the other hand the Maori derive no direct benefit from an excise duty hike,(14) while they have a particularly high proportion of addicted smokers, so their lobbyists have shown some resistance to using tax for tobacco control.)

Generally we know very little about the effectiveness of these various measures on overall tobacco consumption.

Some Issues

Epidemiologists will note that as in the case of many other epidemics, smoking prevalence peaked before public policy took conscious action. Indeed it might be argued that it was only possible to take public initiatives when smoking was in retreat, and the anti-smoking lobby sufficiently augmented by ex-smokers to have the required political weight. But if the smoking epidemic has peaked, the health consequences of smoking have not, for there is often a long lead time between smoking and the resulting disease. While the health consequences for men smoking have probably peaked (or troughed) the damage for women smoking is still rising.

A figure in Tobacco Statistics shows the per capita tobacco consumption between 1921 and 1991. It may be a little misleading in the 1940s, since considerable numbers of New Zealand smokers were in the armed services overseas. Indeed the experience probably encouraged smoking, given the social circumstances, and the cheap tobacco provided to soldiers.

Allowing for this the main trends are the switch from loose tobacco to cigarettes, the rising consumption in the period till immediately after the war, the plateau from then till the early 1970s, and the decline thereafter (steep from the 1980s).

We do not have enough historical data to provide a confident account of the changes, but a best conjecture might go something like this. Before 1940 tobacco consumption, especially cigarettes, became increasingly fashionable as the tobacco companies sought market expansion, including among women’s markets. This probably continued during the war, although the numbers overseas obscure the data. On returning smokers faced much higher tobacco prices because of excise duties which had been imposed during the war for excise duties. This discouraged increasing consumption levels, and discouraged young men from taking up smoking regularly. The momentum of past addiction, plus rising prosperity continued the high levels of tobacco consumption, but the underlying long term consumption was fragile.

The tax hike of 1958 probably led to further quitting, and discouraged the young taking up the habit. (Easton 1967) Probably health and fitness concerns affected social judgements, especially among the better educated (and higher income). It seems possible that higher socio-economic classes took up smoking and smoked more before the 1930s, and they quit earlier, so that generally smoking in New Zealand is a lower socio-economic phenomenon (and hence tobacco taxation is now regressive). The reduced consumption of those in the highest socio-economic classes made smoking less fashionable, and facilitated the anti-smoking policies, which collectively have led to the rapid decline in recent years.

If this account has any veracity, it may have an important implication for developing countries with low tobacco consumption. Their situation is more like New Zealand’s in the early part of the twentieth century. This is not a gloomy prediction of the inevitably of tobacco consumption levels accelerating, but a drawing attention to an earlier use of economic instruments than occurred in New Zealand may prevent consumption levels and disease rising to Western peaks.

But New Zealand cannot be complacent about the reduction in its smoking levels. The latest data suggests there may have been a small rise last year, although this may be a statistical artefact so further data is necessary to make an assessment.(14)

The historical account given here is sketchy. International comparisons may enable gaps to be filled by the pooling of the scarce data. However it lacks one component, characteristic of much economic analysis. Differences in social behaviour are taken as given, rather than explained. But why do men and women have different smoking behaviour? Why do the Maori behave differently from the non-Maori (and why are Pacific Island smoking rates more similar to Maori than non-Maori)? Perhaps some progress can be made with the enormous data base of the 1996 Population Census, but ultimately the economist faces a reality of important phenomenon being explained by other disciplines.

A similar problem applies to the phenomenon of addiction. It may be too much to expect economics to explain addictive behaviour, but until there is an explanation economic policy must tread warily about how its policy recommendations do or may work.

Yet the discipline of economics has more to contribute than just demonstrate the effectiveness of some policy instruments. From its public policy perspective, there is a certain lack of clarity as to what is the policy objective(s). Why eliminate all tobacco consumption? Because all tobacco use is detrimental to health (but there are other poor health practices – bungey jumping, crossing the road, over-eating – we do not treat in this way)? Because it is addictive? Because tobacco smoke and litter offends our sensibilities? Because it is “impure”, that notion which drove much of late nineteenth century policy (and if the equivalent twentieth century notion is that tobacco is a demerit good, why is tobacco one)? Or perhaps it is a convenient way to raise state revenue and the rest is a veil to hide this justification.

What strikes an economist is that there are multiple policy instruments, which suggest that within the anti-tobacco campaign there are multiple policy objectives (or sub-objectives). By better identifying them we may be able to better target the policy instruments or devise new ones. Let me give a couple of examples:

Currently excise duty is levied on tobacco weight, without any adjustment for the chemical content of the tobacco, implying the policy objective is tobacco consumption. However suppose it is tobacco addiction. Since the tobacco companies appears to add nicotine to some cigarettes, which contributes to addiction, there might be a case for the an excise duty related to nicotine content. On the other hand suppose it is health. Then there might be a case for relating the duty to the most health damaging elements in the tobacco, such as tar, encouraging the producers to shift the composition of these elements.

Second, a single instrument of excise duty is trying to target two policy outcomes: discouraging teenagers from starting smoking, and encouraging adults to give up smoking. Addicts can argue that the policy instrument is ineffective for their health but is financially punitive. Their political resistance inhibits the use of the tax instrument for the first purpose. One resolution might be to raise the excise duty to a much higher level, but enable (diagnosed) addicted persons to obtain a quantity of tobacco products on prescription at an excise duty discount. Any entitlement might be only those over 30 years old, and at a level which was adequate rather than generous (allowing them to purchase in the higher price open market if they wish for more).

This leads to a final general theme of this paper. Whatever the technical economic recommendations, there is a political problem in obtaining enough public support for them. Crucial in the New Zealand political experience has been the staunchness of the medical professionals on the health effects of smoking, the desperation of the Treasury for raising fiscal revenue, the significance of passive smoking to the non-smoker, and the destruction of the tobacco growing and manufacturing industry as a part of industry (rather than health) policy. Further progress may require a separating the addicted from the potentially recruitable smoker (as discussed in the previous paragraph).

Bibliography

Broughton. J. (1996) Puffing up a Storm: `Kapai te Torori!’, Department of Preventive and social Medicine, University of Otago.

Chetwynd, J., R. Brodie, and R. Harrison (1988) “Impact of Cigarette Advertising on Aggregate Demand for Cigarettes in New Zealand”, British Journal of Addiction , 83, 409-414.

Easton, B.H. (1967) Consumption in New Zealand 1954/5 to 1964/5, (NZIER Research Paper 10, 1967).

Easton, B.H. (1991) Economic Instruments for the Regulation of Licit Drugs, Paper to Perspective for Change conference, November 1991.

Easton, B.H. (1995) “Smoking in New Zealand: A Census Investigation”, Australian Journal of Public Health, Vol 19, No 2, p.125-128.

Easton, B.H. (1997) The Social Costs of Tobacco Use and Alcohol Misuse, Public Health monograph, Department of Public Health, Wellington School of Medicine.

Evans, L. & C. Meads (1991) An Empirical Study of the Effects of Advertising, Prices and Incomes on Cigarette Consumption in New Zealand, Victoria University of Wellington.

Harrison, R. & J. Chetwynd (1990) Determinants of Aggregate Demand for Cigarettes in New Zealand, Discussion Paper 9002, Department of Economics, University of Canterbury.

Harrison, R., J. Chetwynd, & R. Brodie (1989) “The Influence of Advertising on Tobacco Consumption: A Reply to Jackson and Ekelund,” British Journal of Addiction, 84, 1251-1254.

James, D. (1995) A Review of Tobacco Taxation, NZIER Contract 723, Wellington.

Laugesen, M. & C. Meads (1990) “Cigarette Advertising, Price, Income Publicity and smoking Prevalence in Youth Versus Older Age Groups, New Zealand. 1982-1989”, Tobacco and Health 1990, Proceedings of the Seventh World Conference on Tobacco and Health, Perth.

McClintock, A. H. (1958) Crown Colony Government in New Zealand, Government Printer, Wellington.

Meads, C. (1991) An Empirical Study of the Effects of Advertising, Prices, and Incomes on Cigarette Consumption in New Zealand: 1973-1986, Masters Dissertation, Victoria University of Wellington.

Muldoon, R.D. (1970, 1977) Economic Statement, Government Printer, Wellington.

Peto, D., A. Lopez, J. Boreham, M. Thun & C. Heath (1994) Mortality for Smoking in Developed Countries, Oxford University Press, Oxford.

Public Heath Commission (1995) Tobacco Taxation as a Health Issue, Wellington.

Rutherford, J. & W.H. Skinner (1940) The Establishment of the New Plymouth Settlement, New Plymouth.

Salter, D. (1981) “The Demand for Cigarettes in New Zealand – an Econometric Analysis of the Effects of Increased Taxation”, Economics and Health: Proceedings of the Second Australian Conference of Health Economists, Australian National University Canberra.

Salter, D. (1985) Paying Lip Service to Health: An Evaluation of the National Government’s Policy on Cigarette Taxation, 1975-1983, Department of Community Health, Wellington Clinical School of Medicine.

Simester, D. & R.Brodie (1990) The Effects of Advertising on Brand and Industry Demand for Tobacco: A Meta-analysis of Economic Studies. Working Paper, Department of Marketing and International Business, University of Auckland.

Statistics New Zealand (1997) “1996 Census of Population and Dwellings: Population Characteristics”, Hot of the Press, 7 May 1997.

Taxation Review Committee (1967) Taxation in New Zealand, Government Printer, Wellington. (The Ross Committee)

Thomson, S. (1992a) Evils of `The Fragrant Weed’: A History of the 1903 Juvenile Smoking Suppression Act, Essay in partial fulfilment of MA degree, University of Auckland.

Thomson, S. (1992b) Stubbing out the Social Cigarette, Essay in partial fulfilment of MA degree, University of Auckland.

Tobacco Statistics 1991, Department of Statistics & Department of Health, Wellington.

Endnotes
1. Attendance at the conference was made possible by assistance from New Zealand’s National Heart Foundation and Central Institute of Technology and the conference hosts – the University of Cape Town and the Medical Research Council of South Africa. I am grateful to Sarah Thomson and Murray Laugesen for comments on an earlier draft of this paper.
2. This section is based mainly on Thomson (1992a,b), which is the source for all facts unless otherwise stated. Broughton (1996) contains much useful historical information on the Maori and tobacco.
3. There was a caveat that a youth could so smoke if they had a medical certificate that it would be beneficial.
4. Some respondents give two or more categories – e.g. Maori and Pakeha – so the total of ethnic groups exceeds the total population.
5. The quit rate equals 1 – prevalence/ever-smoking.
6. Another factor to be investigated is whether the respondent is locally or overseas born, which may explain the divergence in Asian rates.
7. Easton (1995) shows the true ever-smoking rates are higher, especially for Maori relative to the total, because the differential mortality of smokers lowers (relatively) the numbers alive who report having ever been regular smokers.
8. Maori anti-smoking organizations became active in the 1990s reflecting a growing commitment by the Maori to deal with their own problems.
9. This section is based on Easton (1991), which also covers alcohol duties, and James (1995)
10. Sometimes the entire GST is included in the calculation of the tax on tobacco products (which would take the total up to 66 percent). This is misleading since GST is imposed upon (almost) all goods and services. Insofar as analysis is concerned with the relative price between tobacco and other products the effect of the GST is neutral. Note that excise duty makes up 62 percent of the pre-GST price, just as it (with the GST levied on it) makes up 62 percent of the post-GST price.
11. It hardly discussed excise duties at all except to note they tend to be regressive.
12. The study uses the convention that private costs are included in social costs.
13. It is interesting to construct a formal model of individual property rights to replace the law a la Coase, although it soon becomes evident that the transaction costs of such an arrangement make it impractical.
14. They could be an indirect beneficiaries, depending how the additional revenue was spent.
15. The data comes from supplies from bond and hence does not allow for changes in commercial inventory levels. Another confounding effect may be purchases by tourists from Asia.