Effects of alcohol controls: Nordic research traditions

Drug and Alcohol Review (March 2004), 23, 43 – 53

SPECIAL SECTION: PREVENTION

Effects of alcohol controls: Nordic research traditions ROBIN ROOM Centre for Social Research on Alcohol and Drugs, Stockholm University, Stockholm, Sweden

Abstract Drawing on a recent review of studies of the impact of alcohol control changes in the Nordic countries (particularly Finland, Norway and Sweden), this paper reviews the development of research traditions of such studies in the Nordic countries. From the Nordic experience, there is evidence of variation in the effects of policy changes by demographic segment, by type of problem and by drinking pattern and amount. Policy changes have often had their greatest effect on heavier drinkers. Big reductions in alcohol taxes in Denmark in 2003 and Finland in 2004 offer a new chance to study whose drinking changes how much, and in what contexts, in a collaborative study comparing northern Sweden with Finland, Denmark and southern Sweden. [Room R. Effects of alcohol controls: Nordic research traditions. Drug Alcohol Rev 2004;23:43 – 53] Key words: Alcohol control, Finland, Nordic countries, Norway, policy evaluation, Sweden.

Introduction The focus of this paper is on studies of the impact of alcohol control changes in Nordic countries, and particularly in Finland, Norway and Sweden, which have been carried out in the last half-century. This tradition of studies has recently been reviewed extensively [1], and this paper relies on that review as its canon of studies—with the addition of several new analyses published in the same volume [2] and the recent Swedish Saturday-opening study [3,4]. Our discussion is limited to studies of discontinuous changes in the physical availability of alcohol. The focus on studies of discontinuous change excludes studies of the effects of gradual changes in alcohol controls—for instance, of year-by-year changes in the number of alcohol sales outlets. Excluded also are studies of the effects of changes in the prices or taxes for alcoholic beverages. Finally, studies of the effects of changes in drinking-driving laws are excluded, although this is an important literature in its own right. However, the effect on drinking-driving casualties is often used as an outcome variable in studies which are included here.

Nordic traditions of studying the impact of changes in alcohol control (this section draws on [5]) The modern tradition of studies of the impact of alcohol controls may be said to have started with the commitment, early in the history of the Finnish Foundation for Alcohol Studies, to study the effects of changes in the Finnish alcohol control system, and in fact to base changes in the alcohol control system on prior studies of the effects of the changes [6]. The commitment to this rationalistic and experimental approach proved fitful in the succeeding decades. However, the early Finnish commitment to this approach established what eventually became a tradition of such studies, starting with Pekka Kuusi’s classic controlled experimental study of the effects of opening of liquor stores in country towns [7], and studies of the effects of the ‘buyer surveillance’ system under which social workers for the Finnish alcohol monopoly made home visits to customers with suspiciously large purchases [8]. Finnish social alcohol research was organized after 1950 as a department of the state alcohol monopoly. With a sociologist, Pekka Kuusi, as the director of

Robin Room PhD, Professor and Director, Centre for Social Research on Alcohol and Drugs, Stockholm University, Stockholm, Sweden. Correspondence to Robin Room, Centre for Social Research on Alcohol and Drugs (SoRAD), Stockholm University, Sveaplan, 106 91 Stockholm, Sweden. E-mail: [email protected] Received 1 August 2003; accepted for publication 27 October 2003. ISSN 0959-5236 print/ISSN 1465-3362 online/04/010043–11 # Australian Professional Society on Alcohol and Other Drugs DOI: 10.1080/095952304100001645547

44

Robin Room

the Finnish alcohol monopoly, it was natural that studies of the effects of alcohol control measures received an early emphasis in the research programme. The present-day Alcohol and Drug Research Group of STAKES in Helsinki carries on the tradition of policy impact studies thus initiated, as the successor to the Social Research Institute for Alcohol Studies. In 1959, what is now known as the National Institute for Alcohol and Drug Research (SIRUS) was founded in Norway, and policy impact studies have been a central task for it. In the other three Nordic countries, alcohol policy impact research has been carried out on a more ad-hoc basis. In Sweden, projects were initiated and funded by government commissions, the alcohol retail monopoly (Systembolaget) and more recently the Public Health Institute (FHI). Different national research councils have also provided funding to individual researchers or groups of researchers for impact studies. In 1999 the Centre for Social Research on Alcohol and Drugs (SoRAD) came into existence as a Swedish equivalent of the Finnish and Norwegian research groups. In Finland, Norway and Sweden, alcohol policies were seen as part of the welfare state, and evidencebased knowledge acquired by scientific methods was seen as a necessary element in the social planning of the welfare state. In line with this belief in science and progress, alcohol policy impact studies were considered to be an instrument for the implementation of effective alcohol policies. The vision was reinforced by the notion that the state had responsibilities in providing funds specifically for alcohol research. Studies of changes in tax or price levels, in opening or closing days or hours for stores and taverns, of the opening of new outlets, of the introduction of new beverages, or of strikes which temporarily limited alcohol availability, have therefore been a part of the Nordic alcohol research agenda. Unplanned disruptions, such as strikes, tend to happen with short notice. If they are to be studied with specially collected data only a short time can be spent on study design, construction of the necessary measuring instruments and data collection. In such cases, existing longitudinal studies and panel data may form the most valuable resources for studying the effects of changes in alcohol control. Such datasets are usually the result of long-standing and coherent research projects, in many cases carried out by national alcohol research institutes. The benefits of settled institutes in Finland and Norway have also been that researchers have been able to attend to research questions that arise from sudden policy changes. A majority of the Nordic alcohol policy impact studies have been carried out by specific alcohol institutes or as governmentally funded projects.

The Nordic studies have varied in the strength of their designs. The recent study of the effects of Saturday opening in Sweden [3,4] was a fully experimental design, with a random assignment of different parts of Sweden to the experimental and the control condition. More common, beginning with Kuusi’s study, have been quasi-experimental designs, with control sites. Such designs have been typical when control changes are being tested in particular communities, as for instance in many of the Norwegian studies. Studies of country-wide policy changes—for instance, the study of the Finnish changes in 1969—have typically had no control site. In studies of the effects of alcohol controls, unlike other areas of social alcohol research, there has been no tradition of collaborative studies across national borders (but see the last section of this paper). A number of studies have collected before and after general-population survey data. The two surveys have usually been of independent samples (the Finnish 1968/ 1969 studies were an exception in collecting longitudinal data from the same respondents, but these were not analysed longitudinally until recently). Reflecting that funding for the study’s fieldwork often had to be found from general institute budgets, the sample sizes in the surveys were often rather small, so that only rather dramatic results would yield significant findings. Relying on the excellent Nordic health and social statistics, almost all the studies have drawn on alcohol sales data, hospitalization data or criminal statistics for measures of the effects of the changes, and many studies have relied primarily on such data. With some notable exceptions (again, Kuusi’s study comes to mind), the strongest evidence in the Nordic studies has generally come from the social and health statistics data. The general findings from the Nordic studies (this and the following two sections draw heavily on [9]) Drawing on and extending a summary table from a review of the Nordic studies [1], Table 1 presents a brief overview of the results of the different studies. Our focus here is not only on the effects in the total population, but also particularly on differential effects in different subgroups of the population. In the first place, in the Nordic experience, big changes in the physical or financial availability of alcohol can clearly produce big effects. The Swedish changes of 1955 and the Finnish changes of 1968 produced immediate large increases in the population’s level of alcohol consumption, and also had substantial effects on indicators of alcohol-related problems. On the other hand, a comparison of the effects of the increased availability of beer in Finland in 1968 with that in Iceland in 1989 suggests that

Effects of alcohol controls: Nordic research traditions

45

Table 1. Summarizing the results of Nordic alcohol impact studies (revised from Ma¨kela¨ et al. [1])

Country and period

Effects on total consumption

Abolition of the Swedish rationing system Sweden 1955

Increase (25%)

Effects of buyer surveillance system Finland 1949 – 52, 1955

Sales of medium beer increased by 242%

Finland 1993/94 (local medium beer ban removed)

Sweden 1967 – 68 (strong beer in groceries) Abolition of beer in grocery stores Finland 1975 – 77 Norway 1972, 1975, 1981

Legalization of beer sales Iceland 1989

Heavy consumers’ consumption increased most after the abolition Alcohol-related harm increased relatively more than consumption No better development in drinking among those abusers who were exposed to sanctions The end of purchase registering resulted in easier access to alcohol among abusers

Introduction of new types of beer in grocery stores Finland 1969 (beer 5 4.8% in groceries) Strong increase in total consumption (46%)

Sweden 1965 ( 5 4.6% medium beer in groceries); 1977 (removed)

Differential effects by demography, by drinking patterns and on harms

Consumption of medium beer increased heavily Total consumption increased by 15%

Sales of strong beer increased by 1124% Total consumption rose by *5% Slight decrease in volume No significant change in total consumption

Rise in total consumption of *15 – 20% when beer introduced Effect dampened by recession

Greater relative increase in consumption for women, in previously dry areas, among younger adults Bigger increases in consumption among those already drinking heavily Frequency of light drinking occasions increased the most Many abstainers started drinking Alcoholics replaced spirits with beer A strong increase in morbidity, mortality and arrests for drunkenness Arrests for drunkenness increased particularly among youth ( 5 18 years) and among women No visible effect on public drunkenness Among the young, beer consumption increased, but not total consumption Medium beer most popular among youth Hospitalizations for alcohol diagnoses fell among teenagers when medium beer removed from groceries Motor vehicle accidents declined significantly among teenagers and middle-aged and older with medium beer removal Persons taken into custody for drunkenness: no clear effect, nor for young people The young not affected In some municipalities the use of illegal and non-beverage alcohol rose Shift from medium beer to other alcoholic beverages Same for youth Purchases of beer decreased particularly among women Increase in beer consumption more due to increase in quantity drunk per occasion than increase in frequency Intoxication frequency decreased among men but not among women Biggest increase in consumption for men (continued overleaf )

46

Robin Room

Table 1. (continued )

Country and period


(Iceland, contd)

Introduction of light beer Norway 1985 Opening of new wine/liquor stores Finland 1951

Norway 1961, 1968, 1971, 1991

Differential effects by demography, by drinking patterns and on harms For men, net increase in mean consumption greater among heavier than lighter drinkers Big increase in consumption for boys (13 – 19 years old) Shift from spirits to beer No significant change in the number of abusers Increase in persons detained without having caused a disturbance

No effect on total consumption

No changes

Introduction of self-service in monopoly stores Sweden 1989 Increase in sales figures of 17%, 12 due to actual increases in purchases by local residents Saturday closing/opening Finland 1977, 1978 Decrease in sales from monopoly stores and an increase in retail sales ? a slight total decrease

Norway 1984

Modest decrease in sales of spirits

Sweden 1981, 1982

No change

Volume and frequency increased among men but not among women or youth Frequency increased more among frequent drinkers; no effect on non-drinkers Older men ( 4 50) and men in white collar jobs less affected Beer and wine replaced spirits and illicit beverages, especially among boys and women No effect on quantities drunk per occasion Slight increase in consumption among women and elderly and in intoxication among women Purchases: more often but less at a time Wine consumption increased and use of moonshine decreased No changes in consequences were observed

Arrests for drunkenness decreased on Saturdays, particularly among people aged 30 + years and among the homeless Illegal sales increased; use of non-beverage alcohol did not No effects on monthly totals of harm rates Purchases: less often but more at a time Detoxification centre: a significant decrease in admission on Saturdays and Sundays, and a larger proportion of illegal and non-beverage alcohol Decrease in police arrests for public drunkenness on Saturdays but increase in the other days Less public drunkenness More frequent customers most affected Purchases: less often but more at a time No change in emergency visits or injuries in traffic accidents Less domestic disturbances Increase in outdoor assaults (continued overleaf )


47


Country and period


Sweden 2000

3.4% increase with Saturday opening

Longer serving hours Iceland 1999

Sales decreased by one-third

Norway 1978

Sales decreased by one-fourth, consumption by 5 – 10%

Norway 1982 (100-day strike)

Spirits and wine sales for year down *20%, total sales of alcohol down 10%

Sweden 1963

Increases in assaults not significant except in northern Sweden. Significant increase in drinking-driving, but may reflect increased policing 13% rise in emergency room admissions, including 24% for accidents, 34% for fights Increases in ER admissions limited to men 80% increase in drinking-driving cases, may reflect changes in policing

Introduction of age-of-purchase limit (15 for off-sales) Denmark 1998

Alcohol store strikes Finland 1972, 1985


Drop of 36% in reported consumption by pupils aged under 15 Drop among pupils aged *15 – 17 of 17% Discussion around new law sensitized parents to teenage drinking? Few reported being affected Men were affected more than women Frequent drinkers were affected more than infrequent drinkers Alcoholics’ consumption decreased Consumption of strong home brews and moonshine increased Young ( 5 35) and white collar workers used alternative sources of alcohol the most Use of non-beverage alcohol increased, especially in 1972 Alcohol-related crimes decreased Arrests for drunkenness decreased by one-half; in 1972 more but 1985 less among socially integrated; more among men aged over 30 Increase in home production and smuggled spirits, especially among heavy consumers Women more affected Number of patients in detoxification centres decreased Police reports on drunkenness and domestic disturbances decreased Decrease in accidents, especially falls and in low socioeconomic status areas 41% fall in detox centre admissions during strike Violent crime rate down 9.2% in year of strike Arrests for disorderly conduct and for drinking-driving for year not significantly affected Moderate drinkers hardly affected Alcoholics strongly affected Effects were similar among men and women (continued overleaf )

48

Robin Room


Country and period


(Sweden contd)

Decrease in police interventions due to drunkenness; no differences between age groups; area closest to Denmark least affected Decrease in accidents among both moderate and heavy drinkers Number of patients in alcohol clinics decreased

the magnitude of the effect also depends on other circumstances of the time. The change in Finland in 1968 has been seen in terms of a belated opening of the doors to pent-up demand from a new ‘‘wet generation’’ [10,11], whereas in Iceland there was a fair degree of de-facto availability of beer before the legal change. At least as importantly for short-term effects, the Finnish economy was flourishing in 1968, while Iceland was moving into a recession in 1989. The effects of smaller changes in availability, however, seem more variable, and often negligible in terms of the effects on total consumption. A series of Norwegian studies of the effects of restricting beer sales to local monopoly stores, and of opening new wine and spirit monopoly stores, found little or no effect on the total alcohol consumption. In terms of total consumption, the effects of opening or closing monopoly stores on Saturdays have been relatively modest—the 3.4% increase from Saturday opening found in the most recent Swedish study [3,4] is at the upper level of the range of results. Differential segments


impacts on different population

The concentration in Nordic policy discussions on the ‘total consumption model’ [12,13] has often meant that the primary attention, in evaluating the effects of changes in controls, has been on the effect on total alcohol consumption. Often, this was taken as the crucial proxy measure for changes in alcohol-related harm due to the change in alcohol controls. However, it is clear that total alcohol consumption is not always a good proxy measure for the effects of the policy change. In the first place, the effect on the drinking of different demographic segments is not always the same. In the second place, there are often variations in the effects on people with different patterns of drinking. In the third place, the effect on

different alcohol-related problems often differs, and may differ from the effect on the total alcohol consumption. We shall discuss each of these points in turn. Variations by demographic segment A clear example of such variation is the effect of taking medium beer out of Swedish grocery stores in 1977. This was the most significant of the restrictive changes in Swedish alcohol policy over a period of about 5 years (1976 – 81), including prohibition of alcohol advertisements in 1978, Saturday closing in 1981, and increases in real prices of alcohol during 1979 – 84. The policy measures were not age-specific, but a major aim of the changes had been to combat youthful heavy drinking. Medium beer had been most popular among young people and its removal, in particular, had some success in reducing young people’s drinking. However, Ramstedt’s analysis [14] suggests that there were also some effects in reducing alcohol-related harm in some older segments of the population. Conversely, the introduction in Denmark of a 15year-old minimum age of off-premise purchase is, clearly, an age-specific measure, but it emerged that there were also declines in drinking among older teenagers. Møller [15] interprets this broader effect as reflecting the public discussion which surrounded the adoption of the minimum age legislation. His findings remind us that the effects of policy changes should not be interpreted in mechanical terms; public sentiment and reaction to the measures may also play an important role in their effects, and a new law may have an effect as much through the public attention, debate and discussion which surround it as through its direct action [16]. There are clear findings in a number of the analyses in this volume that effects of alcohol control changes on women have often been different from their effects on men. The advent of legal beer in Iceland seems to have


had a gender-specific result, as judged from the survey data [17]: men’s consumption of alcohol rose (particularly young men’s), but not women’s. There was a non-significant rise in teenage girls’ consumption, but no net effect for women as a group. The Finnish policy change of 1968, on the other hand, had substantial effects on both women’s and men’s drinking [18,19]. In absolute terms the increase among men was much greater than among women, but the details of the changes among women differed from those among men. In proportion to their drinking patterns prior to the change, men’s median frequency of drinking increased more than women’s, while women’s median annual amount of consumption increased more than men’s. Both the absolute and the relative size of the changes in frequency and amount of consumption varied by age, and were much smaller among older than among younger respondents. There was also a difference in the size of changes, in absolute terms, for both frequency and volume of consumption by level of education. Median frequency increased by 21 units among those in the higher education group, but by only three in those with low education, and there were similar differences in the change in volume. These changes were proportional to the initial frequency and volume levels of the educational groups. There are variations between studies in the effects of policy changes in the city and in the countryside. Among Icelandic men [17], the advent of beer seems to have had a more lasting effect on consumption in the city (Reykjavik) than in the countryside, perhaps reflecting a more severe effect of economic recession in the countryside. On the other hand, in proportion to prior consumption, the median annual consumption rose more in the Finnish countryside than in urban areas, in the wake of the 1968 changes [18], reflecting the much greater difference in availability which the changes made in the countryside. Again, the much greater pre-existing level of consumption in urban areas meant that the absolute amount of increase was much greater in urban areas. Further research is needed to understand better the societal and individual factors behind differences between socio-demographic categories in the effects of policy changes. Variations by drinking pattern and amount Clearly, when policies have a substantial impact on drinking in the population, they tend to have the strongest effect on heavier drinkers, if the effect size is measured in terms of the absolute level of change. In Ma¨kela¨’s analyses [18,20], controlling for the regression-to-the-mean effect, the difference in change between the Finnish and the control samples is greater

49

for the heavier drinkers than for the lighter. Mustonen & Sund’s analysis [19], however, adds the nuance that much of this Finnish increase was actually in relatively low-consumption occasions. Since the data on what happened with the introduction of beer in Iceland are not panel data, we do not have a direct measurement of whose drinking changed and how much. However, it is suggestive that among Icelandic men [17], the consumption level of the highest-consuming 10% of the population had increased between 1988 and 1992 by almost eight times the increase among moderate consumers. Many other phenomena in nature have a distribution which, like the distribution of alcohol consumption among drinkers in a population [21], take a roughly lognormal shape. General discussions of such distributions point out that these distributions are often the net effect of processes where the amount of change on the variable from a given stimulus is roughly proportional to the previous position on the variable [22]. In general, it seems that this is the case for alcohol policy changes: in rough terms, drinkers tend to be affected by policy changes about proportionately to their existing drinking level. Thus, if lighter drinkers increase their drinking by one-half, heavier drinkers also increase theirs by about one-half. From the perspective of the added harm to be expected from added drinking, what happens to heavier drinkers will usually be more crucial. For chronic consequences where the risk curve rises more steeply among heavy drinkers, such as liver cirrhosis, increased drinking by light drinkers becomes relatively unimportant. The same absolute amount of increase in drinking among heavier drinkers would be more important, since the increase in risk is greater due to the steeper risk curve. However, the concentration of the effect in heavy drinkers goes beyond this. Because the actual increase tends to be proportional, so that the increase is greater in absolute terms among heavier drinkers, the effect of a policy change on heavier drinkers becomes the dominant effect for such consequences. For problems related to occasions of intoxication, such as injuries from traffic crashes, violence or other causes, the question of the differential effect of the policy changes on different types of drinking occasions becomes important. Drinking heavily, at least occasionally, is widespread in the population, producing what has often been described as the ‘prevention paradox’: these types of problems, in particular, are also widespread among drinkers [23]. For problems such as traffic crashes, the effects of the 1968 Finnish policy change was presumably muted by the fact that there was a bigger increase in low-consumption occasions than in high-consumption occasions. However, Mustonen & Sund’s analysis [19] shows that there

50

Robin Room

was also, nevertheless, an increase in high-consumption occasions. Indirect evidence that policy changes often have their greatest impact on heavier drinkers can be derived from analyses of the effects of changes on health and social problem indicators. The opening of alcohol monopoly stores in previously dry areas of Finland in the 1950s resulted in a larger increase in drinking among those already drinking frequently than among infrequent drinkers [7]. The 46% increase in consumption in Finland between 1968 and 1969 was accompanied by a 58% increase in deaths from alcohol-specific causes [1]. Rationing in Sweden seems to have had a particularly strong effect in holding down drinking by the most vulnerable heavy drinkers. Thus the end of rationing, which brought a 25% increase in per-capita consumption, brought a 438% increase in deaths from delirium tremens [24]. Variations by type of problem As the examples given above illustrate, the effect of policy changes on rates of alcohol-related health and social harms is often greater than the effect on the total alcohol consumption level. In fact, in a number of instances in the Nordic material, policies appear to have had an impact on the kinds of problems associated with troublesome or social marginal drinkers, even when there was no measurable effect on the overall drinking level. Thus, the analysis of the 1982 Norwegian liquor store strike [25] shows that the strike had a clear effect on admissions to the detox centre in downtown Oslo, and also an effect on reported violent crimes, although the effect on overall consumption was probably modest. These findings are in accordance with findings in several other studies of liquor store strikes in Finland, Sweden and Norway: generally, domestic disturbances and alcohol-related crimes decreased, as did indicators of public drunkenness (arrests or detox admissions), while moderate drinkers were hardly affected by the strike. The results of Nordic experiments with closing or opening liquor stores on Saturdays often had an effect on domestic disturbances or manifestations of public drunkenness, but had little effect on the overall consumption level. On the other hand, introducing round-the-clock serving hours in Reykjavik [26] had a particularly dramatic effect on the numbers of cases of suspected drunk driving, greater proportionally than the effect on numbers of emergency-room admissions for injuries from fights. In this case, the effect was greater for the indicator normally more associated with the settled population, although the possibility that this was due to greater police vigilance cannot be ruled out.

Implications of the Nordic studies for alcohol policy From a policy perspective, it is the effects on drinking problems rates which really matter. From this perspective, the extent to which the drinking of different segments of the population rises and falls in concert remains a question which is intellectually interesting (e.g. 27]) but is not crucial for policy. The crucial issue for some outcomes, such as liver cirrhosis, is the effects of different policy interventions on those who are already or may become heavy drinkers. For other outcomes, such as alcohol-related injuries, the population at risk is much broader, but still considerably less than the whole population of drinkers. Pointing this out has sometimes been taken as an argument for a shift away from general policy instruments that affect all drinkers, and towards more narrowly-targeted instruments [28]. However, the evidence from the Nordic policy impact literature does not support this argument. The changes with the largest effects on alcohol problems rates in the 50 years of Nordic experience we are considering were the Finnish changes of 1968, and probably three Swedish changes: the abolition of the motbok in 1955, the tax increases in the 3 years following that and the policy changes in the era of the repeal of medium-strength beer (1976 – 81). All these policy changes applied to and affected all drinkers, but had their strongest effects on problematic drinkers. Conversely, narrowly focused interventions, such as the adoption of a minimum purchase age in Denmark, often turn out to have effects beyond the population at which they were aimed. Thus, ‘narrow-cast’ policies may have effects on a broader population than their target group and, as the Nordic experience underlines, policies aimed at the total population often turn out to have stronger effects on problematic drinking than on other drinking. The issues of the scope of effect and differential impacts of a policy thus should not simply be assumed. Rather, they are matters for empirical investigation. The future of studies of changes in alcohol policies—Nordic and otherwise The Nordic countries are in a changing alcohol policy environment. The internal free-market provisions of the European Union continue to erode the integrity of the Nordic alcohol control systems. A legal decision in early February, 2003 under EU free-market rules has overturned Sweden’s ban on the advertising of alcoholic beverages in print media [29]. More dramatic changes are under way. The Nordic cross-border allowances within the EU for bringing in alcohol for personal use are to be raised to the full high


51

Table 2. Traveller’s allowances for alcohol imports, in litres of each beverage

Spirits Fortified wine Table wine Beer

Within-EU to Denmark 5 1/2004

Within-EU to Sweden 5 1/2004

Within-EU to Finland 5 1/2004

Estonia to Finland 5 5/2004

General within-EU ‘guidance’: imports for own use

1.5* 20* 90 no limit

5 6 52 64

1 3 5 32

1* 1* 2 16

10 20 90 110

*Either spirits or fortified wine. Source: http://europa.eu.int/abc/travel/shop/index_en.htm

Table 3. Major availability changes in the Nordic countries, October 2003 – May 2004 Date of implementation

Nature and place of change

Mainly affects:

1 October 2003 1 January 2004

Danish spirits tax reduction, 44% Finnish, Swedish, Danish increase in traveller’s allowances Finnish tax reduction (proposed), 44% spirits, 33% overall Finnish increase in traveller’s allowance from Estonia

Denmark, southern Sweden Denmark, Finland, southern Sweden

1 March 2004 1 May 2004

EU levels at the beginning of 2004 (Table 2), and alcohol taxes are being reduced dramatically in Denmark and Finland (Table 3). On 1 October 2003, Denmark’s taxes on spirits were cut by 44%, producing an approximately 25% drop in retail bottle prices. Danish spirits taxes were already considerably lower than those in other Nordic countries, but Denmark was among the high spirits-tax countries in the EU as a whole [30]. The disparity with German taxes had meant for years a flourishing trade in cross-border purchases by Danes in Germany [31], which have grown as the traveller’s allowances for import to Denmark have been brought up by steps to the general EU ‘guidance’ (right-hand column of Table 2). The accession of Estonia to the EU on 1 May 2004 will increase the traveller’s allowance for import between Estonia and Finland also to the general EU level. Low-tax alcohol up to the full EU allowance will thus be brought within an hour of Helsinki. In light of this, the Finnish government has proposed to lower spirits taxes by 44% and alcohol taxes overall by an average of 33% on 1 March 2004. The Danish and Finnish changes have put pressure on the Swedish government to lower taxes also in Sweden. As of October 2003, the official position is that the Swedish government will wait and see the effects of the Danish changes. If Sweden also eventually reduces its taxes, as most experts in Sweden predict [32], this will put pressure on Norway, which has the highest alcohol

Finland Finland

taxes in Europe. There is already substantial Norwegian cross-border purchasing in Sweden. The border traffic threatens not only the relatively high-tax structures in Finland, Sweden and Norway; in the end, it is seen as also threatening the other main pillar of the traditional system of alcohol control in these three countries: the state retail monopoly systems. In Ska˚ne, in southern Sweden, already only about 30% of the alcohol consumed is bought through the state stores. While these further liberalizations are likely to be harmful for public health, they open up opportunities for researchers. A collaborative study involving Danish, Finnish and Swedish researchers is currently under way, with initial surveys conducted in July – September 2003 in Denmark and Sweden and in late 2003 in Finland. A further initial data collection in October – December 2003 in Sweden will allow separation of the effects of the increased traveller’s allowance and the drop in Danish spirits taxes. Northern Sweden— roughly west and north of Stockholm, but not including the Stockholm metropolitan region—will serve as a control for the areas which are primarily affected by the changes: the far south of Sweden (between Gothenburg and Karlskrona), Denmark, and Finland. Respondents from the 2003 surveys in each country will be followed up after 1 year, with new survey samples also collected at the same time. Further follow-ups and new samples are planned for the following years. (If changes in

52

Robin Room

Swedish taxes are announced, the timing of the followup waves may change.) The aim of the study is to focus particularly on a series of questions posed by the current alcohol policy literature and the previous Nordic experience described above: in a situation of quite marked changes in alcohol availability, whose drinking changes, by how much, in what settings, and with what effect on alcohol-related problems? In a recent international collaborative review, the authors offer their collective judgement of the literature’s evidence on the effectiveness of different policy strategies [33]. Looking across the available strategies, those which regulate the availability of alcohol in form, time and place—the alcohol control policies which are studied in the literature we consider here—are among the strongest strategies which are available for limiting the harms from drinking. As the Nordic experience makes clear, this does not mean that each little detail of policy necessarily makes much difference in every circumstance, but there is a consistent pattern of findings that controls of alcohol availability matter in terms of limiting alcohol-related harm, and that they are one of the few strategies which both matter and are potentially politically available in a liberal modern state. Given this, it is an important task to build on the research work we have been considering in future work. In future work, it will be important to give attention to studying differential effects in diverse population subgroups, which holds implications for sample sizes in before-and-after surveys. The message of the literature on the effects of alcohol controls is that total consumption in the population matters considerably, but it is not the only intermediator between alcohol consumption and rates of alcohol problems. In a wider frame, the literature on the effect of changes in alcohol controls offers a good starting point for rethinking public health approaches to limiting the damage from psychoactive substances generally [34]. Neither for nicotine nor for the drugs under international control has there been anything like the length and breadth of experimentation with controlling availability, short of prohibition, which there has been for alcohol. In the context of rethinking control of psychoactive substances more generally, these traditions of studies of the effects of changes in alcohol controls thus have a wider significance.

Acknowledgements This paper is revised from a presentation at an international research symposium, ‘‘Preventing Substance Use, Risky Use and Harm: What Is EvidenceBased Policy?’’, Fremantle, Western Australia 24 – 27 February 2003. The paper draws on the work of my colleagues in a Nordic project and in particular on a

review paper by Ma¨kela¨, Rossow & Tryggvesson. The project was supported by the Swedish Council for Working Life and Social Research and the Nordic Council for Alcohol and Drug Research, with salary support also from SIRUS in Norway and STAKES in Finland.

References [1] Ma¨kela¨ P, Rossow I, Tryggvesson K. Who drinks more and less when policies change? The evidence from 50 years of Nordic studies. In: Room R, ed. The effects of Nordic alcohol policies: what happens to drinking and harm when alcohol controls change? NAD Publication no. 42. Helsinki: Nordic Council for Alcohol and Drug Research, 2002:17 – 70. [2] Room R, ed. The effects of Nordic alcohol policies: what happens to drinking and harm when alcohol controls change? NAD Publication no. 42. Helsinki: Nordic Council for Alcohol and Drug Research, 2002. [3] Norstro¨m T, Skog O-J. Effekter av lo¨rdagso¨ppna Systembolagsbutiker: uppfo¨ljning av de fo¨rst tio ma˚naderna [Effects of Saturday opening of the alcohol monopoly shops: follow-up of the first 10 months]. Stockholm: Social and Health Ministry, 2001. Web address: http://social.regeringen.se/pressinfo/pdf/folkhalsa/rapport_lordagsoppet.pdf [4] Norstro¨m T, Skog O-J. Saturday opening of alcohol retail shops in Sweden: an impact analysis. J Studies Alcohol 2003;64:393 – 401. ´ lafsdo´ttir H, Room R. Nordic traditions of [5] Olsson B, O studying the impact of alcohol policies. In: Room R, ed. The effects of Nordic alcohol policies: what happens to drinking and harm when alcohol controls change? NAD Publication no. 42. Helsinki: Nordic Council for Alcohol and Drug Research, 2002:5 – 16. [6] Bruun K. Interview with Kettil Bruun. In: Edwards G, ed. Addictions: personal influences and scientific movements. New Brunswick: Transaction, 1991:367 – 75. [7] Kuusi P. Alcohol Sales Experiment in Rural Finland. Helsinki: Finnish Foundation for Alcohol Studies, vol. 3A, 1957. [8] Bruun K, Saä¨ski J. Effekten av la¨ttnaden i ko¨parkontrollen i Tamerfors [The effects of lightening buyer surveillance in Tampere]. Alkoholpolitik 1955;18:87 – 92. [9] Room R, Romelsjo¨ A, Ma¨kela¨ P. Impacts of alcohol policy—the Nordic experience. In: Room R, ed. The effects of Nordic alcohol policies: what happens to drinking and harm when alcohol controls change? NAD Publication no. 42. Helsinki: Nordic Council for Alcohol and Drug Research, 2002:167 – 74. [10] Ma¨kela¨ K. Alkoholipoliittisen mielipideilmaston vaihtelut Suomessa 1960- ja 70-luvulla [Fluctuations in the climate of opinion concerning alcohol control measures in Finland in the 1960s and 1970s]. Report No. 98. Helsinki: Social Institute of Alcohol Studies, 1978. [11] Sulkunen P. Abstainers in Finland 1946 – 1976: a study in social and cultural transition. Report No. 133. Helsinki: Social Research Institute of Alcohol Studies, 1979. [12] Sulkunen P, Sutton C, Tigerstedt C, Warpenius K, eds. Broken spirits: power and ideas in Nordic alcohol control. NAD Publication no. 39. Helsinki: Nordic Council for Alcohol and Drug Research, 2000.


[13] Sutton C. Swedish alcohol discourse: constructions of a social problem. Uppsala: Uppsala University Library, Studia Sociologica Upsaliensia 45, 1998. [14] Ramstedt M. The repeal of medium strength beer in grocery stores in Sweden: the impact on alcohol-related hospitalizations in different age groups. In: Room R, ed. The effects of Nordic alcohol policies: what happens to drinking and harm when alcohol controls change? NAD Publication no. 42. Helsinki: Nordic Council for Alcohol and Drug Research, 2002:117 – 31. [15] Møller L. Legal restrictions resulted in a reduction of alcohol consumption among young people in Denmark. In: Room R, ed. The effects of Nordic alcohol policies: what happens to drinking and harm when alcohol controls change? NAD Publication no. 42. Helsinki: Nordic Council for Alcohol and Drug Research, 2002:155 – 66. [16] Hingson RW, Howland J, Levenson S. Effects of legislative reform to reduce drunken driving and alcohol-related traffic fatalities. Public Health Rep 1988;103:659 – 67. ´ lafsdo´ttir H, Leifman H. Legalizing beer in Iceland: its [17] O effects on alcohol consumption in times of recession. In: Room R, ed. The effects of Nordic alcohol policies: what happens to drinking and harm when alcohol controls change? NAD Publication no. 42. Helsinki: Nordic Council for Alcohol and Drug Research, 2002:95 – 106. [18] Ma¨kela¨ P. Who started to drink more? A reanalysis of the change resulting from a new alcohol law in Finland in 1969. In: Room R, ed. The effects of Nordic alcohol policies: what happens to drinking and harm when alcohol controls change? NAD Publication no. 42. Helsinki: Nordic Council for Alcohol and Drug Research, 2002:71 – 82. [19] Mustonen H, Sund R. Changes in the characteristics of drinking occasions resulting from liberalization of alcohol availability: a reanalysis of the 1968 and 1969 Finnish panel survey data. In: Room R, ed. The effects of Nordic alcohol policies: what happens to drinking and harm when alcohol controls change? NAD Publication no. 42. Helsinki: Nordic Council for Alcohol and Drug Research, 2002:83 – 94. [20] Ma¨kela¨ P. Whose drinking does the liberalization of alcohol policy increase? Change in alcohol consumption by the initial level in the Finnish panel surveys in 1968 and 1969. Addiction 2002;97:701 – 6. [21] Skog O-J. Drinking and the distribution of alcohol consumption. In: Pittman DJ, White HR, eds. Society, Culture, and drinking patterns reexamined: alcohol, culture, and social control. New Brunswick, NJ: Rutgers Center for Alcohol Studies, 1991:135 – 56.

53

[22] Aitchison J, Brown JAC. The lognormal distribution: with special reference to its uses in economics. Cambridge: Cambridge University Press, 1957, reprinted 1966. [23] Skog O-J. The prevention paradox revisited. Addiction 1999;94:751 – 7. [24] Norstro¨m T. The abolition of the Swedish alcohol rationing system: effects on consumption, distribution and cirrhosis mortality. Stockholm: Swedish Institute for Social Research, 1987. [25] Rossow I. The strike hits: the 1982 wine and liquor monopoly strike in Norway and its impact on various harm indicators. In: Room R, ed. The Effects of Nordic alcohol policies: what happens to drinking and harm when alcohol controls change? NAD Publication No. 42. Helsinki: Nordic Council for Alcohol and Drug Research, 2002:133 – 44. ´ , Davı´ôsdo´ttir S. Effects of [26] Ragnarsdo´ttir Þ, Kjartansdo´ttir A extended serving-hours in Reykjavik. In: Room R, ed. The Effects of Nordic alcohol policies: what happens to drinking and harm when alcohol controls change? NAD Publication No. 42. Helsinki: Nordic Council for Alcohol and Drug Research, 2002:145 – 54. [27] Gmel G, Rehm J. The empirical testability of Skog’s theory of collective drinking behaviour. Drug Alcohol Rev 2000;19:391 – 9. [28] Stockwell T, Single E, Hawks D, Rehm J. Sharpening the focus of alcohol policy from aggregate consumption to harm and risk reduction. Addict Res 1997;5:1 – 9. [29] Ritter K. Swedish court kills ban on alcohol ads. Seattle Post-Intelligencer, 5 February, 2003. http://seattlepi.nwsource.com/business/apbiz_story.asp?category = 1310&slug = Sweden%20Alcohol%20Ads ¨ sterberg E. A scale of formal alcohol control [30] Karlsson T, O policy in 16 European countries. Nordic Studies Alcohol Drugs 2001;18:117 – 31 [English supplement]. [31] Bygvra˚ S. Border shopping between Denmark and West Germany. Contemp Drug Problems 1990;17:595 – 611. [32] Strandberg L. Systembolaget fo¨rva¨ntas inte overleva [The state alcohol retail monopoly is not expected to survive]. Dagens Nyheter, 28 September 2003, p. 6. [33] Babor T, Caetano R, Casswell S et al. Alcohol—no ordinary commodity: research and public policy. Oxford: Oxford University Press, 2003. [34] Room R. Rethinking alcohol, tobacco and other drug control [with four commentaries and a response]. Addiction 2003;98:713 – 23.