CESifo Area Conference
on
PUBLIC SECTOR ECONOMICS
21 – 23 April 2006
CESifo Conference Centre, Munich
Alcohol Taxation in the European Union
Sijbren Cnossen
CESifo
Poschingerstr. 5, 81679 Munich, Germany
Phone: +49 (0) 89 9224-1410 - Fax: +49 (0) 89 9224-1409
[email protected]
www.cesifo.de
Draft - not to be cited
Comments are welcome
Alcohol Taxation in the European Union
Sijbren Cnossen
Maastricht University and CPB Netherlands Bureau for Economic Policy Analysis
1
Introduction
Alcohol truly permeates every aspect of European culture – it is used before,
during and after meals, to celebrate birth and mourn death, to socialize, as a
relaxant, a deliriant and simply as a means of getting drunk.1 Europeans
(inclusive of infants and abstainers – about 30% of the population) on average
drink 11 liters of pure alcohol per year – twice as much as people elsewhere in
the world (Rehm et al., 2001), and 4-5 times in excess of the level of 2 liters
suggested by the evidence as the lowest population mortality risk (Edwards et
al., 1994). To give the figure some everyday flavor, 11 liters of pure alcohol is
equivalent to 440 bottles of beer (50cl, 5%), 122 bottles of wine (75cl, 12%), or
39 bottles of spirits (70cl, 40%). Some torrent!
Chronic heavy drinking has harmful effects on health (organ damage, birth
defects), causes accidents and domestic violence, and leads to unwanted sexual
encounters. Most crime is alcohol-related. The World Health Organization
(WHO, 2004) has estimated that excessive consumption of alcoholic beverages
is responsible for approximately 9% of the total disease burden in the European
Union (EU), accounting for 10-11% of all illnesses and deaths each year.
Between 40% and 60% of all deaths from intentional and unintentional injury
and poisoning are attributable to alcohol consumption. Overall, the gross costs
of excessive alcohol consumption in the form of lost earnings, material damage,
medical, social and judicial services have been put at between 1% and 3% of
GDP (Klingemann, 2001).
These facts are sufficient reason to look more closely at the economic aspects
of alcohol misuse and taxation. We begin, in section 2, with a review of current
alcohol excise duty structures in the various Member States in light of the EUwide agreed minimum duties, called acquis communautaire. Section 3 proceeds
to highlight the patterns, prevalence and effects of alcohol misuse. In
economics, the alcohol excise is rationalized as a proxy for the costs that
excessive consumers inflict on others. Accordingly, section 4 analyzes the
social or external costs of alcohol misuse, which should form the basis for
1
Throughout this paper, the term ‘alcohol’ covers all forms of alcoholic beverages, including beer, wine and
spirits. There is no common definition in the European Union (EU) of ‘alcoholic beverage’ in terms of units of
alcohol or in % alcohol by volume (abv). Legal definitions vary from 0.1% abv in Italy to 2.8% abv in Finland
(WHO, 2004); the EU average is 1.2% abv. The equivalence of different alcoholic beverages is measured in
term of units of alcohol. One unit is equal to approximately 10gr of pure alcohol, often considered as one
drink, since it is available from 30ml of spirits, one glass of wine (120ml) or one bottle of beer (285ml).
1
calculating the optimal tax rate. Section 5 follows this up by taking a closer
look at various design and enforcement aspects of alcohol excise duties.
Furthermore, alcohol regulation has a ‘duty-equivalent’ effect, so section 6
examines the complementary role of regulations in constraining excessive
alcohol consumption. Section 7 concludes with a summary of the analysis’s
implications for alcohol tax policies in the EU.
2
Excise Duty Structures
Excise on alcohol differ widely between EU Member States and harmonization
measures have mostly been confined to product definitions and agreement on
minimum duties.
2.1
Acquis communautaire
The acquis communautaire for the principal alcoholic beverages, lastly agreed
in 1992 (Council Directive 92/84/EEC of 19 October 1992), is the following.
• The minimum excise duty on beer is €0.748 per hl/degree Plato of finished
product or €1.87 per hl/degree of finished product. Accordingly, half a cent
has to be paid in respect of half a liter of beer with an alcohol content of 5%
by volume (equivalent to 12.5 Plato) or more. Reduced rates apply to
‘independent small breweries’ and to ‘low alcohol’ beer (not exceeding
2.8%).
• The harmonized excise duty on still and sparkling wine (12%) is €0 per
hectoliter of product and so is the duty on fermented beverages other than
wine and beer.
• On intermediate products, such as fortified wines and liqueur wines (under
22% abv and not belonging to the groups of wines or beer), the duty is €45
per hectoliter (a trivial 34 cents per bottle of 75 cl); reduced rates apply to
alcoholic beverages not exceeding 15% by volume.
• The minimum excise duty on spirits (ethyl alcohol) is €550 or €1,000 per
hectoliter of pure alcohol.2 In other words, the price of a 70cl. bottle of
spirits with an alcohol content of 40% includes €1.54/2.80 excise duty.
Lower rates apply to ‘small distilleries’, as well as in certain regions of
Greece, Italy and Portugal.
• All alcoholic beverages are subject to the standard VAT rate which can not
be less than 15%.
• Intra-EU duty paid cross-border shopping allowances are 110 liters for beer,
90 liters for wine (but only 60 liters for sparkling wine), 10 liters for spirits,
and 20 liters for intermediate products.
• The Council of Ministers has placed restrictions on the advertising of
alcohol on television (Directive 89/552/EEC of 3 October 1989) and
encouraged Member States to address the problems caused by alcohol abuse
among young people (Recommendation 2001/458/EC of 5 June 2001).
2
2
The higher rate applies to Member States whose duty rate exceeded €1,000 in 1992.
2.2
Duty structures in Member States
The acquis communautaire leaves considerable scope for significant tax
differentials between Member States for all types of alcoholic beverages . As
shown in table 1, beer, wine and spirits are particularly heavily taxed in
Sweden, Ireland, the UK and Finland, but very lightly in Luxembourg, Austria
and the Mediterranean states, as well as most new Member States. All alcohol
excise duties are specific, although the precise specification of the tax base
differs between beer, wine and spirits. In most Member States, the excise on
beer is proportional to alcohol content, but three states graduate the duty, with
higher rates of duty per unit of alcohol on stronger beer. The duty on spirits is
defined uniformly throughout the EU in terms of an absolute amount per
hectoliter of pure alcohol. By contrast, wine is taxed per unit volume, although
in some Member States the excise is banded according to the alcohol strength
of the product.
[Table 1 about here]
Specifically, the excise duty on beer ranges from merely 4 cents per half liter in
Germany, Luxembourg, Malta and the Czech Republic to almost one Euro in
Ireland, Finland and the UK. Half of all Member States do not levy an excise on
still wine, but in Ireland, Sweden and the UK, the excise duty is close to €2.00
per bottle of 75cl. Generally, sparkling wine is taxed at higher rates (not shown
in the table) than still wine. The excise duties on spirits differ most widely in
absolute amounts, ranging from €1.71 in Cyprus to €15.08 in Sweden per bottle
of 70cl.
In general, little if any excise duty harmonization can be discerned. The agreed
minimum rates have not been adjusted since 1992, implying a reduction in their
real value of about 30%. The European Commission (2004) lastly reviewed
alcohol excise duty policies in 2004. It focused on the proper functioning of the
internal market in view of the wide divergencies in duty levels, the competition
between different categories of alcoholic beverages, the real value of the duty
rates, and paid lip service to the wider objectives of the Treaty, i.e. underlying
health, social and agricultural policies. No action was taken.
Whether alcohol is taxed too lowly or too highly would appear to depend on the
level and variation in the social costs of alcohol misuse across Member States.
And whether a case can be made for further harmonization, particularly
between neighboring Member States, depends on how seriously policymakers
regard the fiscally induced wasteful forms of bootlegging throughout the EU,
i.e. the purchase of alcoholic beverages in excess of shopping allowances in
low-duty states for consumption in high-duty states. Beyond that, the large
differences in pre-tax prices (see the right hand side of table 1) – perhaps
mainly attributable to differences in prices of representative brands chosen for
the survey – suggest that alcohol product competition still has some way to go
in the EU.
3
3
Patterns, Prevalence and Effects of Alcohol Misuse
3.1
Overview of drinking patterns
As table 2 indicates, in 2003 every European (inclusive of infants and
teetotalers) on the average drank 11 liters of pure alcohol. On average, per
capita consumption increased by 30 percent in the period 1965-2000, but since
then consumption has either decreased or stabilized in most old Member States,
attributable perhaps to ageing populations. The decline of alcohol consumption
in Austria, France and the Netherlands is consistent with a decrease in alcoholrelated mortality indicators. By contrast, per capita consumption and alcoholrelated mortality rates are increasing in the Baltic countries (Rehm et al., 2001).
[Table 2 about here]
The common denominator – pure alcohol – hides a wide diversity in the
patterns and prevalence of drinking. The Hungarians, for instance, drink 15.4
liters per capita, more than three times as much as the Maltese. In half of all
Member States the level of per capita consumption exceeds 11 liters per annum,
which is, perhaps, close to some natural saturation level. Predominantly, ‘wet’
Member States with low rates of abstainers (varying from 5% in Denmark to
18% in Sweden) are found in northern Europe, while most ‘dry’ Member States
are situated in the south (27% abstainers in Italy and 37% in Spain) (Rehm at
al., 2001).
There are also wide variations in the type of beverage – beer, wine, spirits – that
are preferred in various Member States. The Czechs, for example, excel in beer
drinking downing 2750 bottles per capita every year, whereas the French win
the wine contest with 660 bottles, and the Latvians the hard liquor drive with 37
bottles. Typically, beer is the beverage of choice in northern Europe, wine in
the Mediterranean states, while spirits consumption is heavily concentrated in
central and eastern European countries. Alcohol production and consumption
largely are local phenomena – only 10% of all wine, 2% of all beer (water is
expensive to move around) and 20% of all spirits produced domestically are
exported.
3.2
Drinking problem groups
Alcohol has its connoisseurs as well as its abusers.3 While 1 or 2 glasses per
day seem to be good for most adults,4 a large number of people drink
3
Researchers have assumed misuse to coincide with binging (8/6 drinks per occasion for men/women),
heavy drinking (50/36 plus units a week for men/women), problematic drinking (a score of one or more on
the CAGE questionnaire) or dependent drinking (based either on the Audit or SAD-Q questionnaire).
According to the CAGE (Cut Annoyed, Guilty, Eye-opener) test, an individual is considered a problem drinker
if her or she answers positive to one or more the following questions: (a) have you every felt you ought to cut
down on your drinking? (b) have people ever annoyed you by criticising your drinking? (c) have you ever felt
bad or guilty about your drinking? (d) have you ever had a drink first thing in the morning to steady your
4
considerably more. Surveys on drinking habits focus on particular problem
groups so that policy measures can be tailored accordingly. Generally, these
groups consist of young drinkers, heavy drinkers and alcohol dependent
drinkers. Table 2 shows their drinking prevalence.
The foundations for alcohol misuse are laid in childhood and adolescence. In 16
Member States, the proportion of 15-year olds who report drinking beer, wine
or spirits at least weekly is 20 percent or higher. In the UK, the Netherlands,
Malta and Denmark, approximately half of 15-year olds drink weekly. But the
habit starts at a much younger age. In most Member States, more than 50
percent of 11-year-old children have already tried alcohol at least once
(Settertobulte et al., 2001) and 12 percent of this age-group reports having been
drunk twice or more.
As Cook and Moore (2002) point out, teenagers and young adults are of special
concern for at least three reasons. First, youths exhibit relatively high rates
(compared with elders) of binge drinking and involvement in violent crime and
motor vehicle accidents. Second, to the extent that drinking is habit forming,
youthful drinking sets the patterns for future consumption. Third, drinking
behavior during the transition from adolescence to adulthood may have
consequences for human capital and family formation.
Heavy or hazardous drinking is generally defined as daily drinkers who drink
over 40gr of pure alcohol on average for men and over 20gr for women (Babor
et al., 2003).5 By this measure, the proportion of heavy and hazardous drinkers
(among drinkers) exceeds 10 percent in half of the old Member States (statistics
for the new Member States seem less reliable).6 Heavy drinkers, i.e. those in the
top decile of the drinking distribution, consume more than half of all alcohol
sold (Cook and More, 2002) and are responsible for most of the external costs
associated with alcohol misuse. Alcohol dependence among the general
population, measured as having four positive CAGE answers, is 5 percent or
higher in France, Belgium, the Netherlands and Poland.
nerves or get rid of a hangover (‘eye opener’)? Similar questions are raised under the Alcohol Use Disorders
Identification Text (Audit) and the Severity of Alcohol Dependence Questionnaire (SAD-Q). See UK Cabinet
Office (2003) for further definitions.
4
It is a well established finding in the medical literature that drinking reduces the likelihood of coronary heart
disease, and moderate drinking in middle age extends life expectancy overall (NIAAA, 2000). But for a view
that one drink a day won’t keep the doctor away, particularly in European drinking cultures, see Lieber
(2004).
5
In Sweden the ratio if 30/20, while Denmark and Ireland define heavy drinking in terms of 21/14 units per
week. Belgium and the Slovak Republic simply count the number of drinks: 6 and 5/3, respectively. In winedrinking Italy, heavy drinking is defined as one-half liter of wine per day.
6
According to Rehm at al. (2001), the figures on heavy drinking that stand out most clearly from the WHO
statistics are the number of women (21%) in Ireland drinking more than 140gr of alcohol per week, the
number of men (29%) in Austria exceeding 420gr per week, and the number of men in the Czech Republic
(16%) exceeding 550gr per week. As noted in footnote 1, 10gr of alcohol stands for one drink.
5
3.3
Effects of alcohol misuse
The (predictable) consequences of alcohol misuse show up in statistics on
diseases, alcohol-related injuries and poisoning in and outside the home, and
road traffic accidents. A traditional indicator of the prevalence of chronic heavy
drinking in a population is the mortality rate from chronic liver disease and
cirrhosis. As table 2 indicates, Austria, Germany, the Baltic countries, the
Slovak Republic and particularly Hungary have high incidences of liver
cirrhosis. While (historically high) liver cirrhosis rates have been declining in
southern Member States, a steep rise has been recorded in the UK (as well as
Denmark, Finland and Ireland) where per capita alcohol consumption has more
than doubled over the past 40 years (Leon and MCambridge, 2006).
Although no direct causal relationship has been established between alcohol
consumption and ensuing injury and poisoning, in the majority of cases alcohol
is strongly implicated. Standardized death rates for external causes of alcoholrelated injury and poisoning vary from a low of 27.3 people per 100,000
population in the UK to a high of 147.8 per 100,000 in Lithuania. High rates are
also found in the other Baltic states, Hungary, Slovenia and Poland (WHO,
2006).
In many Member States, 20 percent or more of all road traffic accidents per
100.000 population can be attributed to alcohol misuse. The mortality rates vary
from a low of 5.1 per 100,000 population in Italy to a high of 41.3 per 100,000
in Belgium. Although the drink drive limit in most Member States is relatively
low, i.e. 50mg% (zero in the Czech Republic, Hungary and Slovakia), since
2000 the number of casualties (of which 40% are innocent victims) has been
rising again. Nine out of ten alcohol-related road fatalities are caused by young
male drivers.
The effects of alcohol misuse involve high social or external costs to which we
now turn.
4
External Costs of Alcohol Misuse
In calculating the costs of alcohol misuse, economists commonly make a
distinction between private or internal costs (borne by the drinker and
presumably taken into account in the drinking decision) and social or external
costs (inflicted by drinkers on third parties). If one accepts the consumersovereignty principle, then only external costs are relevant for policy purposes.
External costs should be computed on a net basis. Accordingly, the value of
‘social lubrication’ or ‘networking’, which may be promoted through alcohol
use, should be deducted from gross external costs. Likewise, the value of the
pleasure of having one or two drinks per day should be deducted from gross
private costs.
6
4.1
Overview
Various researchers, particularly in the US, the UK and Australia (where
historically alcohol misuse is associated with ‘sin’), have attempted to estimate
the external costs of alcohol misuse. An early pioneering US study by Manning
et al. (1989) estimated the net social costs in the mid-1980s at about US$0.48
per ounce of ethanol, double the average state and federal tax per ounce that
was then in place.7 Much of the social costs were borne by victims of drunk
drivers. The costs would have been higher still had nonfatal highway injuries
(Miller and Blincoe, 1994) and intrafamily effects been taken into account. The
latter effects are sometimes left out of consideration, because it is assumed that
the welfare of the family enters the utility function of the alcohol consumer. In
the case of alcohol abuse, this seems implausible, however, and the costs of
domestic violence and injuries, particularly if inflicted on young and unborn
children, would seem best treated as externalities (Smith, 2005).8
In recent years, various EU Member States, at the instigation of the World
Health Organization, have published comprehensive estimates of the costs of
alcohol-attributable output losses, criminal offences and hospitalizations. The
costs differ widely, ranging from 0.6 percent of GDP in Portugal (Lima and
Esquedro, 2003) to 1 percent in Germany (Bühringer et al, 2002),9 1.4 percent
in France (Fenoglio, et al, 2003),10 1.9 percent in Ireland (Strategic Task Force,
7
A similar result has been reported by Pogue and Sgontz (1989). On the other hand, Kenkel (1996)
estimates that the duty rate should be about equal to the pre-tax alcohol price, while Saffer and Chalooupka
(1994) calculate the weighted average optimal US tax on alcohol at 2.3 times the 1991 level. Incidentally, the
variation in excises across US States has created a veritable cottage industry of research devoted to relating
differences in taxes on alcoholic beverages to a wide variety of changes in social conditions. The following
sample is provided: (a) a 1% increase in the price of alcohol decreases the rate of wife abuse by 3.1-3.5%
but has no effect on abuse of husbands (Markowitz, 1999), (b) a 10% increase in the excise tax on beer
would reduce severe domestic violence against children by 2.3% (Markowitz and Grossman, 1998), (c) a
78% increase in the beer tax (restoring the real rate to its 1975 level) would reduce highway fatalities by 78% (Ruhm, 1996), (c) raising the beer tax to the alcohol equivalent tax on distilled spirits would reduce the
drinking of underage drinkers who drink frequently by 32% (Grossman et al, 1994), (d) a 10% increase in
the price of beer reduces the number of high school students who engage in binge drinking by between 25% (Saffer and Dave (2003), (e) college students faced with a $1 increase in the average real price of a
drink would have been 33% less likely to make the transition from being a moderate drinker to a binge
drinker (2002), (f) a 10 per cent increase in the price of alcohol would decrease drunk driving by 7.4 per cent
for men and 8.1 per cent for women (Kenkel, 1993), (g) a 10 per cent increase in the beer tax would reduce
rapes by 1.32 per cent, robbery by 0.9 per cent, murder by 0.3 percent, and assaults by 0.3 per cent (Cook
and Moore, 1993), and (h) a US$0.20 per six-pack increase in the beer tax could reduce the overall
gonorrhea rate by 8.9 per cent (Harrison and Kassler, 2000).
8
Generally, much of the external costs of alcohol misuse are borne by victims of intoxicated consumers. By
way of contrast, most smoking-related costs, including morbidity and early death, are borne by the smokers
themselves.
9
Bühringer et al. (2000) conclude that each year approximately 42,000 Germans die from alcohol-related
causes, that some 238,000 criminal acts are committed under the influence of alcohol and that the costs of
alcohol-related diseases is around €20 billion.
7
2003),11 and 2-5 percent in Italy (WHO, 2004). From an economic point of
view, the problem with most of these studies is that they use a cost-of-illness
(COI) framework, which draws an insufficiently clear distinction between
private and external costs. The value of the loss of output due to reduced
employment often is the most important cost item, but no indication is provided
whether and to what extent this is reflected in lower wages and hence should be
considered part of private costs.12
4.2
Recent UK Cabinet Office Study
Nevertheless, it seems useful to look more closely at a sophisticated recent UK
COI study (UK Cabinet Office, 2003) which provides estimates of prevalencebased alcohol misuse in England and Wales in 2001 over a range of alcoholdependent drinkers (7.9 percent of the English population aged 16 and over)
and heavy drinkers (5.4 percent of the population).13 Table 3 shows the cost
estimates under three broad headings: alcohol-related crime (including drink
driving offences), loss of productive output, and excess use of health services.
[Table 3 about here]
By far the largest proportion of external costs – almost two-thirds – is
accounted for by the valuation of alcohol-related crime. Following Brand and
Price (2000), the study distinguishes three categories of alcohol-related crime
costs: (a) costs incurred in anticipation of crime (excluding insurance premiums
which are not an opportunity cost, although the cost of insurance administration
is), (b) costs incurred as a consequence of crime (cost of property damaged or
stolen, victim support and emotional impact costs and loss of productive output
10
Fenoglio et al. (2003) report that the social cost of alcohol misuse in France was as large as the combined
social cost of tobacco and illicit drugs. The greatest share of the social cost of alcohol came from the loss of
productivity and premature death.
11
In Ireland, total external costs are estimated at €2.9 billion per year. 43% of these costs are attributed to
loss of output due to alcohol-related work absences, 17% to alcohol-related transfer payments, 13% to the
costs of road accidents, 12% to healthcare costs, 10% to taxes not received on lost output, and 4% to
alcohol-related crime.
12
Neither are tax revenue externalities taken into account, as Smith (2005) points out. Afterall the loss in
taxed wages is partly shared by society and hence is not a matter of social indifference.
13
The study follows the international guidelines for estimating the costs of substance abuse (Single et al,
2001) and utilizes the “Cost-of-Illness” methodology (COI) to assess the overall negative impact of alcohol
misuse on society, relative to a counterfactual scenario in which there is no alcohol misuse. It notes that
social surveys consistently under-record consumption of alcohol, because individual respondents
consciously or unconciously underestimate how much alcohol they consume (sometimes by as much as
40% to 60%, IAS (2003)) and respondents reside primarily within private households and hence, students,
publicans or homeless people are excluded. A similar study was done for Scotland (Scottish Executive,
2001), but the results are not directly comparable and therefore have not been added. Incidentally, the UK
government’s sensible drinking message is 4/3 units per week for men and women, respectively. By this
measure, 39% of men and 21% of women exceeded the weekly recommended levels in 2000-01.
8
of the direct victim), and (c) costs incurred in response to crime (costs to the
criminal justice system). The costs of drink-driving offences are shown
separately. In 2000, 6 percent of all road accidents and 16 percent of road
deaths in England and Wales were caused when someone was driving over the
legal limit for alcohol.
The costs of alcohol misuse associated with the workplace and the wider
economy, like those of alcohol-related crime, are extremely difficult to
estimate. The UK Cabinet Office (2003) focuses on three types of costs related
to reduced output (a) employee absenteeism (alcohol misuse increases the
average number of days of sickness), (b) unemployment and lower activity rates
through a “discouraged worker” effect (MacDonald and Shields, 2003), and (c)
lost output due to premature death. As table 3 indicates, social costs under this
heading represent close to one-third of total costs.
As Smith (2005) points out, however, quite how far the productivity effect of
employees’ alcohol abuse is reflected in individual wages (no externality), and
how far it is collectively borne, is unclear. Evidence in the US context shows
that the direct effect of drinking on productivity is small – in fact, self-reported
abstainers earn less on average than drinkers14 – although Cook and Moore
(2002) note that there is persuasive evidence that heavy drinking has an indirect
effect on productivity by interfering with schooling and family formation, both
of which affect subsequent productivity and earnings.
The UK Cabinet Office (2003) also estimates the value of lost output due to
premature mortality as a result of alcohol misuse. In 2001, the number of
premature deaths (excluding ischemic stroke) was estimated to number some
18,600. Interestingly, this number is roughly of the same magnitude as the
number of people that live longer on account of the beneficial health effects of
moderate alcohol consumption. The human costs of alcohol-related social and
physical harm are also discussed, but no attempt is made to quantify them.
However, a short but telling survey of the wider human cost of alcohol misuse,
reproduced in table 4, is provided.15 Clearly, alcohol misuse has dire effects on
family life, while children are often the main victims.
[Table 4 about here]
Finally, health care costs of alcohol misuse were estimated through the use of
“attribution fractions” based on previous established research findings. In most
EU Member States health care costs are largely financed through quasi-tax
contributions. Accordingly, healthcare costs are an important part of external
costs, and, other things being equal, the optimal rate of alcohol taxation should
14
MacDonald and Shields (2001) report an inverse U-shaped relationship between drinking intensity and
mean hourly wages. The turning points for the inverted U-shape curve are in the range of 21-36 units per
month for men and 14-28 units for women.
15
For some indications of alcohol-related social and physical harm from across Europe, see Rehm et al.
(2001).
9
be higher in these Member States than in states where healthcare costs are
largely individually borne.
The UK Cabinet Office’s Study (2003) repeatedly points out that the social
costs of alcohol misuse are difficult to conceptualize and quantify, if only
because of the uncertainty regarding the causality between alcohol misuse and
its negative consequences. This is especially true for the emotional impact costs
for direct victims of crime and the value of lost output. Although the two
categories of costs should not be dismissed too lightly, for expository purposes
table 3 makes a distinction between the social costs with and without the
emotional impact costs and the value of lost output. On this basis, the total
social costs of alcohol abuse (including emotional impact costs and lost output)
amount to three-fourth’s of pre-tax expenditure on alcoholic beverages or more
than three times total excise duty collections.16 If emotional impact costs and
lost output are excluded, total external costs are one-third of pre-tax expenditure
or one-and-a-half times excise revenues (excluding the cost of regulations).
Even in high-alcohol excise UK, this seems to leave some room for further duty
increases to pay for social costs and/or other measures to reduce them, e.g. by
tightening regulations on the physical availability of alcoholic beverages.
5
Excise Duty Design and Enforcement Issues
Alcohol excise duty collections are predominantly, although not exclusively,
related to the level of the duties. As shown in table 5, per capita alcohol duty
collections range from about €16 in Italy to over €260 in Finland. Five Member
States collect more than €100 per capita.17 While alcohol excise collections are
significant in northern EU Member States, they are of negligible proportions in
southern states. The same picture emerges from a calculation of the absolute
duty amounts which are paid per liter of recorded consumption of pure alcohol
across Member States. While Fins have to pay €33 per liter of pure alcohol, the
Hungarian excise authorities are content with merely €2 per liter.
[Table 5 about here]
5.1
Pigovian tax rate considerations
Excise duties on beer, wine and spirits are all specific rather than ad valorem
(apart from the VAT). This is in line with the Pigovian (Pigou, 1918) notion
that the damage caused by alcohol misuse is, at any point in time, independent
16
The VAT, also shown in Table 4, should be left out of consideration, because the alcohol excise may be
assumed to reflect the external cost of drinking and hence the imposition of VAT does not affect the price of
alcohol drinks relative to other goods. As Cnossen and Smart (2005) point out, this may not be the case with
cigarettes if and to the extent the excises exceed the external cost of smoking.
17
With the exception of Finland, in all Member States per capita alcohol duty collections are (much) lower
than per capita cigarette duty collections (see Cnossen, 2006).
10
of the price at which alcoholic beverages are sold, so that correction of
externalities favors specific over ad valorem taxation. The average Pigovian
duty can then simply be calculated as the total net external costs of alcohol
misuse divided by the liters of pure alcohol consumed (subsequently,
appropriately divided over beer, wine and spirits).
Unfortunately, the marginal duty is much more difficult to compute, because
the relationship between alcohol consumption and external costs seems far from
linear. The marginal duty would be considerably higher than the average tax
rate if the external damage caused by alcohol consumption varied across units
consumed by each individual. Applying an average tax rate, therefore, means
that the external costs generated by abusive consumers are compensated at the
cost of reducing the consumer satisfaction of non-abusive consumers (Pogue
and Sgontz, 1989). As noted above, however, nearly all external costs are
caused by 10% of the drinking population who consume half of all alcohol sold.
Perhaps, this suggests that differences in external costs arise from differences
between individuals rather than between units of alcohol consumed by a
particular individual. This provides some support, as Diamond (1973) argues,
for regarding the average external costs as a rough-and-ready indicator of the
optimal externality duty.
A further issue is whether alcohol duties should be related to alcohol content.
As is well-known, the medical profession measures the damage caused by
excessive alcohol consumption in cubic centimeters of pure alcohol. A large
volume of weak drink is just as harmful as a smaller volume of strong drink
(Crooks, 1989). On externality grounds, therefore, there appear to be good
reasons to base the excises on alcohol content, but as table 5 shows , on
average, spirits are taxed 2-3 times more heavily than beer per unit of alcohol,
and 3-4 times more heavily than wine.18 Average relative (tax-inclusive) prices
of beer, wine and spirits, however, indicate that one obtains the best value, i.e.
the largest amount of pure alcohol for a given amount of money, when
purchasing spirits, followed by wine and, lastly, beer (WHO, 2004). So, if
consumption damage control is the objective and assuming that beer, wine and
spirits are close substitutes, spirits should be taxed higher than wine which
should be taxed more heavily than beer. There is also the point that taxation on
the basis of alcohol content would be difficult for wine because the (smallscale) production process gives the producer and the excise authorities less
control over alcohol content.
5.2
Price elasticities of demand
Early on, the relatively higher taxation of alcohol has been defended by
reference to the inverse elasticity rule (Ramsey, 1927) which holds that other
things being equal, efficiency losses from taxation are lower for goods whose
18
In absolute amounts, beer is taxed at about €8.4 per liter of pure alcohol across Member States, wine at
about €6.4 per litre, and spirits at €21.6.
11
price-elasticity of demand is lower than for other goods with higher priceelasticities of demand. Smith (2005) reviews various US studies which
capitalize on the variation of tax-price differentials between states to estimate
demand elasticities. One of the most widely cited studies by Leung and Phelps
(1993) concludes that the price elasticities of alcohol demand are -0.3 for beer, 1.0 for wine, and -1.5 for spirits. Furthermore, young people and moderate
drinkers are more price-responsive than older drinkers and heavy drinkers.
Also, the long-run price elasticity of alcohol demand is significantly higher than
the short-run elasticity.19 Smith (2005) concludes that the reported price
elasticities of alcohol demand are not so low that the inverse-elasticity rule
would seem to justify significantly higher-than-average taxation of alcohol.
The relatively high price elasticities cited above imply that if alcohol prices go
up, consumption goes down (not only of the volume, but also of ethanol), and if
prices go down, consumption goes up (Edwards et al., 1994). Indeed, extensive
studies demonstrate that increases in taxes and prices are related to reductions
in alcohol consumption and harm (Babor et al., 2003). Also, an interesting
finding is that people tend to influence each other in their drinking patterns; in
the aggregate their consumption levels tend to rise and fall together. In this
connection, Skog (1996) and Norström (2002) postulate that the level of alcohol
consumption has a J-shaped relationship to general mortality at the individual
level. This implies that the optimum average drinking level for the whole
population is lower than the optimum drinking level for the average individual
of that population. They conclude that a decrease in overall alcohol
consumption will be attended by a decrease in general mortality (also, Room,
2001).
5.3
Bootlegging and smuggling
The positive effect of a tax-induced decrease in consumption in individual
Member States needs to be weighed against a potential increase in bootlegging
(if other Member States do not follow suit), smuggling or illegal production of
alcoholic beverages. Bootlegging and smuggling (the purchase and
consumption of alcohol without payment of duty) are serious problems in the
EU (European Commission, 2004). In the UK, in 1998, bootlegging involved a
revenue loss of 5% of total alcohol duty revenues and smuggling another 4%.
Smith (2005) believes that there is a case for narrowing differences in alcohol
taxes between EU member states by significantly increasing the agreed EUwide floors to alcohol taxes. This would reduce the economic and fiscal costs
associated with bootlegging (and legal cross-border shopping).
19
Grossman, Chaloupka and Sirtalan (1998) have estimated the demand among individuals between the
ages of 17 and 29 – the age group in which the prevalence of alcohol dependence and abuse is highest.
They report significant and numerically large linkages among past, present and future consumption. The
long-run elasticity of alcohol demand of -0.65 is 60% higher than the comparable short-run elasticity.
12
6
Alcohol Regulation
Invariably, excise taxation goes hand in hand with regulation. In the case of
tobacco excises, the regulations are straight-forward – no smoking in public
places, no advertisements – but in the alcoholic drinking field a more multifaceted approach is common which (besides taxing and pricing) includes
regulating the physical availability of alcohol, modifying the drinking context,
drinking-driving countermeasures, the drinking context, treatment, regulating
alcohol promotion, education and persuasion program, and treatment and early
intervention strategies. These (complementary) regulations are aimed at
reducing the externalities associated with alcohol consumption and have a ‘duty
equivalent’ effect. In theory, they should be deducted from the Pigouvian tax
that would be set in the absence of the regulatory policies.
Nearly all Member States have minimum legal purchase ages (MPA) of 16 or
18 years regarding the sale of beer, wine and spirits in bars and shops. Kenkel
(1996) has shown that MPA reduces the highway fatality rate for the affected
age groups by about 7 percent. Half of all Member States also have legal sale
restrictions on the places of sale of alcoholic drinks (but few on the hours and
days of sale) (WHO, 2006). Random breath testing (RBT) and lowered blood
alcohol concentration (BAC) limits (generally, 40-50mg% but 0% in the Czech
Republic, Hungary and Slovakia) are also common. Furthermore, most states
have partial to complete or voluntary restrictions on advertising, sponsorship or
brand identification. School and work-based programs generally are only poorly
or moderately developed.
Table 6, reproduced from Babor et al (2003), shows how effective these and
other strategies or interventions are in reducing alcohol harm, regardless of the
cost of implementing them. Clearly taxation and pricing, as well as regulations
on physical availability and drink-driving countermeasures should be preferred.
By contrast, the effectiveness of designated drivers, voluntary codes of bar
practice and various forms of education and persuasion is nil.
[Table 6 about here]
Regulatory measures carry a cost, however, which differs from one measure to
the next. Accordingly, it is the cost-effectiveness of the measure in terms of the
number of Disability Adjusted Life Years (DALYS) averted by various
strategies each year that is of greater interest. In a sophisticated recent study,
Chisholm, et al (2004) calculate that in countries with high levels of hazardous
consumption (more than 5% of all drinkers, see table 2), such as the UK and
other northern EU Member States, both individual interventions by doctors and
population-wide interventions like alcohol taxes (even after allowing for an
estimated 10%-15% increase in illicit production or smuggling) can have
significant impact on health at the population level.
Beyond that, reduced access to alcohol (in public building, work places, at
sporting events, etc.), random breath testing and lower BAC limits for young or
13
professional drivers are highly cost effective strategies. By contrast, education
and persuasion or designated drivers (BOB, the non-drinking driver, in Belgium
and the Netherlands) seem to have little effect (and may even be harmful by
suggesting that everyone in the car can be stone drunk as long as the driver is
sober). Overall, individual approaches to prevention are shown to have a much
smaller effect on drinking patterns and problems than do population-based
approaches that affect the drinking environment and the availability of alcoholic
beverages. The same applies to the EU Commission’s faint-hearted restrictions
on advertisements and exhortations to address the problem of young drinkers.
7
Implications for Alcohol Tax Policy
This paper has shown that abusive drinking is one of the most important health
and safety issues in the EU. Europeans drink twice as much as people in
elsewhere in the world and on average five times as much as is commensurate
with a healthy lifestyle. Much domestic violence, many accidents and a large
part of crime is alcohol-related. Drinking is a habit that starts at an ever younger
age, although there is a clear public stake in keeping minors away from alcohol
and in preventing alcohol-induced child abuse and neglect, and fetal damage.
The net external costs of alcohol abuse are high throughout the EU, while the
revenues from alcohol excises are insufficient to meet the costs. Nearly all
external costs are caused by merely 10% of the drinking population who
consume half of all alcohol sold. Cost-effective measures that will have a
noticeable impact on alcohol consumption and abuse include higher alcohol
duties, reduced access to alcohol in public buildings, work places and at
sporting events, and drink-driving tests. Apparently, a decrease in aggregate
consumption will also lead to a decrease in abusive consumption. However,
these measures would reduce the welfare of non-abusive consumers.
Agreement on significant increases in minimum duties on alcoholic beverages
at the EU level would reduce wasteful bootlegging induced by large differences
in duties.
References
Babor, T., Caetano, R., Casswell, S., Edwards, G., Giesbrecht, N., Graham, K., Grube, J.,
Grunewald, P., Hill, L., Holder, H., Homel R., Österberg, E., Rehm, E., Room., R., and
Rossow., I. (2003), Alcohol: No Ordinary Commodity – Research and Public Policy,
Oxford University Press, Oxford.
Barber, J.G., and Crisp, B.R. (1994), The effects of alcohol abuse on children and the partner’s
capacity to initiate change, Drug and Alcohol Review 13/4: 409-416.
Brand, S. and Price, R. (2000), The Economic and Social Costs of Crime, Home Office
Research Study, 217.
Bühringer. G., Augustin, R., Bergmann, E., Bloomfield, K., Funk, W., Junge, B., Kraus, L.,
Merfert-Diete, C., Rumpf, H-J., Simon, R., and Toppich, J. (2000), Alcoholkonsum und
Alkoholbezogene Störungen in Deutschland, Nomos Verlagsgesellschft, Baden-Baden.
14
Chaloupka, F.J., Grossman, M., Saffer, H. (1998), The effects of price on the consequences of
alcohol use and abuse, in M. Galanter (ed.), Recent Developments in Alcoholism, Vol 16:
The Consequences of Alcohol, Plenum Press, New York.
Chisholm, D., Rehm, J., van Ommeren, M., and Monteiro, M. (2004), Reducing the global
burden of hazardous alcohol use: A comparative cost-effectiveness analysis, Journal of
Studies in Alcohol 65: 782-93.
Christiansen, V., and Smith, S. (2001), The economics of duty-free shopping, CESifo, Working
Paper 595.
Cnossen, S. (2006), Tobacco taxation in the European Union, FinanzArchiv, forthcoming.
— and Smart, M. (2005), Tobacco taxation, in S. Cnossen (ed.), Theory and Practice of Excise
Taxation: Smoking, Drinking, Gambling, Polluting, Driving, Oxford University Press,
Oxford.
Cook, P. J. and Moore, M. J. (1993), Economic Perspectives on Reducing Alcohol- Related
Violence, in National Institute on Alcohol Abuse and Alcoholism, Alcohol and
Interpersonal Violence: Fostering Multidisciplinary Perspectives, Research Monograph 24,
Bethesda, MD: 193-212.
— and Moore, M.J. (2002), The Economics of Alcohol Abuse and Alcohol-Control Policies,
Health Affairs 21/2: 120-33.
Crawford, I., Smith, Z., and Tanner, S. (1999), Alcohol taxes, tax revenues and the Single
European Market, Fiscal Studies 20: 287–304.
— and Tanner, S. (1995), Bringing it all back home: alcohol taxation and cross-border
shopping, Fiscal Studies 16/2: 94–114.
Crooks, E. (1989), Alcohol Consumption and Taxation, The Institute of Fiscal Studies, IFS
Report Series No. 34, London.
Cutler, D. (2002), 'Health Care and the Public Sector,' NBER Working Paper 8802, National
Bureau of Economic Research, Cambridge, MA., February.
Diamond, P. (1973), Consumption externalities and imperfect corrective pricing, Bell Journal
of Economics 4: 526–38.
Edwards, G., Anderson, P., Babor, T.F., Casswell, S., Ferrence, R., Godfrey, C., Holder, H.D.,
Lemmens, P.H., Mäkelä, K., Norström, Österberg, E., Romelsjö, A., Room., R., Simpura,
J., Skog, O-J, (1994), Alcohol Policy and the Public Good, Oxford University Press,
Oxford.
European Commission (2001), Tax Policy in the European Union: Priorities for the Years
Ahead, Communication from the Commission to the Council, the European Parliament, and
the Economic and Social Committee, COM(2001) 260 final. Available at
http://europa.eu.int/comm/taxation_customs/publications/official_doc/com/taxation/com20
01260/com2001260_en.pdf.
— (2004), Report from the Commission to the Council, the European Parliament and the
European Economic and Social Committee on the rates of excise duty applied on alcohol
and alcoholic beverages (COM(2004) 223 final, Brussels.
Fenoglio, P, Parel, V., and Kopp, P. (2003), The social cost of alcohol, tobacco and illicit drugs
in France, 1997, European Addiction Research 9:18-28.
Griffiths, S. (2002), Addressing the Health Needs of Rough Sleepers, Office of the Prime
Minister, Homelessness Directorate, London.
Grossman, M. et al. (1994), Effects of alcohol price policy on youth: A summary of economic
research, Journal of Research on Adolescence 4/2: 347-64.
Grossman, M., Chaloupka, F., and Sirtalan, I. (1998), An empirical analysis of alcohol
addiction: results from the Monitoring the Future panels, Economic Inquiry, 36: 39–48.
— Sindelar, J. L., Mullahy, J., and Anderson, R. (1993), Policy watch: alcohol and cigarette
taxes, Journal of Economic Perspectives 7/1: 211–21.
— Chaloupka, F.J., . Saffer, H., and Laixuthai, A. (1993), Effects of Alcohol Price Policy on
Youth, National Bureau of Economic Research Working Paper No. 4385 (Cambridge, MA:
NBER).
Gruber, J., ed. (2001) Risky Behavior Among Youths: An Economic Analysis, University of
Chicago Press, Chicago.
Harrison, P. and Kassler, W.J. (2000), Alcohol Policy and Sexually Transmitted Disease Rates United States, 1982-1995, U.S. Department of Health and Human Services, Center for
Disease Control, Epidemiology Program Office, MMWR Weekly 49/16: 364-9.
15
Health21 (1999), The health for all policy framework for the WHO European Region, WHO
Regional Office for Europe (European Health for All Series, No. 6).
IAS Fact Sheet, Economic Costs and Benefits, 14 November 2003.
Kenkel, D. S. (1993), Drinking, driving and deterrence: the effectiveness and social costs of
alternative policies, Journal of Law and Economic, 36: 877–913.
— (1996), New estimates of the optimal tax on alcohol, Economic Inquiry, 34: 296–319.
Kershaw, C., Budd, T., Kinshott, G., Mattinson, J., Mayhew, P., and Myhill, A. (2000), The
2000 British Crime Survey, England and Wales, Home Office Statistical Bulletin, 18/00.
Klingemann, H. (2001), Alcohol and its social consequences – the forgotten dimension, World
Health Organization Regional Office for Europe.
Leon, D.A. and McCambridge, J. (2006), Liver cirrhosis mortality rates in Britain from 1950 to
2002: an analysis of routine data, The Lancet 367 (January 7): 52-6.
Lieber, C.S. (2003), Alcohol and health: A drink a day won’t keep the doctor away, Cleve Clin
J Med 70:945-53.
Lima, E., and Esquedro, T. (2003), The economic costs of alcohol misuse in Portugal, Working
Paper Series No 24, Nucleo de Investigaçao em Microeconomica Alicada, Universidade do
Minho.
Leung, S., and Phelps, C. (1993), My kingdom for a drink ...? A review of estimates of the price
sensitivity of alcoholic beverages, in M. E. Hilton and G. Bloss (eds), Economics and the
Prevention of Alcohol-Related Problems, NIH Publication 93-3513, Rockville, MD:
National Institutes of Health.
Lynksey, M.T., Fergusson, D.M., and Horwood, J.L. (1994), The effect of parental alcohol
problems on rates of adolescent psychiatric disorders, Addiction 89: 1277-86.
MacDonald, Z., and Shields, M. (2001), The impact of alcohol consumption on occupational
attainment in England, Economica 68: 427–53.
Manning, W. G., Keeler, E. B., Newhouse, J. P., Sloss, E. M., and Wasserman, J. (1989), The
taxes of sin: do smokers and drinkers pay their way? Journal of the American Medical
Association 261, 1604–9.
— Blumberg, L., and Moulton, L. H. (1995), The demand for alcohol: the differential response
to price, Journal of Health Economics, 14: 123–48.
Markowitz, S. (1999), The Price of Alcohol, Wife Abuse, and Husband Abuse, Working Paper
6916, National Bureau of Economic Research, Cambridge, MA.
— and Grossman, M. (1999), Alcohol regulation and domestic violence towards children,
Contemporary Economic Policy 16/3: 309-21.
Maynard, A., Godfrey, C., and Hardman, G. (1994), Conceptual issues in estimating the social
costs of alcohol, prepared for an international symposium on the Economic Costs of
Substance Abuse, Banff, Canada, 11–13 May. Available at
www.ccsa.ca/Costs/maynard.htm.
— and Reilly, A. (1995), The taxation of alcoholic beverages, in C. Sandford (ed.), More Key
Issues in Tax Reform, Fiscal Publications, Bath.
Miller, T.R., and Blincoe, L.J. (1994), Incidence and costs of alcohol-involved crashes in the
United States, Accident Analysis and Prevention 26/5: 583-92.
NIAAA – National Institute on Alcohol Abuse and Alcoholism (2000), Tenth Special Report to
the US Congress on Alcohol and Health, Washington DC, US Government Printing Office.
Norström, T. (ed.) (2002), Alcohol in Postwar Europe: Consumption, Drinking Patterns,
Consequences and Policy Responses in 15 European Countries, National Institute of Public
Health, Stockholm.
— (2002), Per capita alcohol consumption and all-cause mortality in 14 European countries,
Addiction.
OPCS, Office of Population Censuses and Surveys (1996), Psychiatric Morbidity Among
Homeless People, HMSO, London.
Pigou, A. C. (1920), The Economics of Welfare, London: Macmillan.
Pogue, Thomas F., and Larry G. Sgontz (1989), Taxing to Control Social Costs: The Case of
Alcohol,' American Economic Review, 79/1: 235-243.
Ramsey, F. (1927), A contribution to the theory of taxation, Economic Journal, 37: 47–61.
Rehm, N., Room, R., and Edwards, G. (2001), Alcohol in the European Region – Consumption,
Harm and Policies, World Health Organization Regional Office for Europe, Copenhagen.
16
— and Gmel, G. (2002), Average volume of alcohol consumption, patterns of drinking and
mortality among young European in 1999, Addiction 97/1: 105-9.
Royal College of Physicians and the British Paediatric Association (1995), Alcohol and the
Young, Royal College of Physicians, London.
Ruhm, C.J. (1996), Alcohol Policies and Highway Vehicle Fatalities, Journal of Health
Economics 15: 435-54.
Saffer, H., and Chaloupka, F. J. (1994), Alcohol Tax Equalization and Social Costs, Eastern
Economic Review 20/1, 33-43.
— and Dave, D. (2003), Alcohol Advertising and Alcohol Consumption by Adolescents,
NBER Working Paper No. 9676 (May).
Scottish Executive Health Department, Alcohol Misuse in Scotland: Trends and Costs, Catalyst
Health Economics Consultants, Ltd.
Settertobulte, W., Jensen, B.B., Hurrelman, K. (2001), Drinking Among Young Europeans,
WHO Regional Office for Europe, Copenhagen.
Single, E., Collins, D., Easton, B., Harwood, H., Lapsley, H., Kopp, O., and Wilson, E. (2001),
International Guidelines for Estimating the Costs of Substance Abuse, Canadian Centre on
Substance Abuse, Ottawa.
Skog, O-J. (1996), Public health consequences of the J-curve hypothesis of alcohol problems,
Addiction 91: 325-37.
Smith, S. (2005), Economic Aspects of Alcohol Taxation, in S. Cnossen (ed.), Theory and
Practice of Excise Taxation: Smoking, Drinking, Gambling, Polluting, and Driving, Oxford
University Press, Oxford.
Strategic Task Force on Alcohol (2002), Interim Report, Department of Health and Children,
Dublin
UK Cabinet Office (2003), Alcohol Misuse: How Much Does it Cost? Strategy Unit,
September.
Velleman, R. (1993), Alcohol and the Family, Institute of Alcohol Studies, London.
Williams, J., Chaloupka, F.J., Wechsler, H. (2002), Are There Differential Effects of Price and
Policy on College Students’ Drinking Intensity? NBER Working Paper No. 8702 (January).
World Health Organization (2004), 'Global Status Report on Alcohol,'
http://www.who.int/substance_abuse/PDFfiles/who_ncd_msd_2001_2.pdf
— (2006), Global Alcohol Database,
http://www3.who.int/whosis/alcohol/alcohol_about_us.cfm?path=whosis,alcohol,alcohol_a
bout&language=english).
Young, D. J. and Bielinska-Kwapisz, A. (2002), Alcohol taxes and beverages prices, National
Tax Journal 55/1: 57-73.
17
Table 1. European Union: Alcoholic Beverages – Taxes and Prices on 1 January, 2006
Excise duties
Member
Statea
VAT
(%)
Beer per
0.5 literb
Still wine
per 75cl
Spirits per
70cl, 40%
EU-15
Sweden
Ireland
UK
Finland
Denmark
Netherlands
Belgium
France
Germany
Greece
Luxembourg
Austria
Portugal
Spain
Italy
0.33
0.79*
0.99*
0.95*
0.97
0.34*
0.13
0.09
0.13*
0.04
0.06
0.04
0.10
0.08
0.05
0.12
0.58
1.78
2.05
1.85
1.59
0.62
0.44
0.35
0.03
0.00
0.00
0.00
0.00
0.00
0.00
0.00
5.36
15.08
10.99
8.05
7.91
5.63
4.21
4.91
4.06
3.65
3.05
2.92
2.80
2.62
2.32
2.24
EU-10
Malta
Poland
Estonia
Lithuania
Czech Rep.
Latvia
Hungary
Slovenia
Slovak Rep.
Cyprus
0.13
0.04
0.09
0.19*
0.11*
0.04
0.09*
0.10
0.34*
0.06
0.24*
0.14
0.00
0.26
0.50
0.33
0.00
0.32
0.00
0.00
0.00
0.00
EU-25
0.25
0.40
Pre-tax prices (€)
Still wine
per 75 cl
19.8
25
21
17.5
22
25
19
21
19.6
16
19
15
20
21
16
20
0.60
0.22
0.75
1.31
0.18
0.40
0.39
0.69
0.42
0.65
.
0.67
0.45
0.38
0.55
1.30
2.02
1.30
.
2.92
2.22
2.58
1.82
1.67
1.88
1.90
.
3.04
2.52
1.28
0.66
2.50
3.75
1.68
.
.
4.22
2.96
2.42
5.02
5.14
0.67
.
3.51
2.61
1.41
5.32
9.99
2.83
6.52
3.25
2.72
2.59
2.51
2.53
2.46
1.95
2.04
1.71
18.7
18
22
18
18
19
18
20
20
19
15
0.34
0.12
0.27
0.28
0.24
0.15
0.51
0.33
0.79
0.37
.
1.74
1.08
1.53
2.76
2.25
0.99
2.08
0.76
1.88
2.33
.
3.10
9.75
1.67
1.33
1.71
0.33
3.11
1.07
6.08
2.82
.
4.35
19.4
0.50
1.91
3.47
Sources
Excise duties: European Commission, Excise Duty Tables, January 2006.
VAT: European Commission, VAT rates, January 2006.
Prices, including tax: http://data.euro.who.int/alcohol/Default.aspx?TablD=2422
(downloaded 31 January 2006). Prices, excluding tax: author’s calculations.
Notes
a. Ranked in order of the highest duty on spirits.
b. An asterisk (*) indicates that the beer excise is levied per hl/degree of alcohol of finished product.
18
Spirits per
70 cl, 40%
Beer per
0.5 liter
Table 2. European Union: Alcohol Consumption, Prevalence and Effects in 2003
Member
Statea
Consumption
of pure
alcohol
(liters per
capita)b
Drinking prevalences
Alcohol-related diseases and accidents
Youth
Drinking
(% of 15- yr
olds)
Heavy
drinking
(% among
drinkers)
Alcohol
dependence
(% among
drinkers)
Liver
cirrhosis
External
injuries per
100.000
Traffic
accidents
per 100.000
.
4.7
.
3.7
3.8
.
.
8.7
2.2
4.0
.
7.0
1.7
5.5
.
4.9
10.4
12.7
13.9
17.0
10.2
14.1
13.3
18.3
12.4
4.8
11.8
13.6
4.5
5.3
39.4
27.3
51.9
45.9
34.2
32.9
48.4
55.6
44.5
67.8
32.5
54.3
34.3
28.5
40.0
.
20.8
31.3
22.1
29.4
10.6
21.4
.
35.0
19.0
.
41.3
5.1
12.8
12.8
.
4.8
.
26.5
53.5
EU-15
Ireland
UK
Luxembourg
Denmark
Germany
Spain
Portugal
France
Austria
Finland
Greece
Belgium
Italy
Netherlands
Sweden
10.2
11.8
11.6
11.6
11.5
11.2
11.0
10.6
10.3
10.3
9.9
9.7
9.3
8.4
8.4
6.9
31.6
17.4
52.0
.
46.5
39.3
28.3
15.5
16.9
34.5
16.8
27.5
39.1
37.1
51.4
20.1
11.1
26.0
11.3
.
11.7
11.2
2.6
.
12.2
18.3
4.6
3.6
20.1
5.8
10.0
6.5
EU-10
Lithuania
Slovak Rep
Hungary
Latvia
Czech Rep
Slovenia
Estonia
Cyprus
Poland
Malta
12.2
15.5
15.5
15.4
15.1
12.5
11.6
11.5
10.0
9.7
4.9
25.7
12.5
27.0
24.4
17.5
28.9
34.3
23.9
22.0
19.2
47.8
6.3
1.9
8.8
2.8
2.2
19.1
.
1.7
.
7.9
.
EU Average
11.0
29.2
8.6
.
.
1.3
12.2
.
16.7
.
.
.
12.7
147.8
55.4
80.0
136.3
64.2
70.1
129.3
.
62.4
31.7
20.5
24.2
33.4
.
.
.
18.1
.
Sources
Consumption – recorded: World Drink Trends (2005), unrecorded: WHO (2004).
Drinking prevalences: WHO (2004).
Effects: WHO (2004).
Notes
a
Ranked in order of highest per capita alcohol consumption.
b
Including unrecorded consumption of – 0.5 lt: Belgium, Netherlands; 1 lt: Ireland, France,
Germany, Austria, Portugal, Spain, Cyprus; 1.3 lt: Slovenia; 1.5 lts: Czech Republic, Italy; 2 lts:
Denmark, UK, Finland, Greece, Sweden; 3 lts: Poland; 4 lts: Hungary; 4.9 lts: Lithuania; 5 lts:
Estonia; and 7 lts: Slovak Republic, Latvia. No figure is available for Malta and in
Luxembourg, unrecorded consumption is -1 lt due to bootlegging.
19
Table 3. England and Wales: External Costs of Alcohol Abuse in 2001
Category of Cost
€ million
Percent of
(1£ = €0.60)
pre-excise
alcohol
expenditure
Alcohol-related and alcohol specific crime
19.900
36.7
Criminal justice system costs
2.917
Property/health and victim services
4.202
Costs in anticipation of crime (alarms, etc.)
2.492
Lost productive output of victims
1.617
Emotional impact costs for victims of crime
7.798
Drink Driving
875
Criminal justice system costs
128
Lost output
100
Medical and ambulance
52
Human costs
595
Lost Output
10.703
19.8
Absenteeism
2.977
Reduced employment
3.590
Reduced employment efficiency
n.a.
Premature death
4.137
Health Care
2.805
5.2
Hospital inpatient visits
877
Hospital outpatient visits
743
Accident and emergency visits
508
Ambulance services
342
Nurse and GP consultations
113
Specialist treatment services
160
Other health care costs
62
Total including emotional impact costs
and costs of lost output
Total excluding emotional impact costs
and costs of lost output
Pre-tax expenditure on alcoholic drinks
Excise duties
[VAT
33.400
61.7
14.900
27.5
54.100
10.200
9,100]
18.9
Sources: Social costs: UK Cabinet Office (2003). Expenditure on alcoholic drinks: IAS Fact Sheet (14
November 2003), adjusted for England and Wales on a population basis. Taxes and population: OECD
(2005). Figures may not add because of rounding.
20
Table 4. A short survey of the human costs of alcohol misuse in England
Author
Year Content of study-findings
Velleman
1993 Marriages where one or both partners have an alcohol problem
are twice as likely to end in a divorce as marriages where no
alcohol problems exist
Barber & Crips
1994 Children of problem drinkers have high levels of behavioral
difficulty, school related problems, such as under-performance
and truancy, and emotional problems
Lynskey et al.
1994 By the age of 15 young people in families with a problem
drinking parent have higher rates of psychiatric disorder than
other young people
Royal College of 1995 Alcohol problems are also a significant factor in male teenage
Physicians
suicides
OPCS
1996 A third of those using shelters were severely alcohol dependent
Jacobs
1998 Up to 70 percent of men who assault their partners are under the
influence of alcohol when the assault takes place
Kershaw et al.
1999 Data from the British Crime Survey indicate that in 44 percent of
cases of domestic violence, the victims thought their assailants
had been drinking
Farrow & Young 2000 Young women are more likely to have casual sex while under
the influence of alcohol
Griffiths
2002 Half of those sleeping rough are alcohol reliant
Source: UK Cabinet Office, Strategy Unit (2003).
21
Table 5. European Union: Alcohol Excise Revenues and Duties
Member
a
State
EU-15
Finland
Ireland
UK
Sweden
Denmark
Netherlands
Belgium
Luxembourg
France
Germany
Austria
Greece
Spain
Portugal
Italy
Excise Revenues in 2003
(euro)
Excise duty
Effective duty
collections
per liter pure
b
per capita
alcohol
84.8
9.9
261.5
33.1
249.8
23.1
126.4
13.2
122.6
25.0
101.8
10.7
52.1
6.6
53.8
6.1
73.3
5.8
47.8
5.1
44.4
4.4
42.3
4.5
27.9
3.6
27.7
2.8
24.8
1.6
15.7
2.3
Excise Duties
by alcohol content in 2004
Beer
Still wine
Spirits
5%
12%
40%
29
14
50
13
58
68
28
21
7
12
12
25
21
22
41
44
20
60
56
67
38
33
27
23
0
2
0
0
0
0
0
0
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
EU-10
Malta
Poland
Estonia
Lithuania
Czech Rep.
Latvia
Hungary
Slovenia
Slovak Rep.
Cyprus
28.8
.
42.2
.
.
29.4
.
23.5
.
20.3
.
3.2
.
5.5
.
.
2.7
.
2.1
.
2.4
.
64
30
76
44
18
44
44
24
25
16
0
24
55
37
0
44
33
0
0
0
100
100
100
100
100
100
100
100
100
100
EU-25
73.0
8.5
33
20
100
Sources
Excise duties: OECD (2005).
Consumption: table 2, recorded consumption only.
Alcohol content: table 1.
Notes
a
Ranked in order of highest excise duty collections per capita.
b
Recorded consumption
22
Table 6. Effectiveness of Alcohol Policies
Strategy or intervention
Effectiveness
+++ = highest
0 = not effective
? = unknown
Taxation and pricing
Alcohol taxes
+++
Regulating physical availability
Minimum legal purchase age
+++
Government monopoly of retail sales
+++
Server legal liability
+++
Restrictions on density of outlets
++
Hours and days of sale restrictions
++
Drink-driving countermeasures
Random breath testing (RBT)
+++
Lowered BAC limits
+++
Low BAC for young drivers (‘zero tolerance’)
+++
Administrative license suspension
++
Designated drivers and ride services
0
Altering the drinking context
Outlet policy not to serve intoxicated persons
+++
Enforcement of on-remise regulations
++
Training bar staff to manage aggression
+
Voluntary codes of bar practice
0
Promoting alcohol-free activities
0
Treatment and early intervention
Brief intervention with at-risk drinkers
++
Alcohol problems treatment
+
Mutual help/self-help attendance
+
Mandatory treatment of repeat drinking-drivers
+
Regulating alcohol promotion
Advertising bans
+
Advertising content controls
?
Education and persuasion
Alcohol education in schools
0
College student education
0
Public services messages
0
Warning labels
0
Source: Babor et al (2003), downloaded from www.ias.org.uk (Alcohol
Policies).
23