Content uploaded by Ben Baumberg Geiger
Author content
All content in this area was uploaded by Ben Baumberg Geiger on Apr 03, 2019
Content may be subject to copyright.
Page 1 of 38
This is a post-peer-review, pre-copy edited version of an article published in Drug and
Alcohol Review. The definitive publisher-authenticated version of
The global economic burden of alcohol: a review and some suggestions
(Baumberg 2006). Drug and Alcohol Review, 25(6):537-552.
is available online at
http://informahealthcare.com/doi/abs/10.1080/09595230600944479
Page 2 of 38
The global economic burden of alcohol: a review and some
suggestions
Ben Baumberg
1
1
Institute of Alcohol Studies, London, UK
Abstract
Background: economic arguments for acting for health are increasingly important for
policymakers, yet to date there has been no consideration of the likely economic burden
of alcohol on the global level. Method: a review of existing cost estimates was
conducted, with each study disaggregated into different cost areas and the methodology
of each element evaluated. The range of figures produced from more robust studies was
then tentatively applied on the global level. Results: the reviewed studies suggested a
range of estimates of 1.3-3.3% of total health costs, 6.4-14.4% of total public order and
safety costs, 0.3-1.4 ‰ of GDP for criminal damage costs, 1.0-1.7 ‰ of GDP for drink-
driving costs, and 2.7-10.9 ‰ of GDP for workplace costs (absenteeism, unemployment
and premature mortality). On a global level, this suggests costs in the range of $210-
$665bn in 2002. Discussion: these figures cannot be understood without simultaneously
considering six key problems: (i) the methods used by each study; (ii) who pays these
costs; (iii) the ‘economic benefits’ of premature deaths; (iv) establishing causality; (v)
omitted costs; and (vi) the applicability of developed country estimates to developing
countries. Conclusion: alcohol exerts a considerable economic burden worldwide,
although the exact level of this burden is a matter of debate and further research.
Policymakers should consider economic issues alongside evidence of the cost-
effectiveness of particular policy options in improving health, such as in the WHO’s
CHOICE project.
Page 3 of 38
Introduction
However much those working in the public health field may prefer otherwise, the
importance of ‘economic’ arguments
1
when trying to persuade policymakers to act for
health cannot be understated. As Markos Kyprianou, the Commissioner for Health and
Social Protection in the European Commission, has recently put it, “this evidence
provides a powerful argument for European governments to invest in the health of their
populations, not only because better health is a desirable objective in its own right, but
also because it is an important determinant of economic growth and competitiveness”
(1:5).
This is no less true in the alcohol field than in any other field of public health. The recent
resolution on alcohol in the World Health Assembly noted that WHO members are
“concerned about the economic loss to society resulting from harmful alcohol
consumption” (WHA 58.26). Single figure estimates are also useful for comparing the
size of different health issues, often tobacco, alcohol and illicit drugs (2), and are a
platform on which to build economic analyses of particular policy options (see Chisholm
et al, this issue). Finally, the data demands in these analyses acts as a spur to filling gaps
in the evidence base (3), in particular for the costs omitted in cost-effectiveness analyses
(a point to which we will return in the conclusion).
Unsurprisingly then, there are numerous social cost estimates for different social issues at
the European level (4-6). Amongst these are a series of estimates of the social cost of
alcohol in Europe, originally within the European Alcohol Action Plan 2000-5 (see also
7), but also two increasingly sophisticated aggregate level estimates (8, 9), culminating in
a WHO decision to initiate a detailed study of the social cost of alcohol in each of the
countries of Europe.
1
‘Economic’ is here narrowly defined as arguments involving the money economy, rather than the broader
sense that encompasses changes in Quality of Life (which is therefore much closer to a public health
approach); this point is developed below in footnote 21.
Page 4 of 38
However, to date there have been no estimates of the likely economic burden at the
global level. This paper attempts to fill this gap by using a review of existing social cost
studies – conducted for the most recent European estimate – to tentatively outline the
international economic burden due to alcohol. As well bringing previous reviews up-to-
date, this review also disaggregates existing studies to look at individual cost areas,
which gives a much clearer picture of how estimates compare to each other.
At the same time, though, the review also highlights the assumptions on which these
estimates are based, many of which appear both problematic in themselves and are
misinterpreted by policymakers in practice. Furthermore, virtually all of these studies
have been conducted within developed countries, and neither the exact estimates nor the
assumptions within them may be appropriate in developing countries. The paper
therefore goes on to critically review the estimates of economic burden, focusing on five
key problems that need to be addressed. The paper concludes with some suggestions as
to the likely size of the global economic burden due to alcohol, and finally places these
estimates within the wider context of economic evaluations.
Methods
2
Theoretical background
The point of departure for nearly all social cost studies is an unreal ‘what if?’ question –
‘what if alcohol disappeared from the world today?’ Behind this unreal scenario lies an
even bigger assumption – that in this hypothetical world, all of the spending and time
spent drinking alcohol is redirected to something that does not burden society in any way.
Thankfully the utility of these studies does not depend upon the plausibility of this
situation, with this imaginative thinking instead being a way of creating a summary
measure of how much (and in what areas) alcohol burdens human society materially.
2
A number of methodological considerations are not discussed here for both readability and space.
However, readers with unanswered questions should refer to Anderson and Baumberg (2006) or contact the
present author.
Page 5 of 38
Social cost studies also involve a number of other assumptions, some shared between all
studies and others varying between them, of which it is beyond the scope of this article to
cover in full.
3
It is however worth being aware that these studies generally estimate the
economic burden of all past and current drinking on a particular year (prevalence-based),
rather than the burden of new alcohol-related problems on all future years (incidence-
based).
The review also divides between ‘tangible’ and ‘intangible’ costs (see Table 1).
‘Tangible’ costs are those costs that are already in monetary form, such as spending on
healthcare (direct costs), or production losses (indirect costs). ‘Intangible’ costs are those
that do not exist in a monetary form, such as pain, suffering or loss of life itself. The
question of whether intangible costs count as an ‘economic cost’ is discussed below in
footnote 21, and their importance for understanding economic contributions to alcohol
policy is discussed in the conclusion.
Finding and selecting source studies
This paper is based on a review of costing studies undertaken since 1990 (earlier studies
are generally weaker and have been reviewed elsewhere (10)). An initial list of studies
was obtained from PubMed, ETOH and the Web of Science, checked against the WHO’s
Global Status Report on Alcohol 2004 (11) and a search of the Internet through
google.com.
4
These were supplemented by studies from four previous reviews (10),(12-
14) as well as those provided by the European Alcohol Policy Network (APN;
www.eurocare.org/btg/).
5
These studies were selected if they included a new estimate of
the social cost in at least one cost area (the full list of studies included is shown in Table
3
Interested readers can also find further discussion elsewhere (56).(17, 28, 106)
4
The following search terms were used: alcohol*, combined with economic*, cost*, or burden*.
5
Data in languages other than English, French, German or Spanish were translated by the relevant APN
member, using a standard form to extract relevant information only. In two cases the studies were not
publicly available; for transparency purposes, the English summaries have been made available on the APN
website.
Page 6 of 38
1). Due to data limitations, it was not possible to include studies for either West
Germany (e.g. 15) or sub-national regions (13, 16).
Method of analysis
To overcome problems of methodological inconsistency and coverage of different cost
areas, each study was divided into separate cost areas.
6
For each one, costs were then
expressed as the percentage of an appropriate common metric (or GDP if none were
available) – healthcare costs, for example, were expressed as a percentage of total
healthcare expenditure. The methods used in different studies were also compared using
a checklist of methods and transparency, based as far as possible on the WHO Guidelines
for Estimating the Costs of Substance Abuse (17) and shown in Tables 2-5.
[Tables 2-7 from about here]
Results
The results of the review are shown in Tables 2-7. Each table presents the results for a
particular cost area, expressed as a share of a common metric (‘total health costs’ for
health, ‘total public and order and safety costs’ for crime, but GDP in all other cases).
The tables also include additional information on the method used by the authors in
producing the estimate, with studies meeting all of these criteria shown in bold. This
review is therefore ‘systematic’ in the sense that it examines a common set of
methodological criteria within each cost area – but it is also ‘opportunistic’ in the choice
of these criteria, which are only partially based on good practice guidelines (e.g. 17) and
partially based on the limited extent of the information given in the studies.
It should be noted that there is still substantial methodological variation between studies
that appear similar in the tables, to the extent that it is impossible to attribute variations in
6
All transfers between individuals – whether deliberate or stolen – were also removed (17, 107), as have
the health costs of violent crime (due to the risk of double-counting (28)) and non-market costs such as
household work (as these cannot strictly be compared to GDP (17)).
Page 7 of 38
costs to genuine differences in levels of harm. For example, differences in health costs
may reflect the use of different estimates of relative risk for the same condition, while
differences in premature mortality costs may reflect the use of different assumptions as to
future productivity increases (even the linked studies from Canada use assumptions on
productivity growth that differ by 2%).
A global estimate
While the figures presented in Tables 1-7 are interesting for the purposes of comparison,
the easiest way to comprehend what these figures mean is to apply them globally, based
on the common metric used in the review.
7
Clearly this will be a tentative estimate
subject to several qualifications, which are covered in detail in the discussion section of
this paper. Nevertheless, presenting these implied figures from the review gives a first
crude indication of the likely global economic burden of alcohol, representing both
information in itself for policymakers and a platform for future work.
The most sensible basis for a global estimate is to restrict the estimates to results from the
best studies, i.e. those highlighted in bold in Tables 2-7. If these are applied on a global
basis, then we can estimate that the global economic burden of alcohol is between
$210bn and $650bn in 2002.
8
This is made up of $40-105bn for health, $55-210bn for
premature mortality, $30-65bn for absenteeism, $0-80bn for unemployment, $30-85bn
for criminal justice systems (police, prisons, courts) and $15-50bn for criminal damage.
This is equivalent to 0.6-2.0% of global GDP (for the countries included) – or more
simply, somewhere between the total GDP of Austria and India.
One small refinement can be made to these extremely crude figures, by using the existing
estimates of the global health burden caused by alcohol (see Rehm, this volume). If we
7
Crime costs are expressed as a % of GDP given the lack of available data on ‘public order and safety’
expenditure as presented in Table 6.
8
GDP figures taken from the UN Statistical Division 1/2/2006; health spending as % of GDP figures taken
from the World Health Report 2005, published by the WHO. All figures are stated to the nearest $5bn to
avoid giving a misleading imprecision of precision.
Page 8 of 38
adjust for the difference in health burden between the countries where cost studies have
been conducted (all very low child and adult mortality) and other countries worldwide,
then the estimate for the global burden of alcohol on health costs becomes $50-120bn
(compared to $40-105bn for the initial method above). The WHO figures further
suggest that developed countries may have a slightly higher share of homicides and road
traffic deaths due to alcohol than elsewhere.
9
Discussion
While this review allows a tentative first estimate of the global economic burden of
alcohol, it cannot be understood without considering six further points that are discussed
below in turn: (i) the methods used by each study; (ii) who pays these costs; (iii) the
‘economic benefits’ of premature deaths; (iv) establishing causality; (v) omitted costs;
and (vi) the applicability of developed-country estimates to developing countries.
The methods used by each study
The present estimate is a distinct improvement on previous review-based estimates (7, 8)
as it accounts for the different cost types included in different studies, and examines how
robust the methods are within each cost type and within each study. Nevertheless, there
are substantial outstanding concerns as to the accuracy of individual studies, as data
limitations are often only surmountable through extrapolations from other data sources
(for example, assuming that the alcohol-attributable share of outpatient health costs is
equal to that of inpatient health costs). Epidemiological data may be similarly
extrapolated from one country to another and from mortality to morbidity. Such
extrapolations may be an inevitability in social cost studies given their onerous data
demands, but it does mean that no estimate can be seen to be precise.
9
For homicide: 31 (developed countries) vs. 32% (developing) of deaths for males and 31% vs. 22% for
females. For road traffic accidents: 41% of deaths in men aged 15-29 compared to 30% in other regions.
‘Developed countries’ defined as countries with very low child and adult mortality; ‘developing countries’
refers to all other countries. Average figures are obtained by scaling the reported alcohol-attributable
fraction for homicide (Rehm et al 2004) by the numbers of homicides in each region (see
http://www.who.int/healthinfo/bod/en/index.html)
Page 9 of 38
A more theoretical concern is that the review can only follow the data and assumptions
used by primary researchers, the most contentious of which relates to the method of
placing a value on premature death. Studies most commonly assume full employment,
where prematurely deceased workers are not replaced by those who are unemployed.
However, even by the late-1990s it had been widely noted that this could lead to a large
over-estimation of costs, and alternative methods were being recommended (18).
The most common of these – known as the ‘friction cost’ method – has been to assume
that workers who die are replaced at work by a previously unemployed person (19),
which produces a much lower cost of alcohol to society (20, 21). This in turn has been
critiqued for being over-sensitive to macroeconomic variables that are difficult to predict
(22), and more importantly for making unwarranted assumptions that lead to the friction
cost method being an underestimate (see 23, 24). For example, the ‘friction cost’
generally misses the cost of people who cannot be replaced by currently unemployed
people and the likely chains of vacancies that arise by replacing workers with people
employed elsewhere, as well as the cost of training up new workers (23, 25). In practice,
then, the ‘true cost’ is likely to lie between the two estimates – but it is difficult to be
more precise as to exactly where the cost lies.
Who pays these costs?
Perhaps more important than the level of costs per se is whether it is drinkers, other
individuals, government or businesses that pay them. Politicians often see external costs
– those that the drinker imposes on other people, such as taxpayer-funded healthcare
costs in the UK – as a reason to intervene in markets, as people do not (often) take these
into account themselves. On the other hand, private costs – those that are paid by the
drinker or their family, such as private healthcare in the US – may be seen as a matter of
individual choice, given that rational individuals only do something if the (private) gains
are more than the (private) costs (26).
As well as this basic division of costs, there are situations that are ‘private’ in a simple
view but may nevertheless be persuasive in justifying government regulation. One of
Page 10 of 38
these is harm within the household (e.g. child neglect), which is treated as private in
mainstream economics but often seen as requiring intervention in wider society. Indeed,
it seems clear that due to alcohol-related harm within the family, the proper unit of
analysis should be the individual rather than the household.
Another politically important situation is where people cannot make rational decisions,
such as where they are addicted, or if they do not have good information on the costs and
benefits of how much they drink (17, 27). Some researchers have argued that private
decisions made in these conditions should also be included in the total social cost figure –
typically by putting a certain share of total consumption as ‘abusive’ (28, 29). Clearly
this decision adds substantial amounts to the total figure – the total estimated social cost
of alcohol in the EU of €125bn (9) compares to €125bn spent on beer alone in the EU
(30), which using the assumptions of Collins and Lapsley in Australia
10
would imply an
additional cost of €25bn. One consequence of including these costs, however, is that a
poor information campaign – which would increase knowledge of the risks of alcohol
without changing behaviour – would substantially reduce costs without affecting harm,
which would seem to be a slightly counterproductive result of an ineffective policy.
Studies that look only at external costs are rare, and are generally adjustments to existing
major social cost estimates (31-33). These suffer from the exclusion of ‘transfer costs’ in
most studies – where money is moved from one group to another (e.g. from government
to the unemployed), rather than being lost – which are likely to be significant in external
cost studies. The most revealing study that looks at who pays the costs comes from
Australia (28), where the highest costs for business and government were workforce
labour, road accidents and crime.
11
Together these accounted for nearly two-thirds of the
total cost, even after considering the enormous private cost of ‘resources used in abusive
10
This assumption is that 20% of all alcohol consumed was drunk by addicted drinkers and should
therefore be included in the social cost. This excludes the cost of drinking when the drinker does not have
complete information on the risks of alcohol.
11
It should be noted that health costs in the Collins and Lapsley studies are relatively low compared to
other studies as they take into account the ‘savings’ of premature deaths; see below.
Page 11 of 38
consumption’. Other research also suggests that only around one-half of those injured in
these accidents are the drink-drivers themselves (9), which means that much of the cost is
borne by the ‘innocent victims’ of these accidents.
The ‘economic benefits’ of premature deaths
One of the stranger aspects of social cost studies is that they tend to look at present costs
without looking at future savings. For example, in a world without alcohol, a person
dying of alcohol-attributable liver cirrhosis at age 50 may otherwise have contracted
Alzheimer’s disease at age 80. This may partly explain the inconsistent and otherwise
puzzling finding (34) that healthcare costs for abstainers and heavy drinkers are both
much higher than for lighter drinkers (e.g. 35), although this may also reflect
methodological weaknesses (36) and relatively short follow-ups.
12
.
The ‘real’ health burden therefore depends on whether alcohol-related disease is cheaper
to treat than all the diseases that people would otherwise have got in an alcohol-free
world. This means that there is a lower social cost of conditions that kill people quickly,
as opposed to conditions that lead to long periods of illness requiring expensive treatment
(37). In the one study that has investigated this for alcohol, an overall healthcare cost
was found, but this was a much lower value than found in the conventional
methodology.
13
Does this mean that conventional health cost estimates should be discarded as
meaningless? I would argue that this would be an overreaction, for two reasons. Firstly,
12
More theoretically though, there are two reasons why we may not expect individual-level studies to
match the social cost results. Firstly, there may well be systematic biases in how individuals react to
healthcare, with some research suggesting that relatively heavy drinkers may have shorter stays in hospital
as they cannot carry on with their drinking (36). Such biases are particularly important for outpatient
service use, where variations in care-seeking behaviour are likely to explain at least part of the negative
relationship between alcohol use and outpatient service utilization (see summary in 35).
13
Ironically, Collins and Lapsley also found that the health benefits of alcohol increase the total healthcare
burden. This is presumably because cardiovascular disease leads to death relatively more than long-term
disability, and so preventing cardiovascular disease leads to an increase in health costs.
Page 12 of 38
cost studies for all social issues use the same methodology, so that any change must be
made across the board rather than in a single study. Secondly, these estimates are still
meaningful, in that it is accurate to say that an estimated $50-120bn was spent treating
conditions caused by alcohol in 2002. It is unarguable, however, that the figures should
be presented clearly to avoid the all-too-common misunderstandings among
policymakers.
It should finally be noted that a parallel finding can be found for the productivity cost of
premature death, if we take into account the resources that people would otherwise have
used up if they had stayed alive (e.g. pensions). Two studies have estimated this for
alcohol, in both cases finding that it reduced the cost of premature mortality by about
30% (28, 38). However, the implication of these analyses is that an individual’s
expected production is less than their expected consumption from the age of 55 (39),
which is politically troubling, and whose implications are discussed in more detail in the
conclusion.
Establishing causality
The difficulty in dividing between what is due to alcohol from what is merely associated
with it is a common one in research on alcohol, and there is clearly insufficient space to
discuss the issue fully in this paper (for an introduction, see the brief review in 40). It is
nevertheless essential to consider the issue of causality in social cost studies, if only
because it is often brushed over in both the studies themselves and previous reviews.
The most illuminating way of showing these points briefly is to look at a single case
study, that of lowered productivity.
While there is agreement between drinkers (41) and employers (42) that drinking can
lead to lowered productivity at the workplace, there are substantial difficulties in valuing
this precisely. Most of the existing effort has been made by economists looking at the
issue of wages and alcohol, based on the assumption that – in a perfect labour market –
lower productivity or attendance due to alcohol will result in lower wages. Such
assumptions underlie estimates in several countries (e.g. 21, 43, 44), although several of
Page 13 of 38
these (45) simply imported a US estimate (46) into their own country due to a lack of
other data.
14
Yet while this US study looked at the wage penalty in people who have
suffered from alcohol dependence, nearly all studies also show that light drinkers have
higher wages than abstainers (Russia in 47, the US in 48, Australia in 49, the Netherlands
in 50).
The key question here, then, is whether these results show ‘genuine’ costs and benefits of
alcohol, or simply reflect systematic biases in the methods used (51 p360).
Firstly, income could influence alcohol consumption, in that people with more money can
buy more alcohol. Some researchers have claimed that looking only at alcohol use
disorders (rather than consumption levels) reduces this bias, because individuals do not
‘choose’ to suffer from these in the same way that they can choose to change how much
they drink (46).
15
Secondly though, and more problematically, both drinking levels and
wages may be jointly influenced by some other factor, such as how much risk someone is
willing to take in their decisions. There has been a tendency to ignore this problem,
however, presumably because the spurious relationship will be in the ‘right’ direction and
therefore allow a cost burden to be estimated.
These problems are relatively common in economic analyses, and economists have
developed complex tools to deal with them.
16
Yet the unavoidable conclusion from this
14
Even this US evidence is not altogether convincing – a wage penalty was only found for men who had
ever suffered from alcohol dependence (but not alcohol abuse, and not for women at all). The estimate also
assumes that education is a mechanism through which ever-being alcohol dependent can affect wages (i.e.
it uses a reduced-form model). Harwood et al justify this using research that suggests those reporting
youthful alcohol abuse have less education than would be expected from their background, although more
recent research contradicts this (108). If this assumption is dropped, then Harwood et al find no significant
effect in any group.
15
Although the risk of suffering from an alcohol abuse disorder increases at higher levels (and more
detrimental patterns) of consumption, there are more intervening variables (such as genetic vulnerability)
that create a more uncertain probability than found for the level of consumption.
16
Many studies use called an ‘instrumental variables’ technique by looking at the association between
wages and variables that are correlated with alcohol use but not with wages (e.g. alcohol tax changes).
Page 14 of 38
research is that these tools are simply not good enough to produce unbiased estimates of
the effect of alcohol. For a start, the estimates from this method are often implausibly
large (50), to the extent that the National Bureau of Economic Research concluded that
“it is unlikely that previous studies…have produced credible estimates of the effect of
alcohol use on labor market outcomes” (51:370). A more recent study documents these
flaws even more convincingly by looking at the same people over several years, thereby
controlling for all the differences between drinkers and non-drinkers that do not change
over time (52). This found no significant effect of alcohol use at the 5% level on wages
in either men or women, for any level of consumption or for binge-drinking.
Does this mean that there is no relationship between alcohol use and productivity or
absenteeism (51)?
17
While some researchers would seemingly answer yes (53), it is
unclear on further reflection why anyone ever expected wages to have a ‘genuine’
relationship with levels of alcohol use. At a basic level, worker productivity may not be
easily visible to an employer, and even if it is visible then wages may take some time to
reflect changes in productivity (21). More importantly, in a workplace culture where
drinking is accepted, there may even be a workplace penalty where drinking is expected,
such as for the cantineras in Texas who see drinking as a workplace duty (54). Non-
drinking could equally damage prospects where intra- and inter-firm networking is based
on socialising over drinks, as suggested by the £250 per employee spent on alcohol by
UK advertising firms (55).
This case study provides two lessons for future research. Researchers should be careful
in using a proxy variable instead of the measure they are actually interesting in, but more
importantly, the ever-present risk of spurious relationships requires imagination in
research design rather analysis – a point that echoes a call made 15 years ago (27, 56).
Other than better use of natural experiments, an interesting example is a small US study
that looked at the same individuals over four weeks, and found that the relative risk of
17
Similar results have also been found for drinking and unemployment, in that alcohol use disorders are
associated with higher unemployment, light alcohol consumption is sometimes associated with lower
unemployment, and complex methods often produce implausible results (51).
Page 15 of 38
being absent was 10-times greater for individuals on the day after drinking. Although
this study had other methodological weaknesses,
18
it demonstrates how creative
approaches can produce more persuasive results than more expensive and sophisticated
number-crunching.
Omitted costs
Alcohol touches upon many areas of human life, and few studies can claim to covert
these comprehensively. This may be unimportant, however, if the omitted costs are
merely those that are likely to be relatively insignificant. For example, the cost of fires
that occur because people are drunk is only 1% of the total cost in all of the four countries
where this has been estimated (21, 45, 46, 57). It is therefore unfortunate that many of
the frequently omitted cost areas fall into the groups of either ‘miscellaneous productivity
costs’ or ‘other crime costs’, which are both likely to be responsible for a substantial
burden.
‘Miscellaneous productivity losses’ includes many areas where the effect of alcohol is
costed but the impact on people’s ability to work is ignored. For example, the labour loss
due to all workplace accidents accounts for most of the estimated €55bn total cost for the
EU15 (6). Other parallel areas include lost working time from travel delays due to drink-
driving accidents (58, 59), and from people imprisoned due to alcohol-attributable crime
(28, 46, 57) as well as their victims (42, 60). More fundamentally, traditional methods
systematically miss out the work done by people outside the labour force, such as caring,
housework and voluntary work (see Table 3). More recent studies find ways of giving a
value to this work, generally finding that it significantly adds to the total burden of
alcohol (21, 38, 46, 61).
18
For example, alcohol consumption was reported retrospectively at two-week intervals, and the study did
not ask about the quantity of alcohol consumed (or any other features) of the drinking occasion. Such
weaknesses were enough for a systematic review to classify the study as of ‘low quality’, although it should
be noted that only one study worldwide met their definition of even ‘medium quality’ (109).
Page 16 of 38
Most of the crime costs’ considered above concentrate only on the cost of crime in terms
of the criminal justice system, and occasionally also look at the value of criminal damage.
However, the study from the UK Prime Minister’s Strategy Unit was able to build upon
an existing estimate of the ‘cost of crime’ to look at the money spent by private citizens
and companies on preventing crime in advance (burglar alarms, security guards). This
added up to a burden that was equivalent to the total value of criminal damage and only
slightly lower than the cost of police, courts and prisons, demonstrating that the omission
of these costs leads to a serious underestimate of the cost of alcohol-attributable crime
(42).
19
From a global perspective, though, perhaps the most worrying omission is one that has
never been included in a social cost study – the burden of sexually transmitted diseases
(STDs), particularly the enormous burden of HIV/AIDS (62). Certainly there is almost
universal evidence that alcohol is associated with risky sex (e.g. unprotected sex, multiple
sex worker partners) (e.g. 63, 64), which is in turn a risk factor for STDs (65-69). As
discussed at length above, though, the difficulty lies in teasing out the causal link from a
multitude of simple associational data. On the one hand, there are clear theoretical
reasons for seeing this link as ‘genuine’, given the pathways through which drinkers
themselves suggest alcohol genuinely affects sexual risk taking, including reduced
inhibitions and providing a socially acceptable excuse for not using condoms (70, 71).
This is bolstered by experimental evidence that shows how judgement and risk-taking are
affected by drinking in a laboratory setting (72).
On the other hand, there has been a greater problem in demonstrating a link between
risky drinking and actual STD outcomes. A recent systematic review looking at STDs
(but excluding HIV/AIDS) found that there are substantial problems in saying
confidently that there is a causal role of problem drinking (73). Studies looking at the
dose-response relationship have found inconsistent results, and insufficient effort has
19
The same study also valued stolen property as a cost to society, but this has been removed from the
current comparison as it was felt to count as a ‘transfer’ rather than a ‘loss’.
Page 17 of 38
been given to ruling out alternative explanations of an association. Moreover – and
despite the claims of Cook and Clark – the association between problem drinking and
STDs in these studies is simply inconsistent.
20
In contrast, though, a recent National
Bureau of Economic Research paper found that higher alcohol taxes were associated with
lower gonorrhoea rates among young adults in the US, although the results for HIV rates
are less clear (74).
Overall, the implication for future economic studies is that more attention should be paid
to STDs, particularly where the economic burden of HIV/AIDS is high. Despite the need
for further research, there is at least as much evidence for a relationship between STDs
and alcohol use as there is for lowered productivity at the workplace, and the NIAAA in
the US has concluded that “decreasing alcohol use in people who have HIV or who are
at risk for becoming infected reduces the spread of HIV and the diseases associated with
it” (75). Even if the causal relationship is much less than the associational one – such as
for condom use, where a recent meta-analysis suggested that drinking may only be
important for condom use at first intercourse (76) – this may still lead to a considerable
cost, especially considering the effect this may have on long-term equilibrium points.
Are the costs applicable in developing countries?
Perhaps the most difficult question to answer is whether the costs identified in the
countries in Table 1 will be equally applicable in developing countries, as persuasive
cases can be made for both higher and lower costs being likely. Lower costs may arise
due to limited Government spending on e.g. health systems or crime, or a greater number
of potentially employable people to replace those dying or becoming unemployed due to
the effects of alcohol. It can also be argued that “where the God of Productivity has few
20
Cook and Clark argue that “problem drinking is clearly associated with an increased risk of STDs” (73
p159), given that 8 of 11 studies found a significantly increased risk of at least 1 STD among problem
drinkers compared to non-drinkers. However, this conceals the fact that many studies performed multiple
tests separately (e.g. for men and women, for different measure of alcohol use, or for different STDs) – if
we instead look across the full 22 contrasts among problem drinkers, we find that over half were non-
significant, and that 2 further contrasts were insignificant in multivariate analyses.
Page 18 of 38
followers, the negative impact of alcohol on production will not be perceived as a
problem” (77 p103).
Conversely, additional burdens on health systems in developing countries may not be
able to be absorbed in increased spending, and may instead lead to otherwise preventable
deaths. In parallel, individuals with key skills may be effectively irreplaceable in the
workforce, leading to much more economic damage than in richer countries (78).
Workplace-related accidents are also likely to be more common in developing countries
where ‘the extrusion of drinking from the workplace’ (79) is yet to occur, and a variety of
harms may be more likely in the absence of a well-developed treatment system for
alcohol dependents (assuming such treatment is cost-effective). Finally, and as suggested
by the health cost estimate above, the burden of alcohol on human health is greater in
developing countries. Coming down definitively on either side of this argument is
impossible given the current research base, as it is likely that the net effect involves a
balancing of these factors against each other.
There is, however, a further burden that should be considered, although it could never be
included as a ‘social cost’ in the conventional sense – ‘reduced economic development’
(as the World Health Assembly resolution phrased it). Money spent on alcohol can lead
to household impoverishment (80, 81) and the diversion of money away from investment
in economic development (82); it has also been suggested that the move to more
expensive and often international beverages may lead to worsening nutrition in women
and children (83). Clearly there are also development benefits from increasing and
internationalizing alcohol consumption – for example, in employment, government tax
revenues, and technology transfer (84). Yet these can be overstated (78, 85), especially
considering that modernization of brewing and distilling may lead to a reduction in the
number of jobs as productivity increased. And clearly, the economic burdens discussed
throughout this paper must also be taken into account.
Conclusions
Page 19 of 38
Having presented a review and some tentative estimates of the global economic burden
due to alcohol, followed by a discussion of the problems in such studies, there remains
three final points to be made in this conclusion. The first is to draw attention to the limits
of any discussion on the economic burden so-defined, which can only ever quantify a
fraction of the true ‘cost’ of alcohol-related harm. This becomes especially clear when
looking at the ‘cost savings’ from people who die prematurely, which has resulted in a
claim by Philip Morris that tobacco deaths in the Czech Republic save the country money
(86). Such a morally disastrous conclusion can be avoided by putting a value on pain,
suffering and human life itself, although the exact value of these ‘intangible costs’ will
always be contentious. Even so, the various values that have been placed on healthy
human life suggest that these intangible costs are between one- and seven-times the value
of the ‘real money’ costs in the EU (see 9). In other words, there is more to health than
economic arguments alone.
21
Second, few of these studies have evaluated the benefits of alcohol. We can get an idea
of the likely size of these for health costs, using a study in Switzerland that explicitly
compared different methods and alternative scenarios (38). This showed that the health
costs were about 30% lower if health benefits are taken into account relative to a situation
of no consumption – but that the net costs relative to light drinking were only 11% lower
than the gross costs relative to no drinking at all (a much more sensible definition of ‘the
full net social cost’). No studies have quantified other social benefits of alcohol (see 9),
although it should be noted that the size of the alcoholic drinks industry is not an estimate
21
As an aside, it should be noted that ‘health economics’ as a discipline deals substantially with matters
other than production losses, such as Quality of Life and leisure time. The difference between public health
and economic approaches is therefore merely a matter of whether money is used as a metric that enables
otherwise incommensurate areas (work impairment, premature mortality et al) to be combined in analyses.
However, ‘economics’ in popular and policy usage tends to refer to the narrower set of concerns that have
here been labelled ‘economic arguments’ in this article, as seen in e.g. the economic impact assessment for
the prospective European Commission Communication on alcohol (RAND Europe, to be published later in
2006). Many thanks to an anonymous reviewer for stressing the need to clarify this point.
Page 20 of 38
of the social benefit of alcohol, as many of the resources employed in making and selling
these drinks would be used for other purposes (17, 42, 87).
Finally, the main thrust of this article has been to show that alcohol exerts a substantial
economic burden, and that this burden is likely to apply (to some extent) across the
world. This in itself is an important adjunct to the ongoing global debate about acting on
alcohol, although it must clearly take a ‘back seat’ compared to health and welfare
justifications for reducing alcohol-related harm. Yet it can only be a first step in
economic contributions to policymaking, as further research should quantify the share of
costs that can be avoided, the policy investments that should be made to avoid them, and
should then monitor how effective the policies have been in reducing costs (88). In this
vein, the WHO’s CHOICE project (see Chisholm et al, this volume) is an important aid
for policymakers – yet this paper also clearly understates the case for acting on alcohol,
given that it looks only at health outcomes rather than the full burden outlined here.
Further analyses should therefore work towards an elusive goal: a comparative analysis
of the full costs and benefits of different policies that work to reduce the harm done by
psychoactive substances.
Acknowledgements
Many thanks to Christine Godfrey for helpful comments at various stages of the work,
and to Peter Anderson for support during the Alcohol in Europe project (which provided
the foundation for the work in this paper). Thanks also to the anonymous reviewer for
helpful comments for this final version of the paper.
Conflict of interest
I declare that I have no conflict of interest relating to the material presented in this paper.
Page 21 of 38
References
1. Suhrcke, M., McKee, M., Sauto Arce, R., Tsolova, S., and Mortensen, J. 2005. The
contribution of health to the economy in the European Union. European Communities.
2. Single, E., Robson, L., and Xie, X. 1996. The costs of substance abuse in Canada : a
cost estimation study [Full report]. Ottawa, ON: Canadian Centre on Substance Abuse.
3. Midanik, L. T. and Room, R. 2005. "Contributions of Social Science to the Alcohol Field in
an Era of Biomedicalization." Social Science and Medicine, 60(5): 1107-16.
4. Peterson, S., Peto, V., Rayner, M., Leal, J., Luengo-Gernandez, R., and Gray, A. 2005.
European cardiovascular disease statistics (2005 edition). British Heart Foundation
Health Promotion Research Group and Health Economics Research Centre, Department
of Public Health, University of Oxford.
5. The ASPECT Consortium. 2004. Tobacco or health in the European Union: past, present
and future. European Commission.
6. Eurostat. 2004. Statistical analysis of socio-economic costs of accidents at work in the
European Union. Luxembourg: Office for Official Publications of the European
Communities, 2004.
7. Gutjahr, E. and G. Gmel. 2001. "The Social Costs of Alcohol Consumption." Pp.133-44 in
Mapping the Social Consequences of Alcohol Consumption, H. Klingemann and G.
Gmel. Dordecht, the Netherlands: Kluwer Academic Publishers on behalf of WHO-EURO.
8. Andlin-Sobocki, P. and Rehm, J. 2005. "Cost of Addiction in Europe." European Journal
of Neurology, 12(S1): 28-33.
9. Anderson, P. and Baumberg, B. 2006. Alcohol in Europe, a report for the European
Commission.
10. Collins, D. and Lapsley, H. 1991. Estimating the economic costs of drug abuse in
Australia. Canberra: Australian Government Printing Service.
11. WHO. 2004. Global status report on alcohol 2004. Geneva: Department of Mental Health
and Substance Abuse, World Health Organization.
12. Robson, L. and Single, E. 1995. Literature review of studies on the economic costs of
substance abuse. Ottowa: Canadian Centre for Substance Abuse.
Page 22 of 38
13. Jarl, J. and Lyttkens, C-H. 2005. The societal burden of alcohol misuse: literature review
and cost of alcohol related hospitalisation in Skåne, Sweden - 2003. Nationalekonomiska
Institutionen, Ekonomihögskolan vid, Lunds Universitet, Kandidatuppsats 10 poäng.
14. Andlin-Sobocki, P. 2004. "Economic Evidence in Addiction: a Review." European Journal
of Health Economics, 5(S1): S5-S12.
15. Brecht, J. G., Poldrugo, F., and Schädlich, P. K. 1996. "Alcoholism: The Cost of Illness in
the Federal Republic of Germany." PharmacoEconomics, 10(5): 484-93.
16. Aslam, S., Barham, L., Bramley-Harker, E., Dodgson, J., and Spackman, M. 2003.
Alcohol in London: a cost-benefit analysis. London, UK: A Final Report for the Greater
London Authority Prepared by NERA.
17. Single, E., Collins, D., Easton, B., Harwood, H., Lapsley, H., Kopp, P., and Wilson, E.
2001. International Guidelines for Estimating the Costs of Substance Abuse. 2 ed. World
Health Organization.
18. Koopmanschap, M. A. 1998. "Cost-of-Illness Studies: Useful for Health Policy?"
PharmacoEconomics, 14: 143-48.
19. Koopmanschap, M. A., Rutten, F. F. H., van Ineveld, M., and van Roijen, L. 1995. "The
Friction Cost Method for Measuring Indirect Costs of Disease." Journal of Health
Economics, 14: 171-89.
20. Sundhedsministeriet [Ministry of Health]. 1999. "De samfundsøkonomiske konsekvenser
af alkoholforbrug" [The Economic consequences of alcohol consumption" in Denmark].
2nd edition. Copenhagen, Denmark: Ministry of Health.
21. Rehm, J., Baliunas, S., Brochu, S., Fischer, W., Gnam, W., Patra, J., Popova, S.,
Sarnocinska-Hart, A., and Taylor, B. 2006. The Social Costs of Substance Abuse in
Canada 2002. Canadian Centre on Substance Abuse.
22. Tarricone, R. 2006. "Cost-of-Illness Analysis: What Room in Health Economics?" Health
Policy, 77: 51-63.
23. Johannesson, M. and Karlsson, G. 1997. "The Friction Cost Method: A Comment."
Journal of Health Economics, 16: 249-55.
24. Birnbaum, H. 2005. "Friction-Cost Method As an Alternative to the Human-Capital
Method Is Calculating Indirect Costs." PharmacoEconomics, 23(2): 103-4.
Page 23 of 38
25. Godfrey, C. and S. Parrott. 2005. "The Extent of the Problem and the Cost to the
Employer." Pp.21-31 in Addiction at Work: Tackling Drug Use and Misuse in the
Workplace, H. Ghodse.
26. Godfrey, C. 2005. "Economics of Smoke-Free Policies." Going Smoke-free: The medical
case for clean air in the home, at work and in public places, Tobacco Advisory Group of
the Royal College of Physicians. Royal College of Physicians.
27. Godfrey, C. 1991. "Discussion of "The Social Costs of Drinking"." Expert meeting on the
negative social consequences of alcohol use [Oslo, 27-31 August 1990], Norwegian
Ministry of Health and Social Affairs. Oslo, Norway.
28. Collins, D. and Lapsley, H. 2002. Counting the cost: estimates of the social costs of drug
abuse in Australia in 1998-9. Canberra: Australian Government Printing Service.
http://www.health.gov.au/internet/wcms/publishing.nsf/Content/health-pubhlth-publicat-
mono.htm/$FILE/mono49.pdf
29. Easton, B. 1997. The social costs of tobacco use and alcohol misuse. Public Health
Monograph. 2 ed. Wellington, New Zealand: Department of Public Health, Wellington
School of Medicine.
30. Ernst & Young Netherlands. 2006. The contribution made by beer to the European
economy: employment, value added and tax (full report). Amsterdam, Netherlands: Ernst
& Young for the Brewers of Europe.
31. Barker, F. 2002. Consumption externalities and the role of Government: the case of
alcohol. Working Paper 02/25. New Zealand Treasury.
32. Heien, D. M. and Pittman, D. J. 1993. "The External Costs of Alcohol Abuse." Journal of
Studies on Alcohol, 54: 302-7.
33. 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.
34. McMillan, G. P. and Lapham, S. C. 2004. "Does Moderate Alcohol Use Affect Health-
Care Costs? A Propensity Analysis of Female Health-Care Workers." Addiction, 99(5):
612-20.
Page 24 of 38
35. Anzai, Y., Kuriyama, S., Nishino, Y., Takahashi, K., Ohkubo, T., Ohmori, K., Tsubono, Y.,
and Tsuji, I. 2005. "Impact of Alcohol Consumption Upon Medical Care Utilization and
Costs in Men: 4-Year Observation of National Health Insurance Beneficiaries in Japan."
Addiction, 100: 19-27.
36. Rehm, J. 2005. "Selection and Self-Selection: How to Determine the Real Impact of
Alcohol on Health-Care Utilization and Costs? [Commentary on Anzai Et Al]." Addiction,
100(1): 28.
37. Bonneux, L., Barendregt, J. J., Nusselder, W. J., and Van der Maas, P. J. 1998.
"Preventing Fatal Diseases Increases Healthcare Costs: Cause Elimination Life Table
Approach." British Medical Journal, 316: 26-29.
38. Jeanrenaud, C., Priez, F., Pellegrini, S., Chevrou-Severac, H., and Vitale, S. 2003. Le
coût social de l'abus d'alcool en Suisse [The social costs of alcohol in Switzerland].
Neuchâtel, Switzerland: Institut de recherches économiques et régionales, Université de
Neuchâtel.
39. Meltzer, D. 1997. "Accounting for Future Costs in Medical Cost-Effectiveness Analysis."
Journal of Health Economics, 16: 33-64.
40. Rehm, J., R. Room, M. Monteiro, G. Gmel, K. Graham, N. Rehn, C. T. Sempos, U. Frick,
and D. Jernigan. 2004. "Alcohol." Comparative quantification of health risks: Global and
regional burden of disease due to selected major risk factors, M. Ezzati, A. D. Lopez, A.
Rodgers, and C. J. L. Murray. Geneva: WHO.
41. Jones, S., Casswell, S., and Zhang, J-F. 1995. "The Economic Costs of Alcohol-Related
Absenteeism and Reduced Productivity Among the Working Population of New Zealand."
Addiction, 90(11): 1455-61.
42. Leontaridi, R. 2003. Alcohol misuse: How much does it cost? London: Cabinet Office.
43. Lima, E. and Esquerdo, T. 2003. The economic costs of alcohol misuse in Portugal.
Working Paper Series No. 24. Núcleo de Investigação em Microeconomia Aplicada,
Universidade do Minho.
44. Nakamura, K., Tanaka, A., and Takano, T. 1993. "The Social Cost of Alcohol Abuse in
Japan." Journal of Studies on Alcohol, 54(5): 618-25.
Page 25 of 38
45. KPMG. 2001. Excessive alcohol consumption in the Netherlands: trends and social costs.
Hoofddorp: KPMG Economic Consulting.
46. Harwood, H., Fountain, D., and Livermore, G. 1998. The economic costs of alcohol and
drug abuse in the United States – 1992. Report prepared for the National Institute on
Drug Abuse (NIDA) and the National Institute on Alcohol Abuse and Alcoholism (NIAAA),
National Institutes of Health, Department of Health and Human Services (NIH Publication
No. 98-4327). Washington ed. Rockville, MD: National Institutes of Health.
47. Tekin, E. 2004. "Employment, Wages, and Alcohol Consumption in Russia." Southern
Economic Journal, 71(2): 397-417.
48. Zarkin, G. A., French, M. T., Mroz, T., and Bray, J. W. 1998. "Alcohol Use and Wages:
New Results From the National Household Survey on Drug Abuse." Journal of Health
Economics, 17(1): 53-68.
49. Barrett, G. F. 2002. "The Effect of Alcohol Consumption on Earnings." Economic Record,
78(240): 79-96.
50. van Ours, J. C. 2004. "A Pint a Day Raises a Man's Pay; but Smoking Blows That Gain
Away." Journal of Health Economics, 23: 863-86.
51. Dave, D. and Kaestner, R. 2002. "Alcohol Taxes and Labor Market Outcomes." Journal
of Health Economics, 21: 357-71.
52. Peters, B. L. 2004. "Is There a Wage Bonus From Drinking? Unobserved Heterogeneity
Examined." Applied Economics, 36: 2299-315.
53. Cook, P. J. and Moore, M. J. 2002. "The Economics of Alcohol Abuse and Alcohol-
Control Policies." Health Affairs, 21(2): 120-133.
54. Fernandez-Esquer, M. E. 2003. "Drinking for Wages: Alcohol Use Among Cantineras."
Journal of Studies on Alcohol, 64(2): 160-166.
55. Watts, Robert. 26-12-2004. "Advertising is the UK's booziest business." Telegraph.
56. Cook, P. J. 1991. "The Social Costs of Drinking." Expert meeting on the negative social
consequences of alcohol use [Oslo, 27-31 August 1990], Norwegian Ministry of Health
and Social Affairs. Oslo, Norway.
Page 26 of 38
57. Salomaa, J. 1995. "The Costs of the Detrimental Effects of Alcohol Abuse Have Grown
Faster Than Alcohol Consumption in Finland." Addiction, 90(4): 525-37.
58. Miller, T. R., Lestina, D. C., and Spicer, R. S. 1998. "Crash Costs in the United States by
Age, Blood Alcohol Level, Victim Age, and Restraint Use." Accident Analysis and
Prevention, 30(2): 137-50.
59. Miller, T. R. and Blewden, M. 2001. "Costs of Alcohol-Related Crashes: New Zealand
Estimates and Suggested Measures for Use Internationally." Accident Analysis and
Prevention, 33: 783-91.
60. Rice, D. P., Kelman, S., Miller, L. S., and Dunmeyer, S. 1990. The Economic Cost of
Alcohol and Drug Abuse and Mental Illness 1985. Report submitted to the Office of
Financing and Coverage Policy of the Alcohol, Drug Abuse, and Mental Health
Administration. San Francisco: Institute for Health and Aging, University of California.
DHHS Publication No. (ADM) 90-1694.
61. Bergmann, E. and Horch, K. 2002. Kosten alkoholassoziierter Krankheiten: Schätzung für
Deutschland [The cost of alcohol-related illness: estimates for Germany]. Berlin,
Germany: Robert Koch-Institute.
62. Haacker, M. 2006. The Macroeconomics of HIV/AIDS. International Monetary Fund.
63. Krupitsky, E. M., Horton, N. J., Wililams, E. C., Lioznov, D., Kuznetsova, M., Zvartau, E.,
and Samet, J. H. 2005. "Alcohol Use and HIV Risk Behaviors Among HIV Infected
Hospitalized Patients in St. Petersburg, Russia." Drug and Alcohol Dependence, 79(2):
251-56.
64. Madhivanan, P., Hernandez, A., Gogate, A., Stein, E., Gregorich, S., Setia, M., Kumta,
S., Ekstrand, M., Mathur, M., Jerajani, H., and Lindan, C. P. 2005. "Alcohol Use by Men
Is a Risk Factor for the Acquisition of Sexually Transmitted Infections and Human
Immunodeficiency Virus From Female Sex Workers in Mumbai, India." Sexually
Transmitted Diseases, 32(11): 685-90.
65. Kaljee, L. M., Genberg, B. L., Minh, T. T., Tho, L. H., Thoa, L. T. K., and Stanton, B.
2005. "Alcohol Use and HIV Risk Behaviors Among Rural Adolescents in Khanh Hoa
Province Viet Nam." Health Education Research, 20(1): 71-80.
Page 27 of 38
66. Lin, D. H., Li, X. M., Yang, H. M., Fang, X. Y., Stanton, B., Chen, X. G., Abbery, A., and
Liu, H. J. 2005. "Alcohol Intoxication and Sexual Risk Behaviors Among Rural-to-Urban
Migrants in China." Drug and Alcohol Dependence, 79(1): 103-12.
67. Matos, T. D., Robles, R. R., Sahai, H., Colon, H. M., Reyes, J. C., Marrero, C. A.,
Calderon, J. M., and Shepard, E. W. 2004. "HIV Risk Behaviors and Alcohol Intoxication
Among Injection Drug Users in Puerto Rico." Drug and Alcohol Dependence, 76(3): 229-
2345.
68. Simbayi, L. C., Kalichman, S. C., Jooste, S., Mathirti, V., Cain, D., and Cherry, C. 2004.
"Alcohol Use and Sexual Risks for HIV Infection Among Men and Women Receiving
Sexually Transmitted Infection Clinic Services in Cape Town, South Africa." Journal of
Studies on Alcohol, 65(4): 434-42.
69. Kebede, D., Alem, A., Mitike, G., Enquselassie, F., Berhane, F., Abebe, Y., Ayele, R.,
Lemma, W., Assefa, T., and Gebremichael, T. 2005. "Khat and Alcohol Use and Risky
Sex Behaviour Among in-School and Out-of-School Youth in Ethiopa." BMC Public
Health, 5: 109.
70. MacQueen, K. M., Nopkesorn, T., and colleagues. 1996. "Alcohol Consumption, Brothel
Attendance and Condom Use: Normative Expectations Among Thai Military Conscripts."
Medical Anthropology Quarterly, 10: 402-23.
71. WHO. 2005. Alcohol Use and Sexual Risk Behaviour: A Cross-Cultural Study in Eight
Countries. Geneva.: WHO Mental Health: Evidence and Research.
72. Maisto, S. A., Carey, M. P., Carey, K. B., Gordon, C. M., and Schum, J. L. 2004. "Effects
of Alcohol and Expectancies on HIV-Related Risk Perception and Behavioral Skills in
Heterosexual Women." Experimental and Clinical Psycopharmacology, 12(4): 288-97.
73. Cook, R. L. and Clark, D. B. 2005. "Is There an Association Between Alcohol
Consumption and Sexually Transmitted Diseases? A Systematic Review." Sexually
Transmitted Diseases, 32(3): 156-64.
74. Grossman, M., Kaestner, R., and Markowitz, S. 2004. An Investigation of the effects of
alcohol policies on youth STDs. NBER Working Paper No. 10949.
75. National Institue on Alcohol Abuse and Alcoholism (NIAAA). 2002. "Alcohol and
HIV/AIDS." Alcohol Alert, 57.
Page 28 of 38
76. Leigh, B. C. 2002. "Alcohol and Condom Use: a Meta Analysis of Event-Level Studies."
Sexually Transmitted Diseases, 29: 476-82.
77. Morawski, J., J. Moskalewicz, and I. Wald. 1991. "Economic Costs of Alcohol Abuse,
With Special Emphasis on Productivity." Expert meeting on the negative social
consequences of alcohol use [Oslo, 27-31 August 1990], Norwegian Ministry of Health
and Social Affairs. Oslo, Norway.
78. Curry, R. L. 1993. "Beverage Alcohol As a Constraint to Development in the Third World."
International Journal of the Addictions, 28: 1227-42.
79. Room, R. and Bullock, S. 2002. "Can Alcohol Expectancies and Attributions Explain
Western Europe's North-South Gradient in Alcohol's Role in Violence?" Contemporary
Drug Problems, 69(3): 619-48.
80. Saxena, S. 1999. "India." Alcohol and public health in 8 developing countries,
WHO/HSC/SAB/99.9 ed. L. Riley and M. Marshall. Substance abuse department, social
change and mental health, WHO Geneva.
81. Lightwood, J., D. Collins, H. Lapsley, and T. Novotny. 2000. "Estimating the Costs of
Tobacco Use." Tobacco control in developing countries, P. Jha and F. J. Chaloupka.
OUP for the World Bank and World Health Organization.
82. Marshall, M. 1999. "Papua New Guinea." Alcohol and public health in 8 developing
countries, WHO/HSC/SAB/99.9 ed. L. Riley and M. Marshall. Substance abuse
department, social change and mental health, WHO Geneva.
83. Vanesterik, P. and Greer, J. 1985. "Beer Consumption and Third World Nutrition." Food
Policy, 10(1): 11-13.
84. World Bank Group. 2000. World Bank Group note on alcohol beverages.
85. Curry, R. L. 1987. "A Framework for National Alcohol Programmes in Developing
Countries." British Journal of Addiction, 82: 721-26.
86. BBC News Online. 2001. "Smoking is cost-effective, says report [17 July]." BBC News.
87. Lehto, J. 1995. The economics of alcohol policy. WHO Regional Publications, European
Series, No. 61. WHO-EURO.
Page 29 of 38
88. Collins, D. and Lapsley, H. 2000. Economic evaluation of policies and programmes:
further uses of estimates of the social costs of substance abuse. Vienna: United Nations
International Drug Control Programme.
89. Collins, D. and Lapsley, H. 1996. The social costs of drug abuse in Australia in 1988 and
1992. Canberra: Australian Government Printing Service.
http://www.health.gov.au/internet/wcms/Publishing.nsf/Content/health-pubhlth-publicat-
document-mono30-cnt.htm/$FILE/mono30.pdf
90. Muizer, A. P., Reinhard, O. P. M., and Rood-Bakker, D. S. 1996. Externaliteiten van
alcohol gebruik: naar een doeltreffend alcoholbeleid [Social costs: externalities of alcohol
consumption, costs and benefits for third parties]. Rotterdam: Netherlands: Netherlands
Economics Institute (NEI) on behalf of the Dutch Foundation for the Responsible Use of
Alcohol (STIVA) [see also the Condensed Summary by Muizer and Rood-Bakker in
English].
91. Pacolet, J., Degreef, T., and Bouten, R. 2004. Analyse des coûts sociaux et bénéfices en
matière de consommation et d'abus d'alcool en Belgique: synthèse [Analysis of the social
costs and benefits due to the consumption and abuse of alcohol in Belgium: executive
summary].
92. Devlin, N. J., Scuffman, P. A., and Bunt, L. J. 1997. "The Social Cost of Alcohol Abuse in
New Zealand." Addiction, 92: 1491-505.
93. Gjelsvik, R. 2004. Utredning av de samfunnsmessige kostnadene relatert til alkohol
[Counting the economic costs caused by alcohol]. Norway: Directorate for Health and
Social Affairs, the Rokkan Centre, and the Norwegian Institute for Alcohol and Drug
Research (SIRUS).
94. Guest, J. and Varney, S. 2001. Alcohol misuse in Scotland: trends and costs. Scottish
Executive.
95. Kopp, P. and Fenoglio, P. 2000. Le coût social des drogues licites (alcool et tabac) et
illicites en France [The social costs of licit (alcohol and tobacco) and illicit drugs in
France]. Étude n° 22 . Paris, France: Observatoire français des drogues et des
toxicomanies.
96. Fenoglio, P., Parel, V., and Kopp, P. 2003. "The Social Cost of Alcohol, Tobacco and
Illicit Drugs in France, 1997." European Addiction Research, 9(1): 18-28.
Page 30 of 38
97. Kozíková, I. E. 1995.
98. Reynaud, M., Gaudin-Colombel, A. F., and Le Pen, C. 2001. "Two Methods of Estimating
Health Costs Linked to Alcoholism in France (With a Note on Social Costs)." Alcohol and
Alcoholism, 36(1): 89-95.
99. Sesok, J. 2003. Indicators of harmful alcohol drinking in Slovenia in the year 2002 [in
Slovenian]. Ljubljana: Institute of Public Health of the Republic of Slovenia.
100. García-Sempere, A. and Portella, E. 2002. "Los Estudios Del Coste Del Alcoholismo:
Marco Conceptual, Limitaciones y Resultados En España [Studies of the Cost of
Alcoholism: Concepts, Limitations and Results From Spain]." Addicciones,
14(Suplemento 1): 141-53.
101. Byrne, S. 2000. The cost of alcohol-related problems in Ireland. Dublin Institute of
Technology.
102. Johnson, A. 2000. Hur mycket kostar supen? [How much for the booze?]. Stockholm:
Sober.
103. Collicelli, C. 1996. "Income From Alcohol and the Costs of Alcoholism: an Italian
Experience." Alcologia, 8(2): 135-43.
104. Harwood, H., Fountain, D., and Livermore, G. 1999. "Cost Estimates for Alcohol and
Drug Abuse." Addiction, 94(5): 631-34.
105. Baltic Data House. 2001. Economic research about the influence of taxes imposed on
alcohol and influence of price policy on alcohol consumption; research about accidents
caused by alcohol intoxication. Riga, LV: Market and Social Research Group, Baltic Data
House, Baltic Institute of Social Sciences.
106. Godfrey, C. 2004. The Financial Costs and Benefits of Alcohol
107. Maynard, A., Godfrey, C., and Hardman, G. 1994. Conceptual Issues in Estimating the
Social Costs of Alcohol 11-5-1994),
108. Bray, J. W. 2005. "Alcohol Use, Human Capital, and Wages." Journal of Labor
Economics, 23(2): 279-312.
109. Hensing, G. and Wahlstrom, R. 2004. "Chapter 7. Sickness Absence and Psychiatric
Disorders." Scandinavian Journal of Public Health, 32(Suppl 63): 152-80.
Page 31 of 38
Page 32 of 38
Table 1 – Overall comparison of social cost studies included in the review
Year of Cost
Total tangible
costs
Total cost per
capita (inc.
intangibles)
Year of Cost
Total tangible
costs
Total cost per
capita (inc.
intangibles)
% of
GDP
PP € 2003
†
% of
GDP
PP € 2003
†
Australia (28, 89) 1998/9 0.9–1.0 286–315 * N’lands
(90) 1996 0.3 78
* Belgium (91) 1999 2.4 586 New Zealand (29) 1990 4.7 4289
Canada (2) 1992 0.9–1.3 195–265 New Zealand (92) 1991 1.4–2.4 234–386
Canada (21) 2002 0.7–1.7 180-451
‡
New Zealand(59) 1996 - -
Denmark 1996 0.9 218 Norway (93) 2001 1.2–2.1 447–729
Eng. & Wales (42) 2001 1.5–1.7 456–497 Portugal (43) 1995 0.5 73
Finland (57) 1990 1.3–1.8 482–823 Scotland (94) 2001/
2
0.7 296–360
France (95, 96) 1997 1.2–1.4 256–300 Slovak R. (97) 1994 3.1 292
France (98)
1996 - - Slovenia (99) 2002 0.3 50
Germany (61) 1995 1.1 253 Spain (100) 1998 0.7 129
Ireland (101) 2003 1.6 447 Sweden (102) 1998 5.5 1,194
Italy (103) 1994 0.7–0.8 134–153 Switzerland (38) 1998 0.5–0.7 435–482
Japan (44) 1993 1.9 381 USA (46, 104) 1992 2.3 666–731
Latvia (105) 1999 1.8 113 USA (60) 1985 1.7 447
N’lands (45)
2000 0.7 171
‡
USA (58) 1995 - -
Figures may differ from reported headline figures as ranges may be taken from sensitivity analyse; * =
Industry-funded study;
† =
Total cost is inflated to 2003 prices and adjusted for purchasing power;
‡ =
DUI costs only
Page 33 of 38
Table 2 – Review of health costs (selection of most methodologically sound studies)
Cost
% of total
health spend
Accidents,
homicide
& suicide
1
AAF
method
2
Treatment
separate
3
Canada (2)
1.3–1.6 Y Y Y
Canada (21)
2.3 Y Y Y
Denmark
3.4 Y
England &Wales (27)
2.8–3.3 Y Y Y
Finland (29)
0.9–1.4 Y Y
France (31, 32)
2.4 Y
Germany (36)
2.3 Y Y Y
Italy (40)
1.7–1.9 Y
Japan (42)
4.9
Latvia (105)
Y
Netherlands (46)
0.3 Y
* Netherlands (21)
0.7 Y
New Zealand (24)
1.3 Y Y
Norway (26)
0.7–1.3 Y Y
Portugal (28)
0.5 Y
Scotland (30)
1.4 Y Y
Spain (37)
2.4 Y
Switzerland (41)
1.4 Y Y Y
USA (4, 43)
1.5 Y Y Y
1
Indicates whether studies included accidents, homicide and suicide (studies omitting these will produce
noticeably lower results).
2
‘AAF method’ = use of the Alcohol Attributable Fraction (AAF) method (17), where international
epidemiological evidence on relative risk is combined with national surveys on prevalence. Those studies
not using this method either restrict themselves to conditions defined as attributable to alcohol, or are little
more than ‘guesstimates’.
3
‘Treatment separate’ indicates whether the study values specialist addiction treatment separately to the
main figure. Often it is unclear whether these costs are incorporated within the headline health cost, or
whether there was insufficient data to evaluate these.
Page 34 of 38
Table 3 – Review of premature mortality costs using the human capital method
Cost (‰ (per
thousand) of GDP Details of method
Workplace
Non-
workplace
Discount
rate (%)
1
Gross or
net costs
AAF
method
Suicide and
accidents
Canada (2)
2.4-2.6 10 Gross Y Y
3.9 6 Gross Y Y
5.4 4 Gross Y Y
Canada (21)
1.6 5 Gross Y Y
Denmark (20)
1.7 - 3.25 Gross Y Y
England & Wales
(Leontaridi}
2.6-2.8 - 3.5* Gross Y Y
Finland (57)
6.5 - 10 Gross Y
10.3 - 4 Gross Y
France (96)
4.4-6.5 - 6 Gross Y Y
Germany (61)
3.7 0.2 2 Gross Y Y
New Zealand (92)
0.3 - 10 Gross Y Y
0.6 - 5 Gross Y Y
Norway (93)
1.6-9.3 - 3.5* Gross Y
Switzerland (38)
1.4 0.8 6 Net Y Y
2.0 1.2 2 Net Y Y
2.6 1.6 0 Net Y Y
US (46)
4.9 6 Gross Y Y
6.3 5 Gross Y Y
7.2 3 Gross Y Y
1
The discount rate is a way of turning future costs into present-day values (as the same amount of money is
more valuable now than in 10 years). The WHO’s Guidelines (17) suggested that all studies should include
estimates using rates of 5% and 10% to aid comparison; only studies with rates of 4-6% are highlighted in
bold to aid comparison.
* Reduced to 3% after 30 years.
Page 35 of 38
Table 4 – Review of absenteeism costs
Cost (‰ (per
thousand) of
GDP
Method
1
Groups/
conditions
included
Australia
0.1 Hospitalization See discussion of method in
text (demographic model)
England & Wales (27)
1.4
Employee survey
Alcohol dependents
2.0 All drinkers (injury only)
and alcohol dependents
France (31, 32)
0.4-0.5 Hospitalization Excludes accidents and
suicide (see Table 2)
Germany (36)
0.8 Hospitalization & treatment
Long-term absences directly
attributable to alcohol, plus
occupational disability
Italy (40)
1.7-1.8 Hospitalization ‘Alcoholics’ plus all in
alcohol-related accidents
Japan (42)
0.1 Hospitalization & treatment No further detail
New Zealand (92)
0.2 Hospitalization
Assumes alcohol abusers
will be 25% less efficient
than general population
Norway (26)
1.0-1.1 Employee survey No further detail
Slovenia
0.1 Hospitalization No further detail
1
‘Hospitalization’ refers to estimates based on the number of days spent in hospital for alcohol-attributable
conditions as a lower bound for the total alcohol-attributable absenteeism. ‘Employee survey’ refers to
primary research on how levels of absenteeism link to alcohol use disorders in the individual.
Page 36 of 38
Table 5 – Review of unemployment costs (all transparent studies)
Cost (‰ (per
thousand) of
GDP
Excess
unemployment
in which
group?
Other details
England & Wales (27)
2.0 Alcohol abusers and alcohol
dependents – men only Based on MacDonald and
Shields (2004
2.4 Alcohol abusers and alcohol
dependents – both genders
Netherlands (45)
2.4 ‘Problem drinkers’ Assumes ½ of this is due to
confounding
Norway (93)
0.6 Those suffering from
alcohol use disorders
Based on MacDonald and
Shields (2004)
Scotland (94)
1.3 Alcohol dependents
Uses national data on
employment rates in alcohol
dependents v. others
Switzerland (38)
0.6 Heavy alcohol users (>4
glasses/day f, >6 glass m)
USA (46)
0.0
1
Alcohol abusers and alcohol
dependents
Multivariate regression
analysis
1
Harwood et al did attempt to estimate the effect of alcohol use disorders on unemployment, but found no
influence of any of their alcohol measures on any of their employment measures.
Page 37 of 38
Table 6 – Review of crime costs
Cost (% of total
‘public order &
safety’ spend)
Cost (‰ (per
thousand) of GD
Criminal damage
(‰ (per thousand) of
GDP
Crimes covered
Source data
Control for
causality
1
Australia
8.4-15.9 1.4-2.7 - All Arrestee and
prisoner surveys
Arrestee/prisoner
attributions
Belgium (91)
1.4 0.2 - * * *
Canada (2)
10.5-12.3 1.9-2.3 - Violence and
property offences,
alcohol-defined
crime
? ?
Canada (21)
14.4 2.7 - Arrestee and
prisoner surveys
Arrestee/prisoner
attributions
England &Wales
(27)
10.6 2.1 1.6 All
Victim surveys,
supplemented by
arrestee research
None
Finland (29)
12.7-13.9 1.7-1.9 0.6 All Unclear Unclear
Netherlands (46)
1.4 0.2 0.7 * * *
* Netherlands (21)
3.7 0.2 0.7 * * *
New Zealand (24)
25.0 –
51.6 3.2-6.5 - All Brown (1986) Unclear
Norway (26)
1.9 0.2 - All Unclear None
Scotland (30)
14.4 3.0 - All
Flat rate for all
offences from
arrestee survey
None
USA (60)
9.2 0.9 0.4 Violence and
property offences,
alcohol-defined
crime
Harwood et al
(1984)
Interviews with
prisoners
USA (4, 43)
6.4 1.0 -
Update of
Harwood et al
(1984)
Interviews with
prisoners
* These three estimates are based on a single source (90), which appears to produce implausible estimates
when compared to other research in the Netherlands (see discussion in 9).
1
Some studies reduce the figure of association between alcohol and crime by e.g. asking prisoners whether
they feel they committed a crime because they were drunk (see discussion in 9).
Page 38 of 38
Table 7 – Review of drink-driving costs
1
Cost
(‰ (per
thou) of GDP
Australia
1.2
Belgium
6.2
Canada (2)
0.7
Canada (21)
0.7
Denmark
1.2
France (31, 32)
1.4-2.1
Germany (36)
0.4
Japan (42)
<0.1
Latvia (105)
1.7
New Zealand (59)
1.7
Norway (26)
<0.1
Portugal (28)
0.7
Switzerland (41)
0.3
USA (60)
0.6
USA (58)
1.0
USA (4, 43)
1.2
1
Studies typically referred to other cost estimates for all road-traffic accidents, of which a proportion was
attributed to alcohol. Given that this made detailed methodological comparison problematic, only the two
studies (both conducted by the same lead author) that focused entirely on drink-driving (and are also the
most transparent and sophisticated) are highlighted in bold.
