The avoidable costs of alcohol abuse in Australia and the potential benefits of effective policies to reduce the social costs of alcohol
4.3 The effectiveness of interventions
4.3.1 The basis for the choice of effective policiesThe policy interventions approach to estimating the avoidable costs of alcohol abuse requires the identification of policies which, when implemented in Australia, could produce a reduction in alcohol abuse. The policies identified in this study as being useable in estimating avoidable costs have three characteristics:
- Convincing research evidence is available in the relevant literature to show that the policies have been, or would be, effective in all or part of Australia or in at least one other comparable country.
- The resulting reduction in alcohol consumption and/or alcohol abuse is quantified in the research literature.
- The quantification is in a form which is amenable to translation into a reduction of the Australian alcohol abuse costs, as calculated in Collins and Lapsley (2008a) for the financial year 2004/05.
In summarising the conclusions of the research evidence on the effectiveness of specific policies, we have relied upon the most recent comprehensive literature review on the subject. This review (Anderson & Baumberg, 2006) was undertaken by the United Kingdom Institute of Alcohol Studies in a report produced for the European Commission. For each of the policy interventions which they study, they review the literature, provide a brief summary of their conclusions and provide effectiveness ratings based upon the criteria presented in Table 6 below.
Table 6. Criteria used for effectiveness ratings in Anderson and Baumberg
|Rating||Effectiveness||Breadth of research support||Cost efficiency|
|+||Evidence for limited effectiveness||Only one well-designed study of effectiveness has been completed||Relatively high cost to implement and sustain|
|++||Evidence for moderate effectiveness||From two to four studies of effectiveness have been completed||Moderate cost to implement and sustain|
|+++||Evidence for high degree of effectiveness||Five or more studies of effectiveness have been completed||Low cost to implement and sustain|
Source: Anderson and Baumberg (2006, Table 7.1).
Top of Page
For each of the categories of intervention for which estimates of the benefits are presented below, we provide both a brief summary of the conclusions of the literature and the ratings presented in the Anderson and Baumberg study.
In some cases the research literature demonstrates the effectiveness of individual policies but does not adequately quantify their benefits or does not quantify these benefits in a form which is amenable to translation into a reduction in abuse costs (that is, does not satisfy characteristic 3 above). The present study draws attention to these types of policies even when it is not possible to undertake estimation of their potential economic benefits.
The present authors recognise that some of the policies costed may not prove to be currently politically feasible or acceptable. However, all of the policies considered have been implemented in one or more comparable jurisdictions and, therefore, have been proved to be politically feasible in some comparable overseas jurisdiction or Australian subjurisdiction. Furthermore, what is perceived to be a ‘political reality’ at one period of time may well change significantly at a later time. Nowhere is this fact better illustrated than in the area of Australian tobacco policy where initiatives thought in earlier years to be politically impossible not only have now been successfully implemented but also have gained wide public acceptance.
The objective of the present study is not to recommend the adoption of a particular set of alcohol policies. It is to consider from an economic perspective a range of policies which have been shown to be effective and to indicate, as far as the data allow, the economic benefits likely to flow from the implementation of these policies. The information is intended to assist in the development of evidence-based strategies for the reduction in the social costs of alcohol abuse in Australia.
In choosing intervention policies and estimating their benefits, account should be taken of existing policies. It may well be that a particular Australian intervention policy is already close to world’s best practice (for example, driver blood alcohol testing) and so the potential benefits of the adoption of such proven-effective interventions are lower than they would be in other countries. On the other hand, expansion or refinement of policies already implemented (for example, RBT or alcohol excise taxation) may be able to be implemented at very low marginal (that is, additional) cost, and so may be very cost-effective. It is important also to acknowledge that being already at world’s best practice does not necessarily signify that no policy improvement is possible.
Top of Page
4.3.2 Difficulties in aggregating the economic benefits of individual policiesIn presenting what might be considered to be a ‘menu’ of effective policies, with estimates of the value of benefits available from each policy, there is a difficulty involved in aggregating the effects of these individual policies to yield an estimate of the aggregate benefits. The problem here is that different policies may have an impact on the same problem. Techniques such as regression analysis could in the future be employed to address some of these issues.
For example, later in this report we examine two distinct policies designed to reduce hazardous and harmful alcohol consumption. The first is to increase the level of taxation on alcohol. The second is to ban alcohol advertising and other forms of alcohol promotion. Research studies of the impact of specific policies necessarily imply the standard economic assumption of ceteris paribus (all other things being equal)—that other relevant policies remain unchanged. This is necessary in order to identify the individual impact of the policy under study. Once the estimated benefits of different policies are aggregated, the ceteris paribus assumption no longer holds. These two types of interventions can to some extent be considered to be substitutes, with the result that aggregation of the estimated cost impacts of the individual policies would lead to overestimation, if the interventions were implemented at the same time. It will in general be difficult, in the aggregation process, to adjust for policy interactions. It would be problematic, for example, in analysing policies to reduce alcoholattributable violence to adjust for the overlapping effects of reduced opening hours for licensed premises and an increase in alcohol taxation.
4.3.3 Retaining the protective health benefits of moderate alcohol consumptionIt is generally agreed in the literature that the consumption of alcohol in moderation can provide protective effects against certain medical conditions, although in certain risk categories any level of alcohol consumption is hazardous or harmful. Recent studies (see, for example, Begg et al., 2007) suggest that the protective effects are less than previously estimated. In addition, new draft Australian guidelines for non-hazardous alcohol consumption (National Health and Medical Research Council, 2007) recommend significantly lower levels of consumption than did earlier guidelines. Nevertheless, there is good evidence for the existence of protective effects of moderate alcohol consumption.
It may well be possible to target alcohol interventions in a manner which reduces hazardous and harmful consumption while retaining (and perhaps even augmenting) the protective effects. This would be ideal, and specifically targeted interventions such as more intensive enforcement of random breath testing (RBT) will very probably achieve this objective. However, general, or population, interventions, such as excise tax increases, run the risk of reducing the benefits, as well as the costs, of alcohol consumption.
Top of Page