Friday 5th December 2008

Sheffield report analyses effects of alcohol pricing and promotion policies

Sheffield report analyses effects of alcohol pricing and promotion policies

A University of Sheffield report, commissioned by the Department of Health to help Government Ministers decide future alcohol policy, shows that policies which lead to price increases reduce alcohol consumption and can have significant effects on reducing alcohol-related harm.

The findings, which were published as part of a consultation document by the Department of Health, are the results of the second phase of an independent review. As part of the review the Sheffield researchers examined the potential effects of different pricing policies on patterns of alcohol consumption and hence on the nation’s health, crime, absenteeism in the workplace and unemployment.

To compile the report the Sheffield team analysed over 40 separate policy scenarios, including setting minimum prices per unit of alcohol at different levels and bans on price-based promotions in off licences and supermarkets. The research examines how policies affect alcohol purchasing and consumption by different population groups, including moderate, hazardous, harmful and underage drinkers both in the on-trade (such as pubs, clubs and restaurants) and the off-trade (supermarkets, off licenses) sectors.

The terms moderate, hazardous and harmful drinkers relate to how much people drink, and are the standard definitions used by the Office for National Statistics and others. Moderate drinkers are those who drink within current UK guidelines (they do not regularly drink more than 4 units per day for men or 3 units per day for women). Hazardous drinkers are those who drink above these guidelines, but do not consume more than 50 units per week (men) or 35 units per week (women). These levels are associated with increased long-term risks for health and social harms, and many of these drinkers also put themselves and others at risk during binges. Harmful drinkers are those who drink more than 50 units per week (men) or more than 35 units per week (women) – a level of drinking which has been associated with a high risk of both acute and chronic harms.

Dr Petra Meier from the University of Sheffield’s School of Health and Related Research (ScHARR), who led the study, said: “This is the first study to integrate data on alcohol pricing and purchasing patterns, consumption and harm to answer the question of what would happen if government were to introduce different alcohol pricing policies. The results suggest that policies which increase the price of alcohol can bring significant health and social benefits and lead to considerable financial savings in the NHS, criminal justice system and in the workplace.

“Our results also show that targeting price increases at cheaper types of alcohol would affect harmful and hazardous drinkers far more than moderate drinkers. Of course these heavier drinkers, by definition, buy more alcohol, but detailed analysis of data on purchasing patterns also shows that they tend to buy more of the cheaper beers, wines and spirits. The effects of price increases may incidentally be advantageous for alcohol retailers (both in off-trade and on-trade) because the estimated decrease in sales volume is more than offset by the unit price increase, leading to overall increases in revenue.”

The major findings of the University of Sheffield research are summarised below -

1. What are the policy effects of introducing across-the-board price increases?

Across the board price increases can have a substantial impact on reducing consumption, and consequently harm. Such price increases mean that there is less incentive for switching between different types of alcohol or drinking venues (for example by going to the pub if supermarket alcohol is getting more expensive) than in policies targeting price increases at certain products or market sectors. Pubs and supermarkets are equally affected by a general price increase, although it has been argued that supermarkets may be less likely than pubs to pass on such price rises to consumers.

Across-the-board price increases (covering all products in the on-trade and off-trade) tend to lead to relatively larger reductions in mean consumption for the population compared to other pricing options.

Policies targeting price changes specifically on low-priced products or certain product categories lead to smaller changes in consumption, as they only cover a part of the market.

2. What are the policy effects of an introduction of minimum pricing?

Minimum pricing is a policy which sets a minimum price at which a unit of alcohol can be sold. Price increases are targeted at alcohol that is sold cheaply. Cheaper alcohol tends to be bought more by harmful drinkers than moderate drinkers and studies show that it is also attractive to young people. So a minimum price policy might be seen as beneficial in that it targets the drinkers causing the most harm to both themselves and society whilst having little effect on the spending of adult moderate drinkers.

Approximately 27% of off-trade alcohol consumption is purchased for less than 30p per unit, compared to 9% in the on-trade. 59% of off-trade consumption and 14% of on-trade consumption is purchased for less than 40p per unit.

Increasing levels of minimum pricing show very steep increases in effectiveness. Overall reductions in consumption for 20p, 30p, 40p, 50p, 60p, 70p are: 0.1%, 0.6%, 2.6%, 6.9%, 12.8% and 18.6%.

Minimum prices targeted at particular beverages are less effective than all-product minimum prices.

Differential minimum pricing for on-trade and off-trade leads to more substantial reductions in consumption and harm (for example, pairing a 30p minimum price in the off-trade with an 80p on-trade minimum price gives a reduction in consumption of 2.1% compared to 0.6% for off-trade alone.

3. What are the effects of banning off-trade price promotions – buy one get one free, 10% discounts, etc?

Just over 50% of all alcohol purchased from supermarkets is sold on promotion, although many of the discounts are quite small. Only quite tight restrictions on the level of discount offered would have noticeable policy impacts. For example:

Banning only buy-one get-one free offers has very little effect on consumption and harm.

Bans on discounts only for lower-priced alcohol (less than 30p per unit) are not effective in reducing consumption.

A ban of discounts of greater than 20% (which would prohibit buy-one-get-one-free, buy-two-get-one-free and buy-three-get-one-free) leads to overall harm reductions similar to a 30p minimum price.

A total ban on off-trade discounting is estimated to reduce consumption by 2.8%, although this may only prove effective if retailers were also prevented from responding by simply lowering their non-promotional prices.

4. What would be the effect of banning alcohol advertising?

There is some uncertainty over the mechanisms linking advertising to consumption, and thus it is unclear whether advertising restrictions can be expected to have an immediate effect on consumption. The international evidence suggests that effects of advertising may be cumulative over time, and may work through influencing attitudes and drinking intentions rather than consumption directly. Appropriate UK data are not available and we have been limited to exploratory analyses based on the international literature.

5. What are the savings for each policy the review has looked at relating to health harm?

The general pattern here is that the more restrictive the policy, the greater the harm reduction. Higher minimum prices lead to greater harm reductions, and this goes up steeply – for example, there is relatively little effect for a 20p minimum price, but 30p, 40p, 50p and 60p have increasing effects. Similarly, a ban on just BOGOFs (buy-one-get-one-free) does not affect health harm very much, but banning discounts larger than 10%, or even a total ban on sales promotions in the off-trade lead to substantial estimated harm reductions. For example:

A 40p minimum price gives an estimated reduction of around 41,000 hospital admissions per annum.

The financial value of avoided mortality and morbidity is valued using direct (NHS) costs avoided and also using the quality-adjusted life years (QALY) measure. For both, higher prices lead to increased savings. The annual direct costs savings from a 40p minimum price are estimated at £116m, for a 30p (off-trade)and 80p (on-trade) minimum price pairing £72m, and from a total ban on price promotions £111m, the corresponding quality of life-related savings are estimated at £424m, £254m and £415m.

6. What are the savings for each policy you have looked at relating to crime?

The review takes into account estimated reductions in the absolute number of violent offences, thefts, robbery, criminal damage and other crime and the corresponding savings in prevention, detecting and prosecuting. The researchers also valued the effect of reducing crime on the quality of life of victims.

A minimum price of 30p is estimated to reduce total crimes by around 3,800 per annum whereas a 40p minimum price is estimated to reduce crimes by 16,000 per annum and a 30p off-trade paired with a 80p on-trade minimum price by 68,000 per annum. An off-trade discount ban would lead to an estimated prevention of 14,000 crimes per annum, of which 4,000 are violent offences.

Crime harms are estimated to reduce particularly for 11-18 year-olds as they are disproportionately involved in alcohol-related crime and are affected significantly by targeting price rises at low-priced products.

Crime costs are also estimated to reduce as prices increase. A 30p, 40p and 30p(off-trade)/80p (on-trade) minimum price is estimated to lead to direct cost savings of around £4m, £17m and £65m per annum respectively, whereas the value of gains in quality of life associated with decreased crime is estimated at £4m; £21m and £88m per annum respectively. A ban on price promotions in the off-trade decreases direct crime costs by £18m per annum and the cost of quality of life lost by £25m per annum.

It is important to note that different policies emerge as effective when compared to health harms: discount bans, targeting cheap off-trade alcohol and low minimum pricing options, which influence only the off-trade sector, are all less effective in reducing crime when compared to policies that also affect the on-trade sector. This is because many of the offenders are young males who purchase just over 75% of their alcohol in the on-trade.

7. How would absences from work and unemployment be affected by each of the policies?

Generally, all policy options that target harmful and hazardous drinkers are effective in reducing alcohol-related harm in the workplace.

Unemployment due to alcohol problems occurs mainly in the harmful drinker group and is estimated to reduce as prices increase: e.g. 3,800 avoided unemployment cases per annum for a 30p minimum price versus 12,400 for a 40p minimum price.

Absence reductions occur mainly in the hazardous and harmful drinker groups: e.g. for a 40p minimum price, the 100,000 estimated reduction in days absence per annum includes 35,000 days for hazardous and 55,000 days for harmful drinkers.

The largest financially valued component of harm avoided due to policy changes is in the estimated unemployment reductions, valued at £303m per annum for a 40p minimum price and £173m at a 30p/80p off-trade/on-trade minimum price.

8. What are the overall financial savings that might be expected as a result of different pricing and promotion policies?

The majority of the policies appraised have estimated total reductions in harm valued over £500m and some are valued higher than £5billion over a ten-year period.

FAQS prepared by ScHARR Research Team

9. Following your study what policy would you recommend to the government? Is there a clear winner?

We do not see it as our role to recommend a policy to government. Instead, we have looked at a range of policies, and modelled the likely effects each policy option would have on consumer spending, reducing consumption, and reducing health, crime and workplace harms. A number of effective policy options have been identified and it is now up to government to decide which policy option, if any, they feel strikes the right balance between consumer interests and reduced harm to society.

10. Would supermarkets and pubs lose out on revenue with minimum or higher prices or bans on promotions? Would the Government gain more from duty and VAT?

No, both off-trade and on-trade retail sectors would see increased revenue from price-based policies such as minimum pricing. Policies vary, but typically a 1% price increase is estimated to produce around a 0.5% reduction in volume purchased. Therefore, retailers would sell less volume, but at higher prices, leading to an overall increase in sales value. This effect is seen in supermarkets and off-licenses, and also in pubs, clubs and restaurants. Most price polices have only small effects on revenue for the Exchequer, as duty receipts fall (these are related to volume sold) but VAT receipts rise (related to sales value).

11. Would the suggested strategies penalise moderate drinkers?

Most policy options affect moderate drinkers in a very minor way, simply because they consume only a small amount of alcohol and also because they do not tend to buy as much of the cheap alcohol that is targeted by minimum pricing and promotion bans. Harmful drinkers buy more alcohol and also tend to choose cheap alcohol; therefore these would be most affected.

Changes in spending per drinker are proportionate to the price increase associated with each policy.

As might be expected, those who buy the most alcohol are affected the most, so changes in spending are expected to be greatest for harmful drinkers, with hazardous drinkers somewhat affected and moderate drinkers affected very little. This is particularly so for policies that are targeted at cheap alcohol.

Moderate drinkers would be predicted to spend on average about 11p extra per week if a 40p minimum price was introduced.

12. Who would benefit from the strategies, eg. moderate, hazardous or harmful drinkers, young people?

The policies we have analysed do impact on the different drinker groups in different ways, in terms of both effects on consumption and also benefits in terms of reductions to health, crime and workplace harms.

Harmful drinkers are expected to reduce their absolute consumption most. In the more effective policy options they spend significantly more on their alcohol purchases.

Significant health benefits are estimated for harmful drinkers (particularly deaths avoided), and important health gains also occur in hazardous drinkers. Underage drinkers and young binge drinkers benefit less from health harm reductions because their baseline levels of risk for many alcohol-attributable diseases are low at such young ages.

Moderate drinkers are affected in only very small ways by the policy options examined both in terms of their consumption of alcohol and their spending. However, even though moderate drinkers are, individually, at lower risk of health-related harms, they comprise a sizeable element of the population and so the small changes in their consumption feed through to small changes in risk but a considerable cumulative change in population health.

Compared to health harm reductions, a much larger proportion of the crime-related harm is due to reduced drinking in the underage and 18-to-24-year-old hazardous drinker groups. When estimating policy impacts, crime avoided is due more to consumption reductions in the harmful and hazardous drinking groups than the moderate group.

13. Where has all your data come from?

Most of the modelling is based on a fresh analysis of existing UK data sources on consumption, price promotions and harms. For example – via the General Household Survey – the model includes the patterns of consumption for over 14,000 people in England in 2006; and – via the Expenditure and Food Survey – individual alcohol purchases in England between 2001/02 and 2005/06. Whilst we have checked that our findings are consistent with what is known about alcohol policy internationally, a lot of effort has gone into ensuring that the data underpinning the model are both recent and UK-based.

14. Could you summarise your most important findings?

Pricing policies can be effective in reducing health, crime and employment harm.

Pricing policies can be targeted, so that those who drink within recommended limits are hardly affected and so that very heavy drinkers, who cause by far the most alcohol-related harm, pay the most.

Minimum unit pricing and discount bans could save hundreds of millions of pounds every year in NHS, crime and employment costs.

If policy makers wish to see the greatest impact in terms of crime and accident prevention, through reducing the consumption of 18-24 year old binge drinkers, they need to consider policies that increase the prices of cheaper drinks available in pubs and clubs as well as supermarkets.

Notes to Editors:

The report:

Full Title:

INDEPENDENT REVIEW OF THE EFFECTS OF ALCOHOL PRICING AND PROMOTION: Part B

Modelling the Potential Impact of Pricing and Promotion Policies for Alcohol in England: Results from the Sheffield Alcohol Policy Model Version 1.0

Authorship:

Modelling Team: Alan Brennan, Robin Purshouse, Karl Taylor, Rachid Rafia

Principal Investigator: Petra Meier

Research Team: Andrew Booth, Daragh O’Reilly, Tim Stockwell, Anthea Sutton, Anna Wilkinson, Ruth Wong

For a full copy of the report please visit – http://alcoholreports.blogspot.com/2008/12/independent-review-of-effects-of.html.

The full report of the first phase of the project, a systematic review of the international research evidence on alcohol pricing, promotion, consumption and harm, is published alongside this report, and can be found on ….

For further information or to arrange an interview with one of the researchers please contact: Jenny Wilson, Media Relations Officer on 0114 2225339 or email j.c.wilson@sheffield.ac.uk