Minimum unit pricing of alcohol : final business and regulatory impact assessment

7. Options

7.1. Various options were considered when the Bill was going through the legislative process. A key issue in the legal challenge was whether increasing alcohol duty could achieve the same aims as minimum pricing but be less restrictive on trade. The alcohol duty option is considered in this section for this reason.

‘Increase the tax on alcohol products’ option

7.2. The Scottish Government has always been aware that increasing alcohol duty would be likely to reduce consumption at a population level, but it would be unable to target the consumption that causes the greatest harm in the way that a minimum price can, and so using alcohol taxation alone would not meet the aim of the legislation.

7.3. The Scottish Government commissioned the University of Sheffield to model the impacts of a minimum unit price policy. In total, four reports were published between 2009 and 2016. At the time of the first three University of Sheffield reports to the Scottish Government (2009, 2010, 2012), the Sheffield Model ^[168] was able to compare the impacts of a general increase in price with the introduction of a minimum price, but it was not able to calculate the tax required to replicate the estimated benefits from minimum pricing. By the time the last report for the Scottish Government was commissioned, the model had been developed so it was now possible to disaggregate the impact by income group and to use the model to estimate the level of tax rise required to achieve a similar impact on health harms as the introduction of a minimum price (equivalisation).

7.4. The commissioning of the model to produce this particular output was mainly driven by the continuing need to demonstrate to the courts the differential impacts of minimum pricing and taxation.

7.5. Within the countries of the UK, alcohol products are subject to the application of both excise duty (tax) and value added tax ( VAT). The UK must comply with EU Directives 92/84/ EEC and 92/83/ EEC, which make provision, respectively, for minimum rates of excise duty on alcohol and the structure of the duty regime and the basis on which excise duty is calculated. Even if the UK was not subject to these constraints, there are considerable difficulties in designing a tax system which would have the same impact as a minimum price. Whilst an increase in taxation would also be likely to result in an increase in government revenue, the Scottish Government rejects the use of taxation alone for the reasons discussed in the following paragraphs 7.6 to 7.15.

7.6. There is evidence that across the board taxation increases do not have a targeted effect on the consumption of alcohol of those most at risk of alcohol-related harm ^[169] . This finding is replicated in all the modelling carried out by the University of Sheffield ^[170] . This is because those who drink the most (hazardous and harmful drinkers) consume a disproportionate amount of cheaper products ^[171] . Figure 12 illustrates that heavier drinkers pay less across all beverage types, with moderate drinkers paying noticeably more, on average, for spirits, and harmful drinkers paying considerably less, on average, for cider.

Figure 12: Mean prices paid by beverage type and drinker group ^[172]

7.7. The volume of alcohol sales in Scotland is driven by sales in the off-trade. Recent sales data estimate that almost three quarters (73%) of all pure alcohol sold in Scotland in 2016 was sold through the off-trade, with 42% of sales in large multiples sold on promotion ^[173] . The average price of a unit of alcohol sold through the off-trade in Scotland was 53p per unit and, through the on-trade, £1.79 per unit. Minimum pricing and duty increases apply equally to both the on and off-trade. However, given that substantially more alcohol is consumed in the off-trade than the on-trade and the price of a unit of alcohol is far less in the off-trade, a pricing measure that predominantly affects that sector is likely to be more effective at tackling alcohol harms.

7.8. Figure 13 shows that hazardous and harmful drinkers consume proportionately more of their alcohol in the off-trade than moderate drinkers.

7.9. A straightforward increase in existing duty would affect the whole market, and all drinkers ^[174] , impacting on high price products as well as cheap ones and so would have a proportionately greater effect on moderate drinkers than a minimum price.

Figure 13: proportion of alcohol consumed in on and off-trade by drinker type, drinker group ^[175]

7.10. A tax regime allied to the strength of the product (volumetric taxation) has been suggested as a logical way to apply tax to alcohol products ^[176] and is one of the three dominant tax structures used internationally, along with ad valorem tax (based on the value of the product) and a unitary tax (based on product volume). The ability to do this within the UK is constrained, currently, by the relevant EU Directives. But even where the rate of taxation is already allied to the strength of the product, e.g. spirits, taxation cannot target products that have a low cost of production and retail cheaply relative to their strength. A cheap bottle of white spirits (e.g. vodka) attracts the same tax per unit of alcohol as an expensive bottle of malt whisky. To raise the price of the former, through taxation, would also affect the latter – and all products in between.

7.11. Other products (under the EU Directives) cannot be taxed according to the amount of alcohol within them. For some products, the rate of duty remains the same across a range of alcoholic strength, for example, wines between 8.5% and 15%, and the tax due is calculated on the volume of the product.

7.12. The European Commission has conducted a public consultation ^[177] on the structures of excise duties applied to alcoholic beverages as set out in Directive 92/83/ EEC. The aim of the Directive is to ensure the proper functioning of the European internal market, including the avoidance of distortions of conditions of competition, ensuring the free movement of products in this sector. A report ^[178] from the European Commission to the European Council on the evaluation of the Directive states that “In practice, only a few Member States mentioned health policy objectives in connection with the overall relevance of the provisions; accordingly, no definitive conclusions can be drawn in this area. However, public health considerations should be included in any further process.” The conclusion to be drawn here is that the evaluation of the Directive is not designed with public health considerations in mind. The European Council has asked the European Commission to undertake further studies to inform potential legislative change, and this work is ongoing. The current evaluation of the whole Directive is the first that has taken place since the Directive was introduced in 1992.

7.13. An increase in Value Added Tax would not be able to tackle low cost, high strength products as it is applied to the financial value of the products and would have the greatest impact on the most expensive products, which are not those that the policy aims to target.

7.14. For a tax system to result in increases in the price of low-priced products but not in the price of high-priced products, the rate of tax would have to be higher for low-priced products. A tax increase based on price would distort the market whilst not achieving the desired effect because low-priced, low strength products would increase as much as that of low-priced, relatively high strength products and, depending on the level at which the tax is set, only the price of some alcohol products sold cheaply relative to their strength may increase.

7.15. There is evidence that increases in taxation of alcohol will not necessarily be reflected in the price the consumer pays. A study undertaken in 2014 ^[179] concluded that, for lower cost products, tax increases tended to be “under shifted” (price increases are less than the duty increase) and for high cost products “over shifted” (price increases are higher than the duty increase). The Competition Commission’s paper ^[180] on pricing practices noted that ten grocery retailers (nine of whom operated across Scotland) engaged in below-cost selling to varying extents. The Competition Commission also found that, for most grocery retailers, the majority of below-cost sales relates to two or three product groups, with alcohol being one ^[181] . This suggests that tax increases are sometimes absorbed by the retailer, absorbed by the producer or offset against other products. To the extent that prices are offset, customers are paying more for other groceries to subsidise alcohol consumption. Absorption also means that the level of tax needed to achieve the same reduction in harms as a minimum price per unit of 50p is complex.

7.16. The most recent modelling undertaken by the University of Sheffield for the Scottish Government (April 2016), Model-based appraisal of the comparative impact of Minimum Unit Pricing and taxation policies in Scotland: An adaptation of the Sheffield Alcohol Policy Model version 3 ^[182] , was able to estimate the level of tax increase which would be required to achieve a similar impact on alcohol-related health harms as minimum pricing. To achieve a similar reduction and a similar distribution of harm reduction across drinker and income groups, defined either as the change in alcohol attributable deaths or alcohol attributable hospital admissions, would require unprecedented increases in the rates of duty ^[183] .

‘Introduce a prohibition on sales of alcohol below a minimum price per unit’ option

7.17. This is the option being progressed by the Scottish Government as it is considered to be the most robust option available in meeting our goals of reducing alcohol-related harm in Scotland whilst simultaneously targeting where harm is greatest. In addition, it is the simplest price based option to understand, and enforce, and it is transparent.

7.18. Each of the reports commissioned by the Scottish Government from the University of Sheffield, reporting the results of modelling using the Sheffield Model, have estimated that a 50p minimum price will lead to reductions in alcohol-related harms, including health, crime and employment harms ^[184] , with the greatest health benefits accrued from minimum pricing coming from hazardous and harmful drinkers who disproportionately consume more of the lower cost, high strength products. The most recent modelling has further demonstrated that the greatest impact is on harmful drinkers in poverty, who are also those who experience the greatest harm.

7.19. Table 1 summarises some of the main health benefits estimated from a range of minimum unit prices based on the modelling carried out by the University of Sheffield ^[185] .

Table 1: Example of health benefits from various minimum unit price scenarios

Policy: minimum price per unit	Policy impact on deaths per year (full effect[186])		Policy impact on hospital admissions per year (full effect)
	numbers	%	numbers	%
30p	-13	-0.8%	-255	-0.9%
40p	-49	-3.0%	-919	-3.1%
50p	-121	-7.4%	-2,042	-6.8%
60p	-236	-14.5%	-3,812	-12.8%
70p	-393	-24.2%	-6,315	21.1%

7.20. Table 1 also illustrates that, despite the reduction in the effect of a 50p minimum unit price since 2012, as demonstrated by the shift in the price distribution of off-trade alcohol ( Figure 9), 50p per unit is still estimated to have a significant impact on alcohol-related deaths and hospitalisations.

7.21. As a pricing measure, the key features of minimum unit pricing are:

• A minimum unit price is a measure that is targeted at products priced cheaply relative to their high strength.
• A minimum unit price has the advantage of certainty. It is not open to absorption. It does not encourage cross-subsidisation between different products and product groups. Due to its simplicity, it is easier to understand, measure and enforce.
• A minimum unit price impacts only marginally on moderate drinkers. The Scottish Government recognises that many people in Scotland have a balanced, positive and enjoyable relationship with alcohol.
• Evidence suggests that minimum unit pricing may increase revenue across the drinks industry.

Minimum Unit Price vs. Taxation – additional data

7.22. In response to the ruling from the CJEU, and acknowledging the methodological developments that the Sheffield Model had undergone ^[187] , the Scottish Government commissioned the Sheffield Alcohol Research Group ( SARG) to appraise the potential impact of minimum unit pricing on levels of consumption in different population subgroups defined by level of drinking and income: further, to estimate the increase in taxation required to match the reduction in health harms (in these subgroups) as would be achieved by the full effect of a minimum unit price of 50p. The methodology required to carry out this modelling had not been developed at the time of earlier commissioned reports (2009-12).

7.23. The modelling demonstrated that it required a significant level of tax (duty+ VAT) increase in order to achieve similar total health benefits to that predicted by the model from a minimum unit price of 50p. Moreover, in order to achieve the targeting desired, i.e. a disproportionate impact on those who suffer a disproportionate amount of the harm, would require even higher levels of taxation.

7.24. In terms of a reduction in alcohol-related deaths, Table 2 ^[188] illustrates the increasing level of tax required to achieve that targeted impact.

Table 2: Equivalisation of mortality impacts of taxation increases with a 50p minimum unit price

7.25. The table illustrates the differential impact that minimum pricing has on harmful drinkers. For a 50p per unit minimum price, out of a total of 121 deaths prevented, 71 are from harmful drinkers. A reduction of 121 deaths can be achieved by a tax rise of 27%, but to achieve the targeting on harmful drinkers the rise has to be 36%.

7.26. In addition, although in Table 2 the impact of a 28% tax rise on alcohol-related mortality appears very similar to the profile for a 50p minimum unit price, there are differences in the health conditions from which deaths are averted. A 50p minimum unit price has a greater impact on deaths from alcoholic liver disease, while a 28% tax rise leads to greater reductions in cardiovascular mortality. Liver disease, as already described in paragraph 5.48, and deaths associated with it, are far more prevalent in areas of deprivation.

7.27. It should be noted that duty rises in the UK (which already has very high alcohol duty rates when compared internationally) over the last 20 years have rarely exceeded 5%. For the four years from 2013, some rates actually reduced or were frozen. Only in 2017 was there a duty increase across all products.

7.28. A similar equalisation exercise was carried out using alcohol-related hospital admissions.

Table 3: Estimated impacts of taxation and minimum unit price policies on hospital admission rates by drinker and poverty group ^[189]

		Baseline annual admissions per 100,000 drinkers	Change in annual hospital admissions attributable to alcohol per 100,000 drinkers at full effect
Drinker group	Income group		1.1. 50p MUP	27% tax rise	28% tax rise	36% tax rise	55% tax rise	70% tax rise
Consumption breakdown
All drinkers	All incomes	798	-55	-62	-64	-81	-123	-158
Moderate	All incomes	-100	-5	-12	-12	-16	-25	-32
Hazardous	All incomes	1,839	-84	-103	-108	-138	-217	-281
Harmful	All incomes	7,120	-497	-469	-488	-605	-893	-1,123
Income group breakdown
All drinkers	In poverty	1,689	-180	-108	-113	-144	-218	-278
	1.1. Not in poverty	674	-37	-55	-58	-72	-110	-141
Moderate	In poverty	103	-22	-25	-26	-33	-51	-65
	1.1. Not in poverty	-130	-3	-10	-10	-13	-21	-27
Hazardous	In poverty	4,563	-359	-252	-264	-335	-521	-669
	1.1. Not in poverty	1,539	-54	-87	-91	-116	-184	-238
Harmful	In poverty	11,555	-1,440	-641	-667	-861	-1,253	-1,578
	1.1. Not in poverty	6,454	-356	-443	-462	-567	-839	-1,055

Sectors and groups affected

7.29. The proposal is intended to reduce alcohol-related harm through the reduction in consumption which is anticipated as a response to the rise in price for products currently retailing at below a minimum price of 50p per unit. As such, it has the potential to impact on society as a whole. Harms are not experienced solely by the drinker, but also by family and friends, communities, employers and the economy.

7.30. Minimum pricing will directly impact on consumers of alcohol and those involved in the alcohol industry: producers, distributors and retailers in both the off and on-trade. There will also be effects on those in the public sector responsible for enforcing the proposals such as Licensing Standards Officers, Licensing Boards, and the police. Any change in the volume of alcohol purchased will affect the UK Exchequer in the form of the duty and taxation collected by the UK Government. Alcohol-related harm affects rates of ill health, crime and employment. There are therefore potential savings for the NHS through a change in the number of deaths, hospital admissions for acute and chronic illnesses and primary care consultations for alcohol problems; on the justice system including the police, the criminal justice system and victims of crime; and on the workplace in the form of the number of unemployed due to alcohol misuse and the sickness absence rate. Alcohol misuse is estimated to have cost £2.5 billion to £4.6 billion (with a midpoint estimate of £3.6 billion) in Scotland in 2007. The midpoint estimate of £3.6 billion includes estimates of £870 million in lost productivity, £270 million to the NHS and £730 million in crime costs.

Consumers

7.31. Consumers who currently purchase alcohol priced at less than the minimum price per unit will be directly affected. Although all groups of consumers are predicted to alter their behaviour, modelling has demonstrated that those likely to be most affected are those drinking above UK CMOs guidelines ^[190] . This is supported by the findings from literature that young people and hazardous and harmful drinkers (see paragraph 8.8) are most likely to consume low cost alcohol. Consumers will also be affected by any change in the level of societal harm associated with alcohol.

On and off-sales

7.32. Both on-sales and off-sales premises will be affected by the introduction of a minimum unit price for alcoholic drinks. It is likely to have a greater impact on off-sales premises than on on-sales, as the price of off-sales alcohol is generally lower than the price of on-sales alcohol (see paragraph 5.65).

7.33. In the UK, the retail sector (off-trade) consists of a small number of large supermarkets who dominate alcohol sales, a number of smaller supermarkets, a decreasing number of specialist retailers, and a large number of smaller grocers and convenience stores. The hospitality sector (on-trade) consists of a small number of national chains and a large number of small pubs, clubs and restaurants. Independent pubs are increasingly being taken over by large beer producers ^[191] . In Scotland in 2016, there were 16,704 premises licences in force: 11,593 for the off-trade and 5,110 for the on-trade ^[192] .

Wholesalers

7.34. Although not directly impacted, as the price restriction is imposed on the retail price, their pricing structure may be impacted by the changes in the market. Wholesalers holding premises licences will be impacted in a similar way to the off-trade.

Producers

7.35. Producers of alcoholic drinks sold in the domestic market will be affected as the volume of alcohol purchased is expected to decline. The extent of the impact will depend on a number of factors including the quantity of alcohol they produce and sell that is currently priced at less than the minimum unit price.

7.36. Within Scotland, production consists of a number of multinationals producing a range of products for worldwide markets and a large number of smaller producers. These firms source inputs from a number of smaller firms both in Scotland and abroad.

Production supply chain

7.37. Both spring and winter barley are grown in Scotland and the UK. Spring barley is the dominant barley crop grown in Scotland and production is reliant on the strength and long-term confidence of the Scotch Whisky industry.

Local government

7.38. Local government will be affected as it will be the responsibility of Licensing Standards Officers to ensure compliance with minimum pricing and Licensing Boards to take action against businesses breaching the conditions.

UK Exchequer

7.39. The UK Exchequer will be affected through a change in the level of the duty and VAT collected associated with any changes in the volume and pattern of purchasing of alcohol products.