The Alcohol (Minimum Price Per Unit) (Scotland) Order 2018 - Competition Assessment
1. This competition assessment analyses the likely economic impact of introducing a minimum price per unit of alcohol of 50p on the competitive ability of producers and retailers and the consequential impact on consumers.
Definition of competition
2. Competition is a process of rivalry between firms seeking to win customers' business. Effective competition encourages firms to deliver benefits to customers in terms of prices, quality and choice. Competition also provides strong incentives for firms to innovate and to improve productivity  . Where levels of rivalry are reduced (say because a proposal restricts the number of firms active in any market) consumers have less choice because they have fewer firms from which they can buy goods or services.
3. Firms compete for market share using both price and non-price competition. Competition between firms may focus on offering the lowest price, particularly where the product is standardised (either because of the characteristics of the product in question, or because of regulation). Most suppliers will try and compete in a number of ways in addition to price, including developing new 'improved' products, offering products of differing quality or characteristics, branding and advertising the differences in their products relative to their competitors', or using different sales channels.
4. However, left wholly unregulated markets will not necessarily deliver the best outcomes for consumers, companies, or the government  . Government has a legitimate role in intervening and shaping them: it also intervenes more widely to achieve other policy goals and correct market failures  .
Definition of markets
5. Markets and sectors which could potentially be affected both directly (downstream) and indirectly (upstream) have been identified and are listed below.
Directly affected markets/sectors (downstream):
- Sales of alcohol on off-licensed premises
- Sales of alcohol in licensed premises (on-trade)
- Market flows between on and off-licensed sales
- Sales of other products by retailers which sell alcohol, including footfall
- Consumers ability to access low cost products.
Indirectly affected sectors (upstream) include:
- Raw material suppliers
Overview of the Scottish drinks industry
6. The structure of the Scottish alcohol industry is complex. On the manufacturing side, broadly reflecting the global market, multinational companies producing multiple products for different worldwide markets dominate; and there are then a large number of smaller producers. These firms, in turn, use a large number of smaller firms, from Scotland or abroad, to supply the required inputs for the production process and in some cases may subcontract out part of the production process, such as bottling, to other firms.
7. Although global brands tend to dominate in terms of spirits and beer production, in common with the rest of the UK, there is growing interest in both craft distilling and brewing. Between 2010 and 2014, 73 new spirit distilleries opened in the UK with 56 between 2012-14  . UK gin brands more than doubled during this four year period with Scotland responsible for 70% of the UK’s gin production. This trend also applies to the Scotch Whisky industry with 14 distilleries starting production since 2013 and a further 8 were planned to open in 2017. There were also up to 40 new distilleries at various stages of planning and development across Scotland in 2017  . Similarly there has been an upsurge in the number of craft breweries in Scotland, with the Campaign for Real Ale ( CAMRA) listing over 80  .
8. The alcohol retail sector (off-sales) consists of a small number of large supermarkets, a decreasing number of smaller specialist retailers and a large number of other small grocers and convenience stores. In Scotland, in 2016, 73% of sales were through the off-trade. The hospitality sector (on-sales) consists of a small number of national chains and a large number of small pubs, clubs and restaurants. While previously there have been a large number of independent pubs, these are increasingly being taken over by mostly large beer producers. Although in Scotland there remain a higher proportion of independent free trade pubs, compared to England  . The retail sector and the hospitality sector sell products produced both within and outside Scotland  .
9. Table 1 provides data on the Scottish spirits sector which is an important industry for Scotland representing 10% of turnover and 16% of Gross Value Added ( GVA)  in Scottish manufacturing as a whole in 2015. The Scottish share of the spirits sector accounted for the vast majority of UK output (86% of turnover and 91% of GVA)  . This is in contrast to the whole of manufacturing where Scotland accounts for only around 7% to 8% of total UK output.
|Totals||As % of UK spirits sector|
|No. of units||Employment (000s)||Turnover (£m)||GVA (£m)||Turnover(%)||GVA(%)|
10. In addition the average GVA per employee (£254,900) was over 3 and a half times greater than in the manufacturing sector as a whole (£68,800). In terms of the wider sector (Manufacturing of Beverages SIC 11.01-11.07), as shown in Table 2, it contributes by far the most in all measures.
Table 2: manufacture of beverages, Scotland 2015: sector split 
|% of beverages sector|
|Distilling, refining & blending of spirits||58||73||78||89|
|Manufacture of cider, fruit wines* & malt||4||3||4||2|
|Manufacture of beer||30||11||9||3|
|Manufacture of soft drinks #||8||13||10||6|
* none from grapes # includes mineral and bottled water
11. 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 hugely reliant on the strength and long-term confidence of the Scotch Whisky industry. In 2015/16, it is estimated that Scottish grown barley supplied 80 – 90% of the demand from the Scotch Whisky industry and Scottish produced malt around 60% of the needs of the Scotch Whisky sector  .
12. In 2015, the largest industry for exports in Scotland was the manufacture of food and beverages, worth £4.8 billion, 16.8% of all international exports. The majority of this sector continues to be whisky exports which accounted for 80% (£3.8 billion) in 2015  , although this is a slightly lower proportion than in previous years. Approximately 93% of all Scotch Whisky production is exported  .
13. It is not possible to estimate the exact market shares of the biggest companies and brands for Scotland, however by the industry’s own admission, in 2009, the top 10 selling Scotch Whisky brands on the Scottish market by volume were estimated to account for approximately 70% of the market with own/ private label and low cost brands accounting for the remaining 30% of the market  . Based on 2015 market value, Forbes estimate that Diageo dominates whisky production with a 36% share; followed by Pernod Ricard with 19% and William Grant & Sons with 7%  .
14. The number of breweries in Scotland now exceeds 100  although the majority of these are small, employing just a few full time employees. This growth in the craft beer sector has driven growth in sales of ale, up 8.9% in 2016  . Despite this, ales remain a small part of the overall market which remains dominated by lager products (over 87% of sales in 2016  ). The most popular brands in terms of retail sales, remain those produced by multinational companies: Tennents (owned by C&C Group): Stella Artois and Budweiser (both owned by Anheuser Busch Inbev SA/NV).
15. Tennents is brewed in Scotland for both the domestic and export markets. Stella is also brewed, in Scotland, by Tennents on behalf of AB InBev  . Budweiser is brewed elsewhere in the UK but distributed by C&C on behalf of AB InBev.
16. Analysis provided by NHS Health Scotland using Nielsen data  found that in 2016, 10.5L of pure alcohol (1,049 units) were sold per adult (aged over 16) in Scotland (20.2 units per adult per week). The volume of pure alcohol sold per adult in Scotland increased through the 1990s and early 2000s, stabilised between 2005 and 2009, and then declined until 2013. It then increased for 2 years before returning to a similar level as in 2013. Comparable sales in England and Wales were 9.0L in 2016 (17.3 units per adult per week), meaning sales in Scotland are 17% higher than in England and Wales  .
17. The 2016 industry sales data published by NHS Health Scotland further indicates that 46.9 million litres of pure alcohol were sold in Scotland that year. Analysis shows that 73% (34.3 million litres) of the total volume of pure alcohol sold was sold through the off-trade compared with just 27% (12.6 million litres) through the on-trade. The majority of spirits (80%), wine (83%) and cider (74%) were sold off-trade. Beer was the only category of drink for which the majority of alcohol was sold through the on-trade (54%).
18. The Nielsen data published by NHS Health Scotland estimates that the average price per unit of alcohol in Scotland in 2016 across the whole market was £0.87, almost identical to the figure in England and Wales (£0.89)  . The average price per unit in the on-trade in 2016 in Scotland was £1.79, compared to just 53p in the off-trade.
19. Figure 1 displays trends in average price, broken down by on-trade and off-trade, for both Scotland, and England & Wales.
Figure 1: Price per unit of alcohol in Scotland and England & Wales, by trade sector, 1994-2016 
20. The chart show that whilst on-trade prices have risen steadily since 2000, off-trade prices, with the exception of a slight rise between 2007 and 2013, have remained relatively stable. As a result the affordability of alcohol has risen. Alcohol sold in the UK is 60% more affordable in 2015 than it was in 1980  .
21. Average prices are one indicator of the price level in the market but are not sufficient to allow an assessment of the likely impact of the introduction of a minimum price. Data on the distribution of prices (expressed as the price of a unit of pure alcohol) is required. Nielsen data are available for the price distribution across the off-trade sector. This is the sector which will be predominantly affected. The on-trade sector is assessed to be affected only minimally, if at all.
22. Figure 2 shows the price distribution across the off-trade sector, in Scotland, in 2016. As can be seen, just over half the alcohol sold via the off-trade in 2016 (51%) was sold at under 50p per unit.
Figure 2: Price distribution (%) of pure alcohol sold off-trade in Scotland, 2016 
The proportion of alcohol sold under 50p per unit varied with the type of alcohol.
23. An analysis of the cumulative off-sales volume of selected drink types and unit prices based on 2016 Nielsen data is presented in Table 3. The last column in the table (far right) provides market shares for the product categories as a proportion of total off-sales volume. The table demonstrates that the majority of sales for spirits, beer and cider were priced below 50p per unit. A minority of wine sales fell below the 50p per unit threshold.
Table 3: Cumulative volume of off-sales of pure alcohol (volume) by price band and total market share, off-trade, Scotland 2016 
||Price per unit (pence)||<40p||<50p||<60p||<70p||% of total sales volume|
|of which||White/ Strong||97%||100%||100%||100%||1%|
|of which||table wine||9%||33%||69%||85%||28%|
24. Over the period in which this policy has been developed, legislated for, and subject to judicial review, the price distribution illustrated in Figure 2 has been shifting to the right (see Figure 3). As might be expected, as prices have increased the amount sold below any particular price per unit has diminished. In 2008, 81% of all off-trade alcohol was sold at below 50p per unit. Between 2009 and 2013 the percentage declined steadily (e.g. to 73% in 2010). But this decline slowed thereafter with 52% sold under 50p per unit in 2014 and 51% in 2016.
25. The shift to a bimodal distribution is due to the impact of substantial numbers of products clustering around price points e.g. a bottle of spirits ( ABV 37.5%) retailing at £11 is equivalent to 42p per unit; a bottle of wine ( ABV 12.5%) retailing at £5 is equivalent to 53p per unit.
Figure 3: Price distribution (%) of pure alcohol sold off-trade in Scotland, 2009-2016 
26. Since the legislation will be introduced in Scotland only there is some potential for Scottish consumers to purchase alcoholic products in off-licences across the border in England, thereby shifting market demand away from Scottish supply (cross-border effects). The extent to which this might happen is difficult to predict as it will depend on consumers’ willingness to travel for their alcohol purchases and on the scale of the price differential. The products most likely to be affected are high-strength, low price products and potential savings from purchasing these products in England would have to be weighed against the travel and transport costs incurred.
27. The issue is similar to that where countries on either side of a border have different tax regimes for goods. Within the EU, according to Cnossen  , around 12% of the population of the EU live near a border with another member state. Different tax regimes in Scandinavian countries have led to flows of goods across borders. For example, one study analysed sales of alcohol and tobacco in Norway both close to the border with Sweden (where the tax is lower) and further away. Revenue from these products was lower for Norwegian stores near the border but consumers there report higher consumption than those further away. This suggests cross-border shopping by a number of Norwegian households. They also found that measure of externalities were higher near the border. The authors concluded that large tax differentials near borders induce tax avoidance behaviour  . This behaviour was confirmed by 2 other studies  .
28. The issue of cross-border shopping between Northern Ireland and the Republic of Ireland was addressed in a report conducted by the Office of the Revenue Commissioners and the Central Statistics Office for the Irish Department of Finance  . The report noted that the main causes of price differentials between goods in Northern Ireland and the Republic are operating costs, profit margin, taxes and, in particular, the value of Sterling against the Euro (depreciation of around 30% between January and December 2008). These are specific circumstances where it is not just alcohol that is cheaper – people are travelling to do all their shopping. Intertrade Ireland confirm that the main drivers for cross-border shopping are economic factors, such as price differentials and exchange rate fluctuations  . The impact  of fluctuations in exchange rates has been seen again in 2016 when the value of Sterling dropped after the vote to leave the European Union (with the subsequent 10-12 per cent rise in the value of the Euro against Sterling).
29. Another potential consequence of introducing minimum pricing in Scotland only is an increase in internet sales. If the alcohol is despatched from within Scotland, minimum pricing (as it will be a condition of the licence) will apply e.g. weekly grocery shop or local home delivery service. If despatched from outwith Scotland e.g. a wine club based in England, it will not apply. Similar to cross-border shopping, the incentive to buy from outwith Scotland via the internet will be greater the bigger the price differential between the price of alcohol in Scotland and elsewhere, combined with the volume of goods being purchased. At present, from the information available, the Scottish Government believes that the type of alcohol typically bought over the internet, currently, is unlikely to be affected by minimum pricing.
30. However it is acknowledged that the market is diversifying and growing, including the entry of the international internet sales company Amazon . A 2017 report by Profiter  , whilst commenting that alcohol is a much less developed ‘online’ category compared to many others, proportionally, found that more UK consumers are buying alcohol online than in any other market in Europe. The report also noted that Pernod Ricard’s online sales in the UK are 5% compared to 2% globally, fuelled by its collaboration with Amazon, Asda and Tesco to push digital sales. It is worth noting however that Amazon, in 2016, applied for and were granted, premises licences for 2 Scottish distribution centres, bringing them within the scope of the Licensing (Scotland) Act 2005 [377,378] . Nevertheless this remains a market segment which will require careful monitoring.
Impact on retailers, suppliers and wholesalers
31. Guidance produced by the Competition and Markets Authority  recommends the consideration of four key questions in order to discuss whether the legislation on alcohol products would have an impact on competition. Each of these questions is discussed in turn for the proposal of a 50p minimum price for a unit of alcohol.
The four questions are as follows. In any affected market, would the proposals:
1. Directly or indirectly limit the number or range of suppliers?
2. Limit the ability of suppliers to compete?
3. Limit suppliers’ incentives to compete vigorously?
4. Limit the choices and information available to the consumer?
1. Would the proposals directly or indirectly limit the number or range of suppliers?
32. Minimum pricing will not award exclusive rights to supply or restrict procurement processes to a single supplier or restricted group of suppliers. There will also be no direct impact or limitation (quota) on the number of suppliers or retailers as a consequence of any of the proposals.
33. A licensing scheme is already in place for the retail of alcohol in off-sales and on-sales premises. Minimum pricing will affect all off and on-sales licensed premises as it will be a condition of a licence, however, it will not affect the existing licensing scheme or require the introduction of a new licensing scheme.
34. A minimum price will essentially establish a price floor. This could potentially make it harder for firms to enter or exit the market for retailing alcohol if the price floor is binding, i.e. if the free market price for products lies below the price floor. This could prevent low cost producers from using their cost advantage to enter the market. New entrants would no longer be able to attract demand by challenging existing firms on price, and below that price floor would be left with the ability to compete only on non-price factors such as brand, quality, range, advertising, etc. So it may, indirectly, act as a barrier to entry for new firms.
35. Although conversely, for low cost producers, retailers may continue to be attracted to their products. If the low cost of production continues to be reflected in the price charged to the retailer, there will be the potential for increased levels of profit per item.
36. Products that currently retail below the preferred minimum price of 50p per unit will require to raise their price to comply with the legislation. This could result in a number of brands of a similar product retailing at an identical price such as supermarket own/ private label spirits, brands currently associated with a low retail price and those recognised as more premium brands. If there was no price differential it may be that demand for the own/ private label product or value product diminishes leading ultimately to a reduction in the number of suppliers.
37. Minimum pricing may provide an incentive to innovate. One possible effect of minimum pricing could be the introduction of alcohol products containing lower strength alcohol which could be sold at a relatively lower price in larger quantities due to them containing fewer units of alcohol per litre  .
38. The legislation would apply equally to international producers, wholesalers and retailers trying to enter the Scottish market. Any firms wanting to import high strength, low price products would have to raise their retail prices to comply with the minimum price per unit legislation. This could impact on a foreign company’s ability to compete in the domestic market if the company was currently benefitting from low costs of production and selling at very low margins relative to other imports or domestic products.
39. Analysis by the Scottish Government demonstrated that by far the largest share of the impact will be in domestically produced (i.e. from within the UK) products. Whilst we cannot exclude the possibility that there will be both EU and non EU products affected, this is not a disguised restriction on trade. Annex B contains information provided to the Court of Justice of the European Union in 2015, based on the best information available to the Scottish Government at the time, on the extent of the impact in terms of the whole market (i.e. on and off-trade) and the likely source of production for various categories of alcoholic product.
40. The products most likely to be affected, as shown in Table 3, are spirits, beer and cider. The whisky would, by definition, be produced in Scotland; the most popular white spirits (by value) (Smirnoff vodka, Glen’s Vodka and Gordon’s gin  ) remain domestically produced brands; the majority of beer and cider affected is also likely to be domestic production  .
41. The Scottish Government recognises that there are certain imported products which may be disproportionally affected, although they make up a very small part of the overall Scottish market. In 2016, although brandy constitutes only 1% of off-sales, 66% of it was sold under 50p per unit. No cognac, conversely, would be affected by a minimum price of 50p per unit. Analysis by NHS Health Scotland  of wine sales in Scotland in 2013, showed differing amounts sold under 50p per unit by country of origin. The vast majority of imports from Eastern Europe (Bulgaria & Romania) sold at below 50p per unit; their share of the off-trade market was around 0.3%. (The amount retailed below 50p per unit is likely to have diminished as the price distribution has shifted since 2013 as described in paragraph 24).
2. Would the proposal limit the ability of suppliers to compete?
42. Minimum pricing will restrict the ability of retailers to price alcohol products. Since the limitation will act as a price floor, retailers will not be able to out-compete through undercutting one another on price across some or all of their product range or through loss-leading. This could have a weakening effect on competition between retailers.
43. Identifying which part of the retail market will be most affected – supermarkets or small shops – is problematic. Large and small retailers are likely to be affected differently. Larger retailers sell large volumes of popular brands (often priced very competitively) but also, a greater range of products. Convenience stores’ representatives have previously said that they need to maintain low prices to compete with supermarkets, particularly as supermarkets continue to develop their “convenience store” format (such as Tesco Metro and Sainsbury’s Central) putting pressure on independent retailers to compete with them on price. The retail market is continuing to evolve as the importance of the low cost supermarkets (Aldi, Lidl) and their market share, increases.
44. Some small retailers may depend on alcohol sales for a significant proportion of their turnover. The initial consultation response by the Scottish Grocers’ Federation estimated that the imposition of a minimum price of between 40p to 70p per unit could reduce sales by between 10% and 25%  . However, this will have to be weighed up against the additional (off-sales) turnover predicted to be generated.
45. It is very unlikely that the minimum price legislation will force any small retailers out of the market. In the exceptional circumstances where this was the case, there would be a potential competition impact since it could lead to a more consolidated market, and hence less competition between firms even on products where the minimum price floor does not have a direct effect.
46. Table 4 illustrates the potential impact on the price of a selection of products following the introduction of a minimum unit price of 50p. These are examples taken as a snapshot from a comparison website  and represent products at low and medium price range, (including many which are also amongst the most popular brands by value in Scotland 2016) in different drinks categories. The table indicates the minimum unit retail price and those products for which there would be no change.
Table 4: Retail prices of a sample of products from Tesco (on 16 February 2018) and the impact of a 50p minimum price per unit 
|Product||ABV||Units||Prices as at 16 Feb 2018*||Price per unit of alcohol||Minimum price at 50p/unit||Increase (£)|
|Tesco Crofter’s dry cider, 2 litres||5.0||10.0||£2.05||21p||£5.00||£2.95|
|Strongbow dark fruits 4x440ml||4.0||7.0||£4.50||64p||£3.52||Not affected|
|Kopperberg raspberry cider 500mls||4.0||2.0||£2.20||1.10p||£1.00||Not affected|
|Vodka and Gin|
|Tesco Imperial vodka, 70cl||37.5||26.2||£10.00||38p||£13.13||£3.13|
|Glen’s vodka, 70cl||37.5||26.2||£12.50||48p||£13.13||£0.62|
|Smirnoff Red Label, 70cl||37.5||26.2||£14.50||55p||£13.13||Not affected|
|Russian standard vodka 70cl||40.0||28.0||£14.50||70p||£14.00||Not affected|
|Tesco London dry Gin 70cl||37.5||26.2||£11.00||42p||£13.13||£2.13|
|Gordon’s gin, 70cl||37.5||26.2||£14.50||50p||£13.13||£0.13|
|Bombay Sapphire 70cl||40.0||28.0||£21.00||54p||£14,00||Not affected|
|Scots Club 70cl||40.0||28.0||£11.00||39p||£14.00||£3.00|
|Tesco Special Reserve, 70cl||40.0||28.0||£12.50||45p||£14.00||£1.50|
|Bell’s, 70cl||40.0||28.0||£15.00||54p||£14.00||Not affected|
|Whyte and MacKay, 70cl||40.0||28.0||£15.00||54p||£14.00||Not affected|
|Famous Grouse, 70cl||40.0||28.0||£15.00||54p||£14.00||Not affected|
|Glenfiddich single malt 12 yrs, 70cl||40.0||28.0||£35.00||1.25p||£14.00||Not affected|
|Jack Daniels 70cl||43.0||30.1||£26.00||86p||£15.05||Not affected|
|Beer and lager|
|Tesco Lager 4x440ml||3.8||6.7||£2.55||38p||£3.34||£0.79|
|Carlsberg Special Brew 4x440ml||8.0||14.1||£7.60||54p||£7.04||Not affected|
|Tennents Lager 4x 440mls||4.0||7.2||£3.60||50p||£3.52||Not affected|
|Stella Artois 4x568ml||5.0||11.4||£5.10||45p||£5.68||£0.58|
|Tesco everyday value bitter 4x440ml||2.1||3.7||£1.00||27p||£1.85||£0.85|
|Carling 4x440ml||3.7||6.5||£3.60||55p||£3.26||Not affected|
|Wine (75cl bottls)|
|Tesco Spanish white wine||11.0||8.3||£3.65||44p||£4.13||£0.48|
|Tesco Australian Chardonnay||12.0||9.0||£4.15||46p||£4.50||£0.36|
|Echo Falls Pinot Grigio||13.0||9.8||£6.00||62p||£4.88||Not affected|
|Brancott Estate Sauvignon Blanc Marlborough||13.0||9.8||£9.50||96p||£4.88||Not affected|
|Blossom Hill Californian Rose||11.0||8.3||£5.25||94p||£4.13||Not affected|
|Tesco Rioja||13.0||9.8||£5.00||51p||£4.88||Not affected|
|Hardy's Crest Cabernet Shiraz Merlot||14.0||10.5||£7.00||67p||£5.25||Not affected|
|Tesco Cote du Rhone||13.5||10.1||£4.30||42p||£5.06||Not affected|
|Isla Negra Seashore merlot||12.5||9.4||£5.00||53p||£4.69||Not affected|
47. The initial change in the market is likely to be in the quantities sold of a specific alcoholic product if the original price lies below 50p per unit. The change in revenue to retailers and wholesalers will be determined by consumers’ elasticity of demand for that product – the more inelastic the demand, the greater the increase in revenue. This leads to a transfer of ‘rents’ from consumers to retailers. In effect, retailers can charge higher prices for the same goods than they otherwise could under free and unrestricted competitive markets.
48. The Scotch Whisky Association point out that there could be another form of market distortion as a result of obligatory price increases in some of the low price, high strength products. Such an increase would reduce the price gap between low quality products (in this case own/ private label whiskies) and higher quality products such as branded blended whiskies and, to a lesser extent, single malts. The SWA claim this could potentially lead to a ‘commoditisation’ of the market, with consumers expected to switch to alternative, higher quality, but now similarly priced products.
49. An alternative scenario could be a proportionate increase in prices of higher quality products by retailers in order to maintain product differentiation, which would then result in a higher level of prices throughout the alcohol product segment presented to the consumer. Evidence from British Columbia shows that when the minimum price for alcoholic drinks was raised, prices rose across all of the price distribution, including those well above the minimum price. The scale of price increases reduced the higher the original price of the product  .
50. The likely behavioural response to the increase in price is discussed in detail in the section on elasticities (paragraphs 72 & 73). Overall demand for alcohol tends to be inelastic. This mean that an increased price leads to a proportionately smaller decrease in demand and an increase in revenue.
51. The most recent estimates from the Sheffield Model are that, after accounting for duty and VAT, a minimum unit price of 50p will lead to an increase in revenue in the off-trade of around £41m (9.6%) and a marginal decrease in the on-trade of £0.7m (0.7%)  . The change in spend will also impact on Exchequer receipts. It is estimated that there will be a reduction of £15m in duty and VAT.
52. The likely distribution of these increased revenues for the industry across the supply chain is not known. If the majority of profits are retained by retailers, those margins could be used to become more competitive in other areas, e.g. fruit and vegetables. It might lead to loss-leading activities on staple items such as bread and milk. This might put smaller retailers, who would not have the same flexibility of margins, at a competitive disadvantage. If producers raise their prices accordingly following the imposition of a minimum price, this could negate any profit margin increase for retailers  .
53. The Loi Galland, passed in France in 1997, meant that large supermarkets could not pass on discounts negotiated with wholesalers to consumers, the equivalent to allowing industry-wide price floors. Any deals made by retailers with wholesalers would only result in an increase in the retailer's margins, and not benefit consumers. In France, between 1997 and 2002, food prices increased faster than general inflation – 11.8% compared to 6%. Before the Loi Galland food prices increased at a slower rate than inflation  .
54. Similarly, between 1987 and 2005 Ireland's Groceries Act (1987) provided very similar restrictions on retailers' pricing by outlawing below-cost selling in Ireland. Collins et al. (2001)  identified the Act as a key influence on the behaviour of retailers, and as a significant variable in the explanation of retail gross margins. They show a positive relationship between the banning of below-cost selling and retail gross margins, which indicates that the law resulted in a reduction in price competition between retailers. A study by the Irish Competition Authority in 2005  estimated that removing the restriction on below-cost selling could save households nearly €500 per year.
55. An Organisation for Economic Co-operation and Development ( OECD) roundtable in 2005 on resale below cost  further noted that restrictions on selling below cost are associated with slower economic growth and higher unemployment.
56. In some cases, there is a risk that Government-imposed restrictions on pricing could encourage rent-seeking activity e.g. lobbying by firms to maintain or increase restrictions. This could lead retailers to divert resources away from developing and improving their products and services. In the long-run this can result in higher costs.
57. In the case of specialists who sell alcohol products only, there would not be the opportunity to use any increase in revenue to reduce prices of other products such as fruit and vegetables in order to enhance competitiveness. However the aggressive low cost competition between the supermarkets in the off-sales sector is likely to have contributed to the failure of mid-size off-sales chains such as Threshers and Haddows. In terms of lower priced products, a minimum price might increase the ability of independent shops and smaller chains to compete in this market.
Production methods and innovation
58. The producers that will be most affected by a minimum price are those whose production consists of a significant volume of products which currently sell below that minimum price threshold. These producers are likely to be the ones whose main production focuses on own/ private label products, as these generally sell at lower prices.
59. In the case of ciders, some of the cheaper brands are produced by global companies such as Constellation Brands and Heineken which are major drinks companies producing a whole range of alcohol products. These companies are likely to be affected, overall, to a very minimal extent by minimum pricing in Scotland. Some own/ private label cider is produced by Aston Manor, a company with production based in Birmingham whose portfolio includes a range of cider products. 
60. For own/ private label spirits, it appears that there are two companies that are significantly involved in own/ private label whisky production: Whyte & Mackay and Glen Catrine. Whyte & Mackay claim to be a leading supplier of own/ private label whisky for the UK, with an estimated 80% share of that market. Glen Catrine’s website  states it now has the largest independent bottling plant in Scotland. Amongst a multitude of brands, they produce the 5th highest selling whisky in the UK (High Commissioner Scotch Whisky) and the 2nd highest selling vodka in the UK (Glen's Vodka). The strength of these companies in own/ private label spirit production is borne out by the Scotch Whisky Association’s letter of 19 February 2010  to the Finance Committee in which they state that “ while there are a number of companies involved in this trade [cheap or own label] two companies in particular rely heavily on this segment of the market”. The letter goes on to mention Whyte & Mackay and Glen Catrine  .
59. There should be minimal negative impact on innovation or the introduction of new products. New, high-strength products would have to comply with a minimum price, but would not be prevented from being introduced. As mentioned at paragraph 37, there may even be an incentive to innovate. One possible effect of minimum pricing could be the introduction of alcohol products containing lower strength alcohol which could be sold at a relatively lower price in larger quantities due to them containing fewer units of alcohol per litre. This would constitute an introduction of a new product in line with proposed legislation and would not change the characteristics of existing products. However, reducing the alcohol content will not be an option for some products such as Scotch Whisky, where legal definitions dictate that the product has to be of strength of at least 40% or higher  .
61. It is not anticipated that the proposals will limit suppliers' freedoms to organise their own production processes or their choice of organisational form.
63. There is some concern by the industry  that the establishment of minimum price legislation in Scotland sets a precedent which could lead to legislation being introduced in other countries under the protection of a public health rationale. Depending on how these measures were implemented, there could potentially be a detrimental effect on the export segment of Scottish drinks producers, in particular for Scotch Whisky. Scotch Whisky is already subject to a number of imposed duties and restrictions in other countries, so it is difficult to see how minimum pricing introduces a precedent.
3. Would the proposals reduce suppliers’ incentives to compete vigorously?
64. The primary effect of a price floor is to reduce the ability of retailers to compete on price grounds in a certain section of the market. Instead, retailers might switch to competing on other factors, such as customer service, quality, heritage, taste or origin. Some of this could be positive for consumers. However, other forms of competition can be less positive (e.g. competition on advertising). One unintended consequence of the legislation might be an increase in this type of non-price competition facilitated by the increase in revenue and any resultant impact on sales.
65. The previous section (section 3: “limits the ability of suppliers to compete”) established that there could be increases in revenue to retailers following the introduction of a minimum price depending on the elasticity of demand for alcohol. This could remove pressure on retailers to be efficient as it may reduce the ability to compete on price grounds.
66. It is important to ensure that the introduction of a minimum price does not inadvertently allow or encourage competitors to share information on their commercial matters (e.g. future price or demand projections) during the process of setting their price according to the regulations. If this was the case, it could also lead to reduced incentives to compete.
67. Biscourp et al. (2008) found that before the Loi Galland, retail prices were significantly lower in concentrated markets in France but, two years after the enactment of the law, the correlation vanished. This indicates that retail chains were no longer competing fiercely, and consumers would have been losing out. The larger retailers benefited the most in terms of ability to increase prices.
4. Limit the choices and information available to the consumer?
68. A minimum price for a unit of alcohol can be expected to have direct and indirect impacts on consumers. A price floor will lead to price changes for affected products. This means that relative prices of different alcoholic products would change as the minimum price floor would affect some products (whose price would increase), but not others (whose original price was already set above the minimum price per unit).
69. It may limit consumer choice in a particular market segment as the ability to retail alcohol at prices which are cheap relative to the strength of the product will be curtailed. Those who drink most heavily will be most impacted as they are highly likely to buy these products. As shown in Table 3, the volume of alcohol affected will vary with the type of alcohol. (Note, the data does not allow identification of the number of products that will be affected  .)
70. As mentioned in paragraphs 47 and 48, consumer choice may be reduced as, depending on the market response to the imposition of a price floor, products which previously retailed below that may disappear from the market; or they may displace those previously retailing at the new price required. Alternatively, all products may remain in the market with adjustment occurring across a wide range of price points.
71. In terms of pricing information it will be possible for consumers to calculate the minimum price below which a product cannot be sold. It is estimated that the change will result in increased income to the industry via the off-trade. If firms choose to spend this on additional marketing and advertising then consumers could, potentially, have more information about the products that are available.
72. Consumers can be expected to respond to the change in price in either of two ways, either by reducing their consumption of an alcoholic product if the price increases, or by switching to alternative products (substitutes) whose relative price has decreased. The extent to which this happens will depend on consumers’ price responsiveness, i.e. the own-price elasticity ( PED) and cross-price elasticities ( XED) of demand, which will determine change in consumption and switching behaviour. It is not expected that minimum pricing will affect the ease with which customers can switch between competing products.
73. Own-price and cross-price elasticities :
- Own-price elasticity of demand is defined as the measure of responsiveness in the quantity demanded for a commodity as a result of a change in its own price. It is a measure of how consumers react to a change in price.
- If demand for a good is inelastic, a change in the good’s price will invoke a proportionately smaller change in demand for that good (0<PED<1). If the demand for a good is elastic, then a change in price will result in a relatively larger change in quantity demanded (1<PED<∞).
- Elasticities will vary with the level of drinking, and individual’s level of income . Aggregate analysis tends to suggest that heavier drinkers have relatively more inelastic elasticities of demand for alcohol than moderate drinkers, meaning that an overall change in the price of alcohol will cause heavier drinkers to change their consumption behaviour by relatively less than moderate drinkers. However, since heavier drinkers, by definition, consume more in absolute terms, the total quantities of alcohol consumed could change more than for moderate drinkers.
- The Sheffield Model found that heavier drinkers were more responsive to price change. This is because the analysis is based on disaggregated equations rather than aggregated. The model takes into account cross-price impacts which vary in a very complex way between moderate and hazardous/harmful drinkers and across the different drink and price groups of goods.
- Cross-price elasticities of demand ( XED) measure the responsiveness of the demand for one good, to a change in the price of another good. If the XED between two alcohol products is high, this means that consumers would switch easily to an alternative if the price of one product increased.
74. As alcohol is both mind altering and addictive it might be reasonable to suggest alcohol has relatively few substitutes. The PED for alcoholic beverages is therefore likely to be inelastic. Estimates of the PED will vary, however, depending on how the beverage is defined, e.g. it could reasonably be argued the most important substitute products for beer are wine and spirits. As there are relatively few substitute products, it is likely the absolute value of the own-price elasticity of beer is quite low. The same is obviously also true for wine and spirits.
75. The more narrowly defined the market of a product (e.g. alcohol), the greater the flexibility to switch to alternative products, i.e. the greater the elasticity. For any given brand of beer, or beer sub-market category, e.g. imported beer, there are therefore many substitute beer products. As such, it is reasonable to expect the absolute value of the PED for a specific beer brand or beer sub-market category to be relatively high.
76. Estimates of own price elasticities calculated and used in the most recent version of the Sheffield Model for the Scottish Government are shown in Table 5 below. For comparison, examples of price-elasticities from other studies are given in Table 6.
Table 5: own price elasticities for off and on-trade beer, cider, wine, spirits and RTDs in Great Britain 
Table 6: Examples of price elasticities in international studies
|Study||Region||Period/type||Mean own-price elasticities|
|Huang  ( HMRC)(2003)||UK||1970-2002, on-trade||-0.48||-0.75||-1.31|
|1970-2002, off-trade (beer only )||-1.03|
|Fogarty  (2004)||UK||Meta analysis||-0.47||-0.72||-0.76|
|Gallet  (2007)||International||Meta analysis||-0.54|
|Wagenaar  (2009)||International||Meta analysis||-0.51||-0.46||-0.69||-0.8|
|(harmful drinkers only )||(-0.28)|
|Collis, Grayson & Johal  ( HMRC) (2010)||UK||2001-2006, on-trade||-0.77||-0.46||-1.15|
|Sousa J  ( HMRC) (2014)||UK||2007-2012 on-trade||-0.34||-0.24||-1.25|
77. Although there is little consistency in estimates, these tables show that demand for wine and beer is generally inelastic in the UK. Exceptions are on-trade spirits in the HMRC studies; and off-trade cider in the Sheffield Model. (Note that many studies do not split into on and off-trade.) Findings in own-price elasticity for spirits range from very inelastic in the off-trade to elastic in the on-trade. The estimates used in the Sheffield Model are in line with other studies.
78. A possible increase in the price of alcoholic products following the introduction of a minimum price proposal will therefore have different effects on consumption depending on these elasticities. For the more inelastic products, it can be expected that consumers will spend more. For the relatively more elastic products, like off-trade cider, consumers would be expected to reduce their consumption in response to price increases.
79. The own price elasticities in tables 5 and 6 do not take into account switching behaviour. This issue is addressed by the XEDs between different alcoholic products as defined above. The values show both whether products are substitutes or complements and the strength of the relationship. The extent of switching is likely to be limited  .
80. The average increase in consumer spending is estimated to be around £5 per drinker per year. This amount will vary with the level of drinking and income and estimates by level of drinking, and income, and are shown in Table 7.
Table 7: Estimated impact of 50p minimum unit price on consumer spending by drinker and poverty group 
||In poverty||Not in poverty||In poverty||Not in poverty||In poverty||Not in poverty|
|Baseline spending per drinker per year||£230||£378||£1,102||£1,204||£2,484||£2,341|
|Absolute change per drinker per year||£0||£2||£1||£16||-£88||£20|
81. This shows that, on average for the consumer, there is a small impact, particularly if they are moderate drinkers. The largest impact is on those who are most likely to buy the products liable to be affected: those on low incomes who drink at harmful levels.