Minimum unit pricing of alcohol
Andrew Leicester
Independent Scientific Committee on Drugs (ISCD), 2nd February 2012
© Institute for Fiscal Studies
Background
• Minimum unit pricing (MUP) has received considerable support
• Policy is set to be introduced in Scotland this year
– Rate still to be set based on consultation and new evidence
– Legal issues?
• Other policy action towards alcohol pricing
– ‘Below-cost’ ban in April 2012, England and Wales
– Beer tax reform in October 2011 – tax varies with ABV
– 2% real increase in alcohol duty each year to 2014/15
© Institute for Fiscal Studies
Impact of minimum pricing (Leicester, 2011)
• Descriptive picture of off-licence alcohol purchasing in 2010
– Impact on-licence likely to be much smaller
• Average per-unit price on-trade £1.26 (E&W, 2009), off-trade £0.44
• Not modelling any likely behavioural responses
• Data from Kantar Worldpanel (market research organisation)
– 25,000 British households with in-home barcode reader
– Detailed grocery purchase records at the barcode level
• Data on products, stores, prices, household characteristics
– Look at alcohol purchasing and prices paid per alcohol unit
• Actual strength ABV known for beer, cider and alcopops
• Estimated for spirits based on brand and spirits type
• Constant 12.5% ABV assumed for all table wine
– Assume MUP is 45p in December 2010 prices
• How to uprate any minimum price an important question!
© Institute for Fiscal Studies
Impact by off-licence alcohol type
% of units sold
Avg. p/unit
% below 45p
Wine
37.6%
45.3p
65.3%
Spirits
26.9%
42.0p
77.3%
Lager
17.4%
37.9p
80.7%
Cider & perry
8.8%
29.1p
87.4%
Beer
4.3%
49.6p
42.1%
Fortified wine
2.5%
37.7p
73.7%
Sparkling wine
2.0%
79.3p
18.6%
Alcopops
0.6%
83.4p
1.6%
100.0%
42.6p
71.0%
ALL
Source: Leicester (2011), estimates from Kantar Worldpanel data 2010
© Institute for Fiscal Studies
Impact by alcohol consumption level
Increase in budget (%)
All
>35
30-35
25-30
20-25
18-20
16-18
14-16
0
12-14
0
10-12
3
9-10
20
8-9
6
7-8
40
6-7
9
5-6
60
4-5
12
3-4
80
2-3
15
1-2
100
<1
Price (p); % affected
(units per adult per week)
Units per adult per week
Average price per unit (p)
© Institute for Fiscal Studies
% units affected
Increase in food budget (%)
Source: Leicester (2011), estimates from Kantar Worldpanel data 2010
Impact by household income group
(including those who do not buy off-licence alcohol)
2.5
80
2
60
1.5
40
1
20
0.5
All
£70k+
£60k-£70k
£50k-£60k
£40k-£50k
£30k-£40k
£20k-£30k
£10k-£20k
0
< £10k
0
Increase in budget (%)
Price (p); % affected
100
Gross annual household income
Average price per unit (p)
% units affected
Increase in food budget (%)
© Institute for Fiscal Studies
Source: Leicester (2011), estimates from Kantar Worldpanel data 2010
Behavioural responses
• Consumers
– Demand response to changing relative prices (across/within category)
– Substitute towards on-licence consumption, other expenditures
• Responses vary across consumers
– Cross-border purchasing, home-brew, illicit alcohol purchasing?
• Manufacturers and retailers
– Indirect effect on more expensive alcohol products
– Long-term effects on product availability
– Impact on non-alcohol prices? Alcohol as a ‘loss-leader’?
• Current models do not really consider this range of effects
© Institute for Fiscal Studies
Minimum pricing: summary
• Preferable to taxation if taxes are not passed through?
– On average, pass through seems to be more than one-for-one
– But need more evidence on the distribution of pass-through
• Small impact on moderate drinkers?
– At a MUP of 45p, almost all off-licence purchasers directly affected
– Two-thirds of moderate drinker units affected
– Heavy drinkers on average pay less, but not that much less
• Probably slightly regressive
– Low income households buy cheaper products
– But on average effects are small, shouldn’t be main concern
• Transfers from alcohol consumers to producers/retailers
– Higher alcohol taxes at least raise revenue for public purse
– Reform tax system, not just raise rates ...
© Institute for Fiscal Studies
Current structure of alcohol taxes
(on an effective tax per alcohol unit basis)
Effective tax per unit (p)
70
60
50
40
30
20
10
0
0
4
8
Spirits
Source: Leicester (2011)
Note: Cider assumes ‘still’ cider rate
© Institute for Fiscal Studies
12
Beer
16
20
24
ABV (%)
Table wine
28
Sparkling wine
32
36
Cider
40
A flat-rate excise tax on alcohol content?
• Starting point of taxing on basis of alcohol content sensible
– Need compelling evidence to vary rates – area for further study?
– Do the marginal external costs very by drink / strength?
– US studies: if anything, case for bigger tax on beer than spirits
• Floor price through tax system if combined with ‘below-cost’ ban
• EU Directives restrict tax structure
– Precedents for levying additional alcohol taxes if not ‘state aid’
– Convoluted – preferable to reform at EU level!
© Institute for Fiscal Studies