Health, Social Care and Sport Committee National Assembly for Wales Cardiff Bay CF99 1NA 12 March, 2018

Dear Chariman Lloyd, Committee Members and Cabinet Secretary Gething

Re: General Assembly and the principles of the MUP Bill

I am writing to ask you to delay your decision on the MUP Bill. Cebr was asked by the Committee to submit evidence on its assessment of the policy, but was only able to begin its assessment when a sponsor for the research was found. That assessment is due to be produced by mid-April, but we hope to accelerate this to the end of March, so I respectfully request that the decision be delayed until April in order that it might take Cebr’s findings into account.  This letter continues with background on Cebr’s initial findings on MUP in Wales, including that:

1. The evidence from the Sheffield Alcohol Research Group (SARG) exaggerates the benefits of MUP;

2. The benefits are further exaggerated by flawed assumptions, and lack of account taken of unintended consequences; and

3. The evidence from SARG provides an inadequate treatment of the regressive re-distributional consequences of MUP.

Cebr’s contention, based on the findings of welfare analysis studies for other specific parts of the UK, lead it to expect that consequences for Welsh drinkers will reveal disproportionate losses to the poorest Welsh consumer.

I am writing in my capacity as Chief Economist at the Centre for Economics and Business Research (Cebr). Cebr is an independent economics consultancy that is governed by a strict ethics policy to maintain objectivity and transparency. We conduct research on behalf of our clients on commercially agreed terms, including research on the economic impact of current or proposed policies. This has included research for a number of clients on alcohol pricing and taxation.  

Since 2008, Cebr has conducted a range of studies on the potential effects of minimum unit pricing for alcohol. We have appraised the likely impact of the policy and produced reports covering the UK, Scotland and Northern Ireland. On certain impacts of the policy, we have also covered England and the English regions specifically.

We testified on the impact of MUP in the Scottish Parliament in 2012 and we submitted evidence to the Northern Ireland Assembly in 2011.

Our clients have included:

· SABMiller plc and

· The Wine and Spirits Trade Association (WSTA).

Cebr’s services have most recently been retained by Molson Coors in response to the Scottish Government’s public consultation on the proportionality of a minimum unit price of 50 pence per unit of alcohol. The Scottish Government intends to introduce minimum unit pricing on 1 May 2018. 

Cebr was invited, in October 2017, to submit evidence to your Committee. Unfortunately, we had to decline as we had not, by that stage, ever assessed the policy in the specific context of Wales, other than as part of a wider UK analysis. Cebr located a sponsor – Molson Coors – for new Cebr assessments in late December and has been working on the assessments for Scotland and Wales since that time. We aim to complete our assessment for Wales as soon as possible, but it is currently delayed due to a lack of availability of some of the necessary data, some of which we have requested from the Welsh Government.

We have read the Committee’s Stage 1 report and have concerns about how some of the evidence underpinning the recommendation to the Assembly of Wales to accept the general principles of the MUP Bill, has been represented to the Committee. These concerns are explained below.

It is in the context of these concerns that Cebr respectfully requests the Committee and Cabinet Secretary to postpone the Stage 1 debate or at least the Stage 1 vote on whether the Assembly agrees to the general principles of the Public Health (Minimum Price for Alcohol) (Wales) Bill. This would allow more time for Cebr to submit its fuller assessment of MUP, and of the issues raised herein, in the specific context of Wales.

1. The SARG evidence exaggerates the benefits of MUP because they include impacts for alcohol-dependent drinkers

The Committee’s Stage 1 report provides a lengthy synthesis of the work of the Sheffield Alcohol Research Group (SARG) and its Sheffield Alcohol Policy Model (SAPM) – covering its 2014 report, the 2017 interim report and the February 2018 full report. Whilst noting the criticisms, the Committee then states that:

The Cabinet Secretary told us that he was confident in the evidence base that the Sheffield Alcohol Research Group have provided, and he saw this as “an entirely reasonable basis on which to proceed”.

The Stage 1 report goes on to outline concerns about where the impact of the policy will be felt most strongly, with the expectation that this will be the low-priced strong drinks that are favoured by the heaviest drinkers. A prominent theme coming through the evidence is the potential for this to have serious consequences for dependent drinkers and their families.

In Section 4, the idea that alcohol-dependent drinkers should be considered as being a distinct group is introduced, largely through the evidence provided by Alcohol Concern Cymru (ACC). Paragraph 173 then states:

This point was also made in oral evidence by Colin Angus, SARG, who said “It’s very easy to mischaracterise all very heavy drinkers as being dependent, and that’s simply not the case. (…) Dependent drinkers are a different population”.

Paragraph 175 notes that:

“The Sheffield Alcohol Policy Model does not explicitly consider the policy’s potential impact on dependent drinkers…”

Further evidence from stakeholders is provided on the consequences of the impact on this new and apparently separate group of alcohol-dependent drinkers. Paragraph 185 states that “The Cabinet Secretary reiterated that the legislation was not intended to have the biggest impact on dependent drinkers”, and then quotes him as saying in oral evidence:

“(…) the legislation isn’t set up as having the biggest impact on dependent drinkers. (…) we’re not saying that this Bill is a silver bullet for all ills when it comes to alcohol-related harm. Dependent drinkers (…) are a different category, and some of those may be less price sensitive about their behaviour. So, this [MUP] is about hazardous and harmful drinkers in particular.”

We have been reviewing the SARG appraisals of MUP since 2008. The reports on these appraisals have never identified alcohol-dependent drinkers as a separate drinker category or a different population, as indicated by Colin Angus in the quote above. Nor has SARG ever sought to exclude this alcohol-dependent drinker category from its model or results.

According to the ACC evidence, there are 36,000 alcohol-dependent drinkers in Wales. But, whereas Colin Angus acknowledges that they are a separate group requiring separate treatment, SARG does not and never has excluded this group of alcohol-dependent drinkers from its projections of the benefits of the policy. It is simply assumed that their behavioural response to MUP will be the same as that of all other drinkers.

Harmful drinkers are the most important target of the policy and, as such, their predicted behavioural responses to MUP are a key driver of the projected benefits of introducing the policy. The benefits estimated by SARG are based on the full 81,000 harmful drinkers but, by its own admission, up to 36,000 of these could belong to “a different population” that require separate treatment when assessing the policy because they are much less likely to respond to price changes due to their physical dependency.

This means that the health and other benefits of the policy estimated by SARG are exaggerated and, as such, it is Cebr’s contention that the Committee and the Cabinet Minister does not have a complete picture. We are surprised that the SARG team has not revealed the fact that the projected benefits of MUP that it has submitted as evidence to the Committee includes benefits that are predicated on behavioural changes by alcohol-dependent drinkers that are very unlikely to materialise.

The Stage 1 report sets out several submissions expressing concerns about the impact of MUP on this alcohol-dependent drinker group. There is a genuine fear, which Cebr shares, that this type of alcohol-dependent drinker is much less likely to reduce their alcohol consumption in response to MUP. This would mean a significant cost of living impact and, for those on low incomes, cuts to other parts of the household budget, such as food and heating, in order to maintain their levels of alcohol consumption despite the increased price of doing so.

We are likewise of the view that this could cause health-related problems and complexities for the drinker and their family which could contribute to a widening of health inequalities, contrary to the intended purpose of the legislation, which is to narrow them.

Paragraph 155 of the Committee’s report states that:

The Bill’s Equality Impact Assessment suggests that households in poverty have the most to gain from the proposals, because of the anticipated reduction in levels of consumption and the health benefits associated with this.

But this misses the point that, for the alcohol-dependent drinker group, there is no reason to expect reductions in levels of consumption and there are, therefore, no associated health benefits.

This is clearly misleading and means that the Committee is making a recommendation that the General Assembly accept the general principles of the Bill, on the basis of evidence that clearly exaggerates the purported benefits of the policy. However, not only are the benefits exaggerated, but the potential negative consequences in terms of widening health inequalities for the families of dependent drinkers are completely overlooked. The SARG do not and never have taken account of this issue. This means that the net benefits of the policy are also exaggerated because the costs are underestimated (or, more accurately, have not been measured at all).

This is important because the negative implications of any policy will look less detrimental if the benefits are exaggerated. Similarly, the negative implications can be expected to be seen as more detrimental if the purported benefits can be shown to be overestimated and by how much they need to be reduced to eliminate the bias.

2. Benefits further exaggerated by flawed elasticity assumptions and no account of unintended consequences

Cebr’s view has always been that the Sheffield University appraisals are built on weak foundations that most likely lead to exaggerated estimates of the benefits of MUP, quite apart from the specific issue of alcohol-dependent drinkers.

The evidence supporting the proposition that MUP can reduce alcohol consumption is underpinned by a SARG assumption that all drinkers will respond uniformly. This is fundamentally flawed. A uniform response is highly unlikely and any estimates of the benefits of MUP made on the basis of this highly simplifying assumption must be treated with caution because they are more than likely going to be exaggerated.

Cebr’s contention is that hazardous and harmful drinkers can be expected to have lower demand responsiveness to price changes and that the demand elasticity assumptions underlying the appraisal of MUP should be adjusted accordingly. In the context of Scotland, Cebr has demonstrated that the use of a more realistic set of demand elasticity assumptions (elasticities that reflect these differing levels of responsiveness) results in a significant dilution of the expected reductions in alcohol consumption relative to SARG’s estimates, by up to 30%. If MUP is less effective in reducing alcohol consumption, is will also be less effective in reducing alcohol-related deaths and hospitalisations.

The evidence on which the Committee has reached the conclusion that the general principles of the MUP Bill should be accepted could thus be biased. The same concern arises the fact that SARG has never properly considered the unintended consequences of the policy, not least the potential for the policy to widen health inequalities amongst the families of alcohol-dependent drinkers, as already noted.

3. Inadequate treatment of the regressive re-distributional consequences of MUP

The SARG appraisals cannot be considered to provide an adequate treatment of the regressive distributional implications of MUP in terms of its impact on the finances and welfare of drinkers and their families, as reflected in their relative cost and standard of living. The SARG appraisals are bound to paint a favourable picture of the impacts of MUP as they only address whether consumers will spend more or less on alcohol once the policy is introduced and they have decided their behavioural response, if any.

Cebr has, in previous research, highlighted the regressive re-distributional properties of minimum unit pricing for alcohol. The increased alcohol expenditure required due to higher pricing will hit those in the poorest income deciles. This can be understood in terms of an increased cost of living in respect of alcohol consumption that is maintained or a reduction in their standard of living if they have to reduce their alcohol consumption in light of household budget constraints. Either way, the policy has the potential to exacerbate both financial and welfare inequality.

To illustrate the case using evidence from Scotland, SARG estimated that harmful drinkers in poverty will spend about £88 less on alcohol per annum, whereas the harmful drinker not in poverty will spend up to £20 more. But these figures mask a host of other changes that are only revealed through a more thorough application of the principles of welfare economics. What these results do not tell us, in other words, is:

· How much more consumers will have to pay for the alcohol that they do consume (in economic theory, this is known as a transfer from consumers to producers but can be equated with a measure of the value of the cost of living impact); or

· What consumers lose in terms of the erosion of their standard of living as a result of having to reduce alcohol consumption (in economic theory, this is known as the deadweight loss that arises when less consumption is possible, which can be valued by making assumptions about the shape of the demand curve); or

· The financial hardship that it could impose on the families of problem drinkers who do not reduce their consumption in response to the introduction of MUP, as can be expected in the case of alcohol-dependent drinkers.

For the general population, the vast majority of which are non-problem drinkers, the higher alcohol prices that will result from MUP can be expected to have a combination of these negative impacts.

The results of Cebr’s comprehensive welfare accounting exercise revealed much more significant impacts on Scottish consumers. Cebr found a negative net welfare impact on all consumers, with a disproportionate burden of the welfare losses falling on those in poverty. This suggests a loss to harmful drinkers in poverty of over £600 in a year and paints a radically different picture to that suggested by SARG’s representation of this aspect of the policy.

At the income decile level, the picture is even starker. The extent to which Scottish households in the bottom three income deciles would be financially clobbered by a 50p MUP is clearly visible from the table below, especially in the case of hazardous and harmful drinkers. For example, harmful drinkers in the lowest income decile are expected to have to endure a welfare loss valued at over £930 in a year, at £1,380 per annum for those in the second lowest decile and at over £2,200 for those in the third lowest income decile.

Table 1: Total value of consumer losses per household after the introduction of a 50p MUP, by drinker type and income decile, Sheffield University elasticities and 2014 baseline, Scotland

Source: Cebr analysis

The clear conclusion is that the burden of a 50p MUP can be expected to fall disproportionately on the poorest households in Scotland. For instance, the impacts on the families of harmful drinkers in the three lowest income deciles account for between 4% and 8% of the average annual Scottish household budget. These cannot be said to be insignificant. Families with problem drinkers whose demand for alcohol does not respond much to MUP can expect to suffer through cuts elsewhere in the household budget. Alternatively, the drinker seeks cheaper alternatives on the black market or through surrogate products.

Cebr is certain that a similar comprehensive welfare analysis of MUP in the specific context of Wales will reveal a radically different picture to that suggested by SARG. The latter asserts a net financial gain for the most deprived Welsh drinker as a result of reduced alcohol spending. Cebr expects its fuller cost and standard of living impact assessment will reveal losses to the poorest Welsh consumer that are comparable with those estimated by Cebr for Scottish drinkers and their families, as reflected in Table 1. As such, Cebr will provide a much clearer demonstration of how a disproportionate burden of MUP can be expected to fall on the poorest Welsh drinkers and their families.

These deleterious cost and standard of living impacts on Welsh households are of significant relevance in the context of the current economic climate, particularly the challenges presented by the significant depreciation in the value of sterling following the Brexit vote, the resulting rise in inflation and the further squeeze on household disposable incomes (on top of public sector austerity and sluggish overall earnings growth), given the UK’s import dependence in the areas of food and drink. Minimum unit pricing is thus likely to impose further hardship on families already facing significant pressures on their cost and standard of living.

It is in the context of these concerns that Cebr makes its respectful request to the Committee and Cabinet Secretary to postpone the Stage 1 vote on whether the Assembly agrees to the general principles of the Public Health (Minimum Price for Alcohol) (Wales) Bill. This would allow more time for Cebr to submit its fuller assessment of MUP, and of the issues raised herein, in the specific context of Wales.

Yours sincerely,

________________

Oliver Hogan

Chief Economist (& Director)

Centre for Economics and Business Research