National Assembly for Wales

Y Pwyllgor Iechyd, Gofal Cymdeithasol a Chwaraeon

Health, Social Care and Sport Committee

29/11/2017

Aelodau'r Pwyllgor a oedd yn bresennol

Committee Members in Attendance

Angela Burns AM
Caroline Jones AM
Dai Lloyd AM
Dawn Bowden AM
Jayne Bryant AM
Julie Morgan AM
Lynne Neagle AM
Rhun ap Iorwerth AM

Y rhai eraill a oedd yn bresennol

Others in Attendance

Chris Snowdon Sefydliad Materion Economaidd
Institute of Economic Affairs
Colin Angus Grŵp Ymchwil Alcohol Sheffield
Sheffield Alcohol Research Group
John Holmes Grŵp Ymchwil Alcohol Sheffield
Sheffield Alcohol Research Group

Swyddogion Cynulliad Cenedlaethol Cymru a oedd yn bresennol

National Assembly for Wales Officials in Attendance

Gareth Pembridge Cynghorydd Cyfreithiol
Legal Adviser
Philippa Watkins Ymchwilydd
Researcher
Sarah Sargent Clerc
Clerk
Sian Giddins Dirprwy Glerc
Deputy Clerk
Stephen Boyce Ymchwilydd
Researcher

Cofnodir y trafodion yn yr iaith y llefarwyd hwy ynddi yn y pwyllgor. Yn ogystal, cynhwysir trawsgrifiad o’r cyfieithu ar y pryd. Lle y mae cyfranwyr wedi darparu cywiriadau i’w tystiolaeth, nodir y rheini yn y trawsgrifiad.

The proceedings are reported in the language in which they were spoken in the committee. In addition, a transcription of the simultaneous interpretation is included. Where contributors have supplied corrections to their evidence, these are noted in the transcript.

Dechreuodd y cyfarfod am 09:30.

The meeting began at 09:30.

1. Cyflwyniad, ymddiheuriadau, dirprwyon a datgan buddiannau
1. Introductions, apologies, substitutions and declarations of interest

Croeso i bawb i gyfarfod diweddaraf y Pwyllgor Iechyd, Gofal Cymdeithasol a Chwaraeon yma yng Nghynulliad Cenedlaethol Cymru. A gaf i, yn gyntaf oll, estyn croeso i fy nghyd-Aelodau? Mae pawb yma'r bore yma, felly croeso. A allaf i bellach egluro i bawb fod y cyfarfod yma, yn naturiol, yn ddwyieithog? Gellir defnyddio clustffonau i glywed cyfieithu ar y pryd o'r Gymraeg i'r Saesneg ar sianel 1, neu i glywed cyfraniadau yn yr iaith wreiddiol yn well ar sianel 2. A allaf i bellach atgoffa pobl i naill ai diffodd eu ffonau symudol ac unrhyw gyfarpar electronig arall, neu eu rhoi ar y dewis tawel, a hysbysu pobl y dylid dilyn cyfarwyddiadau'r tywyswyr os bydd larwm tân yn canu? Ac mae'r meicroffonau yn gweithio yn awtomatig; nid oes angen i neb gyffwrdd â meicroffon.

Welcome, everyone, to the latest meeting of the Health, Social Care and Sport Committee at the National Assembly for Wales. Firstly, I'd like to welcome my fellow Members. Everyone is here this morning, so welcome. I'd also like to explain that this is, naturally, a bilingual meeting. You can use headphones to hear interpretation from Welsh to English on channel 1, or to hear amplification on channel 2. I'd also like to remind people either to switch off their mobile phones and any other electronic equipment, or to set them onto silent, and also I'd like to inform you that you should follow the ushers if there is a fire alarm. The microphones will work automatically, so you don't have to touch them. 

2. Bil Iechyd y Cyhoedd (Isafbris am Alcohol) (Cymru) - Sesiwn Dystiolaeth 4 - Grwp Ymchwil Alcohol Sheffield, Prifysgol Sheffield
2. Public Health (Minimum Price for Alcohol) (Wales) Bill - Evidence Session 4 - Sheffield Alcohol Research Group, Sheffield University

Gyda chymaint â hynny o ragymadrodd, rydym ni'n symud ymlaen i eitem 2, a Bil Iechyd y Cyhoedd (Isafbris am Alcohol) (Cymru), sesiwn dystiolaeth 4. O'm blaenau mae cynrychiolwyr o grŵp ymchwil alcohol Sheffield, Prifysgol Sheffield. Rydw i'n falch iawn, felly, i groesawu John Holmes, uwch-gymrawd ymchwil, grŵp ymchwil alcohol Sheffield, a hefyd Colin Angus, cymrawd ymchwil, grŵp ymchwil alcohol Sheffield. Croeso i chi'ch dau. Diolch yn fawr iawn i chi am ddarparu'r wybodaeth ysgrifenedig o flaen llaw, ac fel sy'n draddodiadol, rydym ni'n mynd yn syth i mewn i gwestiynau. Mae'r cwestiynau cyntaf o dan law Rhun ap Iorwerth.

With those few words, we move on to item 2, Public Health (Minimum Price for Alcohol) (Wales) Bill, evidence session 4. Before us we have representatives from the Sheffield alcohol research group from Sheffield University. I'm very pleased to welcome John Holmes, senior research fellow from the Sheffield alcohol research group, and Colin Angus, research fellow of the Sheffield alcohol research group. Welcome to you both. Thank you very much for providing the written information beforehand and, as is traditional, we will go straight into questions. The first questions are from Rhun ap Iorwerth.

Bore da iawn i chi. Mi wnaf i ddechrau, os caf i, yn cymharu'r hyn sydd o'n blaenau ni rŵan efo'r hyn a ddatblygwyd yn 2014 yn eich model polisi chi. A allech chi egluro'r newid methodoleg, y newid canlyniadau sydd wedi digwydd rhwng y model 2014 a'r model cyfredol?

A very good morning to you. I will start, if I may, by comparing what's before us now with what was developed back in 2014 in your policy model. Can you explain the change in methodology and the change in outcomes that has happened between the 2014 model and the current model?

Okay. So, there are not too many significant differences in terms of methodology between the two reports. I think, probably, the main difference between them is that in the previous report we considered socioeconomic status in a binary 'in poverty'/'not in poverty' sense, whereas here we have a more detailed treatment of socioeconomic status, so we look at the differences in terms of the baseline consumption, and purchasing, and harm and the policy effects across quintiles of the Welsh index of multiple deprivation. So, we have a more detailed understanding here of the socioeconomic impact right across the spectrum. That's kind of the main methodological difference.

I guess there are a few slight updates in terms of the risk relationships, particularly for some of the cancers and for some of the cardiovascular conditions, which have changed. So, those are the estimates of how much your risk of harm relates to the amount that you drink. So, those are the main differences methodologically speaking, but there are also, obviously, quite a lot of differences in terms of the data that goes into the model to inform it. And I guess the biggest change is the fact that one of the biggest limitations in the previous report was that the consumption data came from relatively old sources, because there wasn't a more recent survey of Wales that had good alcohol consumption data in it. Whereas, in the new report, we have data from the national survey for Wales, which is new, so the consumption data is much more up to date, and it's better. So, that's the main difference. There are also other differences. We have more recent pricing data and newer harms data. So, there are lots of—. The data that goes into the model is quite different. Methodologically, there's not that much that's changed. 

And for a layman—we're all laymen and women here—why should we trust the new methodology more? Why should we believe that what we have in front of us now paints a more accurate picture or makes a better projection, do you think? 

I guess that's how science works, right? Science is continually evolving and you never—. Well, it's difficult to imagine reaching a point where you think that things have advanced as much as they ever could do. So, it's about incremental progression. We can clearly say that having better, newer, more up-to-date estimates of the relationship between consumption and harm is unequivocally a good thing. Having a more up-to-date and more accurate, bigger sample size, and more representative survey data in terms of consumption behaviour, is unequivocally a good thing. These things can always be improved more—more studies can be done, you can do bigger surveys—but I think it's clear that this is an incremental progression and further incremental progression is clearly possible, but—.

09:35

I think the other thing to add is that we've done versions of this model repeatedly over the last nine years. Sorry, I'm getting feedback in my headphones. We've done versions of this model repeatedly over the last nine years and we've continually updated it, adapted it to new countries, brought in new data. And also, we've done a whole range of what we call sensitivity analyses, comparing alternative modelling assumptions, alternative sources of evidence, alternative data, alternative ways of analysing those data. What we found is, although, of course, the results change, the precise numbers change between versions of a model, but the conclusions have stayed robust to all these different ways of running the model. And we still find that the policy's effective, but it's targeted on harmful drinkers and there's relatively little impact on moderate drinkers, and that it reduces health inequality. So, those conclusions seem very robust to different ways of doing the work.

And I have no reason to doubt the calculations that you have made and the conclusions that you have come to. How and where has the theory been tested in practice?

The biggest amount of work has been done in Canada, but before I go on to that, I just want to make the point that it's often said that most of the evidence just comes from Sheffield. That's not really the case. Our model itself is built on a very large body of evidence: on relationship between prices and consumption; the relationship between alcohol consumption and the risks to health from that—there's huge epidemiological literature on those health risks; and also, various bits of evidence on the social patterning of alcohol consumption, so, who drinks what, who buys what and who suffers harm. Our model is not so much a piece of evidence in itself, it's a synthesis of a huge body of evidence. There have also been similar models in Australia and New Zealand. The Institute for Fiscal Studies has done some modelling. There have been various other bits.

So, on to the Canadian work. That work has looked at what impact fluctuations in the minimum prices in Canada over time have had. Canada doesn't quite have minimum unit pricing, it has a slightly different system where the minimum price is different for beer and spirits. It's not directly linked to the strength of the alcoholic drinks, but it's still the same basic principle. You have a floor price, below which alcohol can't be sold to consumers. So, it's a kind of good laboratory for testing what the impact of a policy might be. And they've consistently found that as these fluctuations in the minimum price happen, you get commensurate fluctuations in levels of alcohol consumption, in levels of alcohol-related hospital admissions, mortality and crime. So, there's really good evidence coming out of Canada that when you impose the minimum price, and when you increase the minimum price, all else being equal, you do get the kind of benefits that our model suggests.

Alcohol consumption is driven by culture quite often as well. Are we happy that the relationship between Canadians and alcohol is similar enough to the relationship between people in Wales or the UK and alcohol? I.e. it is known that it's difficult to gather evidence on consumption. Are you happy and what evidence is there that the evidence that you're gathering on consumption here, specifically—partly driven, perhaps, through culture—is similar enough to Canada to warrant drawing those kinds of clear comparisons?

There are differences between Canada and the UK, but their drinking culture is fairly similar. They don't have massive problems. They don't have a huge elicit market, as, say, they do in some of the other countries that have minimum prices but there haven't been studies on, such as Russia and some of the former Soviet countries. Canada isn't like those countries. It has a fairly similar approach to alcohol. Their alcohol consumption data is good quality.

You mention that it is difficult to collect data on alcohol consumption. The effects of minimum pricing on alcohol consumption in Canada are not based on survey data on consumption. They're based on sales data. That data is much more robust. That doesn't suffer from the problems that people talk about. So, those are very good estimates. And I think, yes, we are fairly comfortable that the basic economic relationship between price and consumption is likely to be fairly similar between Canada and the UK.

If I could just add there. So, where you believe that there are differences that might cause your effects to be different of a policy, one way that you can control those differences is by building a model in which you look at the differences between the two countries within the same framework. We've built models for Canada and they came to the same broad conclusions, and they appear conservative with respect to the actual effects estimated from the study—the evaluation studies.

09:40

Yes, well, there are several to go. Angela, a brief intervention here?

Yes, it was just on the model itself. I was curious—or I was keen to understand your basis for using a 20-year period for the modelling and for the payback, and I didn't go and find the piece of research where you said, in your small print, 'Go and have a look at this and it explains it', because I thought it'll be easier for you to explain it to us.

Essentially, it's because many of the harms caused by alcohol are chronic diseases that are the result of the cumulative effects of drinking over time. So, when people change their drinking, you need to wait for the effects of that change to play out and everything to go back to how it—to kind of a new state, a new equilibrium, if you like. So, it's essentially just letting all those effects on chronic disease play through over time—what we call 'time lags'—and cancer in particular takes a long time to develop. So, that's why we run the model for 20 years. It's just that kind of effect of the cumulative effects—

It's not a generational—. No, no, that would be a different analysis. We have done some work of that kind for Cancer Research UK, but it's not in this model. 

Diolch. You mentioned that you've based most of the testing on Canada. Can you tell us a bit about other countries and the research that's been done internationally?

Yes. So, there's a huge literature on the relationship between change in alcohol prices and alcohol consumption, and that comes from a very large range of countries—a lot of European research, particularly in Scandinavia, a lot of American research and quite a bit of research in the UK and from Australasia. So, that's really useful. There's similar kind of evidence from similar countries on the relationship between price increases and rates of alcohol-related harm. The literature on the relationship between alcohol consumption and levels of harm comes from a whole range of countries. It's essentially a very international evidence base. So, this is really evidence from lots of countries.

Specific work on minimum pricing—there is less of it. Canada, obviously, is the country that's looked at it most, but there's been work in Australia—a wholly different model but very similar principles to our work. They found that if they increased minimum prices in Australia, they estimate they would see even bigger falls in consumption than we're seeing in our modelling for the UK. New Zealand found slightly different results. They had some data problems there, and they got slightly different results. I think that that's all the work that is being done in other countries.

One of the things I did want to highlight is some of the work that's been done on drinkers who are dependent on alcohol in Scotland. So, there's been work by a team led by Jonathan Chick surveying 639 dependent drinkers in Glasgow and Edinburgh. These are people who are either in treatment services or who have been hospitalised due to alcohol-related health problems. If I just give you some statistics from that work: among this population, the average weekly consumption of alcohol was 185 units. That's 18 bottles of wine or 4.5 litres of vodka. The average price they paid for that alcohol was 40p per unit and many paid much less. A quarter of them were drinking white ciders—these very strong, cheap ciders—and 11 per cent of them drank those ciders exclusively. Between December 2010, when the data were first collected, and February 2015, when they followed up the sample, 16 per cent of them had died and the mean age of death was just 51 years old.

So, it's often said that there isn't much evidence on minimum pricing. I think there's some really good evidence that hasn't been looked at, that doesn't get talked about a lot and that gives us very compelling reasons to be concerned about the effects of cheap alcohol.

Y cwestiynau nesaf gan Caroline Jones.

The next questions are from Caroline Jones.

Diolch, Cadeirydd. Good morning. Given the complex factors involved in different people's alcohol consumption, to what extent do you think we can depend on a mathematical model to predict behavioural change?

I think there are lots of complex factors, but what we see consistently is that alcohol does follow the basic laws of economics: when prices go up, people buy less of it; when prices go down, people buy more. Similarly, we see, as a general principle, that the more people drink—setting aside debates about whether or not moderate drinking is good for your heart—the higher risk they are from alcohol. Clearly, there are lots of individual variation within that, and one of the strengths of our model is that it takes account of that by, instead of just looking at the average changes in the population and how average changes in consumption affect average risks and average harms, we look at around 100 different groups in the population, defined by their age, their sex, their consumption level and their deprivation level, and we see different patterns among all those groups. They buy different things, they pay different prices, they drink different products and they face different risks of different harms. We take account of all of that complexity. There was always more complexity that we could take account of, but there's got to be a pragmatic approach as well.

But I think, in terms of how much complexity has been looked at in the literature, this is one of the most detailed and sophisticated pieces of work that's been done. That's why we've been able to publish it in some of the most prestigious scientific outlets, so in The Lancet twice, the British Medical Journal and in PLOS Medicine, which is another very high-impact journal.

09:45

If I could just add to that, taking account of variation across the population is our biggest driver, in a sense; that's the thing we've been striving for continually since we began this modelling process nearly a decade ago. And we take account of it even more than John's just described. John's talking about the differences between groups in the population, but we also account for differences within groups in the population. So, we don't just assume that all 18 to 24-year-old extremely deprived, heavy-drinking men drink the same things and pay the same prices. We have a full distribution of consumption and a distribution of prices, even within those groups. So, we're allowing for variation even within groups. We're not saying, 'There are 100 different kinds of people in Wales', we're accounting for the fact that the variation is still richer than that.

Okay, thank you. Next, can I just ask your views on the Institute for Fiscal Studies's suggestion that minimum unit pricing is well targeted at increasing the price of cheap alcohol, but much less so at increasing the price of strong alcohol products?

Okay. So, I guess it depends on how you define strong products. It's very well targeted on strong cider, because that tends to be sold very cheaply, particularly the stuff that is bought in large quantities. Similarly, it's very well targeted on strong beer. If you mean strong products as simply spirits, as a general category, again, it targets the very cheap spirits that are disproportionately purchased by the heaviest drinkers. It doesn't target all spirits, which, of course, are strong drinks.

The other thing to say is that minimum pricing shouldn't be considered as a policy in isolation. I know the Welsh Government have limited powers over what they can do, but it should be considered alongside other strategies that target the things that minimum pricing doesn't target, or that would work in complement with minimum pricing. But I do think it does target the strong drinks within each beverage category, particularly the ones that are bought by the highest risk drinkers.

The minimum price is literally directly linked to the strength of the product. So, in a sense it is targeting the strongest products, because the price is directly correlated to that.

Yes, just to come back to that point: alongside what other things? You said that minimum pricing should be one tool amongst many and that Welsh Government tools are, perhaps, limited. What are the other things, either here or—?

Other effective policies include increases in alcohol taxation and strengthening the licensing system so you ensure that you have fewer alcohol outlets and more well-run alcohol outlets. There's particularly good evidence that increasing the amount of screening for heavy drinking in primary care and in other medical settings and then delivering brief interventions to people who've screened positive is an effective way of reducing consumption.

If you have concerns, particularly about people who are dependent on alcohol, then increasing investment in specialist treatment services is an incredibly important thing, and the Scottish Government have had some success from doing that. There's also a wider, kind of, creating an environment that's conducive to moderate drinking, including things like restricting advertising.

So, there's the public health side. But, briefly on that, can minimum pricing work without the taxation increases that you think is a good part of a set of tools?

Yes, I think it can. It is, perhaps, the other way around that there's a problem—that some of these other tools don't work as well while you still have very, very cheap alcohol available to people who are motivated to seek out cheap alcohol. So, I think minimum pricing will work, because it simply removes that cheap alcohol.

09:50

But you could effectively have minimum pricing through taxation only, could you not?

You could, but it would require very large increases in taxation, and it would hit all drinkers, because, obviously, taxes would go up on all the drinks, including the drinks that are purchased by moderate drinkers as well as heavier drinkers.

But you could specify what you taxed, though. You don't have to tax all alcoholic drinks.

You can, but it gets very difficult within the limits of, particularly, European law. We've seen this through the Scottish Government's court case. It seems open to debate within EU law whether you can target tax increases on very narrowly defined beverage categories. You can certainly up the tax on cider, if you wanted to, whether you—

Whether you can up the tax on certain kinds of cider, particularly very high-strength cider, to very large degrees, is a different matter.

Just on that point of taxation, am I right in saying that if there is an increase in tax, the supplier can deal with that by, actually, not putting the prices up? They can use those as loss leaders, whereas they can't with minimum alcohol pricing. Am I right in saying that?

That's right, yes. So, tax is paid at the point at which alcohol is released for sale by producers, which means that, somewhere along the line, someone has to cover the costs of that tax, but it doesn't necessarily have to be passed on. If you increase the tax on Smirnoff vodka, it doesn't necessarily have to be passed on to the price of Smirnoff vodka. A supermarket could increase the price of lettuces instead. When we looked at this, we found that, when alcohol taxes go up, supermarkets tend to increase the price on more expensive products by more than would be expected and increase the price of cheaper products by less than would be expected. So, to some extent, it seems that supermarkets—when taxes go up, they're subsidising low cost for cheap products by increasing the price of more expensive products. Essentially, they're subsidising the drinking of heavier drinkers by—

But they wouldn't be able to do that with minimum—

Sorry, yes—they wouldn't be able to do that with minimum pricing, because it simply says, 'Any product that's sold below this price threshold has to be sold at least at this threshold.'

Except there are possible loopholes—you can give away all sorts of free things with alcohol to make the price deflate.

I think that's a matter for the Welsh Government to look at carefully and work out what those loopholes might be and which ones could be closed. I know the Scottish Government has looked at this. They've looked at the possibility of closing the opportunity, for instance, to have a meal deal, where you pay for the meal and you get the bottle of wine free. They've looked at that. I think that's in the detail of the policy, of exactly how you write the legislation—

But we actually have the legislation in front of us, and, yes, it would not be allowable to have a meal deal where you have a bottle of wine free, but you can have a wine deal, where you have the meal free.

And, ultimately, in the eye of the purchaser, they're still getting cheap alcohol, because they needed the food anyway.

So, I guess all you can do is look to try and close these loopholes, while at the same time recognising that all of those loopholes aren't necessarily going to cancel out the benefits of the policy. Not every person who is desperately trying to drink 50 units or whatever a week is going to, every time they want a bottle of wine, buy a meal alongside it, because, ultimately, that's going to end up costing them more anyway than alcohol would cost at this time point.

I've developed a very clear loophole in my mind, but, okay. [Laughter.]

Reit, symudwn ni ymlaen. Mae Jayne Bryant wedi bod yn amyneddgar iawn. Jayne.

We'll move on. Jayne Bryant has been very patient. Jayne.

Thank you, Chair. The Scotland example you gave was very powerful, but what evidence—or could you expand on the evidence that there is that harmful drinkers consume the products that will be most affected by the minimum alcohol pricing?

Something that we've consistently found across all of the modelling work that we do is that the heavier drinkers prefer the cheaper products that are more disproportionately affected by the policy. That's almost a universal constant across all of the different countries, the different contexts and the different times that we've looked at.

So, how does that vary across population, age, gender? Have you looked—? Has that been—?

In lots of different ways. Understanding variation across the population is something that we're really interested in. There are so many different ways in which you can look at these variations that it's almost—it's difficult to look across them all at once, I guess. But there are differential impacts of the policy. As well as by drinker group, there are also differential impacts of the policy by age and by sex and by deprivation—

09:55

For example, what we find is that, if you break down the population into moderate, harmful and hazardous drinkers, and if you then further break it down into income groups or deprivation groups, moderate drinkers buy very little of the cheap alcohol that is affected by minimum pricing. It can be as little as a few units a week; two or three units a week, about a pint of beer. In contrast, harmful drinkers, and particularly those harmful drinkers on lower incomes, who are at greatest risk from their drinking—because we know that lower income groups suffer more harm per unit than higher income groups—those low-income, high-risk drinkers buy huge amounts of this alcohol. It accounts for maybe more than 50 per cent of all of their alcohol.

So, what we’re seeing here is that moderate drinkers, including moderate drinkers on low incomes, are relatively unaffected by the policy because they simply don’t buy much of the cheap alcohol. Heavier drinkers, and particularly heavier drinkers on lower incomes, are much more affected because they’re buying large volumes of this alcohol. And that kind of basic pattern is what drives all the results of our model, because all of that plays through into the price increases they face, the consumption changes they make, and then the changes in their risk of harm, which then produces the harm outcomes.

So, one of the points, just thinking about the moderate drinkers—the modelling, I think, is that moderate drinkers’ spending will increase only by about £2 a year, but in the explanatory memorandum it suggests a bottle of red wine, own brand, could increase by as much as 19 per cent. What would you say about that?

The estimated changes in spending in the report are after people have shifted their behaviour. So it assumes that people change their consumption—or it doesn’t assume; it estimates how they change their consumption behaviour, and then once they’ve changed that consumption behaviour, as a result, how much has their spending changed on top of that. So, that’s why you would see some of these quite big changes in prices for some groups, but because they reduce their consumption, their spending doesn’t necessarily increase by as much as you might expect.

Hapus? Rydym ni’n symud ymlaen at Angela Burns nesaf.

Happy? We move on to Angela Burns next.

Thank you, and thank you for your papers. I’ll tell you where my concern lies. I have two. I am concerned that people who are at high risk and on a low income, people we want to target, will actually, because of the desire to drink alcohol, end up forfeiting heat, food, things their kids might need—whatever it might be. So, it’s about unintended consequences. That’s my first thing, and I haven’t seen anything in the evidence that has convinced me or shown me that, actually, that doesn’t happen. So, I’d be really interested in anything you’ve got on that.

The second area where I have real concern is alcoholism and addiction. We understand addiction pathways very clearly and I haven’t yet been overwhelmed by your evidence about where people who are currently having alcohol issues might travel to if they find that alcohol is too expensive. When I go back to your research and the estimated effects of increasing minimum prices from multiple Canadian studies—. You have here, for example, violence or crimes against the person, there are traffic violations, there are alcohol-related hospital admissions, obviously. But what I don’t see is that actually—. Well, I see two things. One is that most of it just seems to come from British Columbia, because the other areas didn’t have the studies taken in them. So I just think, well, how strong, then, is your pool of research? That’s the first thing.

The second thing is: is there any research out there that—? I can accept that if we manage to stop those most at risk from consuming alcohol, then they may not be suffering from alcohol-related hospital admissions, or deaths wholly attributed to alcohol. But do we find that there’s a pathway where they end up on something that is just as bad, if not worse? For example, we have these awful drugs around at the moment, like spice, et cetera. Because there is a very clear correlation in all addictive personality traits and most addictions that there can be a direction of travel, and what you actually need is intervention to stop that direction of travel.

I think this is a really important concern, and it’s one of the things that our model doesn’t cover. There’s been relatively little research in this area, partly because it’s a very difficult population to do research on. One of the most important things to do is to reiterate what I said earlier: that minimum pricing is a policy that’s targeted at heavy drinking in the general population. It’s not specifically targeted at—although it may benefit—dependent drinkers. Dependent drinkers are best helped by investment in good-quality treatment services, and we'd recommend that that is done. 

That said, we do know that people who are dependent on alcohol—and I think you've alluded to this—are not a homogeneous group; they're a very complex group, a very diverse group with very specific multiple and interrelated problems, some of which may be to do with drugs, some of which may be to do with homelessness, poverty or other problems. It's likely that the vast majority of this population will be directly affected by the policy, but they're going to be affected in very different ways, and we can't assume there will be one broad-brush response. So, those who already have problems with drugs or are close to problems with drugs may see substitution behaviour, but we may see positive effects in other dependent drinkers.

So, we may see, for instance, a greater likelihood that people seek treatment for their drinking because of the exact problem that you identified—that they're spending already at their budget, and they realise, 'Well, either I start redirecting spending, or I do something about my drinking'. And what we've certainly heard from some providers of treatment services is that they would expect those kind of responses. We may see improved recovery prognoses because the cheap alcohol isn't there to facilitate relapses. We might also simply just see some of these people reducing their consumption, but the negative possibilities are there as well.

So, what do we do about that? Well, I think the first thing you need to do is monitor it very carefully when the policy comes in. That's been done in Scotland; that's our contribution to the evaluation of minimum pricing in Scotland. We'll be working with both treatment users and treatment providers to see what support and services need to be put in place to ensure that these adverse consequences don't happen. If they do start to appear, what can be done about them? And also, if we start to see positive consequences, how can those be reinforced, sustained and spread to the wider dependent community? I think these are the kind of things that need to be looked at. It shouldn't be a case of a stop-go question with minimum pricing; it should be a case of monitoring and evaluating how it affects these populations and what can be done to support them, because it is likely to be a very complex and heterogeneous response. 

10:00

So, can I ask the question slightly—? It might seem similar, but it isn't, and, in fact, our resident doctor may be able to help me on this. My understanding with alcohol, both addiction and heavy drinking, is you have two facets: you have the psychological need, which in an addicted person is impossible to shake off without major intervention, but within a heavy drinker is also very hard to shake off. You also then have the physical addiction, which is where your response to the alcohol shrinks over time, therefore you have to drink more in order to get more response. So, my question is: is there any research that shows where heavy drinkers are on that scale of physical response to the alcohol compared to the addicted, and where that line is? Because, for example, we may actually as a committee want to make the recommendation to do this, but people who are deemed heavy drinkers—if we can find a way of fishing them out of the system, we must also put an intervention in before you they either tip into the addiction or they tip into going to somewhere else because of that lift, because it is a physical response.

And my understanding from work that I have seen and done with organisations that deal with heavy drinkers is that once your body has travelled that route, and it's an actual physical change in your cell structure and the way that your body physically responds to that alcohol, you can't close that gap. There's a word—it begins with an 'l' I think, and I can't remember what it is—but you cannot make your body physically, with ease, go back down that. It's a bit like if you are a very large person; even if you get skinny, you've still got exactly the same number of cells on your body. It's the same sort of principle; it's already there, so your ability to kick it without help, if you're in the heavy drinker category, is much harder. So, I'm just interested in trying to find the evidence, because I think that this is a good policy intent, but we've got to have the evidence base to make sure that we don't have unintended consequences that we could end up ruining it.

I think the first thing to say is that neither of us are addiction specialists; we do more public health research. But my assumption would be that if that point exists, you wouldn't necessarily be able to say exactly where it is on the line for the whole population. It's likely to be a highly heterogeneous point. Beyond that point, I think you're right that putting in place interventions at an early stage, possibly even before the policy comes in, to make sure you address any potential negative consequences before they arise, is the right approach. But beyond that, I don't think I can really add anything, because it's simply not our area. 

10:05

I haven't seen it, but I haven't looked either, because, as I say, it's not what we do work on. 

Yes, please. Just coming on from what Angela had said, an issue that I've previously brought up is that I'm concerned about the level of homelessness as it is anyway, but what I'm also concerned with is that people who are dependent on alcohol, on cheap alcohol, have a cheque for their rent in their hand, and where are their priorities going to lie? And I don't think that the priority is going to lie in paying that rent if they are dependent on alcohol to this extent. So, I think that we could increase the level of homelessness, and also the impact on family life, where a person is just affording to spend and give children a decent kind of life, when alcohol goes up. That decency decreases, not because of the fault of the individual. And it was brought to me that people have choices. Well, I disagree with this, because some people who are dependent don't have those choices, so I'm very concerned about the impact on family life and the children that will possibly be left behind. And, also, the homelessness section is going to suddenly increase, because we give people the responsibility of paying their rent directly. And I don't think that that is going to be the case. 

I think the thing to reiterate is that this is exactly what we're going to be focusing on in our evaluation of the Scottish policy. 

But, with all due respect, it's going to take time to get a level of how this is impacting upon people, and it's going to be too late for a lot of people in this situation. So, you can't compare something with nothing, can you?

You can't, but we'll be reporting to the Scottish Government on a rolling basis. In fact, we're already planning what our first briefing note to them is going to say before Christmas on the basis of our very early interviews, which actually suggest that, at this point, there may need to be more thinking about this, just in terms of getting services up to speed on what's happening in Scotland. The other thing to say again is that I think these are all legitimate concerns, but there are legitimate potential positives for this same population as well, in terms of treatment seeking, in terms of improved recovery, in terms of reduced consumption. And, of course, what often doesn't get talked about as much is the potential for fewer people to become dependent on alcohol in the first place, because the cheap alcohol isn't there to facilitate that slide into dependence. Obviously, not everyone becomes dependent simply because of cheap alcohol, but it's certainly a facilitator. 

If I can just add a couple of things. The first thing to say is that it's really important to remember that as great and significant a problem as dependent drinking is, dependent drinkers are a relatively small proportion of people even drinking at harmful levels. It's very easy to mischaracterise all very heavy drinkers as being dependent, and that's simply not the case. So, it's wrong to say that we can just say the policy won't be effective in very heavy drinkers because they're all dependent. Dependents drinkers are a different population, with their own issues, but they're not the same as a lot of—. There are a lot of very harmful drinkers who are not dependent. 

And just to reiterate John's point, it's very likely that you will see positives and negatives. So, there may be an increase—. Some people may become homeless as a result, but some people may also stop being homeless as a result. Some family situations may deteriorate, other family situations may improve. And on your point about saying that you can't make a decision based on nothing and the Scottish evidence isn't necessarily here yet, if you're particularly concerned about specific unintended consequences, then it may well make sense to specifically monitor those unintended consequences. So, if you're concerned about homelessness, then put increasing resources into observing, and potentially acting early on those problems. So, if there are specific problems that you're really concerned about, then monitor those problems and take action accordingly. 

One very, very brief point: our evaluation in Scotland will also be looking at the impact on people living in very highly rural areas, which, of course, will be an interesting issue for Wales as well. 

I think the questions on taxation have been covered by Rhun, if that's okay. 

Okay. We'll move on to Dawn, then, with the last couple of questions.

10:10

I just wanted to ask a quick question really on the mechanisms. You talked in your evidence about the level of minimum unit pricing being a key factor in whether or not this will work. I was surprised, actually, that it was a small margin—the difference between 45p and 50p, based on your evidence, was quite significant. But what do you think needs to be the frequency with which that's changed? So, if we introduce something—say we introduce 50p, how long do you think we need to let that run before we review that and see whether that's actually set at the right level?

I think there are two parts to that question. One is: at what point do you evaluate the policy full stop? I think it's worth following the Scottish approach here. They've had a very comprehensive and well-resourced programme. They're giving the policy five years to run, at which point the evaluations will report back, and that seems like a sensible compromise between our desire always to wait longer, because we want more data, and the political need to look at things in a reasonable time frame.

The other part of the question is: at what point should the policy be uprated? Clearly, if you leave it at 50p, or at whatever level it's introduced at, forever, that value is going to decrease in real terms simply through the effects of inflation and rising incomes. The answer is a bit difficult. There's been lots of discussion on this in Scotland. Everyone's kind of agreed that it would need to be periodically uprated, but no-one's really agreed how often or by what mechanism.

My suggestion would be that doing it yearly, or at least every two years, would be a sensible approach. Currently, alcohol taxes go up on an annual basis as a default assumption. Obviously, the Chancellor sometimes makes other decisions, but the assumption is that they will go up in line with inflation each year. You may want to simply do that for the minimum price.

Another option would be to use a more targeted index to uprate it, which might be something like the alcohol affordability index, so that the level of minimum price keeps pace with actual spending power.

Another option would be to say, 'Okay, well, we want to use the minimum price to actually change levels of alcohol-related harm further once we've put it in place.' So, you might look to get estimates from models like ours of what impact you could achieve by increasing it further. So, you could commission that kind of work as well to try and understand what level you might want to increase up to. So, there are different options there. But I think the basic principle of uprating the minimum price over time to ensure it maintains its value is important. We've seen in places like Canada and in the US, where they uprate alcohol taxes very rarely, that, if you don't uprate these things, they simply become ineffective or lose most of their effectiveness over time.

So, you're suggesting then that we would be looking at something along the lines of the minimum unit price in line with inflation, as an example—

Something like that, yes.

So, that could be every year. But the second aspect of that is: how long do you need to evaluate the programme? I think the Welsh Government is suggesting five years, the same as the Scottish Government. You don't think that's long enough.

I think it's a reasonable time frame. As an academic, we always want more time because we like to do things properly—we don't like to make rushed judgments. The other part is, as you've seen in our modelling, we suggest that the effects continue to emerge for as much as 20 years into the future. As a technical point, public health people are increasingly interested in complex—

But you think you'd get a sense of where this takes us after five years.

I think you would. You would have a sense of what it's doing to sales and you would have a sense of what it's doing to alcohol-related mortality and you'd have a sense of what's happening with some of the questions that are being raised over here about the impacts on dependent drinkers. I would continue to evaluate it beyond that point, but it will give you enough to go on.

It seems like a reasonable pragmatic trade-off between all the various concerns, I guess.

You're interested more than anything in the data. What about the legislation itself? You'll have studied the legislation, the Bill, that is being put forward here in the National Assembly. What would be your thoughts on where weaknesses might lie in that legislation to achieve what you think is possible?

We haven't actually looked at the Bill in great detail.

To be honest, we've been a little bit absorbed with the Scottish court case and trying to understand what's going on in the detail of that.

Is there a specific issue that you'd have concerns with?

No. I mean, as we look at this legislation, we need to decide whether we support it in principle and we need to decide on the detail of that legislation. If there are lessons from other similar pieces of legislation elsewhere, perhaps, that we need to consider, we need to take that advice from wherever. So, tell me about other pieces of legislation in other countries that have tried to deal with this and what we need from them to make sure we have included them in this.

10:15

There's nothing that I'm aware of that's come out of the Scottish experience, out of the Irish experience as it stands to date, or the Canadian experience, which is obviously at a provincial level, that kind of flags up major legislative challenges. The key issue, of course, is the matter of legality. I understand that there is some debate going on between the Welsh Government and the UK Government about the Welsh Government's actual powers to implement this policy. That's well outside our competence to comment on.

With regard to the EU issue, my understanding is that the main point that Wales would want to be aware of is this question of proportionality: is the impact on the free movement of goods proportionate to the public health benefit attained? That balance can be different in different countries. So, it is possible that the policy could be legal in Scotland and illegal in Wales, because, for instance, Scotland has a greater level of problems with alcohol than Wales does, thus the public health benefit may be judged to be smaller in Wales.

That said, my reading of the UK Supreme Court's judgment is that they recognise, essentially, this is a value trade-off, not an empirical analysis question, and that means that they give the Welsh Government considerable power to decide how much weight to put on that public health benefit versus the free movement of goods. In a sense, they can make a judgment as to what 'proportional' means, so long as it is a reasonable judgment or any reasonable person would judge it to be a reasonable judgment. 

So, I think, actually, you're on reasonably firm ground there. The alcohol industry may, of course, see it differently and ultimately take the Welsh Government to court, but our judgment would be that it seems you're on reasonably firm ground. 

I think the only thing that I might add to that is that the question of proportionality is all about: is it a proportional approach compared to other approaches to achieve the aims of the policy? I think the only other concern that one might have is in terms of specifying the aims of the policy specifically, because, if you specify them extremely broadly and just say that the aim of the policy is to reduce harm, there are many ways in which you can reduce harm. However, minimum unit pricing is seen as a good policy approach because it's extremely targeted, because it reduces health inequalities— you know, there are many potentially desirable facets to the policy, and the more specific you are about the aims of the policy in terms of those potential outcomes, the more proportionate you might say the policy is compared to other alternatives to achieve it, because the harder it is to design another policy that will achieve the same kind of broad suite of effects.

Is there anything else, as part of the suite of tools that can be used to bring down alcohol consumption, that could be included in legislation? For example, there needs to be, presumably, alongside this, a programme of education in order to try to drive changes in cultural attitudes towards alcohol. Is there room in legislation to demand investment in educating the public about alcohol consumption, for example?

I think there certainly is. Obviously, there are money questions, but you can certainly make those demands/recommendations. Education on its own has been shown not to be particularly effective. The argument is largely that it's a useful thing to support other policies, but the argument has been—in the UK, at least, at Westminster—if that's all the Government does, it's not likely to make a difference, because there are important factors like the availability of cheap alcohol.

Other things you might well include, as I said earlier: investment in treatment services; investing in screening and brief intervention; greater public health involvement in the licensing system to ensure that, for instance, you don't have what used to be called 'an overprovision' of alcohol outlets, such that they compete very strongly on price and perhaps inappropriate behaviour to get customers in and sell them more alcohol. You might look at the role of sponsorship by alcohol companies of sport in particular and things that appeal to young people. There is good evidence that all of these things, and the wider marketing of alcohol, might be effective in reducing the harm caused by alcohol.

If I could just briefly come back to your previous question as well, one other thing that's been debated in Scotland is that, because minimum pricing obviously isn't a tax, the money goes to retailers not to the Government. One thing that Scotland did look at when it first proposed a policy was what they called a 'large retailer levy', which essentially was a way of clawing back that money from the large retailers and to stop them using it, for instance, to increase the promotion of alcohol to try and cancel out the effects. That kind of went quiet during the court case and I'm not sure where that policy is with the Scottish Government now. I'm not sure if they have the political power to bring that in or not at this point, but it's the kind of thing you might look at as a way of ensuring that this kind of windfall that will come to supermarkets—and it is one of the limitations of the policy—might be used more productively by Government.

10:20

Just to highlight a couple of the things that John picked up on, you might—particularly if you're looking at a holistic approach to alcohol policy with minimum pricing as a part of that approach—be particularly interested in focusing other policies on the areas that minimum unit pricing might not touch to such an extent, or when minimum unit pricing might have unintended consequences. I think treatment services is a really important one.

But also, in relation to licensing, minimum unit pricing is unlikely—unless you set the level much higher than the levels that are under discussion, it's very unlikely to have much effect on the price on alcohol in the on-trade. So, it's very unlikely to affect the price of alcohol in pubs. If your biggest concern is harm related to pubs, in terms of particularly—the problem is always portrayed as one of young people drinking in pubs; if your concern is about young people drinking in pubs and harms associated with that, then you may want to do other things specifically to target that, because minimum unit pricing is not the silver bullet that will solve all of the problems, and that is a particular issue that it's unlikely to have much impact on. It might affect preloading, in terms of people drinking lots of cheap alcohol before they go out, but it's unlikely to change very much in terms of pub drinking.

Ocê, diolch yn fawr iawn. Dyna ddiwedd y cwestiynau. Dyna ddiwedd y sesiwn. A allaf i ddiolch yn fawr iawn i chi unwaith eto am eich tystiolaeth ysgrifenedig a gyflwynwyd ymlaen llaw, a hefyd am ateb y cwestiynau ac am eich presenoldeb y bore yma? Fe allaf bellach gyhoeddi y byddwch chi'n derbyn trawsgrifiad o'r trafodaethau yma er mwyn i chi gadarnhau eu bod nhw'n ffeithiol gywir. Gyda hynny o ragymadrodd, fe allaf ddiolch yn fawr iawn i chi unwaith eto ac fe allaf gyhoeddi i'm nghyd-Aelodau y cawn ni doriad byr rŵan am ryw dri munud cyn dechrau'r sesiwn nesaf. Diolch yn fawr iawn i chi.

Okay, thank you very much. That is the end of the questioning and the end of this session. I'd like to thank you again for your written evidence, which was submitted beforehand, and also for answering the questions and for your attendance this morning. You will receive a transcript of these discussions so that you can check them for factual accuracy. With those few words, I'd like to thank you once again and I'll let my fellow Members know that we'll have a three-minute break now before the start of the next session. Thank you very much.

Gohiriwyd y cyfarfod rhwng 10:22 ac 10:28.

The meeting adjourned between 10:22 and 10:28.

10:25
3. Bil Iechyd y Cyhoedd (Isafbris am Alcohol) (Cymru) - Sesiwn Dystiolaeth 5 - Sefydliad Materion Economaidd
3. Public Health (Minimum Price for Alcohol) (Wales) Bill - Evidence Session 5 - Institute of Economic Affairs

Croeso nôl i sesiwn ddiweddaraf y Pwyllgor Iechyd, Gofal Cymdeithasol a Chwaraeon yma yng Nghynulliad Cenedlaethol Cymru. Rydym ni wedi cyrraedd rŵan eitem 3, sef parhau efo'n trafodaethau i fewn i Fil Iechyd y Cyhoedd (Isafbris am Alcohol) (Cymru). Dyma sesiwn dystiolaeth rhif 5, ac o'n blaenau ni mae'r Sefydliad Materion Economaidd. Rwy'n falch iawn o groesawu Chris Snowdon, pennaeth economeg ffordd o fyw y Sefydliad Materion Economaidd. Diolch yn fawr iawn i chi am eich presenoldeb. Diolch yn fawr iawn i chi hefyd am ddarparu papur ymlaen llaw, ac yn seiliedig ar hynny fe awn ni'n syth fewn i gwestiynau. Nid oes angen cyffwrdd a'r microffon, mae'n gweithio'n awtomatig. Felly, y cwestiwn cyntaf—Caroline Jones.

Welcome back to the latest session of the Health, Social Care and Sport Committee here at the National Assembly for Wales. We are now at item 3, continuation with our scrutiny of the Public Health (Minimum Price for Alcohol) (Wales) Bill. This is evidence session 5, and before us we have the Institute of Economic Affairs. Welcome to Chris Snowdon, head of lifestyle economics at the Institute of Economic Affairs. Thank you for your attendance. Also thank you for providing the paper beforehand, and, based on that, we'll go straight into questions. You don't have to touch the mike; it works automatically. The first question is from Caroline Jones.

Diolch, Cadeirydd. Bore da. Good morning. To what extent do you think that the relationship between price, consumption and harm has been evidenced, for example, in the Sheffield model, and also other research and literature that you've come across?

I think the evidence that price affects consumption is very strong. It's one of the most basic economic principles, really, all other things being equal. But, of course, things aren't always equal, and, particularly in a complex area like alcohol and alcohol-related problems, there are a lot of other factors that come into play. But, yes, all other things equal, you put the price of something up, consumption will tend to decline. The interesting thing is who is leading that decline in consumption. Is it, in this instance, moderate drinkers, or is it very heavy drinkers, or is it dependent drinkers? That's more of a difficult question. The link between consumption and harm—yes, sometimes there is a link. As an individual, obviously there is a link. If you go down from drinking 100 units to 10 units, there's a very clear distinction. But, at the population level, again, it's not so clear. There are plenty of examples where harm and alcohol-related deaths have broadly followed trends in overall consumption, but there are several conspicuous examples where they haven't. The United Kingdom, over the course of the last 15 years, is one of them. There has been this big decline in alcohol consumption—the biggest decline in alcohol consumption that we've seen since the 1930s—but not much evidence of a commensurate decline in harm. My view of that is that it's probably because it's not the heaviest drinkers who have reduced their consumption. There has been an increase in teetotalism—the number of people who don't drink at all—although of course you would expect that to have some impact on reducing harm. But I think mainly what it is is people who weren't drinking particularly heavily before drinking a bit less. I see no reason why that should have a positive impact on health, although it's sometimes assumed in public health circles that it would do.  

10:30

The Sheffield modelling work does seem to show that minimum unit pricing would reduce consumption among the most harmful drinkers, the most hazardous drinkers, because they favour the cheaper price—you know, the very high-strength—alcohol, and that would be the drink that would be most affected by the introduction of this measure. So, is there any evidence that you feel contradicts this?

Yes, I think most economic evidence on alcohol contradicts it. I think that one of the flaws of the Sheffield model is that it wrongly assumes that people who are heavy drinkers are more price sensitive than people who are not. That is simply not what the economic evidence shows. It shows that moderate drinkers are most likely to reduce their consumption in the face of a price rise, heavy drinkers less so, and dependent drinkers almost not at all. I was watching the session earlier, and I was watching the sessions here last week, and I was struck by the confusion about who this is supposed to be targeting. If this is a targeted measure, who is it aimed at?

We've heard over the years that almost every group, supposedly, is targeted by this, and then people say, 'Oh well, actually, no, it's not.' So, we heard it was about public disorder, and then people said, 'No, it's not, because these people are spending much more than the minimum price in the pubs and clubs of our cities.' Last week, Duncan Selbie from Public Health England said that it was targeting young people, and Colin Angus, who you just heard from, tweeted, 'No, it's not targeting young people.' And, we've now heard—as was also mentioned here last week—that it's not targeting alcoholics. So, who, actually, is it targeting? It's not targeting moderate drinkers either, supposedly. Apparently, it's targeting people like me, basically—people who drink more than the Government guidelines but aren't dependent on it and don't really suffer any harm or cause any real problem.

It seems to me that if it's not actually affecting dependent drinkers, the very heaviest drinkers who are actually dying from this, I'm not sure what the purpose of it is, other than just trying to launch a doomed attempt to try and get me, people like me, to cut down on their alcohol consumption a little bit. We are only talking, I think, from the model itself, which I think overstates things—. We are only talking about a 4 per cent drop in consumption anyway.

So, you are challenging the basis of the Sheffield research.

Yes, I would challenge any economic model. The reason I first wrote about this is that I just got sick and tired of hearing people talk about this as if it was the final word on evidence-based policy and is science. It's not science. It's statistics. It's built on a bunch of assumptions that may or may not be right. We have economic modelling all the time. We have just had the model changed predicting when real wages are going to rise to above the pre-crash peak last week, which was different to the previous prediction made six months earlier, which was different to the prediction made a year earlier.

Now, I'm not saying there's no purpose to economic modelling, but models are always wrong to some extent, and I got annoyed at the way it was treated as being science: 'Science says that it's going to save exactly this many lives—55 lives in its first year; 62 in the second year', and all of this kind of stuff. This spurious accuracy, I thought, was annoying. The basic assumptions, which you've already mentioned, are not in any way implausible. It was just the way it was presented as being fact, essentially—that these things, which were based on assumptions and predictions, were fact.   

Well, yes, we can come on to the Canadian stuff. I mean, that is really—. The stuff that has come out from Tim Stockwell in Canada is basically worthless. The big claim a few years ago, based on British Columbia, where they don't actually have minimum pricing as we understand it, for a start off—it's a rather different thing over there—but the claim was there was a 10 per cent increase in the minimum price in British Columbia and this led to a 33 per cent reduction in alcohol-related deaths. Now, if that were true, that would be extremely striking. And most normal people, I think, would interpret that to mean three things: (1) that there is minimum pricing in Canada, (2) that the minimum price went up by 10 per cent, and (3) that the number of alcohol-related deaths dropped by a third. That's what most people, I think, would assume from that, but that's not what they mean always in public health.

What it actually involved was modelling what a 1 per cent increase in minimum price would do to the number of alcohol-related deaths, and then extrapolating that by essentially multiplying it by 10 and coming up with this figure of 33 per cent. But if you look at the number of alcohol-related deaths in British Columbia, it didn't fall at all. The numbers actually rose because the population increased, and the actual rates remained exactly the same.

So, the stuff from Canada really is fantasy stuff, similarly claims about crime falling. If you're going to do a study looking at a fall in crime, bearing in mind that crime has been falling all over the western world since the 1990s, you need to compare it to a similar country. And they didn't do that. They just said, 'Well, there was this fairly mild change, actually, in the minimum price round about 2005 or so, and crime fell thereafter'. I mean, this is a basic fallacy. If something happened afterwards, it happened—

10:35

Yes, completely.

Well, because it's not true. Because you can look at the actual number of hospital admissions and the number of alcohol-related deaths and they didn't fall. So, whatever the modelling might say, I don't think that should be of interest to any policy makers if it didn't actually have an effect on people's lives. The Sheffield stuff is different because clearly it's a projection and it could easily be true—any kind of model could be true. I think there are specific reasons to think that that model overstates the benefits of minimum pricing, at the very least, for two reasons that I mentioned in the submission; I'll just mention them again very briefly.

One is this issue I've already discussed about wrongly assuming that the heaviest drinkers are most likely to reduce their consumption in the face of a price rise. But the other thing is that, although the model itself is based on evidence—. It's not being entirely plucked out of thin air. It's not a synthesis of other people's evidence, as was said earlier on, but it does take certain economic estimates from other people's research and it builds a model on them. Now, that's very different to being a synthesis of a broader bank of research, but these numbers are not entirely plucked out of thin air. So, we do have price elasticity figures for all sorts of different drinks from all around the world, and they show us what happens when the price of beer, for example, goes up.

What we don't have, of course, because minimum pricing has never been tried, is any idea of how people react when the entire budget market for alcohol is wiped out. And it will very interesting—I assume you're going to do it here in Wales, and they'll certainly be doing it in Scotland. It will very interesting to see what happens there when you've got the very cheapest drinks—. So, a large bottle of white cider, which you can buy for something like £4, that's suddenly going to go up to £11. The effect of that will be that people will effectively be buying on the basis of taste, rather than price. People aren't drinking white cider because they have an inherent desire for vaguely apple-flavoured alcoholic drinks. They do it because the tax system over the years has hugely favoured cider producers, and cider producers have found a way of making white cider very cheaply.

Street drinkers didn't used to drink white cider. They used to drink spirits or meths or antifreeze or goodness knows what. My concern is that they will return to drinking those products because it will be about getting hold of the cheapest alcohol. But those who don't, those who continue to drink normal, legal alcohol from the off-trade—they will then have a choice. If you're purely interested in getting the most number of units for your buck, you can get brandy, vodka, whisky, cider, beer—everything will be essentially the same price. But I don't think, just because you're making the cheapest stuff less cheap, people will stop drinking it in large numbers. There will be just some kind of substitution effects.

And those are not adequately dealt with in the model, because we don't have the data. So, white cider, for example: they just assume that white cider drinkers—cider drinkers in general—are again the most price-sensitive people. Because that's what the economics say when you're dealing with a pint of cider in a pub, and it's probably quite true if you're out on a Sunday afternoon and you go to the pub and you've got beer, cider, lager—'Oh, I quite fancy a cider'. If the price of that cider goes up by 20p, you may think, 'Oh, I'll have the lager instead.' So, it's true, probably, that the average cider drinker is quite price sensitive, but you can't apply that to people drinking very, very strong cider who are dependent drinkers, and we don't have any evidence on these people's price elasticity. Common sense tells us, I think, that they are actually not very price sensitive and they will drink at any price.

10:40

Can I just ask my last question? Obviously, you've got different views than in the earlier session we had—the people who presented there. Is there any research that backs up what you're saying? Any research that's been done anywhere?

Not on the drinkers of white cider. I'm not aware of any research in terms of price elasticity. There's not even that many price elasticity estimates for cider in general. But on the point about heavy drinkers, yes, there's a lot of research gone into it, and the conclusion is that if you're a heavy, dependent drinker, your elasticity of demand is virtually zero, which is to say you will not change the amount you drink in the face of much larger increases in price than we're talking about here.

Yn symud ymlaen, mae'r cwestiynau nesaf gan Lynne Neagle.

Moving on, the next question is from Lynne Neagle. 

Thanks, Chair. Have you got anything you'd like to share with us in relation to your thoughts about the impact of this proposed legislation on low-income groups?

Yes. I mean, I think it stands to reason that measures like this will hit people on low incomes the most, for two reasons. This is not a tax, obviously, but any kind of sales tax does tend to take the most from the poor as a proportion of their income because they have lower incomes to begin with. But also, in this instance, it's doubly regressive, because the people on low incomes are much more likely to drink these particular products.

Again, I think the Sheffield model here is hugely unrealistic. I think the claims about moderate drinkers only paying an extra two or three pounds a year are enormously unrealistic. They define a moderate drinker in their study as being somebody who drinks five units a week. I think that's a pretty questionable definition of a moderate drinker. They do that because they just take everybody who drinks below 14 or 21 units and average it out. Well, yes, fine, there are plenty of people who drink almost nothing, but I think if you're going to look at the impact on a moderate drinker, you should look at somebody who's drinking 14 or 21 units rather than just average it out to five. But even then, I just don't think it's realistic.

The price of a cheap bottle of whisky, for example, will, I understand, go from about £12 to £14. So, if you drink one of those a year, you're going to be paying £2. If you get two of those a year, you're spending much more than the Sheffield people say people on low incomes will be spending additionally. I think it's because they overestimate how much their consumption is going to decline in the first place. But it's quite simple to define a moderate drinker however you see fit: define them as somebody who gets eight cans of lager from the off-licence each week and those cans are below minimum price—which most beer is, in the off-trade—and you can very quickly do a back-of-an-envelope calculation to see how much extra they're going to be paying. It applies equally to wine. It applies, certainly, to spirits, and it obviously applies to things like strong cider. So, you can do those calculations yourself based on whatever you consider an average low-income, moderate drinker to be.

You've described it as regressive and you've given some financial information to back that up, but do you think it really can be described as a regressive policy when the committee has heard evidence that the burden of ill health from alcohol is higher amongst low-income groups and that heavy drinkers in low-income groups are likely to see the biggest health benefits from this legislation?

Yes, I do, because the word 'regressive' has a very strict economic meaning, and it means taking a disproportionately larger share of income from the poor than it does from the rich. The people who advocate for this policy try and change the subject and say, 'Oh, yes, but the poor get more benefits.' Well, as I say, maybe they will; maybe they won't. We won't know until it's been tried. I think the health benefits are entirely speculative, but the financial impact is inevitable. It's just inevitable. We all know that people are not going to stop buying alcohol altogether as a result of this policy. I've given the figures in my submission, which come from Sheffield, about the share of off-trade alcohol that is currently sold below 50p. It is not a small, niche market of a few white ciders. This is the vast majority of all the beer and cider sold in the off-trade; it's 40-odd per cent of the wine and about two thirds of the spirits. It's not even that it's going to hit people on low incomes most. It's going to affect everybody, bar the rich. It's a policy that actually exempts the rich. So, to call it 'regressive', yes, it is regressive in the economic sense. But it's not just low-income people, it's people on middle incomes as well who will be adversely affected by this. And that is inevitable from an economic financial point of view. Whether it has knock-on benefits to people like me who, apparently, are targeted by this remains to be seen.

10:45

So, if I just ask about young people, because you seem to be suggesting that, if we go ahead with this legislation, our young people are going to be driven to take more drugs. Is that something that you can evidence?

Well, it's certainly possible. I heard one of your witnesses last week saying that it's simply not true that drugs are a substitute product for alcohol. Well, they are. Every product is, potentially, a substitute product for something else, and drugs certainly are. They're fairly close substitutes in actual fact. You can see this throughout history: during prohibition in America, the use of cocaine and heroin went up because people couldn't get hold of alcohol so easily. In the 1990s, when ecstasy became popular with young people, drinking rates declined. So, yes, they are substitutes. I can't give you the exact cross-price elasticity of spice compared to cider and so on, but in some instances it clearly is the same people consuming these products. And if you put the price of alcohol up enough and the price of drugs remains the same or falls, there will be some crossover with people substituting one for the other.

But you've spent this morning sort of trying to debunk the evidence that we've been given from other organisations, yet all you're able to say in relation to young people is, 'It's possible'. It's not just about price comparison, is it? It's also about the accessibility of those products to young people—you know, how easy it is to get cheap alcohol as opposed to a young person being able to go out and buy spice.

Well, young people are able to get—. If you mean underage people, they're often able to get drugs more easily than they are alcohol. There was some research done in Brixton recently where virtually every young person they spoke to said it was much easier to get hold of cannabis than it was to get hold of alcohol. This is one of the problems with the war on drugs. You get rid of all regulation and you end up actually making it easier for people to get things. So, yes, in some instances, it might be easier to get alcohol, but in a lot of instances it's easier to get drugs.

Again, look at America, where the legal age for buying alcohol is 21, and it's pretty strictly enforced—people have fake IDs and so on, but it's fairly strictly enforced—and, as a result of that, huge numbers of people smoke cannabis until the age of 21, because it's a substitute. It's something to do on a weekend. I mean, clearly these things are substitutes.

Yes. Thank you for your evidence, but I do want to press you on this because you do state in black and white,

'It is likely to lead to a shift from cider to spirits for dependent drinkers. A shift to the cheapest illegal drugs is also highly plausible among some groups'.

And this is my fear. My fear is the unintended consequences of a good policy objective. I pick up your point, perhaps, about the lack of clarity as to at whom and how this is aimed, but I am looking for evidence and, therefore, are you able to point us in the direction of any evidential work that's been done on the addiction pathways that are driven by cost factors? We do know that addiction does have a pathway, and I have a concern that people might find it easier to move on to other products if they can't get hold of very cheap alcohol, but I do want to see what the evidence base is on that. The previous people weren't able to point me in that direction, and I wonder if you're able to give us a direction where we can go and have a look for the evidence on this, because I think it's really important to be evidence based when we're looking at this Act.

I agree, and I would be quite happy to e-mail you links to—

—some studies when we've concluded here. You know, this has been looked at. Economists are terrible at predicting the future, but once something's happened, they're pretty good at identifying the unintended consequences, in retrospect. So, yes, there is evidence of that.

10:50

Just before I ask my main question, which is about retailers, or the industry really, and how the industry might respond to this, in terms of illicit street drugs compared to alcohol, because I don’t know—I don’t buy street drugs, so I’ve no idea how much a thing of spice costs, or a line of coke, or whatever—do you know what the relative price difference is between cheap white cider and a pocket of spice or something?

I don’t know how much spice is these days, since it was banned, but I do know that there is what you could probably call an epidemic of it amongst the homeless populations of several cities such as Manchester. And if something is consumed in large numbers by the homeless population, I think it’s safe to assume that it’s pretty cheap.

Yes. It might be worth us getting some information on that, because certainly the evidence that we took last week was suggesting that, actually, street drugs are nowhere near as cheap as cheap alcohol—even the cheapest cheap. So, you know—

Well, I don’t think that’s true. I mean, you can get ecstasy for less than the price of a three-litre bottle of cider.

Yes. But, then it’s the time that you get on the high—how long you can be drunk for and how long you can be high on a tablet of ecstasy. So, that also comes into the equation, as well.

But I just wanted to ask you a little bit about the alcohol industry, and how they might respond to the introduction of minimum unit pricing. Because they stand to make a potential windfall out of this, because it’s not a tax, this is profit and it will go to the industry. Have you given any thought to how they might respond to that and whether there’s any scope for Government, I suppose, to do some work with retailers in terms of using the benefit of that windfall to help achieve these objectives?

Yes. I’m not convinced that there will be a huge windfall for either the manufacturers or the retailers of alcohol. I understand why people assume that there would be, but if you take my example, again, a £4 bottle of cider is suddenly going to be £11. Does that mean that somebody’s pocketing £7? No, it doesn’t. It means that nobody’s going to buy white cider anymore. The bottom is going to completely fall out of that market; it’s not going to exist, I don’t think, anymore.

Effectively what will happen is all the stuff that we now see as cheap—the budget end, Tesco’s own brand beer and things like that—just won’t be made anymore, or if it is made, it will be so fundamentally improved as to take up all that excess profit, or they’ll spend so much more on advertising to justify the higher price. But either way, there won’t be any excess profits. Competitive markets don’t generally allow excess profits of that sort.

There might be some marginal economic benefits for industry if—and I don’t know for a fact that this is true—the mid range, which will soon become the bottom end, if there is more profit per unit at the mid range and if consumption doesn’t decline very much—as I say, it’s only predicted to decline by 4 per cent, even by Sheffield—then, yes, there may be some fairly marginal benefits, but I wouldn’t expect this to hugely benefit the alcohol industry. I know the alcohol industry is split on this, because obviously the Scotch Whisky Association made the legal challenge initially. Most Scotch whisky will not be affected by this—the high-end stuff won’t be affected by it. Quite a few of the premium cider makers are staunchly in favour of this policy, and much of the hospitality industry is in favour of this policy, because they think that it will benefit them. I think they’re wrong about that, actually. So, they’re a bit split on it. You can see why different sides of the industry think that they might benefit and are lobbying one way or the other based on that. But I wouldn’t expect to see much in the way of excess profits for industry; I would just expect to see a lot of the alcohol that we see today disappear.

Good morning. You, like many of us, feel, clearly, invincible; do you think that alcohol is harmful?

Yes, of course it’s harmful. Yes. It’s responsible for a lot of deaths when taken in excess, but it is principally a problem for dependent drinkers, and dependent drinkers are not greatly influenced by price initially. There are huge psychological issues to addiction, not just with alcohol, but with gambling and with drug addiction. These things are not really about the product, they're about the person, and I'm, in no way, any kind of addiction expert, but the people who are understand that you've got to help people get over their problems before they can give up the addiction.

10:55

If we can park dependent drinkers—I think it's quite clear that alcohol consumption for dependent drinkers is very, very damaging—do you accept that alcohol is harmful to moderate drinkers, to you and me? I'm sure I drink too much, certainly some weeks, and I'm pretty sure it's not doing me any good. 

Well, I don't consider myself a moderate drinker. Whether it's doing me any harm or not, I don't know. My liver is in tiptop condition, I believe—

I do, more than 14 units a week, yes, certainly. But people who don't, people who are classified as moderate drinkers under the current guidelines, they are benefiting in their health—there are health benefits from drinking moderately. Moderate drinkers live longer than teetotallers, on average.

What about the category that we read about in our briefing papers—I'm sure you've read them as well—of problem drinkers? Again, I put my hands up, I'm sure I would fall into the category of problem drinkers on some weeks—where you drink too much. Is that harmful? If you're—?

Getting drunk?

No, not necessarily getting drunk. For people drinking 30 units a week, that's classified as—[Inaudible.]

That would be classified as—

Hazardous drinking. Problem drinking is a rather different thing that can't just be defined by the number of units alone.

Do we need to bring down, if possible, through persuasion, legislation, whatever it might be, how much people drink? Do you believe that alcohol is a societal problem, long term, perhaps, that needs somehow to be addressed and for us to try to reduce how much people drink?

Not per se, no. I think there are times when the Government might need to intervene. I think when there are significant negative externalities and excessive costs to the police service and the health service, then it's quite right to have a tax on alcohol to compensate for that, to make up for that. I think if you're dealing—I know you want to park the dependent drinkers, but they're, really, the only people you can justify intervening—. I am not a dependent drinker, even if I drink more than the guidelines, but that's not really any business of the Government. They are only guidelines, at the end of the day. They're only recommendations. Dependent drinkers are a little bit different, because their rational choice is hampered by their addiction, and that is a much more difficult area, and I think, because of that, because they're not always making a rational, self-interested choice, you can justify the Government spending money trying to help them in some way. But I'm not in favour of coercing people or manipulating people who are not asking for help, who are living perfectly happy lives, who just happen to be exceeding some rather arbitrary guidelines.

I think the fact that you're here giving evidence to us—in an earlier session, we had a different viewpoint—shows that there are two sides to this argument. I'm trying to keep an open mind, as we look at this legislation, but in terms of the evidence about the harm caused by alcohol, what I see is pretty universal agreement that even moderate alcohol consumption over a long term can be pretty harmful, and that if we can bring down people's consumption of alcohol, moderate drinkers, that would be pretty good for society and us as individuals.

No, moderate drinkers are not doing any harm to themselves. I don't want to get into the whole debate about the guidelines, but the guidelines are supposed to be based on—. Your 14 units, as it now is, are supposed to represent, basically, the same level of harm as a teetotaller would have. So, if you drink a bit less than that, you actually live longer than a teetotaller because of the benefits to the heart and reducing stroke and diabetes risk, and then you have this J-curve, and it comes round to some point, which I think is a little bit beyond 14 units, but, be that as it may, it then goes up, and it goes up very sharply when you get to the very high levels of consumption. So, there are absolutely no benefits to be had from getting moderate drinkers to reduce their consumption, and I think you would go a fair bit beyond what they call moderate drinking, and you've still got no real problems. At the end of the day, it's down to the individual to decide—me and you can decide to drink more than the chief medical officer says, and that's  no business of the Government's, unless we are driving costs onto other people.

11:00

But we know that alcohol consumption, and excess alcohol consumption—let’s put moderate drinking to one side, if you count moderate as being within Government guidelines, and accept that there’s a significant proportion who are drinking more than the Government would advise you do, and probably because of the difficulty of getting people to be honest about how much they drink, it’s probably a larger number than the statistics tell us. But if we are to bring that consumption down, do you agree that there is a place to look at how the economics of it can be that stick, and whether it’s worth doing for our long-term health? Because, as I say, all the medical evidence I see suggests that people who aren't dependent, aren’t heavy, heavy drinkers, but drink more than they should, are causing themselves harm, costing the NHS, costing public services harm in the long term, and if that can be addressed through taxation, minimum pricing, it’s worth doing.

Well, I think the costs to the NHS are far exceeded by the alcohol duty that we currently receive. The estimates of cost to the NHS in the UK are £3.5 billion. Alcohol duty revenues are about £13 billion. So, teetotallers are making a tidy profit from drinkers. We have amongst the highest rates of alcohol duty in Europe, if not the world. We pay 40 per cent of all the alcohol duty in the EU. We are not undertaxed from that point of view.

In terms of overall health, assuming for a moment that people who are drinking 30 or 40 units are in some way impinging on their long-term health, again, it’s a trade-off that individuals are free to make. Health is not the only thing that matters in life. People want to enjoy themselves, and if that means putting a fairly small risk in front of them, in a free society you should surely be able to do that. Any putative health benefits need to be looked at in the context of the harm that governments do to people by taking a lot of money off them.

Alcohol duty at least can be used for public services. Minimum pricing needlessly puts up the price of a product. I haven’t done the maths in Wales, but it certainly will cost Welsh consumers tens of millions of pounds a year. And for what, really? Even if we assume it’s about £50 million a year, which I think would be an understatement, you’re still valuing these 55 entirely theoretical preventable deaths at £1 million a year. There are surely better ways of spending £1 million. You could certainly prevent more deaths by spending £1 million.

For what it matters, I agree with you that these projections of costing £2 more per moderate drinker seem—

Implausible.

Implausible, yes. Certainly if they continue to drink as they do now. If somebody drinks three cans of lager on a Friday night instead of four because they’ve become more expensive, maybe, actually, you could expect them not to spend any more. Do you think that there could be a drive for people to either drink a little bit less, or drink products with lower alcohol content, or if they want to drink to get drunk, and people do, that they perhaps drink to either get a little less drunk, or that they drink to get drunk less often, that there are things that people could do, which surely would be the purpose of doing this, that would actually be beneficial to them in the long term in terms of their health, and it would mean that they’re not spending more money?

Well, I think that really is the story of the last 10 or 15 years. Alcohol companies have been reducing the strength of a lot of their products. Alcohol consumption has fallen by 18 per cent, and it’s come about, mainly, I think, from the kind of examples you give there—somebody deciding to have three cans instead of four. But it hasn’t had any effect on alcohol-related deaths, and there was never any real reason to believe it would do, because it’s having no effect on the people who are drinking genuinely dangerous amounts. I don’t think minimum pricing will have any effect on those people, either. In fact, all the sessions I’ve heard from this committee have more or less admitted that it won’t have any effect on people drinking vast quantities of alcohol, and they really are the people ending up in hospital with liver cirrhosis. It’s not the people drinking three or four cans on a Friday night.

That’s not what I’ve heard in the committee, I must say. I think you’re absolutely right—I think that it has been a regular topic that this will not hit dependent harmful drinkers. But the modelling, certainly, and the experience from elsewhere, would suggest that non-dependent heavy drinkers would reduce their consumption, which would be a good thing. 

11:05

But, again, we've got real-world evidence from the UK where you've seen alcohol consumption fall by a fifth and alcohol-related deaths stay exactly the same. 

Yes, you would know. There's a bit of a lag with these things, but nowhere near—. It's not like smoking or something, where there's a 20 or 30-year lag. You see most of the effect, or you're supposed to see most of the effect from these things within the first year, and certainly within five years you will see the overwhelming effect from this. A lot of this stuff you actually see overnight, because some of it's acute harm—things like drink-driving and domestic violence and so on. So, you will be able to see, if you do this here in Wales—you should be able to see those harms declining very, very quickly, if they come about at all. 

And that's something that we can and are taking evidence on from doctors—the British Medical Association, people who can look at the long-term effects. Okay.

Are you coming at this from a libertarian angle largely? You started this morning talking about—. You're making pretty good arguments on the economics and what you perceived would happen in terms of people's behaviour, or would not happen as a result of people's behaviour if this became law. You seem to have moved to either being a health expert and thinking alcohol isn't that harmful, or you just don't think that governments should get involved in this kind of thing, and if people want to kill themselves by drinking too much, well, let them do it. 

The economics angle is ultimately libertarian, in a sense, because economists believe that social well-being is maximised by allowing people to choose freely in a free market, with the caveat that if there are negative externalities, or if you have consumers who are severely misinformed or severely irrational then, yes, there could be a role for Government intervention. But generally, markets produce the best outcomes. We assume that people know their interests better than the Government does, and if the Government assumes that we'd all be happier drinking no more than 14 units a week, the Government is probably wrong. 

Has the increase in the price of tobacco helped drive a decrease in tobacco use? 

Yes, it certainly has. I think it's been probably the main thing, in addition to education, that's reduced smoking rates over the years. 

I've never said that it's ridiculous or implausible that minimum pricing is going to reduce consumption. I think it probably will reduce consumption, certainly legal consumption; we haven't even gone on to the issue of people coming over the bridge with a car stocked up with alcohol, and that's not in the model because it wasn't anticipated, but it's a very important thing. Everyone I've spoken to about this have said, 'Well, what about the cross-border traffic between Scotland and England and England and Wales?' That's a major issue. We're talking about crime. Again, the Sheffield model models to the nearest decimal point how much crime is going to fall by as a result of minimum pricing. It doesn't take any account of the possibility of people selling black market alcohol, even though the profit margins are real and quite substantial. 

My understanding is that there has been some modelling on the cross-border trade, for example, and that it's an issue of proximity: if you're living very near the border, yes, you probably will, but if you're living 20 miles away you probably won't travel across the border. 

Well, you'll have to have a word with the guys who run the alcohol shops in Calais about that, because people are prepared to travel quite a long way. 

But I don't go to Calais to buy alcohol; it's too far. 

No, but people do, or they did before they put the taxes up. They go to Belgium now. 

Right, but there's the point, isn't it? If Calais was 5 miles away, yes, I'd probably go there to buy some wine every now and then, but it's too far. So, these kinds of things can be modelled. 

It's not in the Sheffield model. 

We'll move on from your drinking habits, shall we, Rhun, and close this session with some questions from Jayne? 

Thank you, Chair. Some of them have been covered in your answers to Rhun, but the explanatory memorandum part of the Bill acknowledges that the only way to test the model is to introduce it, and you yourself said that we won't know the effects until it's tried. There are provisions in the Bill for evaluation and sunset clauses. To what extent does that address your concerns?  

I'm always happy when there's a sunset clause to a law. So, yes, I think it's a good thing. I think it should be looked at. I hope it will be looked at independently by more than one group of people, and that every aspect of it will be looked at, not just trying to do more modelling on the health benefits, but also looking at these things like drugs and the cross-border trade and crime, and, in particular, the effect on low-income households. 

11:10

A oes unrhyw gwestiynau eraill? Mae pawb yn hapus. Diolch yn fawr iawn i chi. Dyna ddiwedd y sesiwn. Diolch yn fawr iawn i chi am eich presenoldeb. Diolch hefyd am y dystiolaeth ysgrifenedig gwnaethoch chi gyflwyno ymlaen llaw. Gallaf bellach gyhoeddi y byddwch chi'n derbyn trawsgrifiad o'r trafodaethau yma er mwyn i chi allu eu gwirio nhw a chadarnhau eu bod nhw'n ffeithiol gywir. Ond, gyda chymaint â hynny o ragymadrodd, a allaf i ddiolch yn fawr iawn i chi unwaith eto? Diolch yn fawr. 

Are there any more questions? Everyone's happy. Thank you very much. That's the end of that session. Thank you very much for coming today, and thank you also for the written evidence you submitted earlier. I will just let you know that you will receive a transcript of the discussions today, so that you can check them for accuracy. Thank you very much again for coming in today.

4. Papurau i'w nodi
4. Papers to note

I'm cyd-Aelodau, mi wnawn ni symud ymlaen at eitem 4, papurau i'w nodi. Byddwch chi'n gweld bod yna bedwar llythyr yn fanna i'w nodi. Unrhyw sylw? 

For my fellow Members, we move forward to item 4, papers to note. You will see there are four letters there to note. Any comments?

5. Cynnig o dan Reol Sefydlog 17.42 i benderfynu gwahardd y cyhoedd o weddill y cyfarfod
5. Motion under Standing Order 17.42 to resolve to exclude the public from the remainder of the meeting

Symud ymlaen i eitem 5. Nawr, yn naturiol, mae'r pwyllgor iechyd yr wythnos nesaf yn teithio i Gasnewydd—

Moving on, then, to item 5. Now, the health committee next week will be on tour. We'll be going to Newport—

Diolch, Jayne. Rydym yn teithio i Gasnewydd i lansio’n adroddiad ni i unigrwydd ac unigedd, ac hefyd yn mynd i Ysgol Basaleg ynglŷn â chadw'n heini, ac ymgysylltu â'r disgyblion ysgol fel rhan o'n hymchwiliad i weithgaredd corfforol plant a phobl ifanc. Felly, mae hynny er gwybodaeth i bawb.

Thank you, Jayne. We are going to Newport to launch our loneliness and isolation report, and we'll also be going to Bassaleg School in relation to physical activity, and some engagement activity with school pupils there in relation to our inquiry into the physical activity of children and young people. So, that's just for information for everyone.

Cynnig:

bod y pwyllgor yn penderfynu gwahardd y cyhoedd o weddill y cyfarfod yn unol â Rheol Sefydlog 17.42(vi).

Motion:

that the committee resolves to exclude the public from the remainder of the meeting in accordance with Standing Order 17.42(vi).

Cynigiwyd y cynnig.

Motion moved.

Rwy'n cynnig o dan Reol Sefydlog 17.42 i benderfynu gwahardd y cyhoedd o weddill y cyfarfod, a mynd i sesiwn breifat i ymdrin â'r dystiolaeth. Pawb yn cytuno? Pawb yn cytuno. Diolch yn fawr. 

I move the motion under Standing Order 17.42 to resolve to exclude the public from the remainder of the meeting and go into private session to discuss the evidence. Is everyone content? Yes. Thank you very much. 

Derbyniwyd y cynnig.

Daeth rhan gyhoeddus y cyfarfod i ben am 11:11.

Motion agreed.

The public part of the meeting ended at 11:11.

Explore the National Assembly for Wales