A real alcohol crisis [warning: may not actually contain crisis]

By Eric Crampton 26/02/2013

I’d suggested last year that there’s no alcohol crisis in New Zealand.

A lot of the public health crowd disagreed with me rather vehemently, even suggesting that my position was based on that I’d done one bit of work funded by the alcohol industry.

Well, it turns out that there could be a crisis. Here are the latest figures from Statistics New Zealand:

The volume of alcoholic beverage available for consumption in New Zealand fell 3.3 percent in 2012, Statistics New Zealand said today. The decrease was due to a fall in the volume of beer, down 20 million litres. This fall was partly offset by a 4.3 million litre rise in the volume of wine.
“Although the volume of alcoholic beverages available was down more than 3.0 percent, the amount of pure alcohol fell only 0.6 percent,” industry and labour statistics manager Louise Holmes-Oliver said. “This was due to change in the types of beverages available.”
An increase in the volume of higher-alcohol beverages such as wine, spirits, and spirit-based drinks accounts for the smaller fall in pure alcohol available. The volume of high-alcohol beer (over 5.0 percent) also increased. In contrast, all other beer categories available for consumption have decreased.
The volume of pure alcohol available for consumption per person aged 15 years and over fell 1.7 percent, to 9.3 litres in 2012. This is equivalent to an average of 2.0 standard drinks daily per person (aged 15 years and over).
Alcohol statistics are a measure of how much alcohol is available for consumption, rather than actual consumption.

The Radio NZ report is slightly inaccurate: we didn’t have a 15 million litre drop in alcohol production; we had a 15 million litre drop in the volume of alcoholic beverage available for consumption. New Zealand produces a lot of wine for export and imports a lot of spirits: volume available for consumption accounts for that.

A 1.7% drop in pure alcohol available for consumption may be a crisis though, depending on how sensitive a trigger you have for such things. I’m very likely engaging in a bit of hyperbole in calling a 1.7% per capita drop a crisis. But if we have a bit of mean reversion next year, any guesses whether a 1.5% increase would be called a crisis?

Here’s the time path. Some crisis. [If it doesn’t work in Dismal’s WordPress setup, hit it at Offsetting. UPDATE: The interactive chart hates WordPress. The static image here may work.]

It’s also worth noting the data caveats listed at The Ladder.

The anti-alcohol brigade are shifting their efforts to local politics since the alcohol reform bill widened scope for local government control of alcohol use. When your local busybody starts lobbying for earlier closing times, fewer bars, fewer off-licence agents, restrictions on alcohol in supermarkets, or broader areas where open liquor is banned, and present data that pick 1998 as starting point to get a seemingly neutral “over the last 15 years”, please do point your Council to the Stats NZ series.

SciBlogs disclosure: see the link at the top about the one bit of funded work.

0 Responses to “A real alcohol crisis [warning: may not actually contain crisis]”

  • far be it from me to claim a wowser prize – my historical and current consumption of alcohol is such that I am doing my bit for the industry.

    However, for alcohol reform in NZ the issue is less one of the average consumption, more one of the daily peak consumption for individuals, and specifically individuals who are at elevated risk including youth.

    Living in South Auckland as I do, its outstandingly clear that cheap alcohol is consumed in large quantities by a small number of individuals in binge sessions, generally on payday and often as a result of purchases at local retailers. As an employer in this location, we can give stats that show a reliable result – we pay fortnightly on Wednesday and our peak absenteeism is for the three days following payday every fortnight, year in year out.

    The telling part of the quote you include is this: “Although the volume of alcoholic beverages available was down more than 3.0 percent, the amount of pure alcohol fell only 0.6 percent..” In other words, although we are drinking a lower overall volume of products, we are drinking higher alcohol content products- products that present a greater risk to health and personal and public safety.

    Its no surprise this is the case given alcohol industry approaches including the strong move toward alco-pop products that are typically contain 30 to 50% more alcohol than the same volume of beer, and are marketed directly at the youth demographic.

    Your unstated assumption that there is no alcohol crisis or that the crisis is actually one of dropping consumption seems simplistic in this light.

  • I really need to have a cut-and-paste bit that I put at the end of every post on alcohol pointing to all of my prior posts on alcohol. In short, there are many many other bits of evidence all pointing in the same direction. Whatever is going on has not gotten worse in the last 20 years. You can complain about levels if you think the levels are bad. But there isn’t really a worsening trend.

    No crisis in youth drinking:

    No change in potentially hazardous drinking prevalence rates from 1996/1997 – 2006/2007: http://offsettingbehaviour.blogspot.co.nz/2011/07/again-what-crisis.html

    My public lecture on the same topic:

  • 1 – the first link refers to your blog where you make points on the generalised state of youth drinking. It doesn’t seem to address the issue of youth problem ie “binge” drinkers. This is the point I am raising – we may have an overall reduction in alcoholic beverage consumption, but we have specific groups where significant harm is (may be…) being caused by an increase in the alcohol volume in what we drink.

    The risk of the harm (specifically in the youth market) is greatly increased by two things at least – ready access to and availability of alcohol in a retail market characterised by high competition levels, and product design and marketing that targets youth drinkers.

    Looking specifically (and obviously anecdotally) at the markets I am aware of, I note that liquor shops in South Auckland stock low cost, relatively high alcohol volume products and compete fiercely on price. They are never more than easy walking distance within neighbourhoods that have high youth populations. Their windows are entirely covered with marketing material for pre-mix drinks that are formulated for a, shall we say, undiscerning palette.


    2 – how does an overall 2.5% increase in potentially hazardous drinking rates between 1997 and 2007 (in the second linked blog” mean ‘no change’? Especially in the light of DECREASED overall alcohol consumption?

    I agree that additional costs through tax and the like are not a good tool to address the issue of problematic alcohol use in a specific demographic – the raising of costs for everyone would be difficult to justify economically or morally.

    By the same token, denying a problem (in this case by not acknowledging the potential devils in the detail of it) does little to improve the situation either, and categorising those who are looking to improve the situation as “anti alcohol” or “busybodies” is, well, purile…

    • Did you read the first link? I’ll blockquote the relevant bit:

      The same pair of reports have 31% of youths 14-18 drinking heavily in 1998 (reported 5+ drinks at last occasion) and 30% of those aged 12-24 binge drinking now (4% of those 12-14, 27% of those 15-17, and 44% of those 18-24). Again, youth binge drinking is a problem. But it’s harder to say that it’s a problem that’s worse now than it was prior to the change in the purchase age; the proportions seem pretty similar.

      If crisis means “the way things have been at least since 1998”, then fine. To my mind “crisis” means something currently much worse than it has been and not showing signs of improving. Maybe it’s semantics. And you’ve really gotta beat up the data in the table I linked in the second piece to pull any credible increased trend. Look at how the subgroups bounce around between the 96/97, 02/03 and 06/07 reports. I’ve not seen standard errors on their averages, but I’d be a bit surprised if we could reject equality of means.

      I also utterly reject the notion that those pushing population-level controls while taking zero account of the harms they thereby impose on moderate drinkers are working to “improve the situation”. They are harming some to possibly benefit others.

  • Crisis? What Crisis?

    Yeah, a word to avoid in any argument – it runs close to invoking Godwins Law. Nonetheless, if 31% of any age group were self-reporting the problematic use of a drug, you’d have to say it was worrying and probably would be economically and socially efficient to address.

    If, for example, it were another drug in Class B (to which alcohol would probably be attached were it to be classified as a drug) we would (I suspect) not be accepting that manufacturers develop and market a product to the youth sector, or that it would be retailed essentially without control within the community.

    As a case in point, I note the different handling of so-called “legal highs” where the manufacturer now has to categorically prove the safety of the product. How would alcohol fare under that requirement?

    So, crisis? Semantics – I’m not even aware where the term was coined first in this context. Serious problem? I think that if good longitudinal research shows an ongoing and relatively widespread misuse of a drug, particularly in the youth demographic it indicates that we could be doing stuff better.
    I’m not at all sure what that stuff could be, but as in my earlier post, it seems entirely reasonable on the basis of Economics 101 interpretations that supply side factors have a role to play.

    BTW, I am struck by the irony of your “utter” rejection that a group is working to improve a situation because they utterly reject balancing factors…

    Thanks for the “dismal” series too – its nice to have exposure to this level of thinking.

  • Well, either it’s bad and always has been so, or we’re a bit too quick to add people into the “bad stuff” pile. The threshold for problem drinking seems rather low.

    I encourage you to check the rather extensive existing list of regulations around the sale and promotion of alcohol as contrast to “retailed essentially without control”.

    I’d be of the group who’d want to relax controls on all drugs that are at least as safe as alcohol such that they were no more stringently regulated than alcohol is. But I can see the case for having stricter controls around new risks than around ones that we’ve spent the last several thousand years learning to deal with and embedding in our cultural rituals.

    I’m open to the possibility of policies that pass cost-benefit analysis, where the good they do in curbing harms outweighs the harm they do in curbing reasonable consumption. But I really can’t see how those advocating very strong restrictions targeted population-wide, with zero consideration of the costs they thereby impose, are really helping anything.

  • “Nonetheless, if 31% of any age group were self-reporting the problematic use of a drug, you’d have to say it was worrying and probably would be economically and socially efficient to address.”

    Perhaps we read this in different ways – but if I see people voluntarily deciding to do things they both ex-ante and ex-post say they regret doing I like to ask why. In economics, we view this sort of behaviour by individuals as “time inconsistent” behaviour.

    If we are going to try to figure out if there is some set of policies we can introduce to improve outcomes, we don’t then target “getting drinking down” in of itself – we say ok, if people are behaving in a time inconsistent way how can we help them deal with that.

    Fundamentally, people lie on surveys, but they don’t lie with their actions – by helping institute mechanisms that allow people to pre-commit to courses of action, we can ensure that outcomes are “better” without having to be as concerned about “unintended consequences”.

    In the context of alcohol, rather than explicitly tax and limiting the consumption of alcohol because “it is good for them” we can restrict sales, advertise about social norms, and offer other non-alcoholic social activities – these are ways of dealing with one of the perceived causes of time inconsistent drinking.

    And yes, economists tend to hold this attitude for all types of drugs.

    It is easy to sit there and act as if we are taking away people’s choices because they are hurting themselves, and therefore we are good. It is a lot harder to admit that people deserve responsibility and respect and that as a result “society” (through government) should be aiming to work with individuals – rather than legislating against their actions. Yes, if they are hurting someone else society has more of an obligation to get involved. But the seemingly noble goal of “cutting alcohol consumption” comes with the priviso of telling people what to do for their own good.

    My belief is that Eric and many of the Health people want the same thing – different views of what the data represents and how allocation works are the reasons that policy implications differ. Eric specialises in the study of allocation, which is why I tend to place more faith in his overall conclusions on that matter.

  • You discuss the action of regretful drinkers as voluntary as if the act of choosing to continue drinking when you are already moderately intoxicated is rationally made. Its not. By the very nature of the product, the more alcohol you consume in any given period of time, the less capable you become of rational thought processes.

    It is therefore logical that an external force of some sort is required to moderate the consumption behaviour.

    History has shown that the supply side has a very poor ability or desire in this respect. Producers and retailers of alcohol have a long and honourable tradition of observing legislation and their own codes of conduct in the breach. Again, this is natural – there is no real disincentive for them in their actions (and they show a time inconsistent behaviour too – observe the bar managers when they are found to have supplied alcohol to an intoxicated person, or after a fight erupts in a bar at 2.25am…).

    So, if the consumer is unable to act rationally when in extremis, and the supplier is motivated to maximise consumption, its left to the rest of us to put in place actions that moderate behaviour. This is further appropriate since we are where the costs fall.

    So, no problem with allocation. No problem with light regulation. Some difficulty with Eric’s apparent industry blind spot (my interpretation) and offhand treatment of the possibility the figures show steady or slightly increased PROBLEM drinking in the context of reduced overall consumption.

    All as a very lay person and happy wine consumer…

  • Yes, being somewhat intoxicated may make you more likely to choose to consume more than you would have otherwise. And if you knew nothing about alcohol and it was the first time you’d ever had any, this could lead to big problems.

    But people are able to make choices before starting drinking that can bind their later selves. If that were impossible, then drink driving would be a lot more common. Instead, sensible people plan ahead and make other arrangements while sober, knowing that their later self might try something silly. You don’t need the state to be the external force here: people can and do choose to bind themselves. Read Jon Elster’s excellent “Ulysses Unbound”.

  • thats a VERY simplistic understanding of our motivations for consumption of alcohol (and more particularly the decision to consume to harmful excess) Eric.

    For a start it assumes that, having made a decision to NOT drink to excess (for whatever reason) that decision is immutable. This is demonstrably wrong whether you are sensible or not.

    We know the resulting cost of alcohol to the police and justice system in NZ, as well as to the health system in primary, secondary and tertiary settings – this alone is evidence that large numbers of people are unable to reliably be bound by their initial good intentions.

    Estimates of the cost of alcohol misuse in NZ range widely, but seem to be somewhere around $5b dollars a year (http://www.ndp.govt.nz/moh.nsf/pagescm/7752/$File/alcohol-factsheets.pdf). Even if it were half this, that is around $60 per head of population and it disregards non-financial costs eg pain and suffering. Its not clear if it includes an accounting value for loss of life.

    Is that cost balanced by an equivalent value of individual benefits?

  • Ashton:
    1) Jon Elster covers the topic of utility-enhancing constraints really nicely in his book and documents many of the ways that we choose to constrain ourselves. Nothing is perfect, but we don’t need perfect.

    2) I’m somewhat familiar with the $5b figure you cite; you might consider some of the arguments here on the topic. http://www.econ.canterbury.ac.nz/RePEc/cbt/econwp/1129.pdf . Briefly, though, the $5b hardly disregards non-financial costs of pain and suffering; those are included. Further, $1.52b of the $4.8b consisted of intangible non-financial costs of premature mortality. Given how NZ’s VSL is derived, the $1.52b is double-counted with estimated productivity costs of premature mortality. About $967m of the $4.8b might plausibly count as cost in the sense in which economists typically use the term.

    3) I will, again, point you to my public lecture on this topic. It’s an area that clearly interests you. But I am wasting my time in reiterating points here that I made far better in that forum. http://offsettingbehaviour.blogspot.co.nz/2012/12/what-if.html . I also strongly urge you to flip through the posts tagged with the “alcohol” label over at Offsetting Behaviour.
    http://offsettingbehaviour.blogspot.co.nz/search/label/alcohol . Many of the points you are raising I have addressed repeatedly elsewhere.

  • Thanks Eric – as you can tell my knowledge of economic theory is strictly 101 (or lesser) in its a expertise…

    I find the discussion interesting because it does lie at that interface between public and private good. It strikes me that while its entirely appropriate to cost some things in financial terms, we seem not to have (or perhaps don’t apply) tools that would cost our behaviour (individual and societal) in more direct terms – the “happiness index” idea.

    Once again, thanks for engaging a neophyte like me.