Wednesday, 30 August 2017

Alcohol policy in the real world

Neo-temperance dogma says that the best way of reducing alcohol-related harm is to tackle the Three A's: advertising, affordability and availability.

The current available scientific evidence supports prioritization of multiple cost-effective policy actions – the so-called three alcohol policy best buys:

  • Increasing alcohol beverage excise taxes,
  • Restricting access to retailed alcohol beverages and
  • Comprehensive advertising, promotion and sponsorship bans

These policies will supposedly reduce per capita consumption which will supposedly reduce harmful consumption and therefore reduce alcohol mortality. By the same token, relaxing licensing laws, permitting advertising and making alcohol more affordable will have the opposite effect.

The Three A's are the Holy Trinity of alcohol policy, endorsed by the WHO and every 'public health' organisation you can think of. For example, in Health First: An Evidence-Based Alcohol Strategy for the UK, it says categorically...

Long-term success in minimising the harm from alcohol will only be achieved by population measures that reduce the affordability and availability of alcohol products for all drinkers. The research evidence is unequivocal: such population measures are the most effective in reducing alcohol consumption and alcohol-related harm. [My italics]

As neat and tidy as this theory is, drinkers have conspicuously failed to comply with it. In Britain, for example, campaigners claim that alcohol has become 60 per cent more affordable since 1980 and yet per capita consumption of alcohol is no higher today than it was then. Alcohol has become more widely available thanks to the Licensing Act and yet per capita consumption has fallen by nearly a fifth since it was introduced in 2005.

An interesting little study was published last week with some similarly awkward facts, this time from Denmark. The researchers looked at various factors that could influence the Danish drinking culture and looked for evidence of an impact on consumption and harm.

The first thing they noticed was that there was decline in alcohol consumption between 2003 and 2013 which was accompanied by a decline in alcohol-related mortality. So far, so good (although, interestingly, they found evidence of a rise in problem drinking at the same time).

Was this due to alcohol becoming less affordable? Far from it. There was a large cut in the tax on spirits in October 2003 of 45 per cent. Since then, the authors say, 'prices have either remained stable or have decreased'. Since incomes have increased in Denmark since 2003, this means that...

... purchasing power has made alcohol more affordable, along with lower prices through decreased taxes, theoretically this should have enabled, if not encouraged, Danes to buy and consume more alcohol.

But it didn't.

What about advertising? Unlike other Scandinavian countries, Denmark allows alcohol advertising to be published in the print media with relatively few restrictions and it can be broadcast on television and radio at any time. As in the UK, the content of drinks advertisements is self-regulated.

The only substantive change to the advertising environment identified by the researchers in Denmark was a tightening up of the industry's voluntary code in 2010. They conclude that 'advertising has become more restricted over our study period, even if it has been accomplished by self-regulatory measures of the alcohol industry.' These voluntary restrictions emphasised a commitment to avoid advertising to children, but this was already part of the self-regulatory code before 2010 so it is doubtful whether it made much difference to the content of the advertisements. (Incidentally, 'public health' campaigners claim that self-regulation is ineffective and therefore cannot argue that these minor changes reduced alcohol mortality).

Thirdly and finally, there is availability. Did alcohol become harder to obtain? No. On the contrary, the researchers note that...
Two laws affecting the physical availability of alcohol in Denmark were lifted on 1 July 2005. Both concerned off-premise sales of alcohol and point towards liberalising previous restrictions (Danish Health Authority, 2014). One of these was the elimination of a law which had set the closing of sales at 8 PM in grocery stores and a partitioning off of the display area after alcohol sales hours. Additionally, the ‘‘restaurant law’’ was lifted. Previously sales of alcohol to take away from a restaurant had to take place in a room separate from the eating establishment.

Taken together, the situation in Denmark is very much like that of the UK. Licensing laws have been relaxed, alcohol has become more affordable, and advertising appears in all media under a system of self-regulation. According to the wisdom of 'public health', this should have led to a rise in alcohol consumption and a rise in alcohol-related mortality. It has done neither.

The study's authors acknowledge that this is hardly an isolated example and refer to this study from Italy where alcohol consumption fell from 16 litres per person in 1971 to 7 litres in 2005. Over the same period, deaths from liver cirrhosis fell by around two-thirds.

Which of the Three A's accomplished this miracle? The answer, again, is none of them. The authors note that 'there have been very few implemented policies, and these have often been weak and generic'. The most popular drink in Italy is wine but wine duty was 'zero throughout this period.' Although there were some increases in the tax rate of spirits, they were not introduced until the 1990s by which time alcohol consumption had already fallen to less than 9 litres per person. Advertising continued to be largely unrestricted and the only significant change to licensing laws was a ban on the sale of alcohol after 3am, but that did not happen until 2010.

In conclusion, the Italian researchers say:

The ineffectiveness of the preventive policies on the alcohol consumption trends is striking... Alcohol policies, which are usually considered to be the main contributor to changes in consumption and alcohol-related harm, are not able to explain the changes in alcohol consumption which have occurred in Italy during the last decades.

Neither the authors of the Danish study nor the authors of the Italian study are able to explain why alcohol consumption and associated harms declined in their respective countries. The Italian study is even titled 'The puzzle of Italian drinking'. Given that the outcomes seen in these countries are exactly what 'public health' campaigners claim to want, you would think there would be a concerted effort to find out what caused them. Reducing harm without infringing on the rights of drinkers should be the Holy Grail of alcohol policy and yet there does not seem to be much eagerness to learn from real world case studies.

If the 'public health' lobby took evidence seriously, they would want to understand why there is no correlation between supply-side alcohol policies and alcohol consumption. Instead, they ignore what real people do in real countries and focus on computer models which repeat back whatever assumptions they have programmed into them.

When all you have is a hammer, everything looks like a nail.

No comments: