CARBC's Tim Stockwell responds to article in regarding minimum unit alcohol pricing

The Irish government is planning on introducing minimum unit pricing (MUP) for alcohol and has cited CARBC research in support of this legislation. recently ran a "FactCheck" article contesting the goverment's claim that the policy has been “proven to work” in other jurisdictions and critiquing CARBC's research on the topic. CARBC's Dr. Tim Stockwell responded to the article and the "FactCheck" author made some corrections to the article.  However, Dr. Stockwell thinks that there are still "some serious misunderstandings evident in the article and that our work has been misrepresented."  Find his full response below: 

I am grateful to Mr. MacGuill for making some corrections to his FactCheck article in response to concerns that I raised with him about the accuracy of some of his descriptions and criticisms of research published by my team at the University of Victoria in British Columbia (BC), Canada. However, I still have concerns with how our research has been represented, which I outline below. 

I totally support the mission of FactCheck in carefully examining factual claims made by public figures to keep them honest and inform the public. In this case correspondent MacGuill has taken Irish Health Minister Marcella Corcoran Kennedy to task for claiming that the policy of Minimum Unit Pricing for alcohol has been “proven to work” in other jurisdictions to reduce alcohol-related hospitalizations and deaths, citing the example of BC. He concludes that her claim is “mostly false” and cites our research as the basis of the Minister’s claims. In the same interest of public accountability for factual correctness I would like to defend our work from some of MacGuill’s criticisms. However, I do agree with Mr MacGuill that there is a need for our findings to be replicated and for improved studies to be conducted. Our work is certainly not perfect but, on this occasion, I think there are some serious misunderstandings evident in the article and that our work has been misrepresented.

  1. MacGuill describes only two of our five published studies1 on this topic that he suggests are the only ones relevant to the Minister’s claims about health. Two of our other papers demonstrate reduced consumption of alcohol when minimum prices are increased or when a comprehensive new set of minimum prices based on alcohol content is first introduced. Both show clear relationships between the price increases and reductions in consumption, as well as a shift towards consumption of lower alcohol content drinks. There is a strong relationship in the international literature between how much a population drinks overall and their rates of alcohol-related deaths, injuries and illnesses. Thus these other studies are also relevant to the health claims made by the Minister. Furthermore, the fifth paper demonstrates effects on alcohol-related traffic and violent crimes, also relevant because of their many associated injuries. Please note that these papers were accepted for publication after independent peer review by public health scientists unknown to us and accepted by editors of high-profile academic journals, namely the American Journal of Public Health, Addiction and the Journal of Studies on Alcohol and Drugs. The results of these other papers support the Minister’s statement, even if the definitive word ‘proof’ is rather strong for this kind of study.

  2. MacGuill suggests that because alcohol-related deaths and hospital admissions rose and fell across the whole of BC over the period examined in two of our studies, this contradicts the Minister’s statement. But our research looked at how rates of alcohol-related deaths and hospitalizations responded in the short term (the next 3 month periods) to changes in the real value of minimum prices (adjusted for inflation) and for each of 89 distinct geographic areas (not the whole province). Imagine 89 natural experiments occurring in each of the geographic areas studied with 32 observation periods each and detailed local rates of alcohol-related harms. There were periodic increases in minimum prices followed by gradual declines in their value due to inflation. We predicted there would be immediate impacts on "acute" outcomes, i.e. alcohol-related injuries and poisonings, and delayed outcomes for "chronic" diseases (e.g. liver disease, cancers), which may take years to develop. And that is exactly what we found. Thus, we showed that not only did some health outcomes reduce when prices increased, as predicted, but also the reverse: i.e., when minimum price values declined, these health outcomes worsened. We took that as strong evidence that raising minimum prices will reduce alcohol-related harms. “Proof” may have been a strong word for the Minister to use, but I consider there to be encouraging evidence to support her claim and find MacGuill’s ‘mostly false’ verdict a little harsh. To capture the full nuance behind her assertion, the Minister could have said something like “when minimum prices were increased in BC there were immediate reductions in some serious alcohol-related illnesses and deaths in most locations, but these benefits were gradually wiped out as the value of the minimum prices declined due to inflation,” but this would notmake for a good sound bite. However, I believe that the Minister accurately conveyed the essential point that there is evidence for minimum pricing being effective by the crisper “proven to work.”

  3. We also explored whether there were delayed effects of the pricing changes on the health outcomes. We found an immediate effect for acute hospitalizations, for example, which decreased 9% in relation to a 10% increase in the value of the minimum prices. There was no significant immediate association, however, with chronic alcohol-related hospitalizations. Instead, we hypothesized that delayed effects might be found for these chronic cases, which we did, in fact, observe two years after the price increases. MacGuill, however, takes us to task for the fact that only 9 of the 64 exploratory tests of these possible delayed effects were significant for hospitalizations - and only 14 of the 64 tests done for deaths were significant. But we did not predict they would all be significant - we certainly did not expect any significant delayed changes for acute outcomes and very few were found. We did not know if or when we would find delayed effects for chronic outcomes. That not all the relationships were significant in no way counts against our main hypotheses, so I consider this to be another unfair criticism of our study.

  4. MacGuill correctly points out that BC does not in fact have a policy precisely analogous to minimum unit pricing, i.e. one that sets a fixed floor price for a particular quantity of pure alcohol. However, we found much stronger impacts on alcohol consumption in Saskatchewan, which introduced overnight in April 2010 a new comprehensive range of minimum prices adjusted for alcoholic strength - a policy closely analogous to MUP. The BC minimum prices are quite imperfect as MacGuill correctly describes - they are applied per litre of beverage regardless of its alcohol content. Nonetheless, as minimum prices per litre of beverage increase so too will the minimum price per unit of all beverages – just not to exactly the same degree as would occur with MUP.

  5. MacGuill also suggests that the government monopoly on the distribution of alcohol in both the provinces studied means that these alcohol markets are not at all analogous to the free market that exists in Ireland. However, I would point out that in both BC and Saskatchewan, consumers purchase alcohol from bars, restaurants, privately owned liquor stores and government owned stores. Further, the experience of purchasing alcohol in a government owned store is pretty much identical to that in a private store – and in neither case would most people be interested in who was the original supplier of the alcohol. In each case, the relevant laws of economics would apply, namely that amount and frequency of purchases would be sensitive to the prices posted.

It is hard to do these kinds of real world studies but this is the best available evidence and the results are certainly encouraging for the view that new or increased minimum prices will reduce serious alcohol-related health problems. In short, I suggest our research does back up the Minister’s statements – though, to be sure, further studies in other places would be welcome. I think MacGuill chose to take a rather narrow and literal interpretation of the Minister’s remarks to contradict her main message. The Republic of Ireland is a country with more than its fair share of alcohol-related injury, illness and death, so I think that it is important to give every consideration to policies that have the potential to reduce this toll.

Tim Stockwell

Professor of Psychology and Director, Centre for Addictions Research of BC, University of Victoria, BC, Canada

29 August 2016


Stockwell T, Zhao J, Macdonald S, Martin G. The relationship of minimum alcohol pricing on crime during the partial privatization of a Canadian government alcohol monopoly. Journal of Studies on Alcohol and Drugs. 2015;64(4):628-34. 

Stockwell T, Zhao J, Martin G, Macdonald S, Vallance K, Treno A, et al. Minimum alcohol prices and outlet densities in British Columbia, Canada: Estimated impacts on alcohol attributable hospitalisations. American Journal of Public Health. 2013:e1-e7. 

Stockwell T, Auld MC, Zhao JH, Martin G. Does minimum pricing reduce alcohol consumption? The experience of a Canadian province. Addiction. 2012;107(5):912-20. 

Stockwell T, Zhao J, Giesbrecht N, Macdonald S, Thomas G, Wettlaufer A. The raising of minimum alcohol prices in Saskatchewan, Canada: impacts on consumption and implications for public health. American Journal of Public Health. 2012;102(12):e103-10. 

Zhao J, Stockwell T, Martin G, Macdonald S, Vallance K, Treno A, et al. The relationship between minimum alcohol prices, outlet densities and alcohol-attributable deaths in British Columbia, 2002–09. Addiction. 2013;108(6):1059-69.