Alcohol Accords can work: Long-term evaluation of the Norseman Voluntary Liquor Agreement

April 2016

By Professor Richard Midford1,2,3, Dr John McKenzieand Rachel Mayhead2

1. The Menzies School of Health Research
2. Charles Darwin University
3. National Drug Research Institute

This research was funded by the Foundation for Alcohol Research and Education (FARE) and the National Drug Research Institute (NDRI), Curtin University.


The likely prevalence of risky alcohol use in the Indigenous population is twice that of the non-Indigenous population1. This risky consumption is reflected in poorer morbidity and mortality outcomes. In addition to health problems, alcohol is linked to a broad range of Indigenous social problems such as homelessness, unemployment and imprisonment2. People, other than the drinkers themselves are affected, with high levels of alcohol related domestic violence, and intergenerational effects through the teratogenic effect of alcohol on the developing human foetus3,4.

In the early 2000s members of the Australian Indigenous community in Norseman, in Western Australia’s Goldfields region, became increasingly concerned that heavy alcohol consumption was the main cause of chronic health problems in their community5. Reviews of alcohol prevention measures have consistently identified supply restrictions as highly effective6-8, and many remote Indigenous communities in Australia have declared themselves dry so as to restrict access to alcohol. Research with some of these communities indicates that both consumption and related harms declined subsequent to the introduction of this measure7. The Indigenous community in Norseman is distributed throughout the township, so the option of declaring themselves dry was not available. However, recognition that certain beverages were particularly associated with heavy drinking led the Indigenous community to propose restricting the sale of these products5. The Norseman Hotel is the only outlet in town with a licence to sell packaged liquor to the general public, and at a meeting on 13 November 2007 with local community and service agency representatives, the local Licensee agreed to voluntarily restrict the hours of sale and quantities sold to any one individual of products nominated by the Indigenous community.

The Voluntary Alcohol Agreement

The Norseman Voluntary Alcohol Agreement (‘the Agreement’) was implemented on 1 March 2008, with the following sales restrictions between midday and 6pm Monday to Sunday:

  • red and white Lambrusco wine was limited to one five litre case per person per day
  • other non-fortified wine was limited to one four litre case per person per day
  • port wine was limited to one two litre cask per person per day.

At all other times the sale of these products was not permitted. The impact of these restrictions was evaluated in the following year. At that time, there was a decrease in consumption, particularly among Indigenous people and a decrease in a range of offending behaviour and alcohol-related morbidity. Reports from the community also indicated that there had been an increase in healthcare seeking behaviour, more participation in community activities and a decrease in public drunkenness and violence5.

The restrictions were extended from 1 August 2009. In addition to the existing restrictions, people were limited to purchasing:

  • one 750ml bottle of fortified wine
  • one 750ml bottle of full strength beer

These restrictions have remained in place to the present day.

Evaluation over the long term

This research investigated whether the Agreement has been able to maintain its initial benefits. A mixed methods approach was employed, with secular, quantitative alcohol consumption and harm data and qualitative interview data collected from a number of different sources. There is a gap in the alcohol data from May 2009 to October 2012 because of a change in wholesale supplier. This resulted in three distinct periods when data were reported:

  • before the initial restrictions (December 2006 to February 2008)
  • after the initial restrictions (March 2008 to May 2009)
  • and follow-up (October 2012 to December 2014).

Accordingly, consumption, using wholesale sales data, and converted to litres of pure alcohol, was compared between these three periods. Beer consumption has not been included because the wholesale sales data were unreliable.

The average quarterly consumption of cask wine declined significantly from before (305 litres of alcohol) to after the initial restrictions (158 litres of alcohol) (p= 02), and this difference remained at follow-up (161 litres of alcohol) (p= .02). The consumption of fortified wine did not change significantly from before to after the initial restrictions. However, there was significant decline from both before (85 litres of alcohol) and after the initial restrictions (45 litres of alcohol) to follow-up (11 litres of alcohol) (p< 01; p<.01), and here it is important to note that restrictions on fortified wine were increased 15 months subsequent to the initial restrictions. Spirit consumption did not change from before to after the initial restrictions, but increased significantly from both before (245 litres of alcohol) and after the initial restrictions (267 litres of alcohol) to follow-up (387 litres of alcohol) (p<.01; p<.01). Total consumption of all beverages did not change significantly at any point.

Alcohol harm data were collected from January 2004 to December 2014. All comparisons were made between the period before the initial restrictions and the period after the additional restrictions. There was a downward trend in the presentation rate of Indigenous people to the Norseman Hospital emergency department prior to the initial restrictions, which stabilised at a lower level subsequent to the additional restrictions. This was possibly due to the high profile community consultation process on restrictions influencing consumption patterns in anticipation of implementation. The difference was not significant. There was no discernible trend in the non-Indigenous presentations. There was a significant decrease in rates of burglary, domestic violence and assaults by Indigenous people between these two periods. Non-Indigenous burglary and assault rates also decreased significantly. Rates of driving under the influence did not change for either group. Police tasking rates (call outs), decreased significantly post restrictions. These changes suggest the restrictions have led to improved social behaviour, but it is less clear as to their impact on health.

The qualitative data from interviews with the community key informants and focus group participants indicated that the Indigenous community was the driving force for introducing the restrictions, in response to the domestic violence, chronic disease and death that was associated with heavy drinking. The reason given for not allowing alcohol sales, other than between midday and 6pm, was to limit the period of drinking so there was a break for heavy drinkers to sober up. There was almost universal agreement that the behaviour of drinkers, the amount of alcohol consumed and alcohol-related harms had all changed for the better since the introduction of restrictions. The common perspective was that street drinking had decreased and more drinking was occurring in homes during the evening. However, drinking parties at home in the evenings were affecting children’s schooling. The health workers, in particular, considered that the health consequences of drinking had reduced. Community climate was better because of less public drunkenness. Family function had improved and domestic violence had decreased, with a number of comments that drunkenness at home had improved since the restrictions. There was general agreement that the restrictions should remain in place. Most of the suggestions for dealing with alcohol problems in the town in the future went beyond a focus on drinkers and drinking. The need for jobs, employment skills and education was repeatedly mentioned by the key informants and focus group participants.

Discussion – benefits and limitations

The secular, quantitative data indicate that the restrictions were successful in terms of their original intent, in that consumption of the targeted beverages, cask wine and fortified wine, decreased significantly post the introduction of voluntary restrictions. Complementing this was the strong perceptions of improvement from within the community. These perceptions should be privileged because they represent the community voice, and add considerable detail as to what has changed and what problems remain. The other important consideration is that the restrictions have remained in place since their introduction at the behest of the community in March 2008. This is a long period for a voluntary ‘Alcohol Accord’ to continue functioning without external support9.

The initial evaluation of the Agreement stated it demonstrated that a community could achieve change and reduce harm from alcohol misuse through its own action. It was also unique in that government agencies worked with the Norseman community and the Hotel Licensee to enable these changes to occur without regulation or enforcement5. Such a voluntary agreement would, however, be difficult to replicate in larger communities with several licensed premises, and the lack of reliable wholesale beer data means there is a gap in the overall profile of alcohol consumption in Norseman.

While it is important to acknowledge these limitations, the benefits for the Norseman community are clear. The restrictions are still in place; have increased social order; are still overwhelmingly supported by the community, including the Licensee; and have remained effective in keeping in check those beverages identified from initial community discussions as problematic. This does not obviate the need for a reappraisal of the restrictions in light of changing patterns of alcohol consumption and harm, particularly the increase in consumption of spirits and the recent trend of increasing police call outs. These may be early signs that the effects of the restrictions are diminishing. However, any changes have to come with community support, as this underpins the current arrangements.

The full report of the long-term evaluation of the Norseman Voluntary Liquor Agreement is available from the FARE website:
  1. Wilson, M., Stearne, A., Gray, D. and Saggers, S. (2010) The harmful use of alcohol amongst Indigenous Australians. Australian Indigenous HealthInfoNet.
  2. National Indigenous Drug and Alcohol Committee. (2009) Bridges and barriers: addressing Indigenous incarceration and health. Australian National Council on Drugs, Canberra.
  3. Ornoy, A. and Ergaz, Z. (2010) Alcohol abuse in pregnant women: effects on the fetus and newborn, mode of action and maternal treatment. International Journal of Environmental Research and Public Health, 7, 364-379.
  4. Vos, T., Barker, B., Stanley, L. et al. (2007) The burden of disease and injury in Aboriginal and Torres Strait Islander peoples. School of Population Health, University of Queensland, Brisbane.
  5. Schineanu, A., Velander, F. and Saggers, S. (2010)“Don’t wake up angry no more” The Evaluation of the Norseman Voluntary Liquor Agreement. National Drug Research Institute, Curtin University of Technology, Perth.
  6. Babor, T., Caetano, R., Casswell, S. et al. (2010) Alcohol: No Ordinary Commodity: Research and Public Policy. Oxford University Press, New York.
  7. National Drug Research Institute. (2007) Restrictions on the Sale and Supply of Alcohol: Evidence and Outcomes. National Drug Research Institute, Curtin University of Technology, Perth.
  8. Stockwell, T. (2006) Alcohol supply, demand, and harm reduction: What is the strongest cocktail? International Journal of Drug Policy, 17, 269-277.
  9. Manton, E. (2014) Liquor accords: do they work? In: Manton, E., Room, R., Giorgi, C., and Thorn, M. eds. Stemming the tide of alcohol: liquor licensing and the public interest. Foundation for Alcohol Research and Education in collaboration with The University of Melbourne, Canberra.