The effect of liquor licence concentrations in local areas on rates of assault in New South Wales
Neil Donnelly, Patricia Menéndez and Nicole Mahoney
Crime and Justice Bulletin no. 181
Sydney, NSW Bureau of Crime Statistics and Research, December 2014
Release Date: 19 February 2015
Media release, full report [pdf, 567kb]
This paper analyses police, liquor licensing and socio-demographic data to investigate whether the number of locations where alcohol is sold is associated with rates of assault and violence in various NSW Local Government Areas. It discusses the effect of concentrations of liquor licences on domestic and non-domestic assault rates.
Aim: To investigate the relationship between liquor licence concentrations and assault rates in Local Government Areas (LGAs) in New South Wales.
Method: Police, liquor licensing and socio-demographic data were analysed. Spatial regression analyses were conducted to measure associations between liquor licence concentrations and domestic violence (DV) and non-domestic violence (non-DV) assault rates.
Results: The concentration of hotel licences in an LGA, particularly at higher density levels, was strongly predictive of both DV and non-DV assault rates. A similar, but slightly weaker, association was found for the concentration of packaged licences and DV and non-DV assault rates. On-premises concentrations also predicted DV and non-DV assault rates at the LGA level but, unlike hotel concentration effects, in this case there was no evidence of stronger effects at higher density levels. A significant relationship between DV assault rates and the concentration of clubs was also found, but the association between the concentration of clubs and the non-DV assault rate was not as strong.
Conclusion: Regulatory authorities should be concerned about increases in liquor outlet density. In particular, increases in the density of hotels above 2 per 1,000 residents are of greater concern than increases in the density of premises with other types of liquor licence.
Keywords: assault, violent crime, alcohol, liquor licence concentrations, linear regression, spatial autocorrelation, simultaneous autoregressive (SAR) models