The aim of the study was to investigate the relationship between rates of reported assault, alcohol sales and numbers of outlets (differentiated by outlet type) in both Queensland and Western Australia. Counts of assault offences formed the dependent variable in all analyses. In Queensland, the key explanatory variables of interest were counts of outlets by major outlet types and level of total pure alcohol sales. For Western Australia, key explanatory variables included on and offsite outlet counts and alcohol sales. All models included a full accompaniment of potential demographic and socioeconomic confounders. Multivariate negative binomial regression models were created at local government area level based on location, type and time of assault, and victim age and gender. Findings show that offsite alcohol sales and total volume of alcohol sales within a region are important predictors of assault. On this basis, it is reasonable to conclude that policy decisions that ultimately increase total alcohol sales within a community are more likely to exacerbate, rather than ameliorate harms associated with alcohol. This warrants recommendation of a precautionary approach to future liquor licensing policy formulation and application in Australian jurisdictions.