Author Ziyang Lyu, Suzanne Poynton and Scott A. Sisson
Published January 2024
Report Type Crime and Justice Bulletin No. CJB263
Subject Alcohol; Assault; Evaluation reports
Keywords Alcohol; non-domestic assaults; counterfactual; proxy data; intervention model

Download this publication



Between 2008 and 2018, the NSW Government implemented a series of reforms to reduce violence related to licensed premises, including both state-wide and local area strategies. This study aims to quantify the effects of these licensing policies on rates of violent crime in New South Wales (NSW).
Given that the liquor licensing policies were applied to different geographical areas, we first quantify the effects of each individual statewide alcohol policy for the rest of NSW (RNSW; defined as all of NSW excluding the Sydney Local Government Area), and then the effects of each localised set of additional policies in the Sydney Central Business District Entertainment Precinct (CBD) and Kings Cross Entertainment Precinct (KCP).
We use related time series data from Victoria (and areas near the CBD and KCP) to guide understanding on what assaults may have been reported in NSW if the liquor licensing policies had not been introduced. We then identify and quantify the effect of those state-wide and area-specific policies that had a statistically significant impact on reported assaults, both instantaneous and gradually over time, both individually and as a cumulation of all previous policies.

Key findings

Aggregated weekly counts of reported non-domestic assaults in RNSW (black line) and Victoria (grey line) between 1st January 2000 and 31st December 2019
 Aggregated weekly counts of reported non-domestic assaults in RNSW (black line) and Victoria (grey line) between 1st January 2000 and 31st December 2019
The suite of public safety measures introduced in the last decade by the NSW Government have contributed to a significant decline in non-domestic assaults both in Sydney and across NSW. By the end of 2019, the number of non-domestic violence-related assaults in NSW had reduced by an estimated 19% (56,695 cases) from levels that would have been achieved had no alcohol control policies been introduced.
Our results also suggest a significant decline in violent crime in the Sydney CBD, with a 45% (7,677 cases) reduction due to the controls implemented since 2008. Much larger effects were observed in the KCP, though primarily due to lower overall assault numbers, with an estimated 84% (3,117 cases) decline in the number of non-domestic assaults above what would be expected if no policy changes had occurred.
Policies restricting late night (or 24-hour) trading of licensed premises and targeting enforcement toward the highest risk venues contributed most to these declines.


Restricting trading hours can substantially reduce the risks associated with acute alcohol intoxication and can be an effective crime reduction strategy when combined with enforcement that targets the small number of premises that account for most of the harm.