Do flexible opening hours reduce violent crime? : an evaluation of the Licensing Act

• Main Author: Humphreys, D. K.
• Published: University of Cambridge 2011
• Subjects: 364
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.604778

In 2003 the Government enacted a controversial piece of legislation aimed at tackling alcohol-related crime and disorder in England and Wales. The Licensing Act (2003) (referred to hereafter as the ‘Act’) was an effort to simplify a previously complex system for regulating licensed alcohol sales. This study is based on the examination of recorded crime and licensing data for the City of Manchester between February 2004 and January 2008, roughly two years before and after the Act’s implementation (24^th November 2005). Analyses designed to examine the impact of the Act on crime were undertaken in three stages: i) examination of crime trends across the study period at the city-level; ii) measurement and analysis of changes to licensed trading patterns across the study period; iii) evaluation of the relationship between changes to licensed trading patterns and violent crime in small geographic units of analysis. This research found no evidence to support the hypothesis that relaxed trading restrictions have contributed to significant reductions in violent crime. On the contrary, examination of city-level trends in violent crime revealed a small but significant average <i>increase</i> in violent crime (7.6%) following the implementation of the Act. Analysis of temporal trends in night-time violence revealed a significant shift in the timing of offences, suggesting that displacement of crime to later periods of the early morning (3am to 6am) had occurred. However, analyses of patterns of licensed trading showed that changes to the physical availability of alcohol, the temporal distribution of closing times and density of premises over the study period varied considerably across the city as a result of the Act. It was necessary to consider whether changes to trading routines had caused any localised effects on violence (a local dosage effect) that might have been masked by analysis at the city level. By disaggregating to smaller geographic units, it was possible to examine whether a dose-response relationship existed between changes to licensing and violent crime. Using multivariate analyses, this study was able to examine the specific relationship between local changes to licensing and local violence. These analyses concluded that no significant relationship exists between change in opening hours and change in violent crime when controlling for other variables. The thesis emphasises the importance of designing evaluation studies to fit and examine the specific components of the intervention that are proposed to instil change. The collection of detailed process data on pre- and post-intervention trading times enabled a comprehensive model of licensing change to be created.