A response to Dr Humphreys.

Abstract

We thank Dr Humphreys for his thoughtful critique 1 of our paper on liquor licensing restriction and alcohol-related violence 2 and offer this response, structured under the same headings he employs. Dr Humphreys is concerned that we restricted our analysis to actual and grievous bodily harm (ABH and GBH), which might not always be alcohol-related or restricted to night-time events. We focused on assaults occasioning ABH and GBH because assaults resulting in injury are much more likely to be reported to police than assaults that do not result in bodily harm or injury. Although not all ABH and GBH assaults are alcohol-related, we presume that the reduction in total recorded ABH and GBH following the 2008 licensing restrictions is largely alcohol-related. Our discussion section explores other factors that could have caused or contributed to the fall in assaults and why those factors are not likely to have had a large effect. Dr Humphreys is concerned that the level of detail we offer on the three inter-related licensing restrictions we analyse is too low. We agree that it is unfortunate that there were no data on proxy measures of the implementation of the reforms, such as alcohol sales by premise type or data on the prevalence of serving alcohol to intoxicated patrons. Dr Humphreys mentions the need for additional control variables to account for the impact of economic factors on our results. We agree, and it was for this reason that we included a control variable (consumer sentiment) in our original analysis for the impact of the global financial crisis 2. We then re-analysed our results with a measure of unemployment following the suggestions in Shepherd & Page 3. Neither measure had any power to explain the fall in the number of assaults in our study period. Both, therefore, were excluded from the models. Dr Humphreys concludes by observing that: ‘While there is an understandable desire to present the results that demonstrate the effectiveness of restrictive licensing policies, we should not prioritize the message over the science’. We did not set out to present or procure results that demonstrated the effectiveness of liquor licensing restrictions. We asked the question of the data and reported the findings together with the caveats surrounding our findings 2. We looked for other possible effects and tested for them. We concluded that, even though we found an association between the alcohol licensing reforms and the drop in assaults, we could not exclude the possibility that there might be some unforeseen confounding factors that might have also affected the trend in the number of assaults. None.