Impacts of Public Drinking Laws

Research Bulletin no: 
1
Published: 
May 2013
Amy Pennay, Elizabeth Manton, Michael Savic, Michael Livingston, Sharon Matthews, Belinda Lloyd

Policies that restrict the spaces in which alcohol can be consumed are now widely implemented around the world. Bans on the public consumption of alcohol are particularly common in Western countries, including North America, the United Kingdom, Nordic countries, Australia and New Zealand. In Australia, partial or complete bans on drinking in public operate to some degree in all major cities, as well as in many regional and rural towns (Webb 2004).

Introduction

Policies that restrict the spaces in which alcohol can be consumed are now widely implemented around the world. Bans on the public consumption of alcohol are particularly common in Western countries, including North America, the United Kingdom, Nordic countries, Australia and New Zealand. In Australia, partial or complete bans on drinking in public operate to some degree in all major cities, as well as in many regional and rural towns (Webb 2004).

Public drinking bans have different names, including dry areas, alcohol-restricted areas, liquor bans, open container laws, alcohol-free zones and alcohol-exclusion zones (Chikritzhs et al. 2007; d’Abbs et al. 2008; Mast et al. 1999; Webb et al. 2004). There are jurisdictional differences for public drinking laws in Australia, whereby they are a matter of state/territory legislation in some jurisdictions and local council laws in others. In Victoria, public drinking laws are designed, enacted and controlled by local government, but are enforced by state government (police). This means that the provisions of these laws, including which spaces are included and exempt and during what hours the law operates, often differ between local government areas (LGAs), even if they are directly adjacent to one another. It also means that such laws require a considerable degree of cooperation and coordination between local council officers and police (Pennay & Room 2012).

It is only really in the past 10 to 15 years that public drinking bans have proliferated across urban centres in Australia. What is interesting about this timing is that this is also the timeframe in which drinking on the street has become increasingly legitimised in the form of licensed restaurant/bar/hotel footpath trading. Despite the many vested interests involved in public drinking bans, including local council employees, elected councillors, police, licensees, traders, drinkers and community members, and despite the recent proliferation of these drinking bans in urban areas, there have been very few evaluations of their impact or effectiveness throughout the world.

A review of community-based evaluations of public drinking bans in urban areas, which aimed to explore the effectiveness of these policies and their impacts on the community, has recently been published (Pennay & Room 2012). Sixteen evaluations were identified across 13 locations (3 districts had commissioned 2 evaluations at different time points). These included:

  • two evaluations from the United Kingdom (Lancaster and Winchester);
  • four evaluations from New Zealand (Christchurch, Wellington, Havelock North and Auckland); and
  • ten from Australia (Melbourne—City of Yarra, City of Darebin and the suburb of Footscray; Adelaide—City of Adelaide the suburb of Glenelg; New South Wales—a review across the entire state; and Port Augusta).

All evaluations used one or more of the following data collection methods—surveys, in-depth interviews, focus groups, observations and analysis of police data, and hospital statistics. Content and thematic analyses of the evaluations were performed (Joffe & Yardley 2003) and nine key themes were drawn out of the analysis.

These evaluations were not oriented to any single dimension of effectiveness, but they did provide a range of findings on the various impacts of public drinking laws. The most common themes were that public drinking bans often result in negative impacts to marginalised groups, often result in displacement and often improve perceptions of safety among the community. Themes that were noted but less pervasively were concerns about police enforcement and consistency, improvement in the aesthetics of an area (by removing drinkers and/or litter and glass), and variation between stakeholder groups in support of street drinking bans, ranging from strong support from police, traders and older people, through equivocal support from general community members, to disapproval from young people and Indigenous people. Finally, there was little or no evidence that street drinking bans reduced congregations of drinkers, reduced alcohol-related crime or harm or were understood and adhered to by the community.

All LGAs in Melbourne have implemented a ban on public drinking within the last 15 years. However, only three of the 31 LGAs in greater Melbourne have commissioned an evaluation of the public drinking bans that were enacted in their municipality—the City of Maribyrnong, the City of Darebin and the City of Yarra. A third evaluation in these three municipalities was conducted, given the continuing concern about public drinking in these areas. This evaluation allowed the evidence in each LGA to be strengthened by measuring change over time (by comparing the results of this evaluation with the previously commissioned evaluations, some of which are now 7 years old), it also allowed comparison and contrast of the differences and similarities across the three diverse inner-urban districts of Melbourne and allowed generalisation of the findings beyond just one unique geographical area.

The objectives of this project were threefold:

  • to evaluate the implementation of the prohibition of public drinking in the City of Maribyrnong, City of Darebin and City of Yarra (ie the strategies used in the implementation, the effectiveness of these strategies and the barriers and enablers to successful implementation);
  • to evaluate the effectiveness of the prohibition of public drinking in the three LGAs in reducing alcohol-related crime and harm, and improving public amenity; and
  • to evaluate the impact of the prohibition of public drinking in the three LGAs on a range of target groups, including:
    • police
    • residents
    • traders
    • local health and welfare workers
    • local stakeholders (including local and state government representatives)
    • potentially marginalised groups including particular ethnic groups, Aboriginal and Torres Strait populations, homeless people and young people
    • people who visit the area.

Methods

This mixed-methods evaluation involved seven key data collection components.

Media analysis

A search of Australian newspaper articles was undertaken in the ‘NewsBank’ database. Search terms included the name of each of the LGAs, council and variants of alcohol ban. A total of 44 articles were considered relevant and were included in the media analysis.

Observations

Twenty hours of observation were undertaken around selected ‘hotspots’ of each LGA during the day and night, on different days of the week and in summer, autumn and winter. Sessions of observation lasted between one hour and three hours in duration. Observations included the extent to which people were drinking in public, the extent to which there were other visible signs of drinking, the cleanliness of the area, any public disorder problems and the presence and actions of police. Detailed notes were taken during and immediately following sessions of observation.

Interviews with drinkers

In-depth interviews were conducted with 23 drinkers across the three LGAs. Interviewees were both male and female, and aged between 20 and 60 years. These interviews took place at a convenient location. Interviews lasted from 10–45 minutes and were tape recorded and transcribed. The interview schedule was semi-structured, which enabled a certain level of control over the questions while also allowing responses to dictate the flow of conversation and issues arising. Participants received $30 supermarket vouchers for their time and any out-of-pocket expenses.

Interviews with key informants

In-depth interviews were conducted with 15 local stakeholders (including local council officers, local health and welfare workers, and members of other local relevant agencies), 15 police (a mix of senior and junior police) and 15 traders (including licensed venue owners and other traders) across the three LGAs. These interviews took place at a convenient and private location. The semi-structured interviews lasted from 30–90 minutes and were tape recorded and transcribed.

Household survey

A brief household survey was mailed to 15,000 residents in the three LGAs via random selection from the telephone directory. Surveys were sent to all suburbs of each LGA, but suburbs where public drinking was known to occur and areas where the ban applied were purposefully oversampled. Approximately 1,244 surveys were returned as incorrect addresses, leaving a total possible sample of 13,756. Surveys were completed by 1,681 residents (41.5% from Yarra, 29.8% from Darebin and 28.7% from Maribyrnong). This was a lower than expected response rate of 12.3 percent.

The survey included questions about the length of time the residents had lived in the municipality and their general activities within the LGA, their awareness of the law, their views of the law, their experiences relating to public drinking prior to and since the law, and perceived effects of the law.

The demographic characteristics of household survey participants are listed in Table 1. It is important to note that the sample is not representative of the population, given that this method does not selectively target people of different ages, gender and other characteristics, and the low response rate has also resulted in the under-representation of some population groups. In particular, the sample over-represents females (61.3%), older people (mean age 49.9 years) and people with post-secondary school qualifications (77.7%). The sample is also generously represented by retirees (19.4%)

Table 1 Demographic characteristics of sampled residents

Mean age

49.9 yrs (range 18–95)

  n (%)

Gender

Male 637 (38.7)
Female 1010 (61.3)
Aboriginal and/or Torres Strait Islander 13 (0.8)

Highest level of education

University qualification 1061 (64.2)
Certificate/diploma 137 (8.2)
Trade certificate 88 (5.3)
Year 12 182 (11)
Year 7–10 187 (11.4)
Employment
Full-time 579 (34.7)
Part-time 555 (33.2)
Unemployed 32 (1.9)
Student 49 (2.9)
Home duties 84 (5.0)
Retired 323 (19.4)
Housing
Own house 1253 (75.3)
Renting 347 (20.8)
Public/rooming housing 49 (3.0)
Other 16 (1.0)
Household income
$150,000+ 258 (17.0)
$100,000–149,999 254 (16.7)
$75,000–99,999 272 (17.9)
$50,000–74,999 276 (18.2)
$30,000–49,999 208 (13.7)
$0–29,000 251 (16.5)

Focus groups

Focus groups (with between 7 and 9 people per group) were undertaken with residents in each of the three LGAs. These participants were recruited through a question in the household survey. Participants were between the ages of 19–69 years and were a mix of male and females. Focus groups lasted between 60–90 minutes. Focus groups were tape recorded and transcribed. Participants received $30 supermarket vouchers for their time and any out-of-pocket expenses.

Public service case data

Analysis was performed on Melbourne ambulance and Victorian police data, between the years 1999 and 2011.

Ambulance data

Ambulance data were derived from alcohol and other drug-related non-fatal ambulance attendances. Data are obtained from patient care records that are completed by attending paramedics. Patient care records are coded and entered into a database by trained project staff. The database contains information including demographic and location characteristics, clinical signs, treatment details and outcomes. The involvement of drugs in the attendance is determined by paramedic clinical assessment and information available at the scene, and cases were included where alcohol played a causal role in the reason for the ambulance attendance. Data were extracted from a database developed for examining non-fatal drug-related ambulance attendances (Lloyd 2012; Lloyd & McElwee 2011).

Police data

Data on assaults and ‘behaviour in public’ offences that were flagged as being associated with alcohol were sourced from Victorian police to assess whether there was any reduction following the introduction of the public drinking restrictions. Due to relatively low numbers, only annual data were available (due to confidentiality issues associated with accessing police data with a small number of cases).

Analysis

Data collected via media analysis, sessions of observation, in-depth interviews and focus groups were stored and analysed using NVivo9, a qualitative software package that enables thematic and content analysis of large amounts of text (Beekhuyzen 2007). Consistent with an inductive approach, analysis was shaped by the themes that arose, but was informed by the themes identified in the literature on public drinking bans and previous evaluations conducted in each of the LGAs.

Quantitative data from the household survey was analysed using Stata Version 11. Analysis involved descriptive statistics and cross tabulations. Variables such as awareness of the law, approval of the law, concerns about public drinking prior to and after the law, engagement in public drinking prior to and after the law, and changes since the law were analysed for the whole sample, and then analysed by subgroups, including gender, age, education and suburb of residence. McNemar’s tests were performed on variables investigating changes pre and post the law.

A time series analysis was performed on ambulance data using Stata Version 11. Data were analysed using auto-regressive integrated moving average (ARIMA) models, as derived specifically by Box and Jenkins (1976). This technique models the temporal structure of the error term, ensuring that any correlations in the series over time do not bias the model coefficients. Due to the low numbers of police data (meaning only annual data were available), time series analysis could not reliably be undertaken. As a result, descriptive analyses of the police data were undertaken.

Results

The implementation of public drinking laws

There was a long history of public drinking in each of the three LGAs targeted in this evaluation prior to the implementation of the law. However, the laws were triggered partly in response to the arrival of more visible, louder and aggressive drinkers. In each of the LGAs, the catalyst for the ban was the right mix of police, council and trader support.

The three main concerns that were frequently highlighted throughout the evaluation and raised as the grounds for the need for a public drinking law were safety, amenity and crime. Public drinkers were often described as aggressive and threatening, responsible for litter and negatively affecting the ‘look’ of an area. Their behaviour was often linked to crime including drug use, theft and assault.

One of the main themes that arose in each of the LGAs was concern that the public consumption of alcohol had increased over time as the result of gentrification and growing socioeconomic disparity in the three areas. An influx of wealthier residents created new demands on the use of public space and drinkers were seen to disrupt the pace at which the area was evolving and businesses were transforming by bringing down the amenity of the area. Over time, it is expected that these tensions will escalate further. The gentrification occurring in these inner city areas may ultimately lead to the displacement of drinkers to outer LGAs of Melbourne, where there is less access to health care, potentially marginalising them further.

Importantly, successful strategies implemented in two of the LGAs (Darebin and Yarra) were to activate an outreach model of care in conjunction with the local law. In Darebin, this resulted in the facilitation of secured housing arrangements for some of the homeless drinkers. Strategies such as this are crucial in minimising the negative impacts to public drinkers.

More work clearly needs to occur at the local council level to heighten awareness of public drinking laws. Across all LGAs, only half of respondents who received a household survey were aware of the law in their municipality and in some areas as little as 12 percent were aware of the specificities of the law.

Regardless, it appears that public drinking laws have the strong support of the community, with three-quarters of household residents (76.4%) who responded to a postal survey supporting laws prohibiting public drinking in the street and just over half (52.3%) supporting laws prohibiting public drinking in parks. Primary areas of concern in relation to public drinking laws are on the street during the day and night, and around shopping centres during the day. Less concern was expressed in this evaluation in relation to night-time drinking around licensed venues, with the exception of noise and litter. Given the high support for public drinking laws, their continued proliferation is inevitable.

The effectiveness of public drinking laws

This evaluation is inconclusive as to whether public drinking laws reduce the visibility of street drinking. This is likely to always remain context-specific. The three LGAs differed in the visibility of public drinking following introduction of the laws. In Maribyrnong, street drinking continues and is frequent and highly visible. In Yarra, street drinking continues but has been reduced. In Darebin, street drinking is now rarely visible. These findings were consistent across various data sources (observations, interviews, survey and focus groups).

Despite these mixed findings, in all three locations, residents reported a reduction in concern about public drinking in specific locations following the introduction of the law. This reached significance in most locations such as on the street during the day and night, around shopping areas during the day at night, around licensed venues during the day and night, and in parks during the day and night. These findings were even true for a suburb of Maribyrnong (Footscray), where by all reports there has been no change in street drinking pre and post law. This suggests that public drinking laws at the very least create a perception that problems relating to drinking have decreased.

When residents were asked whether they believed public drinking had increased or decreased since the law’s introduction, between three percent and eight percent believed it had increased, between 17 percent and 21 percent believed it had decreased and between 20 percent and 30 percent believed it had stayed the same (with the rest being ‘unsure’). This is generally reflective of a small positive trend in relation to perceptions of the effectiveness of the law.

Police reported a range of approaches to addressing public drinking, from warnings and directing drinkers to tip their drinks out (most common) to frequent issuing of infringements. Police were generally favourable of the law, but it was clear that it was not one of their top priorities and infringement data shows it was rarely applied. However, this is likely to vary from one municipality to another.

Consistent with previous evaluations, there was no effect on ambulance attendances in the three LGAs pre and post law. There was a small drop in alcohol-related assaults in two LGAs (Yarra and Darebin) following the law; however, this finding should be considered with caution given that only one data point was available after the law (given that these laws were recent and only annual police data could only be obtained).

The findings are mixed in relation to experiences of harm and perceptions of improved amenity following a public drinking law. For example, there were no differences pre and post law in Maribyrnong (perhaps due to the continued street drinking occurring in that location), but were significant reductions across many negative outcomes in Darebin and Yarra. Most improvements were around perceptions of safety and amenity such as feeling safer in a public place, feeling safer waiting for public transport, feeling safer using public transport, being less likely to need to avoid drinkers, being less likely to be annoyed at vomit, urination and litter, and being less likely to be kept awake at night or disturbed because of someone else’s public drinking. More severe harms, such as being physically or verbally assaulted remained unchanged, perhaps indicating that the public drinking that occurred prior to the law affected perceptions of safety, rather than experiences of safety.

Impacts of public drinking laws

Consistent with previous evaluations, the primary impact of public drinking laws were negative outcomes for drinkers. These included social, cultural, economic, legal and health impacts. In two locations, drinkers reported loss of social and cultural connections as a result of being unable to congregate in the same space with their friends and family, contributing further to their social marginalisation. Drinkers were also often unable to pay fines, sometimes resulting in increased contact with the criminal justice system. Finally, dispersing drinkers meant they were less likely to access medical, health and welfare treatment because community health workers were unable to locate them once they no longer congregated in the same space.

Also consistent with previous evaluations, there was some evidence of displacement resulting from public drinking bans, with drinkers moving to more covert areas to evade police, such as public housing estates, alleys and railway tracks, or private homes. Drinking at home was anecdotally reported to be related to increased harms, such as drinking more and having fewer people around to moderate antisocial behaviour.

A minority of residents reported a small reduction in positive experiences relating to public drinking after the introduction of a public drinking law. This reduction was only evident in Yarra and Darebin (given that street drinking continued in Maribyrnong) and was apparent across all domains (ie have you enjoyed drinking on the street during the day and at night, enjoyed drinking in parks during the day and at night, and had positive interaction with street drinkers during the day and night?). However, despite the small reduction in positive experiences, residents were generally satisfied with the effects of the law.

Discussion

It is not possible to make a definitive judgement as to whether public drinking laws are effective or ineffective, particularly given that there are numerous ways that ‘effectiveness’ can be measured. The findings are mixed as to whether public drinking laws reduce congregations of drinkers and there is no evidence that they reduce alcohol-related crime or harm, but they do make residents feel safer and improve the amenity of an area (from the perspective of residents and traders). Given the high level of public support for public drinking laws, their continued application is inevitable; however, given the negative impacts to drinkers that occur as a consequence of public drinking laws, it is important that they are carefully considered, implemented and enforced, and are coupled with community-specific social inclusion strategies.

It is recommended that public drinking laws should remain the discretion of local councils, rather than becoming state government legislation in Victoria and local council’s should work closely with police to design and enforce public drinking laws specific to the sensitivities and needs of the individual community.

A warning system is the most adequate enforcement approach in terms of fostering a good relationship between police and drinkers and ensuring that disadvantaged groups are not further marginalised. It was generally recognised by all members of the community that punitive measures are not an appropriate way to deal with a social problem. It is important that there is no discriminatory application of public drinking laws and it is recommended that police pay more attention to enforcing public drinking laws near and around licensed venues at night, including people drinking outside designated licensed areas on the footpath of venues, so that different types of drinkers are subjected to the law equally.

The option of a ‘wet-house’ or ‘wet-zone’ was raised but only in a tentative way. Some residents and most police were opposed to the idea of a wet zone and council officers, while recognising its appeal, did not see it as politically palatable. More realistic options, and those that had the most vocal support, were measures such as reducing access to alcohol (fewer liquor licenses, shorter trading hours, raising the cost of alcohol etc) and more health care services for drinkers, particularly culturally specific outreach and detoxification programs.

Limitations

There are a number of limitations to this evaluation that must be considered when interpreting the findings. First and foremost, the household survey had a very low response rate (12.3%), meaning that the findings are limited in their generalisability. In addition, a random white pages sample is not representative of the population, given that this method does not selectively target people of all ages, gender and other characteristics. The household survey was only distributed in English, meaning that the sample is also missing non-English speaking residents.

Secondly, with the exception of police and ambulance data, no pre-law data were available to compare the results with. For example, survey questions investigating changes ‘before the law’ and ‘after the law’ relied on memory and this is subject to limitations of recall bias. While there were a number of previous evaluations to compare themes with, these often used different data collection methods and different survey questions.

Thirdly, a time series analysis on police data was unable to be run, given that only annual data was available for assault and ‘behaviour in public’ offences that were flagged as being associated with alcohol (due to confidentiality issues associated with accessing police data with a small number of cases). It is also important to note that with both ambulance and police data, analyses were run at the LGA level rather than by postcode (due to a small number of cases and confidentiality issues) and in Darebin and Maribyrnong, the public drinking laws only applied to small areas in the LGA. Thus, it is important to consider that this analysis may not have been sensitive to pick up small changes.

References

  • Beekhuyzen J 2007. Putting the pieces of the puzzle together: Using Nvivo for a literature review. 4th QUALIT Conference Qualitative Research in IT & IT in Qualitative Research: Wellington, New Zealand
  • Box GEP & Jenkins GM 1976. Time series analysis: Forecasting and control. London: Holden-Day
  • Chikritzhs TD et al. 2007. Restrictions on the sale and supply of alcohol: Evidence and outcomes. Perth: National Drug Research Institute, Curtin University of Technology
  • d’Abbs P et al. 2008. Kakadu/West Arnhem alcohol management plan project: Final report. Townsville, James Cook University
  • Joffe H & Yardley L 2003. Content and thematic analysis, in Marks D & Yardley L (eds), Research methods for clinical and health psychology. London: Sage
  • Lloyd B 2012. Trends in alcohol and drug related ambulance attendances in Melbourne 2010–11. Melbourne: Turning Point Alcohol and Drug Centre
  • Lloyd B & McElwee PM 2011. Trends over time in characteristics of pharmaceutical drug-related ambulance attendances in Melbourne. Drug Alcohol Review 30(3): 271–280
  • Mast BD et al. 1999. Beer taxation and alcohol-related traffic fatalities. Southern Economic Journal 66(2): 214–249
  • Pennay A & Room R 2012. Prohibiting public drinking in urban public spaces: A review of the evidence. Drugs: Education, Prevention and Policy 19(2): 91–101
  • Webb M et al. 2004. Banning the bottle: Liquor bans in New Zealand. 3rd Australasian Drug Strategy Conference. Alice Springs, Australia: May