A Civilianised Summary Power to Exclude: Perceptual Deterrence, Compliance and Legitimacy

As a response to alcohol-related disorderly behaviours, the use of exclusion has expanded steadily across Australian jurisdictions but with minimal analysis of its effects. Bans, from public or private locations, are typically imposed summarily and presumed to be a meaningful deterrent to future problematic behaviours. The formalisation of licensee banning powers has created a civilianised police-enforceable power to punish by exclusion. In Victoria, the legislative framing of licensee barring order provisions precludes formal monitoring of their use. This article reports findings from interviews conducted with recipients. The conceptual and situational value of barring orders are acknowledged, but their capacity to act as a tangible deterrent or effective agent for behaviour change is far from


Introduction
Entertainment districts are contested spaces and, over the last two decades, an increasing body of regulation and legislation has sought to govern, manage and control alcohol consumption and associated patron behaviours (Hadfield, Lister and Traynor 2009;Hobbs et al. 2003;Taylor et al. 2018). Across Australia, violence and disorder in and around licensed premises have prompted the implementation of an expansive range of policies and operational policing initiatives. High-profile responses, such as lock-out laws introduced in Sydney in 2014, and Queensland's 2017 implementation of compulsory ID scanners at key licensed venues, have generated extensive debate and analysis regarding their need and effect, and the broader context of expanding governance and control (Donnelly and Poynton 2019;McNamara and Quilter 2015;Menendez et al. 2015Menendez et al. , 2016Miller et al. 2017Miller et al. , 2019. A key element of the regulation of entertainment districts is the use of spatial exclusion and patron banning, which has expanded across all Australian jurisdictions with minimal scrutiny (Farmer 2019Farmer et al. 2018;Palmer and Warren 2014). Banning achieves a practical and political need to address alcoholrelated issues and finds little opposition from the liquor industry and other key stakeholders (Miller et al. 2016). Banning provisions differ across jurisdictions (Farmer and Clifford 2020) but a common presumption is that the experience, or risk, of exclusion from a venue and/or local area will deter troublesome patrons . However, without formal oversight of its use or effects, patron banning has become a politically expedient example of 'being seen to do something' rather than a genuine mechanism to promote tangible and sustainable behaviour change (Farmer 2019Farmer, Clifford and Miller 2020).
This article focuses upon venue specific barring orders in the Australian state of Victoria. Licensee barring orders are discretionary police-enforceable sanctions, imposed on-the-spot by ordinary citizens. Barring orders exclude individuals from a venue, limit freedom to move in its vicinity, and are expected to have a transformative effect upon the behaviour of the recipient and the community more broadly. Barring sits within a largely linear model of undesirable behaviour -response -and attendant deterrence. To date, there has been limited evidence-based analysis of the extent to which any type of patron banning influences the future behaviour of recipients. 1 The study from which this article is drawn was the first in Australia to engage directly with ban recipients, using in-depth interviews to explore how they perceived and responded to their ban. The particular focus of this article is the deterrent effect of licensee barring orders. 2 Key perspectives relating to deterrence theories, displacement, exclusion and legitimacy provide a theoretical and conceptual framework for the study. 3 The rationale for Victoria's licensee barring powers and the core assumptions upon which their implementation is predicated are then briefly examined. An explanation of the research approach is followed by the articulation and discussion of key interview findings. The article concludes by highlighting limitations that are evident with respect to the deterrent effects of licensee barring orders, positions the findings within broader concerns about the attendant civilianisation of punishment, and recommends further examination of the gendered dynamics that are evident in this study.

Deterrence, Displacement and Exclusion
Patron banning in general, and licensee barring in particular, fit within a model of contractual governance (Crawford 2003), which individualises regulation (Room 2012) and draws upon a presumption of rational decision-making. Complex constructions of behavioural theory have been examined in relation to a range of criminological, sociological, geo-spatial and psychological contexts (Tonry 2008;Von Hirsch et al. 1999). Of particular relevance to this article is the notion of perceptual deterrence, whereby individuals weigh up the possible benefits of a given act or behaviour against the probable risks -notably the likelihood, celerity and severity of punishment or sanction (Becker 1968;Mann et al. 2016). To be effective, a deterrent must be sufficient to factor into an individual decision to behave in particular way -but individuals are influenced by a complex blend of personal, social, situational, contextual, psychological, attitudinal and risk-related factors (Miller et al. 2013;Taylor, Keatley and Clarke 2020). For example, surveillance and enforcement mechanisms deployed within the night-time economy (NTE), such as the proximity of police IJCJ&SD 3 www.crimejusticejournal.com officers, visible CCTV, or the use ID scanners to identify patrons as they enter venues have the potential to influence behaviours, but the precise deterrent effects are variable and far from conclusive (Ariel, Bland and Sutherland 2017;De Andrade et al. 2020;Piza et al. 2019). Perceptions regarding risk or certainty of a sanction are individualised and the consequences are not homogeneous. Some people are more deterred than others, with the level of deterrence reflecting a range of variables and circumstances (Matthews and Agnew 2008;McGrath 2009). Individual deterrability embodies the extent to which a person has a propensity to perceptual deterrence in any given situation (Jacobs 2010).
The remit of deterrence is broad, and the effects are neither uni-directional nor siloed. Deterrence can apply to individuals receiving a particular penalty, more generally to the community through the perceived risk of a sanction being imposed and enforced, or it can reflect more intricate intersections of specific and general effects. Within the NTE, the confounding effects of alcohol or drug consumption upon cognisance, rational thinking and risk informed decision-making further complicates the paradigm of perceptual deterrence and deterrability. Across a multi-jurisdictional body of research, clear links have been established between alcohol consumption, anti-social and other disorderly behaviours (Fleming 2008;Graham and Homel 2008;Hadfield, Lister and Traynor 2009;Hughes et al. 2008;Mawby 2017;McNamara and Quilter 2015;Miller et al. 2015). The effects of alcohol on decision-making, violence and disorder are evident, but complex.
Patron banning is one of a range of provisions implemented to manage problematic behaviours across the NTE. As a summary penalty, bans issued on-the-spot immediately remove an individual from a given location. However, the potential for the temporal or spatial displacement of undesirable behaviours has been demonstrated to varying degrees with respect to other mechanisms tackling alcohol-related issues in the NTE, and anti-social or criminal acts in other contexts (De Andrade, Homel and Townsley 2018;Menendez et al. 2015;Sylvestre et al. 2015;Wadds 2019). For example, following the implementation of Sydney's lock-out laws, non-domestic assaults decreased in the affected precincts, but rose by 18% in surrounding suburbs (Donnelly and Poynton 2019). In Queensland, the effect of ID scanners is mitigated by time-limited use, enabling patrons to enter venues without detection (De Andrade et al. 2020). In England, Stafford (2019) found that individuals excluded from Business Crime Reduction Partnership 4 locations (following theft, disorder or other anti-social behaviours) typically continued to offend in other locations. Without comprehensive coverage and/or effective enforcement, the potential deterrent effect of exclusion is fundamentally compromised by the capacity to simply go elsewhere.
Mechanisms to exclude recipients from public or private domains operate across a range of international criminal justice contexts. Common themes underpin the framing of each provision and assume effects upon the future undesirable behaviours of recipients, and of the community more broadly. However, the specific individualised deterrent effects of such localised, summary penalties are far from conclusive. In an examination of dispersal orders and other exclusionary powers in England and Wales, Crawford (2009) depicted a 'net-widening' effect of policing by summary justice but with limited consequences for disorderly behaviours. Since 2014, Public Spaces Protection Orders (PSPOs) have empowered local authorities to apply prohibitions within specific public areas across England and Wales. Heap and Dickinson (2018) contend that PSPOs constitute an unregulated and almost unlimited power to prohibit or exclude, but there is little evidence to support their presumed beneficial effects on behaviour and/or community safety (Johnstone 2017). In Seattle, USA, police officers are permitted to issue on-the-spot exclusion orders from public spaces, which can last for up to a year. In common with Crawford (2009), Beckett and Herbert (2010) found that the Seattle provisions progressively increased the likelihood of infringement and punishment, but without demonstrable effects upon behaviour. Belina (2007) critiqued the arbitrary and reductive nature of summary police banning powers in Bremen, Germany and observed that the resulting dislocation of 'undesirables' reflected a spatialization of risk that did little to change individual behaviour or assure community safety. In a study of responses to police zonal banning powers in Denmark, Sogaard (2018) found that bans were routinely disregarded, had limited value as a specific deterrent, and fear of being banned exacerbated rather than deterred anti-social behaviours.
To date, no study has established the effects of any of Australia's banning provisions on recipients, in terms of their perceptions, compliance, or subsequent behavioural decision-making. Palmer and Warren (2014) and Farmer (2017Farmer ( , 2019 identified and examined key conceptual concerns regarding the use of exclusion in public spaces, the arbitrary application of discretionary powers to punish, and the implications for individual rights. Curtis et al. (2018) explored public understanding of patron banning in Victoria via an online questionnaire. As the majority of respondents demonstrated little knowledge, Curtis et al. questioned the extent to which banning fulfils the expectations of a general deterrent. In a series of interviews with key stakeholders, Miller et al. (2016) examined the perceived effects of patron banning and determined that banning is regarded as a deterrent to anti-social behaviours. However, the study did not engage with recipients of patron bans to understand their perceptions or how they responded to their ban -including whether they simply continued their behaviour in another location.
Licensed venues are private spaces and the right to exclude is embedded within long-standing common law principles. Across Australian jurisdictions, codification has steadily formalised and expanded licensee barring powers. 5 Barring orders are non-criminal sanctions, enforced via a range of infringement mechanisms which may result in a fine or court proceedings. Farmer (2019, 2020) documented conceptual and operational issues regarding the lack of specific training for licensees, and the principle of empowering ordinary citizens to punish without implementing a supporting structure of accountability. Despite their formal nature and the potential consequences of a breach, in most Australian jurisdictions there is limited facility for independent review of a licensee decision to ban, and their use is subject to little scrutiny (Farmer 2019. The propriety of permitting non-law enforcement officers to issue legally enforceable sanctions and attendant perceptions of the legitimacy of such quasi-criminal processes aligns with an extensive body of research examining policing legitimacy (Tyler and Jackson 2014;Tyler and Wakslak 2004;Wadds 2015). Even for routine interactions, such as traffic stops, perceptions of procedural justice and operational fairness are essential prerequisites for individual and community compliance, cooperation, and trust (Huq, Jackson and Trinkner 2017;Mazerolle et al. 2013;Tankebe and Liebling 2013). It is reasonable to presume similar perceptions and expectations with respect to non-police powers to exclude.

Victoria's Licensee Barring Provisions: Rationale, Assumptions and Deterrence
Victoria's licensee barring order provisions are set out in sections 106C-K of the Justice Legislation Amendment Act 2011, which amended the Liquor Control Reform Act 1998, and came into effect in May 2011. Barring orders are intended for those who are '"drunk, violent or quarrelsome" in the licensed premises or its vicinity, and can be imposed by the licensee or another responsible person. The permissible length is up to one month for first time recipients and up to six months where two or more previous orders have been imposed. Recipients do not have recourse to legal advice or representation. Barring order breaches are dealt with as criminal matters, with a potential fine that is court enforceable. The 'vicinity' of a licensed premises is defined as 'a public place that is within 20 metres of the licensed premises, but is not the licensed premises.' (Justice Legislation Amendment Act 2011, part 2 'definitions'). The geographical scope of a licensee barring order extends well beyond the private domain of the venue and may include pavements, thoroughfares, car parks, and any other contiguous public area.
Across the parliamentary debates of the Justice Legislation Amendment Bill, 6 in March, April and May 2011, the operational legitimacy and presumed deterrent and community safety effects of the provisions were discussed in depth. Some concern was articulated in relation to imposition mechanisms, enforcement, monitoring, and the diminution of civil liberties. Proponents of the Bill countered these concerns with broadly framed notions of protection and safety, typified by this assertion: These barring orders … are important in reducing violent incidents outside licensed premises and protecting the safety of patrons within the vicinity… (Legislative Assembly 2011: 586-87 In response to a request to explain how barring orders would reduce alcohol-related harm, the Minister directly linked exclusion with specific and general deterrence, but offered no empirical or other evaluative evidence: The government's intention with respect to this clause is to remove those people who have been causing trouble from those licensed venues… The point of increasing penalties for offences is for the provision to act as a deterrent. Clearly if there is a greater penalty, there will be a deterrent, and that is what the government is seeking to achieve. (Legislative Council 2011: 1066 [Guy]) A licensee-imposed barring order can remove a troublesome individual from a given situation for a fixed period of a time. The ban is also expected to deter the recipient from engaging in similar behaviours again, particularly where subsequent transgressions may lead to more onerous periods of exclusion. Since their implementation, Victoria's licensee barring order provisions have been the subject of no formal scrutiny. There is no monitoring of their use, and any effects of licensee barring are not measured. The 2011 legislation precludes licensees from sharing barring order data other than for the purposes of enforcement (Justice Legislation Amendment Act 2011, s.106K). A direct request to the licensing regulatory body, the Victorian Commission for Gambling and Liquor Regulation (VCGLR), for assistance to establish the quantum of barring orders imposed to date was denied. The 2011 legislation was cited as the reason for the absence of any barring order data (personal email to author, 10 May 2019).
The introduction of licensee barring orders reflects assertion rather than empirical evidence. This is a particular concern in Victoria, where the requirements of the 2006 Charter of Human Rights and Responsibilities ('the Charter') requires all legislation which may limit a core individual right (including the freedom to move, the right to a fair hearing or legal representation) to be supported with demonstrable evidence of need and effect. 7 Licensee barring orders limit freedom to move and are imposed on-the-spot by ordinary citizens. Yet compliance with the Charter expectations is not evident in relation to the legislation which introduced licensee barring. A clear gap exists with respect to the presumed deterrent effects of Victoria's licensee barring orders, and demonstrable understanding and evidence of the actual effect of a ban upon both recipients and the wider community.

Research Approach
This study draws upon in-depth interviews with individuals who have received a licensee barring order in Victoria. Ethical approval was provided by the Deakin University Human Research Ethics Committee: reference 2018-287. There is no central record of Victoria's venue barring orders, as no data are collected or published. Licensees record and retain barring order information for three years, after which it must be destroyed. The VCGLR does not monitor the use of barring powers, gather or maintain any records (email to the author, 10 May 2019). Therefore, there is currently no data through which the use or effects of licensee barring orders can be examined.

Participants
Interviews were conducted with 15 individuals who had received at least one licensee barring order anywhere in Victoria. A range of print, online and social media communication channels were used to publicise the research, including Facebook, Twitter, LinkedIn, and the Victorian newspapers Geelong Advertiser and Herald Sun (shared online across NewsCorp Australia 8 ). Prospective participants were invited to contact the research team via a dedicated project email address. A plain language statement was sent to each prospective participant, and those who wished to proceed were asked to return a signed consent form. IJCJ&SD 6 www.crimejusticejournal.com

Procedure and Analysis
The interviews followed a semi-structured format, with core questions posed to each participant and additional questions dependent upon the responses. Interview length ranged from 25 to 55 minutes; each was recorded and transcribed by the interviewer. The interviews focused upon the process through which the barring orders were imposed, how recipients perceived their ban, as well as their immediate and longer-term responses.
Each of the 15 participants has been assigned a random numerical identifier, from P1 to P15. The interview transcripts were coded by a single researcher using thematic analyses within a grounded theory approach. The coding drew out findings in relation to the themes of compliance; implementation perceptions, process and proportionality; transformational effects upon attitudes and behaviours (for recipients and wider peer groups); practical and emotional responses; legitimacy; and perceived deterrent effects (both specifically and more broadly).

Limitations
Participants self-selected and are not necessarily representative of barred patrons. Legislative restrictions placed upon the sharing of barring order information, and the absence of any formal data, make it impossible to discern a representative sample. It is recognised that those who chose to participate may hold particularly strong views regarding their experience. The sample size for this study is 15. It is not uncommon for in-depth qualitative studies to apply smaller samples. For example, papers exploring nightlife policing in Australia (Wadds 2019) and zonal banning in Denmark (Sogaard 2018) comprised 15 and ten interview participants respectively. While care is taken not to over-generalise from the findings, this study addresses a lack of empirical research in relation to the effects of patron banning. It makes a valuable contribution to knowledge in this area, particularly as the VCGLR, the body responsible for overseeing barring powers in Victoria, has no knowledge of the way or extent to which the provisions are used. 9

Research Findings and Discussion
Key descriptive data with respect to the 15 interview participants are summarised in Table 1. The reasons given for the imposition of each barring order are set out in Table 2. Compliance with the legislated requirements is explored in depth within a separate paper (Farmer 2020) but is considered here in the context of the perceived legitimacy of each ban.
IJCJ&SD 7 www.crimejusticejournal.com Due to the absence of central records, it is not possible to establish any sense of the representativeness of this sample. It is not known how many barring orders have been imposed since the implementation of the provisions, for what reason(s) orders have been given, where, to whom or for how long.
To examine the deterrent effects of licensee barring, the findings are framed around the manner and extent to which each recipient perceived that their ban affected their short and longer term behaviour. Three overlapping themes are explored and discussed: compliance and displacement -including commentary regarding ban enforcement, emotional and practical effects; barring order legitimacy and associated behaviour change; and the perceived potential for barring powers to deter undesirable behaviours.

Compliance and Displacement
Of the 15 participants, eight stated that they complied with their ban (four male and four female), two admitted to actual breaches, and four admitted to attempted breaches. One participant refused to respond to this question.
The compliance of four participants was influenced by the perceived risk of being caught attempting to breach the ban, particularly where the venue utilised an ID scanner. This complements the recent findings by De Andrade et al. (2020) in relation to the use of ID scanners to identify banned patrons in Queensland. It points to the potential for a short-term individual deterrent effect, where barring order enforcement is proactive and perceived to be effective. One female participant (P4) admitted to returning to the venue to test the scanning process to see if she would still be allowed to enter (she was not successful). A male participant (P5) explained that the venue from which he had been barred started using their ID scanners at a fixed time each evening. As a result, he intentionally entered the licensed venue earlier, and was able to avoid detection. Another female participant noted that she was known by the licensee and had been told clearly that the police would be called if she was seen anywhere in the vicinity of the venue. Her compliance was directly linked to fear of reprisal if she was found breaching her order: Yes, I complied with the ban … because I was scared … that they would call the police and I would be fined $2000. (P1-Female) One female participant reflected more broadly and expressed fear that the barring order could affect future job prospects. This was the only example of longer-term thinking that was discerned from the interviews, with respect to potential personal consequences of not complying with a ban: I definitely am more wary … I'm scared of the consequences… It made me really nervous to think that… it could affect my prospects in the future … I thought maybe this could come up as a criminal offence or something. (P10-Female) Embarrassment at being the subject of a barring order was another reason given for compliance. This was particularly the case for female recipients, four of whom specifically mentioned embarrassment, with comments such as: It is not known the extent to which bravado, hindsight and/or the passage of time has affected these perspectives. Nevertheless, while acknowledging the small sample size, there was a notable attitudinal difference between male and female respondents. This reflects the individualised nature of deterrability and the influence of a range of factors, including gender (Matthews and Agnew 2008;McGrath 2009;Jacobs 2010). Du Preez and Wadds' (2016) notion of gendered responsibilisation in the NTE, posits that embedded socio-political and cultural expectations ensure that regret and shame are more likely to promote responsible behaviour in women. The response of the male recipients in this sample aligns with established connections between alcohol consumption, violence, masculine identity and '… the symbolic rejection of respectable social values' (Tomsen 1997, p.100;.
Of the six participants who breached or attempted to breach their barring order, two were female and four were male. Reasons ranged from a lack of awareness that a barring order had been imposed to explicit efforts to circumvent the order. The latter was typified by the male participant who simply ensured that he entered the venue before the time at which the ID scanners were activated. Of the two female respondents who breached their ban, P2 admitted to swapping IDs to avoid detection, while P3 took a more direct approach: … I went into the venue anyway during the ban period … I [just] had to convince the security each week to let me in … (P3-Female) There were no notable differences in compliance or associated perceptions of barring orders between regional and metropolitan locations. Similarly, there were no discernible differences across the age groups of participants. Across the interviewees, the level of compliance was high, but this did not reflect an overwhelming sense of renewed lawfulness or any notable transformational behavioural effect of the ban itself. The most common response was that the recipient simply decided to go elsewhere. Seven participants emphasised that they continued their usual pattern of alcohol consumption and behaviour during their ban, but just went to different locations. The following quotes are typical: This aligns with findings of displacement in relation to other exclusionary provisions (Donnelly and Poynton 2019;Stafford 2019). Spatial displacement is enabled by the operation of Victoria's licensee barring orders, as recipients are typically only barred from one venue. Even where ID scanners are used to manage entry into a venue, individual ban recipients will only be identified if they are barred from that particular establishment. As it is generally very easy to just go elsewhere, 10 licensee barring orders are unlikely to drive tangible behaviour change. To mitigate the effects of displacement, barring would need to be more geographically expansive and comprehensively enforced. This would create logistical challenges, amplify rights-related concerns in relation to exclusion from public spaces (Farmer 2017(Farmer , 2018 and risk further displacement to less visible or domestic settings. The required regimes of governance, policing and private security would further challenge tensions between patrons, venues and police, and heighten concerns about the application, reach and consequences of discretionary civilianised powers to exclude .
Victoria currently does not track the movements of barred patrons between venues or locations. This makes the extent and effect of displacement difficult to monitor. It is not known how many barring order recipients go on to receive orders at other venues, or any other behaviour related sanction or infringement penalty -either concurrently or subsequently. Further research could map the movement of barred patrons and identify the extent to which behavioural issues continue during and after the period of a ban.

Perceived Legitimacy and Behavioural Effects
Only one interviewee accepted the fairness of their order and its imposition without question. Four participants regarded their barring order as generally fair and/or appropriate, given the behaviours they had exhibited. Three of the four articulated some concern about the manner of imposition, typically due to others not receiving the same outcome in comparable circumstances. Across the interviews, the reason for the imposition of the barring order was a common concern. Nine of the barring order recipients were given reasons that could align with the legislated purpose of licensee barring orders (Table 2), that the recipient was 'drunk, violent or quarrelsome.' Six reported being given barring orders for reasons that appear to sit outside of the legislated scope. Examples include being under-age, stealing, or due to misidentification. Farmer (2020) sets out a more detailed examination of barring order imposition and operational expectations. Eleven of the interview participants felt that their barring order was unfair; five of whom cited anger or irritation at the venue for imposing the barring order, and its perceived unfairness. This prompted displacement-driven compliance, but motivated by a determination to punish the venue. For example: [I was] angry and determined to never go back there … I was there with three other people and they saw what happened. None of us will go back there again. We have told as many people as we can. If I can be affected, then anyone can. It's really unfair. (P13-Male) The discretionary nature of barring orders, their civilian imposition, and the absence of review mechanisms (Farmer 2019) may serve to increase perceptions of unfairness and anger, which in turn risks creating an additional layer of tension and potential conflict. At best, disgruntled patrons may take their irritation elsewhere, at worst it may fuel hostility with staff at the barring venue. The nature of the participant recruitment means that the sample may skew towards individuals who harbour a grievance about their barring order. Nevertheless, the interviews suggest an association between perceived legitimacy, compliance and the potential effect of a barring order on the future behaviours of recipients. The effects are nuanced, particularly where fear played a part in compliance for recipients who also regarded their barring order as unfair. However, where barring orders are perceived to lack legitimacy this may exacerbate rather than reduce undesirable behaviour.
Participants were asked to reflect upon the longer-term effect of the ban upon their attitudes -to consider the extent to which the barring order acted as a transitional event. Seven reported no longer-term effect upon their attitudes or behaviours. This correlated closely with those who perceived their barring order IJCJ&SD 10 www.crimejusticejournal.com to be unfair or unreasonable. Eight participants admitted to being smarter about their actions after receiving their ban. This included actively avoiding direct contact with certain venue staff: … it made me have less faith in security guards. It also made me more cautious of what I was doing and who was around me. (P5-Male) It has changed my attitude towards security guards at venues and venues themselves. I'm more sceptical of their capability of basic reasoning and questioning… and less likely to assist them when unruly people are around. (P6-Male) It changed my behaviour at that venue, to be more aware and made me identify certain security guards more than others -to watch their behaviour change. (P9-Female) This finding further challenges expectations that barring orders will drive more positive patron behaviours. More than half of those interviewed actively sought ways to circumvent or modify the effects of their exclusion. When combined with the potential for displacement and lack of effective enforcement, the deterrent effect of barring orders remains far from conclusive.
Three respondents acknowledged a discernible change in their outlook and/or behaviours following their barring order -all were female. When asked about the effect of the barring order upon their peer group, while the responses can only reflect the perceptions of participants, the same three female interviewees were the only respondents to identify a discernible change in attitude or behaviour among their family or friends. Within a general reflection of the effect of banning, one set out the differences between male and female responses more explicitly: I would definitely say the girls over the boys have definitely changed their perceptions and attitude towards everything the most … I would say bans had a stronger effect on the girls. (P2-Female) A different perspective was presented by P1 who, despite the fear she felt, asserted that the way in which her barring order was imposed actively negated any potential longer-term effect. For P1, the perceived lack of legitimacy removed any need for a change in behaviour: … no, it doesn't make me behave any differently. I'm just amazed that licensees can do this out of the blue … (P1 -Female) While these are individual reactions, they support a need to examine regulation and associated behaviours in the NTE through a gendered lens (Du Preez and Wadds 2016;Moore et al. 2020;Nicholls 2018).
The concerns articulated by the interviewees point to a risk that licensee misuse of barring powers may undermine their legitimacy and limit the potential beneficial effects -the issue of misuse is examined in more depth by . This risk to the perceived legitimacy of barring is arguably bolstered by the absence of scrutiny of the provisions, including the lack of any central records. Put simply, licensees know that their use of barring powers is not monitored, and this creates an inherent risk of their inappropriate application. One option to address this deficiency and to strengthen the perceived legitimacy of the provisions would be to introduce formal consequences for licensees found to have misused their powers. In Queensland, in principle support has been given to the introduction of an offence of the vexatious imposition of a barring order (Miller et al. 2019;Queensland Government 2019). Such a mechanism may serve to increase the perceived legitimacy of barring provisions and, in turn, may enhance their deterrent and behavioural effects. IJCJ&SD 11 www.crimejusticejournal.com

Potential Deterrent Effects of Barring
All of the participants acknowledged the potential beneficial effects of licensee barring order provisions. There were no objections to the principle of exclusion, and all interviewees perceived that barring could deter problematic behaviours. A number of common prerequisites were articulated all of which embodied expectations of operational legitimacy. Participants expected bans to be imposed fairly, consistently, and openly, and to be enforced effectively. The potential for licensee barring to have a tangible effect on the behaviour and decision-making both of recipients and the wider community was set out clearly, but with caveats: I think it [a ban] serves a purpose and I think that it works at changing peoples' behaviours, however not so much when they aren't abided by… because if people see someone being banned then they are unlikely to repeat the behaviour in fear of also being banned. (P3 -Female) Sometimes I believe it can have the reverse effect than what is intended. But it may also make people realise that they did the wrong thing and learn their lesson the hard way. (P4 -Female) I think that when complied with and when given to the right people, they change attitudes and behaviours and make people (particularly males) scared and worried about doing the same behaviours. (P5 -Male) For this sample, the operation of licensee barring order powers and associated perceptions of the fairness and proportionality of their imposition are key markers for the potential beneficial effects of the provisions. That barring orders are not currently regarded as inherently fair, appears to be the primary reason for the limited enduring deterrent effects reported by the participants. The way in which barring orders are used, enforced and monitored has a fundamental effect upon their capacity to act as a meaningful deterrent to anti-social, violent and other unacceptable behaviours within licensed venues. This aligns closely with the expectations of procedural justice and legitimacy that we see in relation to policing (Mazerolle et al. 2013;Tyler and Jackson 2014;Tyler and Wakslak 2004).

Conclusion
Patron banning is a tangible, straightforward, low cost, and politically expedient response to alcoholrelated behavioural issues. However, any policy that permits the civilian imposition of a police-enforceable sanction should be supported by demonstrable evidence of effect, and its operation subject to meaningful and ongoing oversight. In Victoria, the Charter mandates empirical evidence to support any measure which limits an individual right, regardless of how undesirable or troublesome that individual may be perceived to be. Licensee barring orders limit freedom to move and circumvent the presumption of innocence, but they have operated for nearly a decade without any evidence of their presumed deterrent effects. There is no data available to enable or inform an examination of the use of licensee barring powers in Victoria. We do not know how many barring orders have been imposed, for what reason(s), where, to whom or for how long. As no data are collected, it is not known what proportion of venue barring order recipients participated in this study, or to what extent the sample mirrors barring order imposition patterns or recipient demographics.
For those interviewed, fear, embarrassment or anger underpinned short-term compliance. There is little evidence of perceptual deterrence or longer term behaviour change. Ban recipients were not notably affected by the likelihood, celerity or severity of a further ban (Becker 1968;Mann et al. 2016). Displacement was the most common short and longer-term effect of the barring order. Going somewhere else was a practical response for the duration of the ban and, for some, a more principled reaction when the recipient objected to the legitimacy of the ban. This response may result in a reduction in behavioural issues at one location, but not necessarily in general. This mirrors findings in relation to other regulatory responses, most notably Sydney's lockout provisions (Donnelly and Poynton 2019). For Victoria's licensee IJCJ&SD 12 www.crimejusticejournal.com barring orders, any short-term deterrent effect within specific venues must be understood within a wider context of the potential displacement of banned patrons to other venues, and/or the continuation of problematic behaviours in other public areas or at home. Patron banning individualises the risk and regulation of problematic behaviours (Crawford 2003;Room 2012). However, if individual deterrence is marked by compliance, the deterrent effect of barring orders appears to be, at best, short-term. Most of the interview participants perceived that their ban had no transformative effect upon their outlook or behaviours. Admissions about being smarter in their interactions with venue staff, and actively concealing their behaviours demonstrate how bans can be circumvented to further diminish their effect. Every participant acknowledged the potential merits of barring order powers and agreed that bans could lead to behaviour change. But their effectiveness, as a deterrent and as a driver for broader behavioural change, is inextricably linked to the perceived fairness of the provisions, the consistency and predictability of imposition, effective enforcement, and an overall sense of justice and legitimacy. These essential preconditions were not evident for, or perceived by, the majority of the study's participants.
None of the participants regarded intoxication as the primary reason for their barring order, and none admitted to an alcohol-induced reduction in cognisance at the time of their ban. The subsequent displacement decisions, proactive attempts to circumvent the bans, and articulation of anger and embarrassment all suggest a high level of ongoing rational thought in relation to the bans, and preplanning of responses -many of which do not reflect the deterrent-related behavioural and community safety improvements asserted by proponents of the provisions during the 2011 parliamentary debates (Legislative Assembly 2011; Legislative Council 2011).
The gendered nature of the findings, particularly in relation to compliance, highlights the need for further research to identify and examine the different responses of male and female recipients. Despite extensive research exploring gendered behaviours in the NTE (Graham and Livingston 2012;Nicholls 2018;Tomsen 1997Tomsen , 2008, gender per se is rarely the focus of alcohol policy Moore et al. 2017Moore et al. , 2020. Responses tend to be generic, targeting alcohol consumption, regulation and general control. The belligerence evident from some male respondents, while moderated by the size of the sample, highlights potential challenges for discretionary, civilianised punishments within the NTE -particularly with respect to enforcement. Licensee barring orders can be imposed on-the-spot by individuals without the implied authority or legitimacy of a police officer. In circumstances where a barring order is perceived as unwarranted, this may increase the risk of conflict between patrons and venue staff. This may be exacerbated should ID scanners be used more widely to enforce barring order provisions, particularly if their use by venue staff is confrontational or provokes aggressive reactions (Graham and Homel 2008;Graham and Livingston 2012;Hobbs et al. 2003;Monaghan 2002;Wadds 2015).
Licensee barring orders have been operational in Victoria since 2011 without monitoring, analysis, or parliamentary scrutiny. Their legislative framing actively prevents the central collection of data. As a result, licensee barring exists in an accountability vacuum -without any understanding of how many have been imposed or what effects they may have had. A similar lack of scrutiny is evident with respect to patron banning more broadly across Australia Farmer, Clifford and Miller 2020). The deterrent effects of spatial exclusion continue to be largely presumed but not subject to meaningful oversight or analysis.
The findings of this study acknowledge the potential value of licensee barring but question its effectiveness as a tangible deterrent or an agent for behaviour change. Licensee barring orders place a policeenforceable summary power in the hands of non-law enforcement and non-judicial officers, without any oversight of its use. This lack of scrutiny creates a clear risk of misuse, with an attendant effect on the perceived legitimacy of the provisions and a greater likelihood of transgression or patron-venue conflict. In addition to further research to examine the particular effects of banning, a review of the operation of Victoria's licensee barring powers is recommended, along with consideration of specific legislative IJCJ&SD 13 www.crimejusticejournal.com changes. These include the creation of a centralised record of barring orders to enable their use to be understood and to facilitate effective ongoing monitoring; the proactive mapping of barred patrons in relation to concurrent and subsequent arrests, orders or infringement provisions; and the introduction of an offence to address the risk of the vexatious use of licensee barring powers.