Kallin Pty Ltd v Independent Liquor and Gaming Authority

The role of the Tribunal

1. A decision on an application for the granting of a packaged liquor licence is a administratively reviewable decision under s 7 of the Administrative Decisions Review Act 1997 ("the ADR Act"): see also s13A Gaming and Liquor Administration Act 2007 and cl 7 Gaming and Liquor Administration Regulation 2016. The Tribunal's role is to decide what is the correct and preferable decision, having regard to the material then before it (ADR Act s 63(3)). That includes material which may not have been before the Authority at the time it made its decision. In determining the application, the Tribunal may exercise all of the functions that are conferred or imposed by any relevant legislation on the Authority. Pursuant to s 63(3) the Tribunal may decide: "(a) to affirm the administratively reviewable decision, or (b) to vary the administratively reviewable decision, or (c) to set aside the administratively reviewable decision and make a decision in substitution for the administratively reviewable decision it set aside, or (d) to set aside the administratively reviewable decision and remit the matter for reconsideration by the administrator in accordance with any directions or recommendations of the Tribunal."
2. Section 3 of the Liquor Act provides: "3 Objects of Act (1) The objects of this Act are as follows: (a) to regulate and control the sale, supply and consumption of liquor in a way that is consistent with the expectations, needs and aspirations of the community, (b) to facilitate the balanced development, in the public interest, of the liquor industry, through a flexible and practical regulatory system with minimal formality and technicality, (c) to contribute to the responsible development of related industries such as the live music, entertainment, tourism and hospitality industries. (2) In order to secure the objects of this Act, each person who exercises functions under this Act (including a licensee) is required to have due regard to the following: (a) the need to minimise harm associated with misuse and abuse of liquor (including harm arising from violence and other anti-social behaviour), (b) the need to encourage responsible attitudes and practices towards the promotion, sale, supply, service and consumption of liquor, (c) the need to ensure that the sale, supply and consumption of liquor contributes to, and does not detract from, the amenity of community life."
3. Section 3(2) of the Liquor Act applies to the Tribunal when exercising its function of review (Auld v Independent Liquor and Gaming Authority [2018] NSWCATAD 25) The Act specifies that I must consider any submissions made to the Authority in relation to the licence (s 43(2) and the Community Impact Statement (s 48(5)(a))

Preliminary issues

1. One issue of contention was whether the ILGA's Guideline 6, published under s 57 of the Liquor Act, had the status of being a Government policy under s 64 of the ADR Act. That section provides: "64 Application of Government policy (1) In determining an application for an administrative review under this Act of an administratively reviewable decision, the Tribunal must give effect to any relevant Government policy in force at the time the administratively reviewable decision was made except to the extent that the policy is contrary to law or the policy produces an unjust decision in the circumstances of the case. (2) The Premier or any other Minister may certify, in writing, that a particular policy was Government policy in relation to a particular matter. (3) The certificate is evidence of the Government policy concerned and the Tribunal is to take judicial notice of the contents of that certificate. (4) In determining an application for an administrative review under this Act of an administratively reviewable decision, the Tribunal may have regard to any other policy applied by the administrator in relation to the matter concerned except to the extent that the policy is contrary to Government policy or to law or the policy produces an unjust decision in the circumstances of the case. (5) In this section: Government policy means a policy adopted by: (a) the Cabinet, or (b) the Premier or any other Minister, that is to be applied in the exercise of discretionary powers by administrators.
2. The Guideline states that it: "…concerns the process by which the Independent Liquor & Gaming Authority ("Authority") considers the likely social impact of a licence, authorisation or approval on community well-being when determining whether to grant certain applications. It provides information to relevant stakeholders (including applicants for a licence, authorisation or approval and members of the community) about the type of information that is likely to be of assistance to the Authority when considering that impact."
3. Guideline 6 was published under s 57 of the Liquor Act which provides: "57 Authority may establish administrative policies and procedures in relation to licensing matters (1) The Authority may approve policies and procedures for administering the licensing scheme under this Act. (2) Any such policies and procedures may be applied by the Authority: (a) in dealing with applications for licences, authorisations or other matters that may be granted by the Authority under this Act, and (b) in determining those applications." There was no evidence that the Guideline was adopted by the Cabinet, the Premier or any Minister. However it is evidently a policy "applied by the administrator in relation to the matter concerned" within the meaning of s 64(4) and it contains matter relevant to the consideration of social impact, therefore it is clearly relevant to the issues before the Tribunal in this case. In Smith v Independent Liquor and Gaming Authority [2018] NSWCATAD 224, the Deputy President had regard to the Guideline on the basis that it came within s 64(4). In Auld v Independent Liquor and Gaming Authority Senior Member Ransome also considered the Guideline to be "unobjectionable" and applicable to the matters before her. I have had regard to the Guideline as a policy under s 64(4).

The matters to be considered

1. The term "overall social impact" used in s 48(5) is not defined in the Act, nor are the terms "well-being" and "detrimental". The Tribunal was referred to the decision of the Victorian Court of Appeal in Macedon Ranges Shire Council v Romsey Hotel Pty Ltd [2008] VSCA 45; (2008) 19 VR 422. That decision considered s 3.3.7 of the Gambling Regulation Act 2003 (Vic) which stated that the Commission for Gambling Regulation was not to grant an application for approval of gaming premises unless it was satisfied that "the net economic and social impact of approval will not be detrimental to the well-being of the community of the municipal district in which the premises are located."
2. At [42-43] the Court of Appeal noted that the legislation did not specify the matters which the decision maker should consider in deciding whether it was satisfied as to that issue, and said: "The identification of the matters which the decision-maker is bound to consider is, accordingly, a matter of implication from the subject matter, scope and purpose of the Act. …"Social impact" means the impact on the society or community (or some part or parts of it) in which the gaming machines are proposed to be located. The 'well-being' of a community is a very broad concept. Dictionary definitions indicate that well-being is to be measured (at least) by the extent to which the community is healthy, happy, contented and/or prosperous."
3. The Tribunal is to consider the "overall social impact" and in Smith, Deputy President Hennessy concluded that this meant both positive and negative social impacts were to be considered.
4. Guideline 6 states that in determining the overall social impact, the Authority will consider: 1. the type of the premises applied for - in this case it is a packaged liquor store, commonly known as a bottle shop; 2. the scale of the proposed premises; 3. the trading hours; 4. the location of the premises; and 5. whether any special measures will be implemented at the premises.
5. I am satisfied that each of these factors are likely to be relevant to the overall social impact, depending upon the circumstances.
6. As suggested in Macedon Ranges Shire Council, the meaning of "overall social impact" will be influenced by the subject matter and purpose of the Liquor Act. Clearly the Act is concerned with the minimisation of harm associated with misuse and abuse of liquor (including violence and other anti-social behaviour), as well as ensuring that the sale, supply and consumption of liquor contributes to, and does not detract from "the amenity of community life" (s 3(2)). Therefore in my view, these factors, where relevant, will inform the meaning of "overall social impact". The Act also refers to other social impacts such as the balanced development of the liquor industry and of related industries such as the live music, entertainment, tourism and hospitality industries.
7. In interpreting the term "amenity of community life" in s 3(2)(c) I note that "amenity" has the following meanings according to the Macquarie Dictionary: "1. the quality of being pleasant or agreeable in situation, prospect, disposition, etc.; pleasantness: the amenity of the climate. 2. (plural) agreeable features, circumstances, ways, etc. 3. (plural) features, facilities, or services of a house, estate, district, etc., which make for a comfortable and pleasant life."
8. Both parties referred me to New Century Developments v Baulkham Hills Shire Council [2003] NSWLEC 154; 127 LGERA 303 , where at [53-54] the meaning of "amenity" was discussed in the context of a development application: "The concept of "amenity" is wide and flexible (Broad v Brisbane City Council & Anor (1986) 59 LGRA 296), transcending the merely physical content (Perry Properties Pty Ltd v Ashfield Council (No. 2) (2001) 113 LGERA 301 at 317 per Bignold J). The following passage, contained in the judgment of Thomas J in Broad at 299, attempts to articulate the notion: The wide-ranging concept of amenity contains many aspects that may be very difficult to articulate. Some aspects are practical and tangible such as traffic generation, noise, nuisance, appearance, and even the way of life of the neighbourhood. Other concepts are more elusive such as the standard or class of the neighbourhood, and the reasonable expectations of a neighbourhood. 54 "Amenity" may embrace the effect of a place on the senses, as well as the residents' subjective perception of their locality. Knowing the use to which a particular site is, or may be, put may affect one's perception of amenity, as explained by de Jersey J in Broad at 305: There is no doubt that the concept of amenity is wide and flexible. In my view it may in a particular case embrace not only the effect of a place on the senses, but also the resident's subjective perception of his locality. Knowing the use to which a particular site is, or may be, put may affect one's perception of amenity."
9. The respondent submitted that "well-being" in s 48(5) is not limited to physical and mental health but also extends to quality of life. The dictionary definition of wellbeing is "good or satisfactory condition of existence; welfare". In Macedon Ranges Shire Council the Victorian Court of Appeal considered that well-being meant: "healthy, happy, contented and/or prosperous", which is fairly consistent with the respondent's submission, and I adopt that interpretation.
10. "Detrimental" has the dictionary meaning of "causing detriment; injurious; prejudicial" (Macquarie Dictionary). "Detriment" is a cause of loss or damage. The applicant submitted that the meaning of the word should be influenced by the context of the Liquor Act; it should not extend to every possible cause of detriment, but only those that the Act concerns itself with. While the meaning of the word is influenced by its context, I do not agree that the meaning is limited only to the causes of detriment that are specifically mentioned in the Act. There is nothing in the Act which suggests this.
11. In Smith, Deputy President Hennessy stated at [25]: "The question of causation (of granting the licence and any particular social impact) is not a question that can be reached intuitively. Common sense and experience are not sufficient to conclude that granting an extra packaged liquor licence in Kurri Kurri will increase the rates of assault in the area or have any other negative social impact. There must be some objective facts from which the inference that there will be a particular social impact can be drawn."

Precautionary principle

1. The parties had opposing views concerning whether the Tribunal should apply "the precautionary principle" to the evidence. The respondent submitted that the restriction in s 48(5) reflects a precautionary approach by recognising that detrimental consequences of alcohol can be difficult to quantify in a particular case, but this should not mean that the licence should be granted. The respondent submitted: "This means that some applications must be refused where the social impacts are uncertain or where the positive and negative social impacts are closely balanced so that the decision-maker cannot be satisfied as to what the overall social impact will be. That is a consequence of the precautionary approach taken in the legislation".
2. The precautionary principle stems from environmental law and has been defined as follows: "The existence of scientific uncertainty becomes a trigger for precautionary measures, and for 'reversing the onus of proof' so that proponents, rather than regulators, bear the burden of demonstrating that there is no need for regulatory action." [1]
3. Unsurprisingly the applicant disagrees with this approach. It submits that the wording of the legislation does not indicate such a principle should be applied.
4. Senior Member Ransome addressed whether one party bore the onus of proof in such proceedings in Auld: "Such characterisations are not necessarily helpful in the context of administrative review. As already stated, the Tribunal's role is to reach the correct or preferable decision after an assessment of the evidence before it. It is sometimes said that an applicant bears an "evidentiary burden" in administrative review proceedings to put sufficient evidence before a tribunal to enable it to reach the requisite level of satisfaction in the case before it. There is some force to this proposition but it does not change the tribunal's role in that a decision must be made after considering all relevant evidence, irrespective of the party who adduces it." [50]
5. See also Smith at [25] and Aldi Foods Pty Ltd v Independent Liquor and Gaming Authority [2019] NSWCATAD 26 at [20]. In my view the legislative test is sufficiently clear and it is not necessary to apply this principle.

The nature of the proposal

1. The following information comes from the Community Impact Statement prepared by the applicant and is not in dispute. The applicant's proposal is to open a bottle shop at 261 Bondi Road on the south side the road. Bondi Road is the main street of Bondi and runs between Bondi Junction and Bondi Beach. The shop would occupy a ground floor area within a two storey building of approximately 85 square metres (excluding the rear storage area but including the cool room). The conditions of the development approval do not allow trolleys to be used in the shop. It is expected that most shoppers will arrive and depart on foot. The proposed trading hours are 10.00 am to 8.00 pm Sunday to Wednesday, and 10.00 am to 9.00 pm Thursday to Saturday.
2. CCTV will be installed at the shop and the development consent conditions include requirements to maintain the amenity of the neighbourhood such as ensuring patrons do not loiter in the vicinity so as to obstruct pedestrians, and control of noise and litter.
3. There was some debate about the type of alcohol the shop would sell. Mr Duane assumed it would specialise in mid to high level wines. However there was evidence from the company's website that Chambers Cellars sells wines from $3.99 and upwards. Its catalogue No. 183 for November-December 2018 was in evidence and this offered discounts and specials. According to my perusal of this document, eighty one of the products offered in the catalogue were wine, eleven were beer (not including one beer product and four wine products offered as a bundle) and twenty four were spirits. Unless a condition is placed on a packaged liquor licence, the kinds of products displayed or sold are not restricted.
4. The respondent's evidence was that the store will not stock items that have a special appeal to minors; will not engage in advertising that promotes irresponsible drinking and will not engage in harmful discounting.

The location

1. It is not in dispute that Bondi Road runs past Waverley Park and Waverley Oval (although these are not in Bondi) and which is about 800 metres from the site. A park is about 600 metres from the site towards the beach. Bondi Beach is 781 metres from the site. A number of shops, cafes and restaurants are located along Bondi Road. Parking was described by witnesses as very limited. The other physical outlets for packaged liquor in Bondi are on this road with the exception of the bottle shop at the Royal Hotel which is entered from Denholm Street around the corner from the proposed site. The entrance is less than 100 metres away from the proposed site but not visible from it.

Other liquor outlets in the area

1. Mr Gavin Duane, an economist who has provided independent advice in the field of market analysis and strategic research for clients in the property industry, was the applicant's expert witness in this area. He drew on sources of Marketinfo estimates prepared by MDS Market Systems. Marketinfo is a microsimulation model which marries data from the ABS Household Expenditure Survey with other data such as the ABS Census of Population and Housing to estimate spending levels across a small area. He had also examined publicly available information about the operation of the other outlets.
2. The following evidence concerning the nature of the outlets was not in dispute. There are a total of licensed liquor outlets in the broader community which offer packaged liquor sales. Of those, 13 are currently operating under packaged liquor licences. Within the local community there are 3 packaged liquor retailers - Bottlemart at the Royal Hotel, Kemenys and Krinsky's Supermarket. The latter offers kosher wine only. Kemenys is estimated to be the largest packaged liquor outlet in the broader community and trades from 8 am to 9 pm daily. The Royal Hotel is adjacent to the proposed site and its bottleshop is open from 10 am to 11 pm Wednesday to Saturday and 10 am to 10 pm Sunday to Tuesday. Mr Duane stated that the Royal Hotel had recently been purchased by the Merivale Group which might be able to grow the liquor market in the area through differentiation of products.
3. There are also four online liquor retailers in Bondi which do not have shopfronts and deliver to the whole of Australia. One only sells premium French boutique wines.

Outlet density in the area

1. The applicant submitted that as the application was for a packaged liquor licence, when measuring the density of liquor outlets in Bondi for the purpose of determining social impact, only the packaged liquor outlets with a shopfront in Bondi should be included.
2. If the four internet retailers are excluded there are only two existing packaged liquor outlets in Bondi itself, and one of those is limited to selling kosher alcohol, although the nearby Royal Hotel's hotelier licence also allows the retail of packaged liquor. I accept the applicant's evidence that internet retailers potentially serve a much wider area than the local or broader community, even if their business address is within the community. In addition there was evidence that one sold exclusive and expensive wines, and one was not operating (although there is no guarantee that it will not operate in the future.)
3. Mr Smith and Mr Duane, expert witnesses for the applicant, expressed the view that the volume of alcohol sold, not outlet density, is the preferable measure of alcohol consumption in an area. However retailers of alcohol such as the applicant are not required by the Government to keep records of how much alcohol they sell and hence this is why Mr Duane's evidence is drawn from estimates of spending levels rather than sales figures. Mr Duane agreed, however, that there is a tendency for people to under-report their alcohol consumption.
4. Mr Whetton, the respondent's expert witness, agreed that volume of alcohol sold would be a better measure if it were available. In its absence, the studies which use licence density to measure the relationship between alcohol and social harm do not distinguish between packaged liquor licences held by online retailers and other retailers. If the online retailer's licences were excluded, this would render any comparison of outlet density in the local region to the State average meaningless, as the State averages would include such outlets. It also makes it difficult to apply research findings where the authors did not distinguish between types of packaged liquor licences. [2]
5. It is hypothesised by one of the researchers on whose work Mr Whetton relied, that increased density of alcohol outlets reduces the real costs of alcohol for those who do not plan their purchase of alcohol ahead, and may impact the drinking patterns of some drinkers. [3]
6. On the one hand, including the online retailers when looking at packaged liquor licence density in Bondi seems disproportionate where there are four such retailers in an area of 10,200 population and only six such licences in total.
7. However the Royal Hotel in Bondi, which is a short distance from the proposed site, will not be included in the number of packaged liquor licences for Bondi because it has a hotel licence, even though it sells packaged liquor. In my view in order to draw conclusions from the research literature it is preferable to adhere to the same parameters used in that literature.

The nature of the local and broader community

1. Mr Duane provided the key evidence on this subject and his evidence was that the local community population was estimated at 10,245 and projected to increase to 10,845 by 2031. The broader community population was estimated at 67,692 and projected to increase to 71,012 by 2031. The socio-demographic profile of the local and broader community is characterised by a young, affluent, worker population.
2. Large numbers of tourists visit the local and broader communities which in 2015/16 was estimated by Tourism Research Australia to amount to a total of 646,400 visitors per year, 342,200 of which were domestic day trippers, 212,600 of which were domestic overnight visitors and 85,800 of which were international visitors.

Expenditure on alcohol in the local and broader community

1. Mr Duane's evidence was that average packaged liquor spending (across all types of venues) per capita on beer in 2016 in the local community was slightly higher than in the broader community and the Sydney Metropolitan area. Expenditure on wine in the local community was the same as in the broader community but significantly higher than the Sydney Metropolitan area. Expenditure on spirits was very similar in the local and broader communities and somewhat higher than the Sydney Metropolitan area. Total packaged liquor spending for the local community in 2016 was estimated at $12.4 million, with $80.2 million for the broader community. This was estimated to increase by 1.9% per annum in the future.
2. Mr Duane estimated the packaged liquor retail sales in the local community to be in the vicinity of $10 to $11 million per year, and in the broader community approximately $45 million per year. The gap between sales of alcohol within the local and broader communities and expenditure of local residents on alcohol as a whole indicated, according to Mr Duane, that residents were purchasing alcohol outside the broader community, which in turn was evidence, in his view, of an unsatisfied demand among local residents.
3. Visitors also spend money on liquor in the local and broader communities. Expenditure on alcohol and drinks (in all types of licensed premises) was estimated at $24.1 million across all categories of visitors.

The Community Impact Statement

1. A Community Impact Statement was prepared by the applicant. The applicant consulted with numerous stakeholders including local police, the Waverley Council, the Local Health District, Roads and Maritime Services, Department of Community Services, the Department of Aboriginal Affairs and Local Aboriginal Land Council, Non-English Speaking community groups, and local schools.
2. It considered a number of risk factors such as the type of outlet proposed, the outlet density in the local and broader communities, the location of any at risk groups or sensitive facilities near the store, the incidence of alcohol-related crime in the area and the incidence of alcohol-related health problems in the area. The submission contained a report by Mr Duane.
3. It concluded that: 1. In the past 10 years alcohol-related crime is stable or trending downward in the local community. 2. Bondi Road is not a major pedestrian corridor or "crime corridor" as suggested by NSW Police. 3. There has been no adverse impact on rates of alcohol-related crime in other areas where a Chambers Cellar store has been granted a licence. 4. There is a genuine need or public expectation for a new store. 5. The store will have reduced trading hours. 6. The proposal will not lower the real price at which packaged liquor is sold in the local and broader communities. 7. The proposal will not expand the market for alcohol in the local and broader community or cause existing consumers of alcohol to consume more than they already are.

Community submissions

1. The applicant engaged in community consultation as required by the Act. A number of community submissions were received as part of the application process. The submissions for and against were fairly balanced in number (14 against and 17 in favour). Two objections were received from the licensee and an employee of the Royal Hotel.
2. The Roads and Maritime Services submission noted that alcohol impairment was a factor in 2.7% of all vehicle crashes in Sydney in 2011-2015. Within the Waverley Local Government Area in 2015 there were 5 crashes resulting in 5 casualties. RMS requested that if the licence was granted the licence contain a number of conditions to increase awareness and reduce incidence of alcohol-affected driving and pedestrian activity.
3. Reasons given for objections included: 1. There were already sufficient retailers of packaged liquor in the vicinity. 2. The shop was close to a bus stop and the attraction of the shop to inebriated people would create a risk for people at the bus stop. 3. The Royal Hotel was very close to the proposed location and it sells packaged liquor. 4. It would increase the incidence of intoxication, undesirable behaviour, noise, litter and nuisances on the street and on resident's property, particularly at night. 5. There would be an increase in people seeking parking in an area where parking is already in short supply.
4. Reasons given for supporting the proposal included: 1. Greater choice of products in the area for local residents. 2. Attracting more shoppers to the vicinity. 3. Creation of local jobs. 4. Chambers Cellars is a respected operator. 5. Providing a local option in Bondi apart from the Royal Hotel and Kemenys.
5. A submission was received from Sgt Peter Bolt of the Eastern Suburbs Local Area Command objecting to the development, contending that there are two crime "hotspots" in the Waverly LGA being Bondi Beach and Bondi Junction, and that Bondi Rd is known to police as a "crime corridor" as it links those two centres. Police relied on BOCSAR data from October 2014 to September 2015 to submit that the rate of alcohol related assault per 100,000 population in the LGA was 172.8, higher than the State rate of 144.3.

Crime Statistics and analysis

1. ILGA Guideline 6 states that for the purpose of its functions, crime statistics and analysis will routinely be obtained by the ILGA from the Bureau of Crime Statistics and Research (BOCSAR). This information may be taken into account when considering the extent of prevailing social problems and crime trends in the relevant communities. BOCSAR statistics and analysis have been considered as part of the evidence in these proceedings.
2. ILGA may also refer to licence density and clustering data and information about the type, location and density of other licensed premises in the local and broader communities.
3. Annexure A to the Guideline states that ILGA may apply a number of general propositions from research literature regarding alcohol related crime. I have not relied on these propositions except where they accord with probative evidence adduced in these proceedings.

Positive social impacts

1. A number of positive social impacts were identified.

Convenience for those living nearby, visitors and commuters

1. Unchallenged evidence from the applicant's witness Mr Duane, drawn from the Australian Bureau of Statistics Census of 2011, was that a greater proportion of residents in the local and broader communities do not own a car as opposed to Greater Sydney (19% and 21% compared to 12%). This suggests that convenience to home and local transport may be important for them. A number of the community submissions in favour of the proposal mentioned that the new site would be convenient for them.
2. It is evident that the location is on a main bus route between two suburbs with large numbers of visitors and commuters. It is also evident that lack of parking in the vicinity is an issue. Therefore I am satisfied that convenience for commuters, local residents and visitors would be increased if the licence were granted, and that is a positive social impact for the local and broader communities.

Greater choice

1. The nature of the other outlets in the local community are dealt with above. It is evident that the addition of a packaged liquor outlet will offer greater choice in the local community, however I note that Chambers Cellars also seeks to differentiate itself by reference to the wine it sells. Greater choice was seen as a benefit by some residents and I accept it is a positive social impact for the local community, while perhaps having less impact on the wider community.

Increase of competition

1. The evidence in this area was inconclusive. Mr Duane stated that he anticipates that the majority of the store's projected sales ($0.8 million out of a projected $1.2 million) would come from the local community, therefore its competitive effect outside that area would be minimal. It was hypothesised by the respondent that other outlets might change their business practices in response to the granting of a new licence. Competition may have the effect of encouraging outlets to drop their prices which can in turn lead to an increase in alcohol consumption, which is not a positive impact in all cases. However, the evidence was not sufficient for me to reach a view on this point.

Creation of jobs

1. The Community Impact Statement stated that the store would create an additional 2 staffing positions and people residing in the local community with the right qualifications and experience would be at "a distinct advantage" when seeking employment at the store, but there was no conclusive evidence that the staff employed would be drawn from the local or broader community. Therefore I do not consider this is relevant to the overall social impact on those communities.

Attracting more shoppers to the area

1. It was stated by the applicant that the store would have the potential to enhance nearby entertainment and socialising by allowing residents and visitors to purchase their alcohol nearby as part of those activities. The respondent agreed and I find that such a benefit is likely and that this would result in a positive social impact in the local and broader communities.

Negative social impacts

1. The main negative social impacts identified in the evidence before the Tribunal were: 1. An increase in the levels of alcohol-related crime or disturbance in the local and broader communities. 2. An increase in alcohol consumption generally with related health impacts. 3. Impacts on amenity.

Police evidence

1. The Eastern Suburbs Police Area Command objected to the application as they hold the view that the consumption of packaged liquor in public places and private residences significantly contributes to harms within the community.
2. The submission stated that in 2016-17 the place of last consumption of alcohol was either a public place or a private residence in some 53% of all policing interventions in Bondi where alcohol was a factor. In the view of NSW Police this indicated that packaged liquor outlets contributed to this.
3. Sgt Peter Bolt of Rose Bay Police Station gave evidence for the respondent. He has been a police officer for 28 years and in the Eastern Suburbs since 1995. Since 2002 he has been the Licensing Supervisor for Eastern Suburbs Police Area Command (PAC). Sgt Bolt expressed concern about the incidence of alcohol related crime in the local and broader community including domestic assault, other assaults or violence, drinking in public areas, street offences and minors gaining access to alcohol.
4. A report on alcohol-related crime in the PAC from 1 September 2017 to 31 August 2018 was tendered. The report contains data from incidents on the COPS system which had "alcohol-related" recorded as an associated factor.
5. The report recorded that of 231 domestic violence offences in the PAC, 98 were alcohol-related. 16 or 16% of these occurred in Bondi which gave it the highest rate of alcohol related domestic violence in the area, followed by Bondi Junction ( 15 or 15%) Rose Bay (11 or 11%) and Vaucluse (11 or 11%). In those domestic violence incidents where the last place where alcohol was consumed was recorded, in 67% it was consumed at a private residence, in 15% it was consumed at a licensed premise and in 9% it was consumed in a public place. It should be noted that apart from occasional tastings, alcohol cannot be consumed at a packaged liquor outlet.
6. Bondi was ranked 5th in the PAC area for non-domestic alcohol-related assaults with 6% or 15 of the total 508 incidents for the PAC. Most such assaults occurred at Bondi Beach (91 or 37%) followed by Bondi Junction (46 or 19%). Where the last place alcohol was consumed was recorded, it was at licenced premises in 104 or 43% of the cases, at a private residence in 54 or 22% and a public place in 24 or 10%.
7. Bondi had 15 of the intoxicated person incidents in the PAC out of a total of 196. In those cases where the last place of consumption was recorded, it was a licenced premises in 18% of the cases, a private residence in 16% and a public place in 16%.
8. Bondi was ranked 9th for alcohol related street offences (2 of a total of 155). In those cases where the last place of consumption was recorded, it was a licenced premises in 37% of the cases, a private residence in 8% and a public place in 21%.
9. Bondi was ranked 8th for alcohol-related offences involving minors (21 or 4% of 556). The most alcohol-related offences involving minors occurred at Bondi Junction (134 or 24%) and Bondi Beach (111 or 20%). Where the last place of consumption of alcohol was recorded as being within the PAC, this was a public place in 43% of the cases, a private residence in 17% and a licenced premises in less than 1%.
10. Under cross examination, Sgt Bolt agreed that in two of the 16 domestic violence assaults in Bondi the incidents should not have been included as a domestic violence assault, in one case because it was rejected and in the other case because it should have been classified as common assault.
11. In relation to the intoxicated persons incidents, under cross examination Sgt Bolt agreed that in three of the incidents the offence was committed outside Bondi and they should not have been included in the Bondi figures. One of the street offences also occurred outside Bondi. Of the offences involving minors, he agreed that five were not an offence, two occurred outside Bondi and one did not mention alcohol. Four COPS reports in this category were not available so could not be examined.
12. The records also showed that in some cases the victim was drinking but the offender was not, or was not recorded as drinking. The applicant suggested that these incidents should be distinguished in determining negative social impact. However, the report deals with offences which are alcohol-related, not necessarily caused by alcohol. In other cases the person who had drunk alcohol had drunk it outside the PAC or it was not known where they drank it. I have not distinguished these incidents.
13. Sgt Bolt stated that if a person appeared to police to be affected by alcohol or said they had consumed alcohol the matter was recorded as alcohol-related. He conceded that the COPS entries made by police who attended the incident sometimes contained errors in recording details related to alcohol.
14. He said drinking alcohol in an alcohol free zone was no longer an offence but the alcohol could be seized or tipped out. This was not endorsed on the COPS system but his evidence was that it occurred frequently. No formal records were kept.
15. I am satisfied that the COPS figures for the following alcohol-related offences in Bondi in the relevant period should be revised as follows: Category of alcohol-related offence in Bondi Total Domestic violence offence 14 Other violence offence 13 Intoxicated persons offence 13 Street offences 2 Offences involving minors 12

1. On these revised figures, Bondi no longer had the greatest number of alcohol-related domestic violence offences in the PAC, as 15 such offences were recorded for Bondi Junction. Of course it is not known if any figures for the other suburbs could also be adjusted if they were subject to a similar examination.
2. It is notable that in the vast majority (67%) of the domestic violence offences across the PAC the last place of consumption was in a private residence, which indicates that packaged liquor had a role. In the other categories of offence, licenced premises were always the most common place of consumption except in the case of alcohol-related offences involving minors, when it was a public place. Packaged alcohol would have had a role in drinking in a public place also.

BOCSAR statistics

1. Bureau of Crime Statistics and Research (BOCSAR) statistics for the Waverley and Bondi areas are a source of information on the levels of crime and alcohol-related crime in the local and broader community over time and how they compare to the State average. BOCSAR data is drawn from the COPS data. The applicant and respondent both relied on BOCSAR data.
2. In 2017-18 the rate of alcohol-related assaults of all kinds in Bondi was 236.9 per 100,000 population as opposed to the NSW rate of 256.4 and marked as "stable".
3. In 2017-18 the rate of alcohol-related domestic violence assault in Bondi was 118.5 whereas NSW rate was 114.0. Alcohol-related non domestic violence assault was 109.4 compared to 127.3 for NSW.
4. Over the period between April 2013 to March 2018, alcohol-related domestic assault was recorded as follows (the number of incidents followed by the rate per 100,000 in parentheses): Year Bondi Waverley LGA NSW 2014 16 (149.1) 60 (85.3) 9977 (134.1) 2015 17 (157.1) 66 (92.9) 9454 (125.3) 2016 13 (119.3) 97 (135.5) 9054 (118.2) 2017 16 (145.8) 63 (87.4) 8857 (114.4)

1. The trend in NSW was downwards by approximately 7% over this period, however the figures for Bondi were stated not to reflect this trend.
2. The figures for 2017-18 are discussed in the evidence of Sgt Bolt above. Based on the evidence above, the actual figures of alcohol-related domestic violence in Bondi may be questioned in relation to 2014-2018 also, but in terms of the rate per 100,000 it can be seen that they were consistently higher than the Waverley rate and the NSW rate except for 2016 when the Waverley figures exceeded those of Bondi.
3. In comparison, the rate of domestic assaults overall (both alcohol-related and not) in Bondi in 2017-18 was substantially less than the NSW rate - 227.8 per 100,000 as opposed to 370 per 100,000 in NSW. This trend was noted by BOCSAR as stable over 2 years.
4. Crime "hotspot" maps from BOCSAR for 2017-2018 show that: 1. Bondi is adjacent to high level hotspots for alcohol related assault and assault overall in Bondi Beach and Bondi Junction. 2. Large high and moderate level hotspots for domestic assault in Bondi Beach and Bondi Junction extend into Bondi. 3. Bondi is within a large hotspot for high levels of malicious damage to property.

Evidence of Mr Paroz

1. Mr Patrick Paroz was an expert witness for the respondent. He was the Commander of Drug and Alcohol Coordination for NSW Police between 2010 and 2015 and has 33 years of policing experience.
2. Mr Paroz observed the site of the proposed licence and the vicinity on 11 February, 27 May and 16 September 2017, all of which were Saturdays and the following mornings. He also commented on the expert evidence of Mr Whetton and Sgt Bolt.
3. In Mr Paroz's view, while it was clear from the BOCSAR data that Bondi Beach and Bondi Junction had high levels of alcohol-related crime and were crime "hotspots" there is no evidence that Bondi Rd is a "crime corridor". Most people moved via private vehicle, bus or taxi. During his observations he did not see any persons carrying packaged liquor down Bondi Rd towards the beach. He did not believe that availability of alcohol in Bondi would affect what happens in Bondi Beach or Bondi Junction, given the much greater availability and choice of liquor and entertainment in those precincts. However he agreed the Royal Hotel was a convenient place to purchase alcohol if walking to the beach and that is just across the street from the site. There was no evidence that street drinking was a concern in Bondi.
4. On the issue of whether people such as transient tourists might buy alcohol from the outlet to drink unlawfully in beachfront or park areas, Mr Paroz found this unlikely. He noted that the outlet would be over 1 km from Marks Park and 800 metres from the beach front. There were three other packaged liquor outlets on Campbell Parade fronting the beach as well as two hotels which can sell packaged liquor. In his opinion it was more likely that people wanting to drink in the park or the beach would purchase their alcohol from those outlets. Parking in Bondi is extremely limited. On his visits to Bondi he did not see evidence of pre-fuelling, drinking in public, or alcohol-related litter either in the evenings or early mornings. He saw relatively few pedestrians on Bondi Rd.
5. He noted that there are no alcohol free zones in Bondi. Therefore tipouts or confiscations would not occur in Bondi Rd.
6. He noted that an unrestricted packaged liquor licence had not been granted in Bondi for 57 years.

Evidence of Mr Whetton

1. Mr Steve Whetton is an economist and the Deputy Director of the South Australian Centre for Economic Studies, with experience in economic modelling, health and environmental economics. He has conducted research and public policy evaluations in the cost and social impacts of alcohol of Western Australia and the Northern Territory. He provided a report reviewing the epidemiological evidence regarding the relationship between liquor outlet density and harms, with particular reference to the specific characteristics of Bondi.
2. The key studies which Mr Whetton drew from did not focus on areas within the Waverley LGA. They included: 1. Michael Livingston, "Alcohol outlet density and harm: Comparing the impacts on violence and chronic harms": Drug and Alcohol Review, (Sept 2011) 30, 515-523. This study examined trends in hospital admission data in Melbourne for assault and for alcohol use disorders over a 14 year period and their relationship with the density of alcohol outlets. The results suggested that the density of packaged liquor outlets is related to the rate of alcohol use disorders. The author considers that hospital admissions are less biased than police data. Each additional packaged liquor licence was associated with an increase in 0.538 in assault hospitalisations per 1000 persons. A 10% increase in packaged liquor licences would increase assault rates by 0.8%. An increase of 1 packaged liquor outlet per 1000 persons increased chronic alcohol disease hospitalisations by 1.9%. 1. Neil Donnelly, Patricia Menendez, and Nicole Mahoney, "The effect of liquor licence concentrations in local areas on rates of assault in New South Wales," Crime and Justice Bulletin, (2014), Number 181, NSW Bureau of Crime Statistics and Research. The authors analysed police, liquor licensing and socio-demographic data in New South Wales in one year and found their results confirmed that there is a relationship between the concentration of licensed premises in a particular area and levels of assault. There was a sharp increase in domestic violence assault rates and an increase in other assault rates once packaged liquor licence concentrations exceeded 0.75 per 1000 residents. Even though the magnitude of these effects will vary across geographical areas with different population characteristics (e.g. socioeconomic disadvantage, age distribution), the form of this relationship will remain.
3. Mr Whetton's opinion was that the relevant literature supported a conclusion that any net increase in the density of liquor outlets in a region will increase the alcohol-related harms in that region, including crime. Significantly, in his view, the estimates of this increase were not dependent upon an increase in overall expenditure on alcohol in the area as a result of the increase in outlet density. He said: "…it is highly like that any net increase in the number of liquor outlets in a region will increase alcohol related harms in that region, with the greatest increase of harms arising from an additional hotel, but packaged liquor outlets and nightclubs resulting in almost as large an increase in harms. The regulatory decision then becomes not whether there will be any increase in social harm as a result of an additional outlet, but rather how significant that harm is likely to be given the characteristics of the local community in question (with particular reference to its existing outlet density) and the proposed venue, and whether the expected benefits for the local community are likely to offset the scale of the harms."
4. The density of packaged liquor outlets had a stronger correlation with domestic violence and chronic health conditions than compared to hotel licence density.
5. The limitations on the literature identified by Mr Whetton are: 1. Almost all studies do not control for the specific characteristics of the premises (such as size, kind of liquor sold etc) which might affect the degree of harm. Where this has been controlled for, volume of alcohol sold is more predictive of harm than number of outlets, although the number still influences the level of harm. 2. The studies do not attribute causation, but two longitudinal studies found that increased outlet density in a specific region led to an increase in harm over time, suggesting that the density was a causative factor.
6. Mr Whetton also provided a cost benefit analysis of the crime and health impact, but overall I did not find this of assistance in determining social impact, as there was no evidence that the costs would necessarily be borne by the local or broader community.
7. Mr Whetton commented that because there can be a lack of accuracy in the COPS data as to whether an incident was alcohol-related or not, it was more valid to look at assault generally in the local and broader communities and compare it to the average to determine impact. He noted that assault, and particularly domestic violence, is recognised as significantly under-reported and the ABS estimates that only 51.5% of assaults are reported to police.
8. In his opinion, the statistical analyses discussed above were robust enough to show average impacts which could be applied to the areas in question.
9. Other relevant studies referred to by Mr Whetton and to which I was referred by the respondent included: 1. E. Lensvelt, W. Gilmore, W. Liang, A. Sherk, and T. Chikritzhs, "Estimated Alcohol-related deaths and hospitalisations in Australia 2004 to 2015", National Alcohol Indicators, Bulletin 16. Perth: National Drug Research Institute, Curtin University, 2018. Nationally and in NSW trends in alcohol-attributable deaths appeared to have remained stable or slightly decreased over time between 2003 and 2015 while alcohol related hospitalisations increased marginally. 1. M. Livingston, C. Wilkinson, and R. Room, Community Impact of liquor licences. An evidence check rapid review brokered by the Sax Institute for NSW Ministry of Health, (2015). This literature review concluded that there is high quality, locally relevant evidence that alcohol outlet density is associated with violence. Most studies found that changes in alcohol outlet density are associated with changes in the rate of health problems. There are weaknesses in the outlet density literature and it was noted that effects attributed to the density of packaged liquor outlets may be mediated entirely by the volume of alcohol sold through packaged liquor outlets. 1. Australian Bureau of Statistics, Crime Victimisation, Australia, 2013-14. (2015) Cat no. 4530.0 This contains findings from the Australia Bureau of Statistics 2013-14 national Crime Victimisation Survey. It found that nationally, 62% of persons who experienced physical assault believed that alcohol or another substance contributed to their most recent incident.
10. Using one approach put forward by Mr Whetton (based on the study by Livingston) a 10% increase in packaged liquor outlet density leads to a 0.8 % increase in assault hospitalisations. As the density of packaged liquor outlets (excluding the online retailers) would increase by 33.3% if the licence were approved, this implies an expected 2.7 % increase in hospitalisations for assault if the impact is line with the average.
11. Alternatively, based on the Donnelly, Menendez and Mahoney model which includes all packaged liquor licences in Bondi an increase in such licences from six to seven would bring the density of outlets from 0.597 to 0.697 per 1000 persons which is close to the point where the impact on domestic assaults and assaults is starting to increase after reaching a turning point, as illustrated in the figures below. According to Mr Whetton the expected number of reported domestic assaults would increase by 1.5 cases and non-domestic assaults would increase by 2.6 cases.

Evidence of Mr Smith

1. Mr George Smith is a qualified town planner who was responsible for preparing the Community Impact Statement for the applicant. He is a consultant to Design Collaborative, a planning and development consultancy. He prepared the applicant's first application to the ILGA for the site and has prepared many community impact statements under the Liquor Act.
2. Mr Smith overall said the vast majority of packaged liquor licences do not result in an increase in assaults and there was very little evidence that assaults occurred in the vicinity of packaged liquor outlets. He did not see any correlation between convenience, low price and assault rates.
3. Regarding density of outlets, he calculated that if the online outlets were excluded the packaged liquor outlet figures for Bondi are 0.299 before and 0.398 after the grant of the licence, which in his view meant that Bondi is still moving from below the State median to above the State median.
4. His opinion was that there was very little evidence that assaults occurred in the vicinity of packaged liquor licence outlets and that assaults associated with such outlets were widespread, while assaults associated with hotels are close by. He did not agree that crime hotspots were associated with the location of packaged liquor outlets. Social impact could only depend on an increase in sales of alcohol in his view and he did not foresee that the granting of the licence would lead to an increase in sales. In his view the store would take a share of the existing market.

Increase in alcohol consumption

1. Mr Duane's expert opinion was the licence approval would make no appreciable difference to the overall level of liquor consumption by local or broader community residents, given the existing number of outlets in the area. He projected that the new store would achieve sales of about $1.2 million and the net growth in the volume of packaged liquor sales as a result of the proposed store would be "negligible, and most likely zero." Of this he estimated approximately 67% would come from "cannibalising" the sales of existing packaged liquor outlets in the local community, and 25.5% from existing packaged liquor outlets in the broader community, with the remainder from beyond the broader community. He said that this view was drawn from data on the market, demographics, sales and alcohol consumption levels. He conceded that consumption might marginally increase for foot traffic but not much in the immediate area. He agreed there was underreporting of alcohol consumption. His estimate was based on an assumption that the store would focus on mid to high level wines, but there was no limit on what they would sell.
2. Mr Whetton agreed with Mr Duane's estimate of sales at $1.2 million and agreed that at least 50% of the sales made by the proposed store would come from "cannibalising" existing outlets. However he questioned Mr Duane's decision not to adjust his model to allow for known household underestimation of alcohol spending and consumption, which studies indicate may be up to 60%. There was no independent data on which to base the assessment of a no net increase in spending in the local community.
3. Mr Whetton and Mr Duane agreed that in NSW it cannot be conclusively established whether sales in a particular geographic area have increased after a licence has been issued, because wholesalers or retailers are not required to record or report liquor sales data.

Influence on pricing

1. Mr Duane's evidence was that it would be difficult for a small liquor store to influence pricing within the local and broader community. There were already retailers offering low prices. If savings were made there was no data to indicate that they would be directed towards purchasing more alcohol. While liquor availability has been increasing, per capita alcohol consumption has remained flat or declined.
2. The respondent submitted that granting the licence may drive down prices at the Royal Hotel's bottle shop as it was so close to the proposed site. The submission by the hotel owner indicated it believed the licence would affect its business.
3. In my view there was insufficient evidence to determine whether the proposed store would have any influence on pricing in the local or broader community.

Health impacts

1. Among the data considered by the ILGA was alcohol-related deaths and hospitalisations data from Department of Health Healthstats figures for 2013. This showed that the number of alcohol-attributable deaths in the Waverley LGA had declined from 11.2 per year (17.1 per 100,000 population) in 2001-2 to 8.6 per year (13.7 per 100,000 population) in 2012-13. On the other hand, the number of alcohol attributable hospitalisations were at 121.50 compared to the NSW average of 100.
2. Mr Whetton's evidence drawn from the research was that an increase in density of liquor outlets has a positive correlation with increase in rates of alcohol related health conditions and alcohol-related presentations to emergency departments (including injuries).
3. He concluded that for an increase in packaged liquor outlet density of 1 per 1000 residents, the rate of hospitalisations for chronic alcohol related disease will increase by 0.84 per 1000 residents and the rate of hospitalisations for alcohol-related assault will increase by 0.21 per 1000 residents.
4. Using another preferred approach led to the conclusion that a 10% increase in packaged liquor outlet density leads to a 1.9 % increase in the rates of chronic alcohol-related hospitalisations. As outlet density would increase by 33.3% if the licence were approved, this implies a 6.33% increase in hospitalisations if the impact is line with the average.
5. There was no submission from NSW Health to the ILGA. While Roads and Maritime Services provided a short submission concerning the role of alcohol impairment in vehicle crashes, it requested the imposition of conditions if the licence were granted; therefore I have not considered this impact as a negative impact in making my decision.
6. The Sax Institute paper concluded that most studies, including two longitudinal Australian studies, showed that increase in alcohol outlet density was associated with poorer health outcomes. These include increased rates of alcohol-related chronic illnesses, and increased alcohol-related presentations to emergency departments. [4]

The overall social impact of the licence if granted

Applicant's submissions

1. The applicant submitted that Bondi has a low packaged liquor licence density compared to the State. No packaged liquor licence has been granted in Bondi in almost 60 years. The population had grown but the local supply of alcohol had not. The rate of population growth in Bondi, however, is not high - it is only projected to grow by 600 by 2031.
2. The rate of alcohol-related crime in Bondi was significantly lower than the rest of the State except alcohol-related domestic violence. The interrogation of the COPS data showed that two of the alcohol-related domestic violence incidents should not have been counted. None of the youth crime matters in Sgt Bolt's report should have been counted, and each area identified for unauthorised drinking had outlets closer to them than the proposed site. The evidence does not show that alcohol related offences are clustered around packaged liquor outlets. The applicant was willing to accept agreed licence conditions dealing with harm minimisation, and there were harm minimisation conditions imposed by Council on the development consent. The risk of harm was stated to be lower for a small store such as this which mainly sold wine, than a larger store selling a wide range of spirits and beer.
3. The applicant submitted that it did not regularly sell cheap products, particularly cheap beer, but had specials from time to time. Customers seeking cheaper alcohol were more likely to go to other retailers. The Royal Hotel and Kemenys were more likely to discount their products.

The respondent's submissions

1. The respondent submitted that I should include the online liquor retailers in assessing density. It agreed that the outlet density was below the State average. It submitted that the Tribunal should consider both the research findings as to the wider background and the local conditions. Outlet density should be treated with care because it did not distinguish between types of outlet, size and product sold. In terms of sales, while the granting the licence may not lead to greater sales in the wider community, there was concern that sales would increase in the local community both among residents and visitors. The concern was that there would be a motivation to sell cheap wine, bear and spirits. The respondent submitted that the higher spend on beer than wine in the local community was a relevant factor.
2. The ILGA submitted that in making its decision the Tribunal must consider the "one-sided" nature of the proceedings as the ILGA does not present the case of the objectors to the proposal. The decision should be based on verifiable data and be consistent.
3. The fact that the alcohol-related health impacts are above average is a relevant consideration. The residents' views on the impacts on amenity are more reliable than those of Mr Paroz. The respondent conceded that the COPS data was not totally reliable but it will never be possible to know all the facts about an incident. The rates for alcohol-related domestic violence in Bondi are still high.

Licence conditions

1. It was drawn to my attention by the applicant that the ILGA had recently approved the granting of a packaged liquor licence in Coogee, which is in the nearby Randwick LGA. That licence had conditions that no alcohol other than wine may be displayed in the basement level of the licensed premises. [5]
2. The respondent also submitted that in another licensing decision a retailer was required to limit the beer it offered to craft beer products. The reasons for these conditions were not immediately apparent.
3. If the Tribunal decides to grant a licence, it also has the power to impose conditions on the licence. The respondent and applicant submitted that if I proposed to grant the licence with conditions, the parties should be notified so that they could make submissions on this issue.

Findings

1. I may not grant the application if I am not satisfied that the overall social impact of granting the licence will not be detrimental to the well-being of the suburb of Bondi or the Waverley Local Government Area.
2. As stated in the Deputy President's decision in Smith, it is necessary to identify objective facts from which conclusions may be drawn as to causation of social impacts. The relevant objective facts which I have found to be established are set out below. 1. At least 50% of the sales to be made by the proposed store would be drawn from existing sales of other outlets in the local or broader community. 2. Notwithstanding this, there is robust evidence from the research literature cited by Mr Whetton that an increase in density of packaged liquor outlets in an area is associated with an increase in crime and health detriment. I find that the studies by Livingston and Donnelly provide cogent evidence of this relationship and those findings can be applied to Bondi and the Waverley LGA. 3. Convenience for commuters, local residents and visitors in the local area seeking to purchase alcohol would be increased if the licence were granted. 4. The addition of a packaged liquor outlet in this location will offer greater choice in the local community. 5. I am not persuaded that the granting of the licence with the conditions imposed under the development consent would have a negative impact upon the amenity of the wider community in terms of noise, litter, loitering or other impact. 6. The granting of the licence would attract more shoppers to the area and the nearby businesses, which would be a positive social impact on the local and broader communities. 7. It is possible that there will be some additional inconvenience to local residents from visitors in vehicles, however the evidence of the lower rate of vehicle ownership in the local area, the availability of bus transport and the lack of parking opportunities indicates this will be minimal. 8. The rate of domestic violence per 100,000 population where alcohol was a factor has been higher in Bondi than in NSW in three out of four years between 2014 and 2017. Even allowing for the inaccuracy in the most recent figures as established by the applicant, these figures cause some concern. 9. Bondi's ranking within the Waverley LGA for other offence categories in Bondi where alcohol was a factor was relatively low. The rate of all alcohol related assaults in Bondi was 236.9 per 100,000 population as opposed to 256.4 for NSW. 10. Assault and domestic violence are under-reported however. Only 51.5% of assaults are reported, according to Australian Bureau of Statistics estimates, and the COPS figures are unlikely to reflect all incidents. 11. While Bondi and Bondi Rd are adjacent to other suburbs in the Waverley LGA which have high levels of alcohol-related assault I am not persuaded that the proposed outlet in this location will substantially impact upon those areas which are well supplied with liquor outlets of all kinds. I consider that the persons who would most be impacted by the granting of the licence (both positively and negatively) will be local residents. It might however attract some persons seeking to drink in public places as it is within a short walking distance of two parks. 12. Expenditure on beer in Bondi is slightly higher than that in the Waverley LGA. Expenditure on wine in Bondi and the Waverley LGA is significantly higher than that of the Sydney Metropolitan area. Expenditure on spirits was very similar between the local and broader communities and somewhat higher than the Sydney Metropolitan area. It is not clear however whether these ratios reflect the relative volume or the relative price of alcohol purchased in Bondi. 13. Packaged liquor outlets have a greater association with alcohol-related domestic violence than other licenced premises. There is a sharp increase in domestic violence assault rates and an increase in other assault rates once packaged liquor licence concentrations exceeded 0.75 per 1000 residents. 14. If the licence were approved the packaged liquor outlet density would increase from 0.597 to 0.697 per 1000 persons and this brings it into the area where the impact on domestic assault rates increases, according to the Donnelly model. 15. Other research findings based on Livingston suggest a 2.7 % increase in hospitalisations for all kinds of assault and a 6.33% increase in chronic alcohol hospitalisations could be expected in Bondi if the licence were granted.

Conclusion

1. The figures identified above lead me to conclude on the balance of probabilities that granting the licence will increase the rate of domestic assault, other assault and chronic illness in Bondi and Waverley, particularly in Bondi, in the above ranges. In determining the overall social impact, I must decide whether the identified detriment to the well-being of the Bondi and Waverley communities outweighs the identified benefits.
2. The impact of domestic assault, other assault and chronic illness in a community extends beyond the persons immediately impacted. The benefits of granting the licence, in my view, are fairly limited in scope in terms of economic impact, numbers of people affected and geographical area. I am not satisfied that the overall social impact of granting the licence would not be detrimental to the well-being of the local or broader community.
3. On the available evidence I am not satisfied that the identified detriments could be remedied by imposing licence conditions.
4. Accordingly the decision to refuse the licence should be affirmed.

Order 1. The decision under review is affirmed.

I hereby certify that this is a true and accurate record of the reasons for decision of the Civil and Administrative Tribunal of New South Wales. Registrar

Endnotes

1. Jacqueline Peel, "Precaution - A matter of principle, approach or process?" Melbourne Journal of International Law, (October 2004), Vol. 5, No. 2, 483-501 at 487.
2. For example Neil Donnelly, Patricia Menendez, and Nicole Mahoney, "The effect of liquor licence concentrations in local areas on rates of assault in New South Wales," Crime and Justice Bulletin, (2014), Number 181, NSW Bureau of Crime Statistics and Research.
3. See for example M Livingston, "Alcohol outlet density and harm: comparing the impacts on violence and chronic harms", Drug and Alcohol Review, (September 2011), 30, at 517.
4. Livingston et al, Community Impact of Liquor Licences, p.6.
5. Decision of the Independent Liquor and Gaming Authority Application for a packaged liquor licence - George's Cellar & Store 20 September 2018. I hereby certify that this is a true and accurate record of the reasons for decision of the Civil and Administrative Tribunal of New South Wales. Registrar DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated. Decision last updated: 11 March 2019