Chief Commissioner of State Revenue v Godolphin Australia Pty Ltd

14] FCAFC 45 Minister for Immigration and Border Protection v SZVFW (2018) 264 CLR 541; [2018] HCA 30 New South Wales Aboriginal Land Council v Minister Administering Crown Lands Act (2016) 260 CLR 232; [2016] HCA 50 News Ltd v South Sydney District Rugby League Football Club Ltd (2003) 215 CLR 563; [2003] HCA 45 People for the Plains Inc v Santos NSW (Eastern) Pty Ltd (2017) 220 LGERA 181; [2017] NSWCA 46 Rich v Australian Securities and Investments Commission (2004) 220 CLR 129; [2004] HCA 42 Robinson Helicopter Company Incorporated v McDermott [2016] HCA 22; (2016) 90 ALJR 679 Royal Agricultural Society of New South Wales v Sydney City Council (1987) 61 LGRA 305 Thomason v Chief Executive, Dept of Lands (1995) 15 QLCR 286 Tweddle v Federal Commissioner of Taxation (1942) 180 CLR 1; [1942] HCA 40 Victoria International Container Terminal Ltd v Lunt (2021) 271 CLR 132; [2021] HCA 11 Warren v Coombes (1979) 142 CLR 531; [1979] HCA 9 Woollahra Municipal Council v Banool Developments Pty Ltd (1973) 129 CLR 138; [1973] HCA 65 Young v Chief Commissioner of State Revenue [2020] NSWSC 330 Texts Cited: Second Reading of the State Revenue Legislation Further Amendment Bill (NSW Legislative Council, Parliamentary Debates (Hansard), 29 November 2005 at 20063-20064 Category: Principal judgment Parties: Chief Commissioner of State Revenue (Appellant)

Godolphin Australia Pty Ltd (Respondent) Representation: Counsel:

T L Wong SC and S Kanagaratnam (Appellant) D Russell KC, J S Emmett SC and Ms D Levi (Respondent)

Crown Solicitor's Office (Appellant) Johnson, Winter & Slattery (Respondent) File Number(s): 2022/00127231 Decision under appeal Court or tribunal: Supreme Court Jurisdiction: Equity Citation: Godolphin Australia Pty Ltd v Chief Commissioner of State Revenue [2022] NSWSC 430 Date of Decision: 13 April 2022 Before: Ward CJ in Eq File Number(s): 2019/00401271

[Note: The Uniform Civil Procedure Rules 2005 provide (Rule 36.11) that unless the Court otherwise orders, a judgment or order is taken to be entered when it is recorded in the Court's computerised court record system. Setting aside and variation of judgments or orders is dealt with by Rules 36.15, 36.16, 36.17 and 36.18. Parties should in particular note the time limit of fourteen days in Rule 36.16.]

HEADNOTE [This headnote is not to be read as part of the judgment] The respondent, Godolphin Australia Pty Ltd (Godolphin), runs a business involving the breeding, sale and racing of thoroughbred racehorses. It applied to the Supreme Court for a review of decisions made by the appellant, the Chief Commissioner of State Revenue (the Commissioner), who refused to apply the exemption for land used for primary production when assessing the Godolphin's land tax with respect to two properties owned by it. The principal question was whether or not the dominant use of the lands in the years 2014 to 2019 could be characterised as for the maintenance of animals for the purpose of selling them, their natural increase (ie progeny) or their bodily produce under s 10AA(3)(b) of the Land Tax Management Act 1956 (NSW). The amount at issue is some $1.438 million. In the proceedings below, the Commissioner sought to distinguish the uses of the land for racing on the one hand, and breeding and stud activities (including servicing of mares for a fee) on the other. In his view the dominant use was the former, or at least not the latter. The properties thus did not fall within the exemption provided by s 10AA(3)(b). Godolphin submitted that in its business model for the relevant land racing and the sale of breeding services, broodmares and progeny "are complementary or mutually reinforcing", and that in its distinctive business model "the purpose of sale and the purpose of racing are two aspects of a single composite purpose". The primary judge, Ward CJ in Eq, upheld Godolphin's objection and revoked the land tax assessments for the years in question. Her Honour found that the two properties were used as part of an integrated operation in which the preparation of horses for racing is with the overall or dominant purpose or objective of increasing or maximising the revenue from the nomination fees (ie the sale of bodily produce) and from the sale of the progeny produced by the broodmares. The Commissioner appealed from that decision. There were two principal issues on appeal. The first concerned the construction of s 10AA(3)(b) and the interaction between the concepts of "use" and "purpose". By a notice of contention Goldolphin asserted that the requirement of dominance attached only to the use made of the land, in the sense of the physical activities undertaken, and did not also attach to the purpose of the use, such that the sale did not have to be the dominant purpose. The second issue was the application of the test in s 10AA(3)(b) to the facts of the case. The Court (per Kirk JA and Simpson AJA, Griffiths AJA dissenting) upheld the appeal, and held as follows: As to the notice of contention on the construction of s 10AA(3)(b) of the Act: Per Kirk JA: Uses and purposes with respect to land are concepts that are distinct but commonly linked. Use is what is done on the land. As the concept of use involves human activity it will be undertaken for some purpose or purposes. The purpose is why (ie to what end) those things are done. Thus the question here is not simply whether the use of maintenance of animals - which both sides accept to be the dominant use of the properties - can then be characterised as for a purpose of sale. Rather, the question is whether that use of the properties can be characterised as having the character of a dominant use for the purpose of selling animals, progeny and produce (ie, here, the sales purpose in the sense identified): at [27]-[32] Chief Commissioner of State Revenue v Metricon Qld Pty Ltd (2017) 224 LGERA 236; [2017] NSWCA 11, Leda Manorstead Pty Ltd v Chief Cmr of State Revenue (2011) 85 ATR 775; [2011] NSWCA, Thomason v Chief Executive, Dept of Lands (1995) 15 QLCR 286, Baulkham Hills Shire Council v O'Donnell (1990) 69 LGRA 404, considered. Per Simpson AJA: In previous decisions under s 10AA(3)(b) this Court has steered away from separating the concepts of "use" and "purpose". It appears also to have been accepted that (to the extent that it is appropriate to separate the two) the requirement of dominance applies to both use and purpose: at [132]. To construe s 10AA(3)(b) in the manner contended for by Godolphin would render any purpose, no matter how insignificant in the overall use of the land, sufficient to attract the exemption provided by s 10AA(3), which is not indicated by either the terms of the provision or by its statutory purpose: at [159]. Leda Manorstead Pty Ltd v Chief Commissioner of State Revenue [2011] NSWCA 366, Chief Commissioner of State Revenue v Metricon Qld Pty Ltd [2017] NSWCA 11, Young v Chief Commissioner of State Revenue [2020] NSWSC 330, followed. As to whether the dominant use was for the sale purpose: Per Kirk JA: Godolphin established that a significant use of the two properties was animal maintenance for the purpose of selling animal produce and progeny. But it did not establish that that should be characterised as the dominant use. A more significant use of the land was animal maintenance for the purpose of racing. True, each purpose aided the other to some extent, and the sales purpose is no doubt more profitable. But taking account of the activities undertaken at the land, the areas directed to those activities, the intensity of those activities and the resources directed to the relevant purposes, the racing purpose was not merely incidental and subservient to the sales purpose. The activities in pursuit of the racing purpose - not the sale purpose - constituted the dominant use of the land at Kelvinside and Woodlands in the relevant tax years: at [125]. Per Simpson AJA: a landowner seeking to establish that one use or purpose of the land is dominant over another or others is called upon to establish more than that two or more uses or purposes of the land are of equal or approximately equal importance. To avail itself of the exemption permitted by s 10AA(3)(b) Godolphin had to establish that the breeding purpose predominated over the racing purpose. It did not do so: at [154]. Per Griffiths AJA (dissenting): The approach applied by the primary judge was consistent with the approach taken in Thomason v Chief Executive, Dept of Lands (1995) 15 QLCR 286 (as approved in two previous decisions of the Court of Appeal): [223]. If one loses sight of the central and unusual feature of Godolphin's business operations on the relevant land, namely that they are related, integrated and symbiotic, the exercise to be undertaken in ascertaining whether s 10AA(3)(b) of the Act applies may miscarry: at [166], [197]-[198]. The primary judge correctly characterised Godolphin's business as an integrated operation in which the preparation of horses for racing was with the overall or dominant purpose or objective of increasing or maximising the revenue from the nomination fees (ie the sale of bodily produce) and from the sale of the progeny produced by the broodmares: at [223].

Judgment

1. KIRK JA: Godolphin Australia Pty Ltd, the respondent in this appeal (Godolphin), runs a business involving the breeding and racing of thoroughbred racehorses. Its strategic motto, at least in recent times, has been "breed to race; race to breed". The most profitable part of its activities involves the selling of "covering" services by stallions, that is, the service of Godolphin's stallions impregnating mares owned by others for a "nomination" fee.
2. Godolphin owns four properties used for racing and/or breeding purposes in New South Wales. Two are the subject of this case: Kelvinside and Woodlands. The other two are Osborne Park and Crown Lodge. The case concerns whether or not Kelvinside and Woodlands are exempt from land tax. The question depends on whether or not the dominant use of the lands in the years 2014 to 2019 can be characterised as for the maintenance of animals for the purpose of selling them, their natural increase (ie progeny) or their bodily produce: s 10AA(3)(b), Land Tax Management Act 1956 (NSW) (the Act).
3. The position of the Chief Commissioner of State Revenue, the appellant, is that the use of the two properties cannot be characterised in that way. He distinguishes the uses of the land for racing on the one hand, and breeding and stud (including servicing of mares for a fee) on the other. In his view the dominant use was the former, or at least not the latter. The properties thus did not fall within the exemption provided by s 10AA(3)(b). Godolphin exercised its right to apply to the Supreme Court for a review of the Chief Commissioner's decisions with respect to the 2014 to 2019 land tax years, pursuant to s 97(1) of the Taxation Administration Act 1996 (NSW). The amount at issue is some $1.438 million.
4. Godolphin in substance did not dispute that there were two purposes which could potentially be distinguished with respect to the activities carried on at the two properties, namely: 1. breeding, educating and training thoroughbred horses, and spelling them between races, with a view to achieving success in horse racing (the racing purpose) 2. the sale of covering services and of (some) progeny produced when bred from Godolphin's own mares (the sale purpose).
5. The evidence indicated that these two types of pursuits were commonly carried out separately by other, competing businesses in Australia. Godolphin's combined business model was unusual. Godolphin submitted that in its model the two purposes "are complementary or mutually reinforcing" and that in its distinctive business "the purpose of sale and the purpose of racing are two aspects of a single composite purpose".

The statutory context

1. Section 7 of the Act provides that land tax "is to be levied and paid on the taxable value of all land situated in New South Wales which is owned by taxpayers (other than land which is exempt from taxation under this Act)". The tax is levied on the land as owned at midnight on 31 December in the year preceding the year in which the tax is payable: s 8. The tax is payable by the owner by reference to the "taxable value of all the land owned by that owner which is not exempt from taxation": s 9(1). The "taxable value of that land is the total sum of the average value of each parcel of that land": s 9(2). The notion of "average value" involves in essence taking account of an average land value of the land in question over three years: ss 9(3) and 9AA. The land value of land is the value, as at 1 July in the previous year, entered in the Register of Land Values maintained by the Valuer-General under s 14CC of the Valuation of Land Act 1916 (NSW): ss 9(4) and 3(1) (see definition of "Register").
2. There is no definition of "parcel" in the Act, nor of "land". The latter word is defined in an inclusive manner in s 21(1) of the Interpretation Act 1987 (NSW), but not in a way that is relevant.
3. The effect of s 9 of the Act is that the Chief Commissioner must use land values as determined by the Valuer-General. Section 14A(1) of the Valuation of Land Act 1916 (NSW) requires the Valuer-General to determine the "land value of each parcel of land in New South Wales", with certain exceptions. Section 14A(4) provides that the Valuer-General "may separately value different parts of the same parcel of land, in which case this Act applies to each such part as if it were a separate parcel of land". There is no definition of "parcel" in this Act. Section 26(1) provides that where "several parcels of land adjoin, are owned by the same person, and where no part is leased, they shall be included in one valuation, unless the Valuer-General otherwise directs" (subject to one qualification).
4. The Chief Commissioner provided this Court, without objection from the respondent, a copy of a Valuer-General's policy entitled "Valuing separate parcels", dated 4 June 2019. That document states that "in general terms, a single parcel of land can be considered as land that is held in the one ownership and used for one purpose". It seems the Valuer-General's practice is, in general, to identify parcels of land without necessarily being limited by each particular lot in a deposited plan registered under the Real Property Act 1900 (NSW). Each parcel of land identified by the Valuer-General is allocated a "PID", that is, a parcel identification number.

Construction of s 10AA(3)(b)

1. From 20 May 2014 onwards s 10AA provided as follows: (1) Land that is rural land is exempt from taxation if it is land used for primary production. (2) Land that is not rural land is exempt from taxation if it is land used for primary production and that use of the land - (a) has a significant and substantial commercial purpose or character, and (b) is engaged in for the purpose of profit on a continuous or repetitive basis (whether or not a profit is actually made). (3) For the purposes of this section, land used for primary production means land the dominant use of which is for - (a) cultivation, for the purpose of selling the produce of the cultivation, or (b) the maintenance of animals (including birds), whether wild or domesticated, for the purpose of selling them or their natural increase or bodily produce, or (c) commercial fishing (including preparation for that fishing and the storage or preparation of fish or fishing gear) or the commercial farming of fish, molluscs, crustaceans or other aquatic animals, or (d) the keeping of bees, for the purpose of selling their honey, or (e) a commercial plant nursery, but not a nursery at which the principal cultivation is the maintenance of plants pending their sale to the general public, or (f) the propagation for sale of mushrooms, orchids or flowers. (4) For the purposes of this section, land is rural land if - (a) the land is zoned rural, rural residential, non-urban or large lot residential under a planning instrument, or (b) the land has another zoning under a planning instrument, and the zone is a type of rural zone under the standard instrument prescribed under section 33A (1) of the Environmental Planning and Assessment Act 1979, or (c) the land is not within a zone under a planning instrument but the Chief Commissioner is satisfied the land is rural land.
2. Prior to 20 May 2014 subsection (4) was in different terms. Nothing turns on this for current purposes. In this case there was no dispute that the two properties are "rural land" falling within s 10AA(1).
3. The dispute was whether the land was "used for primary production". That dispute turns on whether or not the properties fall within s 10AA(3)(b), that is, whether or not the "dominant use of [the land] is for … the maintenance of animals … for the purpose of selling them or their natural increase or bodily produce". There was disagreement about how this question was to be approached.

The reasoning of the primary judge

1. The primary judge's core conclusions in the case are captured in the following: [255] Mr Cox explains the "breed to race, race to breed" model as being that the breeding and supply of stock for the racing stable on the one hand is followed by the racing stable improving the stock and creating the supply of bloodstock assets. This is reinforced by the fact that stallion covering fees are clearly the key revenue driver for the business. … [258] I accept that much of the use of the Land is devoted to breeding, training and preparation of the horses for racing (and that an obvious objective is to maximise prize winnings or success on the track). However, viewed overall, I have concluded that this is an integrated operation in which the preparation of horses for racing is with the overall or dominant purpose or objective of increasing or maximising the revenue from the nomination fees (i.e., the sale of bodily produce) and from the sale of the progeny produced by the broodmares. [259] I do not accept that the integrated operation is for the dominant purpose of generating prizemoney (i.e., of racing). That does not make sense from an economic point of view. Thus (and assuming for the sake of argument for this purpose that one is looking for dominant purpose rather than dominant use - as the legislation provides) I have concluded that the document [sic - dominant] purpose of the stud operations is just that - to run a thoroughbred stud (with revenue generated from the stallions' covering services and, to a lesser degree, from the sale of progeny and any racing success). … [270] The significance of this conclusion is that, if there were two distinct purposes for the activities carried on at the land, then the relevant question would be whether use for any one such purpose was the dominant use. However, to my mind, the objectives of winning races and pursuit of stallion excellence are part of the overall objectives of increasing the value of Godolphin's stud operations (being nomination fees in particular and sale of the majority of Godolphin's progeny) at a time that does not compete with breeders and taking into account an assessment of racing potential but that the sales are sufficiently proximate - if that be a required element - to the maintenance of the animals on the Land. … [272] I consider that factors such as the uneconomic nature of the operations were they to be confined to racing alone; and the evidence that racing prowess (of the horse and of its progeny in due course) is an important factor in the pricing able to be commanded for the sale of stallions' semen, lead to the conclusion that this is indeed an integrated stud operation and that the dominant use of each of the parcels of land for each of the land tax years was for the maintenance of animals for the purpose of selling their bodily produce or natural increase.

The grounds of appeal

1. The Chief Commissioner's first ground of appeal is that the primary judge "failed to correctly apply the legal test called for by s 10AA(3)". Her Honour considered in effect that the purpose of racing - and all of the physical activities involved in achieving that purpose - was secondary and incidental to a dominant purpose of breeding. That is effectively what is conveyed by the language of "an integrated operation". That type of approach is consistent with the planning cases referred to above. Of itself, it does not manifest legal error.
2. The Chief Commissioner argues that her Honour focused excessively on "an economic point of view", had insufficient regard to the actual physical use to which the land in question was put, and erred in not "making findings" in relation to the sorts of matters identified by White J in Leppington Pastoral in the passage quoted above at [34].
3. As explained above at [35], economic factors have some relevance to assessing the purpose/s for which the physical activities are undertaken. The language of "integrated operation" does not suggest that her Honour lost sight of the existence of two possible purposes, one within the exemption and one outside of it. As regards taking account of other factors, in substance the Chief Commissioner's complaint is that her Honour reached the wrong result. It is appropriate to address that complaint in dealing with the other grounds of appeal.
4. The issue of whether or not her Honour's conclusion on the characterisation issue was correct is raised in various formulations by grounds 2-8 in the appeal. It is not necessary or efficient to deal with those grounds individually. Put simply, the issue raised by all of them is whether the dominant use of the properties in the relevant period was maintaining thoroughbred horses for the sale purpose (in the sense explained).
5. Both sides proceeded on the basis that this issue was to be determined on the "correctness" standard of appellate review on this appeal by way of rehearing, and that it was not necessary for the Chief Commissioner to make out error of House v The King kind: note discussion in Minister for Immigration and Border Protection v SZVFW (2018) 264 CLR 541; [2018] HCA 30 at [35]-[50] per Gageler J; cf Metricon at [77]. In other words, the parties proceeded on the basis that the question is whether this Court considers that her Honour's conclusion was in error. That is the approach I will adopt. It was also accepted by both sides that no issues of credibility or reliability arose in relation to any witnesses, and that this Court was in as good a position as the trial judge to address the issues raised.

The operation of Godolphin's business

1. Godolphin was previously known as Darley Australia Pty Ltd. It has been operating in Australia since 2001, initially focusing on breeding operations, in particular by offering stallion servicing but with some broodmares. In 2008 it acquired bloodstock and property assets from Ingham Bloodstocks, apparently including the properties in question. Since 1 August 2014 all horse racing undertaken in Australia by the respondent has been conducted under the Godolphin banner. That banner is associated with the founder of the respondent, Sheikh Mohammed bin Rashid Al Maktoum, who also has horse racing and breeding assets in other countries.
2. It was not in dispute that Godolphin's approach of engaging in both racing and breeding operations was somewhat unusual in this State. More usually businesses would do one or the other. Mr Vincent Cox, the managing director of Godolphin since January 2018, gave evidence describing Godolphin's operations. The primary judge regarded him as "quite candid in his evidence" (at [261]). He said the following: Godolphin's business model is a little different to most commercial thoroughbred breeders. Generally, thoroughbred breeders breed foals and sell them as yearlings (i.e. before they are ready to race). Value is assigned to these foals through an assessment of their appearance and their pedigree. The same businesses will often then purchase tried and proven thoroughbred horses for their breeding program at a significantly higher price than what it would cost to maintain thoroughbred horses that are bred for the same purpose. It costs approximately $50 per day to maintain a thoroughbred horse. Godolphin retains its own thoroughbred horses for its breeding operation. We operate a breed to race, race to breed model as opposed to a breed to market and purchase later model. Selling thoroughbred horses that have a successful racing history can increase the value of the thoroughbred horse (compared to its value as an unraced yearling) because it reduces the risk for the buyer and makes the horse a known commodity. The increase in the value of a thoroughbred horse with a successful racing history can be very significant. In the same way, the value of a thoroughbred horse with a poor racing history is significantly decreased.
3. Both sides took the Court to various iterations of Godolphin's "Flying Start" document, which was used as a presentation for trainees. Mr Cox indicated that this document represented a statement or overview of the business and its objectives prior to his own formulation of a strategy. The last iteration, from 2018, appears representative. It includes the following (underlining in the original): Godolphin Australia breeds to race … Breeding and Racing operation funded by: Commercial Stallion operation (anticipated mares to cover and estimated revenue from this) Prize-money Sale of surplus mares, weanlings and racehorses … COMMERCIAL OBJECTIVE The objective of Godolphin Australia, as an integrated thoroughbred breeding and racing operation, is to enhance the residual value of its bloodstock holdings by achieving success on the racetrack at the highest levels. The objective of racetrack success applies to Godolphin owned racehorses and Godolphin progeny that is owned by third parties. Continued racetrack success will enhance the Godolphin commercial programme and provide an opportunity to create an operation that is self sustainable. … Ingham Bloodstock acquired in 2008 - strategic play for prime bloodstock and property assets Merger Objective - to create Australia's leading thoroughbred operation by combining the history, success and experience of lngham's with ambition, resources and global presence of Sheikh Mohammed Major emphasis on racing operation - strategy shifts to breed to race …

The properties and the activities undertaken there

1. Woodlands is located at Jerrys Plains in the Hunter Valley, and is some 2420 hectares. Most of Godolphin's breeding mares are kept there. Godolphin's foals are born there. Yearlings are kept there for at least some of the time.
2. As at August 2020 - after the relevant period, but the evidence was given by way of illustration - there were some 213 horses on the property, including 92 yearlings, 80 mares in foal, 40 dry mares, 2 foals, and with many more foals on their way.
3. The number of thoroughbred horses "held and maintained" at Woodlands as at 31 December in the years 2013-2019 ranged between 316 and 551. These figures exclude "hacks" and "nanny mares" (mature animals used in educating young horses) along with "teasers" (mares used to arouse stallions for covering purposes). The equivalents numbers for Kelvinside were lower, being 27 to 103 thoroughbred horses.
4. Kelvinside is located near Aberdeen in the upper Hunter Valley. It is approximately 907 hectares. There are 20 stallion boxes there, located next to the stallion covering area. During the breeding season Godolphin generally has 10-14 stallions located on the property, all of which have completed their racing career. Also located on this property are areas and facilities used for educating the yearlings, and undertaking some preliminary training for racing. There is a practice racecourse on Kelvinside. More intensive training for racing commences when the horses are about 18 months old. Most of that training is done at the Crown Lodge property, although when being trained the horses are shuttled between that property and Kelvinside, with this occurring four or five times. The majority of that time is spent at Kelvinside, not at the (small) Crown Lodge property.
5. Horses that have commenced racing also spend significant time at Kelvinside. A series of paddocks on the property are used for "spelling" racehorses, that is, resting them after periods of racing. There is a large old homestead on the property which is used as offices.
6. Only about 10% of one of the four parcels at Kelvinside was focused upon housing the stallions and the covering areas.
7. Godolphin keeps cattle at both subject properties, for purposes which assist in the management of the properties and maintenance of the horses (for health reasons relating to parasites). Godolphin also grows lucerne in some paddocks at Woodlands. Neither side sought to place significant reliance on these activities.

Financial aspects of the business

1. Godolphin's accounts are prepared on a calendar year basis. Over the period 2013-2019 the relevant revenue received was as follows. These figures are taken from the affidavit of Mr Paul Hartmann, the Chief Financial Officer of Godolphin (there were other figures in various sets of accounts, which varied somewhat): 1. The prizemoney received from racing ranged between some $13.6 million to $34.3 million (that being an unusually good year), with an average of $19.6 million. These figures were for all Godolphin horses in Australia, but Mr Hartmann stated that the overwhelming majority of thoroughbreds owned by the company spent some time at one or other of the two subject properties. 2. The revenue received from stallion covering for stallions maintained at Kelvinside ranged from $14.0 million to $29.4 million, with an average of $22.1 million. 3. The revenue from the sale of thoroughbred horses (for all of Godolphin) ranged between $5.0 million and $19.5 million, with an average of $12.8 million. There was also revenue from sale of cattle, but, as noted, neither side sought to place any significant reliance on the role that cattle (or the growing of lucerne) played in operations at the two properties.
2. The primary judge stated at [255] that "stallion covering fees are clearly the key revenue driver for the business". Taken by themselves, however, those fees on average are not substantially higher than the prizemoney from racing (average servicing revenue being some 13% higher than average prizemoney). In two years - 2018 and 2019 - the prizemoney exceeded the revenue from covering fees.
3. Griffiths AJA suggests below at [210] that the average figure for prizemoney is distorted by inclusion of the "aberrational amount" of some $34 million in 2019. I respectfully disagree that high winnings in one year should be disregarded for an undertaking which involves as many contingencies as horseracing does, and where there is a relatively small sample group of seven years. In any event, if that year is excluded then the average prizemoney is $17.135 million. Based on that figure, average servicing revenue would be some 29% higher. That is a significant difference, but not such a dominating one as to suggest that covering fees are clearly the key revenue driver. Prizemoney is an important revenue stream for Godolphin.
4. Adding the average revenue from the sale of covering services and of horses comes to some $34.9 million. This contrasts with the average $19.6 million from racing prizemoney. But it would be too simplistic to conclude that all revenue from the sale of horses should be attributed to a sale purpose as opposed to the racing purpose. As noted above at [61], the horses that are sold tend to be those that are less successful in their first year of racing, and this reflects both racing and breeding imperatives. If the model was simply "breed to race", then the business would still no doubt sell off those parts of its stock that were not proving successful at racing. As Mr Cox said, "[s]ales are necessary to keep control of horse count and quality". The Flying Start document, quoted above at [51], refers to the breeding and racing operation being funded in part by "[s]ale of surplus mares, weanlings and racehorses".

Consideration: one dominant use for the sale purpose?

1. As noted above at [42], there were two main factors at work in the conclusion of the primary judge: 1. the economic driver of the business, in terms of revenue and profit, was the sale purpose and not the racing purpose (where the latter by itself was said to be uneconomic); 2. establishing racing prowess in the progeny of Godolphin's stallions enhances their value for covering purposes and thus indirectly serves the sale process (and, relatedly, Godolphin's racing activities enables it sift through stock to find good breeding material).
2. There is some force in both of these factors. However, they do not persuade me that the dominant use of the land at Kelvinside and Woodlands in the relevant tax years can be characterised as animal maintenance for the sales purpose, that is, selling stallion covering services and some progeny.

The sales purpose as claimed economic driver of the business

1. Godolphin submitted that "there is nothing inappropriate in the Trial Judge having regard to the commercial realities of Godolphin's operations and activities as part of the determination that there is an integrated operation and thus a single use of the Land". As indicated above, I accept that financial matters can be relevant to the characterisation exercise.
2. What emerges from the financial materials considered above, in summary, is as follows: 1. Revenue from stallion covering services in the relevant years exceeds revenue from racing, but not by a large margin. Prizemoney is an important revenue stream for Godolphin. 2. There are substantially greater expenses - including because of the allocation of staff resources - with respect to racing compared to stallion and breeding activities. 3. Stallion operations are run at a net profit; racing operations run at a net loss. 4. The sale of horses also earns significant revenue for Godolphin, albeit behind either of the other two sources in question. However, this revenue source is not divorced from Godolphin's racing purpose, given that those horses which are sold tend to be those that are less successful at racing. 5. Godolphin's breeding assets across its business are valued significantly more highly than its racing assets. 6. Godolphin as a whole is an enterprise run at a substantial loss, and its continued survival depends on the ongoing support of its ultimate shareholder.
3. As stallion covering services are the largest revenue earner, but not by a large margin, I would not conclude that it is the dominant economic driver of the business such as to render other activities - especially racing - merely secondary and incidental.
4. Further, care should be taken about attributing predominance to the pursuit of covering revenue on the basis that it is the most profitable part of the business (as it clearly is). Whilst all of the categories of exemption in s 10AA(3) involve an objective of commercial gain (see above at [30]), there is no necessity that the activities be undertaken profitably. Business activities by private companies may be undertaken for a range of reasons. Horse racing is an activity that may well involve motivations beyond immediate economic reward, including, for example, obtaining prestige for those who own the business.

Establishing racing prowess adds value to the sales side

1. Central to the decision of the primary judge was her Honour's conclusion that Godolphin conducted "an integrated operation in which the preparation of horses for racing is with the overall or dominant purpose or objective of increasing or maximising the revenue from the nomination fees … and from the sale of the progeny produced by the broodmares" (at [258]). It can be accepted that Godolphin did in a sense have an integrated operation. Further, the evidence was clear that, as her Honour said at [272], "racing prowess (of the horse and of its progeny in due course) is an important factor in the pricing able to be commanded for the sale of stallions' semen". However, the significance of these matters should not be overstated.
2. Godolphin's business model was unusual in combining racing and breeding activities, when more usually those two activities are conducted separately. In a sense there was a degree of vertical integration in Godolphin's business. Godolphin participated both at the production level - by selling covering services and, later, selling some of the progeny born of its own mares - and in the downstream racing market. Godolphin's business obtained some synergistic benefits in both directions, because breeding success produced better racing results, and better racing results aided its breeding sales. Thus the two types of activity were linked. But that does not mean that they should be labelled with one overall purpose, let alone a purpose taking its character from just one of those two types of activity.
3. As noted above, Godolphin did not dispute that use for the purpose of racing could been seen, for the purposes of considering the issue at hand, as a distinct use-for-a-purpose. That is why it did not seek the exemption for its other two properties (see above at [81]). That position reflects the practice in these industries, where the two pursuits are generally pursued separately. The distinction was also reflected in Godolphin's own management accounts, as noted above at [86(1)].
4. Combining the two purposes together - achieving a degree of vertical integration - does not convert two types of activities or purposes into one. Godolphin submitted that it: operates a "breed to race, race to breed model" … That is what sets this case apart from other examples. In this sense, the purpose of sale and the purpose of racing are two aspects of a single composite purpose.

Conclusion

1. Significantly, and reflecting industry practice, there was no dispute that the purposes of maintaining horses for racing and for breeding/sale were distinguishable in the context at hand.
2. Godolphin established that a significant use of the two properties was animal maintenance for the purpose of selling animal produce and progeny. But it did not establish that that should be characterised as the dominant use. A more significant use of the land was animal maintenance for the purpose of racing. True, each purpose aided the other to some extent, and the sales purpose is no doubt more profitable. But taking account of the activities undertaken at the land, the areas directed to those activities, the intensity of those activities and the resources directed to the relevant purposes, the racing purpose was not merely incidental and subservient to the sales purpose. The activities in pursuit of the racing purpose - not the sale purpose - constituted the dominant use of the land at Kelvinside and Woodlands in the relevant tax years.
3. What s 10AA(3)(b) requires is an exercise in characterisation. It is not infrequent that reasonable minds may disagree when undertaking such an exercise. I have reached a different view to that of the learned primary judge and to that of Griffiths AJA. In my view the appeal should be upheld.

Orders

1. The orders of the Court should be as follows: 1. Appeal allowed. 2. Set aside the orders of the Supreme Court made in this matter on 13 April 2022, and in lieu thereof order: 1. The Further Amended Summons is dismissed. 2. Pursuant to s 101(1)(a) of the Taxation Administration Act 1996 (NSW), confirm each of the following: 1. the Land Tax Assessment Notice (Correspondence ID 1703980613) issued by the defendant to the plaintiff on 19 March 2020; and 2. the Land Tax Assessment Notice (Correspondence ID 1710080522) issued by the defendant to the plaintiff on 20 July 2020. 1. The respondent is to pay the appellant's costs at first instance and on appeal.
2. SIMPSON AJA: The facts, circumstances and legal principles relevant to this appeal are comprehensively stated in the judgment of Kirk JA, augmented by the judgment of Griffiths AJA, from each of which I have derived considerable assistance. Other than as is necessary to explain my reasons for agreeing with the orders proposed by Kirk JA I forebear repeating what is contained in those judgments. The following assumes a familiarity with the factual background to the appeal.
3. I begin with a statement of the trite: pursuant to s 75A of the Supreme Court Act 1970 (NSW) the appeal is to be conducted as a rehearing. Subject to the constraints stated in Fox v Percy (2003) 214 CLR 118; [2003] HCA 22 ("Fox"); Robinson Helicopter Company Incorporated v McDermott [2016] HCA 22; (2016) 90 ALJR 679; and Lee v Lee (2019) 266 CLR 129; [2019] HCA 28, the duty of this Court is to decide for itself the proper inference to be drawn from undisputed or established facts: Warren v Coombes (1979) 142 CLR 531 at 551; [1979] HCA 9, endorsed by the plurality in Fox at [25]. In this case, those constraints are of little moment: the essential facts found by the primary judge are largely uncontested, and, in any event, did not depend on assessments of the credibility of witnesses. What is in issue is the inferences to be drawn from the established factual substratum. In these circumstances, while it is necessary and appropriate to give respect and weight to the conclusion of the primary judge, this Court is in as good a position as the primary judge to decide on the proper inferences to be drawn, and it is the duty of this Court to give effect to its own conclusions.
4. The ultimate inference on which the outcome of the case, both at first instance and on appeal, depends concerns the purpose for which the land in question is used. By s 10AA(3)(b) of the Land Tax Management Act 1956 (NSW) Godolphin, the owner of the land, was entitled to an exemption from land tax if: "…the dominant use of the land was for: (b) the maintenance of animals … for the purpose of selling them or their natural increase or bodily produce."

Some additional relevant caselaw

1. Kirk JA has helpfully summarised some of the relevant caselaw. His Honour's summary includes several cases involving statutes in other contexts which raise the issue of the use of land, including in the planning context. For my own part, I consider that cases which deal with the legal meaning of the phrase "use of land", as in the case of People for the Plains Inc v Santos NSW (Eastern) Pty Ltd (2017) 220 LGERA 181; [2017] NSWCA 46 referred to by Kirk JA at [25] above, may have some, but only limited, value. As Leeming JA stated in Chief Commissioner of State Revenue v Adams Bidco Pty Ltd [2019] NSWCA 34 at [26]: Much time was occupied in the parties' submissions with invocations of other decisions on the legal meaning of the same words in different contexts. The submissions were of a high quality. However, enough has been said to demonstrate their limited value. It is not likely that a decision on the words in a different statute will capture the precise grammatical context of the word, nor its place in the statute, nor the presence of a quantitative measure in the immediately following subsection applying to the same verb ("comprise"), nor the statute's purpose. It is axiomatic that the starting point for construction is the statutory text considered in light of its context and purpose. It must follow that a decision on the same or similar adverb in a different context is of limited value.
2. That is not to say that decisions regarding the legal meaning of the "use" of land in other statutory contexts are entirely irrelevant to the construction and application of the exemption in s 10AA(3)(b) of the Act. As Barrett AJA (with whom Macfarlan and Ward JJA agreed) noted in Chief Commissioner of State Revenue v Metricon Qld Pty Ltd [2017] NSWCA 11; (2017) 105 ATR 11 at [45], although caution is required in considering cases involving different statutory contexts in which the word "use" is employed with respect to land, "use" in relation to land has "a core meaning independent of statutory context". Barrett AJA referred approvingly in this context to what was said by French CJ, Kiefel, Bell and Keane JJ in New South Wales Aboriginal Land Council v Minister Administering Crown Lands Act (2016) 260 CLR 232; [2016] HCA 50 at [34], namely that while words such as "used" and "occupied" may take much of their meaning from their statutory context, that is not to say that they are devoid of a commonly understood meaning. Rather, they "require an examination of activities undertaken upon the land in question …".

Key aspects of the primary judgment summarised

1. Although Kirk JA has provided a helpful summary of much of the primary judge's reasons, it is desirable to emphasise the following key aspects. The determinative part of those reasons is at PJ[253]-[274].
2. Her Honour acknowledged at PJ[253] the principal difference in the parties' positions, namely how each party viewed Godolphin's racing operations in the context of Godolphin's overall equine operations. The Chief Commissioner contended that the breeding, training, education and spelling were conducted for racing and this formed the dominant purpose and use of the relevant land. Conversely, Godolphin emphasised that it conducted a "breed to race, race to breed" model. Her Honour characterised at PJ[256] the question for determination as follows: The question thus is whether there are two separate activities (such that it is necessary to determine which is the dominant use on each parcel of land) or an integrated or composite activity (involving both breeding and raising [sic - racing]) such that the dominant use of all parcels (whether for breeding, training or spelling) is for the purpose of the ultimate sale of the stallion's semen and the broodmares' progeny.
3. Accepting that much of the use of the relevant land was devoted to breeding, training and preparation of horses for racing, the primary judge concluded at PJ[258]-[259] that this formed part of an integrated operation conducted with the overall or dominant purpose of increasing or maximising the revenue from nomination fees and, to a lesser degree, from the sale of progeny and any racing success. In reaching this conclusion, her Honour rejected the contention that the integrated operation was conducted for the dominant purpose of generating prizemoney, finding at PJ[259] that this did not "make sense from an economic point of view" (see, for example, at PJ[254], where it is noted that Godolphin's stallion servicing operations brought in $17,869,000 in 2019 whereas racing and breeding operations sustained losses in the same year). Another important factor which influenced her Honour's conclusion was the evidence that racing prowess (of the horse and of its progeny in due course) is "an important factor in the pricing able to be commanded for the sale of stallions' semen" (PJ[272]).
4. Her Honour also recited at PJ[261]-[267] some of the evidence given by Mr Cox, which included his candid acceptance of propositions, for example, that training was conducted to best prepare to win in races, spelling was an integral part of racing preparation, yearlings were not sold until their racing potential was able to be assessed, and that those horses retained by Godolphin after their fourth year were those horses which it hoped would earn better prizemoney than those who were sold. However, Mr Cox denied that achieving racing success was the dominant use and purpose of the relevant land; rather, his understanding was that "it was one of two purposes - to strive for stallion excellence and to strive for racecourse success" (PJ[267]). Her Honour found at PJ[269] that, in determining the issue objectively, Mr Cox's evidence supported her earlier conclusion that Godolphin conducted an integrated thoroughbred stud operation.

Weighing the relevant factors

1. In his reasons for judgment, Kirk JA identified the following three topics as being of particular importance to the weighing process in this case, namely: 1. the operation of Godolphin's business; 2. the properties and the activities undertaken there; and 3. the financial aspects of the business.
2. With some important qualifications (which I will soon identify), I gratefully adopt Kirk JA's description of many of the key facts under each of these headings.
3. The qualifications are as follows.
4. First, as to the operation of Godolphin's business, the matters referred to by his Honour support in my view the primary judge's finding that Godolphin conducted an integrated operation on the relevant land, which was conducted with both breeding and racing in mind. That characterisation of Godolphin's operations on all the relevant parcels of land is supported by: 1. Mr Cox's evidence (which was accepted unreservedly by the primary judge). 2. Relevant parts of the Flying Start presentations (as referred to at [167] above). 3. The document described as "Godolphin Australia Report" dated December 2018 (as referred to by Kirk JA at [53]). 4. Godolphin's marketing practices, which include publicising Godolphin-bred race winners whether raced by itself or by others. The marketing is described as some length in an affidavit dated 26 August 2020 by Godolphin's head of sales, Mr Charles Pulford, whose evidence was implicitly accepted by the primary judge. Mr Pulford said that positive information about the performance of Godolphin's thoroughbreds and progeny, whether owned by Godolphin or not, increased the value of Godolphin's brand, demonstrated the positive attributes of Godolphin's thoroughbred horses to the wider market, and increased prices for Godolphin's stallion covering services. Mr Cox gave evidence to similar effect, which was explicitly accepted by the primary judge at PJ[265]. 5. The extensive evidence which highlights the symbiosis between genetics and bloodlines in breeding thoroughbred horses and success in thoroughbred horse racing. This evidence is helpfully summarised by Kirk JA at [66]ff. 6. Finally, and significantly, reference should also be made to the fact that the bloodlines of all thoroughbred horses, in Australia and internationally, must be traced back to only 3 Arab stallions, namely Darley, Godolphin and Byerley Turk. This makes the industry unique. It strongly reinforces the nexus between the breeding and racing of thoroughbred horses. The two activities can and normally are conducted separately, but that is not the case under Godolphin's evidently unique business model.

Net Income (Management Accounts) Year Stallion nominations ($) Racing ($) Horse sales ($) 2013 25,922,000 -13,784,000 3,794,000 2014 17,911,000 -14,189,000 4,255,000 2015 20,925,000 -16,806,000 3,243,000 2016 19,242,000 -13,052,000 11,986,000 2017 22,260,000 -12,899,000 13,264,000 2018 13,401,000 -12,065,000 7,364,000 2019 17,869,000 -167,000 9,773,000

Value of Biological Assets (Financial Statements) Year Bloodstock ($) Racing Stock ($) 2013 Broodmares 89,282,211 82,629,350 Stallion interests 161,065,000 Foals 17,140,100 Yearlings 29,705,000 Total 297,192,311 2014 Broodmares 91,067,800 76,697,885 Stallion interests 152,135,000 Foals 19,600,000 Yearlings 24,057,000 Total 286,859,900 2015 Broodmares 96,096,600 55,795,199 Stallion interests 157,420,700 Foals 19,053,000 Yearlings 20,846,777 Total 293,471,077 2016 Broodmares 84,427,330 57,358,731 Stallion interests 91,654,700 Foals 22,324,502 Yearlings 19,061,500 Total 217,468,032 2017 Broodmares 78,056,679 49,768,599 Stallion interests 71,019,300 Foals 11,230,565 Yearlings 21,860,752 Total 182,167,296 2018 Broodmares 86,637,337 56,492,597 Stallion interests 67,517,400 Foals 12,833,034 Yearlings 11,418,315 Total 178,406,086 2019 Broodmares 94,883,905 49,448,327 Stallion interests 60,301,400 Foals 10,966,836 Yearlings 12,510,284 Total 178,662,425

First table

1. As Kirk JA has noted at [82] above, the average prizemoney for racing during the relevant years is $19.6 million (for all Godolphin horses in Australia).
2. The figures for revenue from stallion covering set out in the first table reflect all of Godolphin's Australian operations. Separate figures, however, were provided in Mr Hartmann's affidavit dated 26 August 2020 in respect of stallion coverings at Kelvinside alone. As Kirk JA has noted at [80], these figures range from approximately $14 million to $29.4 million during the relevant years, with an average of $22.1 million.
3. The figures for sales of thoroughbred horses for all of Godolphin's Australian operations during the relevant years range between $5 million and $19.5 million, with an average of $12.8 million.
4. At PJ[255], the primary judge noted Mr Cox's explanation that the "breed to race, race to breed" model involved the breeding and supply of stock for the racing stable, which is followed by the racing stable improving the stock and creating the supply of bloodstock assets. This is the symbiosis to which I have already referred. Her Honour then added that this is reinforced by the fact that "stallion covering fees are clearly the key revenue driver for the business".
5. At [83] above, Kirk JA queries this statement on the basis that, on average, stallion covering fees are not substantially higher than prizemoney (approximately 13% higher).
6. I see no error in the primary judge's reasons in circumstances where, during the calendar years 2013 to 2019, the revenue from covers by stallions maintained at Kelvinside exceeded the total prizemoney for all Godolphin horses racing in Australia by a minimum of approximately $1 million (2019) and a maximum of approximately $18 million (2013).
7. There is also a difficulty with using averages in comparing the figures for the period 2013 to 2019. The average figure for prizemoney is significantly distorted by the aberrational amount of $34 million in 2019. In any event, in determining whether the dominant use of land for an integrated operation such as that here is for maintaining animals for the purpose of selling them, their progeny or bodily produce, account can be taken of relevant financial figures relating to breeding and sales on the one hand and racing on the other. Financial figures relating to breeding includes not those relating to stallion covering, but also figures relating to the sale of the progeny produced by Godolphin's broodmares.

Second table

1. The second table shows net income for the relevant years under the headings "Stallion nominations", "Racing" and "Horse sales". It may be noted that the table does not include net income figures for what is described in the relevant management accounts as "Breeding Operations". Those figures range from a loss of $8,645,000 in 2016 to a loss of $9,116,000 in 2013. The average loss for breeding operations is approximately $9 million during the relevant years. Those losses are made up in large part by staffing costs in conducting the breeding operations.
2. Accordingly, both racing operations and breeding operations were conducted at significant losses during the relevant years but the net losses for racing were generally somewhat higher than the losses for breeding operations. The fact that there are losses is not determinative by itself. That is because it is well-settled that there is no requirement that there be an immediate productive return from a particular use of land as long as some benefit or advantage accrues (see Metricon at [61] per Barrett AJA and Kirk JA's reference at [99] above to Tweddle v Federal Commissioner of Taxation (1942) 180 CLR 1 at 7; [1942] HCA 40). In the particular and apparently unusual circumstances here, Godolphin's racing operations provide a benefit or advantage to its breeding operation in the manner described by Mr Cox, which was correctly accepted by the primary judge.
3. It might also be noted that the second table does not include figures from the management accounts which concern Godolphin's horse purchases. As has been noted, Godolphin periodically refreshes or expands its cohort of breeding stallions by buying in stallions who have been bred elsewhere. Those stallions may then be used to cover Godolphin's own broodmares or mares owned by other persons. The management accounts, part of which record Godolphin's profit and loss by activity type, record that during the relevant period horse purchases ranged from $1.175 million in 2017 up to almost $17 million in 2014 and were in excess of $18 million in 2015. The average spent on horse purchases during the relevant period is $8.334 million. Although these figures are described as losses in the management accounts, the purchase of new stallions obviously provides some benefit or advantage to Godolphin's breeding and racing operations with reference to the progeny produced by those stallions.

Third table

1. I consider that, consistently with Leda, appropriate weight should be given not only to those parts of Godolphin's financial records to which Kirk JA draws attention, but also to the values attributed in those records to Godolphin's "Biological Assets". Those "Biological Assets" are divided into the following categories:

• Broodmares;
• Stallion interests;
• Horses in training;
• Foals;
• Yearlings; and
• Cattle and other.

1. Each of those biological assets is given a net market value in each of the relevant Godolphin financial statements for the years ended 31 December 2013 to 31 December 2019.
2. Godolphin's aide memoire has deleted the values attributed to "Cattle and other" and created two broad categories for comparison purposes (namely "Bloodstock" and "Racing Stock"). There is no difficulty in grouping the values of broodmares and stallion interests under the heading "Bloodstock" because neither of those kinds of horses will race.
3. It is significant that, in the third table, the values attributed to broodmares and stallion interests, which primarily relate to breeding, far outweigh the values attributed to horses in training. Although the figures vary from year to year, this is the case in respect of all 7 years covered by the table. Generally speaking, the value of bloodstock outweighs the value of racing stock by a minimum of 3:1 in the relevant years and was as high as over 5:1 in 2015.
4. It is true that some allowance may need to be made for the fact that the values attributed to foals and yearlings are included in the bloodstock values. Given their age, a decision would not have been made whether or not any of those horses will ultimately be sold by Godolphin or raced by Godolphin and then a decision made as to whether any of the females or ungelded males will be kept for breeding or sale. As Kirk JA has noted at [59] above, approximately 95% of Godolphin's male horses are gelded, usually after an initial assessment has been made of their racing, and thus breeding, prospects. Moreover, as his Honour has also noted at [60], Godolphin sells approximately 70% of the thoroughbred horses which it breeds, but a decision to sell is not usually made until after the horse is 3 years old. But even if some allowance is made for these matters, it remains the case that the value of bloodstock far outweighs the value of racing stock. This supports the primary judge's conclusion.

Conclusion

1. For these reasons, I am not persuaded that the primary judge fell into appellable error. I respectfully agree with her Honour's conclusion and related reasoning. In particular, I respectfully agree with her Honour's conclusion at PJ[258], which is consistent with the approach in Thomason: I accept that much of the use of the Land is devoted to breeding, training and preparation of the horses for racing (and that an obvious objective is to maximise prize winnings or success on the track). However, viewed overall, I have concluded that this is an integrated operation in which the preparation of horses for racing is with the overall or dominant purpose or objective of increasing or maximising the revenue from the nomination fees (i.e., the sale of bodily produce) and from the sale of the progeny produced by the broodmares.
2. Accordingly, I consider that the appeal should be dismissed, with costs.
3. In these circumstances there is no need to address Godolphin's notice of contention.

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: 20 March 2023