Valuer-General Victoria v WSTI Properties 490 SKR Pty Ltd
[2024] VSCA 157
•4 July 2024
| SUPREME COURT OF VICTORIA COURT OF APPEAL |
| S EAPCI 2023 0081 |
| VALUER-GENERAL VICTORIA | Applicant |
| V | |
| WSTI PROPERTIES 490 SKR PTY LTD | Respondent |
---
| JUDGES: | EMERTON P, KENNEDY and LYONS JJA |
| WHERE HELD: | Melbourne |
| DATE OF HEARING: | 1 March 2024 |
| DATE OF JUDGMENT: | 4 July 2024 |
| MEDIUM NEUTRAL CITATION: | [2024] VSCA 157 |
| JUDGMENT APPEALED FROM: | WSTI Properties 490 SKR Pty Ltd v Valuer-General Victoria (Red Dot) (Land Valuations) [2023] VCAT 734 (Quigley J and Senior Member Jacono) |
---
LAND VALUATION – Assessment of site value under the Valuation of Land Act 1960 – Where land is subject to site-specific heritage overlay due to presence of heritage building – Where other planning controls permit more intensive development of land – Where heritage overlay requires a permit to demolish, remove or carry out works affecting external appearance of building – Whether building is an improvement for the purpose of site value assessment – Relevance of highest and best use – Leave to appeal granted – Appeal dismissed.
Valuation of Land Act 1960, ss 2(1), 2(2), 5A.
Commonwealth Custodial Services Ltd (as Trustee for Burwood Trust Fund) v Valuer-General (NSW) (2006) 148 LGERA 38; Trust Company of Australia Ltd v Valuer-General (2007) 154 LGERA 437; Surfers Paradise Resort Hotel Pty Ltd v Department of Natural Resources and Water (2007) 163 LGERA 14; Spencer v Commonwealth (1907) 5 CLR 418; Morrison v Federal Commissioner of Land Tax (1914) 17 CLR 498; ISPT Pty Ltd v Melbourne CC [2018] VCAT 1647, considered.
Caltex Oil (Australia) Pty Ltd v Chief Executive, Department of Lands (1996) 16 QLCR 435; Maurici v Chief Commissioner of State Revenue (2003) 212 CLR 111, distinguished.
---
| Counsel | |||
| Applicant: | Mr A Finanzio SC with Mr G Kozminsky and Mr J Waller | ||
| Respondent: | Mr S Goubran KC with Mr P Annabell | ||
Solicitors | |||
| Applicant: | Maddocks | ||
| Respondent: | Minter Ellison | ||
TABLE OF CONTENTS
Introduction
Legislative framework
Determining the value of land
Amendments to the VLA concerning heritage-affected land
The planning scheme
Procedural history
First valuation objection
Second valuation objection
Expert evidence in the Tribunal
Heritage evidence
Planning evidence
Valuation evidence
The Tribunal’s decision
Tribunal’s approach to determining site value
Effect of site-specific heritage overlay
Whether the building adds value
Valuation method
Whether VGV’s evidence of a higher value is a new valuation
Tribunal’s conclusions
Proposed grounds of appeal
Grounds 1, 2, 3 & 4 — Assessing the site value of land
VGV’s submissions
WSTI’s submissions
Consideration
Grounds 5 & 6 — The Tribunal’s role on review
Conclusion
EMERTON P
KENNEDY JA
LYONS JA:
Introduction
‘Landene’ is a Queen Anne-influenced, two-storey red brick villa constructed in 1897. It sits on a 1,046m2 parcel of land at 490 St Kilda Road, Melbourne (‘the Land’).[1] As one of the few remaining fragments of the 19th century built environment on St Kilda Road, it is surrounded by much newer commercial and residential towers.
[1]More particularly described as Lot 1 on Title Plan 110409N, contained in Certificate of Title Volume 10114 Folio 065.
The Land is zoned ‘Commercial 1’ under the Port Phillip Planning Scheme. It is subject to a number of design and development overlays (‘DDOs’),[2] along with a site-specific heritage overlay.[3] Among other things, the site-specific heritage overlay prevents the demolition of Landene without a permit, thus constraining the development of the Land in accordance with its zoning.
[2]Schedules 13, 26-5A and 28.
[3]Schedule 331.
The respondent, WSTI Properties 490 SKR Pty Ltd (‘WSTI’), purchased the Land in August 2019 for $8.25 million. The new owner extensively renovated the interior of Landene for use as a private gallery for its collection of art and antiques.
The Land was valued on 1 January 2020 and 1 January 2021 pursuant to the Valuation of Land Act 1960 (‘VLA’). On both occasions, the Land returned a site value of $6.2 million.[4]
[4]On the bases of the assessed site values, the Port Phillip Council issued, respectively, two rates, charges and valuation notices, one dated 7 September 2020 and the other dated 21 September 2021.
WSTI objected to both the 2020 and 2021 site valuations on the basis that they were too high. The applicant, Valuer-General Victoria (‘VGV’), disallowed WSTI’s objection to the 2020 site valuation and WSTI applied[5] to the Victorian Civil and Administrative Tribunal for a review of that decision.[6]
[5]Pursuant to s 22(1) of the Valuation of Land Act 1960 (‘VLA’).
[6]This review application was the subject of proceeding P1013/2021 (‘P1013 proceeding’).
As to WSTI’s objection to the 2021 site valuation, VGV was deemed to have made a decision that no adjustment to the valuation was justified.[7] WSTI also applied to the Tribunal to review that decision.[8]
[7]This was because of the passing of four months within which the valuer was required to provide a decision on the objection but that no such decision was given to WSTI: VLA, s 22(2).
[8]This review application was the subject of proceeding P10049/2022 (‘P10049 proceeding’).
The two Tribunal proceedings were heard together over four days by the President of the Tribunal and a Senior Member with expertise in land valuation. In the Tribunal, as here, the contest turned on whether Landene was to be treated as an improvement for the purpose of assessing site value under the VLA.
On 30 June 2023, the Tribunal agreed that the returned site valuations were too high, and made orders reducing the site value in each of the 2020 and 2021 valuations to $2.925 million.
VGV now appeals the Tribunal’s decision on the grounds that the Tribunal erred in its construction of the provisions of the VLA that governed the determination of the site value of the Land, and that it misapprehended its task on review.
For the reasons that follow, VGV’s application for leave to appeal should be allowed but the appeal should be dismissed.
Legislative framework
The functions of VGV include the performance of duties and functions conferred by legislation,[9] and to cause general valuations and supplementary valuations to be made.[10] Section 11 of the VLA requires a general valuation of rateable land[11] within an area to be made as at 1 January in each calendar year.
Determining the value of land
[9]VLA, s 5(1)(a).
[10]Ibid s 5(1)(ab).
[11]‘Rateable land’ has the same meaning as in the Local Government Act 1989, which is ‘all land’ with certain exceptions, none of which is applicable here.
The ‘site value’ of land is defined in s 2 of the VLA as
the sum which the land, if it were held for an estate in fee simple unencumbered by any lease, mortgage or other charge, might in ordinary circumstances be expected to realise at the time of the valuation if offered for sale on such reasonable terms and conditions as a genuine seller might be expected to require, and assuming that the improvements (if any) had not been made.
This essentially reproduces the test for ascertaining the value of land in Spencerv Commonwealth[12] — the price paid for land as the result of a bargain reached between a willing but not anxious vendor and a willing but not anxious purchaser, both properly acquainted with the land[13] — but based on the assumption that any ‘improvements’ on the land ‘had not been made’.
[12](1907) 5 CLR 418; [1907] HCA 82 (‘Spencer’).
[13]Ibid 432 (Griffith CJ); 436–7 (Barton J); 441 (Isaacs J).
The meaning of ‘improvements’ is given in the VLA as follows:
improvements, for the purpose of ascertaining the site value of land, means all work actually done or material used on and for the benefit of the land, but in so far only as the effect of the work done or material used increases the value of the land and the benefit is unexhausted at the time of the valuation, but, except as provided in subsection (2AA), does not include—
(a)work done or material used for the benefit of the land by the Crown or by any statutory public body; or
(b)improvements comprising—
(i)removal or destruction of vegetation or the removal of timber, rocks, stone or earth; or
(ii) the draining or filling of the land or any retaining walls or other works appurtenant to the draining or filling; or
(iii) the arresting or elimination of erosion or the changing or improving of any waterway on or through the land—
unless those improvements can be shown by the owner or occupier of the land to have been made by that person or at that person’s expense within the fifteen years before the valuation[.][14]
[14]VLA, s 2.
Importantly, in order to be an improvement, the works done or materials used (‘the improvement’) must have the effect of increasing the value of the land. The ‘improvements’ referred to in para (b), essentially involving the removal of vegetation, and landscaping and drainage works, are works that generally increase the value of the land, but they are not to be treated as such for the purpose of determining the site value of the land unless made in the last 15 years.
Section 2(2) of the VLA specifies how the ‘value of improvements’ are to be assessed and relevantly provides:
In estimating the value of improvements on any land for the purpose of ascertaining the site value of the land, the value of the improvements is the sum by which the improvements upon the land are estimated to increase its value if offered for sale on such reasonable terms and conditions as a genuine seller might in ordinary circumstances be expected to require.
Section 5A of the VLA is the central provision for determining the value of land, including land that is affected by heritage controls. Section 5A provides:
(1)Unless otherwise expressly provided where pursuant to the provisions of any Act a court board tribunal valuer or other person is required to determine the value of any land, every matter or thing which such court board tribunal valuer or person considers relevant to such determination shall be taken into account.
(2)In considering the weight to be given to the evidence of sales of other lands when determining such value, regard shall be given to the time at which such sales took place, the terms of such sales, the degree of comparability of the lands in question and any other relevant circumstances.
(3)Without limiting the generality of the foregoing provisions of this section when determining such value there shall, where it is relevant, be taken into account—
(a)the use to which such land is being put at the relevant time, the highest and best use to which the land might reasonably be expected to be put at the relevant time and to any potential use;
(b)the effect of any Act, regulation, local law, planning scheme or other such instrument which affects or may affect the use or development of such land;
(c)the shape size topography soil quality situation and aspect of the land;
(d)the situation of the land in respect to natural resources and to transport and other facilities and amenities;
(e)the extent condition and suitability of any improvements on the land; and
(f)the actual and potential capacity of the land to yield a monetary return.
Amendments to the VLA concerning heritage-affected land
Because of the way in which this appeal was argued,[15] it is necessary to have regard to the provisions of the VLA that were repealed before both the 2020 and 2021 valuations, specifically ss 2(8) and 2(9). Those provisions were directed to the assessment of the value of heritage-affected land and modified, for heritage sites, the approach to determining site value. They were repealed in 2019 by the State Taxation Acts Amendment Act 2019.[16]
[15]See [93]–[98] of these reasons for VGV’s submissions on appeal.
[16]See s 56 of the State Taxation Acts Amendment Act 2019.
Section 2(8) relevantly provided that the site value of any rateable land which was a registered place within the meaning of the Heritage Act 2017 or on which there was situated a building which was included in the Heritage Register established under that Act had to be calculated on the basis of the assumptions set out in paragraphs (a) to (c) of that subsection. Those assumptions included that such land could be used only for the purpose for which it was used at the date of valuation and that the building could not be removed or demolished.
Section 2(9) provided that if a planning scheme under the Planning and Environment Act 1987 prohibited the pulling down or removal of a building, or required a permit before a building could be pulled down or removed and the responsible authority had refused to grant a permit or the Tribunal had directed that no permit be issued, the value of any rateable land that included the building had to be calculated on the basis that the building could not be pulled down or removed.
The second reading speech for the State Taxation Acts Amendment Bill 2019 explained that ss 2(8) and 2(9) of the VLA were to be repealed in response to the decision of the Tribunal in ISPT Pty Ltd v Melbourne CC[17] (known as the ‘GPO case’).[18]
[17][2018] VCAT 1647.
[18]Victoria, Parliamentary Debates, Legislative Assembly, 28 May 2019, 1638 (Tim Pallas, Treasurer, Minister for Economic Development, Minister for Industrial Relations).
The GPO case concerned a dispute about the site value of the land on which the former Melbourne General Post Office (‘GPO’) was located. As the GPO building and land were heritage listed, the GPO’s site value was required to be determined in accordance with the assumptions in s 2(8) of the VLA. Melbourne City Council (as the authorised valuer) had assessed the GPO’s site value as $14,800,850. ISPT, the owner of the GPO land, contended for the figure of $1. In the Tribunal, VGV (standing in for the authorised valuer) contended for the figure of $26,375,000.
VGV argued that the valuation should take into account the commercial trading in the GPO building and advanced a novel ‘rental differential’ methodology, by which the site value of the land would be derived by determining the value of the land if free of heritage restrictions and reducing that amount by the ratio of the rental income actually being achieved to the rental income that might be achieved if the land were developed to its maximum potential free of heritage restrictions. For its part, ISPT relied upon a hypothetical feasibility method, which assessed the cost of developing the land to notionally construct the improvements actually in place compared to the returns that could be derived from doing so. Those costs were so high that the net returns were negative, resulting in a valuation of $1.
The Tribunal held that in the absence of comparable sales, the hypothetical feasibility method (advanced by ISPT) was a permissible approach for the calculation of site value. Although not an ideal basis on which to assess site value, the feasibility method supported a site value of $1. The Tribunal observed that given the practical difficulties of applying the statutory assumptions under the VLA, the heritage provisions were anachronistic, and led to inconsistent outcomes and litigation. The Tribunal suggested that it was desirable that the legislation be clarified to simplify the exercise of valuing heritage registered land.
Thus, the second reading speech for the Bill recorded the following:
Disputes over the valuation of heritage registered properties have led to significant litigation costs for the Valuer-General and municipal councils. As the recent GPO case has demonstrated, these disputes have resulted in unrealistic outcomes such as site values of $1. It is not fair to the rest of the Victorian community that large landowners avoid the payment of land tax on valuable heritage-registered properties due to this anomaly in the valuation regime. This Bill will bring much needed clarity to the valuation of these properties by repealing the specific assumptions that currently must be made when valuing heritage-registered property. This will enable valuers to determine site values for heritage-registered property taking into account its highest and best use on a consistent basis to all other properties. The amendments will ensure that any objections, reviews and appeals to the site values issued for heritage registered properties against the 2018 valuations onwards will be considered in light of the amendments being made by this Bill. This is necessary to ensure consistent treatment of all landowners and to protect the substantial land tax revenue at risk. The Bill does not remove the normal rights of taxpayers to object to site valuations for reasons other than those related to the specific heritage valuation provisions being repealed.[19]
[19]Ibid.
Following the repeal of ss 2(8) and 2(9) of the VLA, land containing heritage-registered properties was to be treated in the same way as other land.
The planning scheme
As discussed, the Land is zoned ‘Commercial 1’ in the Port Phillip Planning Scheme. The purpose of the ‘Commercial 1’ zoning is ‘to create vibrant mixed use commercial centres for retail, office, business, entertainment and community uses’, with ‘residential uses at densities complementary to the role and scale of the commercial centre’.
The Land is, as mentioned, subject to three DDOs, the purposes of which are to implement Port Phillip Council’s planning strategy and planning policy framework, as well as to identify areas that are affected by specific requirements relating to the design and built form of new development.
The three DDOs applicable to the Land are as follows:
(a)Schedule 13, which is aimed at protecting the view line of the Shrine of Remembrance along Swanston Street from outside the State Library;
(b)Schedule 28, which ensures protection of the flight path areas associated with the helicopter landing site at the Alfred Hospital; and
(c)Schedule 26-5A, which concerns the development potential of the land if vacant. The Land being located within Sub-Precinct 5A, a building height of 65 metres, with setback requirements, is permitted.
The Land is also subject to a site-specific heritage overlay (HO331), which seeks to conserve and enhance Landene’s heritage character and ensure that development does not adversely affect its heritage significance.
The heritage overlay means that a permit is required to carry out works on Landene that may affect its external appearance (including such things as replacing the fence, installing solar panels or water tanks, and painting, rendering or sandblasting externally). A permit is also required to demolish or remove a building.
The decision guidelines in cl 43.01-8 of the Port Phillip Planning Scheme make repeated reference to avoiding matters that may adversely affect the significance of the heritage place.
Procedural history
First valuation objection
On 7 September 2020, the Port Phillip Council issued a rates, charges and valuation notice in respect of the Land for the period 1 July 2020 to 30 June 2021 (‘the 2020 valuation’). The notice recorded a site value of $6.2 million as at 1 January 2020, a capital improved value of $7.2 million, and a net annual value of $360,000. The rates payable were $13,923.30.
The 2020 valuation was prepared by Briony Stephen and Lee-Ming Tan, who were contract valuers in the employ of Matheson Stephen Valuations.
On 4 November 2020, WSTI lodged an objection to the 2020 valuation on the basis that the assessed site value was ‘too high’ (‘first objection’). WSTI instead contended for a site value of $1.825 million. The submission in support of the contended value was as follows:
The returned Site Value assessment is too high having regard to heritage constraints of the property being identified as an individually significant heritage property, the estimated cost to replace the existing heritage constrained building, and the definitions as prescribed by the Valuation of Land Act 1960.
Subsequently, between November and December 2020, WSTI and VGV exchanged prescribed information[20] concerning the 2020 valuation.
[20]Pursuant to s 20 of the VLA.
On 11 June 2021, one of the contract valuers, Lee-Ming Tan, wrote to WSTI on behalf of the Port Phillip Council, notifying WSTI that its first objection had been disallowed. The letter stated:
A valuer has revised the valuation objection and considers that no adjustment in the valuation is justified. Please refer to the attached notices of disallowance for particulars of the property and outcome.
The attached disallowance notice was also dated 11 June 2021 and was signed by the contract valuer. The notice recorded the valuation authority as VGV and stated that ‘no adjustment in the valuation is justified’ pursuant to s 21(3)(a) of the VLA.
On 9 July 2021, WSTI applied to the Tribunal pursuant to s 22(1) of the VLA to review the decision to disallow the first objection. That application is proceeding P1013/2021.
Second valuation objection
On 21 September 2021, the Port Phillip Council issued another rates, charges and valuation notice in respect of the Land for the period 1 July 2021 to 30 June 2022 (‘the 2021 valuation’). The notice again recorded a site value of $6.2 million as at 1 January 2021, a capital improved value of $7.2 million, and a net annual value of $360,000. The rates payable were $14,283.95.
The 2021 valuation was prepared by Briony Stephen and Lee-Ming Tan, contract valuers from Matheson Stephen Valuations.
On 12 November 2021, WSTI lodged an objection to the 2021 valuation, again on the basis that the assessed site value was ‘too high’ (‘second objection’). WSTI contended for a site value of $2 million. The reason for the second objection was as follows:
The returned Site Value assessment is too high having regard to heritage constraints of the property being identified as an individually significant heritage property and the definitions as prescribed by the Valuation of Land Act 1960.
There was no response to WSTI’s objection to the 2021 valuation. In the event, the four months within which a decision had to be made on the second objection lapsed and, by virtue of s 22(2) of the VLA, VGV was deemed to have made a decision that no adjustment to the 2021 valuation was justified.
On 9 December 2021, after VGV had provided the prescribed information under the VLA, the valuers retained by each of the parties corresponded and agreed to wait for the Tribunal’s decision in proceeding P1013/2021, rather than engaging in further discussions.
On 10 August 2022, WSTI applied to the Tribunal for a review of VGV’s deemed decision pursuant to s 22(1) of the VLA. That application is proceeding P10049/2022.
On 26 October 2022, the Tribunal ordered that proceedings P1013/2021 and P10049/2022 be heard and determined together, and that evidence in one proceeding be taken to be evidence in the other proceeding.
Expert evidence in the Tribunal
Heritage evidence
Heritage consultants Mr Bryce Raworth, on behalf of WSTI, and Ms Anita Brady, on behalf of VGV, each provided expert reports on the heritage controls affecting the Land.
Mr Raworth opined that neither the Tribunal nor the Council would support an application for a full demolition of Landene. He considered that there was little potential for substantive development on the site with the existing building in its current location, and little opportunity for substantive additions to the existing building.
Ms Brady opined that in any application for redevelopment, the existing building would have to be substantially retained, but the Council might approve applications to extend the building on the ground and first floors.
Ms Brady also gave evidence that without Landene, the heritage value of the Land was effectively extinguished and heritage-related development constraints would fall away, such that the Land’s development potential would be consistent with that of adjoining properties, which are not subject to heritage overlays.
In a joint statement, the experts confirmed that Landene should be substantially retained in any redevelopment and that demolition would not be supported by the Council or the Tribunal. However, some minor demolition and alteration to parts of the building might be possible. They also generally agreed that small-scale additions might be possible to the rear and south of the Land.
Planning evidence
Mr Tim McBride-Burgess and Mr Andrew Clarke, town planning experts on behalf of WSTI and VGV respectively, gave evidence on the planning controls affecting the Land.
Mr McBride-Burgess considered that the heritage constraints and the mandatory setback requirements imposed by DDO sch 26-5A significantly constrained the Land’s development potential, such that the redevelopment of the Land on any substantial scale was not possible. However, absent the heritage constraints, redevelopment of the Land for a mix of uses within a tall building was strategically conceivable. Mr McBride‑Burgess was not instructed to express a view assuming the Land was vacant.
Mr Clarke accepted and adopted all Ms Brady’s recommendations and conclusions regarding heritage issues, including that Landene should be retained and not demolished. According to Mr Clarke, the Land would have significant development potential if it was vacant and free from the heritage impediments. Mr Clarke set out four possible development scenarios, comprising a building of either 15 or 17 stories, with variable floor space ratios. He also proposed two development scenarios if Landene was retained. He considered the possibility of further additions to the existing building, being 30m2 to 140m2 of gross floor area over two floors.
In a joint statement, Mr McBride-Burgess and Mr Clarke agreed with the heritage experts’ opinions that the existing building should be substantially retained and that opportunities for additions were limited.[21] Both planners agreed that if the property did not have heritage constraints, larger-scale development could be considered.
Valuation evidence
[21]While the planning experts disagreed about the requirement for a mandatory setback of 4.5 metres to the western boundary of the Land, they agreed that if a 4.5 metre setback is required, the remnant developable area is confined to the areas in front of the existing garage, which could be developed to two stories, providing a total floor area addition of 18m2. It was also agreed that if the 4.5 metre setback is not required, the remnant developable area is a ground level extension, giving an additional 50m2 of total floor area. In other words, depending on whether there is a 4.5 metre setback requirement, the Land could have an additional development area of between 18m2 to 50m2.
VGV relied upon valuation reports by Mr Nicholas Haines of Matheson Stephen Valuers[22] and WSTI relied upon valuation reports by Mr Grant Jackson of m3property.[23] Each valuer prepared a report in respect of the first objection (in P1013/2021); and each valuer also prepared a subsequent report in respect of the second objection (in P10049/2022). Mr Haines prepared a third report entitled ‘Evidence in Reply’ that, in substance, argued the case for the valuation methodology he used in his two earlier reports.
[22]Valuation reports dated 20 May 2022, 22 December 2022 and 27 March 2023, respectively.
[23]Valuation reports dated 26 May 2022 and 22 December 2022, respectively.
The valuers used starkly different methodologies to assess site value. Mr Haines arrived at a value that was far higher than the returned valuation of $6.2 million; Mr Jackson arrived at a value that was far less.
It is important at this point to note that VGV did not attempt to defend the returned site values. The returned valuers were not called to explain the returned valuations or to otherwise assist the Tribunal. Instead, VGV retained Mr Haines to assess the site values of the Land using the methodology set out in the letters of instruction given to him. That methodology involved, first, determining by a particular means whether Landene (that is, the building and works on the Land) was an improvement and then, valuing the Land in the conventional way, including by reference to its highest and best use.
In determining whether Landene was an improvement, Mr Haines emphasised that this involved considering whether Landene added value to the Land. This, he contended, involved comparing:
(a)the value of the Land as vacant (‘scenario one’); and
(b)the value of the Land with existing building and works (‘scenario two’).
According to Mr Haines, if the value in scenario one is higher than the value in scenario two, Landene cannot be an improvement because it does not increase the value of the Land. This means that it is not to be treated as not having been ‘made’ (for the purposes of the definition of ‘site value’). In other words, it is to be included with the Land when determining site value. When carrying out a comparable sales analysis, the relevant sales are sales of improved land.
In the vacant land scenario (scenario one), Mr Haines posited that the Land was unaffected by any heritage controls and treated it as a prospective development site in accordance with Mr Clarke’s planning evidence that a 15 to 17 storey built-form yield was achievable. Mr Haines analysed comparable sales of development land and arrived at values based on two different floor space ratios (one of which took into account a possible 4.5 metre setback from the western boundary). He chose the midpoint between the two ratios, assessing the vacant land value to be $12.825 million as at 1 January 2020.
For scenario two, Mr Haines assessed the value of the Land with the building and works, having regard to comparable sales in what he described as the ‘heritage market’. Mr Haines concluded that there was no evidence to suggest that the sale of Landene in August 2019 was not ‘at market’. As a consequence, he adopted the transaction price of $8.25 million as the non-vacant land value of the Land as at 1 January 2020.
Mr Haines concluded that as scenario one yielded a higher value than scenario two, the building and works on the Land could not be classified as ‘improvements’ because they did not increase the value of the Land. It followed that the site value as at 1 January 2020 was the value of the house and the Land, being $8.25 million.
In his second report, Mr Haines carried out the same exercise of comparing the vacant land scenario with the non‑vacant land scenario. In relation to the former, he again adopted the midpoint of $12.825 million as the value of the vacant land as at 1 January 2021. In relation to the non-vacant land scenario, Mr Haines again looked at contemporaneous sales and had regard to the 2019 sale of Landene in the context of other sales of heritage‑protected properties in inner Melbourne. He again considered the $8.25 million sale price of Landene in August 2019 to be ‘at market’, but recognised that since January 2020, substantial renovations and works had been undertaken to modernise and secure the interior of Landene. Adding $1.285 million in renovation costs to the 2019 sale price, Mr Haines arrived at a figure of $9.535 million for the non‑vacant land scenario as at 1 January 2021.
Once again, since the vacant land figure was higher than the non‑vacant land figure, Landene was not treated as an improvement. Mr Haines assessed the site value of the Land as at 1 January 2021 as $9.535 million.
Of course, this was later acknowledged to be an error as the renovation costs should, on Mr Haines’ analysis, have been deducted from the figure of $8.25 million, rather than added on to it.
Mr Jackson began his analysis at a different point, being the identification of the highest and best use of the Land, on the basis of which he carried out a direct comparison of sales evidence using an improved sales analysis. He then arrived at a value for the Land as vacant by deducting the replacement costs of the building from the value arrived at by the improved sales analysis, ‘with appropriate adjustments made to reflect the condition of the improvements at the time of sale’.
According to Mr Jackson, the highest and best use of the Land was the existing two‑storey building (Landene), with minor potential extension to the rear boundaries. He considered that Landene provided ‘considerable added value to the Land and that the sale price [in 2019] reflected the added value’. To determine the underlying value of the Land, he deducted from the $8.25 million sale price the replacement costs of the building, which were adjusted to reflect its condition at the time of sale.
At the time of his first report in May 2022 (which concerns the 2020 valuation), the replacement costs were $6,658,103. This left a rounded figure for site value as at 1 January 2020 of $1.59 million.
Mr Jackson also carried out a hypothetical development analysis, which assumed that the Land contained the building with heritage restrictions, but in new condition. He determined the ‘gross realisation value’ of the Land and deducted from that figure the construction costs, selling costs, holding costs, acquisition costs (et cetera), which he said enabled him to arrive at the underlying site value. Using the hypothetical development approach, Mr Jackson assessed the site value as at 1 January 2020 as $1.38 million. This was said to provide support for the figure arrived at using the improved sales analysis.
Mr Jackson’s report concerning the 2021 valuation (in P10049/2022) was based on an updated quantity surveyor’s report for the replacement costs of the building, which had increased to $6,911,078. This resulted in a rounded site value of $1.34 million as at 1 January 2021. Again, this figure was said to be supported by a hypothetical development analysis, which produced a rounded site value of $935,000 as at 1 January 2021.
In this report, Mr Jackson also adjusted the inputs relating to the development costs as at 1 January 2020 (for the purpose of the hypothetical development analysis originally given in the 2020 valuation) in light of the updated quantity surveyor’s report. This produced a revised site value of $1.285 million for 1 January 2020 (compared to the previous figure of $1.38 million).
In his reply report dated March 2023, Mr Haines summarised the differences between his approach and that of Mr Jackson as follows:
My approach and [Mr Jackson’s] differ as to how buildings and works on the land are designated (or not) as ‘improvements’ as defined in Section 2(1) of the Valuation of Land Act 1960, and consequently whether any such buildings and works are to be assumed to have never been made as is required in assessing Site Value. The parties’ valuation methodologies diverge accordingly.
I employ the Direct Comparison approach, having first ascertained whether the buildings and works on the land constitute ‘Improvements’: My analysis concluded that by comparison to the subject land’s value assuming those buildings and works are absent, the property’s value with the intact buildings and works (‘non-vacant’) is lower. As a consequence, the buildings and works cannot be said to increase the value of the land and do not therefore constitute ‘Improvements’ under the relevant definition contained in the Valuation of Land Act 1960. Accordingly, assessment of Site Value at both relevant dates must not disregard the buildings and works, nor the site‑specific Heritage Overlay designation that applies to the land.
Mr Jackson for [WSTI] illustrates two methodologies for each Relevant Date, both of which embrace a hypothetical non‑vacant selling price, deducting a building replacement cost/development cashflow to arrive at a ‘residual’ assessment of Site Value. Both methods automatically assume buildings constitute ‘Improvements’.
As [WSTI’s] two ‘methods’ do not adequately investigate whether the subject property’s buildings are ‘Improvements’ I am unmoved in my approach articulated in my Expert Witness Statements [sic] dated December 2022.
The Tribunal’s decision
The Tribunal identified the ‘focal point’ of the dispute in this case to be ‘the correct approach to determining site value of land which is subject to a site-specific heritage overlay’.[24] It identified five questions for determination:[25]
(a)What is the correct construction of the VLA and thus the correct approach to the assessment of site value?
(b)What is the effect of the site-specific heritage overlay on the subject property?
(c)Does the building on the subject property ‘add value’?
(d)Which valuation method is most appropriate in the circumstances?
(e)Is the VGV’s evidence of a higher value than the returned valuation a new valuation and is it relevant?
Tribunal’s approach to determining site value
[24]WSTI Properties 490 SKR Pty Ltd v Valuer-General Victoria (Red Dot) (Land Valuations) [2023] VCAT 734, [34] (‘Reasons’).
[25]Ibid [35].
The Tribunal held that the correct approach to determining the site value of the Land was to first apply s 5A of the VLA, which ‘is the primary guiding provision for determining the value of land’.[26] The Tribunal considered to be ‘untenable’, VGV’s contention that the starting point is to ask whether the ‘improvements’ add value because the answer to that question is that ‘it depends’ on the factors set out in s 5A.[27]
[26]Ibid [105].
[27]Ibid [121].
The Tribunal recognised that while s 5A(3) sets out six factors that a valuer must, where relevant, take into account, it provides no express guidance as to sequencing or weight to be given to those factors.[28] However, the Tribunal was of the view that
it would be logical to consider for example, the specific attributes of the subject property — such as location, land area, topography, shape, extent and condition of improvements, planning controls, other instruments etc. — before being able to consider the HBU [highest and best use of the Land]. This would then enable the consideration of the suitability of any improvements, which can only be made in the context of the HBU. This in turn would inform what sales might be appropriate (at the relevant time) and the degree of comparability of those sales.[29]
[28]Ibid [119].
[29]Ibid.
The Tribunal concluded that the correct approach to determining the site value of land that is subject to a site-specific heritage overlay is to follow four steps, each one of which informs the step that follows:[30]
132.1Firstly, to ascertain the specific attributes of the subject property such as location, land area, topography, shape, extent and condition of improvements, planning controls, other instruments, etc;
132.2Secondly, this would then enable the valuer to consider the HBU [highest and best use of the Land] and the suitability of any improvements, which can only be made in the context of the HBU;
132.3This would then inform the third step in ascertaining what sales might be appropriate (at the relevant time) and the degree of comparability of those sales; and
132.4 Finally, the statutory definitions can be assessed and determined.
[30]Ibid [132].
As to the principal cases referred to by the parties that set out the contrasting approaches to the sequencing involved in determining site value — Commonwealth Custodial Services Ltd (as Trustee for Burwood Trust Fund) v Valuer-General (NSW),[31] Trust Company of Australia Ltd v Valuer-General,[32] and Caltex Oil (Australia) Pty Ltd v Chief Executive, Department of Lands[33] — the Tribunal considered the approach in Caltex Oil to be ‘most similar’ to the approach that was required ‘in this case and in the context of the legislation of this jurisdiction’.[34]
[31](2006) 148 LGERA 38; [2006] NSWLEC 400 (‘Commonwealth Custodial’).
[32](2007) 154 LGERA 437; [2007] NSWCA 181 (‘Trust Company’). Trust Company is the appeal to the New South Wales Court of Appeal from one of the appeals in Commonwealth Custodial.
[33](1996) 16 QLCR 435 (‘Caltex Oil’).
[34]Reasons, [122].
In Caltex Oil, the owner sought to reopen the annual valuation of a plot of land that it owned in the parish of Yeerongpilly. The land had been developed for service station purposes and, before the relevant valuation date of 31 March 1992, had been entered as a ‘probable site’ in the Contaminated Sites Register maintained under s 24 of the Contaminated Land Act 1991 (Qld).[35] As between the parties, it was agreed, among other things, that the highest and best use of the land was as residential land and that the figure of $135,000 was the starting point for determining whether the unimproved value of the land should be reduced by reason of the operation of the Contaminated Land Act.[36] The owner contended that in arriving at the land’s value, the relevant zoning of the land and other statutory restrictions affecting the land ought to be considered.[37]
[35]Caltex Oil, 440–1 (Fryberg J and Mr Neate).
[36]Ibid 441–2.
[37]Ibid 452.
In the course of their reasons, the majority observed that an ‘improvement’ was something that added value to the land and said:
It cannot be said whether a structure on or other quality of land amounts to an improvement in this sense until it is known whether the structure or quality enhances the value of the land. Any enhancement in turn will usually depend upon whether the structure or quality advances or inhibits a particular use. Thus, whether the structure or quality constitutes an improvement cannot ordinarily be known until the highest and best use of the land is determined.[38]
[38]Ibid 458.
To reason otherwise, the majority said, was to assume the conclusion.[39]
[39]Ibid.
According to the Tribunal, Caltex Oil supports the view that the highest and best use of the land will inform the question of whether the improvements add value to that use.[40]
Effect of site-specific heritage overlay
[40]Reasons, [123]–[124].
The Tribunal rejected VGV’s argument that Landene has a ‘chilling effect’ on land value because its removal would diminish the effect of the site-specific heritage overlay, such that a very significant development opportunity would arise.[41] According to the Tribunal, the fallacy in VGV’s argument was that it equated ‘the theoretical assumption of removal of the building, for the purposes of site value assessment, as being the same as if the building were in fact not there’.[42]
[41]Ibid [136]–[137].
[42]Ibid [137].
The Tribunal held that a ‘realistic’ as opposed to a ‘highly hypothetical approach’ ought to be adopted when considering the effect of the site-specific heritage overlay.[43] It held that the effect of the planning control was to be taken into account, rather than ‘the theoretical removal of the building to effectively retrofit an argument that no building equates to no heritage control’.[44]
Whether the building adds value
[43]Ibid [138].
[44]Ibid.
The Tribunal determined that there was unequivocal evidence before it to indicate that Landene added value to the Land and was therefore an improvement for the purpose of determining site value.[45] The correct approach to ascertain if ‘improvements’ add value is
to compare the market value in its full current state, that is, including its limited development potential and existing structures with the market value that the land would have with the same development potential, without the buildings on the land.[46]
Valuation method
[45]Ibid [153].
[46]Ibid [162].
After considering the valuation methodologies adopted by Mr Haines (for VGV) and Mr Jackson (for WSTI), the Tribunal rejected[47] Mr Haines’ methodology of comparing an unencumbered site, being vacant land where no heritage controls apply, with an encumbered one, with the existing building and works in place and the relevant heritage controls in force.
[47]Ibid [214].
Following a careful examination of Mr Jackson’s valuation methodology and after making adjustments to his calculations, the Tribunal determined the site value of the Land to be $2.925 million at each of the relevant dates.[48]
Whether VGV’s evidence of a higher value is a new valuation
[48]Ibid [262].
The Tribunal stated that its role was to stand in the shoes of the (returned) valuer and to determine, on the merits of the evidence before it, whether the valuer’s decision on the objection was the correct or preferable decision.[49] The Tribunal referred to its earlier decision in S & JG Investments Pty Ltd v Valuer-General Victoria (Land Valuation),[50] in which it determined that ‘evidence of a higher valuation must be in addition and not in substitution of the returned valuation’.[51]
[49]Ibid [265], [267].
[50][2023] VCAT 246 (‘S & JG Investments’).
[51]Reasons, [270].
Having considered the new valuation evidence led by VGV, the Tribunal concluded that Mr Haines’ evidence was ‘an entirely new valuation, untethered in any way to the returned valuation’.[52] The Tribunal held that although VGV could have contended for a new valuation by an independent expert to demonstrate that the returned valuation was not too high, this should only be done in addition to the evidence of the valuer who made the returned valuation and not in substitution for the evidence required to review the decision.[53]
Tribunal’s conclusions
[52]Ibid [282].
[53]Ibid.
After addressing the five issues that arose for determination, the Tribunal returned to the three-step process that it said it would undertake in review applications brought under s 22 of the VLA.[54] Having undertaken these steps, the Tribunal determined that the site value of the Land as at 1 January 2020 and 1 January 2021 was $2.925 million.[55] Comparing this figure with the returned site valuation of $6.2 million, the Tribunal concluded that the returned valuation was ‘too high’,[56] which meant that the valuation ought to be reduced pursuant to s 25(1)(a) of the VLA.[57]
[54]Ibid [283], citing S & JG Investments.
[55]Ibid [284].
[56]Ibid [285].
[57]Ibid [286], [293].
Proposed grounds of appeal
VGV now seeks leave to appeal the Tribunal’s decision on the following grounds:
1.The Tribunal erred in holding that the highest and best use must be identified before determining what constitutes an improvement.
2.The Tribunal erred in its analysis of the legislative history.
3The Tribunal erred in finding that VGV’s approach required the identification of two highest and best uses, and otherwise misunderstood the effect of the heritage overlay on the different steps in the valuation process.
4.The Tribunal erred in finding that the building was an improvement.
5.The Tribunal misapprehended its task on review.
6.The Tribunal erred in finding that VGV’s valuation evidence was irrelevant or impermissible.
Grounds 1, 2, 3 & 4 — Assessing the site value of land
It is convenient to consider grounds 1 to 4 together as they address the fundamental question of the correctness of the Tribunal’s finding that the building and works (Landene) were ‘improvements’ for the purpose of determining the site value of the Land.
VGV’s submissions
As discussed, VGV submits that based on the text of the VLA, to determine site value, one must first identify what constitutes ‘improvements’ (step 1), and then notionally remove those ‘improvements’ (step 2) for the purpose of determining site value (step 3). Only after the completion of steps 1 and 2 is it clear what must be valued. A valuation can then be carried out in accordance with s 5A of the VLA.
According to VGV, step 1 does not involve the valuation of land; it does not require an assessment of how much more (or less) the land is worth with or without the works or material. Whatever the highest and best use of the land might be for the purpose of assessing site value, it has no part to play in determining whether work done or material used on the land qualify as an improvement.
VGV submits that there is no dispute that the Land without the building (which VGV
described to us as being ‘in its natural state’) is worth considerably more than the Land with the building. In this case, the heritage overlay does not restrict intensive development other than in relation to the building. It is the presence of the building that creates the barrier to greater development and suppresses the value of the Land. Because the building suppresses value, it cannot be treated as an improvement when assessing site value.VGV does not contend that the heritage overlay is to be ignored when assessing site value. To the contrary, it must be taken into account. However, it is only once steps 1 and 2 have been completed that the valuation process commences. Determining ‘site value’ is step 3 in the sequence of tasks and this step is distinct from the identification of ‘improvements’ to the land. VGV contends that the ‘highest and best use’ of the Land only comes into consideration when the valuation process proper commences, that is, as part of step 3.
VGV relies on Maurici v Chief Commissioner of State Revenue,[58] Commonwealth Custodial,[59] Trust Company,[60] and Surfers Paradise Resort Hotel Pty Ltd v Department of Natural Resources and Water[61] to support the proposition that it is necessary to first identify any ‘improvements’ on the land before carrying out the site valuation. According to VGV, to determine the highest and best use of the Land as the first step in the process — as the Tribunal did — is erroneous, given that such an approach precludes the comparison of the two scenarios. That comparison, it argues, elucidates the effect of the building on the land value.
[58](2003) 212 CLR 111; [2003] HCA 8 (‘Maurici’).
[59](2006) 148 LGERA 38; [2006] NSWLEC 400.
[60](2007) 154 LGERA 437; [2007] NSWCA 181.
[61](2007) 163 LGERA 14; [2007] QLAC 0127 (‘Surfers Paradise’).
Furthermore, VGV argues, to approach the exercise otherwise than according to the sequencing of steps 1, 2 and 3 would frustrate the legislative purpose of the State Taxation Acts Amendment Act2019, which repealed ss 2(8) and 2(9) of the VLA. The express purpose of the repeal of those provisions was to overcome the effect of the decision in the GPO case. The Tribunal’s construction would defeat that stated purpose because by adopting, as the starting point for the assessment of site value, the highest and best use of heritage-constrained land, and then deducting the cost of the reconstruction of a building on the land, the result in the GPO case would not be altered.
WSTI’s submissions
WSTI argued that there are three reasons why grounds 1 to 4 do not reveal error on the Tribunal’s part:
(a)First, VGV has impermissibly raised an argument on appeal that it did not advance in the Tribunal, namely, that the ‘land in its natural state’ is that to which the building and works (Landene) must add value in order to qualify as an improvement.
(b)Secondly, in any case, VGV led no evidence in the Tribunal going to whether Landene is an improvement based on the ‘natural state’ analysis now advanced.
(c)Thirdly, irrespective of which stream of case law is applied — whether it is the ‘natural state’ body of case law or the ‘highest and best use’ body of case law — the Tribunal’s determination that the building constitutes an improvement was not erroneous.
According to WSTI, the definitions of ‘site value’ and ‘improvements’ in the VLA, together with s 2(2), require the assessment of the impact of the ‘work done or material used’ on land value. It is not possible to conduct such an assessment without considering the highest and best use of the Land. The concept of highest and best use is implicit in the assessment of value based upon the Spencer test, and has been described as ‘a fundamental principle of valuation’.[62] It was common ground that s 5A of the VLA, which frames the methodology for determining the value of land, is to be taken into account when determining ‘site value’. The operation of s 5A, WSTI argues, is not limited to any particular step in the assessment of ‘site value’ and there is nothing in the definition of ‘improvements’ that excludes the application of s 5A at that stage of the enquiry.
[62]Citing ISPT Pty Ltd v Melbourne City Council (2008) 20 VR 447, 458 [37]–[38] (Warren CJ, Kellam JA and Osborn AJA); [2008] VSCA 180; Challenger Property Asset Management Pty Ltd v Stonnington City Council (2011) 34 VR 445, 465 [43] (Croft J); [2011] VSC 184.
WSTI submits that this position is supported by well-established authority.[63] In contrast, the identification and comparison of two highest and best uses of land is an unorthodox approach to valuation and is contrary to authority.
[63]Citing Australian Postal Commission v Melbourne City Council (2005) 14 VR 678, 683 [17], 684 [20] (Charles and Nettle JJA); [2005] VSCA 295; Pattas v Stonnington City Council [2010] VSC 487; F.A. Springall Pty Ltd v Boroondara CC (Land Valuation) [2011] VCAT 388.
WSTI further submits that on the correct application of Commonwealth Custodial and Trust Company, there can be no doubt that the building on the Land is an improvement. The relevant question, identified in those cases, is ‘whether the improvements enhance the land’s value compared with its natural state’. The enhancement of ‘value’ should not be conflated with the enhancement of a particular use (including VGV’s hypothetical 15 to 17 storey residential tower). It is sufficient if a purchaser has a use for the ‘improvements’.
According to WSTI, Mr Haines’ approach does not address the question of ‘whether the improvements enhance the land’s value compared with its natural state’. His reports do not identify whether a purchaser may have a use for the ‘improvements’, nor whether the ‘improvements’ enable an existing use of the Land to be continued.
WSTI also argues that VGV’s valuation evidence contains a number of flaws:
(a)It compares an unencumbered site with an encumbered site, which is artificial and has the consequence that a building would never add value in an intensive development zone.
(b)Its reliance on other heritage sales on an ‘all in’ basis was based on the incorrect assumption that the Land and building were indivisible.
(c)It is inconsistent to conclude, as Mr Haines did, that the renovations to Landene added $1.285 million in value and were accordingly ‘improvements’, but the rest of the building was not.
According to WSTI, Landene is a substantial and functional heritage building in excellent repair and condition. As the Tribunal found, the building has value to potential purchasers, including for beneficial occupation and long-term capital growth. As Mr Jackson opined, the building is structurally intact and has value to a potential purchaser. Even Mr Haines acknowledged that the building was desirable and highly functional. Any of those matters is sufficient for the purposes of the tests in Commonwealth Custodial and Trust Company. There is no realistic suggestion that Landene does not add value to the Land compared with its natural state.
Consideration
Grounds 1 to 4 all concern, in substance, how the site value of the Land as at the relevant dates was required to be determined under the VLA. More specifically, grounds 1 to 4 challenge the Tribunal’s rejection of VGV’s approach to determining whether Landene is an improvement for the purpose of assessing the site value of the Land.
Apart from ground 3, which addresses the Tribunal’s understanding of VGV’s submissions, the grounds of appeal raise for consideration the proper construction of the provisions of the VLA governing the assessment of site value. Thus, ground 1 concerns the sequencing of steps required by the definition of ‘site value’. Ground 4, said to be consequential on the other grounds, simply alleges that the Tribunal misconstrued the definition of ‘improvements’ to arrive at the wrong conclusion about whether Landene was an improvement. Ground 2 strikes more generally at the construction adopted by the Tribunal, on the basis that it frustrates the legislative intention behind the repeal of ss 2(8) and 2(9) of the VLA.
As to ground 3, we accept that the Tribunal may have conflated the steps identified by Mr Haines when it found that his valuation methodology required the establishment of two highest and best uses for the Land at each of the relevant dates. In fact, Mr Haines sought to separate the task of identifying ‘improvements’ on the Land from the task of assessing the site value of the Land (by applying s 5A and the Spencer analysis). The latter task — the valuation task — involves identifying a highest and best use of the Land, which Mr Haines did by reference to all of the existing planning controls, including the heritage overlay.
However, nothing flows from any such conflation.
It is true that, technically, Mr Haines arrived at values for the Land on different bases in scenarios one and two. He carried out a comparable sales analysis for each scenario, having regard to different highest and best uses of the Land. However, this analysis merely states the obvious. Mr Haines might just as well have simply asserted that the Land would attract a higher purchase price if its use and development was not constrained by one of the applicable planning controls — the heritage overlay. So much is self-evident. But, the central question is whether it was legitimate to conclude that Landene is not an improvement because it does not relevantly add value to the Land as the Land would be worth more without the heritage overlay.
Mr Haines attempted to establish that Landene did not increase the value of the Land by comparing the value of the Land assuming that Landene did not exist with the value of the Land containing Landene. This exercise was premised on the proposition that without Landene, the heritage overlay, which prevents the development of the Land to its full potential under the Commercial 1 zoning, had no work to do and was irrelevant. In fact, Mr Haines was simply comparing the value of the Land with and without the heritage overlay, which he tied to the existence of the building. That assessment took place on a highest and best use basis, using two highest and best uses of the Land — one with the heritage overlay in play, the other without.
The Tribunal considered VGV’s position to contain a logical fallacy. According to the Tribunal, the ‘key element’ in ascertaining the role that Landene plays in enhancing or diminishing the value of the Land is the Land’s highest and best use. This is because the suitability or value of any ‘improvements’ can only be assessed having regard to the Land’s highest and best use. That highest and best use is to be determined by reference to the planning controls in force as at the relevant dates. Mr Jackson’s valuation method correctly treated Landene as an improvement because Landene added value to the Land when viewed in the context of the applicable planning controls, including the heritage overlay.
The contest was therefore formulated in the Tribunal as one of sequencing. VGV argued its case in terms of what it said was the correct sequencing when applying the definition of ‘site value’. In this respect, VGV relied on two principal authorities — Maurici[64] and Trust Company[65] — to contend that the first step to be taken when assessing site value is to identify the ‘improvements’ to the land, and that this step must be taken before considering what the highest and best use of land is.
[64](2003) 212 CLR 111; [2003] HCA 8.
[65](2007) 154 LGERA 437; [2007] NSWCA 181.
In Maurici, the High Court considered the assessment of ‘land value’, which is the New South Wales equivalent to ‘site value’. Section 6A of the Valuation of Land Act 1916 (NSW) (‘NSW Act’) relevantly provided that the ‘land value’ of land is
the capital sum which the fee‑simple of the land might be expected to realise if offered for sale on such reasonable terms and conditions as a bona-fide seller would require, assuming that the improvements, if any thereon or appertaining thereto, other than land improvements, and made or acquired by the owner or the owner’s predecessor in title had not been made.
As is apparent, s 6A is very similar to the definition of ‘site value’ in the VLA. However, the NSW Act, unlike the VLA, does not contain a definition of ‘improvements’. This has some significance, which we explain below.
In Maurici, the High Court said this about the application of s 6A:
The first step to be taken under s 6A is to identify what is capable of being regarded as improvements, ‘other than land improvements’. The second step is notionally to remove the improvements from the land. It is at the third point that difficulties arise. How is the land in its notionally unimproved state to be valued?[66]
[66]Maurici, 120 [16] (McHugh, Gummow, Kirby, Hayne and Callinan JJ).
Maurici was concerned with the ‘third point’, namely, how to value land in its notionally unimproved state. The High Court observed that the traditional and usually unexceptionable method was to seek out relatively contemporaneous sales of comparable properties and to use those sales as a yardstick for the valuation of the relevant land.[67] The High Court held that the method adopted by the respondent in that regard was ‘unduly selective’ in that it used, almost exclusively, sales of unimproved (vacant) land in circumstances where these sales were not representative of sales in the area because vacant land was scarce, if not very scarce, in Hunters Hill,[68] and held that in the circumstances, the respondent was required to also look at reasonably contemporaneous sales of improved land in the area.[69]
[67]Ibid.
[68]Ibid 120 [17].
[69]Ibid 121 [19].
The High Court in Maurici was concerned with the identification of comparable sales for the valuation of unimproved land. In our view, Maurici is of limited assistance in this case.
Trust Company is a decision of the New South Wales Court of Appeal, once again concerning ‘land value’ within the meaning of s 6A in the NSW Act. The land for which the ‘land value’ (site value) fell to be determined was a parcel of land in the Sydney suburb of Ashfield.[70] At the relevant time, the Ashfield land contained two five‑storey office buildings that did not conform to the highest and best use of the land; under the land’s mixed-use zoning, the highest and best use was a residential tower. It was common ground that if the Ashfield land was to be developed for mixed use, the existing office buildings could not be converted to that use and would have to be demolished.[71]
[70]Trust Company, 441 [3].
[71]Ibid 442 [9].
The appellant, the owner of the Ashfield land (‘owner’), contended that in deciding whether particular structures are improvements, one should ask whether they make land better for the purpose of the highest and best use.[72] The owner argued that because the presence of the office buildings on the site detracted from its value as a site for mixed use, there needed to be deducted from the value found by the trial judge amounts representing the cost of demolition and removal of debris, as well as interest or holding charges.[73]
[72]Ibid 442–3 [12].
[73]Ibid 443 [14].
This circumstance gave rise to the following question of law:
if at the base date for valuation a parcel of land has structures on it, but those structures are not suited for the best and highest development of the land and would need to be demolished if the best and highest development were to be achieved, are those structures ‘improvements’ within the meaning of the definition of ‘land value’ in s 6A of the [NSW] Act?[74]
[74]Ibid 442 [9].
The trial judge (Biscoe J) held that the office buildings were improvements and the Court of Appeal upheld his Honour’s decision.[75]
[75]Ibid 453 [68] (Campbell JA, Beazley JA agreeing at 440 [1], Ipp JA agreeing at 440 [2]).
While there is no definition of ‘improvements’ in the NSW Act, Campbell JA (with whom Beazley P and JA agreed) considered that there was ample recognition in the case law of the need for an improvement to ameliorate the land,[76] citing McGeoch v Federal Commissioner of Land Tax,[77] and Commonwealth v Oldfield,[78] in which Jacobs J (with whom Gibbs, Stephen and Mason JJ agreed) quoted with approval the following passage from the judgment of Griffith CJ in Morrison v Federal Commissioner of Land Tax:
What operations of man are improvements? When I say ‘operations of man’, I think the term should be limited to what is done by the owner for the time being, that is, after the land has ceased to be Crown land. Any operation of man on land which has the effect of enhancing its value comes within the definition of ‘improvement’.[79]
[76]Ibid 445 [25]–[26].
[77](1929) 43 CLR 277.
[78](1976) 133 CLR 612, 619.
[79](1914) 17 CLR 498, 503 (‘Morrison’).
Thus, although the NSW Act contains no statutory equivalent to the definition of ‘improvements’ in the VLA, the requirement that the works or materials in question ‘enhance’ or ‘increase’ the value of the land is common. In both cases, the effect of the improvement must be to increase or enhance the value of the land.
In response to the owner’s submission, Campbell JA first confirmed that while the NSW Act contains no express recognition of the concept of ‘highest and best use’, it was well established that the highest and best use of land was to be used for the purpose of ascertaining its value. His Honour then addressed the question of sequencing when determining ‘land value’:
It is particularly important, for present purposes, that there is a particular order of operations that s 6A requires to occur in ascertaining the ‘land value’. In carrying out the thought task that s 6A(1) calls for, first, the ‘improvements’ other than land improvements are notionally removed. Only then does the notional sale occur. And it is by reference to that notional sale that the highest and best use is determined. Thus, it is necessary to determine the identity of the improvements that are to be removed before the highest and best use can be ascertained.[80]
[80]Trust Company, 446 [33] (emphasis in original).
Campbell JA said further:
According to the [owner’s] submission, the identity of the improvements is ascertained by reference to what is the highest and best use. That requires the highest and best to be known before the identity of the improvements is ascertained. But the temporal sequence of events that s 6A requires simply does not enable that to occur.[81]
[81]Ibid 447 [36].
VGV relies on these passages to argue that the Tribunal was wrong to hold that the assessment of site value must commence by determining the highest and best use of the land in question, and as support for its preferred methodology of first identifying what qualifies as an improvement and notionally removing it from the land before carrying out the valuation.
However, in Trust Company, the New South Wales Court of Appeal endorsed the approach to identifying improvements articulated by the trial judge, Biscoe J, in Commonwealth Custodial.[82] In our view, this approach is not reflected in the methodology deployed by Mr Haines.
[82]See ibid 443 [17], 453 [68].
In Commonwealth Custodial, Biscoe J considered two appeals, one in relation to land in Burwood brought by Commonwealth Custodial Services Ltd, the second concerning the Ashfield land. His Honour described the submissions of the parties as to the meaning of ‘improvements’ in s 6A of the NSW Act as follows:
The Valuer-General’s submission is that all structures on land are improvements. The [owner’s] submission is that ‘improvements’ are any human operations on land which have the effect of enhancing its value for its highest and best use.[83]
[83]Commonwealth Custodial, 51 [39].
Biscoe J rejected both submissions.[84] His Honour said:
In my opinion, ‘improvements’ within the meaning of s 6A(1) of the [NSW] Act are human operations of persons on land which have the effect, as at the date of valuation, of enhancing its value compared with its natural state. A structure on land is not an improvement if it does not enhance the land’s value compared with its natural state. The commercial buildings on the Ashfield Land enhance the land’s value compared with its natural state. They are valuable to a purchaser who has a use for them.[85]
[84]Ibid 51 [41].
[85]Ibid 51 [42] (emphasis added).
In rejecting the proposition that improvements had to add value to the land for the purpose of its highest and best use, Biscoe J said that s 6A ‘should not be construed so as to attribute to the word “improvements” a meaning which changes according to the use of the land’.[86] The meaning that he favoured — something that enhances the value of land compared to its natural state — was consistent with the statutory definition of ‘land improvements’ in s 4(1) of the NSW Act,[87] namely, such things as vegetation removal, drainage and other earthworks.[88] Biscoe J held that ‘[a]ll of the activities within that definition appear[ed] to have as a common element the notion of rendering land in a better state than its natural state’, which ‘tend[ed] to support the conclusion that the same common element exist[ed] in respect of the undefined term “improvements”’.[89] His Honour observed that the ‘land improvements’ listed in s 4(1) appeared to be destructive or invisible improvements such as clearing the land; as distinct from constructive or visible improvements, such as buildings’, which fell to be considered in the context of the undefined term ‘improvements’.[90] It was also consistent with the meaning given to improvements by the High Court in Morrison.[91]
[86]Ibid.
[87]Section 4(1) of the NSW Act defines ‘land improvements’ to mean:
(a)the clearing of land by the removal or thinning out of timber, scrub or other vegetable growths,
(b) the picking up and removal of stone,
(c) the improvement of soil fertility or the structure of soil,
(d) the restoration or improvement of land surface by excavation, filling, grading or levelling, not being works of irrigation or conservation,
(d1) without limiting paragraph (d), any excavation, filling, grading or levelling of land for the purpose of the erection of a building, structure or work, not being for the purpose of irrigation or conservation,
(e) the reclamation of land by draining or filling together with any retaining walls or other works appurtenant to the reclamation, and
(f) underground drains.
[88]Commonwealth Custodial, 52 [44].
[89]Ibid.
[90]Ibid.
[91]Ibid 52 [45].
It will be recalled that in Morrison, Griffith CJ described ‘improvements’ as ‘[a]ny operation of man on land which has the effect of enhancing its value’, by which his Honour meant compared with the ‘primitive condition’ of the land, illustrated by ‘a jungle so dense that it is almost impenetrable by four-footed animals, even the smallest’.[92] Thus, in Morrison, the operations of clearing timber and tussocks were held to be improvements because they increased the value of the land from its natural state.[93]
[92]Morrison, 503.
[93]Ibid 504–5 (Griffith CJ, Barton, Isaacs, Powers and Rich JJ agreeing at 505).
Biscoe J gave detailed consideration to the owner’s submission that the office buildings on the Ashfield land were ‘worsements’,[94] these being ‘the antithesis’ of improvements.[95] Biscoe J accepted that a ‘worsement’ would have to be taken into account in quantifying site value under s 6A(1).[96] His Honour gave, as an example of a ‘worsement’,[97] the condition of land in Kiddle v Deputy Federal Commissioner of Land Tax,[98] where a pastoral property consisted of land that had been partly improved to its full capacity and partly rendered practically useless because, having once been partially improved by ringbarking, the process of destruction had not been continued. The cost of improving the land by destroying the timber would have been at least equal to the value of the land when so improved.
[94]See Commonwealth Custodial, 57–9 [67]–[77].
[95]Ibid 57 [68], citing Brisbane City Council v Valuer-General for the State of Queensland (1978) 140 CLR 41, 51; [1978] HCA 40, in which Gibbs J said a ‘worsement’ is that to which ‘something done on or appertaining to land which reduces rather than enhances its value is not an improvement for the purposes of the Act, any more than it would be in the ordinary sense of the word’.
[96]Commonwealth Custodial, 57 [68].
[97]See ibid 58 [70].
[98](1920) 27 CLR 316.
Biscoe J did not accept that the buildings on the Ashfield land were worsements, even though they did not represent the highest and best use of the land. This was because they added value to the Ashfield land, as at the valuation date, compared with its value in its natural state.[99]
[99]Commonwealth Custodial, 58 [71].
VGV also referred the Court to Surfers Paradise, where three members of the Queensland Land Appeal Court considered a similar submission to that advanced by the owner of the Ashfield land. Surfers Paradise considered the site value of land occupied by a five-star hotel that was not the highest and best use of the land. The parties agreed that if the land was in its natural or vacant state, its highest and best use would be for high-rise residential development and that its value on that basis would be $31 million.[100] They agreed that the hotel could not be successfully converted to that use and that the cost of demolition of the hotel and the termination of an associated management agreement would be $8.3 million.[101] They also agreed that the value of the land as it stood with the structures on it (ie, the hotel building and facilities) would exceed $31 million.[102] It had sold on dates before and after the relevant date for $59 million and $64 million, respectively.[103]
[100]Surfers Paradise, 17 [4] (White J and QLC Members Scott and Jones).
[101]Ibid.
[102]Ibid.
[103]Ibid.
The hotel owner submitted that ascertaining the unimproved value of the land involved ascertaining its highest and best use if there were no structures on it and then asking whether the structures added value for the highest and best use.[104] If not, they could not be described as ‘improvements’, the land would not be ‘improved land’ and should be valued ‘as it exists’.[105] That would be done by deducting the costs of $8.3 million from the land value of $31 million to produce an unimproved value of $22.7 million.[106] The approach proposed by the hotel owner was, in effect, one whereby the hotel building would be characterised as a ‘worsement’, whereas the respondent treated it as an improvement.[107]
[104]Ibid 17 [6].
[105]Ibid.
[106]Ibid.
[107]Ibid 18 [11].
The Queensland Land Appeal Court rejected the hotel owner’s submission. Applying Morrison, it held that ‘improved land is land upon which operations of man have taken place which enhance the value of the land to the extent that its value is greater than it would have been had no such operations taken place’.[108] The Court went on to say that the hotel owner’s approach ‘confuses the requirement to identify whether the land is improved or not by introducing a requirement to identify the highest and best use of the land first – a step not required until the valuation process is being undertaken’,[109] and that they agreed with the reasoning in Commonwealth Custodial and Trust Company as to why this approach was to be rejected.[110] The Court ultimately held that ‘[a]t a practical level, the question whether a structure or other work of man is an improvement is answered by asking whether a prudent purchaser would pay a higher price for the land together with the structures than he would to put the land to a use which rendered the structures redundant’.[111] On the facts of that case, the land was worth $31 million without the hotel, but much more with the hotel.[112] The hotel was therefore an improvement.
[108]Ibid 18 [10].
[109]Ibid 18–9 [12].
[110]Ibid 19–20 [13]–[14].
[111]Ibid 21 [18].
[112]Ibid.
In Trust Company and Surfers Paradise, the office buildings and the hotel respectively, while not advancing the highest and best use of the lands in question, were nonetheless ‘improvements’ because each increased the value of the lands on which they stood by advancing or accommodating a use. Although these buildings did not attract a planning control that inhibited the development of the respective lands to their full potential, the physical presence of the buildings did so, which is why the owners sought to have the site value reduced to incorporate the costs of demolition. However, both the New South Wales Court of Appeal and the Queensland Land Appeal Court accepted that while the buildings or structures may have inhibited the development of the lands in accordance with their highest and best use, this did not mean that they did not enhance the value of the lands.
In our view, these authorities are not helpful to the case that VGV seeks to make based on the methodology used by Mr Haines. As discussed, in the Tribunal, VGV expressly disavowed any attempt to identify an improvement by reference to the Land in its ‘natural state’ or to rely upon the approach in Commonwealth Custodial, and cautioned the Tribunal against reliance on authorities considering the NSW Act. Moreover, VGV led no evidence as to whether Landene was an improvement relative to the ‘natural state’ of the Land. It did not seek to adduce any evidence relevant to this form of analysis, and nor did it suggest that Mr Haines’ evidence engaged with it. In fact, Mr Haines’ method for identifying whether Landene was an improvement asked whether Landene added value to the Land having regard to its highest and best use, but assuming that the heritage overlay did not exist.
As a consequence, the Tribunal did not engage with the ‘natural state’ test either. However, it postulated that for works or materials to effect an ‘increase in value’ of land, there must be a way of identifying the value of the land without the works or materials. Any ‘increase’ in value is relative. In the Tribunal’s view, the base value in respect of which there may be an increase in value due to works or materials is a product of the specific attributes of the land — its location, land area, topography and shape — as well as the planning controls that apply to it. This base value, by including the effect of the planning controls, reflects its highest and best use.
The statement in Caltex Oil upon which the Tribunal relied, that ‘[i]t cannot be said whether a structure on or other quality of land amounts to an improvement … until it is known whether the structure or quality enhances the value of the land’,[113] may be considered benign so far as it goes. However, in that case, the Queensland Land Appeal Court went on to observe that ‘[a]ny enhancement in turn will usually depend upon whether the structure or quality advances or inhibits a particular use’ and that, as a consequence, ‘whether the structure or quality constitutes an improvement cannot ordinarily be known until the highest and best use of the land is determined’.[114] We are not persuaded that the second proposition necessarily follows from the first. The analysis moves from asking whether a particular use is advanced or inhibited, to postulating that it is necessary to determine the highest and best use of the land in order to answer this question. The question posed in the ‘natural state’ cases is whether a use is advanced, not any particular use and not necessarily the highest and best use.
[113]Caltex Oil, 458, cited in Reasons, [124].
[114]Ibid.
The cases decided in other jurisdictions to which we were referred, and the analyses of improvements to land in its ‘natural state’ undertaken by courts in those jurisdictions, are useful for understanding the way in which the concept of an ‘improvement’ to land has developed, absent a statutory definition such as we have in Victoria.
In our view, however, the answer to whether Landene is an improvement lies in the definition of ‘improvements’ in the VLA and, in particular, the requirement not only that the improvement increases the value of the land, but that the benefit to the land conveyed by the improvement be ‘unexhausted’ at the time of valuation.
It will be recalled that the definition refers to ‘all work actually done or material used on and for the benefit of the land, but in so far only as the effect of the work done or material used increases the value of the land and the benefit is unexhausted at the time of the valuation’.[115] This formulation distinguishes ‘the work done or material used’, which must increase the value of the land, and the ‘benefit’ to the land, which must be ‘unexhausted at the time of the valuation’.
[115]VLA, s 2 (emphasis added).
The definition of ‘improvements’ in the VLA therefore speaks of two points in time.
First, it refers to the time the work is actually done or the material is used. It imposes the qualification that at that time, the work or material must increase the value of the land. Secondly, it refers to the time of valuation. It imposes the qualification that the benefit must be ‘unexhausted’ at that time. This is not a requirement that the improvement increase the market value of the land at the date of valuation. It is simply a requirement that there be a continuing benefit to the land at the date of valuation.
The construction of Landene in 1897 constituted ‘work actually done or material used on and for the benefit of the [L]and’. At that date, the effect of Landene was to increase the value of the Land. More particularly, the work benefitted the Land in comparison to the hypothetical unimproved ‘natural’ state of the Land at the date of the ‘improvements’. Importantly, the benefit of the work done was unexhausted at the time of the 2020 and 2021 valuations, in that Landene continued to serve a variety of economic purposes, including the provision of accommodation and as a gallery for the owner’s art and antiques.
Land may be benefitted, and hence improved, without achieving its highest and best use. Thus, the erection of a detached dwelling upon land that could accommodate multi-unit development may benefit the land without achieving the highest and best use of the land. This reflects the meaning of ‘improvement’ adopted in Trust Company and Commonwealth Custodial.
In any event, even if it were necessary for land to achieve its highest and best use in order to become ‘improved’, there is no reason to doubt that Landene constituted the highest and best use of the Land when it was added to the Land. The critical question now is whether the benefit to the Land conveyed at that time is ‘unexhausted’ at the date of the 2020 and 2021 valuations, not whether if the works and material were supplied to the Land today, they would increase the value of the Land.
The fact that the benefit of an improvement to land may be seen to persist until ‘exhausted’ was recognised in Morrison. There, Griffiths CJ referred to the value of improvements as a ‘constant’, as opposed to improved value, which is a ‘variable’. Describing operations on land ‘the full effect of which has been obtained’ — such as the removal of stones making the land fit for agriculture — his Honour observed that while the operation has been completed the benefit remains, unless and until exhausted.[116] This is consistent with what we consider to be the proper construction of the definition of ‘improvements’ in the VLA.
[116]Morrison, 504.
The benefit of ‘improvements’ will be exhausted if they no longer facilitate the economic use of the land, as for example: a poppet head and other mining works at the head of a mine that has been worked out; an underground fuel tank on the site of a demolished petrol service station; a residential building with substantial quantities of asbestos that is no longer fit for habitation or other use.
Section 2(2) does not derogate from this construction. It is expressed to operate ‘for the purpose of ascertaining the site value of the land’ and, as such, must be given some work to do in the present context. However, it does not govern the initial inquiry as to what constitutes an improvement, but is directed to the subsequent valuation exercise. More particularly, s 2(2) relates to ‘the value of the improvements’. It does not relate to the value of the work done or material used on the land and is not expressed to govern the question of whether the effect of the work done or material used raises the value of the land. Rather, it is concerned with the valuation of matters that meet the definition of ‘improvements’.
To the extent that s 2(2) uses the phrase ‘ascertaining the site value of the land’, it must be read as contemplating circumstances in which estimation of the value of ‘improvements’ is a necessary step in adjusting comparable sales evidence relating to the value of the site. However, if contrary to the analysis above, s 2(2) is read as requiring the first qualification in the definition of ‘improvements’ to be satisfied by reference to the market value test, it still does not govern the question of whether the benefit of the ‘improvements’ is ‘unexhausted’ at the time of valuation.
The construction that we favour is, as a matter of reality, the only sensible one. It would be anomalous if an increase in the building bulk permissible within a zone (or other relaxation of planning controls) had the capacity to transform entire neighbourhoods of existing buildings into encumbrances (or ‘worsements’) overnight, despite the fact that they still facilitate substantial economic use of the lands in question. This is a potential consequence of requiring the question of improvement to be assessed by reference to the highest and best use of the Land at the date of valuation.
On the evidence before it, the Tribunal was correct to reject the approach of Mr Haines, who declined to characterise Landene as an improvement by focusing on the effect of the heritage overlay, rather than whether the benefit of Landene was ‘unexhausted’ at the time of valuation. Landene is a valuable structure accommodating a number of uses that continues to benefit the Land. On the evidence before it, the Tribunal was correct to hold that Landene is an improvement for the purpose of assessing the site value of the Land.
For Landene to be recognised as an improvement is not to undo the effect or undermine the purposes of the repeal of ss 2(8) and 2(9) of the VLA. It is to treat the Land in the same way as all other land in Victoria.
There is no challenge to the manner in which the Tribunal otherwise went on to determine the site value of the Land by reference to the relevant expert evidence, including the evidence of the quantity surveyors.
Accordingly, although the Tribunal was not correct to base its method of ascertaining whether Landene was an improvement on first establishing the highest and best use of the Land, grounds 2 and 4 are not made out, and the remaining grounds, whether or not made out, are inconsequential. We will not interfere with the Tribunal’s orders.
Grounds 5 & 6 — The Tribunal’s role on review
Together, grounds 5 and 6 challenge the Tribunal’s method of conducting a review under s 21(3)(a) of the VLA. Ground 5 is that the Tribunal misapprehended its task on review, which is to carry out a merits review. Ground 6 is that the Tribunal erred in finding that the valuation evidence tendered by VGV was irrelevant and impermissible, and that, insofar as the Tribunal relied on S & JG Investments, that case was wrongly decided.
Nothing turns on these grounds. They do not affect the outcome of this appeal. We shall therefore consider them in very short compass.
VGV submits that the Tribunal’s role was to ‘do over again’ what the returned valuer was required to do under s 21(3) of the VLA, which entailed consideration of whether an adjustment to the original valuation was justified. As the original (returned) valuations were flawed, it was necessary to present a new valuation to enable the Tribunal to be positively satisfied as to whether an adjustment was justified. To ignore the flawed nature of the original valuation would render the Tribunal incapable of properly determining whether an adjustment was justified.
VGV submits that the Tribunal therefore erred in finding that the valuation evidence tendered by VGV was irrelevant.
Section 22 of the VLA provides for the Tribunal to review the valuer’s decision to disallow an objection, rather than to review the original (returned) valuation. The Tribunal’s task was therefore to determine whether the valuer was correct to disallow the objection. To discharge this task on review, the Tribunal cannot begin by simply conducting an entirely new valuation for itself. The Tribunal must first understand the basis for the returned valuation and the evidence considered by the valuer in arriving at that valuation.
The original (returned) valuations were not before the Tribunal and the Tribunal had no opportunity to consider whether they were ‘flawed’ and should be disregarded. Although VGV was uniquely placed to lead relevant evidence about the returned valuations, it decided not do so, despite being invited to do so by the Tribunal.
The Tribunal did not say that VGV was not entitled to adduce evidence of a higher valuation. Indeed, the possibility of adducing evidence of a higher valuation was contemplated by the Tribunal in S & JG Investments. What the Tribunal considered to be impermissible was the proffering of a new valuation prepared on an entirely novel basis while ignoring the returned valuations (in respect of which the objections were taken).
However, we do not consider that this rendered Mr Haines’ evidence irrelevant, as appears to be suggested in the Tribunal’s reasons at [279]. Mr Haines’ evidence was not accepted, but it was not irrelevant.
Conclusion
Leave to appeal will be granted, but the appeal must be dismissed.
---
7
14
1