The Judicial Committee of the Privy Council Decisions

You are here: BAILII >> Databases >> The Judicial Committee of the Privy Council Decisions >> Rodney District Council & Ors v Attorney-General (New Zealand) [2002] UKPC 47 (7 October 2002)
URL: http://www.bailii.org/uk/cases/UKPC/2002/47.html
Cite as: [2002] UKPC 47, [2003] RA 180

[New search] [Printable RTF version] [Help]

Rodney District Council & Ors v Attorney-General (New Zealand) [2002] UKPC 47 (7 October 2002)

Privy Council Appeal No. 29 of 2001

(1) Rodney District Council

(2) Manukau City Council

(3) Hutt City Council and

(4) New Zealand Local Government Association

Incorporated Appellants

v.

(1) The Attorney-General and

(2) The Valuer-General Respondents

FROM

THE COURT OF APPEAL OF NEW ZEALAND

---------------

JUDGMENT OF THE LORDS OF THE JUDICIAL

COMMITTEE OF THE PRIVY COUNCIL,

Delivered the 7th October 2002

------------------

Present at the hearing:-

Lord Bingham of Cornhill

Lord Nicholls of Birkenhead

Lord Hope of Craighead

Lord Hutton

Lord Hobhouse of Woodborough

[Delivered by Lord Hope of Craighead]

------------------

This is an appeal from a judgment of the Court of Appeal of New Zealand (Richardson P, Thomas, Keith, Blanchard and Tipping JJ) dated 18 September 2000 allowing an appeal and dismissing a cross-appeal from a decision of Fisher J in the High Court of New Zealand dated 16 August 1999: [2000] 1 NZLR 101. Conditional leave to appeal from this judgment to the Privy Council was given by the Court of Appeal on 23 November 2000. The appellants are three local authorities and the New Zealand Local Government Association Incorporated. The respondents are the Attorney-General and the Valuer-General.

The issue which the appellants have raised relates to one of the key components of the district valuation roll prepared under the now repealed section 8 of the Valuation of Land Act 1951 (“VLA 1951”). Section 8 of that Act set out the particulars which the valuation roll was to include “in respect of each separate property”. The question is whether the expression “separate property” in this context means a property as defined by a certificate of title, with the result that a single entry must be made even in cases where the land to which the certificate of title relates is occupied by two or more separate occupiers as defined by the Rating Powers Act 1988 (“RPA 1988”), or whether it means a property as defined by the unit of occupation.

The Valuer-General, Mr Quinn, explained his approach to this problem in this way in paragraph 4.1 of his affidavit:

“The term separate property is not defined in either the VLA or the RPA. That is left over to the Valuer-General. In summary, the Valuer-General’s basic trigger to create a separate property entry on the district valuation roll is the creation of a separate unit of ownership. As New Zealand uses the Torrens system, the existence of a separate certificate of title has been a key element in that decision as that identifies the owner’s land to be valued and that which would sell on the market. The main trigger to create a rating apportionment for a territorial authority has been the existence of a lease of 12 months or more. This is driven by the territorial authority’s desire to maintain and identify occupier details on the rate record. However, identification of the owner’s land (the separate property) is quite different from needing to record on the rates record different occupiers of that land (rating apportionments).”

As the quotation from Mr Quinn’s affidavit indicates, the Valuer-General’s contention is that for there to be a separate property within the meaning of section 8 of VLA 1951 there had to be certificate of title or something similar such as a separate strata or unit title. For the local authorities and the association it was contended that occupation was a sufficient criterion for the identification of a property as separate property. The argument for the local authorities and the association that the approach intended by the legislation was that based on the unit of occupation was upheld by Fisher J. His decision on this point was reversed by the Court of Appeal. They held that the expression “separate property” meant a property as defined by the certificate of title: para [57]. That conclusion was based on their view as to its ordinary meaning in its land law and statutory context: para [23]. On their approach the relevant criterion was the unit of ownership.

The appellants submit that the decision of Fisher J that occupancy was a sufficient criterion was correct and that the declarations which he made to that effect should be reinstated. But they did not seek to argue before the Board the point which they raised in their cross-appeal to the Court of Appeal, which was that the declarations which the judge made did not go far enough. They had contended before Fisher J that any form of separate occupation was enough, irrespective of whether it was occupation within the meaning given to that expression by section 2 of RPA 1988. An “occupier” as defined by that section is a person with a right to occupy the land by virtue of a tenancy granted for a term of not less than 12 months certain. The judge said, at p 16 of his judgment, that the more limited definition of this expression in section 2 of RPA 1988 lay at the heart of the “occupation” concept as imported into section 8 of VLA 1951. The appellants now accept that this is so, and it was on this basis that they invited the Board to hold that the approach based on the unit of occupation was the approach which was intended by the legislature.

The importance of the issue

The rating powers of local authorities are very wide. At the relevant time they were set out in Part II of RPA 1988. Territorial authorities had power under section 19 to make and levy uniform annual general charges on every separately rateable property in the district or a subdivision of the district. The expression “separately rateable property” was defined as “a property entered as a separate property in the district valuation roll which is rateable property in terms of this Act:” RPA 1988, section 2. Uniform annual general charges are to be contrasted with general and special rates, which were levied either as a uniform rate in the dollar on every rateable property or on a differential basis under sections 79 to 83 of the Act: RPA, sections 12 and 16. They are also to be contrasted with separate uniform annual charges for the ordinary supply of water and refuse collection and disposal, which attached to “each separately used or inhabited portion of the property or building”: RPA 1988, section 24(1). Those charges could be levied on each separately used or inhabited portion of the property, irrespective of whether the person against whom they were levied was the occupier of that portion as defined by the Act. The problem which has been identified in this case relates only to the system for levying uniform annual general charges.

The Valuer-General’s approach was to determine a single value for the property described in a single certificate of title. He then apportioned that value on a pro rata basis between the various premises, if any, within that unit of property which were occupied by different persons with a right to occupy the land by virtue of a tenancy granted for a term of 12 months or more. Where any property for which there was a single certificate of title was in multiple occupation his practice was to make a single entry in the valuation roll for the property as a separate property and then to make a sub-entry for each of the multiple occupancies. Where these apportionments were made most councils treated the sub-entries as separate properties. They did so with a view to achieving a result which appeared to them to be appropriate and equitable: see Wellington City Council v Woolworths New Zealand (No 2) [1996] 2 NZLR 537, 545 per Richardson P. They levied uniform annual general charges against each of the sub-entries which were the result of the process of apportionment and not on the properties identified in the valuation roll as a separate property. The charges were paid by the occupiers on the assumption that the properties on which they were levied were separately rateable under VLA 1951.

Questions then arose as to whether the councils were right to treat these sub-entries as separately rateable properties and, if not, as to whether by levying uniform annual general charges on sub-entries they had acted illegally. The effect of the Court of Appeal’s decision is that only the head entry, which reflected the unit of ownership represented by a separate certificate of title, was a separately rateable property within the meaning of RPA 1988, section 2. As uniform annual general charges could only be levied against a separately rateable property, the decision appears to have put at risk the councils’ aim of achieving equity and consistency between occupiers. The problem was described by Fisher J in his judgment at p 9:

“There is an inherent anomaly in exacting six uniform rates or charges from six households occupying the land in six separate certificates of title, but only one uniform rate or charge for the land contained in one certificate of title notwithstanding occupation by six otherwise identical households.”

The anomaly is most clearly demonstrated where the two sets of properties are physically indistinguishable and are used and occupied in the same way. Various examples of this situation are described in the Record. They include industrial premises as well as self-contained residential flats and dwellings. The Court of Appeal were not asked to consider the legal consequences of their decision, but the appellants have indicated that the territorial authorities concerned may be liable to repay charges levied by them amounting to about $36m. The estimated total liability nationwide is said to amount to 1.2% of the councils’ overall annual operating income.

The Court of Appeal said when it decided to grant leave that the question of law as to the meaning of the expression “separate property” was important not only for the past but also for the future. This is because the system of valuation described by the Rating Valuations Act 1998, which repealed and replaced that laid down by VLA 1951 with effect from 1 July 1998, was also based on “separate property”: see section 7 of the 1998 Act. But sections 14 and 17 of and Schedule 2 to the Local Government (Rating) Act 2002 have introduced an entirely new system for defining categories of rateable land. So the focus of attention in this appeal has been throughout on the meaning which must be given to this expression in the context of the now repealed legislation.

The complexity of that legislation is such that the question which has been raised in this case is far from easy. The Board are also conscious of the fact that it must be answered in the light of the role which is given to a certificate of title under the Torrens system in New Zealand, and that there are important differences between the legislation which gives power to local authorities to levy rates and charges in New Zealand and the legislation in the United Kingdom with which they are familiar. This is a highly technical field of law, and their Lordships’ task has been greatly assisted by the careful judgment of the Court of Appeal delivered by Keith J to which they wish to pay particular tribute.

Historical background

Although the Board is concerned in this case with the meaning of the expression “separate property” in section 8 of VLA 1951, it is of some importance by way of introduction to examine the historical background. The significance of land as a source of taxation had been recognised both in England and in Scotland from the earliest times. Originally it was the principal source from which the Crown derived its revenue. The Domesday Book was a valuation roll: O’Keefe, The Legal Concept and Principles of Land Value (Butterworths, Wellington, 1974), para 2.1. This can be attributed to the fact that land was the most significant element in the national economy. It later became the basis for the assessment of a wide variety of taxes for local and municipal purposes such as relief of the poor. No doubt this was because it was easier to impose these burdens on the value of land rather than the value of other kinds of property. But the system grew up in a haphazard fashion. The various public bodies charged with the imposition of rates for local and municipal purposes each had to make up their own valuation roll, and these were often prepared from insufficient data and on different principles. Radical reform was needed, and it came in the nineteenth century.

The foundation of the modern system of land valuation for rating in Scotland, whose system has always been separate from that in England and Wales, was laid by the Lands Valuation (Scotland) Act 1854. It established a uniform system for the valuation of lands and heritages in Scotland according to their yearly rent or value by means of which all local assessments could be collected. It also required details to be given in the valuation roll of the nature of the subjects and the names of their proprietors, tenants and occupiers. As a result there was produced a reliable inventory of the heritable property in each local authority area. This became available for use for other purposes such as the voters’ roll, as at that time the right to vote was dependent on the ownership of land. But its principal purpose was to provide the basis for obtaining revenue by means of rates levied on the value of the property.

The basic principles which that Act established remained unaltered until the system was revised and brought more closely into line with that of England by the Valuation and Rating (Scotland) Act 1956. Liability for rates, which had previously been charged partly on owners and partly on occupiers, was transferred by that Act entirely to occupiers. The actual rents passing were no longer to be conclusive as to annual value, which is the basis upon which the rateable value of property is assessed under the Act except in cases where it is determined by means of a statutory formula. Annual value was to be assessed instead on the basis of a hypothetical rent, with a view to ensuring that the same level of value was used for all comparable properties in the same valuation area. As the entire liability for the payment of rates was transferred by the statute to the occupier, it came to be recognised in practice that the unit of property which was to be valued and entered in the valuation roll was to be measured by the unit of occupation and not by the unit of ownership. But, following the position which had long been established in England, occupation had to have the character of permanence for it to be rateable: R v The Assessment Committee of St Pancras (1877) LR 2 QB 581, per Lush J at p 588; Westminster Council v Southern Railway and Others [1936] AC 511, per Lord Russell of Killowen at pp529-530; John Laing and Son Limited v Assessment Committee for Kingswood Assessment Area [1949] 1 KB 344.

The development of legislation as to valuation and rating in New Zealand has a similar history: O’Keefe, The Legal Concept and Principles of Land Value, para 2.1; O’Keefe, The Law of Rating (Butterworths, Wellington, 1975), paras [102]–[103]. The general idea of levying rates on the occupiers of land was derived from England. Auckland, for example, was constituted a borough under the Provincial Councils Ordinance by charter in 1851 with the ordinary powers of an English town council. This included the power to levy rates. In Dunedin the Town Board had power under its Ordinance of 1855 to levy rates on the net annual value of land within its area. The Wellington Town Board Ordinance set up a local authority in 1862 with power to rate on the capital value of land. As the system spread, different provinces used different valuation systems. Rates were all levied on land, but the English approach of valuing the land by reference to its annual value had been departed from in many cases by 1876 when the system of provincial government in New Zealand came to an end. In the same year, recognising the importance of local finance from rates, the General Government introduced a uniform system for the making and levying of rates on annual value: O’Keefe, The Law of Rating, para [104]. But it was not long before annual value as a uniform system was displaced by capital value. That system was made compulsory by the Rating Act 1882, except in the case of boroughs established as municipal corporations which were given the option of bringing themselves within that system by resolution.

The Municipal Corporations Act 1867 introduced some uniformity into the government of new municipalities, and it is from that date that the development of the modern system in New Zealand can be traced. Section 205 provided that rates were to be levied “upon every person who occupies or if there be no occupier then upon the owner of any rateable property”. Section 213 of that Act provided that the valuation of the rateable property was to be computed at its annual value, “that is to say at the rent at which the same might reasonably be expected to let from year to year”. Section 216 provided that the valuation was to be in the form stated in the Sixteenth Schedule, which set out the particulars to be given “with regard to each rateable tenement or hereditament” in the valuation list. These provisions appear to have been designed according to the English model. While they describe the property itself as rateable, the primary liability attaches not to the property as such but to the occupier: see Ryde on Rating and the Council Tax (Butterworths), para B[52]–[53]. They also indicate that, as the occupier was to be primarily liable for the rates assessed on rateable property, the unit of valuation was not to exceed the unit of occupation.

It should be noted that the basic elements of this system were in place some years before the introduction into New Zealand of the land transfer system adopted by the Land Transfer Act 1870. Moreover the introduction of that system did not result in a reconstruction of the system for making up the valuation roll. There is no sign in the legislation of this period that it was the intention that these two systems should be linked to each other. That would not, of course, have been practicable at the outset. As late as 1923 it was estimated that about one-fifth of the land in New Zealand was still held under the Deeds system: Butterworths Land Law in New Zealand (1997), para 2.026. It was not until 1951 that the task of bringing all Deeds system titles under the Land Transfer Act was officially regarded as complete.

The system which was laid down by the Municipal Corporations Act 1867 was amended from time to time by subsequent enactments, but the essential features remained the same. Section 2 of the Rating Act 1882 defined the expression “rateable property” as meaning, with certain exceptions, “all lands, tenements or hereditaments in the colony”. Section 4 of that Act provided that the valuation roll was to state for all rateable property in the district “belonging to each separate owner, or occupied by each separate occupier” the rateable value of the same, with the names, occupations and addresses of the owners or occupiers. Section 28 provided that the occupier was to be primarily liable for all rates. The expression “occupier” was defined in section 2 as meaning “the person by whom or on whose behalf any rateable property is actually occupied, if such person is in occupation by virtue of a tenancy which was for not less than six months certain” or, in other cases, the owner. This gave statutory recognition to the principle that a degree of permanence was necessary for occupation to be rateable. From now on, only occupiers as so defined were to be treated as in rateable occupation of the property.

The Rating Act 1894 retained the system by which the occupier was primarily liable and section 2 of the Act included the same definition of the expression “occupier”. But it was more explicit as to the treatment which was to be adopted where a property was occupied by more than one person with different degrees of interest in the property. Section 11 provided that in that event “a separate valuation shall be made of the interest of each such occupier, and his name entered in the column of occupiers in the valuation-list”. Here again there is a clear indication that the unit of valuation was not to exceed the unit of occupation as defined by the Act. Where two or more persons were found to be in separate occupation of a property, separate valuations had to be made of the interest of each occupier, and separate entries had to be made in respect of each of them in the valuation roll. Both annual value and capital value were available to enable the rateable value of their interests to be assessed.

It was not until the office of the Valuer-General was established by the Government Valuation of Land Act 1896 that the phrase “in respect of each separate property” was introduced into the legislation. Section 4 of the Act provided that the Valuer-General was to prepare a general valuation-roll of all landed properties in the colony. But continuity with the system laid down by the 1894 Act was preserved by the provision in section 6 that the general valuation roll was to be compiled from the district valuation-rolls prepared under that Act. Section 5 provided that the general valuation roll was to set forth certain particulars “in respect of each separate property” including “the name of the occupier and owner within the meaning of the Rating Act 1894”. The expression “separate property” was not defined. There is nothing in the context in which it appears to indicate that it meant a property as defined by a certificate of title under the Land Transfer Act. On the contrary, the instruction in section 11 of the 1894 Act that a “separate valuation” was to be made where property was occupied by more than one occupier indicates that the intention was that the interest of each person who was in rateable occupation of the property was to be entered separately in the general valuation roll.

The conclusion which their Lordships draw from this brief historical review is that at least up to this point, when the land transfer system introduced by the Land Transfer Act 1870 was still in its relative infancy, the unit of valuation contemplated by the rating legislation was the unit of occupation. This was to apply not only to a property which was occupied by only one person or by joint occupiers but also to cases where a property in a single ownership was occupied by more than one occupier.

The system under VLA 1951 and RPA 1988

The question then is whether the system laid down by VLA 1951 required a different approach to be taken to identify the unit which was to be valued and entered as a separate entry in the valuation roll in cases where there was more than one occupier. This Act, according to its long title, was “An Act for the compilation of certain enactments relating to the periodical valuation of landed properties”. It provided for the appointment of officers to administer the valuation system, for the creation of district valuation rolls, the hearing of objections to valuations and the general administration of the valuation system. It was a consolidating statute.

As a general rule an investigation of the statutory predecessors of provisions in a consolidating statute is to be avoided, especially where the issue concerns the meaning of a single word or a single expression. It is apt to reduce the benefit which the process of consolidation is designed to achieve. As Lord Simon of Glaisdale said in Maunsell v Olins [1975] AC 373, 392, the very purpose of consolidation is to enact a compendious code standing on its own with a view to making it unnecessary to scrutinise the repealed statutes which are consolidated. But the rule is not an absolute one. There may be cases where is it necessary to consider the earlier, consolidated provision in its original context, bearing in mind that the overriding aim is to give effect to the intention of Parliament as expressed in the words used: R v Secretary of State for the Environment, Transport and the Regions, ex p Spath Holme Ltd [2001] 2 AC 349, 388, per Lord Bingham of Cornhill. In their Lordships’ judgment this is such a case. The true meaning of the expression “separate property” cannot be properly understood without examining the factual context in which those words were first enacted. It requires to be read in the light of the historical background.

Some important amendments were made to VLA 1951 by subsequent legislation. Section 2 of the Rating Act 1967 altered the definition of occupier, which was now defined in relation to any land as “the owner thereof, except where a person other than the owner has a right to occupy the land by virtue of a tenancy granted for a term of not less than twelve months certain, in which case the term ‘occupier’ means that other person”. Section 62 of that Act provided that the occupier of any rateable property was to be primarily liable for the payment of all rates becoming due and payable while his name appeared on the rate records as the occupier of the property. The Valuation of Land Amendment Act 1981 added a new subsection (2) to section 8 of LVA 1951 which set out the particulars to be entered in the district valuation roll. The Valuation of Land Amendment Act 1988 inserted into the valuation legislation the provisions which had previously been contained in the rating legislation dealing with annual value valuation rolls. One of the amendments made by this Act was the insertion by section 5(2) into section 8 of VLA 1951 of section 8(1A), which set out the particulars to be entered in the district valuation roll in the case of valuations made on the basis of annual value. Finally, RPA 1988 consolidated and amended the previous legislation dealing with the rating powers of local authorities together with that dealing with various procedural aspects relating to the making and levying of rates.

Section 8 of VLA 1951, in its amended form, was as follows:

“(1) A district valuation roll shall be prepared for each district by the Valuer-General, and shall be in the prescribed form, and shall set forth in respect of each separate property the following particulars

(a) The name of the owner of the land, and the nature of his estate or interest therein, together with the name of the beneficial owner in the case of land held on trust:

(b) The name of the occupier …

(c) The situation, description and area of the land:

(d) The nature and value of the improvements:

(e) The land value of the land:

(f) The capital value of the land:

(ff) Where applicable, the special rateable value or the rates-postponement value of the land:

(g) Such other particulars as are prescribed.

(1A) An annual value valuation roll shall also be compiled by the Valuer for any district of a territorial authority where the annual value rating system is in force, and shall in the prescribed form contain for each separate property the following particulars:

(a) The name of the owner:

(b) The name of the occupier:

(c) The situation and description of the property:

(d) The annual value:

(e) Where applicable, the rates postponement value or the special rateable value, as the case may require:

(f) Such other particulars as may be prescribed.

(2) For the purposes of this section any land that is capable of separate occupation may, if in the circumstances of the case it is reasonable to do so, be treated as separate property whether or not it is separately occupied.”

The following expressions which appear in that section were defined:

(a) “Owner” meant “the person who, whether jointly or separately, is seised or possessed of or entitled to any estate or interest in land:” VLA 1951, section 2.

(b) “Occupier” had the same meaning as in section 2 of RPA 1988, which provided that, in relation to any land, it meant “the owner thereof, except where a person other than the owner had a right to occupy the land by virtue of a tenancy granted for a term of not less than 12 months certain, in which case the term ‘occupier’ means that other person.” VLA 1951, section 2 incorporating RPA 1988, section 2.

(c) “Land” meant “all land, tenements and hereditaments, whether corporeal or incorporeal, in New Zealand, and all chattel or other interests therein, and all trees growing or standing thereon”: VLA 1951, section 2.

(d) “Capital value” of land meant “the sum which the owner’s estate or interest therein, if unencumbered by any mortgage or any charge thereon, might be expected to realise at the time of valuation if offered for sale on such reasonable terms and conditions as a bona fide seller might be expected to require:” VLA 1951, section 2.

(e) “Annual value”, in relation to any rateable property, meant “the rent at which the property would let from year to year, deducting therefrom 20 percent in the case of houses, buildings, and other perishable property, and 10 percent in the case of land and other hereditaments; but in no case shall it be less than 5 percent of the capital value of the fee simple of the property.” VLA 1951, section 2, inserted by section 2(1) of the Valuation of Land Amendment Act 1988.

The expression “separate property” was not defined either by VLA 1951 or by RPA 1988. Its meaning must be ascertained from its text and in the light of its purpose: Interpretation Act 1999, section 5(1). So far as the text is concerned, the following points can be made in the light of its immediate surroundings in VLA 1951, section 8 as amended.

The first point is that the expression “separate property” was used both in subsection (1A) which set out the particulars to be entered in the valuation roll under the annual value system and in subsection (1) which set out the particulars to be entered in the valuation roll under the capital value system. This suggests that the expression was equally serviceable, and that it was intended to have the same meaning, irrespective of the method of valuation that was being used. The second point is that some guidance as to what the expression means can be found in subsection (2). This subsection provided that, if land was capable of being separately occupied, it could be treated as separate property if it was reasonable to do so.

There is no doubt that the purpose and effect of section 8(2) of VLA 1951 was to widen the circumstances in which a property might be treated as separate property. It permitted land to be treated as separate property if it was reasonable to do so, whether or not it was separately occupied. The assumption on which it appears to have proceeded was that a property which was in fact separately occupied already fell to be treated as separate property. According to the definition of the expression “occupier” in section 2 of RPA 1988, if there was no person other than the owner who had a right to occupy the property by virtue of a tenancy granted for a term of not less than 12 months certain, the owner was the occupier. If there was such a person, that other person was the occupier. Subsection (2) provides a clear indication that separate occupation as so defined is the key to the existence of a separate property.

In Findlay v Valuer-General [1954] NZLR 76, 78 Judge Archer made these observations on the effect of section 8 of LVA 1951 and the relevant definitions:

“It is clear from the foregoing provisions of the Act that, although the valuation roll is described as a roll of ‘separate properties’ any estate or interest in land which is held in separate ownership may be a separate property for roll purposes, and may be valued accordingly. Conversely, it would appear that nothing can be entered as a property in a district valuation roll which is not an estate or interest in land. The statutory definition of ‘land’ appears to be the decisive factor in determining what may be entered in a district valuation roll …

It will be noted that the definition of ‘land’ includes ‘all chattels or other interests therein.’ Leasehold interests in land may, therefore, be the subject of separate entries in a district valuation roll.”

These observations, with which their Lordships respectfully agree, were followed in Valuer-General v Radford [1993] 3 NZLR 721. They are important in the present context. The definition of “occupier” in section 2 of RPA 1988 appears to march hand in hand with the definition of ‘land’ in section 2 of VLA 1951. A person who occupies land by virtue of a tenancy granted for a term of not less than 12 months certain has a leasehold interest in the property which he occupies. The property which he occupies is capable of being made the subject of a separate entry in the district valuation roll as it is a leasehold interest in land. It need not be co-extensive with the freehold title to the land over which that interest extends. It need not be comprised in a certificate of title.

In Valuer-General v Alfred Kohn Family Trust (1990) LVC 867 the High Court was concerned with the question whether the land values of four contiguous freehold properties in a commercial property in central Wellington were to be arrived at by considering them separately or as one entity. The crux of the appeal was the meaning which was to be given to section 8 of LVA 1951. After setting out the provisions of section 8(1) and section 8(1A), Greig J said:

“What we think is essential in the preparation of the district roll is, first of all, to identify the separate properties. That phrase is not defined but it must be the case that separate occupation is one aspect of that. Subsection 8(2) necessarily implies that separate occupation and the capability of separate occupation are two of the ways in which the separate occupation can be identified. Other matters which the appellant submits, we think correctly, to be among the criteria for that identification include separate ownership, different or distinct land tenure, separate land use and the availability of separate title.”

Turning to the guidance which is to be found in RPA 1988 as to the purpose of the expression “separate property”, its significance for the purpose of the making and levying of rates is clearly demonstrated. Section 121, which re-enacted section 62 of the Rating Act 1967, provided:

“The occupier of any rateable property shall be primarily liable for all rates becoming due and payable while his or her name appears in the rate records as the occupier of the property, and all rates levied under this Act shall be recoverable in the manner hereinafter provided.”

Section 106 of RPA 1988 dealt with the situation that arose on a change of owner or occupier. The outgoing owner or occupier was required to give notice of the change to the territorial authority in whose district the land was situated, to every other local authority that made and levied rates on the land and to the Valuer-General: section 106(2). He was deemed to be the owner or occupier, as the case might be, for the purposes of recovery of rates under the Act until the requisite notice had been given: section 106(5). Section 106(6) made the following provision as to what was to happen if the transaction, whether by way of sale or the granting of a tenancy, related to part only of the rateable property:

“Every owner or occupier of rateable property who –

(a) Sells or otherwise transfers a part only of any property identified in the valuation roll as a separately rateable property; or

(b) Grants a tenancy of such part for a term of not less than 12 months certain or a renewal of any such tenancy; or

(c) Surrenders any tenancy of such a part –

shall, for the purposes of recovery of rates under this Act, be deemed to be the owner or occupier, as the case may be, of such part of the property for the balance of the financial year, unless within that period either-

(d) The rateable value of the whole property is apportioned in accordance with section 120 of this Act; or

(e) The Valuer or the Valuer-General authorises the amendment of the valuation roll to record such part of the property as a separately rateable property.”

Section 120(1) of RPA 1988 was in these terms:

“Where during a rating year but before the making of a rate for that year a local authority receives notice under section 106 of this Act of –

(a) The sale or transfer of any part of any rateable property; or

(b) The surrender or termination of any tenancy of any rateable property so far as it relates to part only of the property comprised in the tenancy, -

the local authority may apportion the rateable value of the property among the several parts thereof, and the amounts so apportioned to each part shall be deemed to be its rateable value for the purposes of this Act until a valuation of that part made under the Valuation of Land Act 1951 comes into force.”

These provisions show that the grant of a tenancy of part of any land for a term of not less than 12 months certain was treated in the same way as the sale of part of that land. The person to whom the tenancy of the part was granted became the occupier of the land within the meaning of section 2 of RPA 1988. But, for the purposes of recovery of rates in respect of the part, the owner or occupier by whom the tenancy was granted remained the owner or occupier until steps were taken to provide the part with its own rateable value. The unit to which the rateable value was to be apportioned under section 120 or valued under VLA 1951 was the part of the land which was subject to the tenancy. In other words, it was the unit of occupation. There is no indication here that it was to be defined by a certificate of title.

The Solicitor General said that there were several provisions in both VLA 1951 and RPA 1988 which assumed that there could be more than one occupier of a separately rateable property. The Court of Appeal said that these provisions treated the owner and the occupier equally or closely linked them: para [29](6). For example, sections 25A-25G which were originally enacted in 1988 introduced a system of rates postponement and the use of special rateable values in a variety of circumstances. Typical of these provisions was section 25D, which provided for the determination of special rateable values of single or double-unit dwellinghouses situated in an area where the rateable value of residential land was influenced by a demand for multi-unit housing. Subsection (1)(a) of this section provided that land, “being a separate property”, was to be deemed to be used for single-unit housing or double-unit housing if, inter alia:

“There is erected on the land a building or a group of buildings used or intended to be used solely or principally for residential purposes and occupied or intended to be occupied exclusively as the home or residence of a single household or, as the case may be, 2 households.”

Fisher J said at p 16 of his judgment that, viewing the picture overall, he would conclude that in broad terms an occupation approach was intended. The Court of Appeal said that they did not see the references to “occupier” and “occupation” in the legislation as having the significance accorded to them by Fisher J: para [31]. In their view some of the provisions, especially those which affected the owner, supported what they described as the ordinary, certificate of title meaning. They referred to section 136 provided that rates were to constitute a charge on the land, and to section 139 which enabled the local authority to recover from the owner if the occupier was in default. In various other respects, such as in regard to the objection provisions in sections 18 to 20 of VLA 1951 as enacted in 1988, owners and occupiers were given equal treatment.

There is, of course, no doubt that ownership is closely identified in land law, and in particular in the land transfer system, by a distinct certificate of title. It was for this reason that the Court of Appeal regarded the fact that separate ownership was identified by a certificate of title in land law as the starting point: para [26]. But the view which Fisher J reached as to the meaning of “separate property” seems to be much more in keeping with the importance which has always been attached to rateable occupation as the criterion of liability in rating law. In this context it is rating law and not land law which provides the best guide to what is meant by that expression. There is no indication in the rating legislation that it was intended to have the same significance as it undoubtedly now has in the wider land law context.

In Telecom Auckland Ltd v Auckland City Council [1999] 1 NZLR 426 objection was taken to the Valuer’s decision to enter the utilities as a separate property in the valuation roll. It was held that, as Telecom had an exclusive right to occupy the portion of soil in which their lines and booths had been installed, they had an interest in land for rating purposes. It followed that they were to be regarded as the owner of that portion of the soil within the meaning of section 2 of RPA 1988 and that the land which they occupied was correctly entered in the roll as a separate property even though it was not contained in any certificate of title. This decision followed earlier decisions of the Court of Appeal about the treatment for rating purposes of gas mains and electricity lines: Auckland City Corporation v Auckland Gas Co Ltd [1919] NZLR 561; Hutt Valley Electric-Power Board v Lower Hutt City Corporation [1949] NZLR 611. It supports the argument that the expression “separate property” was not tied to the existence of a certificate of title for rating purposes, as it demonstrates that the “owner” of land within the meaning of section 2 of RPA 1988 did not have to be the owner of a freehold interest which a certificate of title had identified.

Apportionments

The Valuer-General accepted that it was necessary for him to distinguish between different occupiers when he was entering the particulars required by section 8 to be set forth in the valuation roll in respect of each separate property. His contention is that this could be done without dividing a property which was in multiple occupation into separate properties. He said that the result could be achieved by apportioning the rateable value under section 202 0f RPA 1988 between the parts which were separately occupied. This is an important argument. If that section did indeed confer a general power of apportionment, it would enable the rates and charges for which each occupier “of any rateable property” was liable under section 121 of RPA 1988 “as the occupier of the property” to be identified. That would meet the problem which this case has identified. But if it did not, this would strengthen still further the appellants’ argument that the unit of occupation was the criterion that was to be used to identify the extent of the rateable property.

Section 202 of RPA 1988 was in these terms:

“(1) Where it is necessary to apportion the rateable value of any rateable property between 2 or more portions of the property, the rateable value shall be apportioned in such manner as the Valuer-General, or, as the case may be, the Valuer for the district, thinks fit, so that the rateable value of each portion, when added to the rateable value of the remaining portion or portions of the property shall equal the rateable value of the whole property.

(2) Each such occupier may object to such apportionment as if it were a valuation, and the provisions of the Valuation of Land Act 1951 relating to objections, as far as they are applicable and with the necessary modifications, shall apply accordingly.

(3) Notwithstanding anything in the foregoing provisions of this section, where the occupier of a portion of any rateable property is the lessee or licensee under a lease or licence or has entered into an agreement with the owner, and the lease or licence or agreement specifies the portion of the rates in respect of the whole property that are to be paid by that occupier, the rateable value of that portion of the property shall be the sum which bears to the rateable value of the whole property the same proportion that the portion of the rates payable by the occupier pursuant to the lease or licence or agreement bears to the total amount of the rates payable in respect of the whole property.”

It is to be noted from the opening words of the subsection that the power to apportion in section 202(1) applied only where apportionment was “necessary”. The context for that expression was provided by the following provisions elsewhere in RPA 1988 which require an apportionment of the rateable value to be made between parts of the property:

(a) Section 4(4) and 6(2) required this to be done where part of a separately rateable property was deemed not to be rateable property. In that situation the rateable value of the whole had to be apportioned in order to determine the part of the rateable value for which the occupier of the part which was rateable was primarily liable. Sections 4(4) and 6(2) provided that in that event section 202 was to apply in relation to the apportionment.

(b) Section 105(4)(b) provided that where land was differentially rated the rateable value of the property was to be apportioned among its different parts. Here again it was necessary for this to be done in order to identify the extent of the liability of the occupier of each part. Section 105(5) provided that section 202 was to apply to any apportionment under section 105(5)(b).

(c) Section 120(1), to which reference has already been made, enabled the same exercise to be carried out where, during the rating year but before the making of a rate in that year, there was a sale or transfer of part only of a separate property.

(d) Section 179(1) gave power to a local authority to remit or postpone the payment of any rates in respect of certain types of land, and section 179(2) directed a local authority to remit half of the payment any rates and any uniform annual charges in respect of land of certain other types. Section 179(5) provided that, where in any case part only of any separately rateable property fell within these types, section 202 was to apply.

The reason why these provisions required an apportionment to be made was to enable the rates liability of each part of a separately rateable property to be identified in situations where the liability of each part was different. The purpose of section 202 appears then to have been to lay down the methods by which the apportionment was to be done where this was required elsewhere in the Act. Section 202(2) gave a right of objection to an apportionment by the Valuer-General to each occupier. The words “where it is necessary to apportion” in section 202(1) do not seem to be apt if the intention was to confer a separate power on the Valuer-General to apportion the rateable value in situations where an apportionment was not directed by the Act whenever he thought that this was appropriate. The Court of Appeal said that the power in section 202(1) of RPA 1988 was not presented as an ancillary power: para [53]. But that indeed is what it appears to be when it is read in the context of the Act as a whole. Their Lordships agree with the conclusion of the High Court in Auckland City Council v The Big Fresh Food Co (Auckland, M165/94, 6 July 1995), pp 8-9 that section 202(1) did not create a power to apportion where none already existed, but rather that it outlined the formula to be used where an apportionment had become necessary for some other reason.

Section 202(3) of RPA 1988 dealt with the situation where the occupier of a portion of any rateable property had entered into a lease or licence or any other agreement with the owner which specified the portion of the rates that were to be paid by that occupier. In that situation the apportionment was to be determined by the agreement with the owner and not by the Valuer-General. Fisher J said at p 12 of his judgment that it seemed to him to be implicit in section 202(3) that a given rateable property might have more than one occupier. At first sight this is difficult to reconcile with an approach based on the unit of occupation under which each separate property would have its own occupier. But the word “occupier” appears to have been used here in a more general sense than that indicated by the definition of this expression in section 2 of RPA 1988. This is because it is not confined to occupiers under a tenancy for a term of not less than 12 months certain, but includes an occupier under a lease of whatever term and an occupier under a licence. For this reason the better view is that section 202(3) does not cast any useful light on the meaning which is to be given to the expression “separate property”.

Practical considerations

As the Court of Appeal pointed out in paras [19]-[22] of their judgment, there are a variety of different fact situations that could arise under the provisions of section 8 of LVA 1951. There were three relevant variables. The number of occupancies within any given property might be greater than the number of certificates of title. Or the number of certificates of title might be greater than the number of occupancies. Or one person or more might own different certificates of title if there was more than one of them. Various situations within these variables might be envisaged which are of greater or lesser complexity.

It would, of course, be highly significant if there was evidence that an approach based on one or other of the competing interpretations of the expression “separate property” would have resulted in a system that was impractical. But all of these situations would appear to have been capable of being dealt with satisfactorily once the unit that was to be adopted as separate property had been identified. If it was to be measured by the unit of ownership, the number of distinct uses or occupancies could be disregarded for the purposes of the valuation exercise. All one needed to do was to identify the extent of the “land” held under each certificate of title. Each unit of ownership would then be the subject of a separate entry in the valuation roll, and it was also be the unit of valuation. Conversely, if the unit to be adopted was the unit of occupation, the question whether it was co-extensive with a certificate of title would not have been decisive of the unit which was to be entered as separate property. If the owner was the occupier, the extent of his ownership as disclosed by the certificate of title would determine the extent of the separate property. But if a person who was in occupation by virtue of a tenancy granted for a term of not less than 12 months was the occupier the extent of the separate property would be determined by the extent of his leasehold interest in the land, not by the certificate of title of the person who granted the tenancy.

This approach is consistent with what section 8 required as to the entries to be made of the other particulars. Section 8 of LVA 1951 used the expressions “owner” and “occupier” in the singular. It did not say what particulars were to be entered in the roll under each heading if there was more than one owner or more than one occupier, as the case might be. But the normal rule of interpretation is that the singular includes the plural. So the section may be taken to have contemplated the entry of more than one name under each of these headings where this was appropriate.

Nor is there any reason to think that either of the two approaches would give rise to difficulty at the stage of valuation. As Fisher J said at p 13 of his judgment, to comply with particulars (e) and (f) of section 8 of VLA 1951 the separate property had to be capable of attracting its own individual land value and capital value. But this does not mean that it would have been impractical to treat as separate property parts of the land that were smaller than the land contained in the certificate of title. Smaller areas of land might not have been saleable at law, but this does not mean that they were incapable of valuation. The valuation exercises which were directed by the Act did not require evidence of actual transactions relating to the property that was to be valued. They were based instead upon a hypothesis. The definitions in section 2 of VLA 1951 show that a hypothetical rent had to be determined where an annual value was required, and that a hypothetical capital value had to be determined where this was to be the basis of rateable value.

The witnesses were agreed that, while the valuation of different types of property might involve different degrees of difficulty, all areas of land and all forms of tenure are ultimately capable of valuation. Mr Robert McGough for the appellants said in his affidavit that there is no doubt that a piece of land which is not contained in its own certificate of title can be valued and that valuers frequently undertake such tasks. Mr Allen Pratt Pegler, who was until 1998 the Assistant Valuer-General, said that he accepted the argument that anything can be ascribed a value, although he made the point that a valuation is more likely to be challenged if it cannot be validated by reference to reliable market data. The Valuer-General, Mr Warwick Quinn, said that he agreed with Mr McGough that there was no valuation reason why properties without a certificate of title could not be valued, although he too drew attention to the issues that may arise where valuers are working in an evidentiary vacuum.

There is no doubt that the approach based on a certificate of title is better suited to a system which seeks to rely on the comparative principle of valuation, which is the one with which the Valuer-General plainly feels more comfortable. Figures based on sale prices which have received wide acceptance in the open market are likely to provide the best indication of the capital value of comparable properties. But the valuer’s art is not incapable of providing other solutions if no such evidence is available. As Lord Dunedin said in Port of London Authority v Assessment Committee of Orsett Union [1920] AC 273, 295:

“If the subject is an ordinary one similar in character to other subjects which have stood the test of the markets, or better still, if it has stood the test of the market itself without disturbing circumstances, the inquiry is simple. But when the nature and circumstances of the hereditament in question do not admit of such a test, some other way must be found.”

The Solicitor General said that the concept of a separate property involved something that was tradeable, and that a tenancy granted for a term of not less than 12 months certain could not be regarded as consistent with the type of transactions of which records would be kept for the purposes of giving effect to the legislation. But this submission overlooks the fact that the valuation that was required by section 8 was the capital value of the land or the annual value of the property as the case may be, not of the tenant’s interest in it. It was the sum which the owner’s interest in the land, if unencumbered, might be expected to realise if offered for sale on reasonable terms on the open market, or the rent at which the property would be expected to be let from year to year subject to the statutory deductions. This principle was capable of being applied to each part of a property in a single ownership which was in multiple occupation in the same way as it was applied to cases where each of the parts in separate occupation were in separate ownership.

There is therefore no compelling reason for rejecting the argument that a separate property was to be measured by the unit of occupation on the ground that it would have created a situation which was unworkable. In any event, as both Fisher J at p 14 and the Court of Appeal at para [56] observed, the ultimate test is the meaning of the statute rather than the convenience of those administering it.

Conclusion

The expression “separate property” has to be given a meaning which is appropriate to its context. Its purpose was to describe the extent of the property which was to be entered on the valuation roll with a view to its being identified as rateable property so that rates could be levied on the occupier. The starting point is to be found in the rating legislation and not, as the Court of Appeal thought, in the function which a certificate of title has in land law.

The legislative origin of the expression shows that it could not have been intended to mean a property as defined by a certificate of title when the expression was first used in 1896. Subsequent changes in the legislation up to and including the changes introduced by RPA 1988 do not reveal an intention that it should be given that meaning when it was re-enacted in section 8 of VLA 1951. Read as a whole the legislation as it had developed indicates the contrary. Primacy was given to the concept of rateable occupation and to the liability of the occupier of rateable property. In this situation very little, if any, weight can be attached to the meaning which the expression has in the wider land law context. The overriding consideration is that the unit of occupation determined the extent of the property which was to be valued for the purpose of levying rates on the occupier.

For these reasons their Lordships are unable to agree with the Court of Appeal that the expression meant a property as defined by a certificate of title. They consider that Fisher J was right to conclude that separate occupation rather than a separate title was the basic criterion.

Their Lordships will humbly advise Her Majesty that this appeal should be allowed and that the declarations which were made in the High Court should be restored. Neither side sought an order for costs in the courts below, and it was agreed that there should be no order for costs before the Board.