Stevely-Cole v Dilworth Trust Board
[2020] NZHC 2843
•30 October 2020
IN THE HIGH COURT OF NEW ZEALAND AUCKLAND REGISTRY
I TE KŌTI MATUA O AOTEAROA TĀMAKI MAKAURAU ROHE
CIV-2020-404-1186
[2020] NZHC 2843
UNDER the Arbitration Act 1996 IN THE MATTER
of an appeal against an arbitration before Raynor Asher QC
BETWEEN
SUSAN MARY STEVELY-COLE, CHRISTOPHER GRAEME COLE and
BRETT AVON LEAVETT in their capacity as trustees of the Mangahewa Trust
Plaintiffs
AND
DILWORTH TRUST BOARD
Defendant
Hearing: 9 September 2020; further submissions on 17 and
21 September 2020
Counsel:
JG Miles QC and KJ Jarvis for plaintiffs
DM Salmon and AW McDonald for defendant
Judgment:
30 October 2020
JUDGMENT OF FITZGERALD J
[As to application for leave to appeal against arbitral award]
This judgment was delivered by me on 30 October 2020 at 11:30am, pursuant to Rule 11.5 of the High Court Rules.
Registrar/Deputy Registrar
Date……………
Solicitors: Webb Farry, Auckland
LeeSalmonLong, Auckland
Stevely-Cole v Dilworth Trust Board [2020] NZHC 2843 [30 October 2020]
Introduction
[1] The applicants (the Trust) are the lessees of a property in Dilworth Avenue, Remuera in Auckland (the Property). The respondent (Dilworth) is the owner and lessor of the Property. The lease is a perpetually renewable ground lease. It provides for rent reviews every seven years.
[2] The most recent review period, and that in dispute in the present proceedings, is the annual rent to be paid for the seven-year period commencing on 12 August 2017 (the review period). The new rent is to be set by reference to the “fair annual rent” for the land. The rent for the previous seven-year period had been fixed at $125,000 per annum (plus GST). The parties were unable to agree on what the new fair annual rent should be. They agreed to refer that dispute to arbitration and appointed a retired Court of Appeal judge as arbitrator.
[3] The arbitrator subsequently issued an award dated 20 April 2020 setting the fair annual rent for the review period (the Award). In the arbitration, Dilworth contended that the new rent should be $364,000 per annum (plus GST), and the Trust said it should be between $125,000 and $157,500 per annum (plus GST). After a nearly two-week hearing in which a range of experts gave evidence, the arbitrator fixed the rent for the review period at $226,337 per annum (plus GST). The new rent accordingly fell slightly below the mid-point of the range suggested by the parties.
[4] The Trust is not happy with the outcome of the arbitration. But under the Arbitration Act 1996 (the Act), recourse against an arbitral award is limited to an application to set aside the award,1 and in some cases, to appeals on questions of law.2 Further, and unless otherwise agreed, a party must obtain leave of this Court to bring an appeal on a question of law.3 The parties did not agree in advance that they could appeal to this Court on a question of law and the Trust accordingly seeks leave to do so.
1 Arbitration Act 1996 [The Act], sch 1 art 34.
2 Schedule 2 cl 5(1).
3 Schedule 2 cl 5(1)(c).
[5] Consistent with the Act’s purpose of encouraging the use of arbitration to resolve commercial disputes and to clarify the limits of judicial review of the arbitral process and awards,4 the Act expressly excludes certain matters from the scope of an appeal on a question of law. Clause 5(10) of sch 2 to the Act defines a “question of law” as follows:
(10)For the purposes of this clause, question of law—
(a)includes an error of law that involves an incorrect interpretation of the applicable law (whether or not the error appears on the record of the decision); but
(b)does not include any question as to whether —
(i)the award or any part of the award was supported by any evidence or any sufficient or substantial evidence; and
(ii)the arbitral tribunal drew the correct factual inferences from the relevant primary facts.
[6] The effect of cl 5(10)(b) is accordingly to exclude what are often referred to as “perverse” factual findings from the scope of an error of law.
[7] The Trust seeks leave to appeal against the Award on what it says are two questions of law:
(a)First, the arbitrator erred in law in failing to apply the principle of the “prudent lessee” when setting the new rent for the review period.
(b)Second, the arbitrator failed to comply with art 31(2) of sch 1 to the Act, which requires an award to state the reasons upon which it is based. The Trust say the arbitrator failed to give any or adequate reasons for certain findings made in the Award and which were central to setting the new rent.
[8] Dilworth opposes the granting of leave. Among other matters, it says that the Trust’s real challenge (summarised at [7(a)] above is to factual findings made by the arbitrator and which cannot therefore form the basis of an appeal on a question of law.
4 Sections 5(a) and (d).
Dilworth further says the Award is well reasoned and the arbitrator’s chain of logic as to why he reached the conclusions he did is clear.
[9] Before going any further, it is important to recall that the Trust’s application is for leave to appeal against the Award only. Thus the central issue for determination is whether the relevant threshold for granting leave (discussed at [31]–[36] below) has been met, rather than resolving the merits of the proposed appeal.
[10]The balance of this judgment is structured as follows:
(a)First, I summarise the contents of the Award.
(b)Second, I set out the principles to be applied when determining whether leave ought to be granted for an appeal on a question of law.
(c)Third, I address the first proposed ground of appeal, including the threshold question of whether it in fact gives rise to a question of law.
(d)Finally, I address the second proposed ground of appeal, as to whether the Award adequately states the reasons upon which it is based.
The Award
[11] The arbitration extended over approximately two weeks of hearing time, during which a number of valuers and other experts gave evidence. A number of other rent review awards were also put before the arbitrator, including rent reviews in ground leases. In this context, and referring to commentary in the leading arbitration text Williams & Kawharu on Arbitration, the arbitrator noted that other arbitral decisions are a guide to but not a precedent for future arbitrations.5
[12] The arbitrator explained the approach to be taken to fixing the fair annual rent under a ground lease, namely that an assessment was to be made of the rent for the land only, and that buildings or other improvements on the land, and the actual use to
5 Arbitral award dated 20 April 2020 [the Award] at [13], citing Williams & Kawharu on Arbitration
(2nd ed, LexisNexis, Wellington 2017) at [13.9.2].
which the land is put by the current lessee, are to be disregarded. It was common ground before the arbitrator that the rent for the land was to be ascertained by reference to what a hypothetical, prudent lessee would pay for the land in question, on the terms and conditions of the actual lease of the subject land.
[13] The arbitrator then explained that there are generally two accepted approaches to determining ground lease rents, being the “classical” and “traditional” approaches:
(a)The classical approach involves considering comparable lettings, and then making adjustments for differences in time, physical factors (such as location, size and dimensions) and lease terms (including duration).6
(b)Where there is no or insufficient evidence of comparable lettings, the traditional approach involves determining a market value for the land, and then applying to that value a percentage appropriate to the circumstances to arrive at a rental figure (the rent rate).7
[14] The arbitrator noted that in Granadilla Ltd v Berben, Blanchard J had observed that because of the relatively few long-term ground rent leases, the traditional approach “may have become the predominant method of fixing ground rents on renewals of perpetual leases”.8 The arbitrator accordingly adopted the traditional approach to determine the new rent. There is no dispute it was appropriate for him to do so.
[15] It was also common ground that under the traditional approach, the market value of the property is set by reference to its “highest and best use” (that is, rather than by reference to how it is actually being used – in this case, as a childcare centre). Dilworth’s position was that the Property’s highest and best use was a commercial use. The Trust’s expert evidence was that the highest and best use was residential (and in particular, subdividing the land into four residential units). The arbitrator accepted the Trust’s position.
6 Granadilla Ltd v Berben (1999) 4 ConvC 192,963 (CA) at [14].
7 At [7].
8 At [7].
[16] The next step was for the arbitrator to arrive at a market value for the (unimproved) Property on the basis of its highest and best use, as summarised above. After assessing a considerable body of expert evidence as to comparable property sales, the arbitrator adopted a market value of $2,800 per square metre, resulting in a total property value of $4,765,600. No issue is taken with this aspect of the arbitrator’s conclusion.
[17] As highlighted at [13](b) above, the next step under the traditional approach is to fix the rental by applying a rent rate to the property’s market value. It is the steps taken by the arbitrator in this regard which lie at the heart of the Trust’s first proposed ground of appeal. Most of the experts agreed that commercial rent rates are traditionally higher than residential rent rates. The approach taken by the Dilworth valuer who advanced a specific rent rate (Mr Smithies) was to adopt a commercial rate of 6.5 per cent, based on a number of other new lettings he had reviewed (all of which were based on a 6.5 per cent commercial rate). The Award records Mr Smithies as stating that the differential between commercial and residential rates had “traditionally been 1 – 1.5 per cent”.9
[18] Given aspects of the Trust’s application discussed later in this judgment, it is useful to set out one paragraph of the Award which records Mr Smithies’ evidence on the appropriate rent rate:
[129] Consistent with this, Mr Smithies, in the course of cross-examination, agreed that a residential rate would require a different figure from commercial use. He was cross-examined on a report he had written for a ground rental assessment on the Sofitel site at 21 Viaduct Harbour Avenue, Viaduct Harbour. In the course of that cross-examination, he put forward figures of 4.5 – 5 per cent for residential properties. In the circumstances, he adopted a 4.5 per cent rental factor so as to reflect the highest and best use being residential or visitor accommodation, with only a small portion available for commercial use.
[19] The valuers called on behalf of the Trust did not give evidence of a commercial rate, but of what they considered to be an appropriate residential rate. One of the experts, a Mr Gamby, concluded that the range for developed residential property was between 2.01 per cent and 4.05 per cent, and that an appropriate residential rate in this case would be 3.5 per cent. The other valuer called by the Trust, Mr Dean, also
9 At [128].
adopted 3.5 per cent, based on comparative returns for developed residential leasehold property and the low risk associated with owning residential property. But he then discounted that rate further to a figure of 3.15 per cent, because of what he considered to be the “stigma associated with residential leasehold land”.10
[20] While Mr Gamby and Mr Dean did not advance a commercial rate, Mr Gamby stated in his reply brief of evidence that:
I have not provided an opinion on the commercial rental factor in my evidence but, were I to do so, it would likely be no higher than 5 per cent.
[21] The arbitrator ultimately adopted a “top down” approach by determining an appropriate commercial rate and then discounting it to reflect that the highest and best use in this case was residential. In doing so, the arbitrator also had regard to a number of other arbitral awards and the rent rates adopted in them. The arbitrator said he did not treat the other awards “in any way as precedents”, but noted that they showed how arbitrators approached the residential discount in a way similar to Mr Smithies.11
[22] The arbitrator accepted there was good reason for the residential discount and that it was widely recognised. He noted that Mr Smithies’ acceptance of a differential was “consistent” with the evidence of the Trust valuers.
[23] Again, it is useful to set out in full one aspect of the Award on this topic, given the issues discussed later in this judgment. The relevant aspect of the Award states:12
Mr Gamby stated that there was a basis for the differential for five reasons. First, residential lessee’s pay rent from tax paid income with the exception of investment apartments. Second, residential uses of property cannot claim back GST. Third, historically, residential returns on properties have been lower than commercial returns. Fourth, there is no aversion to undertaking a residential development on leasehold land. Fifth, a consideration of historic leases shows that residential lease rates are historically lower than those for commercial leases.
[24] The arbitrator also took into account evidence of a Mr Winstanley, called by Dilworth, stating as follows:
10 At [131].
11 At [135].
12 At [137].
[138] Mr Winstanley explained that if capital growth increases are not translated into equivalent rental increases at reviews, the lessor does not, in fact, benefit from capital gains. Mr Winstanley's evidence that there has been, of recent times, no material difference in total return between commercial and residential property in the catchment area if inflationary land value increases are taken into account, has no bearing on the factors requiring the residential rate to be lower than the commercial rate.
[139] Mr Winstanley, in his evidence, did not consider that the factors referred to by Mr Dean and Mr Gamby supported their conclusion reached about the rental rate for residential land. As I have indicated earlier, he takes the view that the return profiles for residential and commercial property have been very similar in respect of total returns for the medium to long term. This is because of the greater rate of inflation of residential property against commercial property, which meant that the owner of a residential property was enjoying a significant extra financial benefit which, in the end, evened out the differential and actual rent returns. He did not think it valid to say that residential investment property was less risky than other land. He considered that the tax issues were specialist areas which he certainly was unqualified to comment on.
[25] The arbitrator also referred to the evidence of a Mr Bagrie, an economist called by the Trust, which the arbitrator recorded as being “helpful”.13 Mr Bagrie did not agree with Mr Winstanley’s evidence to the effect that the stability of the 6.5 per cent commercial yield factor was broadly accepted. Mr Bagrie saw no reason why the yield factor on commercial land should remain constant, particularly in light of evolving economic considerations (including the large material fall in interest rates), and supported Mr Gamby and Mr Dean’s conclusions that residential land is worth less in income to a lessor than commercial land.
[26] Having discussed the experts’ evidence, the arbitrator then explained why he did not view the evidence on a commercial rate of 6.5 per cent to be persuasive (including that while that rate had been included in leases for a number of decades, it did not appear to be responding to market changes and rates of return, and thus may not reflect a “true rate”).
[27] The arbitrator did not accept Mr Dean’s evidence that the unpopularity of residential ground leases should be a “key factor in assessing a ground rental percentage”, because of the hypothetical nature of the exercise. The arbitrator also did not accept Mr Dean’s position that the rent rate would effectively be capped by (or
13 At [160].
“closely linked” to) the rate of return on occupied developed residential properties, given the hypothesis turned on the unique factor of the underlying lease being a ground lease only, with the actual developments on site being put to one side.14 In reaching this view, the arbitrator accepted Mr Winstanley’s evidence that landlord and tenant, when entering into a ground lease, would be aware of and accept that a hypothetical traditional approach would be applied to rent reviews in the event of a lack of comparable evidence. But given the tax consequences of commercial and residential property are different, the arbitrator did not accept Mr Winstanley’s view that there was no basis for any distinction between commercial and residential rates.15 Noting that Mr Smithies had also accepted a deferential between commercial and residential rent rate, the arbitrator accepted that a differential ought to be recognised. The key issue was then, what should that differential be?
[28] Given the Trust’s arguments (and its proposed appeal on an alleged failure to give adequate reasons), it is useful to set out the balance of the arbitrator’s discussion of the applicable rate in full:
[163] However, I am less influenced by historic factors referred to by Mr Gamby, or, as I have said, the unpopularity of ground leases emphasised by Mr Dean. I know that in other awards, a differential between 1–1.5 per cent has been accepted between commercial and residential, but taking into account my reservations about some of the factors relied on by the Mangahewa valuers, and the points made by Mr Winstanley, I would fix the differential as somewhat lower at 0.75–1 per cent.
[164] As I have said, Mr Smithies has referred to a number of ground lease rentals where the rate was 6.5 per cent or to be fixed in the future at 6.5%. For reasons that I have already traversed, I cannot accept that they are a sound indication of the real market ground percentage rates for residential properties, any more [than] the lower rates referred to by the Mangahewa Trust. Although there were general statements made by the Dilworth Avenue experts about the bona fides of these transactions, I was given no particular detail that could satisfy me of those bona fides, and I can only place limited weight on this evidence. I share the caution shown by the arbitrators quoted in the Lusk report, and the comments in that report. Similarly, Mr Gamby’s evidence about Melanesia Road was also so general and untested that I can place little weight on it.
[165] It is of course possible in some circumstances to get some reliable evidence on ground rental percentage rates. This can be done when there is evidence before an arbitrator proving a genuine arms-length transaction, where the rent has clearly been negotiated on a free market basis, and where
14 At [161].
15 At [162].
a land value can be fixed with confidence. This was to an extent available in the way referred to in the Award in D v M arbitration, in relation to what was known as the Giltrap settlement. In this case, as I have set out, I have no such relevant evidence that I am confident is reliable.
[166] What is clear evidence before me is first, the fact of falling interest rates and by and large investment returns in 2017, and second, the fact of the low yields on residential and commercial properties (although I accept that these are not to be treated as commensurate with ground rental percentage rates). Land values have surged, but as Mr Bagrie pointed out, the other core economic variables such as rents, inflation and wages, have not.
[167] To all this I must apply the obligatory end consideration of what it would be fair to expect a prudent lessee to pay, and a willing but not anxious lessor to accept.
[168] I am mindful of the fact that if Mr Smithies’ valuation was accepted, the tenant would have to meet an increase in rent of well over 150 per cent over seven years, in a context where inflation and the cost of living index have increased far more modestly and investment returns by and large have fallen on a relatively permanent basis. Equally, on Mr Dean’s evidence, there would be no increase at all, but that ignores the very significant increases on land value during that seven years. I bear in mind Mr Gamby’s analysis of the costs of running a rest home. A prudent lessee could not accept a rent so driven by high land values that it would make the investment uneconomic. Equally, a willing lessor can expect a reasonable return on the value of the land. These two contrary tensions must be recognised.
[169] Balancing the figures and these various considerations, and the lack of a reliable comparable in terms of rental percentage rate, in my view the correct rental rate, if this was a commercial property, would have been in the vicinity of 5.5 per cent. This being a residential property, and the land being valued on a residential basis, for the reasons I have given, it is appropriate to apply a discount to that 5.5 per cent percentage. In my view, that discount should be more modest than that indicated by the valuers who were prepared to recognise a differential, because as I have set out, I do not accept all the reasons for the differential put forward by Mr Gamby and Mr Dean, and I do not ignore Mr Winstanley’s evidence as to how the differential can smooth out when inflation and commercial reality are considered. I am also aware that on occasions hybrid rates have been fixed. Taking into account all the circumstances, I fix the differential at 0.75 per cent. The ground rental percentage factor will therefore be 4.75 per cent.
(footnotes omitted)
[29] Applying the residential rate of 4.75 per cent to the land value of $4,765,000, the arbitrator arrived at the new annual rent for the review period of $226,337.50 (plus GST).
[30] The arbitrator recognised that this represented a large increase (of approximately 80 per cent over seven years) from the earlier rental of $125,000 per annum. The arbitrator nevertheless concluded:16
The figure is justified in terms of what is fair, by the very great increase in land value, but considering the prudent lessee, the figure is far short of reflecting all of that increase. The land has increased in value at a considerably greater percentage rate than this rent reflects. Not all the burden of that increase must be accepted by the lessee, but part of it, inevitably under the approach taken to ground leases under the traditional approach, falls on it.
Applications for leave to appeal on a question of law – approach
[31] The principles applicable to an application for leave to appeal on a question of law were authoritatively stated by the Court of Appeal in Gold and Resource Developments (NZ) Ltd v Doug Hood Ltd.17 After a comprehensive survey of the history to the Act as well as the approach taken in other jurisdictions, Blanchard J, delivering the judgment of the Court, stated: 18
Parliament has made clear its intentions that parties should be made to accept the arbitral decision where they have chosen to submit their dispute to resolution in such a manner. It plainly intended a strict limitation on the involvement of the courts where this choice has been made.
[32] As discussed further below,19 amendments to the Act in 2007, which narrowed the scope of what constitutes a “question of law”, provide further evidence of Parliament’s intention that there are strict limitations on the Court’s involvement when parties have chosen to have their dispute resolved by arbitration.
[33] Turning to the Act’s provisions on appeals on questions of law, and in particular, when leave is required for such an appeal, cl 5(2) of sch 2 to the Act provides that the High Court:
… shall not grant leave … unless it considers that, having regard to all the circumstances, the determination of the question of law concerned could substantially affect the rights of 1 or more of the parties.
16 At [171].
17 Gold and Resource Developments (NZ) Ltd v Doug Hood Ltd [2000] 3 NZLR 318 (CA).
18 At [52].
19 At [48]–[49].
[34] In Gold and Resource Developments, the Court of Appeal stated that “[o]nce the statutory threshold has been passed, the Court should in each case exercise its discretion in a disciplined way”.20 It set out a number of factors to be taken into account in exercising that discretion, which it described as “guidelines to, rather than as governing, the exercise of the discretion.21
[35]The factors identified by the Court are as follows:
(a)The strength of the challenge/nature of point of law. The Court stated that a court should consider in a preliminary way the strength of the argument that there has been an error of law and the nature of that point. The Court noted:22
If it is a one-off point, in the sense that it is unlikely to occur again and cannot be seen as having any precedent value, either generally or to the parties on another occasion, then unless there are very strong indications of error leave should rarely be given. In other cases, the Court will be looking for a somewhat less stringent assessment. In those cases a strongly arguable case would normally be required for leave to be granted. The existence of conflicting decisions will also be relevant.
(b)How the question arose before the arbitrators. It will be relevant “whether the question of law arose incidentally, or whether it was the very point of the arbitration”.23 The Court commented that where the parties had full knowledge that the very point of the arbitration centred on a question of law and still chose to submit their dispute to arbitration before an arbitrator who was not legally qualified, the parties should generally be held to their choice.24
(c)The arbitrator’s qualifications. The Court stated that “[w]here the arbitrator chosen by the parties is legally qualified, it will be harder to obtain leave to appeal”.25
20 At [54].
21 At [54].
22 At [54(1)].
23 At [54(2)].
24 At [54(2)].
25 At [54(3)].
(d)The importance of the dispute to the parties.
(e)The amount of money involved.
(f)The amount of delay involved in going through the courts.
(g)Whether the underlying contract provides for the arbitral award to be final and binding, which “will indicate that the parties did not contemplate becoming involved in litigation over the arbitral award”.26
(h)Whether the dispute before the arbitrators is international or domestic.
[36] The first factor (the strength of the challenge/nature of point of law) was stated by the Court to be the most important, but other than that, the relevant factors were not listed in any particular order.27
Should leave be granted – preliminary points
[37] I am satisfied the statutory threshold that determination of the question of law could substantially affect the rights of one or more of the parties has been met. The first error of law alleged by the Trust in particular goes to the heart of the arbitrator’s approach to determining the fair annual rent for the Property for the review period. Further, the new rent determined in the Award represents a significant increase on the earlier rent, as the arbitrator acknowledged. Given the lease contains a ratchet clause, the new rent will also impact future rent reviews.
[38] I am also satisfied that whether an award adequately states the reasons on which is it based is a matter central to the arbitral process and its outcome, such that a failure to comply with this requirement could substantially affect the rights of one or more of the parties (and in particular, the unsuccessful party). The key issue is, therefore, whether I ought to exercise my discretion to grant leave to appeal on one or both of the proposed grounds of appeal.
26 At [54(7)].
27 At [54].
[39] As a second preliminary point, I am also satisfied that the matters in issue in the proposed appeal, at least in relation to the proposed first ground, are not a “one off point” in the sense described in Gold and Resource Developments. I say this because:
(a)The new rent for the review period will impact the parties for the next seven years, a reasonably lengthy period of time.
(b)As noted, the lease contains a ratchet clause, such that the rent determined by the Award will be “locked in” to future rent reviews.
(c)The childcare centre located on the Property straddles the Property and an adjoining piece of land, which is also subject to a lease on virtually the same terms as that in issue in this case. It is unrealistic to expect that the rent set by the Award will not be taken into account in the forthcoming rent review for the adjoining land.
(d)Finally, though not as material as the preceding points, there is no doubt that rent review awards such as that in this case have some “precedential” effect in relation to other similar (Glasgow) ground leases. The pool of valuers assessing rents under such leases in Auckland is relatively small. Despite the confidentiality attaching to arbitral awards, the outcome in Glasgow ground lease rent reviews become known in the valuation industry; and as is evident from the face of the Award itself, outcomes in other rent review processes are referred to and taken into account in Glasgow ground lease rent reviews generally.
[40] Because the first alleged error of law is not a “one off”, I will adopt the test of a “strongly arguable case” when assessing whether to grant leave.
First ground of appeal – alleged failure to apply the principle of the prudent lessee
The parties’ submissions
[41] The Trust argues the arbitrator reached a conclusion that was so outside the range of available outcomes based on the evidence before him that it can be inferred he misapplied the principle of the prudent lessee. The Trust says the misapplication of a legal principle is itself a question of law. It relies in this context on observations in decisions such as Milk New Zealand (Shanghai) Co Ltd v Miraka Ltd and Todd Petroleum Mining Company Ltd and Shell (Petroleum Mining) Company Ltd v Vector Gas Trading Ltd (discussed later in this judgment at [57] to [58] below).
[42] This aspect of the Trust’s proposed appeal focuses on the rent rate adopted by the arbitrator, namely 4.75 per cent. This was based on a commercial rate of 5.5 per cent (which Mr Miles QC, counsel for the Trust, accepts the Trust cannot take issue with, falling between Mr Gamby’s assessment of 5 per cent and Mr Smithies’ 6.5 per cent) and then applying a discount of 0.75 per cent. The Trust says this outcome falls well outside the range of two reasonable or possible outcomes on the evidence, namely:
(a)a residential rent rate of between 3.15 per cent (Mr Dean’s evidence) and 4.5 per cent (said to be Mr Smithies’ evidence) (with Mr Gamby’s figure of 3.5 per cent falling much nearer to Mr Dean’s assessment); and
(b)a discount on commercial rent rates ranging from 1 to 1.5 per cent (Mr Smithies’ evidence).
[43] The Trust says that given there was no evidential basis upon which the arbitrator could have adopted a discount from commercial rates of only 0.75 per cent, it can be inferred that something must have gone wrong in his application of the law. The Trust says that this issue is at least seriously arguable.
[44] In response, Dilworth first says that, irrespective of the merits of the Trust’s arguments, the matters raised by this aspect of the Trust’s proposed appeal do not give
rise to an error of law. Mr Salmon, counsel for Dilworth, says that the arbitrator’s assessment of the experts’ respective evidence and his adoption of the percentage rates discussed above plainly involved factual findings and assessment of expert evidence which simply cannot give rise to an error of law.
[45] Second, Dilworth disputes that it is seriously arguable the arbitrator adopted a discount from commercial rates, or a residential rate, that fell outside the available range in any event, even if this could give rise to a question of law. Mr Salmon emphasises Mr Winstanley’s evidence, which the arbitrator plainly took into account,28 that there ought to be no differential between commercial and residential rent rates. Thus Mr Salmon says the “floor” of the available range for the differential was not
1.0 per cent as suggested by the Trust, but in fact zero per cent.
[46] And while it is clear the arbitrator accepted there ought to be a differential, Mr Salmon says the arbitrator plainly took into account Mr Winstanley’s evidence in adopting a discount to the commercial rate that was “more modest than that indicated by the valuers who were prepared to recognise a differential”.29 Mr Salmon notes that the arbitrator explained that he did so because, among other matters, he did not accept all the reasons advanced by Mr Gamby and Mr Dean for the differential. Mr Salmon also notes that the overall result in the arbitration, namely a new rent for the review period, fell squarely within the range contended for by the parties, and in fact closer to the rent proposed by the Trust than Dilworth.
Is the proposed first ground of appeal a question of law – principles
[47] Despite Parliament’s clear intention that only questions of law are capable of being appealed to this Court from an arbitral award, what is and is not a question of law for this purpose has been (and arguably remains) somewhat unclear.
[48] As the authors of Williams & Kawharu on Arbitration note, prior to amendment of the Act in 2007 (to include cl 5(10) of sch 2, as set out at [5] above),
28 See that extract of the Award set out at [28] above.
29 At [169].
there were three categories of issue in respect of which the courts have allowed appeals under the label “question of law”:30
(a)a conventional legal question on unchallenged facts;
(b)a positive factual finding unsupported by any evidence; and
(c)a failure to draw the only reasonable [factual] inference from primary facts.
[49] Given the confusion and conflicting decisions on what did and did not amount to an error of law, the topic was considered as part of the Law Commission’s review of the Arbitration Act in its 2003 Report “Improving the Arbitration Act 1996”.31 The Commission noted that in Gold and Resource Developments, the Court of Appeal had seen force in the argument that a perverse finding of fact (characterised by a tribunal reaching factual findings without evidence, or upon a view of the facts which could not reasonably be entertained) ought to be excluded from the scope of an appeal on an error of law.32 Reflecting parties’ choice of arbitration rather than litigation as the means of resolving their dispute and the benefits of finality, the Commission recommended that perverse factual findings be expressly excluded from the scope of a question of law.33 As noted, this led to the enactment of what is now cl 5(10)(b) of the Second Schedule to the Act.
[50] But difficulties remain, particularly in relation to findings or decisions which can be characterised as involving mixed questions of fact and law.
[51] A useful starting point in this context is the Supreme Court of Canada’s identification of the differences between questions of fact and law and what constitutes a mixed question of law and fact. The Court explained the differences as follows:34
30 Williams & Kawharu on Arbitration, above n 5, at [18.4.2], citing Auckland City Council v Wotherspoon [1990] 1 NZLR 76 (HC) at 85–86.
31 Law Commission Improving the Arbitration Act 1996 (NZLC R83, 2003) [Law Commission Report].
32 At [116], citing Gold and Resource Developments (NZ) Ltd v Doug Hood Ltd, above n 17, at 335.
33 At [122]–[125].
34 Canada (Director of Investigations and Research) v Southham Inc [1997] 1 SCR 748 at [35].
Briefly stated, questions of law are questions about what the correct legal test is; questions of fact are questions about what actually took place between the parties; and questions of mixed law and fact are questions about whether the facts satisfy the legal tests. A simple example will illustrate these concepts. In the law of tort, the question what “negligence” means is a question of law. The question whether the defendant did this or that is a question of fact. And, once it has been decided that the applicable standard is one of negligence, the question whether the defendant satisfied the appropriate standard of care is a question of mixed law and fact.
[52] Mustill J (as he then was) articulated a similar approach in Finelvet AG v Vinava Shipping Co Ltd; The Chrysalis, dividing the arbitrator’s task into three stages:35
(1) The arbitrator ascertains the facts. This process includes the making of findings on any facts which are in dispute. (2) The arbitrator ascertains the law. This process comprises not only the identification of all material rules of statute and common law, but also the identification and interpretation of the relevant parts of the contract, and the identification of those facts which must be taken into account when the decision is reached. (3) In the light of the facts and the law so ascertained, the arbitrator reaches his decision.
[53]Mustill J went on to state:36
In some cases, stage (3) will be purely mechanical. Once the law is correctly ascertained, the decision follows inevitably from the application of it to the facts found. In other instances, however, stage (3) involves an element of judgment on the part of the arbitrator. There is no uniquely “right” answer to be derived from marrying the facts and the law, merely a choice of answers, none of which can be described as wrong.
…
Whether stage (3) can ever be the proper subject of an appeal, in those cases where the making of the decision does not follow automatically from the ascertainment of the facts and the law, is not a matter on which it is necessary to express a view in the present case. Pioneer Shipping Ltd v BTP Tioxide Ltd [1982] AC 724 and Kodros Shipping Corp v Empresa Cubana de Fletes, The Evia [1983] 1 AC 736, show that where the issue is one of commercial frustration, the court will not intervene, save only to the extent that it will have to form its own view, in order to see whether the arbitrator's decision is out of conformity with the only correct answer or (as the case may be) lies outside the range of correct answers. This is part of the process of investigating whether the arbitrator has gone wrong at the second stage. But once the court has concluded that a tribunal which correctly understood the law could have arrived at the same answer as the one reached by the arbitrator, the fact that the individual judge himself would have come to a different
35 Finelvet AG v Vinava Shipping Co Ltd; The Chrysalis [1983] 1 WLR 1469 (QB) at 1475.
36 At 1475.
conclusion is no ground for disturbing the award. (emphasis added)
[54] Coming back to the amendments to the Act in 2007, the Commission made it clear that its recommendations did not extend to appeals on questions of mixed fact and law:37
This chapter is not directly concerned with appeals on questions of mixed fact and law; that is, where it is alleged that the facts found by the arbitral tribunal cannot support the legal description given to them. Such appeals do not challenge the tribunal's findings of fact; rather its application of the law to these facts. Not every appeal involving an allegedly erroneous legal conclusion will involve an error of law. This is because it is possible (and not uncommon) for judges and arbitrators to come to different legal conclusions on the same facts, without making any error of law. Whether there will be an error of law depends on the reasonableness of the legal conclusion. As to the test to be applied see Edwards (Inspector of Taxes) v Bairstow [1956] AC 14, 36 (HL) Lord Radcliffe: there would be an error of law if "the true and only reasonable conclusion contradicts the determination”.
(emphasis added)
[55] This demonstrates that a legal conclusion or decision reached by an arbitrator (or to frame it another way, the outcome arrived at by applying the law to the facts) which is inconsistent with the only true and reasonable outcome based on the arbitrator’s factual findings, or which falls outside the only reasonable and possible range of outcomes based on the arbitrator’s factual findings, is probably not excluded from the scope of a question of law under the Act. Rather, such outcomes point to something having gone wrong at the “second stage” of the arbitrator’s task, as set out in the extract from the judgment in Finelvet at [52] above.38 But plainly an arbitrator’s factual findings or inferences which are inconsistent with the primary evidence, or which fall outside the range of reasonable findings or inferences to be drawn from the primary evidence, are expressly excluded by cl 5(10). This distinction between perverse factual findings and perverse legal conclusions is important, given Parliament’s express intent that even perverse factual findings are no longer amenable to challenge.
37 Law Commission Report, above n 31, at [115], fn 167.
38 And the “something going wrong” at the second stage might be an incorrect statement or identification of the legal principle, or as Walker J in Milk New Zealand (Shanghai) Co Ltd v Miraka Ltd [2019] NZHC 2713 observed (at [54]), if one or more of the principles was misapplied or not applied at all.
[56] The authors of Williams & Kawharu on Arbitration thus summarise the position as follows:39
The Commission only recommended exclusion of perverse factual findings from cl 5, hence whether the arbitral tribunal drew the correct factual inferences from the relevant primary facts is no longer a potential ground of appeal by virtue of cl 5(10)(b)(ii). Therefore, an inquiry into whether the factual findings of the tribunal support its legal conclusion may be permitted under cl 5 on the basis of the Edwards (Inspector of Taxes) v Bairstow principle left untouched by the 2007 reforms to the NZ Act. If the legal decision of the tribunal cannot be supported in light of its findings of fact, the assumption would be that this flows from a misunderstanding of the law.
(emphasis in original)
[57] This approach was recognised by Venning J in Shell (Petroleum Mining) Company Ltd v Vector Gas Contracts Ltd.40 One of the issues Venning J considered was whether the Tribunal’s findings on the relevance of the parties’ post contract conduct on the proper interpretation of contract could give rise to an error of law. Having referred to Mustill J’s observations in Finelvet, Venning J stated:41
The [Tribunal’s] underlying factual findings about the post contractual conduct were that the parties held a variety of conflicting views. On the basis of those primary factual findings, which this Court cannot go behind, it was open to the Tribunal to find it could not draw any useful conclusions from such conduct, indeed such a finding was inevitable.
Even if this were to be regarded as a question of mixed law and fact, it is one which lends itself to more than two possible outcomes. In other words, the Tribunal could have taken a number of views about the effect of the various instances of post contractual conduct. The authorities suggest in such a case that the Court should only interfere in circumstances where the outcome challenged is effectively outside the realm of reasonable outcomes.
[58] Williams J adopted similar principles in Todd Petroleum Mining Company Ltd v Vector Gas Trading Ltd.42
39 Williams & Kawharu on Arbitration, above n 5, at [18.4.3].
40 Shell (Petroleum Mining) Company Ltd v Vector Gas Contracts Ltd [2014] NZHC 31.
41 At [47].
42 Todd Petroleum Mining Company Ltd v Vector Gas Trading Ltd [2017] NZHC 1166 at [48]–[51].
Proposed first ground of appeal a question of law – discussion
[59] I do not consider the Trust’s proposed first ground of appeal involves a question of law, or that it is seriously arguable that it does. Further, even if the first ground does involve a question of law, I do not consider it to be seriously arguable in any event.
[60] The arbitrator’s overall task was to determine the new rent to be paid over the review period. There is no dispute the arbitrator identified the correct principle to be applied, namely the prudent lessee test. And there is no doubt that determination of what a prudent lessee might pay by way of rent during the review period required an element of judgement on the part of the arbitrator; that is, there was no uniquely right answer to be derived from marrying the facts and the law.43
[61] Ultimately the Trust’s evidence and submissions were that a prudent lessee would pay no more than $157,500 (plus GST) per annum during the review period. Dilworth’s evidence and submissions were that a prudent lessee would pay $364,000 (plus GST) per annum. The arbitrator’s conclusion that a prudent lessee would pay
$226,337 (plus GST) per annum accordingly fell well within the range of available outcomes identified by the parties and based on the evidence advanced by each of them. In other words, it is not the case that had the prudent lessee test been applied correctly, the only possible outcome was something less than the new rent level determined by the arbitrator.
[62] Turning to the arbitrator’s finding that the differential between commercial and residential rates was 0.75 per cent, this was a factual finding based on the arbitrator’s assessment of the experts’ evidence. Mr Miles accepted that this, and the other rates and percentages adopted by the arbitrator, were factual findings. His central proposition was nevertheless that the finding fell so far outside the range of available conclusions that it gives rise to an error of law.
[63] In my view, the Trust’s arguments on this aspect of its application blur the important distinction between perverse factual findings (excluded by cl 5(10)), and a perverse outcome once the law has been applied to the facts (as they have been found
43 Finelvet AG v Vinava Shipping Co Ltd; The Chrysalis, above n 35, at 1475.
by the arbitrator). As the discussion at [55] above highlights, factual findings which fall outside the range of available conclusions arising from the raw or underlying evidence cannot give rise to a question of law for the purposes of the Act. In this case, the Trust wants to argue that there was no evidential foundation for the arbitrator’s finding that the differential between commercial and residential rates was 0.75 per cent. But such an argument is expressly excluded by cl 5(10)(b) of sch 2. Even if it were a perverse factual finding, it is one which the parties are bound by.
[64] If I am wrong that the first ground of appeal does not give rise to a question of law (or that it is not seriously arguable that it does so), I consider the Trust’s argument that the residential rate of 4.75 per cent fell outside the available range is weak.
[65] As noted, the arbitrator adopted a top down approach, adopting a commercial rate before applying a differential to arrive at a residential rate. The starting point of
5.5 per cent was clearly within the available range. As noted, Mr Miles accepts that. And once Mr Winstanley’s evidence (that there should be no differential) is factored into the analysis, it is not seriously arguable in my view that the differential of 0.75 per cent fell outside the available range. It is correct that the arbitrator did not accept Mr Winstanley’s evidence that there should be no differential. But he clearly (and expressly) attached some weight to Mr Winstanley’s evidence that the differential can smooth out when inflation and commercial reality are considered, which (in part) led him to conclude that the differential should be less than that proposed by the valuers who recognised a differential.
[66] Mr Miles further submits that it is seriously arguable that the arbitrator’s finding of a residential rate of 4.75 per cent also fell outside the available range of residential rates discussed in the evidence, being 3.5 per cent (as per Mr Dean and Mr Gamby) and 4.5 per cent (as per Mr Smithies). But it is not clear on the face of the Award that Mr Smithies’ evidence was in fact that a residential rate in this case ought to be 4.5 per cent. Rather, the relevant extract of the Award could equally be a reference to Mr Smithies’ evidence on the appropriate rate in another arbitration concerning different properties. Mr Smithies adopting a residential rate in this case of
4.5 per cent would certainly have been inconsistent with his evidence that the appropriate commercial rate was 6.5 per cent and that a differential for residential rates
would be no more than 1 to 1.5 per cent. Given this, if Mr Smithies had indeed accepted a residential rate of 4.5 per cent in cross-examination, it is surprising that the arbitrator did not address this in the Award, despite his otherwise detailed discussion of the experts’ respective positions.
[67] For completeness, Mr Miles also urged me to accept that the Award indicates that the arbitrator considered Mr Smithies’ position on the appropriate residential rate as being consistent with that of Mr Gamby and Mr Dean, namely at or around 3.5 per cent. This was on the basis of the arbitrator’s statement that:44
I have already referred to Mr Smithies’ evidence, where he accepted a differential between a residential and a commercial rental percentage factor. His acceptance of a differential was consistent with the evidence of the Mangahewa Trust valuers, Mr Gamby and Mr Dean.
[68] Mr Miles says this reinforces that the arbitrator’s adoption of a residential rate was clearly outside the range, given the consistency between Messrs Gamby, Dean and Smithies. But I do not consider it seriously arguable that the Award can be read in the way Mr Miles suggests. All the above extract suggests is that Mr Smithies’ accepted that there should be a differential between commercial and residential rates, but not what that differential should be. That this is so is reinforced by the fact that this aspect of the Award falls under the heading “…whether there should be a differential between a commercial and residential rate”, rather than under the separate and later heading and discussion of “[t]he rate”. And again, if Mr Smithies had adopted a residential rate similar to that of Mr Gamby and Mr Dean, this would have been quite inconsistent with his evidence of a commercial rate of 6.5 per cent and a differential of 1 to 1.5 per cent, or alternatively, the Trust’s argument that Mr Smithies adopted a rate of 4.5 per cent.
[69] Standing back, I view this aspect of the Trust’s proposed appeal to be primarily directed at the arbitrator’s assessment of and weight given to the various experts’ evidence. The Court cannot and should not intervene in such assessments under the guise of a question of law.
44 At [137].
[70] Turning to those other factors to be considered in exercising the discretion on whether to grant leave:
(a)How the question arose before the arbitrator: This also weighs against granting leave. The suggested point of law did not arise incidentally or by chance at the hearing; rather the assessment of competing expert evidence on the factors to be taken into account under the traditional approach to ground lease rent reviews was central to the issue between the parties, which they chose to submit to arbitration. As noted by the Court of Appeal in Gold and Resource Developments, in such circumstances, the parties should generally be held to their choice.45
(b)The qualifications of the arbitrator: The arbitrator is legally qualified (and is in fact a retired Court of Appeal judge), and well versed in hearing and assessing competing expert evidence. In such circumstances, it will be harder to obtain leave.
(c)The remaining factors: I accept the amount of money involved is not insignificant, and in the overall context of the rent review process, the delay involved in an appeal to this Court would not be determinative. But I do not consider these factors tip the balance in favour of granting leave, at least when balanced against the issue of whether the alleged error is an error of law, and even if so, the strength of the Trust’s argument.
(d)The lease itself does not expressly state that the Award was to be final and binding. This can weigh in favour of granting leave. Mr Salmon noted that the Public Bodies Leases Act 1969 provides that any arbitration of a dispute on a rent review will be binding on the parties. But in this case, the parties had in fact entered into a “bespoke” arbitration agreement referring the determination of the fair annual rent for the review period to arbitration before a single arbitrator. This bespoke arbitration agreement did not include a statement that the
45 Gold and Resource Developments (NZ) Ltd v Doug Hood Ltd, above n 17, at [54(2)].
award would be final and binding. But I do not see this factor as particularly material or determinative. Even if words such as “final and binding” are included in an arbitration agreement, a party disappointed with the outcome of an arbitration will always be able to seek leave to appeal on a question of law (unless the parties have expressly contracted out of such appeals). This factor carries much less weight, in my view, than the strength of the challenge and those factors discussed at (a) and (b) above.
[71] Taking these factors into account, I decline to grant leave for the Trust to bring an appeal in terms of paragraph 2(a) of the application for leave dated 17 July 2020.
Second proposed ground of appeal – alleged failure to give reasons
Requirement to give reasons – approach
[72] Pursuant to art 31(2) of sch 1 to the Act, an award is required to state the reasons upon which it is based (unless the parties have agreed otherwise).
[73] The Court of Appeal considered the adequacy of reasons in an arbitral award in Casata Ltd v General Distributors Ltd46 and Ngāti Hurungaterangi v Ngāti Wahiao.47
[74] The arbitration in Casata, like in this case, was a rent review arbitration. The arbitration was before a panel of three arbitrators. While it is not stated in either the High Court or Court of Appeal judgments, it appears the Tribunal may have been comprised of three experts (presumably valuers), rather than legally qualified arbitrators. A number of alleged errors of law were pursued by the appellant, one of which was whether the arbitrators had erred in failing to give reasons for preferring the evidence of certain of the lessees’ witnesses over the competing evidence of the lessor’s witnesses.
46 Casata Ltd v General Distributors Ltd [2005] 3 NZLR 156 (CA).
47 Ngāti Hurungaterangi v Ngāti Wahiao [2017] NZCA 429, [2017] 3 NZLR 770.
[75] Ellen France J in the High Court found no error in the arbitrators’ approach, on the basis the evidence had been considered in detail and in that context, it could not be said that more was needed in terms of the challenged aspects of the reasoning.48 On appeal, the Court of Appeal summarised Ellen France J’s decision on this aspect of the claim (which it endorsed) as being that it was not a case where the arbitrators “gave no reasons at all for their overall approach and the preference for [the evidence in question was, as the Judge] said, explicable in the context of the decision as a whole”.49
[76] The Court of Appeal also referred to the English Court of Appeal’s decision in Flannery v Halifax Estate Agencies Ltd, in which that Court had emphasised that the nature of the case would affect the nature and extent of the reasons needed.50 The Court in Casata also observed that reasons given by an arbitral tribunal must not be “so economical that party is deprived of having an issue of law dealt with by the Court if necessary”.51 But, the Court confirmed that there is no requirement for arbitrators to “give elaborate reasons for each and every component of the award.”52
[77] In Ngāti Hurungaterangi v Ngāti Wahiao,53 the three-member arbitral tribunal comprised a retired Supreme Court Judge (as chair of the Tribunal), a widely respected leader of Ngāti Hine and Ngāpuhi and an authority on tikanga Māori, together with a member of Ngāti Hine, Ngāti Whātua and Tainui (who the Court described as “another distinguished leader with impressive credentials in charitable and environmental governance”)54. The essence of the dispute referred to arbitration was a dispute between Ngāti Whakaue and Ngāti Wahiao as to which of those iwi was entitled to certain lands which the Crown had agreed to return to iwi. The Court of Appeal noted that these matters were accordingly of great significance to the iwi involved.55
48 General Distributors Ltd v Casata Ltd [2004] 2 NZLR 824 (HC) at [131].
49 Casata Ltd v General Distributors Ltd, above n 46, at [33].
50 At [91], citing Flannery v Halifax Estate Agencies Ltd [2000] 1 WLR 377 (CA).
51 At [89].
52 At [90].
53 Ngāti Hurungaterangi v Ngāti Wahiao, above n 47.
54 At [4].
55 At [70].
[78] The hearing before the Tribunal extended over 13 sitting days and extensive evidence was called of both an oral and documentary nature.
[79] In the High Court, the respondent had alleged a number of errors of law, including that the Tribunal had failed to give adequate reasons for the conclusions it reached. The High Court found (by a “fine margin”56) that the Tribunal’s reasoning was sufficient. On appeal, the Court of Appeal described the duty to give reasons in the following way:57
[61] The purpose of the arbitral obligation to give reasons merits restatement. Within the arbitral framework for determining competing rights and obligations, the reasons explain how the adjudicator progressed from a particular state of affairs to a particular result. The reasons are the articulation of the logical process employed by a person whose particular skills, expertise or qualification the parties have chosen to decide their dispute. The reasons expose to the parties the disciplined thought pattern of the specialist adjudicator, thereby dispelling any suggestion of arbitrariness. A requirement to give reasons concentrates the mind; if it is fulfilled, the resulting decision is much more likely to be soundly based on the evidence than if it is not.
[62] In Flannery v Halifax Estate Agencies Ltd the English Court of Appeal observed that the duty to give reasons is a function of due process, and therefore of justice. Fairness requires that the parties, especially the disappointed party, should be left in no doubt why they have won or lost or their expectations have otherwise been frustrated. Without reasons the disappointed party will not know whether the panel has misdirected itself, and thus whether he or she may have an available right of appeal. These observations were made in the context of determining whether a Judge at first instance was required to give reasons for a conclusion essential to his decision
— in that case for preferring one expert witness to another. The principle holds equally true, however, for the arbitral process. As we are about to explain, the nature and extent of the reasons required to fulfil this function varies according to the context. But the underlying purpose for which reasons are necessary remains common to both processes.
(footnotes omitted)
[80] The Court of Appeal nevertheless emphasised that the leading authorities on the requirement to give reasons stand for a common theme, namely that the nature and extent of the duty to give reasons necessarily imports a degree of flexibility according to the circumstances, including the subject matter arbitrated, its significance to the
56 Ngāti Hurungaterangi v Ngāti Wahiao [2016] NZHC 1486, [2016] 3 NZLR 378 at [136].
57 Ngāti Hurungaterangi v Ngāti Wahiao, above n 47.
parties and the interests at stake.58 The Court stated that there was no minimum criterion (or need to “satisfy the curial standard”), except that the reasons:59
… must be coherent and comply with an elementary level of logic of adequate substance to enable the parties to understand how and why the arbitrator moved in the particular circumstances from the beginning to the end points. They must engage with the parties’ competing cases and the evidence sufficiently to justify the result…
[81]The Court accepted that:60
…brevity is often acceptable in an arbitral panel’s assessment of evidence and factual findings, reflecting the principles of arbitral finality and party autonomy underpinning the Arbitration Act. … However, … brevity is acceptable provided that the circumstances justify it and there is reasonable clarity about the core basis for the conclusion.
[82] Ultimately, the Court stated that an award’s reasons need to be “sufficiently full for the parties to understand the pathway taken by the panel to explain the result.”61
[83] Applying these principles to the award in issue, the Court of Appeal concluded that the Tribunal’s reasons were inadequate and allowed the appeal. The Court described the Tribunal’s reasons as:62
…essentially conclusory in nature and to the extent that they purport to explain the result they are so inadequate and inconsistent that they fall short of discharging the panel’s mandate to give a reasoned award. The reasons are not commensurate with the importance of the subject matter and the panel’s conclusion.
[84] The Court set aside the award. It observed that, absent agreement otherwise, “the trustees will be required to constitute a new panel to determine [the dispute] in accordance with the requirements of the Trust Deed and the Arbitration Act”.63
58 At [63].
59 At [63].
60 At [69].
61 At [75].
62 At [104].
63 At [110].
Adequacy of reasons – the parties’ submissions
[85] The Trust says the arbitrator erred in failing to provide any or adequate reasons for:
(a)fixing the residential rate at 4.75 per cent;
(b)setting a notional commercial rate at 5.5 per cent; and
(c)applying a discount on the notional commercial rate of 0.75 per cent.
[86] It is clear, however, that the residential rate at (a) above was the product of deducting the percentage rate at (c) from (b). As such, it was the arbitrator’s articulation of why he adopted the rates at (b) and (c) above which was the focus of the Trust’s written and oral submissions.
[87] The Trust highlights the Court of Appeal’s observation in Ngāti Hurungaterangi that a lengthy recital of the parties’ respective cases and of some of the evidence does not assist in understanding the arbitrator’s reasoning, nor is it a substitute for identifying the true issues.64 The Trust acknowledges that the Award is lengthy, but says much of this is a recital of the parties’ cases and evidence, rather than the arbitrator’s reasons. The Trust also emphasises the Court of Appeal’s statement in Ngāti Hurungaterangi that an arbitrator’s discretion as to the admissibility and weight to be given to certain evidence does not absolve the arbitrator from stating why he or she preferred certain evidence and disregarded other expert evidence. The Trust accepts that the arbitrator’s task in this case involved evaluation of a range of expert opinions, but says that the arbitrator did not articulate, either at all, or adequately, why he rejected some aspects of Mr Gamby and Mr Dean’s evidence, while accepting other parts of it.
[88] In the context of the authorities referred to earlier, Mr Miles emphasises that the Award might describe how the arbitrator reached the residential rate of 4.75 per cent, but fails to provide adequate reasons for why that outcome was arrived at.
64 Ngāti Hurungaterangi v Ngāti Wahiao, above n 47, at [80].
[89] In terms of the commercial rate of 5.5 per cent adopted by the arbitrator, while the Trust accepts this fell within the available range, it submits the arbitrator failed to deal with and give reasons why Mr Gamby’s evidence that he would have adopted a commercial rate of no higher than 5 per cent was rejected. The Trust says that this was particularly necessary given the arbitrator accepted other aspects of Mr Gamby’s evidence.
[90] The Trust further submits that the arbitrator failed to give any or at least adequate reasons for why he adopted the differential of 0.75 per cent. Acknowledging that the arbitrator stated he had reservations about some of the factors relied on by the Trust’s valuers, and that he did not ignore Mr Winstanley’s evidence, it says the arbitrator did not articulate why he rejected the historical factors relied on by Mr Gamby in particular. The Trust also says that while the arbitrator might have given some reasons for why he rejected Mr Dean’s approach based on the unpopularity of residential leases, that matter did not go to the heart of Mr Dean’s argument.
[91] The Trust also points to the arbitrator’s acceptance of Mr Bagrie’s evidence, who had disagreed with Mr Winstanley’s evidence. In those circumstances, the Trust says it was incumbent on the arbitrator to articulate more fully why he nevertheless relied on Mr Winstanley’s evidence in arriving at the differential rate. The Trust also says the arbitrator failed to provide any reason why he rejected Mr Smithies’ evidence of a differential of 1 to 1.5 per cent, and did not even refer to that evidence when fixing the 0.75 per cent.
[92] Dilworth describes the Trust’s argument on the adequacy of reasons as a “stretch at best”. Dilworth says that given the considerable reasoning and analysis of the evidence in the Award, coupled with the nature of the decision-making in question (requiring balancing judgements to be made on competing quantitative and qualitative factors), “it is difficult to conceive of any proper basis for interfering with the arbitrator’s determination”.
[93] Dilworth notes that the relevant context to the requirement to give reasons in this case is the assessment of a fair market rental for a ground lease. It notes that such arbitrations are regularly conducted by non-legally trained valuers in informal
arbitration settings without lawyers present, with accordingly very limited arbitral reasoning provided. It says that detailed workings of all determinations, many of which weigh competing factors in the round, and include judgements made by the arbitrator in assessing valuation positions, are generally not appropriate in this context and in some instances cannot be provided.
[94] In terms of the arbitrator’s assessment of the differential rate, Dilworth emphasises that the arbitrator’s assessment necessarily involved weighing up various quantitative and qualitative factors, and that the arbitrator specifically reasoned that:
(a)the position advanced by Mr Smithies of a commercial rate of 6.5 per cent was too high;
(b)there were falling interest rates and large investment returns in 2017;
(c)there were low yields on residential and commercial properties (though noted that the yields are not a proxy for ground rental percentage rates); and
(d)land values had increased sharply in recent times, but core economic variables such as rents, inflation and wages had not.
[95] Dilworth submits that when the Award is read as a whole and in context, it is very clear the arbitrator has identified and grappled with the issues arising from the competing cases, and the pathway taken by the arbitrator to explain his overall result is clear.
Requirement to give reasons – discussion
[96] In my view, three principles provide important context to assessing the adequacy of reasons in an arbitral award:
(a)First, the context of the arbitration in question. The authorities discussed earlier make this clear. I bear in mind that the context to this arbitration is a rent review, in which the arbitrator is to determine what
a prudent lessee would pay by way of rent in the review period. As the expert evidence given at the arbitration made clear, there could be a range of reasonable alternative outcomes, depending on the overall assessment of and weight given to each expert’s evidence.
(b)Second, the courts must in my view be cautious about requiring arbitral awards to exhibit the same characteristics and detail, in terms of reasoning, as a judgment of a court. I consider this to be so even if a retired judge has been appointed as the arbitrator. Despite appointing a retired judge as arbitrator, the parties have chosen not to go down the litigation path, and want their dispute to be resolved in a different and presumably more efficient and flexible way.
(c)Third, it is also clear that an arbitrator (like a judge) is not required to give reasons, or detailed reasons, on every aspect of the evidence or case put before them. Ignoring this qualification invites a microscopic analysis of arbitral awards to see if an arbitrator has commented on and given reasons for accepting, or rejecting (as the case may be), each item of evidence, and each proposition put to him or her by the parties. Such an approach would be undesirable, and inconsistent with the principles of party autonomy and the finality of arbitral awards.
[97] Against this background, I do not consider it seriously arguable that the arbitrator failed to give any or adequate reasons. My own reasoning follows.
[98] First, it is correct that Mr Gamby expressed a view in his written (reply) brief that had he assessed a commercial rate, he would have assessed it at no more than 5 per cent. But Mr Gamby’s evidence was not directed to a commercial rate, but rather to a residential rate. It was nevertheless quite open to the arbitrator to adopt a “top down” approach, by determining a commercial rate and then applying a differential to arrive at the residential rate.
[99] In this context, I do not consider it seriously arguable that the arbitrator erred, or otherwise breached his mandate, by not expressly discussing Mr Gamby’s
suggested commercial rate of 5 per cent. As noted, this did not form a key part of Mr Gamby’s evidence. It did not in fact form any basis for Mr Gamby’s assessment of fair market rent for the review period. As discussed earlier, an arbitrator is not required to address every aspect of the evidence put before them. Mr Gamby’s reference to a commercial rate of “no more than 5 per cent” was a somewhat tentative and very high-level view.
[100] Second, the Trust’s argument that the arbitrator needed to explain much more clearly why he adopted a differential of 0.75 per cent was against the backdrop that that figure fell well outside the available range of 1 to 1.5 per cent. For the reasons discussed earlier, I do not consider it seriously arguable that the figure of 0.75 per cent did fall outside the available range. In this context, the Trust’s criticism of the extent of the arbitrator’s reasoning on this point loses some of its force.
[101] Third, it is also clear from the Award overall that the very significant increases in land values in more recent years (but which have not carried through to other core economic variables such as rents, inflation and wages) was a significant factor in the arbitrator’s reason for placing less weight on the historical factors adopted by Mr Gamby and Mr Dean. The arbitrator’s acceptance of aspects of Mr Winstanley’s evidence also contributed to his view that the differential rate should be somewhat lower than the figures advanced by the other valuers,65 including Mr Smithies. The arbitrator also took into account that hybrid rates have been fixed in some cases.
[102] Fourth, the arbitrator adequately explained why he did not consider Mr Dean’s reference to the unpopularity of residential ground leases to be a relevant factor in driving a residential rate, namely because of the hypothetical exercise being conducted. The arbitrator also explained that he did not accept Mr Dean’s argument that the ground percentage rate should be closely linked to the rate of return on occupied developed residential property, because:66
The approach in this valuation exercise is not driven, most distinctly, by the rate of return on developed residential properties, because the whole hypothesis turns on the unique factor of this being ground lease only, with the actual developments on the site being put to one side.
65 At [169].
66 At [161].
[103] Finally, the arbitrator overlaid his conclusions on the various rates with the ultimate question of what it would be fair to expect a prudent lessee to pay, and a willing but not anxious lessor to accept, during the review period. He reasoned that while not all the burden of the considerable increase in land value should fall on the prudent lessee, under the traditional approach, a part of it inevitably must.67
[104]Standing back and bearing in mind the principles discussed at [72] to [82] and
[96] above, I consider the nature and extent of reasoning in this case to fall much closer to that assessed in Casata than that in Ngāti Hurungaterangi.
[105] I therefore view the Trust’s argument on the adequacy of the arbitrator’s reasoning to be relatively weak. And the observations above (at [70]) on the other factors to be taken into account when determining whether to grant leave apply equally to the second proposed ground of appeal.
[106] I accordingly also decline to grant leave to appeal in relation to the adequacy of the Award’s reasons.
Adequacy of reasons – further observations
[107] For completeness, and while not determinative in this case (given I have found it is not seriously arguable that the arbitrator failed to give adequate reasons), there may be a question for determination in a future case as to the manner in which a claim of inadequate reasons comes before the Court.
[108] In this case, the Trust says that a failure to give reasons gives rise to an error of law. This is the manner in which such claims have also been advanced in cases such as Casata and Ngāti Hurungaterangi.68 But it is not clear to me whether an alleged failure by an arbitrator to give reasons for their award gives rise to an appeal
67 Award, at [171].
68 In Ngāti Hurungaterangi v Ngāti Wahiao (above at n 47) there was a separate claim of breach of natural justice, but that concerned matters other than the alleged inadequacy of reasons. The issue of adequacy of reasons also appears to have been addressed as an error of law in leading Australian decisions; see for example, Westport Insurance Corporation v Gordian Runoff Ltd [2011] HCA 37, 224 CLR 239 at 262. Though this turned on the particular wording of the statutory provisions in question, and it was noted (at 262) that a different approach may be required under the Australian Federal Act and Model Law – the First Schedule to the Act in this jurisdiction being the Model Law.
on a question of law, or more properly gives rise to an application to set aside the award pursuant to art 34 of sch 1 of the Act. One of the grounds to set aside an award pursuant to art 34 is that:69
…the composition of the arbitral tribunal or the arbitral procedure was not in accordance with the agreement of the parties, unless such agreement was in conflict with a provision of this Act from which the parties cannot derogate, or, failing such agreement, was not in accordance with this Act…
[109] The Law Commission, in its 1991 report and recommendations which led to the Act, stated that the failure by an arbitral tribunal to give reasons:70
…would mean non-compliance with the agreed arbitral procedure, and scope for an application to set the award aside, which would include giving the tribunal an opportunity to remedy its default: see Article 34(4).
[110] This view appears to be shared by the authors of Williams & Kawharu on Arbitration, who refer to this extract of the Law Commission’s report without comment or dissent.71
[111] It strikes me that a failure to give reasons is more akin to a failure by an arbitrator to comply with their statutory or agreed mandate, rather than erring in their statement or application of legal principles. Certainly, in Ngāti Hurungaterangi itself, the Court characterised the failure to give adequate reasons as “the panel’s failure to discharge its agreed mandate by giving a reasoned award”.72
[112] That the procedural route of an application to set aside the award may be the correct route to challenge the adequacy of an award’s reasons is supported by the overall scheme of the Act. Unless sch 2 to the Act applies to the arbitration (see [113] below), the only recourse to a court against an arbitral award is through an application to set aside the award pursuant to art 34.73 The requirement that an award state the reasons on which it is given is found in sch 1 to the Act. Accordingly, unless sch 2
69 Article 34(2)(a)(iv).
70 Law Commission Arbitration (NZLC R20, 1991) at [388]. I note that the Court of Appeal in Ngāti Hurungaterangi referred to this particular paragraph of the Law Commission’s Report at [60], albeit with reference only to the Commission reporting strong support for including a mandatory requirement in the Act to provide reasons, and not to the Commission’s view that a failure to do so would give rise to an application to set aside pursuant to Article 34.
71 Williams & Kawharu on Arbitration, above n 5, at [14.10].
72 Ngāti Hurungaterangi v Ngāti Wahiao, above n 47, at [109].
73 Article 34(1).
applies to the arbitration in question, the only way to challenge the award for an alleged failure to give reasons would be through an application to set aside.
[113] Schedule 2 of the Act contains the additional ground upon which an arbitral award can be challenged, namely an appeal on a question of law (which requires leave in some cases). Under the scheme of the Act, sch 1 (which is based heavily on the Model Law) applies to all arbitrations which take place in New Zealand (including international arbitrations which are conducted in New Zealand). Pursuant to s 6 of the Act, sch 2 applies to domestic arbitrations taking place in New Zealand, but only to an international arbitration if the parties so agree. Thus, if an international arbitration took place in New Zealand but the parties did not agree that sch 2 applied to it, the arbitrator would be subject to the obligation to give reasons for his or her award, but there would be no mechanism to appeal from that award on a question of law. The only recourse available in those circumstances would be an application to set aside pursuant to art 34. Leave would not be required.
[114] It seems somewhat odd that the same legal requirement to give reasons pursuant to art 31 of sch 1 gives rise to an application to set aside the award in some arbitrations, but an application to appeal on a question of law in other arbitrations. Moreover, it would seem odd that a party to an international arbitration taking place in New Zealand would not need leave to challenge the adequacy of an award’s reasons, but in many cases, a party to a domestic arbitration seeking to mount the very same challenge would.
[115] A further point follows. Again, both the Law Commission and the authors of Williams & Kawharu on Arbitration observe that an application to set aside the award “would include giving the tribunal an opportunity to remedy its default: see article 34(4).” A similar view was expressed by the majority of the High Court of Australia in Westport Insurance Corporation v Gordian Runoff Ltd, who stated that “[t]he remedy in respect of the inadequacy of reasons which then would be appropriate would be an order for remitter to the arbitrators for reconsideration pursuant to s 38(3)(b).”74
74 Westport Insurance Corporation v Gordian Runoff Ltd, above n 68, at 272.
[116] While the issue of adequacy of reasons was addressed through the lens of an error of law in Casata and Ngāti Hurungaterangi, cl 5 of sch 2 also includes provision for remission to the tribunal.75 It is not clear from the latter decision (in which the reasons were found to be inadequate) whether remission was considered, but it would have avoided the very unfortunate scenario of a newly constituted tribunal having to start from scratch in determining the parties’ dispute.76
[117] Ultimately, assessing a failure to give reasons in an award through the lens of an application to set aside, with a corresponding opportunity for the tribunal to remedy, would appear to be consistent with the overall scheme of the Act, as well as the Court of Appeal’s observations as long ago as 1988 in Money v Ven-Lu-Ree Ltd that:77
… reasons given by an arbitrator or umpire should be read fairly and as a whole. Awards should not be vitiated by fine points; the modern approach is in favour of sustaining awards where that can be fairly done, rather than destroying them.
Result and costs
[118]The application for leave to appeal is dismissed.
[119] The parties are encouraged to agree the costs on the application. Should they be unable to do so:
(a)Dilworth may file a costs memorandum within 15 working days of the date of this judgment; and
(b)the Trust may file a memorandum in response with a further 5 working days.
[120]No memorandum is to be longer than 3 pages in length.
75 Clause 5(4)(b).
76 And having been given an opportunity to provide adequate reasons, the parties would then be in a position to consider whether that reasoning gave rise to an alleged error of law.
77 Money v Ven-Lu-Ree Ltd [1988] 2 NZLR 414 (CA) at 417.
[121]I will thereafter determine costs on the papers.
Fitzgerald J
3
6
0