BEST ORIGIN LTD v. COMMISSIONER OF RATING AND VALUATION

LDGA 14/1998

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

GOVERNMENT RENT APPEAL NO. 14 OF 1998

______________________

______________________

______________________

J U D G M E N T

______________________

The Hon. Mr Justice LAM, President:

1.  The Government Rent (Assessment and Collection) Ordinance [“the Rent Ordinance”] Cap. 515 provides for the assessment and collection of rents on certain Government leases extended beyond 30 June 1997 by virtue of grants and renewals made pursuant to Annex III of the Joint Declaration.  Such leases included what one can broadly be described as leases for development sites, viz. vacant sites acquired for development purposes.  Under the Rent Ordinance, the annual rent is fixed at 3% of the rateable value of the land leased.

2.  The adoption of rateable value as a reference for the calculation of the Government rent gives rise to a special problem in the context of development sites.  Under rating law, vacant land is not rateable.  That principle continues to apply when building work is in progress.  On the other hand, Regulation 2 of the Government Rent (Assessment and Collection) Regulation [“the Rent Regulation”] specifically sets out how the rateable value shall be ascertained in such circumstances,

“

Where any leased land has not been developed after the commencement of the term of the applicable lease under which it is leased, the rateable value of the leased land at any time before any part of it is developed shall be ascertained as if the leased land were a tenement liable for assessment to rates under the Rating Ordinance.”

3.  In Commissioner of Rating & Valuation v Agrila (2001) 4 HKCFAR 83, the Court of Final Appeal held that Regulation 2 is valid and the above mentioned principle of rating law concerns rateability of a property, not the rateable value of such property.  Because of Regulation 2, a development site is, for the purposes of the Rent Ordinance, deemed to be rateable.  Hence, Sir Anthony Mason NPJ said at p. 103I,

“	… even where land subject to a lease is not assessable to rates under the Rating Ordinance because there is no rateable occupation, it would still be assessable for the purpose of Government rent and that it was the Legislature’s intention so to provide.”

4.  As regards how the rateable value is to be ascertained, the Court of Final Appeal had to consider this issue in the context of Point 4 of the preliminary issues,

“	Whether, in making a valuation under s.8 of the Rent Ordinance of leased land (being a development site) before or during development, the Commissioner is required or authorised to take into account the likelihood at the relevant date of development being carried out?”

5.  Sir Anthony Mason NPJ resolved this issue with reference to sections 7 and 7A(2) of the Rating Ordinance and the rating principle of rebus sic stantibus.  In view of the submissions of counsel, it is important to pay particular attention to the judgment of His Lordship at p. 107 to 109.  Instead of quoting that in full, we respectfully endeavour to summarize the key propositions in that part of the judgment as follows,

(a)	It is permissible to have regard to probable future developments, viz. a present probability of a future happening, as it is something which influence the mind of a hypothetical tenant (p. 107 B to D);
(b)	The valuer must consider every intrinsic quality and every intrinsic circumstance which tends to push the rental value either up or down (p. 107E to G);
(c)	Though the rating hypothesis prescribes the hypothetical tenancy as a yearly tenancy as opposed to a fixed term tenancy, the valuer should also take into account the possibility of a longer (or a shorter) duration of the yearly tenancy.  It is open to the Tribunal to adopt a finite figure for the duration of a tenancy if the facts of the case warrant the same (p. 107G to J);
(d)	The prospect of profits in later years after construction period had been completed can be taken into account in the light of the prospect of the continuation of the tenancy for a number of years (p. 108A to E).

6.  Counsel disagreed with each other as regards the correct interpretation of the judgment at p. 108E to H and it will be useful to quote the same here,

“

It follows that, in ascertaining the rateable value of the sites, it is permissible to have regard to their character as development sites for that is an intrinsic characteristic of each property.  Having regard to that characteristic entails taking account of the likelihood of development taking place and proceeding to completion.  But this does not mean that the sites should be valued as completed developments.  Nor does it mean that either of the Commissioner’s methods of valuation or what has been described as ‘the contractor’s method’ of valuation should be adopted.  The appropriate mode of valuation, is a matter for the Lands Tribunal to determine.  It is not for this Court to express an opinion about valuation or about the appropriateness of any method of valuation.”

“

In these cases much will depend on the estimated duration of a yearly tenancy which the hypothetical tenant might secure.  It might be sufficiently long to allow for completion of the relevant development, so that the hypothetical tenancy would extend eventually to such a situation.”

7.  The formal answer of the Court of Final Appeal to the issue can be found at p. 114B to C,

“	The state of each site should be taken as it was on the relevant date, having regard to all the intrinsic characteristics of each site.  When determining those characteristics, the Lands Tribunal should take into account evidence as to the likelihood at the relevant date of development being carried out on each site by the hypothetical tenant amongst other relevant considerations.”
“	The evidence referred to in (b) above should be taken into account in determining the mode or category of occupation for each site at the relevant date and in particular whether the site was being occupied as a development site.”

8.  The present case is the first case that the Tribunal has to carry out a valuation to arrive at the Government rent for a development site following the guidance from the Court of Final Appeal.

9.  Inland Lot No.8874, the subject of this appeal is a development site at Electric Road, Hong Kong.  The site was sold in a land auction to the Appellant on 11 December 1996 at a premium of $760 million.  On the site, there used to be buildings which were the former Causeway Bay Police Station and Accommodation Block.  They were vacated before the auction.  The successful bidder had to demolish the buildings within 9 months of entering into the Conditions of Sale.  Further, the successful bidder covenanted to develop the site by erecting building or buildings to be completed on or before 31 December 2000 with a minimum gross floor area of 11,000 square metres.

10.  As a matter of fact, a 35-storey building with 2 floors of retail spaces and 5 floors of car-parking facilities, one refuge floor, one floor for plant rooms and 26 floors of office units was erected at the site and occupation permit was issued on 6 December 2000.  It is now known as 148 Electric Road.

11.  This appeal is in respect of the Government rent for the site for the year of 1997/98.  Pursuant to Section 11(5) of the Rent Ordinance and Gazette Notice No.3055 of 20 June 1997, the Government Rent Roll came into force on 28 June 1997.  The date 1 July 1996 was designated as the relevant date (see Gazette Notice 2753 of 1997).  Further, the Respondent, the Commissioner of Rating and Valuation (“the Commissioner”) is required by Section 7A(2) of the Rating Ordinance to take the state of the tenement, the relevant factors affecting the mode or character of occupation and the locality in which the tenement is situated as they were on the date the list came into force, viz. 28 June 1997.

12.  The rateable value of the site originally entered in the Government Rent Roll for 1997/98 was $37,050,000.  Pursuant to Sections 16 and 17 of the Rent Ordinance, the Appellant made a proposal to alter the entry by amending the rateable value to “nil”.  On 28 February 1998, the Commissioner decided to alter the rateable value to $29,640,000.

13.  On 18 March 1998, the Appellant filed the present appeal.  At the early stage of the appeal, the focus was on the preliminary issues.  Those issues were settled by the decision of the Court of Final Appeal in Commissioner of Rating & Valuation v Agrila (2001) 4 HKCFAR 83.  After that, the Appellant filed a report of Mr George Doran of 21 May 2002.  The valuation of Mr Doran followed the legal advice given by counsel.  In that report, Mr Doran reached a conclusion that the only possible hypothetical tenant is a building contractor and such a tenant would not pay anything to occupy the tenement.  Hence, the rateable value for Government rent is a nominal figure of $1.  By way of alternative, Mr Doran carried out a valuation by having regard to evidence derived from tenements being used for purposes which fall into a different mode or category of use.  He referred to comparables for open storage and work sites and arrived at a value of $764,000 per annum.

14.  Mr Doran’s report remains as the primary position taken by the Appellant in this appeal although by the time of hearing, Mr Lynch replaced Mr Doran as the expert giving viva voce evidence due to the retirement of Mr Doran.

15.  To be fair to Mr Doran and Mr Lynch, their instructions were to do a valuation in accordance with the legal advice rendered by counsel.  The legal advice was contained in a document produced as “GD-7” in the appendix to Mr Doran’s report.  That document was an attempt by counsel for the Appellant setting out what counsel regarded as the correct legal basis for the valuation exercise in the light of the Agrila judgment.

16.  The correctness of the legal advice is very much in issue.  Mr Holgate QC submitted on behalf of the Commissioner that the primary approach of the Appellant artificially excludes the site’s development potential from the valuation and unduly restricts the valuation exercise by proceeding on the premise that building works at the site were undertaken solely for the sake of building.  As such, counsel contended that the Appellant’s primary approach conflicts with the object of the Rent Ordinance and the Agrila judgment.

17.  As at 28 June 1997, demolition works regarding the disused police station were in progress.  The process was completed on 9 July 1997.  In respect of the mode or character of occupation of the site as at 28 June 1997, paragraph 40 of the Statement of Agreed Facts set out the agreement between the parties,

“

As at 28 June 1997, demolition of the existing Police Station and Accommodation Block had commenced and comprised an early phase of the development of the Site by the construction of a new building.  Therefore, the mode and character of occupation was that of a development site, which the Government lessee had the right to build a commercial building with a maximum gross floor area of 18,339 square metres.”

18.  The first issue we should resolve is what are the implications flowing from the agreed fact that the mode and character of occupation was that of a development site.  We believe that would involve an examination of the validity of the legal advice of the counsel for the Appellant.

The implications of the site being a development site

19.  At the time when the Court of Final Appeal decided on the preliminary issues, the Tribunal had yet to find the relevant facts pertaining to the character or mode of occupation.  We have now crossed that hurdle in view of the agreed facts between the parties.  The question is how to arrive at a rateable value for the purpose of Government rent in respect of a development site.

20.  In this connection, the Court of Final Appeal clearly stated that the development potential should be taken into account.  At p. 108C, Sir Anthony Mason NPJ held,

“	The prospect of continuation of the tenancy for a number of years enabled the Tribunal to have regard to the prospect of profits in later years and set them off against losses in earlier years.”

21.  The “later years” in that sentence must refer to the time after construction has been completed.  It is only when the development had been completed that there could be the profits generated from the sale or letting of the completed building.  “Later years” were in contrast with “the construction period” in the first sentence in that paragraph at p. 108A during which no profit could be generated.

22.  It is also plain that Sir Anthony Mason NPJ was referring to someone in the position of a developer as opposed to a contractor as the hypothetical tenant.  A developer could derive profit from the later years and suffer losses in the earlier years.  A contractor would not.

23.  The test for ascertaining the extent to which development potential should be taken into account in the valuation is set out by the Court of Final Appeal at p. 108D,

“	If there is a sufficient likelihood of a change of use of the property as would affect the mind of a hypothetical tenant and alter the rent he would pay for it in its existing state, that is a matter to be taken into account in the valuation.”

24.  A change of use could change the character or mode of occupation.  It means that there could be an end to the existing hypothetical tenancy and the commencement of another one based on the new character or mode of occupation.  But such a change should not prevent a present prospect of such change from being taken into account in assessing the current rateable value if such a prospect would affect the rent the hypothetical tenant would pay under the current hypothetical tenancy.

25.  In the context of development site, a change of use occurs upon the completion of the development.  When the units are sold or let and used for commercial or retail purposes, there would be new hypothetical tenancies based on the new character or mode of occupation.  However, the Court of Final Appeal held that if the valuer is satisfied that the present prospect of such future uses would affect the rent the hypothetical tenant would pay under the current hypothetical tenancy, this must be taken into account.  Were it otherwise, there could never be any set off of losses in earlier years (during the development stage) against the prospect of profits in later years (after the completion of development). 

26.  In our view, it is almost self-evident that once it is concluded that the site is a development site, there must be a sufficient likelihood of change of use so as to affect the rent that a hypothetical tenant would pay.  Ex hypothesi, a person who takes up a tenancy for a development site must have done so with an intention to develop it.  Given the intention to develop, the change of use must be contemplated.  A developer would not intend to keep the site in its existing state forever.  In the context of the facts of this case, it is difficult to imagine any person other than someone with intention to utilize the development potential taking up a tenancy for the site as a development site in the state as it was on 28 June 1997.  That would include a person who instead of developing the site himself, intends to sell the site to another person who is a developer.  We will deal with the issue regarding the ability to sell in the context of the hypothetical tenancy below. 

27.  Further, given the intention to develop, the change of use and the profit that could be generated from such change must be the most relevant matters that a developer would take into account in determining the rent that he would be willing to pay.  In this respect, we have evidence from the Appellant’s own witness to support this proposition.  Mr Lee testified how the “Chinese developers” used the “back of the envelope” method to arrive at the prices they would bid at an auction for development sites.  Granted that the auction bids were for sale of the site as opposed to a yearly tenancy of indefinite period for the same.  However, subject to the legal viability of having a tenancy for development purposes under the rating hypothesis of yearly tenancy, we do not see any reason why Mr Lee’s evidence regarding how developers operated in the real world could not be taken into account as how a hypothetical tenant would behave in deciding the rent for a development site.

28.  Regarding the difficulty posed by the rating hypothesis of yearly tenancy of indefinite duration, it is important to understand the reason why there is such a rating hypothesis.  The role of the rating hypothesis is to set a universal standard to assess the rates payable in respect of every tenement.  Lord Pearce explained this in Dawkins (VO) v Ash Brothers & Heaton Ltd [1969] 2 AC 366 at 381H to 382A,

“

Rating seeks a standard by which every hereditament in this country can be measured in relation to every other hereditament.  It is not seeking to establish the true value of any particular hereditament, but rather its value in comparison with the respective values of the rest.  Out of various possible standards of comparison it has chosen the annual letting value.  This is appropriate since the tax is charged annually. … This standard must be universal even though in many cases it demands various hypotheses.”

29.  Obviously, there are tenements that no tenant would take up a tenancy from year to year in the real world.  Lord Pearce gave the examples of sewage works, portion of railway lines.  Development sites fall within the same category.  In the real world, no developer with his right mind will take up a site for development based on a yearly tenancy.  In this sort of situation, the rating hypothesis of yearly tenancy cannot be permitted to operate in such a way to exclude a tenancy for that particular mode or character of occupation.  Otherwise, it would be putting the cart before the horse and the rating hypothesis would frustrate instead of facilitating a proper valuation for rating purposes.  Lord Pearce succinctly pinpointed the limit of the rating hypothesis at p. 382B to C,

“

So one must assume a hypothetical letting (which in many cases would never in fact occur) in order to do the best one can to form some estimate of what value should be attributed to a hereditament on the universal standard, namely a letting ‘from year to year.’  But one only excludes the human realities to a limited and necessary extent, since it is only the human realities that give any value at all to hereditaments.  They are excluded in so far as they are accidental to the letting of a hereditament.  They are acknowledged in so far as they are essential to the hereditament itself.”

30.  At p. 383D to F, Lord Pearce highlighted the distinction between a hypothetical tenant having a yearly tenancy for an indefinite time and a tenant who expected to be turned out at the end of a year.  See also the judgment of Lord Wilberforce at p. 387G to 388B and that of Lord Pearson at p. 393A to E.  The statutory rating hypothesis does not bar a hypothetical tenant from anticipating that his occupation of the subject site could be for a duration much longer than one year.  This is of critical importance in the present context.

31.  In Hoare v National Trust [1998] RA 391 at p. 415, Peter Gibson LJ emphasized the limitation of rating hypothesis and endorsed the following comments of Lawton LJ in Trocette Property Co Ltd v Greater London Council [1974] RVR 306 at 311 as equally applicable in the rating context,

“	It is important that this statutory world of make-believe should be kept as near as possible to reality.  No assumption of any kind should be made unless provided for by statute or decided cases.”

Then at p. 415, Peter Gibson LJ said,

“

In particular I would emphasise the necessity to adhere to reality subject only to giving full effect to the statutory hypothesis, so that the hypothetical lessor and lessee act as a prudent lessor and lessee.  I would call this the principle of reality, which is, to my mind, of fundamental importance in this case.”

To the same effect is the dicta of Schiemann LJ at p. 408,

“	The statutory hypothesis is only a mechanism for enabling one to arrive at a value for a particular hereditament for rating purposes.  It does not entitle the valuer to depart from the real world further than the hypothesis compels.”

32.  In the end, it boils down to how the principle of equality operates in the context of rebus sic stantibus.  In this regard, the judgment of Lord Pearson in Dawkins (VO) v Ash Brothers & Heaton Ltd [1969] 2 AC 366 at p. 393F to G is enlightening,

“

In the expression rebus sic stantibus which are the res? In other words which are the factors to be taken into account in order to produce equality of rating? There is, in this case, a present probability of a future happening, and the present probability affects the present value of the hereditament.  There is inequality of actual values if of two otherwise identical hereditaments one is likely to have part of it demolished within about a year and the other is likely to remain intact.  If they had to be deemed to be of the same value, although in fact one is worth less than the other, there would be artificiality and fiction and unfairness in the valuations.”

33.  What is essential to a letting of a development site? In our view, it must be capable of being developed in a commercially viable manner.  In particular, the duration for which the tenant would have the occupation of the land must be of such a length to make development worthwhile although it may not be occupation under the same hypothetical tenancy.  To achieve that, not only must the total duration of occupation be long enough to permit the development to be completed, it must also be long enough to permit the developer to recoup his costs of development plus a reasonable profit through sale or letting of the developed units.  There cannot be any sale or letting if the duration of occupation after completion is short.

34.  In other words, for every development site, there must be a present probability of the site being held for a duration that is long enough to make development worthwhile.  If there is no such probability, development would not take place and the mode or character of occupation of the site would not be a development site.

35.  How does that fit into the hypothetical yearly tenancy? The yearly tenancy would be a yearly tenancy of a development site.  It is well settled that for rating purposes a valuer is entitled to take into account of the longer duration of the occupation (see Humber v Jones [1960] 53 R & IT 293; (1959) 5 RRC 23), though the rateable value could be varied from year to year.  Thus, Scott LJ said in Robinson Brothers (Brewers) Ltd v Houghton & Chester-le-Street Assessment Committee [1937] 2 KB 445 at p. 476,

“	The idea of a lease for years is excluded, but the prospect of the tenancy from year to year continuing must be assumed.”

36.  As explained by Lord Pearson, this would satisfy both the principle of equality as well as rebus sic stantibus.

37.  Further, it is necessary to distinguish between two concepts: (i) the duration of occupation of the site as a development site under the hypothetical tenancy; and (ii) the anticipated total duration of occupation by the hypothetical tenant including its occupation as development site and as completed development during the post-development period.  In their written closing submissions, the Appellant made the point that as a matter of law, the duration of the hypothetical tenancy cannot be longer than the completion of the development.  That is because after the completion of the development, the occupation would be of another mode or character.  However, it does not follow from this proposition that in deciding the rent payable under the hypothetical tenancy, a valuer cannot take into account of the prospect of continued occupation of the site after completion of the development.  For reasons given above, we are of the view that it is an intrinsic quality of a development site that the tenant contemplates continued occupation after the completion of the building in order to generate profit from sale or letting.  This contemplation is a present probability at the time when the hypothetical tenant takes up a tenancy for the site as a development site.  Hence, even though the duration of the hypothetical tenancy would not be longer than the development period, the rent that could be achieved must take into account of the prospect of continued occupation after completion of development.

38.  In other words, if no account is taken of the present prospect of continued occupation after the completion of the building, the valuer will not be assessing the rent for a tenement with development site as its character or mode of occupation. 

39.  Mr Roots referred to the judgment of Sir Anthony Mason NPJ in Agrila at p. 108G-H, in particular the following observation regarding the duration of the tenancy,

“	It might be sufficiently long to allow for completion of the relevant development, so that the hypothetical tenancy would extend eventually to such a situation.”

40.  As explained above, we have no difficulty with this observation.  But we do not regard such observation as preventing the valuer from taking into account of the development potential of the site by reference to a present prospect of remaining in occupation after the completion of the development.

41.  Sir Anthony Mason NPJ also stressed that the site should not be valued as completed development.  This we must bear in mind.  But taking into account of the development potential of the site in assessing the annual rent for a development site is a far cry from assessing the annual rent for the completed development.  In the former case, the site has yet to be developed and the hypothetical landlord is not to be treated as if he were the owner of the building to be erected thereon.  The costs of construction are to be credited to the hypothetical tenant and the annual rent has to be reduced accordingly.  In the latter case, the mode or character of occupation is that of the actual units as built and used, viz. commercial or retail units as the case may be in the present context.  The hypothetical landlord is to be treated as the owner of the units and no credit will be given to the hypothetical tenant for the construction costs.  

42.  Counsel for the Appellant urged us to bear in mind that the hypothetical tenant does not have security of tenure.  The same argument had been raised in Humber v Jones (1960) 53 R&IT 293.  On the evidence before the Tribunal, it was decided that there should not be deduction from the actual rent on account of lack of security.  The decision of the Tribunal is reported at (1959) 5 RRC 23.  At p. 33, the Tribunal dealt with the issue of lack of security.  The approach was summarized at the bottom of that page,

“	I think that the question whether, in attempting to relate a rent for a term of years to a rent on the statutory tenancy, there should be a deduction from the actual rent to allow for the factors of security must in the end depend upon the facts of each case …”

The Tribunal found as a fact in that case the evidence of actual rent approximates in its terms to those of the statutory tenancy because of the evidence of the surveyor of the ratepayer on the value of the option attaching to the lease with only 5 years to run.  That led the Tribunal to come to view that the hypothetical tenant could expect to have the same sort of period of occupation as that entered into by the actual tenant (see p. 36).

43.  The decision was upheld on appeal at (1960) 53 R&IT 293.  Willmer LJ considered that the lack of security must not be allowed to push too far, at p. 296,

“

Bearing in mind that the statutory hypothesis is a tenancy from year to year, everybody agrees that, in the world as it is, no sane manufacturer would take a tenancy from year to year of a factory in which he is going to install valuable machinery and for which he is going to assemble a skilled labour force to work for him.  It seems to me that, if one had to value such a hereditament in the conditions of the world as they are, the only result would be that there would be a nil valuation, because there would be no possibility in practice of finding a tenant for such premises at all.  That seems to me to be the reduction ad absurdum of the argument presented on behalf of the ratepayers.”

44.  His Lordship further said at p. 297,

“	I accept that one has to postulate that the hypothetical tenant and the hypothetical landlord will be reasonable people, and will behave as reasonable men in the real world would behave in the circumstances which have to be postulated.”

45.  The reasonable men test is also set out by Lord Denning in R v Paddington Valuation Officer [1966] 1 QB 380 at p. 412E to F.

“	The rent prescribed by the statute is a hypothetical rent, as hypothetical as the tenant.  It is the rent which an imaginary tenant might be reasonably expected to pay to an imaginary landlord for a tenancy of this dwelling in this locality, on the hypothesis that both are reasonable people …”

46.  Mr Holgate submitted that it follows that (a) neither hypothetical party can expect to impose a requirement in the imaginary negotiations which the other acting reasonably, would find unacceptable; and (b) having arrived at a hypothetical letting based upon reasonable negotiating positions, a valuer must not assume that either party would act unreasonably thereafter.  We agree.  In the context of a development site, unless the statutes expressly provide otherwise, one must assume that the hypothetical landlord would not frustrate a development by exercising his right under the yearly tenancy in a manner that would prevent the hypothetical tenant from having a reasonable duration of occupation in order to enable the latter to reap a reasonable profit for developing the site.  We do not think there is any provision in the legislation that calls for a departure from this position.  

47.  In the end, it is a question of fact as regards what a hypothetical tenant for a development site could expect in terms of the duration of occupation and Sir Anthony Mason NPJ had pinpointed this as the crucial question in this case.  It must also be built into the hypothesis that the hypothetical landlord for the development site would behave as a reasonable landlord in the real world would behave even though he has the power to terminate the yearly tenancy by six months’ notice.  For reasons already given, the expected duration of occupation could not be so short as to frustrate any incentive for development with reference to the realization of the fruits of development.  That would simply negate the existence of any hypothetical tenancy for a development site. 

48.  Adopting this approach, we do not see any reason why the development potential of a development site cannot be taken into account in assessing the Government rent.  Such development potential is one of the intrinsic quality of the tenement that tends to push the rental value upwards (as opposed to a tenement with no development potential) and it ought to be taken into account, see Robinson Brothers (Brewers) Ltd v Houghton & Chester-le-Street Assessment Committee [1937] 2 KB 445 at p. 469.

49.  In so saying, we are not suggesting that the development potential of all unoccupied sites have to be taken into account for rating assessment.  The mode or character of occupation of a particular site is a question of fact.  If the mode or character of occupation is not a development site, development potential is irrelevant.  However, once it is concluded that the mode or character of occupation at the material time is a development site, the development potential is an intrinsic advantage of that occupation which must have an impact on the rental value under the hypothetical tenancy.

50.  The situation is not dissimilar to the taking into account of the expectation of coal mining trade being improved in subsequent years when the factual matrix warrants a finding that the hypothetical tenant could reasonably assume his tenancy would last beyond one year as in Consett Iron Co v Assessment Committee [1931] AC 396.  The House of Lords held such finding is essentially a question of fact.  See also R v Mirfield (1808) 10 East 219.  However, as explained above, in the context of development site, the crucial question is the expected duration of occupation (which must go beyond the period of construction) and its current impact on the rent of the hypothetical tenancy.  So long as the expected duration of occupation has an impact on the current rent, it would be relevant and it does not matter that the expected duration of occupation is longer than the duration of the hypothetical tenancy.

51.  We must therefore reject Mr Roots’ submissions at paragraphs 199 and 200 of his written opening to the effect that the tenement should not be valued by reference to a prospective change of use outside the existing mode or character of occupation and that the valuer should ignore any value that reflects the value of the development when complete and the future receipt of rents or profits form the completed development.

52.  Since the rationale for taking into account of the development potential of the site stems from the present prospect of the development being worthwhile having an impact on the rent paid by the hypothetical tenant, it matters not even if the continued occupation after completion will be under another hypothetical tenancy for rating purposes.  It also matters not whether the hypothetical tenant would generate profit or income by letting or by sale.  As explained by Professor Baum, price is just a function of yield and rent.  Both of them could reflect the development potential.  Either way, there would be a new hypothetical tenancy or tenancies after the completion of the building.  But the effect of such prospect on the hypothetical rent during the development period would be the same.  This present prospect must be economically relevant to the inquiry as to the real value of tenement as a development site (see Robinson Brothers (Brewers) Ltd v Houghton & Chester-le-Street Assessment Committee [1937] 2 KB 445 at p. 470-1).

53.  In this connection, the evidence of Mr Lynch, the Hong Kong valuer called by the Appellant testified under cross-examination that this was what a rating specialist would have done if he were unconstrained by the legal advice rendered by the Appellant’s counsel.  He said he would have to look at the position from the perspective of reality and ask what in the real world would be long enough to enable a person occupying a development site to recoup their moneys in developing it.  This can be achieved either by renting the units or by sale.  In either case, the expected duration of occupation must be long and 40 years would not be unreasonable.

54.  It is difficult to see how the development potential of a development site can properly be taken into account if the rating hypothesis is to be construed in such a way to confine the expected duration of occupation to the period of construction or a short time afterwards for marketing of the units.  If the valuation of the rateable value under the hypothetical tenancy of a development site were to be constrained by the prospect of the termination of occupancy upon the completion of construction or at the end of a short marketing period thereafter, the value achieved would not be the real value of the occupancy of a development site.  The intrinsic quality in terms of the development potential of the site (which is the actual purpose of occupation in the real world) will not be taken into account.  It would become the value of a bare construction site without any development potential.  Such an approach is inconsistent with the judgment of the Court of Final Appeal in Agrila and it is also against the principles set out in the other authorities mentioned above.  Yet this lies in the heart of the stance taken by the Appellant.  It permeates throughout the propositions set out in the legal advice based on which the Appellant’s surveyors were instructed to conduct their valuation.

55.  In the legal advice of March 2002, the surveyors were instructed as follows,

(a)	“	Alterations which the hypothetical tenant might make to the tenement during the hypothetical tenancy are not to be taken into account in the valuation unless they are minor … Any prospective change of use is to be ignored.” (Para. 8);
(b)	“	It is not permissible to value the tenement as though it was in a state different from that on 28 June 1997, nor to attribute any value to alterations to the physical state which a potential tenant might carry out, nor to value it taking account of profits which might be expected to be received when the land is in a physical state different from that on 28 June 1997.” (Para. 9);
(c)	“	The valuer should ignore any value which might be attributable to any use of the land other than as a development site.  In particular, the value ascertained should not reflect (a) the value of the development when complete and the tenements thereby created are occupied or (b) the future receipt of rents or profits from the completed development.” (Para. 14);
(d)	“	An open market capital value which reflects (i) anticipated changes (other than minor ones) to the state of the tenement and/or (ii) a different mode or character of occupation (for example, following completion of development of the land and occupation of the tenements thereby created), and/or (iii) anticipates the receipt of rents or profits following completion of development would not of itself be evidence of value for the purposes of the Rating Ordinance, ss 7 and 7A.” (Para. 23).

56.  With respect, these propositions fail to take proper account of the nature of a development site and its development potential.  The net effect of these propositions is to strip the site of its intrinsic quality as a development site.  As explained above, the consideration of the development potential in the valuation is premised on a real present prospect of such potential being realized in the future as having an impact on the current rental value of the site as a development site.  It is an inherent quality of a development site.  Hence, it does not infringe Section 7A of the Rating Ordinance to take it into account even the mode or character of occupation has to be the same as it was when the list comes into force, viz. as a development site.

57.  On the contrary, adopting the Appellant’s approach will enjoin the valuer to value the site as a bare construction site with no development potential.  That is not a valuation on the same mode or character of occupation as required by Section 7A.

58.  Not surprisingly, with their hands tied by the legal advice, Mr Doran and Mr Lynch came up with a valuation of $1 for the site, treating the building contractor as the hypothetical tenant as opposed to the developer.  We have no hesitation in rejecting such valuation as it is based on a mis-application of the rating hypothesis.

59.  A related hurdle put forward by the Appellant is the proposition that the hypothetical tenant could not sub-sell or sub-let the tenement or parts thereof to third parties.  This point was raised in Paragraphs 219 of the Appellant’s written opening.  Mr Holgate argued that the point does not arise and it is an artificial construct as the subject matters of sale and letting by the hypothetical tenant are the units in the building (which in the hypothetical rating world must be regarded as the asset of the hypothetical tenant) as opposed to interests in the tenement itself.  Further, counsel referred to some old cases to demonstrate that the fact that a yearly tenant could not grant an interest in perpetuity is not an objection to rate the rector’s ‘occupation’ by reference to the capital receipt he obtained through selling freeholds of graves or easement in perpetuity: North Manchester Overseers v Winstanley [1910] AC 7; R v Abney Park Cemetery Co (1893) LR 8 QB 515.  The same point was illustrated by reference to a grant by one railway company to another to use part of its rail system for 999 years by way of joint user, R v Fletton Overseers (1861) 3 E & E 450.

60.  Mr Roots tried to distinguish these cases by contending that, unlike the developer, these ratepayers remained in occupation of the land.

61.  We agree with Mr Holgate that under the rating hypothesis in the present context, one must segregate the occupation of the building from the occupation of the land.  If one were to distinguish between the occupation of the building and the occupation of the land, the developer’s building would still be occupying the development site.

62.  Ultimately, it goes back to the issue of the likely duration of occupation.  If the likely duration is long and comparable to the duration of the actual grant under the Conditions of Sale, or for any duration which makes it commercially viable to develop the site by building a new building thereon, there is no reason in principle why the development potential could not be assessed by reference to the prospective sale prices or the prospective rents of the units of the building in the real world as at the relevant date of valuation.

63.  Mr Doran and Mr Lynch also put forward an alternative valuation based on comparables derived from open storage and work sites.  We agree with the submissions of Mr Holgate that this alternative valuation is based on a completely different mode or character of occupation and it fails to take into account of the development potential of the appeal site.  We would therefore reject the alternative valuation as well.

64.  At this juncture, it should be recorded that we have submissions from the parties regarding the permissibility of referring to all realistic alternative uses the sites could be put for rating valuation based on Commissioner of Rating and Valuation v Lai Kit Lau Mutual Aid Committee [1986] HKLR 93.  That being a Court of Appeal decision is binding on us.  Our attention was also drawn to the English Court of Appeal’s decision in Williams v Scottish and Newcastle Retail Limited [2001] RA 41.  Insofar as there is any divergence between these two decisions, we are duty bound to follow Lai Kit Lau.  However, in neither case was the court concerned with the application of the rating hypothesis in the rating valuation of a development site.  We have explained why we consider the present prospect of the development potential being realized should be taken into account as part of the existing character and mode of occupation as development site and we do not think it is necessary to refer to these cases at length.  Unlike Lai Kit Lau, we are not dealing with a situation where the alternative uses would result in a higher hypothetical rent for the tenement as compared with its actual use as a development site.  The Lai Kit Lau point only arises if we have to assess the Government rent by reference to its use as a bare construction site without any development potential, an approach rejected by us.

65.  On a general note, we are very much indebted to the assistance offered by eminent specialist counsel in the present case and we have been referred to many authorities on rating.  At the same time, we find the following observations of Robert Walker LJ (as he then was) at Para. 56 of the judgment in Williams v Scottish and Newcastle Retail Limited [2001] RA 41 must be borne in mind in reading earlier cases,

“

However, respect for ‘intellectual freight’ from earlier centuries must not be carried too far.  In particular, it is not helpful to fasten on isolated pronouncements by judges, however eminent, without regard to the context in which they were made and to seek to apply them to an issue which was not before the court.”

66.  Further, as pinpointed by the learned judge in Para. 54 of the same judgment, the proper approach depends ultimately on statutory construction.  In the present context, the object of the exercise is to give effect to the legislative intent as gathered from the combined effect of the rating hypothesis set out in Section 7A(2) of the Rating Ordinance and Rent Regulation 2 in the light of the Agrila decision.

67.  It is common ground that Section 7A(2) incorporates the well- established rebus sic stantibus rule.  In the conventional understanding of the rule as identified by Robert Walker LJ at Para.17 of Williams, it has two limbs: physical state and use.  The Lands Tribunal in Williams formulated the approach (and upheld on appeal) as follows,

“

In carrying out a valuation under the rating hypothesis the following assumptions are to be made about the hereditament: (a) that the hereditament was in the same physical state as on the material day.  Alterations which the hypothetical tenant might make to the hereditament may be taken into account if, taken overall, they are minor.  All other prospective alterations to the hereditament are to be ignored; (b) that the hereditament could only be occupied for a purpose within the same mode or category of purpose as that for which it was being occupied on the material day.  Any prospective change of use outside that mode or category is to be ignored.”

68.  This approach was in turn derived from the rule formulated by the Lands Tribunal in Fir Mill Ltd v Royston UDC (1960) 7 RRC 171,

“

The second assumption … is that the mode or category of occupation by the hypothetical tenant must be conceived as the same mode or category as that of the actual occupier[1].  A dwellinghouse must be assessed as a dwellinghouse; a shop as a shop, but not as any particular type of shop; a factory as a factory, but not as any particular type of factory.  Some alteration to an hereditament may be, and often is, effected on a change of tenancy.  Provided it is not so substantial as to change the mode or category of use, the possibility of making a minor alteration of a non-structural character[2], which the hypothetical tenant may be assumed to have in mind when making his rental bid, is a factor which may properly be taken into account without doing violence to the statute or to the inference we draw from the authorities.”

69.  We recite these judgments at length because the legal advice from Appellant’s counsel to Mr Doran and Mr Lynch has obviously relied heavily on these passages.  It was also relied upon in the Opening Submissions of Mr Roots to contend that it would not be correct to value the tenement by taking into account of profits which might be expected to be received upon development being completed (see Paras.194 to 200).

70.  The Fir Mill approach (with slight modifications or qualifications as explained in the footnotes) was held to be correct in the context of a conventional rating case.  In that context, as highlighted by Lord Wilberforce in Dawkins v Ash Brothers & Heaton [1969] 2 AC 366 at p. 385 H to 386A,

“

The principle was mainly devised to meet, and it does deal with, an obvious type of case where the character or condition of the property either has undergone a change or is about to do so: thusa house in course of construction cannot be rated: nor can a building be rated by reference to changes which might be made in it either as to its structure or its use.”

71.  But His Lordship went on to say,

“

But it would surely be unreasonable to suppose that the hypothetical tenant is so inescapably imprisoned in the present that no anticipation is permitted of what is to come.  Whether the test is what would influence his judgment, or what intrinsic qualities the hereditament possesses, any occupier in real life has to ascertain and to consider whatever may make his tenancy more or less advantageous over the period for which he takes it.”

72.  In the present case, we are dealing with a novel situation, the application of the hypothesis under Section 7A(2) in the context of development site.  Reading Section 7A(2) together with Rent Regulation 2 in the light of Agrila, we are of the view that it would not be right to apply the same approach without further modifications.  To apply the rating hypothesis in the manner as the Appellant did would fail to take into account of the development potential of the site.  In effect, the Appellant’s approach fails to value the tenement as a development site, which is the mode or character of occupation subsisting at the material time.

73.  It has to be noted that there are differences in the wordings in the English statutes and our Section 7A(2).  Section 7A(2)(b) refers to “any relevant factors affecting the mode or character of occupation” whilst para.2(7)(b) of Schedule 6 of the Local Government Finance Act 1988 refers to “the mode or category of occupation of the hereditament”.  Factors affecting the character of occupation can encompass matters beyond the category of occupation.  Development potential must be an important factor affecting the character of occupation of a development site.  Although different developers may have different views on the precise quantification in money terms of the development potential of a site, there are objective factors that developers must take into account in assessing development potential and deciding how the site could be developed in a economically viable manner.  Take the instant case as an example, whilst it may be open to debate whether the site could be suitable for development by erecting a Grade A or Grade B commercial building, there is no dispute that it was suitable for the erection of a commercial building.

74.  For reasons already given, we think that in the context of a development site, Section 7A(2) must be construed as permitting the present prospect of the development potential being realized in the future to be taken into account.  Any other construction would be inconsistent with Agrila and would frustrate the legislative objective as reflected in Rent Regulation 2.

75.  The Commissioner contended that the development potential should be taken into account by a different route.  It is argued that the tenancy is likely to endure for the whole of the economic life of the building and in the event of the tenancy being determined at an earlier state, the hypothetical tenant will receive the market value of the building (excluding the site value).

76.  Rent Regulation 6 provides for deletion upon the development being completed.  The Appellant contends that this means the hypothetical tenancy could not extend beyond the development period.  The Appellant further submitted that upon completion of development, the mode or character of occupation of the tenement would be changed and therefore the occupation would no longer be under the same hypothetical tenancy.  We see the force of this argument and we therefore prefer to rest our decision on our own analysis set out above, viz. that the taking into account of the development potential is based on the expected duration of occupation as opposed to the expected duration of the same hypothetical tenancy in order to apply the statutory hypothesis meaningfully to a development site.

77.  On the second part of the Commissioner’s argument, the assumption of full compensation to the hypothetical tenant for his asset is derived from the following dicta of Lord Hailsham in The Railway Assessment Authority v The Southern Railway Co [1936] AC 266 at p. 286-7,

“

Then it was said that the Court has assumed that the hypothetical tenant might well be left at the end of the year with a huge amount of unrealizable rolling stock and that he would require very heavy compensation to make up for that risk … I think it must be assumed that the hypothetical tenant would be able to realize his stock at the end of the year as easily as he could acquire it at the beginning of the year at the market value prevailing at the respective dates.”

78.  Mr Holgate developed this part of his submissions in paragraphs 5.68 and 5.69 of his written opening.  It is common ground that the building erected on the site should be treated as the property of the hypothetical tenant for present purposes.  The costs of construction was to be paid by the hypothetical tenant and there must be built into the exercise a model whereby the hypothetical tenant does not have to face any risk of the building being confiscated by the hypothetical landlord without compensation.  Further, the hypothetical tenant must be able to realize its value.

79.  Mr Roots did not dispute the proposition that upon termination of the tenancy, the tenant would be entitled to receive the market value of the building at the date of termination.  However, counsel did not accept that the witnesses of the Commissioner has worked out the market value of the building correctly.  Counsel also stressed that the hypothetical tenant has to take the risk of the market value of the asset having been fallen between the beginning of the hypothetical tenancy and the termination thereof (see Paras.33 to 43 of the Appellant’s Supplemental Written Opening).  These submissions addressed the detailed valuation of the development potential attributable to the site itself (as opposed to the building) and they are relevant in the receipt and expenditure valuation.  We will come back to them later.  On a general level, the points made by Mr Roots do not present any objection to the development potential being taken into account in the assessment of Government Rent for a development site.  

80.  In the Appellant’s Written Closing, a further point was made in Para.551 that it is difficult to apply this proposition to building since it is physically fixed to the land and cannot be offered on the open market without the land on which it stands.  Whilst the exercise may not be easy since there is no real life comparable, we do not regard this as an impossible task.  The exercise is artificial since it would not happen in the real world.  But this is an inherent difficulty arising from the application of the rating hypothesis to a development site.

“Likelihood of development being carried out” vis-à-vis “development potential”

81.  In paragraph 541 of the Appellant’s Written Closing, counsel invited us to steer away from the expression “development potential” as it is a loose term and the decision of the Court of Final Appeal was expressed by reference to “likelihood of development being carried out”. 

82.  Whilst the formal answer given by the Court of Final Appeal did use the expression “likelihood of development being carried out” instead of referring to development potential, that was in the specific context of the question asked by way of preliminary issue.  For obvious reasons, the Court of Final Appeal did not wish to pre-empt how this Tribunal would perform its task in assessing the Government rent once it is concluded that the tenement in question is a development site.

83.  However, in the judgment of the Court of Final Appeal, Sir Anthony Mason NPJ extracted specifically this proposition from Consett Iron at p. 108C, “The prospect of continuation of the tenancy for a number of years enabled the Tribunal to have regard to the prospect of profits in later years and set them off against losses in the earlier years.” As discussed above, reading that sentence together with the opening sentence in the same paragraph at p. 108A, the “later years” that His Lordship referred to in the last sentence must be those years after construction has been completed and “the prospect of profits” cannot be anything other than prospect of profits generated from the sale or letting of the units in the completed building. 

84.  This should be crystal clear if one were to read on.  At the next paragraph at p. 108D, Sir Anthony Mason NPJ explained why the profits in the later years should be taken into account despite the change of use in those later years.  We have already quoted the relevant dicta above.  In short, it was based on the present prospect of the future profits having an impact on the rent that the hypothetical tenant would be willing to pay for the tenement in its existing state.  In order to properly evaluate the impact on the quantum of hypothetical rent, we fail to see how a valuer could have ignored the present knowledge of the hypothetical tenant (viz. the knowledge as at the relevant date for valuation) as to the likely quantum of prospective profits that may be generated in those later years.  We see no difficulty in referring to such knowledge as the development potential.

85.  Mr Roots submitted that Consett Iron did not go so far.  That may be so.  But it is neither here nor there.  Consett Iron was not dealing with the novel situation of applying the rating hypothesis to a development site.  In any event, given the clear guidance provided by the judgment of Sir Anthony Mason NPJ, this Tribunal is duty bound to apply it.

86.  We acknowledge that the task involves projecting into the future where nothing can be certain.  However, it is a matter of valuation judgment and given our conclusion on the proper construction of the legislation, it is a task that we must perform.

How to value the development potential?

87.  The judgment of Member Lo will deal with the valuation of the development potential.  I have read the same and I agree with his conclusions.

88.  I shall confine myself to some additional observations on the expert evidence of the Appellant’s experts.  I have already commented on the evidence of Mr Doran and Mr Lynch.

89.  Regarding Mr Charman, as explained by Member Lo, he unduly confined the value of the site to the hypothetical tenant to the rental income he could expect to derive from it for a period of 6 years (and we can forget about the compensation regarding the value of the building since it would be set off by the costs of construction).  And the rent under the hypothetical tenancy was worked out as 1/6.692 of that income obtainable by him during that 6 years.

90.  For reasons already canvassed, there is no justification for confining the income of the hypothetical tenant to 6 years’ income and the compensation to the construction costs.  Perhaps Mr Charman had subconsciously been influenced by the Legal Advice of the Appellant’s counsel that the valuation should not take into account of the prospect after the completion of the development even though he somehow gave another 27 months to the hypothetical tenancy after the completion of construction.  We have already alluded to the fallacy of treating the end of hypothetical tenancy as the limit to the development potential that could be taken into account by a prospective hypothetical tenant.

91.  Mr Charman justified his approach by postulating that the hypothetical landlord would probably terminate the hypothetical tenancy in six years’ time as this course would be most advantageous to the hypothetical landlord.  In our judgment, this postulation is a misapplication of the rating hypothesis.  In effect, Mr Charman put forward a scenario where the hypothetical tenant would have to invest a lot of money in the construction of the building in return for very limited income for a very short duration after the end of the construction period.  This grossly distorts the mode and character of the site and artificially suppresses its development potential.  A landlord who harbours an intention to terminate the occupancy of the hypothetical tenant soon after the completion of the construction of the building does not behave reasonably in respect of the tenancy for development site and as such cannot be the hypothetical landlord for the present purposes.  A landlord with such intention will fail to attract any tenant bidding for his site as a development site.

92.  In a nutshell, we are of the view that Mr Charman got it wrong in his estimate of the likely duration for which the hypothetical tenant would remain in occupation of the site.

93.  This also solves the difficulty regarding the use of the adjusted auction price for the contractor’s basis valuation.  Given that the development potential cannot be properly taken into account without considering the likely duration of the continued occupation of the site under another tenancy after the end of the hypothetical tenancy of a development site, the objection stemming from real option value is misconceived. 

94.  Turning to the evidence of Professor Hughes, Member Lo will elaborate on the flaws in his PDM model.  I have no hesitation in rejecting his evidence.  Whilst there is no rule prescribing that only a valuer can give evidence in a valuation appeal, it strikes me as remarkable that the Appellant has chosen an economist as their lead expert in this appeal.  The Appellant obviously had access to experienced valuers and I do not see any reason why this court should not proceed on the basis that it has been properly advised by experts in the valuation field.

95.  It is said that we are dealing with a novel situation and the service of an economist is required to devise a workable model to cater for this situation.  The result is the PDM model of Professor Hughes.  I must say I do not buy this argument.  As demonstrated by the evidence of Mr Hatchwell and Mr Tang, valuers are quite capable of putting forward workable valuation models by using conventional valuation methods to quantify the development potential of the site for the purposes of arriving at the rent under the hypothetical tenancy.

96.  As pointed out in the judgment of Member Lo, there was simply no need to devise a PDM model which, on careful analysis, seriously and unrealistically slanted in favour of the Appellant in the resultant figures put forward as the rent payable to the hypothetical landlord.

97.  In Cala Homes v Alfred McAlpine Homes East [1995] FSR 818, Laddie J deprecated expert acting as a partisan hired gun at p. 843,

“

Of course the court will be aware that a party is likely to choose as its expert someone whose view is most sympathetic to its position.  Subject to that caveat, the court is likely to assume that the expert witness is more interested in being honest and right than in ensuring that one side or another wins.  An expert should not consider that it is his job to stand shoulder to shoulder through thick and thin with the side which is paying his bill.  “Pragmatic flexibility” as used by Mr Goodall is a euphemism for “misleading selectivity”.  According to this approach the flexibility will give place to something closer to the true and balanced view of the expert only when he is being cross-examined and is faced with the possibility of being “found out”.  The reality, of course, will be somewhat different.  An expert who has committed himself in writing to a report which is selectively misleading may feel obliged to stick to the views he expressed there when he is cross-examined.  Most witnesses would not be prepared to admit at the beginning of cross-examination … that he was approaching the drafting of his report as a partisan hired gun.  The result is that the expert’s report and then his oral evidence will be contaminated by this attempted sleight of mind.  This deprives the evidence of much of its value.”

98.  I have commented upon the importance of the impartiality of an expert witness in L v L HCMC 1 of 2003, 18 Nov 2003, at paras. 154 to 155, and in particular I said, 

“

One important consideration the court will bear in mind in resolving the difference between the experts is the impartiality of the expert.  When it can be demonstrated that an expert had been partial to his client in giving his opinion, the court would inevitably have serious misgivings about attaching great weight to his evidence in areas that depend a lot on his judgment which is not verifiable objectively.  The court will naturally be reluctant to accept opinions from an expert who has permitted his zeal in putting forward a plausible theory that serves his client’s interest to override his duty of impartiality.”

99.  When that case reached the Court of Appeal (L v C [2007] 3 HKLRD 819), Stock JA made the following observations at paras. 110-111,

“

Although there are of course many experts who come before the courts knowing full well the proper duty of an expert, the courts here, as elsewhere, are too often encumbered in the performance of their duty by experts who do not assist.  An expert stakes his reputation on the assistance he gives a court, regardless of the fact that he or she is engaged by one of the parties; and the expert who does not assist in as impartial and professional a manner as he is able does little more than waste judicial time and add to the expense of litigation. … The courts must take a firm hand.  The overriding duty of an expert is to the court: it overrides, in other words, any obligation to the person from whom he has received instructions or by whom he has been paid.  His opinion is to be ‘uninfluenced as to the form or content by the exigencies of litigation.’ Where such a breach is clearly threatened, or becomes manifest, the court may if necessary appoint, at the parties’ expense, an assessor pursuant to the provisions of section 53 of the High Court Ordinance, or disallow the relevant costs of the party that puts forward the defaulting expert.”

100.  I do not wish to recite at length the criticism levelled by Mr Holgate against Professor Hughes.  Speaking for my part, I am driven to the conclusion that Professor Hughes had regrettably embarked on an exercise which is partisan and misleading.  Given his knowledge and his intelligence, it is difficult to believe that he had not noticed the “second bite” and the “third bite” built into his PDM.  It should equally be plain to him that his PDM cannot properly be described as a R & E valuation.  Yet when he was cross-examined on the “second bite”, he tried to dodge the issue by saying he was only doing “what is close to an exact analogy to what Mr Tang has done” in the latter’s R & E calculation (Day 16, p. 1371).  As Mr Holgate submitted, the gap point had already been flagged by Mr Hatchwell in his Third Report served in April 2006 and reiterated in his Fourth Report of August 2006.  Professor Hughes had been forewarned.

101.  In view of the observations by the other side’s expert, a conscientious impartial expert would have addressed the point on his own volition.  If there were flaws in his earlier report, the expert should take the initiative of acknowledging the mistake and correct the same in a supplemental report as soon as practicable.  At least, he should communicate his views to his counterpart.  Instead of doing so, Professor Hughes played on dodging tactics and left the subject to Mr Holgate’s cross-examination.

102.  When it was clear to everybody that the point would not go away, the professor made a feeble attempt to create further smokescreens by claiming that the gap was due to the third bite and the difference in methodology and by producing AAD Tab 25.   When he was questioned by the President at the beginning of Day 17 (p. 1390), the professor categorically maintained that there was no double counting.  It was only later in the day, he admitted that there was double counting, but “it’s not double counting on the overall capital cost; it is double counting, in your terminology, of giving a return on the tenant’s return that has been charged.” The professor further claimed that it was the characteristic of any DCF analysis.  It then took several pages of transcript of cross-examination by Mr Holgate to lead to an admission by the professor at p. 1431 that the 34 % discount rate has been applied twice in his column 12.

103.  The existence of a second bite in Professor Hughes’s PDM should not have taken such a tortuous route to be exposed.  As in other court, this tribunal expects an expert witness to be forthcoming in giving his expert assistance to it and mindful of his duty in testifying in the witness box.  If it dawns on an expert witness that a position he has previously taken in his reports is wrong, it will not do him much credit by adopting diversionary tactics and creating smokescreen with a view to brush the problem aside.  The proper course is for the expert to come clean and admit candidly that a mistake has been made and revise his opinion accordingly.

104.  Regrettably, the way in which Professor Hughes dealt with this gap point demonstrated clearly to this Tribunal that he chose to adopt the dodging strategy.  As Stock JA said, it is necessary for the court to take a firm stand.  Too much costs and judicial time have been wasted on winnowing the chaff from the wheat in dealing with expert evidence.

105.  It has to be borne in mind that it is usually more difficult and time consuming for the court to identify the irrelevant materials in the expert evidence on a technical subject because ex hypothesi, the court requires the assistance of an expert in understanding the subject in the first place.  Thus, the court will generally leave it to an expert to decide what materials should be put forward to the court to support his opinions.  Unfortunately, this also means that an expert who set out with a hidden agenda to confuse rather than to help the court in grappling with a technical subject can easily exploit the latitude afforded to him and generate a lot of irrelevant and highly technical issues.  Not only must time and costs be spent on hearing such evidence, it is most likely that the expert on the other side have to respond to the same and issues are proliferated and hearing is lengthened.  This is particularly so when the first report of the defaulting expert is less than transparent.

106.  I am sad to say that this is precisely what had happened here.  Professor Hughes had not clearly explained his PDM in his first report.  E.g. he had not clearly explained that there was a “second bite” and “third bite” hidden in his PDM.  Issues were proliferated in the several rounds of evidence and new materials were generated throughout the trial.  Such a course of conduct has made it very difficult not only for the Tribunal, but also for counsel.  Mr Roots had been at pain to indicate on several occasions that he had found the matter very difficult to grapple with.

107.  Given that the PDM of Professor Hughes is a novel model, the need to carry out some sensitivity analysis could not have escaped his attention.  Professor Baum had conducted such an exercise after removing the “second bite” and the “third bite”, thereby converting the PDM into a conventional DCF analysis.  Professor Baum’s sensitivity test demonstrates that the PDM is highly sensitive to the exit yield, the discount rate and the duration of the tenancy.  Thus, Mr Holgate submitted that even if no real growth in rents is assumed for the sake of argument, the adoption of a sensible discount rate and a sensible terminal value would produce result that strongly support the Commissioner’s valuation of $26.88 million.  In other words, with the removal of the “second bite” and “the third bite” the crux of the dispute is the variables input into the model.

108.  Another illustration of Professor Hughes’ opportunistic approach is his evidence about the duration of the hypothetical tenancy.  This has to be considered against the background that it must have been known to him that under his model the rateable value would increase from $4.56 million to $8.78 million if the duration were changed from 5 years to 4 years.  Initially, the professor accounted for a 5-year duration by adding 18 months (to achieve a 90% occupancy) to the 42 months’ construction period.  Subsequently, the parties reached agreement on a 32-month construction period.  It follows that the total duration should be reduced by 10 months.  However, Professor Hughes maintained his 5-year duration by referring to “the uncertainties that would be in the minds of the parties about when the property might have reached a stage at which it would be feasible to consider terminating, refinancing, whatever explanation leads to the end of the tenancy”.  In the end, the professor prayed in aid his “reasonable judgment” of the distribution of risks and consequences of delay as well as the benefits of progressing faster.

109.  What the professor has been unable to explain is why he did not consider such uncertainties when the construction period contended for was 42 months.  Mr Holgate hit the bull’s-eye in his submission that this new argument on uncertainties was wheeled out in order to resist a shorter 4 year period of analysis.

110.  When the professor was cross-examined about it, he again tried to deny that he had changed his case.  He maintained at p. 1173 lines 17 to 29 that his argument was not new.  He said,

“

It represents what I had originally said, which was that it would be in the minds to use a period of five years.  Five years is a very standard period to use for many kind of investment appraisals which do not strictly operate on the basis of being rigidly tied to a particular timetable for construction …” (my emphasis)

111.  In so saying, the professor seems to have forgotten about how he worked out the period of five years in his reports.  In paragraph 8.2.1 of his first report of 11 May 2005, Professor Hughes referred to the period as “a period of 5 years from the purchase of the land until it is expected that the building is 90% occupied.” At paragraph 8.2.9, he further elaborated that there was a “time difference of about 18 months between the issuance of an occupation certificate” and the end of the 5 year period.  Reading that together with his appendix GH-06, it is quite clear that he was aggregating a 42-month construction period (see the figures under the construction costs column) and a 18-month post-construction period to come up with a total of 5 year duration.

112.  The same calculation is reflected in appendix GH-16 to his supplemental report of 10 July 2006, see the figures under the “Year after completion” column and compare with the figures under the same column in the revised GH-16 submitted at the trial.  The professor also referred to the 5-year duration at para.10.13 of his supplemental report as the aggregate of the construction period and the initial period of occupation.  There is no trace of adopting a 5-year duration as “a standard period for investment appraisals”.

113.  Even though he admitted there were changes in his assumption on the effect of duration on the valuation, he steadfastly denied that he had changed his approach and reasoning for the 5-year duration, see transcript p. 1174.

114.  I do not have much confidence on an expert who is less than forthcoming in admitting changes in his case and explaining why such changes are made.  A reasonable inference and it is an inference I draw here is that the professor could not offer a good explanation for the changes.

115.  Professor used a yield of 7% to arrive at the terminal value for the purpose of his model.  In the report, he gave an impression that the yield was derived from his detailed statistical analysis.  Thus, at para. 2.1.6 of his First Report, he said,

“	My analysis … based on detailed statistical analysis in Appendix GH-02 --- indicates that a property developer who was properly advised would have anticipated a decline in the ratio of office prices to rents for both Grade A and Grade B office over the period of development of the site.”

116.  Appendix GH-02 is a document called “statistical evidence on office rents, prices and rental yields”.  In that appendix, the professor undertook statistical analysis called the ARIMA models and VAR models on some data published by the Rating and Valuation Department.  Based on such analysis, in paragraph 58 of the Appendix the professor related the forecasts he made with the “back-of-the envelope” approach of Mr Lee and adopted the assumption of 7% yield and said that was in line with his forecast.

117.  It took quite a bit of cross-examination by Mr Holgate to lead the professor to a clarification that he did not actually derive the yield from statistical analysis (see Transcript p. 1461-2).  As pointed out in Mr Holgate’s closing submissions, the unreliability of statistical projections was exposed by Professor Kalok Chan’s Second Report and Appendix GH-02 is no more than a “smoke and mirrors” exercise designed to give some illusory support to Professor Hughes’ views.

118.  In paragraphs 8.2.7 to 8.2.10 of his First Report, Professor Hughes explained the alternatives he had considered in constructing the terminal value for his model.  Regarding the first alternative, he concluded in para. 8.2.8 that it had to be rejected because “it seems highly unlikely that any developer would have been willing to pay even $550 million for the site at the valuation date if it had acted on that expectation.” By his second approach, the professor was able to reach a higher terminal value of $1,140 million and he adopted the same for his model.  However, even on this alternative approach, it represented zero real growth and Mr Holgate justifiably put to the professor that no property developer is going to go ahead with a project on this basis (see transcript p. 1456-1458).  The professor had to resort to his theory about option value to support his approach and he accepted that there was no real-life evidence to show that property developers behaved in such a way.

119.  The professor should be well aware of the lack of realism in his approach because he admitted under cross-examination that he had done a cross-check of his approach in paragraph 8.2.9 against the $550 million auction price and he was not prepared to say that the result was satisfactory (see transcript p. 1456-1457).  Yet he refused to go into details of such cross-check and chose to duck the issue by claiming that he had done so long ago and he did not have a power calculator to enable him to work out the figures again in the witness box.

120.  It seems the professor was content to advance his case on the basis of the slender proposition of option value notwithstanding the difficulty revealed in the cross-check.  I must say it does not strike me as a re-assuring approach.

121.  I categorically reject the evidence of Professor Hughes.

The summons to strike out

122.  There were pre-trial skirmishes on the question whether it was open to the Appellant to advance a case along the line set out in the report of Mr Doran in view of the Agrila decision.  On 21 April 2005, the Commissioner issued a summons seeking to have certain parts of Mr Doran’s report struck out as abuse of process. At the hearing on 17 June 2005, that summons was stood over to the beginning of the substantive hearing.  On 31 March 2006, the Commissioner issued another summons seeking to have parts of the Legal Advice to Mr Doran struck out on the ground of res judicata.

123.  The arguments on the merits of the summons overlapped with the discussion on the merits of the appeal and at the highest, the disposal of the summonses at this stage may only have costs implications.  I therefore do not intend to devote a lengthy part of this judgment to that question.

124.  The grounds of the striking out applications can broadly speaking be identified as follows,

(a)	The challenged materials (viz. parts of Mr Doran’s report and the Legal Advice) are inconsistent with the Agrila decision;
(b)	Some of the points advanced by the Appellant were inconsistent with Lai Kit Lau and that could and should have been canvassed in the context of the determination of the preliminary issues in Agrila.

125.  I have already discussed the implications of the Agrila decision and why the valuation of Mr Doran and Mr Lynch should be rejected at the earlier part of this judgment.  It can be seen that I am in agreement with the Commissioner on the first ground whilst I do not think Lai Kit Lau has much relevance for present purposes.

126.  In the circumstances, I do not deem it necessary to make any order on the summonses apart from making a costs order nisi that the Appellant shall pay the Commissioner’s costs of the summonses. 

Mr W. K. LO, Member of the Lands Tribunal:

Background information on valuation of the Appeal Tenement

127.  The Appeal Tenement, known as Inland Lot No. 8874, is situated at Electric Road, Hong Kong.  It was a piece of vacant site sold by public auction on 11 December 1996 by the Government to the Appellant at a sale price of $760 million.  It is not disputed that the Appellant purchased the Appeal Tenement for development into an office building.  At the time of hearing, the office building was completed and occupied for some years.  The Tribunal, accompanied by the legal representatives and the experts for the parties carried out a site inspection during the hearing.  A summary of the physical description, lot particulars, development potential as well as restrictions on the development of the lot, and the completed development, which are not disputed, can be found in pages 7 to 9 of Bundle PKT.

128.  The appeal by Best Origin Limited (“the Appellant”) was made under section 26 of the Government Rent (Assessment and Collection) Ordinance (“the Rent Ordinance”) against a Notice of Decision issued by the Commissioner pursuant to section 21(1) of the Rent Ordinance on 20 February 1998 altering the rateable value of the Appeal Tenement from $37,050,000 to $29,640,000.  The Appellant submits that if the Tribunal determines that the rateable value at a figure below that entered into the Government Rent Roll following the Commissioner’s Notice of Decision, then the appeal will succeed.  Since the figure now contended by the Commissioner through its expert witness, Mr P. K. Tang (“Mr Tang”) at $26,880,000 is lower than the figure in the Notice of Decision, the appeal must, according to the Appellant, technically succeed and the Tribunal should direct that the figure currently in the Government Rent Roll be reduced.  If the Tribunal is persuaded to determine a figure lower than $26,880,000, then it should determine that lower figure to be the correct rateable value and should also direct to amend the Government Rent Roll accordingly.

129.  Under section 26(1) of the Rent Ordinance, the question to be determined by the Tribunal on the appeal is whether the rateable value shown in the Government Rent Roll following the Commissioner’s Notice of Decision accords with the definition of rateable value in the Rating Ordinance (as applied by section 8 of the Rent Ordinance) according to the interpretation of all relevant judicial decisions and taking into account all relevant evidence adduced to the Tribunal.

130.  The Appeal Tenement falls to be assessed under section 8(2) of the Rent Ordinance, section 2 of the Rent Regulation and section 7(2) and 7A(2) of the Rating Ordinance.  It is agreed by the parties that the relevant date within the meaning of section 7A(2) of the Rating Ordinance for the purposes of assessing the rateable value of the Appeal Tenement is 1 July 1996 (“the Relevant Date”).  Under section 7A(2) of the Rating Ordinance, the Commissioner is required to take certain matters as they were on the date the list comes into force, which for the purpose of this Appeal is 28 June 1997.  These matters are: (a) the state of the tenement, (b) any relevant factors affecting the mode or character of occupation, and (c) the locality in which the tenement is situated, including the occupation and use of other premises in the locality and the transport and other facilities of the locality.

131.  This case is the first case of an appeal by an owner of a development site against the determination of Government rent for the site by the Commissioner, following the Court of Final Appeal’s (“the CFA’s”) decision on Commissioner of Rating and Valuation v Agrila Ltd. [2001] 2 HKCFAR 83 (“Agrila”).  Therefore, although not strictly a test case, both parties have spent a considerable amount of time and effort in exploring various issues surrounding the assessment by the Commissioner of the Government rent payable by the Appellant in respect of the Appeal Tenement for the period from the date of purchase and prior to the completion of its development into a 35-storey office building whereupon the building would be assessed to rates, which will also form the basis of assessment of the new Government rent.

132.  The parties in this case have called a number of expert witnesses who have among them produced a multiplicity of valuation reports.  Before the hearing, the parties managed to agree on certain issues so that there were no longer the need to call for the evidence of experts on the construction period for the completion of the building on the site and there were no longer any dispute on the market capital value of the site and the market rental value of the office and retail space in the completed building on the site, both at the relevant date of valuation.  It is convenient at the beginning of this part of the Judgment to set out a summary of the valuations produced by their experts in Table 1 below.

Table 1 – Summary of assessments of Rateable value of the Appeal Tenement

	Basis of valuation & mode or character of occupation	Appellant			Commissioner
		Mr Doran/ Mr Lynch	Mr Charman	Professor Hughes	Mr Tang
1.	Comparative Method: use by the Building Contractor as the Hypothetical Tenant (“HT”)	Nominal at $1* (Primary valuation)	-	-	-
2.	Comparative Method: for   Open storage Use	$764,000* (Alternative valuation)			-
3.	Comparative Method: for   Open Car Parking Use	$1,930,000* (Alternative valuation)			-
4.	The Contractor’s Basis (“the CB”): as a development site	-	$3.62M (Estimated site value of $95.267M x decap rate of 3.8%)	-	$26.88M (Estimated site value $611M** x decap rate of 4.4%) (Primary valuation)
5.	The   Receipt & Expenditure Method   (“    the  R & E Method”): as a   development site	-	-	-	Ranging between (i) $22.2    9  M (assuming HT’s return at 18% nominal  , or the “WACC”***    ) and  (ii) $30.    11  M (assuming HT’s return at 14.1  3  % nominal, or the project return) (Check valuation)
6.	Professor Hughes’ “Property Development Model”     (“the   PDM”): a    s a development site	-	-	$4.56M (revised Appendix GH-16)	N/A

*	Values agreed by the parties
**	Estimated site value was based on the auction price of $760 million adjusted for time
	(which adjustment was agreed by the parties after negotiation prior to the hearing) and the gross up factor to account for the liability to pay Government Rent
***	WACC = Weighted Average Cost of Capital

133.  The valuation figures shown in the above table are extracted from the main valuation under the respective basis put forward by the expert valuers of either party.  In most situations, they are the figures adopted by the valuers prior to the issues of the “Speaking Notes” of the valuers giving evidence in the Tribunal.  In addition, the expert valuers instructed by the parties have both prior to and during the hearing undertaken dozens and dozens of sensitivity analysis types of valuation, either supporting the valuation of one party or refuting the valuation of the other party.  These will be considered, where necessary, under the main valuation basis as outlined in the above table.

134.  The Appellant in its Final Closing Submission submitted that its primary position was the valuation of $1 (nominal value) arrived at by Mr Doran and Mr Lynch.  If the Appellant’s primary position were not accepted, the Appellant also submitted that the valuation of Mr Charman (at $3.62 million p.a.) and Professor Hughes (at $4.56 million p.a. as per his revised valuation at Appendix GH-16) would assist the Tribunal in this valuation exercise.  Furthermore, it is the Appellant’s case that although the Appellant might not have positively advocated valuation on the CB or R & E Method in their traditional formats, the Appellant’s case as a whole should not be rejected.

135.  The Appellant submitted that there was no legal rule or limit on the appropriate method of valuation, which should be based on the mixed findings of law and facts in each case.  The CFA’s decision in Agrila also clearly did not specify a particular method of valuation.  More importantly, the Appellant stated that “many of the areas of dispute arise in the context of both the Commissioner’s and the Appellant’s valuations.”  Therefore, the Appellant has throughout the hearing drawn the Tribunal’s attention to the unreliability of adopting the valuation of Mr Tang or his colleagues either on points of law or the basis of evidence adduced by the witnesses, or on both.

136.  On the other hand, the Commissioner summarized in the Final Closing Submission the following:

(1)	the primary and alternative valuation of Mr Doran and Mr Lynch, being the direct comparison approach, should certainly be rejected in their entirety because they were based on invalid legal directions (as set out in the Legal Advice in App. 7 of Mr Doran’s report) as a result of which the Appeal Tenement was not valued on the mode or character of occupation as a development site;
(2)	the Appellant did not set out any valuation using the traditional CB of valuation;
(3)	on the other hand, the Commissioner has set out a valuation using the CB of valuation from day 1 (i.e. Mr Tang’s first report) and unless the Tribunal accepted the Appellant’s submission that the CB could not be used, the method should be adopted as a primary method in this valuation;
(4)	the Appellant did submit a valuation by Mr Charman which was neither a valuation using the CB nor the R & E Method but a postulation of Mr Charman for which the Commissioner contended, was simply faulty in construction;
(5)	the Appellant also submitted a valuation by Professor Hughes in his revised Appendix GH-16, his Property Development Model, which, the Commissioner contended, was also neither a valuation using the R & E Method nor the Discounted Cash Flow Method but was a hybrid of both, and was faulty in the structure as well as in the assumptions of the inputs of the Model;
(6)	on the other hand, Mr Tang for the Commissioner submitted a check valuation using the R & E Method of valuation; and
(7)	finally, Mr Tang’s primary valuation using the CB, at a value of $26.88 million p.a. (using the agreed adjusted land value figure of $593 million, based on the auction price as adjusted for time and grossing up to reflect the liability to pay Government rent, and a decap rate of 4.4%), which was supported by his check valuation using the R & E Method (ranging in values of $22.29 million based of a 18% WACC or $30.1 million based on a 14.13% project return) should be accepted by the Tribunal.

The Contractor Basis of valuation

137.  Before we consider in details the various valuation put up by the experts for the valuation of the Appeal Tenement, it is convenient to set out below “The Approach to Valuation” in the CB of valuation, based on the Guidance Notes prepared by the Joint Professional Forum:

“	2.	The Approach to Valuation
		2.1	The recommended approach to valuation comprises the five component Stages listed below which are dealt with in more details in Section 3.
			Stage 1 –	Estimate the replacement cost of the siteworks, buildings, rateable structures and rateable plant and machinery.
			Stage 2 –	Adjust the replacement cost to reflect any deficiencies in the buildings etc.
			Stage 3 –	Value the land.
			Stage 4 –	Decapitalise the sum of Stage 2 and Stage 3 by the appropriate interest rate.
			Stage 5 –	Stand back and look at the result of Stage 4 and make any further adjustments considered appropriate.
		2.2	It should be noted that in some cases a sixth stage has been added under the heading ‘negotiations’ to reflect the ‘higging’ process deemed to take place between landlord and tenant.  However, if Stages 1 to 5 are properly applied, it may not be necessary to have a Stage 6.
		2.3	It is important to note that whilst the valuation process can usefully be broken down into five (or six) stages, it is the property as a whole which is to be assessed.  The valuer should therefore be conscious of the need to adopt an integrated approach and not be diverted into regarding each stage in isolation.”

The R & E Method of valuation

138.  Similarly, we set out below the steps that should be taken in the other method employed by the experts, the R & E Method of valuation, based on the similar set of Guidance Notes prepared by the Joint Professional Forum:

“	4.1	An explanation of the R & E method as applied by valuers can be found in the case of Kingston Union AC v. Metropolitan Water Board (1926):
		‘	From the gross receipts of the undertakers for the preceding year they deducted working expenses, an allowance for tenant’s profit and the cost of repairs and other statutable deductions and treated the balance remaining (which should presumably represent the rent which a tenant would be willing to pay for the undertaking) as the rateable value of the entire concern.’
	4.2	It is necessary to look at this explanation in the light of modern practice and development of the R & E method through case law.  The methodology is based upon the approach set out below:
		(a)		Gross Receipts should be determined by taking into account all income reasonably able to be derived from occupation of the property.
		(b)		The proper Costs of Purchases made in order to produce those receipts should be deducted to determine the Gross Profit.
		(c)		From the Gross Profit the Working Expenses should be deducted to determine the Divisible Balance.
		(d)		The Divisible Balance is the sum available to be shared between the landlord and the tenant.  It comprises two main elements:
				(i)	the Tenant’s Share – to provide a return on any tenant’s capital employed and a reward to the tenant for his venture reflecting the extent of the risk and the need for profit.  This is deducted from the Divisible Balance to leave;
				(ii)	the Landlord’s Share, i.e. the rent payable (which becomes the rateable value).”

Primary valuation of the Appellant

139.  As we said before, the Appellant’s primary valuation is the direct comparative approach undertaken by Mr George Doran and Mr Lynch.  Mr Lynch, a co-signer of the reports also gave evidence in the hearing.  However, it is agreed by the parties that whether the valuation by Mr Doran / Mr Lynch, should be accepted by the Tribunal really depends on whether the Appellant’s interpretation and application of the judgment of CFA in Agrila, is correct.  Simply put, the Appellant contended that the decision of the CFA prevented the realization of the “development potential” of the Appeal Tenement to be taken into account whereas the Commissioner argued otherwise.

140.  The framework within which the Appellant contended that any valuation of the Appeal Tenement should be prepared was set out in the Legal Advice in Appendix 7 to Mr Doran’s report (“Legal Advice in App. 7”).  On the other hand, the Commissioner submitted that the Legal Advice in App. 7 was incorrect.  This is the root of the main differences in the valuations undertaken by the experts for the two parties.  In both the Openings and the Final Submissions, counsel for the parties gave lengthy submissions as to whether the said Legal Advice in Appendix 7 is correct.  We have addressed this issue in a separate section above and will not repeat here.

141.  On the basis of the Appellant’s Legal Advice in Appendix 7, Mr Doran / Mr Lunch carried out their valuations.  They concluded that (a) a building contractor would be the only possible hypothetical tenant; (b) a building contractor would not pay anything to occupy the Appeal Tenement; and (c) the rateable value for the Appeal Tenement would be a nominal sum of $1.  Mr Lynch confirmed these conclusions in the hearing.  He further accepted that following the Legal Advice in App. 7, the same results would apply to all other development sites in Hong Kong.

142.  In the alternative, Mr Doran / Mr Lynch also considered if they could find any other evidence of value comparable to the value of occupation of the Appeal Tenement for construction activity by the building contractor as the HT.  They opined that the best alternative comparables would be the value of the comparable sites used as a contractor’s work site or for open storage purposes, which was valued at $760,000 per year.  Mr Lynch admitted during the hearing that these uses fell into a different “mode or character of occupation” to the use of the Appeal Tenement as a development site.  He explained in his evidence in chief (Day 6, 407:15-26) that this alternative valuation exercise was done because

“

… we couldn’t find any evidence within the existing mode and character, so as a valuation practice, we cast our net wider to try to find alternative uses that we could use as a substitute, and that’s what the second exercise is doing.  We recognize it is not within the same mode or character, but it would give a useful guide in the absence of the first approach to reference the value for the purposes of this assessment.”

143.  However, Mr Lynch accepted in cross-examination that his alternative valuation was also based on the Appellant’s Legal Advice in App. 7, and that he had not attempted to value the Appeal Tenement on the basis of the Commissioner’s assumptions, i.e. on the assumptions that his valuation is not constrained by the Legal Advice in App. 7.  He further accepted that:

(1)	although in his alternative valuation, he looked for other evidence of value, a building contractor would not need to pay more than $1 to occupy the site for that purpose;
(2)	in doing his alternative valuation, he was exposing the site to competition for another type of land use;
(3)	no one would pay $760,000 per year for open storage unless they could actually use it for that purpose for the entire site, which would be inconsistent with using the site for putting up a 36-storey building; and
(4)	neither the primary nor the alternative valuation of Mr Doran and himself would take into account the potential of the site to accommodate any kind of development.

144.  The Commissioner did not take issue with the valuation of the Appeal Tenement on the basis of (i) occupation by a building contractor for construction activity on the Appeal Tenement and (ii) occupation as a work site or for open storage, in the sum of nominal $1 and $760,000 per year respectively, both on the assumption that the Legal Advice in App. 7 was valid.  In addition, the Commissioner submitted that if the Tribunal accepted Mr Lynch’s alternative valuation approach, the Appeal Tenement could equally be used as an open car park, as set out in Mr Brooke’s valuation report (NB Rep. 2, 81) in which the agreed value was $1,930,000 p.a.

Tribunal’s decision on the primary valuation of the Appellant

145.  We already explained earlier that based on our understanding of the CFA’s decision in Agrila, the Ordinance as well as the basic principles of rating, we do not accept that the Legal Advice in App. 7 sets out the correct legal direction for the valuers undertaking the primary and alternative valuation.  We therefore reject Mr Doran / Mr Lynch’s valuation in their entirety.  For the same reason, we do not find Mr Brooke’s valuation for open car parking use, as an additional alternative to the Appellant’s primary valuation as a fallback approach to be relevant.

The Appellant’s valuation on the CB - Mr Charman’s approach

146.  Neither Mr Doran nor Mr Lynch was instructed by the Appellant to carry out valuation on the Contractor’s Basis.  The other local valuer, Mr David Lee (“Mr Lee”), himself not being familiar with rating, was also not asked to carry out any valuation of the Appeal Tenement on either the Contractor’s Basis, the R & E Approach, or any other basis.  His main areas of work and evidence included the market valuation of the Appeal Tenement on the relevant date, the evidence of occupational rental value of space in the completed building on the Appeal Tenement (i.e., the initial occupational rent), the grading of the completed building by the Commissioner, the depreciation rate of the office space as well as the growth rate of rental value of space in the completed building, if any.

147.  Of the other two witnesses called by the Appellant, Professor Hughes did not in his various reports carry out any valuation on the Contractor’s Basis.  As explained below, although Mr Charman was a rating valuer with wide experience in UK and familiar with both the comparative method and the CB of valuation, he has not carried out any valuation of the Appeal Tenement on either of these 2 bases in the traditional way.

148.  Mr Charman criticized the valuation by the Commissioner’s expert, Mr Tang using the CB of valuation.  He has instead devised his own valuation model in his reports for which we will consider in details below.

149.  In the Final Closing Submission, the Appellant clarified that the Appellant had attempted to produce 2 valuations on the Commissioner’s approach, prepared by Mr Charman and Professor Hughes respectively.  They rejected the Commissioner’s allegation that although both experts were originally instructed to apply the Commissioner’s assumptions, they had not done so in every respect.  According to the Appellant, the misunderstanding between the parties probably arose because whilst Mr Doran and Mr Lynch were instructed “to adhere to the legal framework” in the Legal Advice in App. 7 of Mr Doran’s report, Mr Charman and Professor Hughes were given instructions “to reflect the Commissioner’s overall objective”, which was to bring into assessment the development potential of the tenement.  Therefore, the instructions given by the Appellant to both experts were intended to mean “the legal framework and the terms of the hypothetical tenancy as defined by the Commissioner.”   The Appellant blamed the Commissioner in not putting up a precise definition of the latter in the first round of reports filed and exchanged by the parties so that the experts might have a misunderstanding when carrying out valuation.

150.  As we said before, notwithstanding the possible misunderstanding by the Appellant’s experts on the precise legal framework and the terms of the hypothetical tenancy of the Commissioner, the Appellant has proposed a valuation that was carried out by Mr Charman in his First Report, as amended in his Synopsis and then further amended in pages 94 & 95 of his Supplementary Report.  Mr Charman’s valuation is in any event the only valuation that has the label of valuation on the CB made by any of the Appellant’s experts.

151.  In his First Report, Mr Charman set out, in his own words, to assess what the HT could afford to pay by way of site rent on the assumption that the payback will start when six years have elapsed so that, if the tenancy would come to an end at that point, he would not lose money.  If the hypothetical tenancy would continue for longer than 6 years, the HT would begin to make a profit.  He concluded that the HT would take the view that he ought only to base his bid for the Appeal Tenement on the assumption that the tenancy might terminate after 6 years. In other words, his valuation was based on his assumption that the HT would not expect the hypothetical tenancy to last for more than 6 years.

152.  On this basis, Mr Charman set out in page 36 of his First Report the following methodology:

“	Value of the site to the HT at the relevant date equals:
	Present value of the total rent which will be received by the tenant for the completed development for 6 years
	minus
	present value of the rent for the site which the HT will pay over the six years of the tenancy
	plus
	market value of the building at the time that the tenancy is terminated
	minus
	the cost of erecting the building.”

153.  Mr Charman said that the last 2 items in the above equation cancelled out as a result of which he only needed to deal with the first part of the equation.  He assessed the “site value” as being the net present value of the HT’s right to the net cash flow over the initial six years (being the difference between the average annual rack rent for the first 6 years and the site rent) on the assumption that the rental values in these six years would only keep up with inflation but without any growth in real terms.  The rental values retained by the HT was then capitalized for six years using a “discount rate” (or a years’ purchase figure as accepted by him) of 8% resulting in the net present value of the Appeal Tenement to the HT, or the site value to the HT, at the relevant date.  Although Mr Charman said in his report that his choice of the discount rate figure was backed up by Professor Chan, he subsequently corrected that this was not so.  He clarified during cross-examination that his 8% was derived from yields for Grade B offices, which was about 6% at the relevant time.  He adjusted the yield from 6% to 8% using his judgment as a valuer to reflect a riskier nature of the investment.

154.  Mr Charman further converted his estimated site value of the Appeal Tenement to the HT, in the sum of $95,267,000 (para. 15.55 of JC Report 1 as amended by AAD Tab10) to an annual rent by using a decapitalisation rate (“decap rate”) of 3.8%, the latter of which was derived from Professor Hughes.  On this basis, Mr Charman arrived at a rent for the Appeal Tenement in the sum of $3.62 million.  This, the Appellant submitted, should be the rateable value for the Appeal Tenement.  In addition, the Appellant said in the Final Closing Submission that,

“

…. since Mr Charman prepared his valuation, Professor Hughes has reviewed his evidence on the decapitalisation rate and there has been much other evidence concerning decapitalisation rate.  If the Tribunal were to find for a different rate, then Mr Charman’s valuation could be adjusted accordingly.”

155.  Mr Charman explained in the hearing that he had used a method of trial and error in his formula to arrive at the rental value by inputting an estimated site value a number of times until the said inputted site value equaled to the calculated site value in his formula.  He also did not agree with the Commissioner that the fact that his two different calculations could simply be expressed by one formula (i.e. X = Y / 6.692 where X = site rent payable by the HT to the HL and Y = average of annual rack rent receivable by the HT during the 6 years, per Commissioner’s Additional Document Tab 10, “RAD Tab 10”) would invalidate his method of valuation.

156.  However, during the hearing, Mr Charman did agree with the Commissioner the following:

(1)	his valuation was, as he himself described, a “worst case scenario” on the basis that the tenancy would not last for more than 6 years but the average annual rent figure would increase the longer the tenancy;
(2)	his valuation assumed no real growth in occupational rents;
(3)	his valuation was originally put forward based on a construction period of 3 years and 9 months (i.e. 45 months), a further period of 18 months to reach stable occupancy and an additional period of 9 months during which the rents would be received by the HT assuming 90% occupancy before sale of the building;
(4)	even though he had later been aware of the agreement between the parties of a 32-month construction period, he did not change the valuation to reflect this change;
(5)	his valuation did not require a site value to be found at all in order to get to the site rent as the annual site rent would be driven entirely by changes in the average annual occupational rental value which was dependent on the period chosen for analysis; and
(6)	his valuation was dependent on (i) the assessment of the average annual rack rent over the first 6 years; (ii) the adoption of a year’s purchase (“Y.P.”) factor of 4.623, being the Y.P. for 6 years at 8%; and (iii) the adoption of Professor Hughes’s decap rate of 3.8% (the combination of the adopted Y.P. factor and the decap rate of 3.8% would give rise to the factor of 6.692 as revealed by the Commissioner in RAD Tab 10).

157.  Mr Charman in his First Report also produced what he described as an alternative valuation using Mr Hatchwell’s method of valuation of assessing market value.  However, he subsequently withdrew this in his Second Report saying that it was unworkable and that his Depreciated Replacement Cost valuation was the only practicable method of valuing the building.

The Commissioner’s criticisms of Mr Charman’s valuation

158.  The Commissioner criticized Mr Charman’s valuation mainly on two grounds: (a) Mr Charman’s assumed duration of the tenancy; and (b) the methodology of his valuation.

159.  We first consider the issue of the assumed duration of the tenancy in Mr Charman’s valuation.  Firstly, the Commissioner pointed out that Mr Charman had failed to substantiate his assumed duration of a 6-year tenancy.   As Mr Charman repeatedly admitted, his assumption of a 6-year duration was the HT’s “worst case scenario” as it was based on the fear that a notice to quit might be served by the hypothetical landlord (“HL”) terminating the tenancy at the end of year 6.  However, the Commissioner submitted that Mr Charman had not considered the possibility that the HL might be prepared to have the tenancy to last longer because, for example, the longer the duration of the tenancy, the higher would be the site rent.  In addition, the Commissioner pointed out that even though after the parties had agreed on the duration of the construction period as 32 months, Mr Charman still did not reduce his assumed construction period of 45 months in his valuation.  Further, the Commissioner submitted that Mr Charman’s opinion that the tenancy would not last for more than 6 years because the HL would serve a notice of quit at a low point in the market was not supported by the market evidence at the time, in 1996.

160.  Summing up, the Commissioner submitted that Mr Charman’s assumption of a fixed, 6-year duration of the tenancy between the HT and the HL was so unrealistic in both the real world and the rating world that this valuation should be discarded.  In the context of a development site in the real world, it would not be realistic to assume that a developer, whether developing for investment or for sale, would not have required a long-term interest in the site without which it would not be possible for him to realize the development potential of the site.  During cross-examination, Mr Charman accepted this assumption for the developer in the real world.  The Commissioner submitted that the same should be assumed for the rating world.

161.  Regarding the methodology employed by Mr Charman, he adopted the process of firstly assessing the “site value”, certainly not in the conventional meaning but as the net present value of the HT’s right to the net cash flow (i.e. the difference between the net rack rent and the site rent) over the initial 6 years, and secondly, multiplying his assessed “site value” by a decap rate of 3.8% (as advised by Professor Hughes in the latter’s report).

162.  Although Mr Charman said that he had used a method of trial and error in his computations to arrive at the rental value for the site, however, the Commissioner illustrated by simple algebra in RAD Tab 10 that his two formulae could simply be reduced to one whereby at a fixed duration of a 6-year lease (which was the only scenario envisaged by Mr Charman), the site rent to the HL would always be a fixed fraction (i.e., 1/6.692) of the rack rent.  Expressed as a rough percentage, this is equivalent to that about 15% of the rack rent would be received by the HL with the remaining 85% received by the HT.  With Mr Charman’s experience in valuation, we do not accept that he could not have foreseen this relationship in his formulae and the simple deduction in algebra.  We are quite surprised that he had not explained this in more simple terms in his reports and in the hearing, before the Commissioner’s expert managed to reveal this in RAD Tab 10.

163.  The Commissioner submitted that in Mr Charman’s valuation, his adopted decap rate was not applied to the HL’s interest, as it should be in the CB of valuation, but to the value of the site to the HT.  The Commissioner submitted that this was fundamentally wrong.

164.  Also, Mr Charman agreed during the hearing that although he was an advocate of using a money market approach (as in Ebdon which he relied on) in the construction of a decap rate, he had not himself constructed one using such an approach.  Instead, he relied on Professor Hughes for the decap rate he used in his valuation.  We will discuss further the issue of the choice of the decap rate under the section of CB valuation by Mr Tang.

165.  To conclude, we agree with the Commissioner that Mr Charman’s valuation, which was purported to be a valuation using the CB should be rejected on the grounds of its unrealistic assumption of the 6-year fixed duration of the lease and the faulty methodology.

Mr Charman’s “check valuation”

166.  In his Supplementary Report dated 7 July 2006, Mr Charman attempted to carry out an exercise to check “what position the HT would be in at a future date if he commits himself to a ground rent of $26 million per annum, as proposed by the Commission” (JC page 83).  He adopted the Commissioner’s assumptions regarding rental values, construction period and occupancy rates.  His analysis show that for the first 11 years, the HT would be making a substantial loss and from the 12th year, the HT would start to break even, albeit at a minimal 0.03% p.a.  He came to the conclusion that the HT simply would not enter into a tenancy when he had to pay such a high proportion of the rack rents but only had such paltry returns for such a long time.  Mr Charman then up-dated his analysis by producing his Revised Table JC-13 at JC pages 90-91.

167.  Mr Hatchwell however commented in LH pages 231-232 that the defect of Mr Charman’s analysis was due to: (i) his assumption of no real growth in rack rents or ground rents; and (ii) his reliance of the Depreciated Replacement Cost (“DRC”) in the formulation of the capital sum which the HT received for his building, both of which contributed to the seemingly untenable results that indicated that the tenement would fail to be let at all.  Mr Hatchwell suggested that the result would change considerably, if, for example, the developer’s profit was added to the DRC for passing to the HT.  It would change the outcome giving the HT a substantial profit.  And, the assumption of no real growth in rack rents or ground rents, Mr Hatchwell opined, was obviously not consistent with the yields that the Appellants were applying to the office building as an investment property.

168.  For reasons stated above, we agree with Mr Hatchwell’s conclusion that Mr Charman’s check valuation did not produce the correct result of the site rent payable by the HT.

Primary valuation of the Commissioner – using the CB

169.  The Commissioner submitted that the well-known case of Garton v Hunter [1969] 2 QB and [1969] 15 RRC 145had confirmed 2 points in principle- (1) all relevant valuation evidence was admissible; and (2) the 3 different methods (i.e. rental test, contractor’s test and the profit’s test), if properly applied, should lead to the same answer. This was not disputed by the Appellant who stated that, “a valuation on the Contractor’s Basis is admissible but the Tribunal must decide what weight to attach to it”.

170.  It is generally known that the Contractor’s Basis (“the CB”) is one of the 3 main traditional methods employed in assessing rateable values, the other two being the Comparative Method and the Receipts & Expenditure Method (“the R & E Method”).  For tenements that are not commonly let in the market, the employment of the Comparative Method is usually not possible because of the lack of suitable comparable lettings.  Therefore, for these tenements, which usually include special industrial plants built for special purposes, oil and chemical installations, telecommunication facilities, institutional buildings such as schools, universities, town halls and utility undertakings, valuation for rating purposes have to be carried out by using the other two methods.

171.  Mr Tang quoted in page 32 his First Report the passage from the Tribunal in the case of Agrila Ltd & Ors v Commissioner of Rating and Valuation [1999] 2 HKC 168, 179H – 180C:

‘	When there is no satisfactory comparable rental evidence or other direct evidence, contractor’s test will be resorted to; as for in instance in the following cases:
	“	(1)	Mobil Oil Hong Kong Ltd. v Commissioner of Rating and Valuation [1993] HKDCLR 77 – where contractor’s test was admitted in evidence as a check on the estimate on comparable basis and the Tribunal eventually preferred the estimate by using the comparable basis.
		(2)	Liverpool Corp v Chorley Union Assessment Committee & Withnell Overseers [1912] 1 KB 270 – where the owner kept the land as gathering ground for water rendering the ordinary test inapplicable, - the price paid for the land was admitted as evidence, though not conclusive, to estimate what a hypothetical tenant would pay.’

172.  The Commissioner submitted that the primary valuation of the Appeal Tenement was the valuation by the CB.  It involved the estimation of 2 factors, the value of land and the estimation of the appropriate decap rate.  Mr Tang for the Commissioner carried out valuation on this basis from the beginning of his various rounds of reports.  Mr Tang concluded that he relied primarily on the CB because “(a) it implicitly reflects regular rent reviews, and does not require any specific adjustments for arriving at an initial rent and is therefore more direct; and (b) it is based upon the market evidence of the site value and property yield, thus obviating the need to make specific assumptions to the details of the cashflows.”

173.  On the other hand, the Appellant submitted that in the past, the justification for using the CB of valuation had arisen in 2 situations: - “(a) Where the cost of providing alternative premises may be treated as a guide to the value of occupation to the occupier and may be converted to an annual sum.  The is the “classic” situation described in Dawkins (VO) v Royal Leamington Spa [1961] 8 RRC 241; (b) Where the actual occupier of a tenement has recently incurred the cost of acquiring it, this may be evidence that he considered it worthwhile to do so and consequently that the cost he incurred may represent the value to him of permanent ownership which may be converted to an annual sum.  This was the situation in Liverpool Corporation v Chorley Union Assessment Committee [1912] 1 KB 270.”

174.  In particular, the Appellant submitted that, prior to this case, the CB had only ever been used in the circumstances where “the occupier’s motivation for acquiring the tenement is to use it for its existing purpose for the foreseeable future, and he is not motivated by the prospect of selling the tenement in order to receive the proceeds of sale.”  Also, the Appellant submitted that the CB had not previously been used in a rating context “where the tenant has to incur substantial expenditure over a period of time to make the building ready before it can be used for its intended purpose.”

175.  Therefore, the Appellant alleged that Mr Tang’s valuation was fundamentally wrong in principle and the inputs adopted by Mr Tang were also erroneous.  Mr Tang was wrong in his assumption that the decapitalisation stage could properly convert the benefits of permanent ownership to the rent payable under a year-to-year tenancy.  The Appellant decided not to instruct its experts to provide any valuation on the CB but submitted that in the event that the Appellant’s primary valuation as carried out by Mr Doran / Mr Lynch was not accepted by the Tribunal, Mr Charman’s valuation, or Professor Hughes’ should be taken as the proper substitute valuation for the Appeal Tenement.

176.  It may be correct that this case is novel in that there was no precedent case of valuing development land for rating purposes both in Hong Kong and elsewhere.  But as explained earlier in this Judgment, that is the result of the development land not being rateable in the past.  For the same reason, the R & E Method or the PDM was also never used in the rating of development land before. With the enactment of the Rent Ordinance, we have to deal with the situation such as the present case.  Therefore, that CB was not used in the rating of development land in the past should not be a ground for rejecting the method.

177.  Towards the end of the hearing, the President invited Mr Tang to address a few questions regarding the details of his CB valuation.  Both parties then adduced a number of documents giving comments on the various inputs that are appropriate for a valuation of the Appeal Tenement on the CB.  At the end, mainly because the parties were desirous of not delaying the Final Closing Submissions, they agreed not to call for the evidence of any of the experts who produced these additional documents but asked the Tribunal to consider the documents without the benefit of hearing these experts again.

178.  Before going through Mr Tang’s valuation in details, we first consider the various issues raised by the Appellant who contends for one reason or another that the CB is in principle not an appropriate method of valuation in the present case.

Is Mr Tang’s “CB valuation” a CB valuation?

179.  Mr Lynch, the Appellant’s own witness, accepted that but for the legal directions given to Mr Doran and him in Appendix 7 of their Report, he would have decapitalised the auction price (after making adjustment to reflect the time difference) to arrive at the rateable value for the Appeal Tenement.  He did not carry out such a valuation because he was directed to ignore the value of any future rents or sale proceeds from the building to be completed on the site.  Also, although Mr Lynch was a valuer with rating experience in Hong Kong, he was not asked by the Appellant to prepare a CB valuation on the Commissioner’s legal assumptions. Of course, as said earlier in this Judgment, we have already decided that this legal direction given to both Mr Doran and Mr Lynch by the Appellant was wrong.

180.  Mr Lynch also agreed with counsel for the Commissioner that he would have arrived at the rateable value by firstly deriving a decapitalisation rate from the property yields as a primary source and secondly applying that rate to the adjusted auction price (Day 6, 448).  He also agreed the following: (1) there was no difference between decapitalising a land value to arrive at a rent and the reverse process of capitalising a rent to arrive at a capital land value; (2) the CB valuation as performed by Mr Tang would not produce a ceiling value; (3) the CB valuation was more reliable than the R & E Method as the latter was dependant upon a large number of inputs and judgments as well as discounting over a number of years (Day 6, 454 – 459); and (4) the CB valuation as used by the Commissioner “could easily be replicated” in the case of other development sites in Hong Kong, whether for office, residential or industrial purposes, and reflected the development potential in each case (Day 6, 458:40 – 459:8).

181.  However, the Appellant drew to our attention that Mr Lynch had answered in cross-examination that he would have used the CB of valuation but that was “subject to finding the right de-capitalization rate” and that in re-examination, he said that he would not expect to find any evidence of a yield for bare land.  Moreover, the Appellant submitted that it would not be fair to Mr Lynch as a witness giving evidence other than his own valuation in the present case since he did not have the benefits of reading any of the reports on either side.

182.  The Appellant reminded us that Mr Charman had considered that the “CB valuation” as presented by Mr Tang was not a proper CB valuation because (1) Mr Tang only used Stage 3 and Stage 4 of a 5-Stage CB valuation, as set out in the Joint Rating Forum’s Guidance Notes on the CB; (2) in a normal CB valuation, the land, which was taken as the value for which the tenement was used at the valuation date, usually represented a relatively small proportion of the total capital costs but in Mr Tang’s CB valuation, the capital land price became the only component for decapitalisation; and (3) the Commissioner’s valuation was “value based” rather than “cost based”, which was the essence of a CB valuation (this being the reason for the CB commonly known elsewhere as “the Cost Approach” or the “Depreciated Replacement Cost Method”).  In emphasizing that Mr Tang’s “CB valuation” was not a CB valuation in the usual sense of the words, Mr Charman preferred to describe it as a “de-capitalization of a capital land price” or “the reverse of the Income Approach Method” but to avoid confusion, he continued to use the same description as that of Mr Tang.

183.  In response to these criticisms, the Commissioner admitted that as suggested by Mr Charman, in most cases where the CB was used, the Stage 3 element, land, might only be a relatively small contribution to the total value whilst in the present case, the land formed the only capital component in the valuation process.  However, the Commissioner said that there was no basis for saying that the CB would then be less reliable in the present case.  On the contrary, the Commissioner submitted that the CB was in fact more reliable for the following reasons:

(1)	The Lands Tribunal had endorsed the use of the CB where the land value represented the major component in the valuation (Royal Hong Kong Golf Club v. Commissioner of Rating and Valuation [1977] HKLTLR 236);
(2)	The typical complications of Stage 1 and/or Stage 2, such as the problems relating the tenant’s alternative to the existing installations at the valuation date and the issues of obsolescence, which were typically associated with many of the CB valuation cases concerning complex oil or chemical installations, did not arise at all in the present case;
(3)	Therefore, when compared with the typical CB valuation involving all 5 stages, the present case which does only involve Stage 3 and Stage 4 was a simple and very direct method of valuation involving only two components: the land value and the decapitalisation both of which were based on market evidence;
(4)	“In most reported cases where the contractor’s basis has been used, it was the sole method of valuation employed.  Therefore no question would have arisen as to how the result from the contractor’s basis should be compared to results from other methods, such as the receipts and expenses method”; and
(5)	The cost of land in the valuation was based on the evidence of what the actual occupier paid for the site in December 1996 with adjustments made to reflect the said date of transaction and the valuation date so that this was precisely a case where the CB valuation “proceeds on the direct basis of what it is worth to the real world occupier or lessee of the site” and “there is therefore no need to construct a tenant’s alternative or proxy in order to substitute for the lack of evidence of value to the occupier, which is the typical circumstance in which the contractor’s basis is used.”

184.  The Appellant however rebuked that the Tribunal should not be misled by the Commissioner that because the CB valuation was superficially attractive, involving only 2 steps, it should necessarily be preferred to either (i) the valuation proposed by Mr Charman, in his version of an alternative to the CB valuation, or (ii) to the much more complicated PDM (a non-orthodox “income” approach) propounded by Professor Hughes.  In short, the Appellant stressed that, “correctness must not be sacrificed to convenience”.  The Appellant and its experts gave lengthy written evidence and submissions on how the two components in the Mr Tang’s CB valuation should be assessed and treated with caution.  The Commissioner’s side also assisted the Tribunal by giving further evidence in writing and submissions.  We will deal with that below.

185.  However, at this stage, we do not find that in a rating valuation such as the present case or indeed in any valuation with an attempt to assess certain historical value based on market evidence, in accordance with the provisions of statutes and case laws, we should aim at conducting complicated valuation but reject simple valuation just because they require the estimates of 2 inputs.  That cannot be right.  On the contrary, we do find that a simpler valuation involving just 2 variables certainly have merits over very complicated valuation involving many variables and a very long time span. In the former situation, the effects of the changes of any one of the 2 explicit variables could be more easily understood and measured whilst in the latter situation, the varying estimates in any combination of these variables could easily lead to very different results.  This is what we have found in the R & E Method or any other income approach such as Professor Hughes’ PDM which we will cover later.

186.  Dawkins v. Royal Leamington Spa Corporation and Warwickshire County Council [1961] 8 RRC 242, 251 (“Dawkins”) provided the classic explanation of the theoretical basis for the CB of valuation, which was given by the then Solicitor General in England and accepted by the Lands Tribunal. It was later endorsed by Lord Denning MR in Cardiff City Council v. Williams [1973] RA 46, 50and then again by the English Court of Appeal in Imperial College of Science & Technology v Ebdon and Westminster City Council [1986] RA 233,237 as well as by the Hong Kong Lands Tribunal in Royal Hong Kong Golf Club v. The Commissioner of Rating and Valuation [1977] HKLTR 236, 241.  The rationale given in Dawkins is:

“

…the argument is that the hypothetical tenant has an alternative to leasing the hereditament and paying rent for it; he can build a precisely similar building himself.  He could borrow the money, on which he would have to pay interest; or use his own capital on which he would have to forgo interest to put up a similar building for his owner-occupation rather than rent it, and he will do that rather than pay what he would regard as an excessive rent – that is, a rent which is greater than the interest he forgoes by using own capital to build the building himself.  The argument is that he will therefore be unwilling to pay more as an annual rent for a hereditament than it would cost him in the way of annual interest on the capital sum necessary to build a similar hereditament.  On the other hand, if the annual rent demanded is fixed marginally below what it would cost him in the way of annual interest on the capital sum necessary to build a similar hereditament, it will be in his interest to rent the hereditament rather than build it.”  (Dawins, p. 251)

187.  In Monsanto Plc v. Farris [1998] RA 107, the English Lands Tribunal continued to give general approval to the passage of Dawkins cited above, but also added that, in operating the CB, the tenant’s alternative “is neither built nor valued, it is a total fiction…”

188.  From what we have seen and in light of the Guidance Notes published by the Joint Rating Forum and the case laws involving CB valuation, we do not agree with the suggestion that Mr Tang’s CB valuation is not a proper CB valuation.

Can the adjusted auction price be used in the CB valuation?

189.  Mr Tang gave evidence that he had used the auction price of the subject site as the base value, taking into account the adjustment in time as agreed between the experts for the parties (i.e., Mr Lee for the Appellant and Mr Brooke for the Commissioner) in order to arrive at the estimated market value of the subject site, at $593 million at the relevant valuation date of 1 July 1996.  Mr Tang then grossed-up the adjusted auction price to reflect that in the hypothetical letting, the Government rent would be borne by the HL.  After grossing-up, he arrived at the figure of $611 million, which he adopted as the effective capital value for the purpose of the CB of valuation.

190.  In his Synopsis of evidence, Mr Charman maintained the view originally expressed in his First Report that the price paid at auction should not be an appropriate starting point for assessing the annual rental value to the HT, on the following grounds:

(a)	the auction price reflected the right to build and retain the building for 50 years certain or to sell it in whole or in part at a time of the purchaser’s choosing;
(b)	the developers, in arriving at their auctions bids, were motivated by factors that would not be in the minds of the HT;
(c)	the auction price was not explainable simply by reference to the right to receive the rents from the building for 40 years following construction.

191.  The Appellant’s other experts, Mr Lee and Professor Hughes also gave evidence in support of Mr Charman’s opinion.  The evidence of the Appellant’s experts could be grouped as follows: (1) criticism in the use of the real world auction price in principle; (2) the developers’ bids in land auction were motivated by other factors; (3) the auction price could not be explained by the rights of the HT to receive the occupational rents for the next 40 years as shown in Mr Tang’s R & E Method of valuation; and (4) “the Morton’s Fork” argument as put up by Professor Hughes.

Was the auction price tainted by the benefits of ownership and “real option value”?

192.  Mr Charman’s first reason (as set out in para. 15.15 of JC Report 1) was that the auction price was tainted because it included the benefits of ownership and real option value.  Professor Hughes raised essentially the same argument.

193.  Mr Charman and Professor Hughes suggested that, (a) the adjustment to eliminate the option value from the auction price must be done before Stage 4 of the CB (i.e. application of the decap rate to the total value of the land and the replacement cost of the buildings); and (b) since there is no evidence what the adjustment is (with Professor Hughes admitting that he has “no clear way of putting values on it”, at transcripts page 1485, lines 4-9), the CB must be rejected altogether.  However, both Mr Charman and Mr Lynch agreed that if a decapitalisation rate based on a property yield is applied to a capital land value, the benefits of ownership (including real option value), which are said to be “peculiar” to the freeholder or a Government Lessor, as compared with that of a tenant, are removed.  In other words, the rental value so arrived at would be free of any such benefit.  Professor Hughes also agreed with both Mr Charman and Mr Lynch on this point.

194.  Also, it was pointed out by the Commissioner that if Professor Hughes’ argument on the “real option value” was right, it would have the effect of destroying the use of capitalization and decapitalisation methods not only in rating, and in other general valuation situations.  Professor Hughes agreed with the Commissioner’s suggestion that if he was right that a decap rate could not be applied to the auction price in this case, the same would apply to any decapitalisation of a capital value in England (freehold) or Hong Kong (50-year lease).  When cross-examined, he accepted that the effect of his argument was that the capital value of the right to recover a ground rent assessed by him at $4.56 million (under his PDM at the Updated App. GH-16) was no more than $120 million, even though the landlord had paid $593 million for the land.  According to him, the difference of $473 million would be attributable to the real option value and the special bid arguments (Day 18, 1576:30 – 1577:9).  It was put to him and he further accepted that, if, on his analysis, a valuer should wish to arrive at the ground rent from the real world evidence of the capital value of land (at $593 million), he would need a wholly different decap rate of 0.76% in order to arrive at his assessed ground rent of $4.56 million.  He agreed that, “there is no evidence in the real world available to enable valuers to find a figure like 0.76%”.

195.  Mr Hatchwell put up a convincing argument in his RAD Tab 31, para. 11.  He said, “in any event, the structure of Professor Hughes’ own valuation in GH-16 shows that the HT ends up with the developed site.  Having bought out the HL, it is his to do with as he likes.  In other words he ends up with the real option, having paid the HL a price in terms of very low rents and a correspondingly low capital sums based on those rents…  But if it be supposed that at the end of the lease the HT pays a capital sum large enough to reflect the rights over the land which the HL is relinquishing, then the rental value of the land ought to be based on that sum, just as in Updated App. GH-16 it is the decaptalised cost of buying out the landlord.  And if it can be based on that sum, why should the initial rent not be the decapitalised $593 million which is agreed would have been paid for the Government Lease at the valuation date?”

196.  We agree with the Commissioner that the acceptance of Professor Hughes’ argument would have the effect of destroying the use of capital values as a basis for valuation of land.  Therefore, we do not agree with the hypothesis of Professor Hughes that we cannot continue to use the property yield as a link between the capital value and the rental value.  This is because, as Mr Hatchwell has put it in RAD Tab 31, para. 4, since capital values reflect option values whilst the rental values do not, “valuers do not have to concern themselves with adjusting the values and costs in Stage 1 and Stage 3 of the Contractor’s Basis in order to eliminate option value.  It is a process which is done by the application of a property yield as the decapitalisation rate in Stage 4”.  Hence, we agree with the Commissioner that the Appellant fails to demonstrate why the auction price could not be used as the starting point because of the existence of the so-called “real option value” within that price.

Were the developers’ auction bids motivated by other factors?

197.  Mr Charman’s second reason (para. 15.16 of JC Report 1) that the developers’ auction bids were based on factors that would have been in their minds but could not have been in the minds of the HT originated from Mr Lee’s First Report.  Mr Lee referred to this as the motivation of the “Chinese Developers” bidding up in the land auctions in order to lever up property prices elsewhere in Hong Kong, which would hence bring beneficial effects on other sites owned by them, whether kept in their land bank or under development.  The Appellant submitted that the desire to influence the value of other land holdings did not therefore reflect the value of occupation of the Appeal Tenement.  However, Mr Lee himself did accept that the said motivation, even if assumed to be in existence and had the effect of raising the land values elsewhere considerably, did not mean that the bids by these developers were not evidence of open market value.  Mr Charman agreed with Mr Lee on this point too.  Also, although Mr Charman drew a distinction between the present case and the facts of Robinson Bros (Brewers) Ltd v Houghton and Chester-Le-Street Assessment Committee [1938] AC 321case, he failed to identify any rating principle that would justify the exclusion of a “special bid” of a purchaser if it was accepted as to be part of the open market value.  We therefore agree with the Commissioner that even if the argument of levering up of prices by the “Chinese Developers” as claimed by Mr Lee existed, the Appeal Tenement would also increase in value in the process.  But since Mr Lee already agreed that the bid for the Appeal Tenement had to be considered as evidence of open market value, the Appellant failed to justify why we could not use the auction price as the starting point in finding the open market value of the land, or the effective capital value of the land in the CB valuation.

Is the auction price not supported by Mr Tang’s R & E valuation?

198.  Mr Charman’s third reason (para. 15.17 of JC Report 1) that the auction price should not be used was because that price was not explicable by Mr Tang’s R & E Method since the latter’s assumptions as to (a) the occupational rental value for the completed building; and (b) the rate of rental growth rate (assumed by Mr Tang to be 9.5%) were too high.  Mr Charman said that if Mr Tang had used a lower estimate of occupational rent and a lower rental growth rate, the resulting ground rent would be lower.  Mr Charman clarified that although he had made the point that a growth rate of 9.5% could not be justified by the auction price, he agreed in cross examination that he had not himself carried out any exercise to substantiate this view.  As to the compatibility of the results of Mr Tang’s CB valuation with his R & E valuation, since it was similar to the “Morton’s Fork” argument put up by Professor Hughes, we will consider it below when that argument was analyzed.

“Morton’s Fork” argument raised by Professor Hughes

199.  Professor Hughes submitted in his Speaking Notes that the Commissioner’s experts “have consistently failed to provide any method of assessing the value of the Best Origin site – or any other site – as a cleared site once the auction price is no longer relevant”.  Also, he suggested that, “the difficulty of assessing land value is so serious that reliance upon the Contractor’s Basis simply can not be sustained as a general method of valuing development and redevelopment sites”.

200.  Further, Professor Hughes said in his Speaking Notes that the evidence of an auction price “is only relevant if it can be demonstrated that the auction price provides a reliable guide to what a hypothetical tenant would have been willing to pay for the tenement under the assumptions of the Rating Hypothesis.”  He concluded that after reviewing the Commissioner’s valuation, the contrary was true.  Using the original discount rate of 18% adopted by Mr Tang, the Commissioner’s R & E valuation show that the value of the site had fallen to $279 million (based on Mr Brooke’s valuation at NB74), which was only 47% of the agreed adjusted value of the site of $593 million at 1 July 1996.  Professor Hughes put forward 2 possible reasons to explain the discrepancy: (a) the special bids of the developers that might have driven up the auction price and (b) the successful bidder at a Government auction acquired a set of property rights that were substantially more valuable than that obtained by the HT under the Rating Hypothesis.  Therefore, Professor Hughes submitted that the auction price could not be used as a starting point in the CB valuation.  We have already dealt with these two points before.  In this section, we will consider the issue of “Morton’s Fork” that was raised by Professor Hughes throughout his reports.

201.  Professor Hughes raised this issue of “Morton’s Fork” in his First Report dated 11 May 2005 (GH page 44).  “Morton’s Fork” refers to a choice between two equally unpleasant alternatives, often used by bridge players.  Professor Hughes compared Mr Tang’s valuation in his First Report, i.e., his valuation on the CB and his check valuation using the R & E Method.  Professor Hughes said that the Commissioner’s CB valuation had started from the assertion that the value of the development site at the relevant date was the adjusted auction price.  However, Mr Tang’s own cash flow analysis show that the present value of the landlord’s share was about $343 million, obtained by applying a discount rate that is equal to the standard cost of capital used in the cash flow model (PKT page 47).  Professor Hughes therefore alleged, “It follows that even though the rental values generated by Mr Tang’s two valuations are very close, the explicit or implied capital values for the development site are far apart.  Clearly, there is some kind of problem of consistency between two valuation methods.”

202.  In replying to this allegation of inconsistency and “Morton’s Fork”, the Commissioner submitted that, “Professor Hughes has exploited the way in which both parties have operated their R & E models”, and that the “Morton’s Fork” argument was wholly fallacious for the following reasons:

(1)	The NPV of the cash flows in Mr Tang’s R & E model is simply part of a calculation “which enables the HL’s cash flows in PKT App. 22 to be redistributed as rent in PKT App. 24 using the HT’s discount rate and the projected rate of rental growth.  It has not been put forward as an estimate of the market value of the HL’s interest in the land”;
(2)	“The R & E method is residual in nature…The Commissioner has always made his case clear, namely that the use of the 18% WACC is conservative, hence the size of the HT’s share in relative terms and the effect of discounting are both increased, thereby leading to a smaller NPV for the HL.  If a less conservative rate of return for the HT and discount rate were to be applied (based upon the IRR of the development project for the appeal site), the NPV of the HL’s cash flows rises to $580.91 m” (see RAD Tab 4 revised PKT App. 31); and
(3)	“If it is desired to compare our NPV figure for the HL’s share of the cashflows in the artificial construct of an R & E valuation with a real world open market value of the land, it would be nonsensical to do so using a NPV calculated with a conservatively high discount rate.  A comparison of that nature is bound to be false.”

203.  The Commissioner submitted that another expert, Mr Hatchwell, had given in his Third Report dated 4 April 2006 (LH Rep. 3, pages 152-155) a response to Professor Hughes’ criticism to Mr Tang’s inconsistency in valuation.  The Commissioner said that there was no answer to that response.

204.  In addition, the Commissioner submitted that Professor Baum had also set out calculations in his First Report arriving at rates of return first for the completed property and then for the land, without using the auction price at all, whether directly or indirectly (see RAD 38 pages 13-15 and 18-20).  And, although Professor Baum had also refuted Professor Hughes’ Report 2 in his Third Report (AB Rep. 3, paras. 6.1 to 6.2), he had not been challenged on that point too.

205.  We agree with the Commissioner that the apparent difference in the HL’s interest in the R & E model of Mr Tang’s First Report (when using a discount rate of 18%) and the auction price of the land does not necessarily lead to the conclusion that the CB valuation is fallacious.  On the contrary, in our view, it alerts us on the volatility of the R & E valuation bearing in mind that, as we have noticed, the large number of variables that have to be adopted and the long time span in the R & E model.  As we have seen in these income models, discounting rents and costs over a long period of time (over 40 years as in Mr Tang’s model) could easily lead to varying results depending on what variables as to the rents, the costs and the discounting rates that one chooses to use.  However, since we are considering the development process for a development site which lasts for a long period of time and for reasons we have given in the earlier part of this Judgment, there is no reason not to consider all the revenue and the cost figures for the whole period of the life of the building in order to fully reflect the development potential of the land in these income models.

Conclusion in the use of the adjusted auction price in the CB valuation

206.  To conclude, we do not find that the Appellant had provided sufficient justification to reject the use of the CB valuation in principle, and the use of the adjusted auction price in the present case.  For this reason, we accept the adjusted figure of $593 million (agreed between Mr Lee for the Appellant and Mr Brookes for the Commissioner) as the open market value of the subject site at the relevant valuation date of 1 July 1996.  We further accept the adjustment made by Mr Tang, to reflect that in the hypothetical letting, the Government rent is to be borne by the HL, i.e., to gross up the adjusted auction price to $611 million as to be the effective capital value of the site for the purpose of the CB valuation.  So, the next question before us is whether the decap rate adopted by Mr Tang is correct.

Decap rate for the CB valuation

207.  Mr Tang explained in his First Report that as there was no direct evidence of decap rate for development sites such as the Appeal Tenement, he had adopted the property market yield for the office building proposed to be completed on the subject site as the starting point for deriving the decap rate.  Based on the estimated capital value of the space in the office building to be developed on the subject site and the estimated initial market rental value of the same space, he estimated this to be 5.5%.  However, Mr Tang admitted in his First valuation report that an adjustment would be required to reflect the differences between the property yield that is appropriate for the Appeal Tenement as a bare development site and that applicable to an office building to be erected on the site.  He said that he had “made a downward adjustment of 20% to reflect (a) that the Tenement comprises land only and is therefore more secure than a building if held as an investment; (b) that the Tenement as a bare site is free of such outgoings as repairs/maintenance and insurance costs; and (c) that the Tenement is not a depreciating asset as compared with a completed building.”  Applying the said downward adjustment of 20%, he arrived at the decap rate of 4.4% for use in his CB valuation.

208.  Mr Tang has reviewed the appropriateness of the said 20% downward adjustment in his Second Supplementary Report.  Mr Tang disagreed with Professor Hughes’ proposition that the ground lease was equivalent to a Government index-linked bond and that the HL’s tenement was risk-free.  Mr Tang concluded that his adoption of 4.4% implied “conservatism” and that the CB valuation was indeed “a very robust approach”.  The other experts for the Commissioner, Mr Hatchwell and Professor Baum have also separately commented and supported Mr Tang in his derivation of the decapitalisation rate.

209.  Mr Tang has, in his Third Supplementary Report, after reviewing the 20% downward adjustment in the decap rate previously adopted by him and Mr Brooke’s estimates of the market yields maintained the same view as before.  Mr. Brooke’s estimate of the market yields were derived from (i) estimating the open market rental value and capital value of the completed building upon the Appeal Tenement as at the valuation date; and (ii) estimating the yield associated with the development of the Appeal Tenement from the viewpoint of a property developer.  Mr Tang said his conclusion was supported by Mr Hatchwell who regarded his CB valuation as conservative.

210.  On the other hand, the Appellant submitted that the Commissioner had failed to substantiate the use of any appropriate decap rate for the CB valuation.  In particular, the Appellant severely criticized Mr Tang because firstly, there was no justification for his use of the property yield of 5.5% as the base rate; and secondly, there was no basis for his downward adjustment of 20% resulting in the use of the decap rate of 4.4% in the CB valuation.

211.  Mr Charman for the Appellant continued to use in his version of CB valuation (in ADD Tab 10) a decap rate of 3.8% that was said to be based on Professor Hughes, who remarked in his Speaking Notes dated 31 October 2006 that his original estimate of a decap rate of 3.8% was extremely conservative.  In addition, Professor Hughes said, “the decapitalisation rate should certainly be no more than the rate of 3.55% associated with the real internal rate of 4.8% generated by the gross rental yield on office properties in mid-1996”.

The use of property yields or borrowing rates as the basis for decap rate

212.  In his First Report, Mr Tang said he had followed the approach of Mobil Oil Hong Kong Ltd. v Commissioner of Rating and Valuation [1993] 2 HKDCLR 77(“Mobil”) case in the choice of the property yield instead of the bank interest rates in deriving the decapitalisation rate in the CB valuation.  In the Mobil case, the Commissioner of Rating and Valuation submitted that the decapitalisation rate should be based on property yields instead of the bank interest rates.  Judge Cruden decided in favour of the Commissioner and his reasoning was as follows:

“

… We are satisfied that property market yields are a more accurate and reliable basis, than bank and other financial market rates.  We are also satisfied that an important advantage of property market yields is that they will normally already include, at least some of the factors, for which a Cardiff deduction is made.” (page 102)

213.  Mr Tang adopted Mr Brooke’s gross rental yield of 5.5% as the starting point.  This was conceived as a market yield since it was calculated as the ratio of the estimated market rental value of the building developed on the subject site at full occupancy divided by the estimated Gross Development Value of the building, both as at the valuation date. 

214.  The Commissioner submitted that apart from Mr Tang, another local valuer/expert retained by the Commissioner, Mr Brooke also considered it reasonable to adopt the derived market property yield of 5.5% as the starting base rate for the valuation purpose.  In addition, the Commissioner said that a third local valuer/expert, Mr Lynch, when cross-examined also agreed to the use of the property yield as the basis for computing the decapitalisation rate.

The President’s questions to Mr Tang and the parties’ responses

215.  During the hearing, the President asked Mr Tang a number of questions on the composition of the 20% adjustment which he had applied to the property market yield in arriving at the decap rate of 4.4% in his CB valuation.  More specifically, Mr Tang was asked whether calculations could be done with reference to the figures in the R & E valuation.  Mr Tang submitted a Note dated 1 December 2006 as amended on 3 December 2006.  This was followed by a Joint Response dated 2 January 2007 prepared by Mr Tang and Mr Brooke.  In response to these, the Appellant submitted a Response by Professor Hughes dated 11 December 2006 and another Response, also by Professor Hughes dated 9 February 2007.  And in response to the two Responses of Professor Hughes, Mr Tang and Mr Brooke jointly submitted another Response for the Commissioner on 12 March 2007, before the hearing for the final closing submission of the parties commenced. 

216.  Since the parties wished the hearing of this case to be completed as scheduled, they have agreed not to call the experts to give further evidence (in chief and for cross-examination).  Instead, they asked the Tribunal to decide on the basis of the written Responses submitted by the experts.  Since the Tribunal finds that the CB valuation is the appropriate approach to be adopted in the present exercise, we will consider the various issues arising in the CB valuation below, including the adjustments propounded by the experts for the parties in their original reports and the subsequent written “Responses”.

Submission from the Appellant on Mr Tang’s 20% adjustment

217.  In the written Supplementary Submission dated 19 March 2007, counsel for the Appellant set out the Appellant’s positions in the light of various documents produced by both parties in relation to the topic of Mr Tang’s 20% adjustment to the property market yield as derived by Mr Brooke.  The Appellant submitted the following points:

(a)	Although Mr Tang said that he had followed the Lands Tribunal’s decision in Mobil case, he had not actually followed Mobil case. In Mobil, the property market yield was derived from the open market whilst Mr Tang’s approach was not;
(b)	The Appellant did not accept Mr Tang’s starting point that Mr Brooke’s 5.5% was “an open market yield” as it was not a yield based on some directly observed evidence but was based on valuer’s estimate;
(c)	When cross-examined, Mr Tang’s only explanation of his 20% adjustment was “valuer’s judgment”.  But the Appellant said that even “if the areas of disputes were resolved by the Tribunal wholly in Mr Tang’s favour, that would not alter the fact that there is no open market evidence in order to verify the calculations”.  However, the need for such verification was well illustrated by the Mobil case itself where “it became necessary to make a very large adjustment of 27% to the property market yield in order to derive a yield which, when applied to the effective capital value (i.e. the outcome of stages 1 – 3), would produce a figure in line with the rental evidence”;
(d)	Similarly, for the Scottish cases cited in the Mobil case, there was, in respect of the comparable premises, the evidence of the ratio of cost to rent, which was the reason that caused the Lands Tribunal to reduce the yield derived from other sources;
(e)	Therefore, even if the Tribunal could calculate the adjustment percentage for converting a yield for a completed office building into one for development land, there was “no confidence at all that such a figure will be even close to the open market yield if one were available”.  For this reason, the Appellant maintained that there was no justification for attaching any weight to the outcome of the adjustment exercise;
(f)	The Appellant also pointed out that Mr Tang’s valuation was “valued based” but not “cost based” even though Mr Tang claimed to have been conducting his valuation on the CB.

Professor Hughes’ adjustments to the market rental yield

218.  Professor Hughes set out in his Response dated 9 February 2007 (AAD Tab 34) that Mr Tang had either omitted to make adjustments for many factors (for example, voids, letting and marketing expenses, building depreciation) or incorrectly estimated other factors (default risks and in particular, rental depreciation).  He sought to make his own estimates for some of the factors.

219.  Professor Hughes summed up in para. 50 of his Response dated 9 February 2007 (AAD Tab 34) the following estimate of the rental yield for the land in the present case,

“

Even without making an adjustment to reflect a lower risk premium for an investment in land, the adjusted rental yield for land would be either (a) 36.9% (58.1% - 3% - 18.2%) of gross rental income at full occupancy for the lower rate of 1% per year rental depreciation, or (b) 18.7% (39.9% - 3% - 18.2%) of gross rental income at full occupancy for the higher rate of 2% per year rental depreciation.”

220.  However, Professor Hughes made it absolutely clear in the Conclusion section of the Response (AAD Tab 34) that in his opinion, although it was possible to construct a set of adjustments used to “translate a market rental yield on office property into an approximate estimate of a market rental yield on land”, it was a completely futile exercise for the following reasons:

(1)	The magnitude and the potential margin of error of the cumulative adjustments were large in relation to the final figure;”
(2)	Even if it were possible to arrive at a reliable estimate of the market rental yield for a development site, it could not be applied in the present exercise as there was “neither an estimate of a market rent for the tenement nor an estimate of the market value of the rental stream”; and
(3)	Most importantly, he did not believe that Mr Tang’s procedure of attempting to adjust a market yield could be regarded “as providing a reliable basis for applying the Contractor’s Basis method of valuation in this case.”

The relevance of the Mobil case and the Scottish cases

221.  The Commissioner submitted that it was clear that in the present case, nobody had put forward any evidence to derive a decap rate from the evidence of borrowing costs.  Mr Charman confirmed that he had relied on Professor Hughes for the estimate of a decap rate and Professor Hughes had used property yields in order to derive a decap rate (see his GH-03 and GH-11).  Although Professor Hughes equated the return on index-linked bonds as the decap rate, it only represented the rate at which the Government borrowed but could not be equated to the borrowing costs of the HT.  On the other hand, all the witnesses for the Commissioner supported the use of the property yield as the basis for deriving the decap rate.

222.  The Commissioner also reminded us that (1) the LT in Scotland preferred the use of the property market yields to the borrowing rates (see Shell UK Exploration and Production v Fife Region Assessor [1989] RA 158 (LT) (“Shell”) and BP Petroleum Development Ltd v Lothian Region Assessor at [1988] RA 176 (LT) (“BP”)); (2) the LT took the view that it was bound by past decisions of the LVAC which had not accepted the borrowing approach as the sole approach and emphasized on the need to consider the valuation from the view point of both the HT and the HL (see BP at p. 176); and (3) in any event, a property premium would have to be added to a risk free rate derived from an index linked stock, to cover depreciation, illiquidity and other property risks.  In BP, the LT assessed a property premium (net of growth) of 3.5%, to be added to a Risk Free Rate of 3.5% (see BP at [1988] at 175-6).

223.  The Commissioner further show that in the earlier Scottish cases (the Distillers (Bottling Services) Ltd v Fife Regional Assessor [1987] RA 209, Fife Regional Assessor v Distillers (Bottling Services) Ltd [1989] RA 71 (LVAC), BP and Shell cases), cross checks were employed in order to test a standard 8% decap rate used by assessors for all commercial and industrial properties.  That is, the cross checks were carried out involving the calculation of the ratios between the rateable values (as proxies for rents) and the capital values of other properties.  These were considered necessary because the assessed premises and the “comparables” were not the same type of property but were properties with widely different characteristics even though they were broadly considered to be specialized properties.

224.  Similarly, in the Mobil case, the Lands Tribunal used flatted factory yields as the basis for valuing the oil tank farm, a very different type of property.  Therefore, it was necessary for adjustment to be made between the different types of properties.

225.  On the contrary, the Commissioner submitted that, in the present case, the Appeal Tenement, a development site, was only compared to the office development that was to be completed on the site.  There would be no advantage in making a comparison with other types of properties and, in fact, no evidence had been led to that effect too.

Professor Hughes’ adjustment for rental depreciation

226.  In RAD Tab 46, Mr Tang and Mr Brooke jointly alleged that Professor Hughes had failed to support his assertion that Mr Tang had underestimated the percentage adjustments required to allow for the rental depreciation.  They said that Professor Hughes suggested alternative to Mr Tang’s approach was in fact based on his assumed uniform depreciation rate of 1% to 2% per year throughout the life of the building to be completed on the site.  This led Professor Hughes to estimate much higher rental depreciation rates of between 18.2% and 36.4%, as compared to the Commissioner’s 11%.

227.  However, according to Mr Tang and Mr Brooke, the assumed uniform 1% yearly depreciation rate was based on Professor Baum’s rough assessment that was not, the subject of any detailed calculations.  Moreover, they pointed out that Professor Hughes’ alternative assumption of a 2% yearly depreciation rate was based upon the analysis he carried out in GH-14.  That analysis was flawed as it was based on wholly inadequate data of relying on the regress analysis of 7 buildings only. 

Adjustments for factors such as void period, letting and marketing expenses

228.  In the Response made on 2 January 2007, Mr Tang and Mr Brooke stated that there was no need to take account of factors that were common to both the office building and the land.  These included the void period, letting and market expenses.  The purpose of estimating the quantum of adjustments was to reflect the differing nature of the two property assets.  Therefore, the attempt by Mr Tang was only to make an allowance for rental depreciation, and for those repairs, maintenance and vacancy expenses which related specifically to the upkeep, maintenance, management and operation of a completed development that were not applicable to land.  Similarly, they explained that in Hong Kong, as both the land (which was held on Government lease) and an office building were “wasting assets”, there was no need to make allowance for the duration of the lease of the land over which the office building stood.

Points beyond the scope of the question raised by the President

229.  The Commissioner submitted that Professor Hughes had attempted to bring in various points that were beyond the question of the President.  These included: (i) the issue of the gross rental yield and net investment return, i.e. the derivation of a net investment return to the landlord out of the gross rent he received from the tenant, whether in respect of the development land or the completed office building and the relationship of that to the yield that should be applied to the development land; (ii) whether the CB valuation undertaken by Mr Tang was “cost-based” or “value-based”; (iii) an attempt to re-open the “real option value” point; (iv) the depreciation of land as an asset; (v) depreciation of building/physical depreciation; (vi) deduction of Government rent and (vii) additional risk premium.

230.  Mr Tang and Mr Brooke opined that the additional adjustments suggested by Professor Hughes were all irrelevant for the purpose of adjusting the yield from the market yield for the office building (which was based on the ratio of the estimated market rent and the estimated market price of the building) to arrive at the yield for the development site.

The parties’ adjustments made to the basic open market yield

231.  Mr Tang and Mr Brooke helpfully summed up in their note (RAD Tab 46) the adjustments made by the Appellant and the Commissioner to the basic “open market yield” derived from the office space in the building completed on the site.  This is shown in the following table:

	Appellant	Commissioner
Rental depreciation	18.2 % - 36.4%	11%
Repairs/maintenance/vacancy expenses	5.1%	6%
Smaller default risk	No estimated figure	3%
Total adjustment	No estimated figure	20%

Tribunal’s decision on the approach adopted by Mr Tang

232.  The Appellant did not recognize the 5.5% derived by Mr Brooke as an open market yield.  This is something we do not understand.  Both parties agreed, after much negotiation, the figures of the capital and rental values of the office space in the office building to be completed on the subject site.  Presumably, these values are based on the comparable sales and lettings using direct comparative method.  As such, we fail to understand why the ratio of the capital value and the rental value agreed for the office space cannot be taken as the estimated open market yield (albeit indirectly) for the office space in the building on the site.

233.  We agree with the Commissioner that in the present case there is no evidence that the decap rate should be based on anything but market property yield.  It is true that there is no actual market property yield for the subject site, which is a piece of development land.  But we cannot agree with the Appellant who said that even when if one could calculate the adjustment percentage for converting a yield for a completed office building into one for development land, there is “no confidence at all that such a figure will be even close to the open market yield if one were available.”  This proposition is against the fundamental precepts for the CB valuation.  In the CB valuation, the adjusted property yield will form one of the 2 inputs that lead to the final rateable value figure in this valuation exercise.

234.  The Appellant challenged Mr Tang’s adjustment on the basis that when he was cross-examined, he answered that the 20% adjustment was his “valuer’s judgment”.  We do not find this to be a sufficient reason for rejecting Mr Tang’s valuation in its entirety.  A valuer’s judgment is, we find, part and parcel of every day valuation by valuers in Hong Kong and elsewhere.  If a valuer’s judgment per se is not allowed, it seems to us that the Appellant is advocating that every step of a valuation must be carried out with mathematical precision. But we all appreciate that valuation is an art instead of a science.

235.  Having considered the various adjustments to the yield of the office building as proposed by Professor Hughes and Mr Tang, we are satisfied that Mr Tang’s adjustments are more reasonable and appropriate.  As such, we accept his conclusion as to the derivative of the property yield for the Appeal Tenement.

Stage 5 and Stage 6 review in CB valuation

236.  The Appellant submitted that the Rating Forum’s Guidance Notes recommended that valuers would have to perform a review of their valuations and reflected the “haggling” of the market.  Therefore, the Appellant criticized Mr Tang that after carrying out the fifth stage of the CB valuation, he had concluded that no further adjustment was required.  When asked about this in the cross examination, Mr Tang said that he had looked at the result of the R & E valuation and had come to the conclusion that his R & E valuation as a check valuation did support his primary valuation by CB.  However, the Appellant pointed out that it must be clear to Mr Tang that after the first round of reports, his R & E valuation in fact did not support his CB valuation.  Also, the Appellant submitted that Mr Charman had reviewed the outcome of Mr Tang’s valuation and opined that, a developer would not enter into a deal where he had to pay about 40% of the rack rent to the HL in return for a substantial loss for the first 11 years and would not break even until the twelfth year.

Tribunal’s decisions on the Commissioner’s primary valuation, the CB valuation

237.  We have earlier decided that Mr Charman’s valuation, which was purported to be a valuation using the CB, should be rejected on the grounds of its unrealistic assumption as to the 6-year fixed duration of the lease and the faulty methodology.

238.  We agree with the Commissioner that the Appellant failed to justify its rejection of the CB as the primary valuation of the Appeal Tenement on the grounds that (1) the auction price did not provide the basis of assessing the effective capital value of land in the CB valuation; and (2) the adoption of the decap rate by Mr Tang was erroneous.

239.  Mr Lee and Mr Charman both suggested that the developers might have different motivation when bidding for the site.  There was no evidence to support their assertions.  Therefore, the Appellant’s specific attacks on the relevance of the auction price of the subject site have not been substantiated.

240.  Professor Hughes opined that the auction price of the subject site, generally accepted as the best indicator of the market value of the site, could not be used as it contained a substantial element of real option value.  However, even though Professor Hughes was challenged in the hearing, he failed to exclude or quantify the so-called real option value.  As it is trite knowledge in valuation of a property that yield always expresses the relationship between capital value and rental value, we fail to understand why the existence of the real option value in the value of a development site will disable the yield to provide the link between the capital value and the rental value of the site.

241.  We therefore disagree with the Appellant that in principle, the estimated market value of the site (which in this case was originated from the auction price) after adjusting for the time could not be used as the effective capital value of the land in the CB valuation.

242.  Regarding the decap rate, the Appellant contended that Mr Tang’s adjustments were wholly incorrect.  However, we are satisfied that by and large, Mr Tang succeeded to have reflected the most important differences between the property yield of a development site and that of a completed office building.  Therefore, we are satisfied that for the purpose of the CB of valuation of the Appeal Tenement, we could use Mr Tang’s adjusted decap rate.

243.  To conclude, we accept Mr Tang’s primary CB valuation that the rateable value of the Appeal Tenement as at the relevant date should be in the order of $26,880,000.  We will consider below that after finding the rateable value by the primary valuation, how and to what extent should this be checked by other methods of valuation?

The Appellant’s valuation using the R & E Method

244.  The Appellant did not submit any valuation carried out on a conventional R & E Method even though the experts for the Appellant have commented on the R & E valuation carried out by Mr Tang.  However, given the Appellant’s criticism of Mr Tang’s various assumptions, in particular the assumption as to the duration of the hypothetical tenancy for the Appeal Tenement used in his R & E Method of valuation, the Appellant’s stance in not carrying out a similar R & E valuation is understandable.  For convenience of discussion, the Appellant’s comments on Mr Tang’s valuation will be addressed when we consider his valuation in details below.

Professor Hughes’s PDM (latest version at the Updated App. GH-16)

245.  Unlike Mr Tang who used cash flow analysis merely to give additional support to his primary CB valuation, Professor Hughes, the principal expert of the Appellant, relied exclusively on his PDM in assessing the rent, i.e. the rateable value for the Appeal Tenement.  Professor Hughes first set out his PDM in App. GH-06 of his report and described that in section 8 of that report.  The essential features included a cash flow with a terminal calculation.  Professor Hughes’ analysis depended upon: (i) an estimation of the HT’s return at 25.8% in real terms, or 34% in nominal terms; (ii) a presumption that the residual ‘development profit’ left after the deduction of the HT’s required return would be further divided between HL and HT (assumed by Professor Hughes to be in equal shares); and (iii) at the end of the 5 years, the HT would acquire the HL’s interest and sell the merged interest.

246.  Mr Hatchwell commented that Professor Hughes’ analysis in GH-06 was dependant upon his conclusions in GH-03 as to the HL’s return, and GH-04 as to the HT’s return, which were all fundamentally wrong, for reasons stated in LH, pages 134 to 152.  Subsequently, Professor Hughes replaced his App. GH-06 by his App. GH-16, which was later revised as Updated App. GH-16 (Bundle GH at pages 140-1 to 140-5).

247.  It is clear from the Updated App. GH-16 that the PDM assumed a 5-year duration for analysis.  It first estimated, on a quarterly interval, the rent revenue, construction costs and operating expenses.  After allowing for the HT’s required return, it estimated the development profit before rent at column (13) (summation of (i) operating cash flow excluding rent at column (7) and (ii) the HT’s required return at column (12)).  It assumed at the beginning of year 6 that the HT would sell the building when the net value of the building was calculated as the difference between (i) the value of the building and land and (ii) the value of the land (i.e. $1,286.86 million - $164.52 million = $1,122.34 million).  The value of the building and land was taken to be the full rental value of the building at the beginning of year 6 divided by the estimated yield of 7% (i.e. $22.52 million per quarter x 4 / 0.07 = $1,286.86 million).  And the value of the land was taken to be the value of ground rent at the beginning of year 6 divided by the estimated decap rate of 3.8% (i.e. $1.56 million per quarter x 4 / 0.038 = $164.21 million.  The actual figure shown in the Updated App. GH-16 arrived at by using the “Goal Seek” function of the “Excel” program was $164.52 million).  In column (13), the summation of the negative cash flows in year 1 to year 5 and the proceeds from the net value of the building together gave the net present value of $28.08 million for sharing between the HT and the HL.  Professor Hughes suggested that this was split equally between the HT and HL, giving an initial ground rent of $1.14 million per quarter, or $4.56 million per year.

248.  The Appellant suggested that Professor Hughes had devised this PDM “as a perfectly clear and reasonable combination of the R & E and DCF models designed to deal with a very unusual valuation problem”, a problem that had not arisen in any rating valuation in the past.  Also, the Appellant entirely accepted that Professor Hughes’s PDM was not a standard method of either R& E or DCF Method but it sought “to address the consequences of adopting conventional valuations models which imply that the hypothetical tenant would agree to pay a large ground rent from the outset…”

249.  The Appellant added that the reason for combining the 2 approaches in a single model such as the PDM as devised by Professor Hughes was to address the problem of the need to apply the hypothetical tenancy to a property development project.  A normal R & E valuation had only a one-year perspective: the tenant got a return on his capital and a return of its capital immediately in that year, not on a deferred basis.  However, the Appellant submitted that in the present case, the Commissioner had moved away from a one-year perspective to the duration of the hypothetical tenancy for 43 years whereupon the return of capital and return on capital was spread over the said 43 years.  In contrast to a normal R & E valuation, a DCF model would work only if the analysis period was fixed at the outset, whether at 5 years or 43 years. 

250.  Hence, the Appellant submitted in the Written Closing Submission that, “the PDM introduces the central element of the R & E approach into a cash flow framework by allowing the HT a credit in respect of his return on capital which is deducted from the development profit that has to be divided between the parties.  This has the effect of bringing forward in time its return on capital.  The return of capital is accelerated by adopting the cash flow measures of the tenant’s investment, which reduces the level of the tenant’s return as capital is recovered earlier.”

251.  The Appellant remarked that the real issue was not the structure of the valuation model but the level of discount rate that was used, which lied at the heart of the different valuations proposed by the Appellant and the Commissioner.

Criticism of Professor Hughes’ PDM

252.  The Commissioner reminded us that, during the hearing on Day 16, Professor Hughes had accepted the following:

(1)	A negative NPV result did not mean that the DCF Method was unreliable, rather it was consequence of the inputs;
(2)	His criticism of the negative results produced by running the R & E and the DCF Methods over 43 years would also apply to his PDM when run over a 5-year period (for example, it produced a “degenerate result” with the same inputs in columns 7 and 8 of his Updated App. GH-16);
(3)	The PDM could be “binned” on the same basis of producing negative NPV results, if run over a period exceeding only 7 years;
(4)	The R & E method was not to be binned.

253.  Therefore, the Commissioner submitted that by the above admissions, the Appellant had failed to substantiate that there is any need for an alternative to Mr Tang’s R & E Method of valuation.  The Commissioner questioned that if it was true that the PDM was so close to the R & E Method, why was it ever necessary to introduce the former to replace the latter?  And, since Professor Hughes agreed that the soundness of Mr Tang’s valuation was only affected by the quality of the inputs used, it was not understood why Professor had to introduce his PDM, which according to the Commissioner, did not “mimic the R & E Method” (as claimed by Professor Hughes) but “turned out to be a hybrid or cross-bred between the R & E and the DCF methods.”

254.  Comparing the PDM with a standard DCF Method, the Commissioner pointed out that Professor Hughes had agreed (Day 17, 1422:36 – 42) that in a standard DCF, the depreciation charges would not have been deducted (for example, as in column H of PKT App 22), but the construction costs would instead have been put in as negative cash flows.  Also, a DCF analysis (as in the Updated App. GH-15) did not involve deducting in a separate column the HT’s share.  Instead, the HT’s share was allowed for simply through the process of discounting.

255.  In addition, the Commissioner submitted that although it was an essential assumption of the PDM that the tenancy will end after 5 years when the HL bought out the HT’s interest, Professor Hughes failed to justify his 5-year period of analysis.  At the date of termination of the tenancy, it will also be necessary under the PDM to estimate the terminal value of the completed development.  The accuracy of that estimation would in turn affect the accuracy of the PDM.  Again, the Commissioner said that Professor Hughes had failed to justify his choice of the yield for calculating the terminal value of the completed development (which was adopted by him at 7% whilst the yield on a grade B building using the RVD indices would only be 5.4% at the relevant date).  Also, the Commissioner disagreed with Professor Hughes that (i) a separate asset beta of zero should be ascribed to the HL and (ii) the occupational rents would experience zero % real growth over the assumed 5-year duration.

256.  The Commissioner contended that the PDM had used a 34% nominal rate of return and discount rate (equivalent to a real rate of 25.8%) whilst the Commissioner had used a rate derived from the real world WACC of property companies of 18% in nominal term or a project risk rate for the development of the Appeal Tenement at 14.12 % in nominal term.  This is a very major issue between the parties.  In short, Professor Hughes believed that the real world WACC of 18% for a property company estimated by Professor Chan (which figure he accepted) should be decomposed into separate rates of return for the HL and the HT.  However, the Commissioner submitted that the Appellant “has not pointed to any authority for splitting a real world rate of return between the two hypothetical parties, or indeed, criteria for determining how that should be done.”  Therefore, the Commissioner said that the rate of return for the HT should follow the usual practice of applying the rate of return appropriate to the real world occupier or user of the tenement, or alternatively companies operating the same type of business.

257.  In summary, the Commissioner pointed out that the following basic propositions of Professor Hughes were wrong:

(1)	that the ground lease is equivalent to a Government index-linked bond and that the HL’s tenement is risk free since the HL has the first claim on the HT’s cash flow;
(2)	treating development land as an alternative for investment with index-linked bonds as there is no guarantee for the annual rent (on the contrary, the value of development land is versatile, as explained by Professor Baum in his report);
(3)	the HT does not have the burden of assembling the site, which falls upon the HL; and
(4)	the HL cannot develop the site himself, following the principle approved by Wilmer LJ in Humber v Jones [1960] 6 RRC 1671 at 170,
	“	Rating is not the every day world, it is the domain of a HT who is faced with the fact that, try as he may, he cannot become the owner of the Premises, neither can he get a lease for a term of years, short or long.”

Duration of the tenancy assumed in the PDM

258.  The Commissioner submitted that since Professor Hughes’ analysis at the Up-dated App. GH-16 was made on the assumption that the HT would buy the HL’s reversionary interest so that the two interests would merge on that point of time, that procedure could happen at any time and would not have to occur on the anniversary of the yearly tenancy.  During the hearing, Professor Hughes clarified that the 5-year period of tenancy assumed in his PDM was “a reasonable judgment” made by him taking into account the distribution of risks, the consequence of delay as well as the benefits of progressing faster.

259.  The Commissioner questioned why Professor Hughes did not revise the period of analysis in his valuation even though after the parties had agreed on the reduction of the period of construction from 42 months to 32 months.  In response, Professor Hughes said that an extra risk margin of 10 months should be added so as to arrive at the same period of 5 years.  The Commissioner however submitted that there is no reason why all of a sudden, an extra risk margin of 10 months had to be allowed for.  Instead, the Commissioner suggested that the refusal to revise the valuation was simply because the PDM valuation was sensitive to the period of tenancy assumed, as Mr Hatchwell already demonstrated in the sensitivity analysis at RAD Tab 13 that by simply altering the duration of the letting from 5 years to 4 years (but otherwise retaining all of Professor Hughes’ assumptions in the Updated App. GH-16 as well as the “second” and the “third bites”), the initial rent would rise from $4.56 million to $8.78 million, a rise of over 92%.

260.  In fact, the Commissioner said that if, for instance, the building that was completed on the site were to be sold immediately upon completion, the initial rent payable under the PDM would be higher still.  As a result, the Commissioner queried how, under the circumstances, the HL and HT could agree on the expected duration of the letting.

Yield for estimating the terminal value for the completed development

261.  The Commissioner submitted that although the yield on a Grade B office building using the relevant RVD indices should be 5.4%, Professor Hughes assumed, in valuing the terminal value for the completed development: (i) a yield of 7% and (ii) zero % real growth over the 5-year duration.

262.  The Commissioner also said that “having spent many hours of ploughing through “GH02” and related material,” all that time was wasted because Professor Hughes is really relying on straight forward averaging of RVD indices for yields.  In response, Professor Hughes did accept that the average Grade B yield of 7% was derived from yield of offices in previous years across the whole of the territory.  He further added that he had reduced the average yield of 7.5% to 7% in order to be “conservative”, i.e. to reflect the point that the yield for a relatively new building of 18 months old (i.e. 5 years after the valuation date, after taking into account the period of construction) should be lower than the average yield figure of older buildings that produced the RVD indices.

263.  On the other hand, Mr Brooke put forward a valuation method, which was a well-established market practice of arriving at the yield for an office building.  We have no hesitation to accept Mr Brooke’s valuation and reject Professor Hughes’ method.

Division and proportion of GDV “cake” allocated to HL

264.  Mr Hatchwell stated in his Second Supplementary Report dated 4 April 2006 that although Mr Tang’s valuation at PKT App. 22 and Professor Hughes’ analysis at App. GH-06 were essentially residual valuations, Professor Hughes had failed to see the overall picture that had led him to adopt positions that were totally unrealistic.  In particular, Mr Hatchwell said that Professor Hughes did not identify the slices of the overall “cake” that he generously allowed for the HT.  Therefore, Mr Hatchwell summed up in his Table 5 (LH p. 129-1) the proportion of “cake” allocated to HL, under the various approaches of Mr Tang and Professor Hughes.  Subsequently, Mr Hatchwell, in commenting Professor Hughes’ revised PDM valuation at the Up-dated App. GH-16 (replacing the App. GH-06) on the similar issue of division of GDV “cake” allocated to HL, produced the following table at LH page 27.  Mr Hatchwell’s main point was that Professor Hughes’ calculation was very unrealistic as it only gave the HL a mere 3.04% of a slightly larger “cake” in his revised PDM valuation.

Division and Proportion of GDV “cake” allocated to HL  (LH page 27) (in Millions)
	Gross development value	Return of HT’s capital	Reutrn on HT’s capital	Landlord’s Share
Position of parties at April 2006 Mr Tang at 18% discount rate Mr Tang at 14.75% discount rate Prof Hughes At 34% discount rate	$526.02 100% $789.32 100% $281.74 100%	$47.41 9.01% $65.00 8.20% $159.59 56.64%	$167.62 31.87% $153.85 19.51% $104.02  36.92%	$310.98 59.12% $570.47 71.74% $18.13 6.44%
Current position of parties Mr Tang at 18% discount rate Mr Tang at 14.12% discount rate Prof Hughes At 34% discount rate	$479.6 100% $793.61 100% $299.78 100%	$46.19 9.63% $68.56 8.64% $181.40 60.51%	$158.87 33.13% $144.35 18.19% $109.28 36.45%	$274.54 57.24% $580.70 73.17% $9.10 3.04%

265.  Alternatively, Mr Hatchwell show in his analysis at LH page 208 his so-called the “unreality of GH-16”, in finding the loss in value, which the HL suffered on granting the hypothetical tenancy.  First, Mr Hatchwell said that the value of the HL’s interest before the granting of the tenancy to the HT was $593 million because (i) this was the value of a 50-year Government lease; and (ii) the HL was either the Government or the owner of a very long lease.  Second, the value of the HL’s interest after the letting, on Professor Hughes’ assumptions, would be the sum of (i) the estimated value in the first 5 years of the hypothetical tenancy; and (ii) the estimated value of the HL’s interest in 5 years’ time, as follows:

	(i)	For the 1st 5 years
		Initial rent	$2.96M
		YP of 5 years @3.8%	4.57
		Estimated value		$13.52M
plus	(ii)	Capital value of HL’s interest in 5 years’ time
		Initial rent of $2.96M ÷ 3.8%	$77.89M
		PV in 5 years @3.8%	0.83
		Estimated value		$64.64M
		Total value of HL’s interest (after letting) = sum of (i) and (ii)		$78.17M

266.  Therefore, Mr Hatchwell said that the above calculations show that the difference between the value of the HL’s interest before and after the granting of the tenancy was $514.83 million (i.e. $593 million less $78.17 million).  This was the loss to the HL arising purely from the granting of the hypothetical tenancy that Professor Hughes envisaged would take place.  Mr Hatchwell alleged that no rational landlord would have accepted such a loss.  The only possibility was that the landlord would have received a premium, to the extent of $514.83 million from the tenant, in addition to the rent.  But this assumption is not permissible in rating valuation since the consideration passing between the HT to the HL must only be in the form of the rent.

267.  Mr Hatchwell concluded that his analysis had demonstrated the internal inconsistency within the App. GH-16 and the Up-dated App. GH-16 in that the HL had, on Professor Hughes’ assumptions, given away almost $515 million out of his original $593 million asset by entering into the hypothetical tenancy with the HT.  The Commissioner submitted that the PDM is flawed on this further ground that no real cross checking or “stand back and look” exercise had been undertaken.

“Second bit” and “third bite” of the tenant’s share

268.  More importantly, the Commissioner said that because of the “faulty” structure of the model, the PDM allowed for the “second bite” and “third bite” of the tenant’s share of the rent hence leaving an unreasonably small share for the HL.  The Commissioner even went so far as to allege that Professor Hughes had introduced his PDM in order to disguise the repeated additional “bites” of the rent by the HT.  For this reason, the Commissioner gave lengthy submissions in the Written Closing Submission from para. 4.37 to 4.70 about how Professor Hughes was not forthcoming in respect of the “second” and the “third bites” of the HT’s share in his PDM (App. GH-16 and the Up-dated App. GH-16).  In short, the Commissioner submitted that Professor Hughes was not candid in giving evidence since during the cross-examination, he denied a number of times the double counting of the HT’s share.  The Commissioner alleged that this obviously affected his credibility as an expert witness.

269.  In particular, the Commissioner said that Professor had by Day 17 accepted that there was a gap but he still denied that it was as large as $8 to $10 million.  He claimed that the gap was explained by two causes: (a) the “third bite”; and (b) the “difference between a cash flow analysis and my version of the R & E analysis.”  He suggested that about half, about $4 million was due to cause (a) and the other $4 million related to cause (b).  However, Mr Hatchwell responded to this by presenting an analysis at RAD Tab 34 that show that the “third bit” only accounted for $1.23 million out of a gap of $10 million.  Therefore, the remainder of the gap of $8.77 million arose because a DCF valuation as carried out by Mr Hatchwell, using all the inputs of Professor Hughes at his Up-dated GH-16 (including a discount rate of 34%) would arrive at an initial rent of $14.56 million.  The DCF valuation could only arrive at an initial rent as low as $4.56 million by using a discount rate of 54.2%, or 52% (if the “third bite” was removed).  Only at that stage, the Commissioner submitted, Professor Hughes did concede that there was a methodological difference producing this gap, namely the double return for the HT in his PDM.

270.  In response to the accusation of “double counting” of the HT’s share in Professor Hughes’ PDM, the Appellant simply replied that, “the HT’s share is high but that is a consequence of the discount rate and has nothing to do with the structure of the model”.  We do not accept this explanation at all.

Sensitivity analysis on PDM valuation

271.  Professor Baum produced various tables at RAD Tab 16 to show the effects of removing the “second bite” in columns (11) and (12) and the “third bite” (50:50 sharing at the end) of the Up-dated App. GH-16.  All the Tables in RAD Tab 16 assumed no rental growth.  Professor Hughes accepted that Table 1 in RAD 16 replicated columns (1) and (8) of his GH-16.  Professor Baum show in Table 2 of RAD 16 that after removing the “second” and the “third bite”, but simply changed the yield for estimating the terminal value (i.e. the sale value) of the completed development at the beginning of year 6 from 7% to 5.5%, and otherwise continued to use Professor Hughes’ assumptions of (i) no real growth in occupational rents during the 5-year period and (ii) the 34% nominal discount rate, the initial ground rent would still rise from $4.56 million (as in his Up-dated App. GH-16) to $21.57 million.  Therefore, the Commissioner said that this show that Professor Hughes’ 5-year valuation is very sensitive to the “exit yield” upon the sale of the completed development.

272.  Subsequently, Professor Baum extended the sensitivity exercise carried out in RAD Tab 16 by producing an additional series of 24 tables in his AB App. 4, which was later included in RAD Tab 34.  The Commissioner said that none of the calculations had been challenged.  The Commissioner submitted that, “the results (of Professor Baum’s analysis) in RAD Tab 38 (entitled AB Appendix 4: GH 16 amended) show that once the “second” and “third bites” are removed, then even if no real growth of rents is assumed for the sake of argument, the adoption of a sensible discount rate or a sensible terminal value for the sale of the completed development, produces results which strongly support the Commissioner’s valuation at $26.8 million.  If two or more such sensitivities are combined, then the Commissioner’s valuation is distinctly conservative.”

273.  In fact, the summary results of the said sensitivity exercise carried out by Professor Baum show that for the required return of 33.98%, the rents vary between $14.56 million to $30.34 million, depending on a combination of different exit yields of 7% or 5.5%, a 5-year or a 4-year model, and the land exit yield of 3.8% or 4.4%.  Similarly, the summary results range between $21.23 million and $36.26 million for the required return of 18.00% and, between 22.42 million and $37.37 million for the required return of 14.75%.  This shows that Professor Hughes’ own PDM valuation actually supports the Commissioner’s valuation after correcting his wrong deduction in the PDM of an extra return for the HT and the adoption of sensible discount rate and terminal value for the sale of the completed building.

Tribunal’s decision on Professor Hughes’ PDM valuation

274.  We have considered the parties’ evidence and the submissions on the structure and the construction of Professor Hughes’ PDM carefully.  We come to the conclusion that we agree with the submissions of the Commissioner.

275.  On the structure of the PDM, we do not agree with Professor Hughes that he could freely choose to mix up the inputs of an otherwise R & E model or DCF model.  We agree with the Commissioner that it is apparently clear that the inclusions of columns (11) and (12) of the Updated App. GH-16 (i.e. HT’s investment excl. return carried forward and HT’s required return respectively) had resulted in “double counting” because whilst columns (1) to (7) of GH-16 (i.e. rent revenue, construction costs, operating expense and operating cash flow excluding rent) contain inputs suitable for analysis by a DCF Method, columns (11) and (12) do not belong to a DCF Method (i.e. these are the columns in which the HT’s return on his investment is specifically deducted).  Since in a DCF Method, the discounting would take place in column (13), a DCF Method should only contain the inputs that are based on columns (3) to (7) but not columns (11) and (12).  We also note that only when Professor was asked “whether the 34% discount rate had been applied twice in relation to column 12” did he reply “yes that’s right. But column 12 only.” (Day 17, 1431:44-46).  Therefore, it is clear that columns (11) and (12), which belong to the R & E Method but not to the DCF Method, were wrongly inserted in the PDM.

276.  We do not agree with the Appellant that the Commissioner has overblown these issues of the “second bite” and the “third bite” of the HT’s share of the rent.  We note and agree with the Commissioner’s Written Closing Submission in para. 4.60 that, “Once the concealed “second bite” is removed, along with the smaller “third bite”, it is clear that even with the absurdly high discount rate of 34% (and accepting for the sake of argument all the other assumptions they use), the Appellant’s valuation (ignoring GD App. 7) should have arrived at a rent of at least $14.56 million, or more properly $18.83 million for a 4-year duration (instead of the 5-year duration assumed by Professor Hughes).  This contrasts with Professor Hughes’ valuation of 4.56 million.  Also, on that basis, GH makes a non-sense of Mr Charman’s valuation on CB of $3.63 million.”  We also agree with the Commissioner’s criticism that although Professor Hughes knew that his PDM valuation only gave the HL a mere 3.04% of the GDV “cake”, he failed to “to stand back and check” if his valuation was realistic.

277.  To the extent that Professor Hughes criticized Mr Tang’s adoption of various inputs in the latter’s R & E valuation, we are not satisfied with Professor Hughes’ assumptions of (i) a 5-year tenancy period with zero real growth; (ii) a yield of 7% for calculating the terminal value of the completed development and (iii) a 34% (nominal) discount rate.

278.  We concur with the analyses by the Commissioner’s experts that in the PDM valuation, the site rent is sensitive to the duration of the assumed tenancy.  Basically, the shorter the duration, the higher the resulting site rent.  In that case, we fail to understand why even after the parties’ experts have agreed that the sum of (i) the construction period and (ii) the period to reach 90% occupancy was 50 months, Professor Hughes still did not revise his duration of the assumed tenancy.  Also, we do not agree on the adoption of 7% for valuing the terminal value of the completed development.  Professor Hughes failed to satisfy us on the adoption of the chosen yield.  Finally, we have already stated above that we do not agree with Professor Hughes’ analysis of the rates of return for the HT & HL in arriving at the discount rate.

279.  To conclude, we agree with the Commissioner that the PDM is flawed and therefore unsuitable for valuation purposes for reasons set out in the Commissioner’s Written Closing Submission the important parts of which are summed up above.  We further agree that the PDM is thoroughly unreliable in valuing the ground rent of the Appeal Tenement.

The Commissioner’s valuation on R & E Method as “a check”

280.  Mr Tang opined that his primary valuation by CB was supported by his valuation on the R & E Method.  However, it is well known that in the R & E Method of valuation, it involves a multiplicity of variables.  If the parties failed to agree on the various aspects of the CB valuation, which in this case only involves two variables, it is not difficult to foresee that in the R & E Method, the parties would have far more differences.  This is exactly what we find in this valuation exercise.  Now, we will first of all find out the extents of agreement and the remaining differences in these variables before we touch upon the methodology of the R & E valuation as carried out by Mr Tang.

281.  Mr Tang summed up in para. 9.2 of his Speaking Notes the various components used in the Commissioner’s R & E valuation.  Mr Tang reminded us that the inputs and estimates used in the R & E valuation were set out in para. 8.8 to 8.40 of Mr Brooke’s Original Report, which were subsequently amended and set out at App. NB71 of Mr Brooke’s 4^th Supplementary Report.  Many of the inputs, including the initial occupational rents, had already been agreed between the parties, as set out in the Statement of Agreed Facts and Document SAF8.  The remaining main differences between the parties were in the following areas:

(1)	the revenue estimation (in terms of the varying occupancy rates during different periods in the life of the completed building, the future growth rates of occupational rents and the whether the proposed building at the valuation date was expected to be Grade A or Grade B) and the more general issues underlying this and other factors, such as the general state of the property market, particularly the office sector in the part of North Point surrounding the Appeal Tenement, the economy of Hong Kong and the demand for the development sites ;
(2)	the effects and treatment of depreciation on the rents, and
(3)	the discounting rate to be applied in arriving at “the tenant’s share”.

282.  The parties’ inputs and estimates to be used in the R & E valuation are summarized in the table below (see page 10 of Mr Tang’s Speaking Notes):

Take-up and Occupancy Rate	Appellant’s Assumption	Respondent’s Assumption
At the date of completion (OP Date)
6 months from OP Date	10%	10%
12 months from OP Date and 19 years thereafter	50%	50%
18 months from OP Date and for the life of the building	/	95%
After 20 years from OP Date	90%	/
	/	92%

Future Growth in Rental Values	Appellant’s Assumption	Respondent’s Assumption
Real growth	Nil	3% p.a. over first 10 years after completion of the building 2% p.a. over next 10 years 1% p.a. over balance of lift of the building
Effect of building age	2% p.a. over first 10 years after completion of the building 6% p.a. over balance of lift of the building	Reflected in declining rates of real growth set out above

283.  Therefore, notwithstanding the agreements of large number of inputs (including even the initial occupational rent at the relevant date), the seemingly slight differences in some of the inputs between the parties still produced drastically different results in the R & E valuation.  This is because the R & E valuation, when run over a long period of time such as 43 years as suggested by Mr Tang for the Commissioner, will be sensitive to various inputs such as the occupancy rate, the growth rate of rent and the depreciation rate.  Therefore, when deciding on whether the results of the Commissioner’s R & E valuation, as a check, support the Commissioner’s primary valuation on the CB, we have to bear this in mind.

284.  Mr Tang summed up his R & E valuation in his First Report dated 11 November 2003.  Based on the valuation inputs provided to him by Mr Brooke in App. NB 28 and the Weighted Average Cost of Capital (WACC) provided by Professor Chan, Mr Tang carried out his R & E valuation, which were summarized in PKT pages 42 – 48.

285.  Mr Tang prepared at PKT App. 21 and 22 spreadsheets showing the derivation of various components in the R & E valuation.  He calculated the inputs, “Tenant’s Asset (Development Cost)” and “Depreciation” in PKT App. 21, which were then used in PKT App. 22.  There would be no income until the completion of the building after which receipts and expenses relating to the occupation of space in the building were estimated.  Following the format of a usual R & E valuation, Mr Tang calculated in PKT App. 22 the “Divisible Balance”, being the difference between the Gross Rental Income (after deducting Letting Expenses and Regular Repairs and Maintenance as well as Allowance for Major Repairs) and the Tenant’s Asset (after allowing for Depreciation).  He then allowed for the Tenant’s Share before arriving at the Landlord’s Share.  In this manner, he calculated the total Present Value of the Landlord’s Share for the whole period of analysis (being the sum of the period of construction and the assumed life of the building completed on the subject site) in the sum of $342.95 million.  Mr Tang said that this sum of the Present Value of the Landlord’s Share was the amount the HT could afford to pay to the HL over the economic life of the building.

286.  Mr Tang said in PKT page 45 that, “according to the calculation at Appendix 23, the HT will be prepared to pay a fixed annual rent of $56.42 M in each year during the continuation of the tenancy.  This is the annual equivalent of the NPV of $342.95 M”, as computed in PKT App. 22.  However, after Mr Tang considered the comments of Mr Hatchwell’s report that it would be more realistic to assume that the rent would vary during the course of the tenancy, he decided to convert the NPV of future rents into a rental stream, which would be reviewed and revised regularly upon the following assumptions: “(a) the rent is to be reviewed every year…; (b) the anticipated long-term rental trend for development site is 9.5% p.a., in line with the expected initial trend for occupational rents; (iii) the discount rate for the expected rental stream is the same as the nominal WACC rate i.e. 18.0%; and (d) the rent is payable monthly in advance.”  On that basis, Mr Tang produced calculations at PKT App. 24 showing a succession of rising rents with an initial rent of $27.73 million.

287.  Mr Tang then reviewed and revised his R & E valuation in his Second Supplementary Report dated 6 April 2006.  He has agreed with and taken into account the various comments made by the Commissioner’s other experts and the revised valuation assumptions proposed by Mr Brooke.  In particular, Mr Tang did not agree with Professor Hughes’ in his adoption of a WACC of 34%.  Instead, he considered and agreed with Mr Hatchwell in the latter’s analysis of the appropriate rate of return for the HT (at PKT pages 71 to 74).   He therefore agreed with Mr Hatchwell in rejecting Professor Hughes’s proposition that the typical Hong Kong property company was a combination of the HL and the HT and that the return for the developer should be decomposed between a HL with a lower risk and a HT with a higher risk.  In the final analysis, he agreed with the following statement from Mr. Hatchwell’s report (LH2 para. 10.5):

“

the best evidence of the hypothetical tenant’s required return is supplied by the analysis of the price paid by the real world developer; and that indicates a nominal required return of close to 14.75%.  That developer has all the essential characteristics of the HT except that it does not pay rent, except Government rent.  Whilst rejecting the proposition that the return to the HT will be more than 18%, I embrace the probability that it will be substantially less.”

288.  Mr Tang concluded in para. 4.31 of his Second Supplementary Report the following,

“

In view of the comments made by the Respondent’s experts, with which I agree entirely, the nominal WACC of 18% previously adopted is indeed conservative in favour of the Appellant.  I would therefore prepare two revised R & E valuations in Section 6 below: one using the previously adopted nominal WACC of 18% (real WACC at 10.8%) to arrive at the minimum rent the hypothetical tenant can afford to pay and the other, as a sensitivity test, using a 14.75% nominal WACC to show what the result would have been if I had adopted a less conservative rate of return for the HT’s share.  The 14.75% nominal WACC is close to the IRR of 14.79% calculated by Mr Brooke.  The real WACC is 7.75% (i.e. 1.1475/1.065-1).”

289.  On the above bases, Mr Tang revised his valuations in PKT Revised App. 22 & 24.  In PKT Revised App. 22, assuming the HT’s return at 18% nominal (10.8 % real), the NPV of the HL’s Share became $310.98 million and the initial rent that the HT could afford to pay changed to $25.15 million.  Similarly, on the HT’s alternative return at 14.75% nominal (7.75 % real), the respective NPV of the HL’s Share and the initial rent became $570.47 million and $32.06 million respectively.  Using the results of both analyses to check against his then valuation on the CB of $26.62 million, Mr Tang concluded that the rateable value of the Appeal Tenement, at $26.62 million, was supported by the R & E Method of valuation.

290.  In his Third Supplementary Report dated 24 August 2006, Mr Tang carried out an exercise similar to that in the Second Supplementary Report.  Following Mr Hatchwell’s conclusion (in LH Third Supplementary Report para. 10.6) that there was no evidence to support a return and discount for the HT that was anything higher than that adopted by Mr Hatchwell and Professor Baum, then revised to 14.12% nominal (7.15% real), Mr Tang carried out the R & E valuation using this revised rate of return for the HT, in addition to the R & E valuation using, as before, the HT’s return at 18% nominal (10.8% real).  In addition, he revised his valuation inputs using the latest valuation parameters then agreed between the parties (including the newly agreed development period).  On the basis of the HT’s return at 18% nominal (10.8% real), Mr Tang calculated the NPV of the Landlord’s Share and the initial annual rent payable by the HT to be $274.54 million and $22.20 million respectively (see PKT App. 28 and 29).  Similarly, on the basis of the less conservative HT’s return of 14.12 % nominal (7.15% real), he estimated the respective sums of the NPV of the Landlord’s Share and the initial rent payable by the HT at $580.70 million and $30.02 million respectively (see PKT App. 30 and 31).  Mr Tang drew attention to the results that, as before, the R & E valuations show that “if a less conservative rate of return for the HT is adopted, the initial annual rent payable by the HT is higher.”  Similarly, using these 2 figures from the R & E valuation to check against the revised valuation on the CB, then revised at $26.88 million (using an estimated site value of $611 million, after grossing up to account for the payment of Government Rent and the decap rate of 4.4%), Mr Tang concluded that the rateable value should be that derived from the CB of valuation.

291.  To assist the Tribunal, both parties have undertaken various sensitivity analyses to the valuation for the Appeal Tenement, especially for the R & E valuation carried out by the Commissioner.  In particular, the Commissioner submitted that the R & E valuation as carried out by Mr Tang was ‘(a) a proper adaptation of the traditional R & E “single year look” for the purposes of analyzing cash flows over a number of years beyond the completion of the building; and (b) supported by the analogous DCF techniques’.

292.  The Commissioner said in the Written Opening Submission that it was clear from Mr Tang’s workings at PKT App. 21-24 that:

“	(i)	The object of App. 23 and 24 is to arrive at an estimate of the site rent payable at the beginning of the tenancy.  It has never been suggested that the figures in the NPV column of those tables should be treated as a pre-agreed schedule of fixed rents.  The object is simply to arrive an initial level of rent;
	(ii)	The higher the discount rate the lower the NPV of the landlord’s share and the lower the initial rent;
	(iii)	The assumption of a uniform rent results in a substantially higher initial rent than if it be assumed that the ground rent rises over time in accordance with a projected trend line…., but the assumption of a real rate of growth in site rent in Appendix 24 is simply being used to arrive at an initial rent for the ground.”

Criticism of the Commissioner’s R & E valuation

293.  The major criticisms of the Commissioner’s R & E Method of valuation were that (i) the valuation had been run over 43 years and there were no actual developer’s accounts to base on; (ii) the choice of the discount rate; (iii) the assessment of the Tenant’s Share and (iv) the results of the R & E Method did not support the valuation on the CB.

294.  The Appellant pointed out at the outset that for historic reasons, the R & E Method had been little used in England and Scotland since the Second World War.  The Appellant said that Mr Hatchwell had agreed in cross examination that there had been virtually no hereditament assessed on the profits basis and there was no decision of the Lands Tribunal to give us guidance on how it should be applied.  Of the list of cases in the Rating Form Guidance Notes, only eleven have been decided from 1950 onwards, of which only 2 cases have been cited in the Lands Tribunal before.  The Rating Forum Guidance Notes is now some 9 years old and the only subsequent case that has been mentioned is British Telecommunications Plc v Central Valuation Officer [1998] RVR 86 (“British Telecom”) determined at the level of tribunal in England below the Lands Tribunal and settled before the appeal to the Lands Tribunal was heard.

Rationale for running the R & E Method for 43 years

295.  The Appellant submitted that although both parties had examined numerous cases including those listed in the Rating Forum Guidance Notes they had found no case in which either the receipts and expenses had been drawn from accounts of the business of more than 5 years, or the future accounts had been projected for more than 5 years.  For example, in Hong Kong, in the Cross-Harbour Tunnel Co Ltd v Commissioner of Rating and Valuation [1977-1979] HKC 81, the valuation date was taken as the 3^rd quarter of 1976 and both parties’ valuations were for one year whilst in the China Light and Power Co. Ltd. v Commissioner of Rating and Valuation [1996] RA 475 (LT & CA) (“China Light and Power”), the valuation date was 1 July 1990 and both parties’ valuation were again for one year.

296.  The Appellant said that Mr Tang’s R & E valuations set out in PKT App. 22 and revised in PKT App. 28 arrived at the totality of the Landlord’s Share for a period of 43 years, i.e. the whole period of business over which the rents would be received by the landlord.  There was, however, nothing in the Rating Forum Guidance Notes that gave advice that a rating assessment should be carried out over the whole of the expected life of the business.  Therefore, although Mr Tang might argue that his valuations at PKT App. 22 and 26 provided the evidence of the totality of the rents paid over 43 years, there was no evidence to support or verify his assumption that the first rent’s rent as calculated at PKT App. 24 and PKT App. 29 would be agreed between the HL and the HT.  The Appellant submitted that in the cross examination, Mr Tang was put to this fact but he was still only able to say that this was a reasonable assumption.  In the re-examination, he was reminded about this point but he could reiterate that the real world developer had agreed to pay the auction price before he got any return.

297.  Also, the Appellant suggested that whilst the developer in the real world, by paying the auction price, virtually fixed the consideration for a lease for 50 years, it provided no evidence of what the HT would pay in the Commissioner’s hypothetical tenancy.  This goes back to the fundamental challenge raised by the Appellant that in Mr Tang R & E valuation, it is based on the assumption that the HT will be willing to pay, for the first year, a rent that has been calculated directly by reference to the projected receipts and expenses and therefore the profitability in years 1 to 43.

298.  In particular, the Appellant submitted that, “there is nothing in the Rating Forum Guidance Notes which explains how the rent should be assessed in the first year of the venture when the HT is incurring heavy expenditure and will not receive receipts for some time.”  Therefore, the Appellant reminded us that in the Consett Iron Co. Ltd. v Assessment Committee for No. 5 or North-Western Area of Country of Durham [1931] AC 396 (“Consett”)case, “the actual occupier of the mine was willing to pay a rent which was more than nil, but which was, nevertheless, very much lower than the rent payable when the mine was profitable.”

299.  However, the Commissioner said that a period of 43 years was assumed in the R & E Method in the present case because (1) that ties in with the expected economic life of the building; (2) it was necessary to project the receipts and expenses for the whole period of development (including the construction period and the life of the building) in order to reflect fully the development potential of the land; and (3) it is clear from the accounts of Sino (the holding company of the Appellant) as well as contemporaneous statements by its Chairman that the building was intended to be held for investments, rather than sale upon completion.  In addition, the Commissioner said that the R & E valuation was actually not dependent upon its being run over 43 years because “the HT will be entitled to the market value of the building in the event of the tenancy being determined earlier” (see Mr Hatchwell’s report LH Rep. 4 189/190).  As highlighted in the Written Closing Submission, the Commissioner said that Mr Hatchwell had already reminded us that although the R & E valuation had been assumed to run over 43 years at the relevant date of valuation, the assessment of rent for a development site,

“

will not need to be carried out over very many years because, of course, once the development is complete, the rateable values of the completed units on the development takes the place of the value we are trying to find in this case.  So, it’s not a case of having to provide many valuations far into the future at dates very, very divorced from the date on which the site was acquired.”

Choice of the discount rate in the Commissioner’s R & E valuation

300.  The Appellant also severely criticized the Commissioner’s choice of the discount rate that was adopted by Mr Tang in the R & E valuation.  Professor Hughes agreed to the use of the WACC developed by Professor Chan.  Otherwise, Professor Hughes did not agree to the use of other discount rates, not even the discounting rate of the project calculated by Mr Brooke and others.

301.  However, we find that even though Mr Tang and the Commissioner’s other experts have actually used both sets of discount rates when undertaking the R & E valuation, Mr Tang agreed at the end of the valuation exercise to adopt the results from using the WACC.  Mr Tang and others have described the rate from the WACC as a “conservative” rate but the important point is that notwithstanding this remark, the Commissioner has still agreed to adopt, in the final analysis, this discount rate in the check valuation.  Therefore, we have decided not to go through the details of discount rates further in this Judgment.  Indeed, we find that far too much of the time have been spent on this topic during the hearing.

302.  In addition, the Appellant submitted that the Commissioner’s R & E valuation did not, as claimed by the Commissioner, support its CB valuation.  It was indeed Professor Hughes who introduced in his First Report the “Morton’s Fork” argument faced by the Commissioner after reading Mr Tang’s comparison of CB valuation and R & E valuations in the latter’s First Report.  We have already covered this point when discussing the Commissioner’s valuation on CB.

The Tenant’s Share in the Commissioner’s R & E valuation

303.  The Appellant criticized that in the Commissioner’s R & E valuation, the HT would earn no more than its hurdle rate of return.  As a result, the Appellant said that there was no provision for the “haggling of the market”, or negotiation between the parties that might lead to the HT obtaining something that was more than the minimum return.  Therefore, the Appellant submitted that it was misleading for Mr Hatchwell to describe it as “a second bite of the cherry” when Professor Hughes assumed that the HT would seek and obtain a share of the divisible balance over and above its minimum return. And it was therefore simply wrong for Mr Hatchwell to deny that the HT had any right to earn a profit over and above its hurdle rate of return, particularly in the present case when we are not dealing with a large regulated monopoly as in the cases of China Light and Power in Hong Kong, and British Telecom in the U.K.

Tribunal’s decision on the check valuations carried out by the Commissioner

304.  Since this is the first case before the Tribunal after the CFA decided in Agrila case that the development site was “rateable”, both the Appellant and the Commissioner have spent considerable efforts in preparing their case, not only in attempting to produce their primary valuations, but also in seeing if it could be supported by check valuations and to refute the other party’s valuations.

305.  We have already decided that the primary valuation of the Appellant was made on the wrong legal basis.  We have rejected Mr Charman’s valuation, which, as rightly pointed out by the Commissioner, was only an attempt to value the HT’s right to receive the net rent for the assumed 6-year period.  Further, we have also found that Professor Hughes’ PDM valuation was flawed both in its structure and in its details.

306.  In the Commissioner’s case, it was clear from the beginning that they had considered the CB valuation as to be their primary case.  We have already decided above that this is the correct approach.

307.  In addition, what the Commissioner’s experts have tried to do was to carry out check valuations using the R & E Method (mainly by Mr Tang, with supporting valuations by Mr Brooke and Mr Hatchwell) and check valuations using the DCF Method (mainly by Professor Baum).  What we should not have lost sight of is that in the R & E Method, it is to a large extent, novel in the sense that (1) there was no actual account at all to base upon; and (2) the period of tenancy is an assumed duration of 43 years, including the period of developing the land and the physical life of the building to be completed on the land.  Also, mainly because of the long duration of tenancy that one must have regard to in order that the development potential of the subject land as a development site could be realized, the differences in terms of a few variables under dispute (i.e. the growth rate of the occupational rent (if any), the take up and the occupancy rate, the depreciation rate as well as the discounting rate) will lead to very different results in the final assessment of the Net Present Value of the Landlord’s Share to be shared between the HT and the HL, and the resulting ground rent payable by the HT.  This is exactly the situation in the present case.

308.  Of the criticisms in principle that have been raised against Mr Tang’s R & E valuation, the major one is that one could not and should not carry out the valuation over a period of 43 years.  We already covered earlier the necessity of discounting the receipts and expenses over the whole period of construction and the life of building in order to reflect the development potential of the land.  Otherwise, as pointed out by Mr Tang in the re-examination, discounting the receipts and expenses for a shorter period of time would inevitably lead to the conclusion that the rent for year 1 would be nil.  We reject the suggestion that a valuation over a period of 43 years cannot be right.

309.  As to the question of whether the HT is entitled to seek and obtain a share of the divisible balance that is above the hurdle rate, we agree with the opinion of Mr Hatchwell.  We find that the way the hurdle rate was arrived at (either based on Professor Chan’s WACC for property companies or on the analysis of the rate of return for a property developer as the HT) already reflected the reward to the HT in terms of both an allowance for risk and a return for the tenant’s capital.  No additional reward is needed.

310.  To summarise, as we have set out earlier in this Judgment, despite of the time and efforts of the experts and the parties in reaching agreements on a large number of variables, the remaining differences in just a few variables could lead to drastically differing results in the valuation on the R& E Method.  So, the question is, is the check valuation helpful?

311.  In the present case, we do find that the rate of return as well as the other valuation inputs assumed by the Commissioner are far more reasonable than those assumed by the Appellant.  On the whole, we accept that the Commissioner’s valuation by the R & E Method does lend support to the Commissioner’s valuation on the CB.

312.  However, on a more general level, we do not find it really necessary, as in the present case, to indulge in the valuation on the R & E Method as such since we have a much simpler valuation, which is robust and is dependent on only two variables both of which are based on market evidence.  There is no point of carrying a check valuation that is less reliable than the primary valuation in terms of the structure of the valuation method (requiring discounting of revenues and expense for a very long period of time) and the greater number of variables.

Final conclusion on rateable value

313.  We have said in the earlier part of this Judgment that we do not agree that the Appellant’s Legal Advice to the Valuer was correct.  As a result, we have decided that the Appellant fails to substantiate their case in their primary valuation in the present case.

314.  Likewise, the Appellant also fails to substantiate the valuation of Mr Charman and that of Professor Hughes.

315.  On the contrary, we have decided that Mr Tang’s CB valuation is a correct method for valuing the rateable value of the Appeal Tenement.  We have also found that that the Appellant fails to substantiate the case that Mr Tang’s CB valuation was erroneous in principle or in any of the inputs.

316.  We therefore uphold the primary valuation of Mr Tang and hold that the rateable value of the Appeal Tenement should be reduced from $29,640,000 in the Rent Roll to $26,880,000.

The Hon. Mr Justice LAM, President:

317.  Section 27 of the Rent Ordinance gives the power to the Tribunal on hearing an appeal to determine the appeal and to do the following,

(a)	make such order as it thinks proper;
(b)	awards costs to any party;
(c)	direct the Commissioner to amend the Government Rent Roll in any manner; and
(d)	make such other direction as to the payment of Government rent as may be necessary.

318.  Mr Roots submitted that notwithstanding the wide-ranging power provided for under Section 27, it has to be confined in scope with reference to the appeal.  Counsel submitted that since the appeal is against the decision of the Commissioner seeking to reduce the Government rent payable, the Tribunal should not increase the rent payable even if it comes to the conclusion that the figure in the Rent Roll is too low after hearing all the evidence.

319.  That contention would be correct if the Tribunal’s power is confined to the determination of the appeal.  However, Section 27 is not that restrictive.  If the power of the Tribunal is so confined, there is no need to have Section 27(1)(c) and (d).

320.  Section 27(1)(c) is worded in such a manner to give the power to the Tribunal to amend the Rent Roll “in any manner”.  It does not depend on the final outcome in terms of the success of the appellant.

321.  I think Mr Holgate is correct in his submission that bearing in mind the principle of equality and the public interest in Government rent appeals, provided that the rule as to natural justice and fairness has been complied with, there is no reason why the Tribunal should not direct the Rent Roll be amended in a manner to reflect the true and correct valuation even though it is not a figure advocated by an appellant.

322.  We have concluded that in the present case, the Government rent should be reduced from $29,640,000 to $26,880,000.  Hence, we direct the Commissioner to amend the Rent Roll accordingly pursuant to Section 27(1)(c).

323.  However, as far as the appeal is concerned, the Appellant’s case, whether based on the primary valuation of Mr Lynch or the valuation of Mr Charman or that of Professor Hughes, has been rejected by us.  In the circumstances, the appeal should be dismissed.

324.  We would also make a costs order nisi that the Appellant shall pay the Commissioner’s costs for the appeal, and such costs will be taxed if not agreed.  The costs order nisi will become absolute if there is no application for variation within 14 days.