Earl Cadogan and another v Sportelli and another and similar appeals

DECISION ON PRELIMINARY ISSUES Introduction                         1. The Leasehold Reform Acts give the holders of long leases of residential premises the right to “enfranchise” their interests by extension of their leases or purchase of the freeholder’s interest. The statutory provisions for the assessment of the price payable on such enfranchisement of the higher value houses brought within the leasehold enfranchisement provisions by section 9(1A) of the Leasehold Reform Act 1967, and of flats, whether by way of collective enfranchisement or a single extended lease, under the Leasehold Reform, Housing and Urban Development Act 1993, all require the ascertainment, upon assumptions to which we will return later, “of the amount which at the valuation date [the freeholder’s] interest might be expected to realise if sold on the open market by a willing seller”.                         2. The ascertainment of this sum is only part of the valuation process required under the Acts. By valuers’ convention, the value of the right to receive a ground rent during the term has been assessed separately from the value of vacant possession at the end of the term. The former is ascertained by capitalisation of the ground rent to arrive at the present value of the flow of income as at the valuation date. The latter has been arrived at by ascertaining the open market value of the freehold interest with vacant possession as at the valuation date and then adjusting that value to reflect the fact that vacant possession will not be available until the end of the term. The adjusting factor is called the “deferment rate”. The valuers in the present cases explain it thus: it is “the annual discount applied, on a compound basis, to an anticipated future receipt (assessed at current prices) to arrive at its market value at an earlier date” [that is to say the valuation date].                         3. For thirty or more years after the passing of the 1967 Act valuers dealing with leasehold enfranchisements agreed a deferment rate of 6% within the prime central London area (PCL). Somewhat higher rates were adopted in less high quality areas. In about 2003, however, freeholders in the PCL area began to question why, when yields on all other investments had been falling, the deferment rate should be assumed to be static. In Cadogan Holdings Ltd v Pockney (LRA/27/2003, 19 May 2004) the Tribunal (N J Rose FRICS) held that a deferment rate of 5.25% in respect of a house in the PCL area was not too low and applied this rate in the valuation. Following this a number of decisions by leasehold valuation tribunals became the subject of appeal to the Lands Tribunal on the deferment rate issue, and the Tribunal decided to hear together a number of these. In order, if possible, to receive evidence as to the effect upon the rate of whether the valuation was of a house, a flat or a block of flats, five cases were heard together. The Tribunal’s decision dated 15 September 2005 is usually referred to by reference to the name of the appellant (on an issue other than the deferment rate) in the lead case, Arbib v Earl Cadogan [2005] 3 EGLR 139.                         4. In Arbib the Tribunal (His Honour Judge Michael Rich QC and P H Clarke FRICS)                    (i) rejected the established rate of 6% (para 180(3));                    (ii) arrived at a norm in the PCL area of 4.5% for houses and 4.75 % for flats (para 180 (11) and (15));                    (iii) in the absence, as they found, of reliable market evidence (para 180(2)) arrived at those figures on the basis of financial market evidence (para 180(10));                    (iv) took as a starting point a risk-free rate represented by index-linked gilts (para 145) and arrived at a deferment rate of 4.5% by allowance for risk (para 152);                    (v) arrived at such figures however, as they recorded, without the benefit of specialist evidence as to the analysis of such evidence (para 126, and see para 142 as to the reasons therefor).                         5. By the time that the decision in Arbib was published there were appeals outstanding in a number of cases where the LVTs had followed the established rate which the Tribunal in Arbib had held to be wrong. The decision, however, left room for further evidence and argument as to what other figure was right. In those circumstances the Tribunal sought to collect up such further cases for hearing together, on the issue of the appropriate deferment rate. This, however, proved more difficult than might have been hoped from the point of view of the Tribunal’s providing further guidance on the issue, having heard evidence which filled the gaps upon which it had commented in Arbib. A number of the outstanding appeals included other issues which were, in some cases, of greater importance to the parties than the deferment rate. In general, respondents who were concerned with the enfranchisement of a single dwelling were unwilling to incur heavy costs in opposing appeals by freeholders who had more extensive interests, and several appeals were settled.                         6. The Tribunal’s order dated 22 February 2006 directed that “the proper deferment rate to be applied to vacant possession value” should be determined as a preliminary issue in seven appeals to be heard together. In the event only five appeals were the subject of our hearing. Of these only two were contested appeals. They are both cases of collective enfranchisement, where each party called both valuation and expert financial evidence in accordance with the directions which had been given. We heard evidence and argument over a period of 11 days. The three uncontested appeals relate to a lease extension of a flat, a further collective enfranchisement and the acquisition of the freehold of a house.                         7. A second preliminary issue was ordered in the two contested appeals “as to the proper valuation of any ‘hope value’”. Hope value arises from the option that the freeholder has in the real market to sell the freehold or a lease extension to the tenant and thus realise the whole or part of the freeholder’s share of such marriage as exists at the date of the sale. This was referred to as “an early profit”. In Arbib the Tribunal allowed for hope value by taking account of it in the deferment rate. In the present cases it is a matter of dispute whether hope value is excluded by the statute in valuing the reversion in a collective enfranchisement; and, if it is not excluded, how it should be valued.                         8. These, therefore, are the two preliminary issues. We were not concerned with the other element in the value of the reversion, the ground rent and how this should be capitalised. Nothing that is said in this decision has any direct application to capitalisation rates. Market evidence should be more readily available for those, and in any event such rates, applying as they do to an element of static value, are determined by different criteria from those that are relevant to the deferment rate. The appeals                         9. LRA/50/2005. Earl Cadogan and Cadogan Estates Limited v Sportelli This was an uncontested appeal by the landlord (“Cadogan”) in respect of Flat 14, 85 Cadogan Gardens, London SW3 2RD against the LVT decision of 9 March 2005 relating to the tenants’ claim for a lease extension under Chapter II of the 1993 Act, with the valuation being made pursuant to Schedule 13 to that Act. The original application was on five grounds, one of which (terms of lease) has subsequently been agreed. Of the remaining four, only the issue of the appropriate deferment rate was heard in this preliminary issue hearing, and the remaining grounds are to be heard at a later date. The appellant landlord was represented by Mr Kenneth Munro (as in all the Cadogan cases) who called Peter James Clokey, a partner in the UK firm of PricewaterhouseCoopers LLP to adduce financial evidence, and Julian Mansfield Clark BSc MRICS, a partner in Gerald Eve, who gave valuation evidence. The relevant valuation date was 22 December 2003, the term remaining on the underlease was 21.25 years and there was an intermediate lease in place. We refer to this case as “Sportelli”. LRA/9/2005. Cadogan v (1) 27/29 Sloane Gardens Limited and (2) Wayhil Mahdi This appeal, from an LVT decision of 26 November 2004, related to a collective enfranchisement under Chapter I of the 1993 Act with the valuation being made pursuant to Schedule 6. The property comprises two adjoining houses, 27/29 Sloane Gardens, London SW1W 8EB, converted to 7 flats (one of which was caretaker’s accommodation), with 4 of the tenants participating and 2 non-participating. The unexpired term was 41.04 years, there is an intermediate lease, and the agreed valuation date was 14 September 2004. The two grounds of appeal were (1) whether the headlessee is entitled to recover a market rent for the caretaker’s flat from the lessees, and if he is, the amount of that rent and, (2) the appropriate deferment rate to be applied. The first ground, which was contested by the second respondent, was heard as a preliminary issue by the Tribunal, (His Honour Michael Rich QC) on 30 March 2006, and the decision, determining the issue in the respondents’ favour was dated 7 April 2006. The deferment rate issue, which was uncontested, is determined here. Mr Clokey and Mr Clark gave evidence for the landlord. We refer to this case as “27/29 Cadogan Gardens”. LRA/63/2005. (1) Bircham & Co (Nominees) (No 2) Limited and (2) Ms S E Stowell v Mr M C and Mrs S Clarke The third uncontested appeal, from an LVT decision dated 15 March 2005, relates to the enfranchisement of a house, 13 South Terrace, Kensington, London SW7, pursuant to section 9(1A) of the 1967 Act. 22.58 years remained on the lease at the valuation date of 18 May 2004. The only issue relates to the appropriate deferment rate. Mr Munro appeared for the landlords and called Mr Clokey, and Roland Cullum FRICS, a partner in Cluttons, gave valuation evidence. This case is referred to hereafter as “13 South Terrace”. LRA/117 & 131/2005. Cadogan v Grandeden Property Management Limited This is an appeal by the landlord and cross-appeal by the nominee purchaser from a decision of the LVT dated 16 September 2005 in respect of the collective enfranchisement (Chapter I and Schedule 6 to the 1993 Act) on 59 Cadogan Square and 105 Cadogan Gardens, London SW1X 0HZ, an 8 storey mansion block consisting of 24 flats (including 1 caretaker’s flat). There are 17 participating and 7 non-participating tenants, the unexpired term at the valuation date of 7 July 2005 was 71.05 years, and there was an intermediate lease in place for the term plus 10 days. Outstanding issues other than those relating to the deferment rate and hope value will be heard at a later date. Mr Munro called Mr Clokey (financial) and Mr Cullum (valuation). For the respondents, Mr David Holgate QC and Mr Tom Jefferies called Professor Colin Martyn Lizieri PhD, BA (Hons), FRICS, FRGS, Head of the Department of Real Estate & Planning at the University of Reading Business School as the tenants’ expert economist, and Robert Orr-Ewing FRICS, a partner in Knight Frank, who gave valuation evidence. This case is referred to as “Grandeden”. LRA/38 & 40/2005. Howard de Walden Estates Limited v Maybury Court Freehold Company Limited This also is an appeal by the landlord and a cross-appeal by the nominee purchaser. It is from an LVT decision of 14 January 2005 relating to Maybury Court, Marylebone Street, London W1Q 8JE, 5 conjoined blocks totalling 68 flats converted from former nurses’ accommodation in the mid 1980s. The collective enfranchisement of the remaining 57.4 years term of the leases (Chapter I and Schedule 6 to the 1993 Act) is by 44 participating tenants (24 are non-participating) with a valuation date of 22 October 2004. There is an intermediate lease. A number of grounds remain outstanding for determination at a later date, the sole issue before us at this hearing being the deferment rate and hope value. Miss Judith Jackson QC and Mr Michael Pryor appeared for the appellant landlord and called Charles Edward Dumas BA (Hons), Head of the World Service at Lombard Street Research who gave financial evidence, and Mr Clark who gave valuation evidence. Mr Stephen Jourdan, for the respondents, called Robert John Francis of Frontier Economics Ltd who gave financial evidence, and Peter Beckett FRICS, a partner in Beckett and Kay LLP, Chartered Surveyors of London W1, who gave valuation evidence. We refer to this case as “Maybury Court”. The LVT Decisions                         10. In each of the decisions on the above appeals the LVT determined a deferment rate of 6%. In Grandeden, where the decision was dated one day after Arbib, the LVT concluded (para 25): 25. “In the judgment of the Tribunal, to move away from 6% in the present circumstances would be unwise. As demonstrated above, a small change in yield [as they called the deferment rate] would have a large effect on valuation and there is no convincing market evidence that investors in the £2M to £5M reversionary market have changed their views or overturned the established settlements. As far as the Tribunal is aware, the 6% generally adopted, accepted and reflected in agreements for prime residential reversions has never been supported by factual evidence of transactions in the substantial sums of money put forward by both experts in this case. In this Tribunal’s view, it would be unsafe to overturn years of established agreements purely on the economic arguments relied upon by the Freeholder, which suggest a lower figure from that being contended on his behalf….Before departing from the established 6% this Tribunal would require market evidence…” The 6% deferment rate that the LVT determined did not include any allowance for hope value, the question of which they had dealt with in para 20: “20. Hope Value – this is a difficult issue as to which there is scant reported case law of direct assistance, and this is mostly in the Lands Tribunal. However, reference may be made to Becker Properties Limited v Garden Court NW8 Property Co Ltd [1998] 12 EG 135 LT, where the President (Judge Marder QC) considered expert evidence as to the appropriate percentage yield rate in valuing the freehold interest but inclined to the view that the value might be increased by an assumption of ‘hope value’. In that case, as in the present case, this was as to non-participating tenants seeking lease extensions at a premium, but no concluded view was expressed because the evidence did not justify finding any such value. Similarly, the present Tribunal is of the view that no quantified sums should be added for ‘hope value’ and that this aspect should be regarded as taken into account as an element in the yield.” In this appeal (Grandeden), the appellant landlord seeks a deferment rate of 3.6% inclusive of hope value. The respondent nominee purchaser seeks 5.5%, and argues that hope value is to be excluded under the provisions of the 1993 Act.                         11. In Maybury Court, the LVT specifically included hope value and said, at para 65: “65. …the Tribunal concludes that there is a prospect of non-participating tenants enfranchising in the future. After some deliberation the Tribunal has determined that this hope value should be included in the all-risks yield rate of 6%. There is therefore no need to specify a different sum in the valuation.” The appellant landlord seeks a deferment rate of 3.5% inclusive of hope value and net of transaction costs, or 3.64% if transaction costs are excluded (see paragraph 97 below). (We deal below with the question of whether a deduction should be made for transaction costs.) The respondent seeks a deferment rate of 6.25%, again upon the assumption that hope value is to be excluded.                         12. In Sportelli, the LVT said that it did not accept that the market value could logically include, in addition to the open market value, any hope value arising out of the possibility that at some future date the tenant might wish to extend his lease or purchase the freehold (para 144). It preferred the submission of counsel for the tenants (para 148), who had said that the expression [in para 3(2) to Schedule 13 of the 1993 Act] that the tenant is not “buying or seeking to buy” should be treated as meaning that the purchaser was not interested now or in the future in purchasing any additional interest in the premises. That, he had said, would produce an entirely fair result, as the landlord would get a share of the marriage value separately (para 147). The appellant landlord seeks a deferment rate, inclusive of hope value of 3.6% if transaction costs are allowed, or 4% if they are not.                         13. In the LVT decision on 27/29 Sloane Gardens, hope value was not specifically referred to. The LVT had analysed the evidence of Mr Clark, who sought 5.25% for the landlord on the grounds that due to the steady reduction in yields from gilts, returns on short-term lettings and those indicated in the prospectus of the Freehold Investment Trust, the accepted 6% yield for deferment was now too high. It also considered the evidence of Mr Charles Boston, for the tenants, who said that 6% had been established for many years and was still supported by an overwhelming body of evidence. The LVT concluded (at para 43) that it did not find the arguments for moving from the established deferment rate of 6% sufficiently supported. In this appeal, the landlord contends for a deferment rate of 3.5% if transaction costs are allowed, and 3.7% if they are not.                         14. In 13 South Terrace, the landlord seeks a deferment rate of 3.75% in this appeal. However, Mr Cullum, for the landlords, had argued for a rate of 5.25% before the LVT, and Mr Robin Sharp had been seeking 6% for the tenants. Again, the question of hope value was not specifically considered. Financial evidence                         15. The Tribunal’s primary difficulty in Arbib had been the lack of expert evidence from which to identify the nature of the products from which statistical records used by the Tribunal in their decision had been obtained (see for example para 143). We have been grateful for the evidence, much of it agreed, of the four financial experts who were called in these appeals. We think it would be helpful for the understanding of this decision and comparing it with Arbib to set out from the experts’ statement and glossary of terms and abbreviations, their agreed definitions and the particular component parts that go to make up the risk-free rate and the risk premium.                         16. The risk free-rate is defined as: “the return demanded by investors for holding an asset with no risk, often proxied by the return on a government security held to redemption,” They agreed at paragraph 3 of their statement that “The deferment rate can be calculated from the addition of an appropriate risk free rate and an appropriate risk free premium, with a deduction for capital growth” although this was not the method adopted by Mr Dumas (who gave evidence on behalf of Howard de Walden). They further agreed that for valuation methods that did require a risk-free rate of return, Government bonds are the usual source of information. The risk premium is: “The additional return required by investors to compensate for the risk of not receiving a guaranteed return.”                         17. Three of the experts accordingly gave evidence as to the “Equity Market Risk Premium” (EMRP) which is described as: “A component of the cost of equity in the Capital Asset Pricing Model (CAPM) which is an accepted model relating asset returns to systematic risk in the investment market,” to which they applied a multiplier known as Beta: “the measure of the correlation of returns on a particular asset with the returns on a portfolio of investments or assets”. Using the model called CAPM, therefore, the product of the EMRP and the beta is the risk premium that is relevant for the purposes of the investment that is being evaluated. Mr Clokey (Cadogan), Mr Francis (Maybury Court residents) and Professor Lizieri (Grandeden) agreed that the beta is best assessed from the empirical assessment of quoted companies, but significant judgment was required to select a suitable figure for the purpose of valuing the reversionary freehold interest. They thus used this common methodology to arrive at a deferment rate, and they arrived at the following figures: Clokey Lizieri Francis Risk-free rate 1.4% 2.5% 2.5% EMRP 4.5% 4.5% 5% Beta 0.4 1.0 0.5 Risk Premium 1.8% 4.5% 2.5% Real Growth 1.5% Deferment rate 3.2% 5.5% 5% Thus they were able to agree figures that were not too disparate for each of the first two elements of the assessment of a deferment rate, and that were within limits to which Mr Dumas also subscribed.                         18. On the risk-free rate, Mr Clokey maintained that, in order to make a valuation at a particular date, the spot rate yield at that date for a term commensurate with the term of the asset to be valued should be chosen. His 1.4% was a rounding of the spot rate for the longest dated unit-linked gilt available at the valuation date. None of the other financial experts thought a spot rate appropriate to assess the risk-free rate, although they differed as to their opinion of the length of period to be analysed. They took the view that the market for index-linked gilts was currently distorted by institutional factors (such as what was described as the “pensions panic”), so that it would not provide an accurate assessment for the risk-free rate for investors in freehold residential conversions. Professor Lizieri said that the apparent decline in yields from long-term gilts should be compared with the more consistent rates for 10-year gilts before drawing conclusions as to the risk-free rate. He said that the average index linked yields on a five year rolling basis over the last decade suggested that a more stable market based risk-free rate would be in the range 2.0% – 2.5%.                         19. Although Mr Dumas did not regard it as necessary to assess a risk free rate he was, having regard to the Tribunal’s approach in Arbib, asked to give his opinion as to the risk-free rate. He proposed 2% as being the lower end of a bracket of 2.0 – 2.2% best supported by zero-coupon real yield evidence, and therefore a “representative long-run real fixed zero-coupon interest rate”, but he reiterated that it was unnecessary to ascertain such rate in order to arrive at a deferment rate. There was unanimity that inflation was best stripped out by reliance on index-linked gilts. Zero-coupon gilts (that is “a debt instrument that does not make coupon payments”) were selected because the investment to be valued is without any income flow. Index-linked gilts have, however, been issued only since 1983, and although there has been a market off and on in “splits”, indicating the zero-coupon rate, since 1986, the experts were agreed that the information was inadequate to provide a clear single figure for a period commensurate with the terms for which a risk-free rate has to look forward.                         20. On the risk premium, the experts all agreed that, where applicable, the assessment of the EMRP for use in the CAPM required judgment that relied upon a broad selection of empirical analyses, academic studies, legal and/or regulatory precedent and survey information. There was a range for the EMRP which, for the purposes of assessing the deferment rate, was between 4.5% and 5.0%. Mr Dumas, who did not use this approach, said that if he had done so, 4.5% would have been appropriate. The experts were rather further apart in respect of the appropriate beta to be applied. Whilst Mr Clokey and Mr Francis in arriving at their betas of 0.4 and 0.5 respectively, both placed emphasis on the Grainger Trust as an investment company that had a percentage of its portfolio in residential reversions, Professor Lizieri considered also Berkeley Group, Development Securities and Helical Bar. He said that it could be seen that the betas of companies whose profitability depended more upon volatile capital values were much higher than those that produced a regular income stream. Bearing in mind the length of time to reversion (in Grandeden’s case, 71 years) and that betas tended to move back towards a value of 1, he considered that a beta of 1 was appropriate. Mr Dumas, however, had adopted a different approach, seeking to ascertain a deferment rate from what he described as “the all-in return” on residential property. He criticised the use of the EMRP as it was an analytical tool that was not a directly measured value, and its application by way of “beta” involved an analysis suitable to the equity market as opposed to the quite different and significantly larger property market. In a supplementary report he compared the size of the property market with the equity market, out of which the returns of a limited number of allegedly comparable companies were considered. He described the application of CAPM as “the tail wagging the dog”.                         21. Professor Lizieri also used his expertise in the field of property (he was the only financial expert who was also a chartered surveyor) to provide an alternative estimate of the risk premium, without using CAPM, by assessing the adjustment to be made to a risk-free rate to reflect the particular disadvantages of investment in freehold reversions: volatility, illiquidity, depreciation and obsolescence. He considered the volatility of house prices as compared with equities and produced two graphs derived from ODPM Office for National Statistics showing movements, on annual and quarterly bases since 1970. These showed that real house prices were prone to shocks and were highly cyclical, there being evidence of persistent periods of negative growth as well as sustained periods of positive growth. Despite the overall anticipation of positive long-term real price growth, an investor in a reversion would need to consider whether the market might be depressed either at the point when he might wish to sell the interest, or at the point of reversion. The considerable volatility in house prices thus created a risk for investors, albeit price volatility was even greater for equities. Property was also, he said, an illiquid asset, in that the ease of turning a property asset into cash was affected by volatility, restrictions in the market, the length of time it took to sell and high transaction costs which, unless the asset was held for the long term over which those costs could be amortised, could significantly affect the return. He suggested an illiquidity risk premium of 1.1% to 1.3% 22. Whilst physical depreciation of the property could be a negative factor, Professor Lizieri said that he appreciated that lessees were on the whole obliged to maintain their properties and that it was in their interests to do so, but there could be unrecoverable costs incurred in monitoring the maintenance of those obligations and in remedying situations where lessees defaulted. He thought also that there could be an additional risk of functional as against physical obsolescence in that tastes and styles change and buildings that are currently in vogue, or the areas in which they were situated, might be much less attractive to the market by the time of the reversion. Overall, offsetting the extra volatility of equities against property’s relative illiquidity and some additional allowance for the possibility of physical and functional depreciation, he said he felt an appropriate risk premium would fall in a range of 4% to 5%, and that 4.5% was appropriate.                         23. As to real growth, neither Mr Clokey nor Mr Francis was prepared to offer a view as to the rate to be assumed, although Mr Francis accepted in his initial report that an allowance would have to be made, and Mr Clokey agreed in answer to the Tribunal that it should be incorporated into the deferment rate but that amount was for the property experts to determine. Professor Lizieri said that house prices had increased at a real rate of some 3.7% between 1970 and 2005, but as the ODPM graphs had shown, growth had been highly uneven and he was of the view that real growth was likely to be much more muted in the future. Growth in private renting as tenure choice, a decline in household size, reducing impact of the ‘baby-boom’, an ageing population and the eventual release of more land to accommodate housing needs, together with an analysis of international trends, all consorted to cap long term growth, he thought, at no more than 2%. From this there needed to be deducted the capital input required to maintain the property’s value, leading him to the conclusion that realistic real growth over the next 70 years would be in the region of 1.5%. Mr Dumas identified an annual increase in real (that is, inflation adjusted) house prices of 2.9% historically from 1953 and thought that growth at the long-run real GDP growth trend of 2.5% seemed likely.                         24. Mr Dumas’ approach, as we have said, was to derive a deferment rate directly from rental yields, and he considered a figure of 3 – 3.25% for the deferment rate to be appropriate, independent of real growth. He treated the rental yield as equal to the deferment rate, because in his judgment, all that the owner of a reversion was giving up, as compared with the owner of a freehold in possession, was the right to occupy as represented by such rental yield. It was, therefore, he maintained, unnecessary to assess the real rate of growth of the value of the freehold because that remained the same whether the reversioner had possession or not. He further sought to confirm his approach by the construction of an artificial “real mortgage rate”, but this was not his primary method and although conceptually attractive was not persuasive as being factually supportable. Valuers’ evidence                         25. Each of the financial experts accepted that any conclusion at which he arrived should be subject to adjustment to reflect specific features of residential reversions that were outside his expertise, and of the valuers only Mr Orr-Ewing felt able to follow the conclusions of the financial expert (Professor Lizieri) who had been instructed by his clients. The other three valuers adopted an element of averaging to produce the figures that they considered appropriate as a matter of valuer’s judgment.                         26. All eight of the experts made great efforts to assist the Tribunal by identifying issues between themselves, and agreeing matters within their respective areas of expertise. We appreciate this assistance, and that of the parties, which has been of considerable help not only in containing the length of the hearing of these appeals, but also to the Tribunal in the preparation of this decision. Furthermore, in the course of the hearing, the Tribunal raised a number of additional questions for the experts and all of them gave careful and considered answers for which the Tribunal is also most grateful.                         27. The valuers produced their statement of agreed matters only after the financial experts had produced theirs. We consider it helpful to set out those parts essential to the understanding of this decision, and include, where appropriate, some reference back to the financial experts’ statement. Agreement was reached, firstly, as to historic real growth: “1. Taking at least a 25-year view and a near-term view, the growth in residential values has exceeded inflation. There have been periods, of varying duration, during which growth has exceeded inflation and periods when growth has fallen short of inflation, as shown in Mr Dumas’s report (chart 3 at his paragraph 4.9), which shows real growth since 1953.” This chart supports Mr Dumas’s figure of 2.9% annual real house price growth over that period.                         28. Secondly there is agreement as to the primacy of the market: “2. The 1993 Act requires an assessment of the value of the landlord’s interest in accordance with Schedule 6 or Schedule 13. Deferment rate is one component of the assessment of the value of that interest.                     4. The market value of an interest in real estate is best derived from evidence of market transactions in that interest, but see the reservations expressed at statements 13-17 below…” Those paragraphs set out the difference of opinion between Mr Beckett on the one hand and Mr Cullum and Mr Clark on the other, as to whether or not there was a market in residential reversions conformable with the Act and whether reasonably plausible conclusions could be drawn from such comparables. At that stage Mr Orr-Ewing remained, as he accepted, an agnostic.                         29. Thirdly, as to the alternative source of evidence, namely financial markets, all four valuers agreed: “11. Financial data are an interesting and useful backdrop to the determination of deferment rate, but cannot be definitive [by which it was agreed at the hearing they meant ‘determinative’].” Mr Cullum substantially, and Mr Beckett (who used it as a check only) entirely, rejected the financial data as part of their assessment of the deferment rate.                         30. Fourthly the valuers sought to agree a definition of deferment rate but differed as to the relevance of “hope value”, as it was referred to in the definition of the issues in these appeals: “3. In the context of Schedule 6 and Schedule 13 valuations, deferment rate can be defined as: the annual discount applied, on a compound basis, to an anticipated future receipt (assessed at current prices) to arrive at its market value at an earlier date, subject to the comments at statement 20 below.” Paragraph 20 of their statement, agreed between Mr Beckett and Mr Cullum, but not agreed by Mr Clark or Mr Orr-Ewing (although for different reasons), related to hope value and said: “20. The value of the anticipation of being able to release latent value [i.e. marriage value] through early dealing with the lessee should in principle be separately identified and assessed, but this is difficult to do in practice when analysing comparable evidence.” Mr Clark rejected the desirability of separate identification, but his assessment of the deferment rate included a specific allowance for it. He said: “Whilst I consider that the expectation of the release of early profit is an important contribution to the positive returns that a purchaser will expect from the investment in a reversion, given that those opportunities are entirely dependent upon voluntary negotiations, I am not able to quantify ‘how many, when or how much’ in terms of the additional proceeds that may be realised over the life of the reversion. I have incorporated the anticipation of the release of early profit in my assessment of the deferment rate for the properties in which I am involved.” Mr Orr-Ewing disagreed: “..because the present value of the freehold is based on an assessment of the freehold vacant possession value, and extra value brought about by lease extensions is incompatible with that.”                         31. The valuation experts’ conclusions as to deferment rate were: Mr Cullum (Cadogan (Grandeden)) 3.5% Mr Clark (Cadogan and Howard de Walden) 3.5% Mr Orr-Ewing (Grandeden) 5.5% Mr Beckett (Maybury Court Residents) 6.25% Mr Clark’s percentage was reached after deducting 0.25% for hope value and the equivalent of 0.14% for transaction costs and rounding up from 3.42%. Neither Mr Clark nor Mr Cullum made any adjustment for flats as against houses as, in their appeals, headleases are in place. Mr Beckett’s percentage was inclusive of hope value, which he put at 1%, and reflected a 0.25% addition for the fact that what was being valued was flats. Mr Orr-Ewing added 0.5% for flats and a further 0.25% to reflect the length of lease remaining. He then deducted 0.75% for the low risk of obsolescence in the property with which he was concerned, bringing him back to 5.5%. Making the necessary adjustments so as to express each valuer’s deferment rate net of hope value, transaction costs and property-specific factors the figures become: Mr Cullum 3.5% Mr Clark 3.81% Mr Orr-Ewing 5.5% Mr Beckett 7%                         32. Although Mr Cullum said he had benefited from meeting the financial experts and reading their reports, he remained as cautious as he had been when he gave evidence in Arbib in placing reliance upon their analyses. He tested Mr Clokey’s opinions against his own view of the market by firstly adding a 0.5% allowance to Mr Clokey’s 1.38% risk free rate to reflect what he said was the majority view that the risk free rate may rise. He then took an equity risk premium of 4.5% – that figure having been agreed by 3 of the 4 financial experts, Mr Clokey’s beta of 0.4, and an additional 0.5% which he described as the increase needed for property specific factors to reflect the difference between owning shares and owning a freehold reversion. Finally he deducted 2%, which was his own opinion of long-term real growth, to arrive at a deferment rate of 2.18%. Such a figure, and Mr Clokey’s own net figure of 1.68%, if the 0.5% for property specific factors were excluded, was in his opinion far too low. He went on to produce worked examples, making various assumptions as to leasehold relativity, that proved such a low rate would create negative marriage value, something all the valuers agreed should not happen.                         33. Mr Cullum also said that he did not believe an investor in a freehold reversion would pay a price that reflected a deferment rate (as Mr Clokey’s figure, adjusted for growth and other property specific risks would do) that was lower than the net yield available in the residential rack rental market, an opinion with which Mr Clark agreed. Although not specifically using rack rental yields to arrive at his opinion, he commented upon Mr Dumas’s analysis on that basis and said he did not agree that this method of arriving at the deferment rate needed adjustment for property specific factors – these already being catered for in the net rental yield. He noted that Mr Dumas said the correct deferment rate lay somewhere between a gross rental yield of 4.8% and a net rental yield of 2.9%, and that the value of occupation was economically equivalent to the rental yield of housing. In cross-examination by Mr Jourdan, Mr Cullum said that he thought Mr Dumas was looking for the occupational yield rather than the investment yield, thus accounting for the difference between the published net yield of 2.4% and Mr Dumas’s opinion of 3% to 3.5% – say 3.25%. Mr Cullum produced, on day 6 of the hearing, alongside his second supplementary report of 9 June but with no explanatory note, a comparison of gross and net yields as between an owner/occupier and an investor (document RWRC1) based upon a recent letting of a flat at 59 Cadogan Square. This, and the similar one produced by Mr Orr-Ewing, has assisted in showing by how much deductions from gross to net returns will be less for owner/occupied property than for investments.                         34. In the example Mr Cullum gave, based upon a spot gross yield per Savills indices of 4.8% at the valuation date, the net investment yield became 2.45%, whereas without the deductions for management, letting commission, higher repairs and renewals costs and allowance for voids, the owner/occupier net yield became 3.66%. Although he acknowledged that this information had not determined the deferment rate for which he contended, Mr Cullum said that he drew comfort from it, bearing in mind his initial view that an investor would not accept a yield lower than that obtainable from the residential rental market. He therefore rejected Mr Clokey’s figure of 3.2% and suggested that an appropriate deferment rate was 3.6%. This was the figure for which Cadogan have contended in their appeal on Grandeden, and which he said “seems right”. He stressed in cross-examination that individual percentage adjustments for positive and negative factors must, by the very nature of the task in hand, include some “finger in the air” guesswork. In his supplemental statement of 11 June he suggested the deferment rate might be adjusted to 3.5% to reflect the fact that changes in Minimum Lending Rate (MLR) tend to be in quarter percentage point shifts. Although he had argued for a deferment rate in Arbib of 4.5%, he said that now he had had the opportunity to consider the financial experts’ evidence, it was clear that investors in those markets were willing to accept much lower returns and, furthermore, interest rates had fallen by about 0.75% between the valuation date in Arbib and that applicable to the subject property, July 2005.                         35. On the question of hope value, Mr Cullum considered Professor Lizieri’s opinions and agreed that any allowance, if following the legal arguments it was to be made, ideally should not be included within the deferment rate. It should be dealt with by way of a lump sum that would vary depending upon the length of unexpired term and should be added to the mathematically calculated value of the reversion prior to the statutory marriage value apportionment. The amount of this “option to deal” was dependent upon the amount of marriage value to be released by early dealing, and marriage value as a proportion of freehold value, he said, was greatest in mid-term of a long lease (as all the valuers agreed), and least at the very beginning and very end of the lease. The quantum of hope value could only, he said, be calculated once the freehold and existing lease values were known.                         36. From an analysis of the financial experts’ evidence Mr Clark took the risk-free rate to be 2.18% (by averaging their views, including Mr Dumas’s orally given range). He then added 2.49% as an “interim” risk premium again as an average and, as far as Mr Dumas was concerned, deducting his proffered 2.1% risk free rate from his all risks yield of 3.25%. A further 1.5% was added for other specific property factors such as risk of tenants holding over, unforeseen management issues, further legislation and illiquidity in comparison with equities. From the resultant 6.17% he then deducted the 2.5% real growth rate that he expected over the unexpired term and a further 0.25% to represent the opportunity for “early profit” on deals with the leaseholders (hope value). That gave him a deferment rate of 3.42%, which he rounded to 3.5%. He thought this was consistent with a rack rental yield of between 2.5% net and 4.8% gross Prime Central London (PCL) flats, taken from Savills published quarterly research, having regard to a comparison between investment in long-dated reversions and short-term lettings, which recognised their common characteristics as both being underpinned by growth in open market vacant possession values.                         37. In relation to the hope value deduction, Mr Clark said that an investor in a long dated reversion might achieve a release of marriage value well in excess of the value of his interest if the tenant came forward for a lease extension at an early stage (say 60 years plus still remaining). However, it was more likely that the tenant would come forward at a much later stage when the proportion of the value of the freehold reversion that marriage value represented had fallen. Thus, there was a balance to be struck between lower probability and higher returns as against higher probability but lower returns. It was appropriate therefore to account for hope value evenly throughout the term, and the deferment rate should likewise remain the same regardless of the remaining length of term.                         38. As a separate issue, Mr Clark was alone in assuming that a deduction should be made to reflect transaction costs on the sale and purchase of the property, but as he had made that deduction from the vacant possession value (7.525% of the present value of the freehold reversion after the application of the deferment rate), it did not go directly to the deferment rate. Without such assumption, which we consider later in our conclusions, Mr Clark said the arithmetical equivalent for the deferment rate would be 3.64%.                         39. On the question of the application of growth assumptions to the deferment rate, Mr Clark said that the investor in the freehold reversion was likely to be more interested in the medium to long term growth prospects than was the owner occupier, who whilst anticipating short to medium term growth would be more driven by other factors such as location and employment needs. The investor, he said, would take a more sophisticated view as to how today’s market prices might perform over the long term and it would not, therefore, be double counting to include a factor for growth in the deferment rate. Commenting on Mr Beckett’s market evidence, Mr Clark said that he, along with Mr Cullum, considered that any transaction evidence was not conformable with the Act, and furthermore, he thought that Mr Beckett’s sample of transactions was too small even to derive a baseline rate for deferment.                         40. Mr Orr-Ewing spoke to the same figure of 5.5% as Professor Lizieri and adopted his conclusions as to the risk-free rate (2.5%), risk premium (4.5%) and growth (1.5%) to arrive at a generic deferment rate of 5.5%, which he then proposed to adjust for property specific risks in accordance with his judgment as a valuer. Whilst accepting the financial evidence (although he did not agree with the use of a spot rate for assessing the risk-free rate), he expressed his view that the CAPM approach involved reliance on a “correlation between a reversion in a single property and a quoted stock market share [where] even the closest is not very similar [which] does seem to throw the whole exercise in doubt.” Having reviewed the property specific risks, he said he would add 0.5% to the deferment rate in respect of a block of flats as opposed to a house. He was also of the view that whilst the deferment rate would remain constant for unexpired terms of 20 up to 50 years, interest in the reversion would wane where the term was longer, and in such circumstances a further 0.25% should be added for the length of the reversion (71.05 years). However, he regarded these two adjustments as balanced out by the low risk of obsolescence of the particular building, for which he made a deduction of 0.75%, and thus spoke to the rate of 5.5% for 59 Cadogan Square and 105 Cadogan Gardens.                         41. Hope value had not been included in Mr Orr-Ewing’s calculation of a generic deferment rate, partly for the reason that his instructing solicitors had advised that no account should be taken of the opportunity to release early profit, as that would be inconsistent with the assumption of vacant possession of the whole building at the end of the term. Furthermore, the headlease had a 10-day reversion, and without the assistance of the Act, the freeholder would be prevented from granting an extended lease with an uninterrupted right to possession. He said that if account was to be taken of it anywhere, it would be in the risk-free rate.                         42. Mr Orr-Ewing commented in his supplementary report on Mr Dumas’s rack rental yield approach and also produced a comparison between gross and net yields as applicable to owner/occupiers and investors, on a similar basis to Mr Cullum’s calculations (his RWRC1), but not based, he said, on any particular property. From a 5% gross yield, he concluded that deductions for an investor would amount to 34% of the rent received, giving a net investors’ yield of 3.3%. For an owner/occupier, with reduced outgoings of only 6.7% of the rent payable, the yield became 4.6%. He rejected Mr Dumas’s contention that the rental yield was equivalent to the right to occupy on the basis that it could not be right that a lessor letting out his flat and receiving a substantial income from it would require the same rate of return as the owner of a reversion. Commenting on Mr Orr-Ewing’s owner/occupier rate, Mr Cullum had said in cross-examination that he thought not enough had been allowed for repairs and renewals, and the service charge allowance was also understated, these differences accounting for the disparity between Mr Orr-Ewing’s 4.6% owner/occupier yield and his own figure on that basis of 3.66%.                         43. Addressing Mr Beckett’s market based approach, Mr Orr-Ewing said he thought that the methodology was basically sound and, as all the valuers had agreed, transaction evidence was always to be preferred if it was available. Having briefly considered the generality of Mr Beckett’s conclusions, and having analysed a small sample of the transactions he had referred to, he said that he was, in the main, in agreement with them. He also produced his own analyses of the recent (May 2006) auction sales of 6 and 7 Ovington Gardens and 2-4 Ovington Square, two properties with which he had personally been involved as joint agent. The achieved prices of each of these he dissected in two different ways to take account of the hope value which was likely to have existed in these Act-world sales. Six and 7 Ovington Gardens comprised 13 flats, 12 of which had 50.5 years remaining and one had had its lease extended by 90 years. A headlease was in place and the property sold to a private investor for £450,000. Having assessed the leasehold values by comparing with recent sales and consulting with his sales colleagues, he firstly calculated an all-risks deferment rate (to include hope value) to be 4.98% and, secondly, separated hope value out at 20% of the marriage value (based upon a relativity of 74.5% in line with Gerald Eve’s 1996 graph), which produced a deferment rate of 6.9%. The same exercise for 2-4 Ovington Square produced 4.85% and 5.9% respectively.                         44. Whilst he accepted that one auction was of little significance, such evidence, along with that provided by Mr Beckett constituted “work in progress” that so far indicated all-risks deferment rates in the range 4.85% to 7.03%. This range showed, in his view, just how wrong Mr Clark’s and Mr Cullum’s (revised) deferment rates at 3.5% were. He said that he appreciated that tenants in the market had rights under the 1993 Act, but he thought that that had not corrupted the sales evidence. If tenants did exercise their rights, that would be a benefit to the purchaser, in releasing marriage value and providing rollover relief, and that would serve to reduce the deferment rate. In the light of this additional evidence, and Mr Beckett’s range, Mr Orr-Ewing said that he felt his conclusion at a rate of 5.5% was correct.                         45. Mr Beckett did not accept what was “common ground in the [Arbib] appeals [that there was] no market in reversionary interests on terms conformable with the assumptions required by the statutes” (see Arbib para 88). He therefore set about collecting and presenting evidence from auctions of reversions, which, of course, he accepted had been transactions in the real world rather than upon the assumptions required to be made by the Act, but which he proposed with some caveats to apply directly to the ascertainment of a deferment rate, which he placed in the range of 5.5% to 7.0%. He is to be thanked for his efforts to fill the gap in the evidence in Arbib, which cut the Tribunal off from its preferred approach of a market based assessment.                         46. He considered that the conclusions he drew from the market were supported by the evidence of the financial experts, and, although he placed no special reliance on Mr Francis, he said the deferment rate should be 6.25% on the basis of his assessment of the evidence of experts whose expertise he did not claim to share. The deferment rate, he said, would be tenuously influenced by, but in no way linked to, the generality of returns on financial instruments, interest rates and the like. These changed over time, but the effects would be muted and slow in their influence on deferment rates.                         47. Mr Beckett examined altogether some 200 transactions concerning sales of freehold reversions of flats and houses, including some mixed commercial properties, mostly at auctions, which took place between 2000 and 2004. He found the data available to him in 38 of them useable to derive deferment rates in the real world, in that he had information as to the leasehold interests and the prices paid, and was able to estimate a vacant possession value, from which a deferment rate could be calculated. He capitalised the ground rent in each case at 10%, in order to avoid understating the influence of the income stream, and he chose to use only transactions where the rental income was unlikely to have been the main driver of the transaction.                         48. Discussing how the hope of marriage value interacted with the deferment rate, Mr Beckett concluded that it could be usefully subsumed in the deferment rate – its contribution to it being, he tentatively thought, 1%, meaning that, if it had to be totally excluded for the purposes of this decision, his figure for deferment would become 7.25%. If the hope value was separated out, at 1% it became approximately 20% of marriage value. However, he said he was not convinced that the Act tainted the market in freehold reversions to the extent that had been suggested. He thought it unlikely that a purchaser would pay significantly more for the freehold of a property on the basis that at some point in the future, when tenants started to want to buy lease extensions or acquire the freehold, he would be able to act as an unwilling seller and extract more than 50% of the marriage value. Moreover, actual transactions in the market seemed to be virtually clear of “contamination” by Tribunal decisions, there being a wide scatter of rates both above and below those derived from such decisions with no evidence of rates hugging them. Whilst he accepted that in the real world the tenant’s bid would be taken into account, and would serve to push up the price to a point at which the value became too high, that did not, in his opinion, preclude using the transaction as a ceiling for the value or, conversely, as a floor to the deferment rate.                         49. With regard to the exercise that Mr Cullum had undertaken to test Mr Clokey’s figures, Mr Beckett agreed with him that negative marriage value must be a sign of valuation failure. He said that there was not a scrap of evidence in the real world to suggest deferment rates as low as 3.6%, and in his view leasehold relativity was much higher than the 5% – 10% that Mr Cullum had used to get to that figure. The deferment rate and the nature of the investment                         50. It is important to be clear at the outset what it is that the deferment rate represents and the nature of the interest to which it is to be applied. We have identified the three elements of value in the landlord’s interest that need to be considered – the right to receive the ground rent, the right to vacant possession at term, and the option of realising a share of the marriage value by an earlier sale to the tenant. Each of these elements has to be valued separately. The future income stream of the ground rent and the postponed realisation of vacant possession are such separate elements that their separate valuation is obviously appropriate. The right to vacant possession at term and the option of an earlier sale to the tenant are more closely linked, although they are distinct elements and are capable of separate valuation. It is in any event unavoidable that they should be valued separately since it is in dispute whether as a matter of law the latter, hope value, falls to be included, and we are therefore obliged under rule 50(4) of the Lands Tribunal Rules 1996 to give alternative values, with and without hope value.                         51. The deferment rate is an annual discount of a future receipt, the vacant possession value of the house or flat at term. It incorporates a rate to compensate for deferment of enjoyment, what in the course of the hearing was referred to as “the time-preference rate”, for which the risk-free rate was a proxy. The purpose, however, is ascertainment of the present value of an asset that consists, and consists only, of the right to vacant possession of a particular residential property at the end of the lease to which the freehold is subject. In the present cases the leases had 21, 41, 23, 71 and 57 years respectively to run. The asset is a hypothetical one in that, in the real world, a reversion that does not carry with it an entitlement to ground rent and the opportunity to make a profit by selling the interest to the tenant is never in practice encountered. As was pointed out in Arbib (paras 88 and 148) the deferment rate is a valuation tool to enable the vacant possession value as at the valuation date to be used for the purpose of arriving at the present value of the freeholder’s reversionary interest.                         52. It is implicit in this valuation process that there would be a market for such assets if they existed, and we are satisfied that there would indeed be a market. The nature of the investment being assumed may be analysed thus: the value of the asset consists of its prospect of appreciation; it will appreciate through the lapse of time as the term date gets nearer (inherent growth); if, however, the vacant possession value of the property increases in real terms, the reversion will appreciate through real growth in the same way as the property in possession. On the other hand, set against the relatively secure long-term nature of the investment that the asset represents are the risks and disadvantages that are associated with it. There is volatility in the market in residential property, and there are prolonged periods of downturn. The property is illiquid in the sense that any sale will take time to achieve. While it is a tradeable asset, therefore, the reversion is subject to the risk that a sale may only be achievable after a delay and at a time when the market is low. It is an asset also that may become obsolescent and deteriorate physically.                         53. In seeking to establish a deferment rate, the experts have approached the matter by establishing a generic deferment rate applicable to long-term residential reversions generally before adjusting this, as necessary, to the individual property under consideration. We shall approach the matter by considering first the generic deferment rate and then turning to specific factors, both ones that impact on the generic deferment rate and ones that may apply so as to require its adjustment in relation to individual properties. Deferment rate: methodology                         54. We were presented by the evidence with what were essentially four different methods of reaching the deferment rate. These were as follows: (a) A form of financial valuation known as CAPM. This was the method used by three of the financial experts, Mr Clokey, Mr Francis and Professor Lizieri. It can be expressed by the basic formula DR = RFR – RGR + RP (deferment rate equals risk-free rate minus real growth rate plus risk premium). CAPM assesses RP by applying to the equity market risk premium (EMRP) (the addition for risk that an investor in equities generally would make) a factor, known as beta, that reflects the comparative riskiness of the particular investment under consideration. This would normally be an individual company, but here has to be long-term residential reversions as a whole. (b) Determination by reference to rack rental yields. This was the method adopted by one of the financial experts, Mr Dumas. (c) Derivation from an analysis of market sales of long-term residential reversions. This was the method as adopted by one of the valuer witnesses, Mr Beckett. (d) As in (a) above, but, instead of reaching RP by the application of beta to EMRP, making an independent assessment of what the RP should be. This was the method adopted by this Tribunal in Arbib, and it was the method followed by the other valuer witnesses, Mr Cullum, Mr Clark and Mr Orr-Ewing, and Professor Lizieri used this as an alternative method. 55. The conclusion we have reached is that the first three methods, though they are in some respects instructive, have deficiencies that render them inappropriate for use, and we shall explain why before dealing with method (d), which we adopt. We should note that Mr Cullum also employed, as a check only, what was referred to as the “back-to-front” method, which reaches a conclusion on the deferment rate by inferring the value of the reversion from the relativity of freehold and leasehold values. It is sufficient for us to say that we do not find this persuasive. If it were thought to be a reliable method, there would, of course, be no need of a deferment rate, since the reversion would have been valued already. But in reality it is not reliable since it relies on assumptions as to the marriage value, which can only be determined once the deferment rate is known, and as to the relativity of the freehold and leasehold values, which is not established. Methodology (a): CAPM                         56. Three of the four financial experts employed CAPM. Mr Clokey took an equity market risk premium of 4.5%, a beta of 0.4, and thus arrived at a risk premium of 1.8%. Professor Lizieri took an EMRP of 4.5%, a beta of 1.0, and reached a risk premium of 4.5%. Mr Francis took an EMRP of 5% and a beta of 0.5, giving a risk premium of 2.5%. The widely differing end results were thus primarily due to the choice of beta. On the EMRP the difference (between Mr Francis and the other two valuers) was small, and Mr Dumas, although he did not use CAPM, agreed with the other two valuers on 4.5%. We think it appropriate in the light of this measure of agreement to adopt this as the EMRP.                         57. Assessing the risk premium by the application of CAPM requires the choice of “beta”, which, since it operates as a multiplier of the broadly agreed EMRP, is crucial to the result. It depends on a judgment of the comparability of an investment in a residential freehold reversion to “a portfolio of investments or assets”. The choice of beta was the most important point of disagreement amongst the financial experts. We do not doubt the conscientiousness with which each of them sought to make that comparison nor the care with which the valuers have reviewed the businesses upon which the financial experts relied, but the difficulty is that the exercise involves the comparison of unlike investments.                         58. The choice of beta for the purpose of applying CAPM requires the selection of the returns of a single company, or a range of companies, as sufficiently representing the return from a quite different asset, namely a single property investment. Although the details of the business of the public companies selected are more transparent than was the case of the Freehold Investment Trust considered in Arbib (paragraphs 135 to 137), ultimately the problem is the same: it involves treating a business as equivalent to an investment in an asset with quite different characteristics, for example, of liquidity. It is in our judgment as impossible to arrive at an appropriate deferment rate for freehold reversions by seeking to assess the risk premium attaching to companies whose business includes some dealing in such reversions as compared with the equity market as a whole as it proved to be for the Tribunal in Arbib to draw conclusions from the price of a “financial product”. We therefore reject reliance on CAPM as a direct means of assessment of the deferment rate.                         59. CAPM has nonetheless helped in our understanding of the nature of the deferment rate. Also, as an indicator of the returns required by the equity market as a whole, it provides assistance for the assessment of the rate. Adjusting the EMRP to reach a risk premium appropriate to residential reversions lies at the interface between the expertise of the financial experts and the valuers; and, while we note the apparent consensus of judgment amongst the financial experts that the return from equities would be unlikely to be less than the return from residential freehold reversions, we do not regard this view as conclusive. It is merely one factor to be borne in mind, together with the valuers’ evidence of the particular risks associated with investment in long-term reversions and such conclusions as can be drawn from market transactions, in reaching a judgment on the risk premium. Methodology (b): rack rental values                         60. On the face of it Mr Dumas’s identification of rack rental yield with compensation for not enjoying possession provides a conceptually elegant means of assessing the deferment rate. It is however incomplete in that the value of possession, in terms of control, opportunity to deal and even pride of possession are all elements in the difference between the value of a reversionary interest and an interest in possession, which are not reflected in the rack rental value. These factors would appear to indicate that at least evidence of the rack rental value would provide a minimum for the deferment rate, and Mr Cullum thought that it would serve that purpose. But it is clear, we think, that rack rental values can be seen by the way that they move in relation to vacant possession values to be unrelated to the deferment rate. When the residential property market is high (that is, above the historic long-term trend) rack rental values fall, and they rise when the market is low. The deferment rate, on the other hand, being related to the prospect of long-term growth, would (if there were no counter-balancing-effects – see below paragraph 73) tend to do the opposite. It would fall when the market was low, because the prospect for long-term growth would be improved, and would tend to rise when the market was high.                         61. Rack rental yields generate a market that is quite different from the hypothetical market in long-term reversions that we have to consider. The focus is essentially relatively short-term, with the different risk considerations that apply in consequence, and we can see no comparison that would enable rack rental yields to be usefully employed in establishing the deferment rate. Methodology (c): market evidence                         62. In Arbib (paragraph 100) the Tribunal concluded that: “The absence of transactions which can be used to derive yields from investments in freehold reversions, whatever dispute there might be as to analysis, is, we think demonstrated by the failure of the valuers in these appeals to identify comparables whose facts could usefully be agreed, subject to analysis”. This led the Tribunal, after reviewing such evidence as was adduced, to say at paragraph 109: “We are grateful to the witnesses who have done their best to produce market evidence which might have been of assistance. Their failure to do so confirms us in the view that there is none, and the circumstances which now exist make it virtually impossible that there can be any.”                         63. Mr Beckett doubted the correctness of this conclusion, and he carried out an analysis of a large number of market transactions as summarised above. For the most part his assessments of value have limited the reliance on leasehold transactions in order to arrive at vacant possession values, thus avoiding assumptions as to relativity, which produce a circularity. In any case, he makes the point that a 10% error in the vacant possession value would, on a 50 year reversion (which was the average with which he was dealing), produce only a 3% error in the deferment rate.                         64. Although criticism was levelled at Mr Beckett’s evidence on the basis that the factual evidence was uncertain – for instance in relation to the terms of the leases and the factual circumstances of each transaction – and indeed Mr Beckett himself gave significance ratings to the transactions that did not suggest a high degree of confidence in their reliability, we are satisfied that the material is sufficient to establish the range of relativities in the with-Act world between vacant possession freehold values and the prices in fact paid for reversions of different lengths. However, to treat this factor as a deferment rate applicable to the valuation of reversions in the no-Act world would, in our judgment, be mistaken. The market in the real world is substantially different from the one to be envisaged in the hypothetical no-Act world.                         65. For the purpose of determining the deferment rate we have to consider a reversion that can be held until the end of the lease and can be sold at any time with this potential. Its attraction, therefore, is as a secure long-term investment, and its value consists entirely in the vacant possession value to be realised at the end of the lease. In the real world, by contrast, vendor and purchaser know that, because of the Acts, the reversioner can have no expectation that he will be able to retain the reversion until the end of the lease. Indeed the likelihood is that it will be enfranchised at some point, possibly at an early date. What drives the market, in consequence, is not the perception of the value of a secure long-term investment but the expectation of profit through the early realisation of marriage value. We consider that these differences are so fundamental as to mean that the market evidence cannot be used as the basis for calculating the deferment rate. This was the view that the Tribunal took in Arbib, and it is one that in our own minds is confirmed now that we have had the benefit of Mr Beckett’s outstandingly thought-provoking evidence.                         66. A second reason for rejecting the market evidence is that the market is inevitably influenced by expectations of what an LVT might determine as the enfranchisement price under the statutory assumptions. Although there is a wide scatter in the results deduced from the transactions by Mr Beckett, certain of the results are weakly suggestive of a decline in the percentage following the decisions of the Tribunal in Pockney and Arbib. It would be surprising if such an effect did not exist, and it might be expected to be to be stronger in the PCL area, with which both those decisions were concerned, than elsewhere. Mr Beckett pointed out that, contrary to his expectations, the results did not show any difference in the rate according to location. Since, as he showed, LVTs, particularly in London and the south-east, made substantial adjustments to the rate for location, this suggested, he said, that LVT decisions did not influence the market. It may be that other considerations are more influential than the expectation of what an LVT would decide, but the possibility, indeed, as we think, the likelihood, of there being such an influence constitutes a further reason for rejecting an approach based on the market evidence.                         67. A third reason why it is not possible to rely on the market evidence gathered by Mr Beckett in order to calculate a deferment rate is that the comparison has to be of the whole price paid for the reversion as shown by the market evidence with the vacant possession value. That must include whatever is paid for hope value, which, as we have pointed out, is likely to be the prime factor driving the market. That will be a speculative part of the vacant possession value, determined as a share of the marriage value in accordance with the relevant statutory provisions. Although, as we have concluded for reasons set out below, such hope value is to be disregarded except in a case where section 9(1A) of the 1967 Act applies, use of the market evidence results in its being used to determine the rate for the deferment of the full vacant possession value. Notionally, of course, the prospect of early profit ought to enhance the price paid for the reversion, so that, as Mr Beckett claimed, the factor calculated from such market evidence should provide a minimum figure for a deferment rate for the purpose of a valuation under the Acts. But in our judgment Mr Beckett’s calculation is not only rendered unreliable, because it is not the same thing that is being calculated, but it is inevitably distorted as a result. Deferment rate components: risk-free rate                         68. The first component of the deferment rate that we need to establish is the risk-free rate. We have said above (paragraph 16) what it represents. We have recorded at paragraphs 18 and 19 the different figures at which the financial experts have arrived for a risk-free rate and the basis on which each of them did so. Mr Clokey took 1.4%, Mr Dumas 2-2.2% and Professor Lizieri and Mr Francis 2.5%. Despite these differences they were in agreement that the risk-free rate is proxied by the return on a Government security held to redemption. There was also unanimity that inflation was best stripped out by reliance on zero-coupon (that is “a debt instrument that does not make coupon payments”) index-linked gilts. Index-linked gilts have, however, been issued only since 1983, and, although there has been a market off and on in “splits” indicating the zero-coupon rate since 1986, the experts are agreed that the information is inadequate to provide a clear single figure for a period commensurate with the terms for which a risk-free rate is to look forward.                         69. Mr Clokey was unconcerned by this. He maintained that, in applying CAPM to make a valuation at a particular date, the spot rate yield at that date for a term commensurate with the term of the asset to be valued should be chosen. That may well be persuasive for such purpose. We, however, have rejected the CAPM method for the reasons summarised at paragraphs 56 to 59 above.                         70. It seems to us that, in order to arrive at a suitable long-term rate which is to be used as a valuation tool to derive the value of the reversion from the current vacant possession value: (a) It is unnecessary to match the length of the record to the length of the unexpired term in order to arrive at a representative long-run risk-free rate; (b) It is however necessary to choose a period sufficient to avoid, or to be capable of allowing for, distortions such as the so-called “pension panic” which three of the experts agree is presently distorting the bond market; (c) On the other hand if one allows the inadequacy of pre-1983 market information to justify reliance on “undated consols” as proposed by Mr Dumas, one introduces new distortions, as Professor Lizieri identified, in equating the effect of inflation observed retrospectively with the market expectation of inflation, which is what is material for the purpose of assessing the risk-free rate. (d) The choice of period for assessing the risk-free rate should be as long as is necessary for identifying from self-consistent information a steady long-run rate. We therefore agree with Professor Lizieri that the apparent decline in yields from long-term gilts should be compared with the more consistent rates for 10-year gilts, before drawing conclusions as to the risk-free rate. We regard his conclusion that: “the average index linked yields on a five year rolling basis over the last decade suggests a more stable market based risk-free rate would be in the range 2.0% – 2.5%” to be more helpful than assessments of the cost of capital by regulatory authorities upon which he also partially relied and upon which, it seemed to us, Mr Francis placed excessive reliance. The improved evidence available to us as compared with that produced in Arbib leads us to the conclusion that the risk-free rate should be taken as 2.25% rather than the 2% assumed in Arbib. Deferment rate components: real growth                         71. Real growth in house prices since 1953 has been at an average rate of 2.9%. None of the experts, however, suggested that the deferment rate should be assessed on the assumption that this rate of growth would continue over the sort of periods that we are considering. Of the financial experts, Mr Dumas took 2.5% on the basis of the long-run GDP growth trend and Professor Lizieri 1.5% in the light of a number of factors that he spelt out. Neither Mr Clokey nor Mr Francis advanced a figure. Of the valuers, Mr Cullum took 2%. Mr Clark 2.5% and Mr Orr-Ewing used Professor Lizieri’s 1.5%. Mr Beckett did not put forward a figure.                         72. Two features of the real growth rate assumption, in our judgment, require particular consideration. The first is the very substantial difference over a long period that can arise from different rates. A house worth £1m now would be worth £3.437m in real terms after 50 years at a growth rate of 2.5%. At 1.5% it would be worth £2.105m. It is clear that an investor who based his assessment on a particular rate would be conscious of the consequences of this rate failing to materialise, and we can see that for this reason there may well be an interaction between the growth rate assumption and the third component of the deferment rate, the risk premium. Of course the investor would also be conscious of the additional gain that would result from the assumed rate being exceeded, but the likely type of investor in a long-term reversion producing no income would be someone looking for long-term security – Mr Cullum had pension funds particularly in mind – and it is unlikely that, for such an investor, the hope that the assumed rate might be exceeded would outweigh the fear that it might not be achieved, unless the assumption as to growth was pitched at a sufficiently conservative level. In general we see force in the evidence of Professor Lizieri in the light of the factors that he referred to, but we think a realistic, or neutral, assumption would be 2%, the one made by Mr Cullum, with any concern on the part of the investor that this rate might not be achieved being reflected in the risk premium.                         73. The second feature of the assumption that has to be considered, in our judgment, is that real house prices fluctuate significantly, so that at any one time they may be substantially above or substantially below the level that would have resulted from a projection of past average growth rates. Thus at the present time the recent sustained strength of the market has pushed prices well above the long-term trend. To base the deferment rate on the same real growth rate when prices are above the trend as when they are below the trend would run the risk of overvaluing the reversion in the former case and undervaluing it in the latter. This was not, however, a matter to which evidence was directed. Nevertheless it seems to us likely that the optimism of a buoyant market would be likely to feed through, to some extent, to investors in the sort of long-term reversions that we are considering, and that the same would be true when the market was low and pessimistic. It could therefore be expected that this would be reflected in a reduction or increase, as the case might be, in the risk premium, which would tend to counterbalance any adjustment that might otherwise seem appropriate to the growth rate. It is a reasonable working assumption, we think, that there would be an effective counterbalancing, so that the deferment rate would not need to change according to whether prices were above or below the trend. We therefore adopt 2% as the assumed growth rate. Deferment rate components: risk premium                         74. In Arbib (para 150) the Tribunal treated the risks inherent in an investment in a freehold reversion as only marginally offset by the prospect of growth then identified. Here, we have been able to arrive at a more confident assessment of the growth rate to be assumed and, subtracting the 2% that we have adopted for real growth from the risk-free rate, determined at 2.25%, leaves 0.25% to be added to such assessment of the appropriate risk premium as we are able to make.                         75. Having rejected the use of CAPM, it is necessary for us to assess the risk premium by a consideration of the individual components of the risks of investment in long reversions (volatility, illiquidity, deterioration and obsolescence) and to form an overall assessment of the premium that would be required by investors in the type of asset that we have to consider. This was in fact the alternative approach that Professor Lizieri adopted in support of his CAPM conclusions. We have stated above what we see as the nature of the investment, and we have expressed our satisfaction that, were such assets to exist, there would be a market of substance for them. Of the components to be considered, physical deterioration and obsolescence are factors that require to be reflected in the generic deferment rate to the extent that the risk related to them is common to all residential property viewed in the long term. To the extent that they are property specific we consider them further below.                         76. It is, in our judgment, the combined effect of the other components, volatility and illiquidity, that must have the major impact on the risk premium. If the market was composed, or contained a substantial number, of people who intended to hold the reversion to term the fluctuations in residential property prices and the illiquidity of the investment would have very little influence. The investor would simply lock away the investment, and the passage of time would iron out the fluctuations. The illiquidity would have no influence because the investment would not be sold. Mr Cullum’s assessment was based, it appears to us, on the assumption of a market very much of this sort, and he identified pension funds and the great estates as the likely purchasers. We do not, however, accept that in the market that we have to envisage there would be any significant number of investors who would be looking to hold these very long term assets throughout their lives. The attraction of the investment would be its relative security, the prospect of growth and the opportunity for both long-term retention and earlier sale. Tradeability would, we think, be important as one of its components, and it is this that would make the volatility of the housing market and the relative illiquidity of the investment significant factors in the mind of a purchaser.                         77. In this assessment of the characteristics of the market and the factors that would influence it we therefore prefer the evidence of Professor Lizieri, which we have summarised earlier. We think that the landlord’s witnesses substantially underestimated the risks. We agree with Professor Lizieri that, since real house prices are shown to be prone to shocks and to be strongly cyclical, with persistent periods of negative growth, an investor in a long-term reversion would be very conscious of the risk that the market could be depressed at the point at which he wished to sell his interest, even though, as compared with equities, the residential property market is rather less volatile. Reversions would suffer in comparison with equities from illiquidity resulting from high transaction costs and the length of time to complete a transaction, and the latter factor would, we think, be perceived as adding substantially to the risk associated with volatility.                         78. As investment comparators we were offered long gilts, equities, individual property companies and the Act-world market in reversions. The first is the base-line from which the further assessments relating to real growth and the risk premium can be made. We follow this course. Equities are a different sort of investment, but, provided the differences are borne in mind, we accept the views of three of the financial experts that they form a useful guide to the risk premium that we have to establish. We can derive little from the individual property companies. The market evidence, though related to investments bought for shorter term profit and subject to the risk of compulsory purchase at any time, is, we think of some use to the extent that it suggests that yields in the longer term, more secure, investments that we have to consider would be substantially lower than those derived from the market transactions.                         79. With these particular considerations and the totality of the evidence in mind we have reached the conclusion that Professor Lizieri’s assessment of the risk factor is about right. Our conclusion is that the market in such investments would require a risk premium of 4.5%. Overall, therefore, we would take a risk premium of 4.5%, in combination with a risk-free rate of 2.25% and a real growth rate of 2%, producing a generic deferment rate of 4.75%. Specific factors                         80. So far we have concentrated in identifying what we have called a generic deferment rate, one that is applicable to long-term residential property reversions in general. We must now consider whether such a generic rate needs to be adjusted in any particular case for specific factors. It is to be borne in mind that the valuers’ evidence has been directed at establishing the appropriate rate for the particular reversions that fall to be valued in the present appeals and that these are flats and houses in the PCL area. We have, however, had the benefit of the answers that they and the financial experts gave to questions of more general application, and we draw on this in what follows. Specific factors: length of term                         81. Mr Clokey took the longest dated index-linked bond as representing the estimate of the risk-free rate for the terms of over 30 years that he was asked to consider for the purpose of the appeals with which he was concerned. However, he said, yield curves showed the gross redemption yield falling at progressively longer maturities. He noted also that the recently introduced 50-year gilt had experienced yields lower than those on the 30-year index-linked gilt. In addition the Government’s Green Book showed discount rates declining for periods beyond 30 years. His assumption of a rate that, at the date of valuation, was level beyond 30 years was thus, he said, a conservative one since there were good reasons to argue for longer terms equating to lower risk.                         82. Mr Dumas considered that a single deferment rate would hold good for terms of 20 years and over, but for terms shorter than that it would require adjustment to reflect the position in the property cycle at the time of valuation. Mr Francis thought that there would be no variation according to the unexpired term of the reversion. Professor Lizieri said that below 20 years the deferment rate would need to be assessed by reference to market data for the appropriate matched maturity and to include an explicit real residential capital growth forecast. These would vary over time. Where the length of the reversion exceeded the period for which reliable and robust market evidence of gilt yields was available, ie in excess of 30 years, he assumed that investors would require an additional premium on the risk-free rate to account for future uncertainty on yields. As a result the deferment rate would curve gently upwards.                         83. Of the valuers, Mr Cullum considered that the deferment rate was constant between 5 and 75 years. Mr Clark also considered that the rate was constant between 5 and 75 years, except to the extent that it was appropriate for it to reflect transaction costs. Mr Orr-Ewing thought that beyond 50 years 0.25% ought to be added to the deferment rate to reflect a falling-off in investor interest, as evidenced in recent auction sales. Below 20 years he agreed with Professor Lizieri that the deferment rate would vary according to short-term forecasts for growth and market data on risk-free rates. Mr Beckett considered the length of term in relation to three groups of terms unexpired: 0-20 years, 20-60 years and 60-plus years. He thought that 0.25% should be deducted from his 6.25% for the first group and that 0.25% should be added for the third group. In each case the adjustments reflected the different prospects of realising marriage value.                         84. The evidence of the financial experts does not persuade us that there is any need to assume other than a constant deferment rate for terms in excess of 20 years. Both Mr Dumas and Mr Francis thought that it was constant. Mr Clokey suggested that longer terms might equate to lower risk, and Professor Lizieri thought the opposite, but we do not regard either of these views as sufficiently firmly based to justify a conclusion that the rate would vary. Of the valuers, both Mr Cullum and Mr Clark considered that the rate was constant between 5 and 75 years. Mr Orr-Ewing’s view that it was higher beyond 50 years was based on an inference drawn from auction sales, but this we regard as wholly unpersuasive in view of the particular nature of the market, which we have discussed above. Mr Beckett’s variability was related to hope value, which, as we conclude below, we must leave out of account in the assessment of the deferment rate.                         85. Our conclusion is that the deferment rate is constant beyond 20 years. Below 20 years we accept the view of Mr Dumas, Professor Lizieri and Mr Orr-Ewing that the rate would need to have regard to the property cycle at the time of valuation. Beyond 75 years we see no reason on the evidence before us to conclude that the rate would be either higher or lower. Specific factors: location                         86. We invited the witnesses to say whether they thought that the deferment rate would vary according to the location of the property. Of the financial experts Mr Clokey thought that there would be no such variation as beta was applied to reflect the risk relating to a portfolio of investments. Mr Dumas and Professor Lizieri both said that the real growth rate would not vary between locations, since the compounding effects of different growth rates would eventually lead to differences in value that were unsustainable and would accordingly correct themselves. Professor Lizieri said that the convergence effect applied most clearly within an urban area or region but would apply more generally across regions. Of the valuers Mr Cullum said that locational considerations did not affect the application of the deferment rate to the appeal properties. He noted, however, that there were regional variations within national figures relating to house price growth. Mr Clark said that he agreed with the Tribunal’s conclusions in Arbib (para 185) to the effect that the comparative risk of obsolescence, volatility in values or reduced growth rates were accounted for in the deferment rate, with other purely locational differences being reflected in freehold vacant possession values. His area of expertise was central London, but he could very well imagine that different deferment rates might apply in different locations in London as well as in other parts of the country as compared to central London, having regard to local prospect for growth and local factors affecting risk.                         87. Mr Orr-Ewing and Mr Beckett both said that they would expect location to make a difference. Mr Orr-Ewing said, however, that he could not provide any statistical justification for such differences, and Mr Beckett said that the transactions that he had analysed, which included properties in Dagenham, Hampstead, Earls Court, Catford and Hillingdon in addition to Mayfair and Chelsea, did not show any difference according to location.                         88. While we accept the view of the valuers that the deferment rate could require adjustment for location, on the evidence before us we see no justification for making any adjustment to reflect regional or local considerations either generally or in relation to the particular cases before us. The evidence of the financial experts suggests that no adjustment to the real growth rate is appropriate given the long-term basis of the deferment rate, and locational differences of a local nature are, in the absence of clear evidence suggesting otherwise, to be assumed to be properly reflected in the freehold vacant possession value. Specific factors: obsolescence and condition                         89. Of the valuers Mr Cullum considered that the age of the property and structural issues were likely to be reflected in the vacant possession value. Mr Clark said that he was not averse to varying the deferment rate between buildings when the circumstances warranted an adjustment. For instance a dilapidated building where the covenants in the leases and the service charge regime were insufficient to ensure that the building remained in good condition could justify a higher deferment rate. Mr Orr-Ewing considered that, while age, construction, condition and type of building were factors that were reflected in vacant possession values, they might also result in different exposure to the risk of depreciation and obsolescence. In relation to 59 Cadogan Square and 105 Cadogan Gardens, he made a deduction of 0.75% (balancing additions for flats and length of reversion) to reflect the low risk of obsolescence of the particular building.                         90. Professor Lizieri said that, beyond the generalised risk of obsolescence, building quality and condition had an impact on obsolescence and depreciation, and both of these should be in the vacant possession value. There could, however, be minor, property-level adjustments to the deferment rate to account for differential risk of depreciation and obsolescence, with some buildings being more prone than others to diminution in value, but that was a matter outside his expertise.                         91. As with location, while we do not rule out the possible need to adjust the deferment rate to take account of such matters as obsolescence and condition, we think that it would only exceptionally be the case that such factors were not fully reflected in the vacant possession value and the risk premium. Evidence would be needed to establish that they were not fully reflected in this way. Although Mr Orr-Ewing made a deduction in respect of 59 Cadogan Square and 105 Cadogan Garden to reflect what he said was the low risk of obsolescence, no such deduction was made by Mr Clark on behalf of the freeholders, and we can see no justification for doing so. Specific factors: flats v houses                         92. In Arbib the Tribunal concluded that there should be 0.25% differential between houses and flats (increasing the deferment rate for the latter) having regard to the lesser management problems of a single house and the possibility that there might be a better prospect of growth in the house as opposed to the flat market. The Tribunal also determined that it was not necessary to assume a different risk factor for the reversion upon a single flat compared to a block of flats. The relativity between growth in house prices against flats was not a matter specifically in issue in this case, but was dealt with by the experts in answer to questions raised by the Tribunal during the course of the hearing. However, it was one of the elements that had led the Tribunal in Arbib to its conclusion, and is of course a factor to be taken into account in determining the generic deferment rate.                         93. Mr Beckett and Mr Orr-Ewing both said that an upward adjustment of 0.25% was appropriate for flats, due to the inherent increased management liabilities. Mr Orr-Ewing also placed reliance on the fact that there had, according to Savills PCL index, been lower capital growth of flats over houses for a considerable period of time. A graph reproduced in his report showed houses growing at consistently higher rates since 1980, and especially in the late 1990s, and he applied an additional 0.5%.                         94. Mr Clark said that the growth differential that is currently evident in the flats and houses market in central London was likely to even out in the long term – say 20 years plus. Although for the past 13 years the Savills PCL flats and houses indices had shown house values to be growing more than flats, there had been earlier periods when demand for flats was perceived to be higher. No adjustment to the deferment rate was therefore appropriate to reflect the growth factor, he said, and Mr Cullum expressed the same view. Mr Clark said that for the blocks with which he was concerned in these appeals there would be no additional management problems sufficient to warrant an adjustment to the deferment rate. This was because in all the properties there was a headlease and a comprehensive service charge regime was in place. He accepted, however, that where a freeholder managed a block directly, or where there was no headlease, there could be problems, but he thought that an upwards adjustment of no more than 0.125% was justified in such cases. Mr Cullum thought that a 0.25% rise would be appropriate only in the case of flats where there was no headlease in place.                         95. In Arbib the adjustment of 0.25% was intended to reflect both the greater management problems associated with flats and the possibility that there might be a better prospect of growth in the house as opposed to the flat market. As to the second of these factors we accept Mr Clark’s view that any disparity between growth rates for houses and flats is likely to even out over the longer term. We think, however, that an adjustment needs to be made to reflect the management problems, although we do not consider it appropriate to differentiate between flats that are the subject of headleases and those which are not. Nor do we think that the management concerns are necessarily so much less for a single flat than for a block to warrant a different adjustment. Even where flats are efficiently managed, service charge and repairs problems inevitably occur, and the management exercise in itself is, we feel, sufficiently more complex to warrant a generalised 0.25% addition for flats. We do not consider that any fine-tuning below this percentage is justified.                         96. Because what we are considering is a long-term investment it is the prospect of management problems arising during the course of the tenancy that is the important consideration rather than the state of affairs at the time of valuation. Our view is that the potential for problems to arise is inherent in all leases and that standard adjustment is therefore appropriate. We do not rule out the possibility that there could be a case for an additional allowance where exceptional difficulties are in prospect, but this would need to be the subject of compelling evidence. Specific factors: transaction costs                         97. To reach his deferment value Mr Clark made a deduction from the vacant possession value to reflect transaction costs on the sale and purchase of the property. Although no doubt, in comparing a freehold reversion with an investment in equities, the cost of management and dealing will affect the relative rate of return, Mr Clark’s approach must we think be wrong in principle, as a substitute for valuing the freeholder’s reversionary interest by using a deferment rate to discount the vacant possession value. There can, in the first place and in any event, be no justification for a deduction to reflect the cost of purchase as opposed to sale, and Mr Clark was unable to propose one save the practice of his firm in valuing investment portfolios. Secondly, since the statutory requirement is to arrive at “the amount which … the interest might be expected to realise if sold on the open market”, the requirement is to arrive at the price to be paid, not that price less deductions for the cost of sale. This indeed makes practical sense, since in regard to the actual transaction for which the valuation is required, namely the collective enfranchisement by the respondents, the costs of enfranchisement are borne by Maybury under section 33 of the Act of 1993. Hope value: the law                         98. Hope value, as the term has been used in the present appeals, consists in the option that the freeholder has to sell the freehold or a lease extension to the tenant and thus realise the whole or part of the freeholder’s share of such marriage value as exists at the date of the sale. The issue between the landlords and the nominee purchasers is whether, as the nominee purchasers contend, hope value is excluded under the terms of Schedule 6 to the 1993 Act.                         99. Paragraph 3 of Schedule 6 to the 1993 Act provides: “(1) Subject to the provisions of this paragraph, the value of the freeholder’s interest in the specified premises is the amount which at the valuation date that interest might be expected to realise if sold in the open market by a willing seller (with no person who falls within sub-paragraph (1A) buying or seeking to buy) ….” A person falls within sub-paragraph (1A) if he is: “(a) the nominee purchaser, or (b) a tenant of premises contained in the specified premises, or (ba) an owner of an interest which the nominee purchaser is to acquire in pursuance of section 1(2)(a), or (c) an owner of an interest which the nominee purchaser is to acquire in pursuance of section 2(1)(b).” The interests referred to in (ba) and (c) are interests in other parts of the premises or in appurtenant property and other interests in the premises themselves.                         100. The contention for the landlords was that the exclusion contained in the words in brackets in paragraph 3(1), while it excluded bids from tenants on the valuation date, did not exclude the possibility of tenants seeking voluntary lease extensions after the valuation date from the new notional owner. The nominee purchasers said that the purpose and effect of the provision was to exclude the tenants from the market for all purposes of the valuation. They said that the price payable for the freehold, as provided for, contained two elements, the value of the freeholder’s interest, to which paragraph 3 related, and marriage value, provision for which was contained in paragraph 4. If the freeholder were able to include hope value in the value of his interest under paragraph 3, this would introduce a substantial element of double counting.                         101. In our judgment the extent to which hope value is excluded by the statutory provisions is most helpfully considered in relation to all the relevant provisions and the sequence in which they have been made. The exclusion of the tenant as an assumed bidder appears in three places in the legislation – section 9(1) of the 1967 Act as amended by the Housing Act 1969, paragraph 3 of Schedule 6 to the 1993 Act and paragraph 3(2) of Schedule 13 to the 1993 Act. Section 9(1) as originally enacted provided: “(1) Subject to subsection (2) below, the prices payable for a house and premises on a conveyance under section 8 above shall be the amount which at the relevant time the house and premises, if sold in the open market by a willing seller, might be expected to realise….” Section 82 of the Housing Act 1969 (which came into force 18 months after Part I of the 1967 Act) inserted after the words “a willing seller” the words “(with the tenant and members of his family who reside in the house not buying or seeking to buy).” The words “who reside in the house” have since been repealed to accord with the removal of the residential qualification for purchase. The insertion effected by section 82 followed two Lands Tribunal decisions, Custins v Hearts of Oak Building Society (1969) 209 EG 239 and Haw v Peek (1969) 210 EG 347, which held that under the terms of section 9(1) the tenant himself was in the market to buy the freehold so that his overbid had to be taken into account in ascertaining the purchase price. In other words, marriage value was part of the price. The purpose of the insertion, quickly made after these decisions, was, clearly, in our judgment, to exclude any element of marriage value from the purchase price.                         102. Since, however, the valuation under section 9(1) is as “if sold in the open market” the exclusion of the tenant’s bid at the valuation date excludes hope value (that is the value of the prospect of some future tenant’s bid) only if the assumption required by the words in brackets applies not only at the relevant time but also at any future time and, furthermore, applies not merely to “seeking to buy” the freehold but also to seeking any such interest as would have the effect of including in the valuation the special value which it was the clear purpose of the amendment to exclude. If, however, the provision is not so construed, it would be negated because it would have to be assumed that, although the tenant was not buying on the valuation date, the purchaser could assume in accordance with the facts that he would wish to buy the next day. In our judgment the amendment to section 9(1) of the 1967 Act must be construed so as to achieve its undoubted purpose, and it does exclude all hope value as well as marriage value.                         103. The 1993 Act dealt with marriage value in three separate cases. It inserted subsection (1A) in section 9 of the 1967 Act. This provided for the valuation of higher value houses, on terms different from those applied to the houses originally included in the 1967 Act, and it did not include the words “(with the tenant and members of his family not seeking to buy)” that had been inserted in subsection (1) by the 1969 Act. In the case of the valuation of such higher value houses, including 13 South Terrace, therefore, there is no doubt that the valuation in the open market includes the tenant’s bid.                         104. In the case of the grant of new leases of flats to which qualifying tenants became entitled under section 56 of the 1993 Act, the premium payable to the landlord included, under paragraph 3 of Schedule 13, the diminution of the landlord’s interest as the result of the grant and, under paragraph 4, a share of the marriage value realised by the grant of the extended lease. Paragraph 3(2) as there enacted provided: “Subject to the provisions of this paragraph, the value of any such interest as is mentioned in sub-paragraph(1)(a) or (b) is the amount which at the relevant date that interest might be expected to realise if sold on the open market by a willing seller (with neither the tenant nor any owner of an intermediate leasehold interest buying or seeking to buy) ….” Again the words in brackets are so manifestly derived from section 9(1) as amended of the 1967 Act that it seems to us inescapable that the same effect is to be given to them in each case.                         105. In regard to the valuation of the freeholder’s interest for the purposes of collective enfranchisement paragraph 3 of the Schedule 6 provides: “Subject to the provisions of this paragraph, the value of the freeholder’s interest in the specified premises is the amount which at the valuation date that interest might be expected to realise if sold in the open market by a willing seller (with neither the nominee purchaser nor any participating tenant buying or seeking to buy) ….” The freeholder’s interest which is to be valued is in this context the reversion of the block of flats. The words are not therefore well drawn in order to exclude from the valuation the prospect of realising marriage value by the grant of extended leases to tenants, although the marriage value so realisable in the case of participating tenants is to be shared in accordance with paragraph 4 of Schedule 6. Nevertheless we cannot accept that the intention of these words was other than to have the same effect as the similar words first used in the amendment of the 1967 Act by the Act of 1969. We therefore feel constrained to construe the words in Schedule 6 of the 1993 Act as having the same effect of excluding hope value.                         106. By the Housing Act 1996 those words as originally enacted were amended paragraph to the form set out in paragraph 99 above, not limiting the excluded special purchasers to those actually joining in the purchase but adding the non-participating tenants. In using the same formula derived, as we have seen, from the 1969 Act, there can, in our judgment be no doubt that Parliament was assuming that the words had the same effect, although apparently applied to the sale of the reversionary interest rather than the grant of extended leases. For these reasons we conclude that, in the valuation of the landlord’ interest under Schedule 6 as well as under Schedule 13 to the Act of 1993, no account is to be taken of any hope value or, save as is specifically provided, of marriage value.                         107. We were referred to a number of Lands Tribunal cases in which these provisions had been considered (in particular Becker Garden Properties Ltd v Garden Court NW8 Property Co Ltd [1998] EGLR 121; Re Shulem B Association Ltd [2001] 1 EGLR 105; Blendcrown Ltd v Church Commissioners for England [2004] 1 EGLR 143; and Arbib). We accept Mr Jourdan’s submission that they fall into two categories: those in which the point was not argued at all, it being assumed that the legislation permitted a payment in respect of hope value for non-participating tenants, and those where the point was argued but the Tribunal found it unnecessary to pronounce on the point because, on the evidence, it was held that there was no hope value. In view of this there is no need for us to analyse the decisions.                         108. Our conclusion, therefore, is that in all the appeals before us, with the exception of 13 South Terrace, where the price is to be determined under section 9(1A) of the 1967 Act, hope value is to be excluded from the valuation. Since, however, having come to this conclusion of law, we shall be required to produce alternative valuations, with and without hope value, in those cases where we have held it to be excluded (see Lands Tribunal Rules 1996, rule 50(4)), and since it falls to be evaluated in 13 South Terrace, we must address the question of how hope value is to be dealt with in the valuation. Hope value: treatment in valuation                         109. Two different ways of treating hope value in the valuation of the appeal properties were advanced by the valuers. The first involved making an adjustment to the deferment rate, and Mr Clark made a reduction of 0.25% to it to reflect the prospect of achieving an early profit. Mr Beckett’s deferment rate of 6.25%, being derived from market transactions that were driven by the prospect of achieving marriage value, was inclusive of hope value, and he tentatively suggested that if it were to be excluded, 1% would need to be added to his deferment rate.                         110. The second method of treating hope value was that adopted by Mr Cullum. He said that any allowance should be dealt with by a lump sum. The value of the “option to deal” depended upon the amount of marriage value that would be released by early dealing, and marriage value (as a proportion of freehold value) was not a constant, but was likely to be at its highest mid-term. It would therefore vary dependent upon the term unexpired, and it was thus not appropriate to apply such an allowance to a generic deferment rate. In any event, he said, market custom valued the option in that way. Mr Orr-Ewing agreed that, if it were to be accounted for (and he had not done so), it should not be by adjustment to the deferment rate. He said that the underlying assumption in the calculation of the deferment rate was that the landlord would obtain vacant possession of the property at the end of the term, and the attribution of value to reflect the possibility of granting extended leases to tenants was incompatible with that assumption.                         111. We agree with Mr Cullum that an adjustment to the deferment rate is not the right method to allow for hope value for the reasons that he gave. The value of the option to deal will be at its highest at about 30 years from the end of the lease, as demonstrated by a graph that Mr Beckett produced. In comparing the with-Act world with the hypothetical situation with which we are currently concerned, Mr Cullum said that in the real world both the assignor and assignee were aware that there was latent value to be to be released, and the power to compel the freeholder to deal was in the tenants hands. The assignor would not abandon his latent value, and the assignee would be willing to pay a share of it, as he would be giving himself the option of releasing the whole of the marriage value. That was why market evidence of the value of leasehold interests had to be discounted in enfranchisement valuations to arrive at the no-Act world value – in other words, the value of the option had to be stripped out. In the no-Act world, the negotiating boot would be on the other foot, and the tenant would not be in a position to demand an extended lease or seek the freehold. With the purchaser of the freehold reversion being in control and knowing that it was most likely that the tenant would come forward at some stage, he would pay a price that was the sum of the reversionary value together with an amount that represented the value of that option to deal.                         112. In our view this sums up the reasoning quite succinctly, and we conclude that it is appropriate to apply a lump sum to the value of the reversion prior to the statutory marriage value apportionment. It would be related to the unexpired length of the lease. As Mr Cullum said, this lump sum could only be calculated once the freehold and existing leasehold values were known, and because in his case those elements were to be decided at a future hearing, the mathematics to incorporate the hope value element into the deferment rate could not, therefore, be undertaken at this stage. Mr Orr-Ewing’s point about adjustment to deferment rate being incompatible with the underlying assumption that the landlord would be obtaining vacant possession at the end of the term is also, in our view, well made and adds weight to the conclusion we have reached. General effect of conclusions                         113. In Arbib paragraphs 110 to 117, the Tribunal explained why decisions of this Tribunal on matters of fact cannot be determinative of such facts in other cases. Nevertheless it sought to set out its decision as to the deferment rates that it found to be appropriate in the cases which it was considering in a way which would provide guidance for other cases. It set out in paragraph 179 its reasons for not varying the deferment rate for different valuation dates over a period of 9 months, in terms which have more far-reaching implications: “In our judgement, it is not until a change in the trend in risk-free yields has been established over a period of years, so that it is recognisable in the market for long term investments, or the continuance of the trend establishes a new level of yields, as can be seen in yields to conventional zero coupon gilts since the 1990s, that there will be changes that can properly be reflected by step gradations in deferment rates, assessed for the purpose upon which we are engaged. Changing yields in the property market, such as we have commented on above, would be helpful in identifying the time for adjustment of deferment rates once credibly established, but short term fluctuations should be discounted in making use of what we have identified as a valuation tool.” The words which we have underlined show that the Tribunal was expecting, in effect, that once a deferment rate had been established it should be generally applied, at least in comparable circumstance, until evidence of a change in the risk-free rate justified a “step gradation”.                         114. It is, we think, appropriate that we should be rather more explicit about the status of our present decision on the deferment rate. The starting point is to observe that it is the function of the Lands Tribunal, as a specialist tribunal with an England and Wales jurisdiction, to promote consistent practice in the application of the law in decisions that are made within its particular jurisdictions. The purpose of the Tribunal was to provide “a single consistent jurisdiction” in relation to those matters assigned to it (see Sir Hartley Shawcross QC A-G on the Second Reading of the Lands Tribunal Bill, HC Hansard 1948-49 col 41).                         115. In Shephard v Turner [2006] EWCA Civ 8 in the Court of Appeal, a case on the Tribunal’s original jurisdiction under section 84 of the Law of Property Act 1925 to discharge and modify restrictive covenants, reference was made to previous decisions of the Tribunal on the matter in issue – whether the prevention of disturbance during construction operations might amount to a benefit of substantial value or advantage (within subsection (1A) of that section). At paragraph 57 Carnwath LJ said this: “In reviewing these decisions, it is important to keep in mind that Tribunal decisions are not to be regarded as setting any precedent in relation to what must be essentially a question of fact and degree. However, one of the functions of a specialist tribunal such as the Lands Tribunal (made explicit by s 4(1)(b) of the Lands Tribunal Act 1949) is to promote consistent practice in the application of the law to its specialist field. Unexplained inconsistency of approach may in certain circumstances amount to an error of law.” 116. In another recent case, R (Sinclair Gardens Investments Ltd) v Lands Tribunal [2006] 3 All ER 650, the Court of Appeal had to consider the circumstances under which the refusal by the Tribunal of permission to appeal from a decision of a leasehold valuation tribunal might be susceptible of judicial review. At 664 paragraph 57 Neuberger LJ said this: “I think it appropriate to say, that there could, in my view, be cases, which would be wholly exceptional, where it would be right to consider an application for judicial review of such a decision on the basis of what could be said to be an error of law. A possible example would be if the Lands Tribunal, despite being aware of the position, refused, without any good reason, permission to appeal on a difficult point of law of general application, which had been before a number LVTs which had taken different views on it, and which cried out for a definitive answer in the public interest. In that connection, it seems to me that one could say that it was not so much the point of law itself which justified judicial review, but the failure of a public tribunal to perform its duty to the public, as well as what one might call its duty to the parties in that particular case.” 117. The function of the Tribunal is thus to make decisions on points of law and on what may be called principles of practice to which regard should be had by the first-tier tribunals and by practitioners dealing with claims in any of the Tribunal’s original or appellate jurisdictions. Such principles of practice are not, in our view, confined to valuation methodology (for example, in rating, whether local authority leisure centres should be valued on the contractor’s basis or by some other method: see Eastbourne Borough Council v Allen (VO) [2001] RA 273) but may extend to matters of quantification if the considerations underlying the quantification are of general application. For many years the Tribunal’s decision in Stokes v Cambridge Corpn (1961) 13 P & CR 77, a case on compensation for compulsory purchase, was treated by valuers as establishing that the appropriate share of the development value of a piece of land to attribute to the land over which the only access to such land could be obtained was one-third. This was in fact based on a misreading of the Tribunal’s decision, which rejected a share of one-half in favour of one-third to reflect the fact that development of the land for which access was required would increase the prospect of development of other land owned by the owner of the ransom strip. However, had the decision been treated as establishing that the ransom value for access should be one-half of the development value but that this should be varied as necessary to reflect the particular circumstances of the case, that would have been appropriate, and the decision could have been applied as determining such quantification as a matter of general application. 118. A similar problem to the one in the present case used to confront local valuation courts (the predecessors of the present valuation tribunals) and the Lands Tribunal when considering rating assessments made on the contractor’s basis, the method used for valuing public buildings, schools and universities, sports stadia and other hereditaments that are rarely let from year to year. In such a valuation the rateable, annual, value is reached by decapitalising the adjusted cost of construction, and the decapitalisation rate, of considerable importance to the end result, gave rise to many disputes. In Imperial College of Science and Technology v Ebdon (VO) [1984] RA 213 the Lands Tribunal (C R Mallett FRICS) heard in the course of an 18-day hearing detailed evidence, from economists and valuers, on the economic and market considerations that would influence the hypothetical tenant’s choice between construction and ownership on the one hand and renting on the statutory terms on the other and what decapitalisation rate should as a consequence be applied. The Tribunal’s decision that the appropriate decapitalisation rate was 3½% was challenged in the Court of Appeal as being wrong in law. The Court of Appeal ([1986] RA 233) dismissed the appeal. In his judgment Glidewell LJ said this (at 242): “I emphasise that this is not a decision that 3½% is necessarily the correct rate to be applied in decapitalising a capital value in the application of the contractor’s basis to the rating assessments of all universities or colleges. Each case depends upon its particular facts and the evidence called. My judgment is that the facts established and the opinion evidence called justified the Lands Tribunal’s decision in this case.” 119. Although the other members of the court, Purchas and Watkins LJJ, did not expressly associate themselves with this observation, the prospect of the sort of detailed evidence that had been called in that case being required to establish the appropriate decapitalisation rate whenever a contractor’s basis valuation was in dispute led to the view that the rate should be prescribed by statute. The Local Government Finance Act 1988 Schedule 6 paragraph 2(8) gave the Secretary of State for the Environment power to prescribe assumptions for the valuation of hereditaments in a prescribed class; and since 1990 decapitalisation rates for contractor’s basis valuations have been prescribed, with one general rate and a lower rate for educational and medical establishments. 120. The function of this Tribunal in determining a deferment rate to be treated as a guideline in other enfranchisement cases is similar to the courts’ determination of a discount rate for the purpose of reaching the present value of the loss of future earnings in personal injury cases. What the discount rate should be was the subject of consideration by the House of Lords in Wright v British Railways Board [1983] 2 AC 773 and later in Wells v Wells [1999] AC 345 before the rate was statutorily prescribed by the Damages (Personal Injury) Order 2001. In Wright Lord Diplock said (at 785C): “A guideline as to quantum of conventional damages or conventional interest thereon is not a rule of law nor is it a rule of practice. It sets no binding precedent; it can be varied as circumstances change or experience shows that it does not assist in the achievement of even-handed justice or makes trials more lengthy or expensive or settlements more difficult to reach. But though guidelines change, too frequent alteration deprives them of their usefulness in providing a degree of predictability in the litigious process and to facilitating settlement of claims without going to trial.” Later on (at 785H) Lord Diplock said: “…the guideline, if it is to serve its purpose in promoting predictability and so facilitating settlements and eliminating the expense of regularly calling expert evidence at trials of personal injury actions, should continue to be followed for the time being…”                         121. While the words in these passages were spoken in the context of a claim for interest, as to the rate of which the judge had an obvious discretion, the same considerations in relation to the promotion of predictability apply, in our view, to the deferment rate. It is obviously undesirable, and indeed it would be impossible, for the sort of financial and valuation evidence that we have heard to be called and considered in every enfranchisement case. It is, in our judgment, unnecessary that it should be, because LVTs and this Tribunal are entitled to rely on their own expertise, guided by this decision. The prospect of varying conclusions on the deferment rate in different cases reached on evidence that was less comprehensive than that before us can therefore be avoided by LVTs adopting the practice of following the guidance of this decision unless compelling evidence to the contrary is adduced. This is justified because, as we have explained above, the deferment rate is unlikely to vary according to factors particular to the individual case. Some factors, including in particular the prospect of long-term growth, will not vary from case to case, while other factors, such as location and obsolescence, will already be reflected in the vacant possession value. Hope value would be a factor that could lead to different deferment rates for different lengths of term if it was not reflected elsewhere in the valuation; but we have concluded that hope value is excluded as a matter of law. The case for adopting a single deferment rate (with a standard adjustment for flats) for all reversions in excess of 20 years is thus, in our view, strong. Indeed we think that statutory prescription could well be appropriate and could usefully give a greater certainty to the market than a decision of the Lands Tribunal setting a guideline is capable of doing.                         122. It is a necessary part of the concept of a guideline applicable to future cases that the deferment rate should be stable, although clearly its potential for change needs to be recognised. We sought the opinions of the experts in these appeals as to whether and to what extent the deferment rate could be expected to vary over time. Mr Clokey, of course, would vary his deferment rate as the yield on index-linked zero-coupon gilts moved from day to day. That approach, however, would, in our judgment, be inconsistent with recognising the function of the deferment rate as a valuation tool to assess a long-term investment in a particular form of asset, one of the features of which is stability rather than volatility. In our judgment the deferment rate may be treated as stable over time unless a trend movement in the risk-free rate can be identified or it can be established that the long term prospects of growth in residential property have changed or that, for some other reason, the attraction of investment in residential reversions can be shown to have increased or diminished.                         123. The application of the deferment rate of 5% for flats and 4.75% for houses that we have found to be generally applicable will need to be considered in relation to the facts of each individual case. Before applying a rate that is different from this, however, a valuer or an LVT should be satisfied that there are particular features that fall outside the matters that are reflected in the vacant possession value of the house or flat or in the deferment rate itself and can be shown to make a departure from the rate appropriate. Decision on each appeal                         124. In the light of our findings above, we determine that the deferment rate applicable in each of the appeals relating to flats (LRA/50/2005, LRA/9/2005, LRA117 & 131/2005 and LRA/38 & 40/2005) is 5%, and hope value, where in issue, is to be excluded from the valuation. For the house, 13 South Terrace (LRA/63/2005), the deferment rate is determined at 4.75%. Unless the landlords in the contested appeals accept our decision that hope value is excluded as a matter of law, there will be a need for alternative valuations that reflect its inclusion in the manner that we have determined, and these will require consideration at subsequent substantive hearings. In relation to 13 South Terrace, since under the provisions of section 9(1A) of the 1967 Act hope value is not excluded, we invite the appellant in this unopposed appeal to submit a revised valuation reflecting the deferment rate that we have determined. Where hope value is to be included in the valuation this should be done by the addition of a lump sum to the value of the reversion. Dated: 15 September 2006 George Bartlett QC (President) His Honour Michael Rich QC P R Francis FRICS