17 May, 2016
by: Cripps Pemberton Greenish

Sloane Stanley Estate v Munday
Flat 3, 36 Elm Park Road London SW3 (Flat 3)

Sloane Stanley Estate v Lagesse
Flat 11, 26-28 Elm Park Road (Flat 11)

Aaron v Wellcome Trust
Flat 5, 17 Cranley Gardens London SW7 3BD (Flat 5)

The Upper Tribunal (Lands Chamber) has now handed down its long awaited decision in these relativity cases. Each of the three cases (heard together) was concerned with a new lease claim made under Chapter II of Part I of the Leasehold Reform, Housing and Urban Development Act 1993 (“the 1993 Act”). All the flats are in Prime Central London; two in Chelsea and one in Kensington. The unexpired terms and valuation dates in each case were: (i) Flat 3 – 23 years – 20th March 2014 (ii) Flat 11 – 37.71 years – 9th April 2014 and (iii) Flat 5 – 41.32 years – 25th February 2014. The cases had been referred directly to the Upper Tribunal.

The principal issue [1] in each case was the appropriateness of using a hedonic regression model (“the Parthenia model”) to determine relativity for different lease lengths on the statutory assumption that the lease had no right under the 1993 Act to acquire any interest in any premises containing the tenant’s flat or to acquire any new lease [2]. The Parthenia model had been constructed from a large amount of data of market transactions (7,969) taken from a period prior to the introduction of the 1993 Act (the period taken was between 1987 and 1991) at a time when transactions were not influenced by any statutory rights under the 1993 Act. The intention was that, by use of hedonic regression, it is possible to isolate from that data the effect on value of the single variable of lease length for different lengths of lease. The proposition was that the model was accurate in indicating relativity in the market for the period 1987 to 1991 and that it remained accurate in the current market because there had been no material alteration in market forces since then [3].

The Parthenia model had previously been considered by the Upper Tribunal in Kosta v Carnwath [2014] UKUT 0319 (LC). Although criticised and not relied on in that case, it continued to be put forward as evidence of the appropriate relativity in subsequent applications to the First-tier Tribunal.

These three cases were heard over nine days before Mr Justice Morgan and Mr Andrew Trott FRICS. There was valuation evidence from four expert valuers; in addition, six witnesses gave evidence on hedonic regression and on the Parthenia model and there were eight further witnesses of fact who gave evidence about the construction of the various Graphs of Relativity; nine different graphs were considered [4]. [Appendix C to the decision]

The determination of the Upper Tribunal can be summarised as follows:

1.    The relevant market for the purpose of determining the open market value of the existing lease on the statutory assumptions is the real world market. It is not a hypothetical market. What is hypothetical is that the lease is on the market, is sold on the market and does not have rights under the 1993 Act. It is therefore the sale of a hypothetical asset in the real market. [17]

2.    The function of the tribunals in determining values under the statute is to decide market value at a past valuation date. As part of that function, the tribunal needs to make findings as to how the market then behaved. It is not a function of the tribunal to direct a market how to behave in the future nor to “correct” past behaviour; it is only actual market behaviour that is relevant [5]. [143]

3.    The use of the Parthenia model for the purpose of determining relativity in these cases was rejected. Furthermore, it is not capable of being revised or modified to make it effective [6] [122]. It should not therefore be put forward in any future case [165]. Although rejected primarily because the model was incompatible with market evidence App B [60], it was also held that a number of the technical criticisms (set out at some length in Appendix B to the decision) were justified App B [61]. It was also doubted in any event that relativities from the period 1987 to 1991 would be relevant to the market that existed at these valuation dates; relativities now were probably lower than they were during the period covered by the Parthenia model App B [61].

4.    As regards the various graphs,

a.    the tribunal gained no assistance from the College of Estate Management Graph App C [67]; the John D Wood Graph App C [68}; the WA Ellis Graph App C [69]; the Charles Boston Graph, the Cluttons Graph or the Knight Frank Graph App C [70] [7]. None of them were considered to be influential in the real world market.

b.    Savills Enfranchiseable (2015) Graph. This is a hedonic regression model based 10,887 real world transactions that occurred between 2010 and 2015. Its purpose is to estimate real world relativity in that period. Although subject to technical criticisms, it is considered to be a significant improvement on the Savills Enfranchiseable (2002) Graph App C [59]. If the technical criticisms could be overcome, then this graph could be beneficial and replace the 2002 equivalent [170] [8].

c.    Savills Enfranchiseable (2002) Graph. This graph is based solely on the subjective opinion of a panel of valuers. It was accepted that it was in common use by valuers at and before the valuation dates to determine FHVP values from real world leasehold transactions. App C [59].

d.    Gerald Eve Graph. The tribunal accepted that this graph was in the most common use at the valuation dates to determine relativity and that it influenced the market for leases with rights App C [62] [9].  It was the “industry standard” App C [63]. At the valuation dates, prospective purchasers acquiring leases would most likely have referred to the Gerald Eve Graph App C [71]

5.    The tribunal considered that there was no reason to believe that relativity is immutable. App C [63]. Indeed, the evidence suggested that the Gerald Eve Graph may well overstate current relativities [153], a proposition supported by the fact that the Savills Enfranchiseable (2002) Graph shows higher real world relativities that the Savills Enfranchiseable (2015) Graph App C [64].

6.    The suggestion in Kosta that prospective purchasers of leases, acting prudently, would have regard to the average of the graphs in the 2009 Relativity Report produced by the RICS was wrong and should not be followed. The evidence showed that such purchasers would likely have utilised the Gerald Eve Graph App C [71]

7.    The determinations for relativity in each case are:

a.    Flat 3. FHVP £950,000 (determined [10]) Existing lease £448,400 (47.2%) [159]

b.    Flat 11. FHVP £581,000 (determined[12]) Existing lease £370,155 (63.71%) [13] [154]

c.    Flat 5. FHVP £2,750,000 (agreed) Existing lease £1,800,000 (65.55%) [14] [148]

8.    Although not specifically called a “guidance case”, there is a section in the decision headed “Future Cases” [15]. Having rejected the Parthenia model and after expressing reservations about the Gerald Eve Graph [153], the tribunal has set out those matters which it considers might be useful in determining relativity in future cases.

a.    First, the Parthenia model should not be used [165]

b.    Secondly, tribunals and experts must focus on the real market at the valuation date, what influenced it and how it performed. It is not open to a party to suggest that the market was historically badly informed or operating illogically or inappropriately in order to invite a tribunal to replace actual market forces with what might be suggested as more logical or appropriate considerations. [166]

c.    Thirdly, markets and what influences them over time can change [16]; regard must be had to that [167]

d.    Fourthly, if there has been a true market transaction with the existing lease on or around the valuation date then that transaction is likely to be a useful starting point for the purpose of determining what is the value of that lease without rights. An expert valuer will be able to express a view as to the proper deduction to me made against the price achieved with rights to reflect the value without rights. [168]

e.    Fifthly, where there is no such market transaction, then valuers would be encouraged to adopt more than one approach. One method would be to use what the valuer considers to be the most reliable graph to ascertain the relative value. Another method would be to use a graph to ascertain the relative value of the existing lease with rights and then make a deduction from that value (the amount of the deduction would be a matter of expert opinion) to reflect the value of rights. If the two methods produce different results, the expert valuer will need to weigh up the strengths and weaknesses of the two methods in reaching his conclusion [169]. If the Savills Enfranchiseable (2015) Graph could overcome the technical criticisms made of it [17] and become accepted in the market as a true reflection of real world relativities of leases with rights, then it could become the “industry standard” for adjusting the FHVP value in the real world in order to ascertain a value for the existing lease with rights from which a deduction for rights can then be made.

So, where do we go from here? This decision clearly gives substantial guidance on the difficult issue of relativity. The Parthenia model is now dead and is not capable of resuscitation. Most of the other graphs analysed in the decision are also likely to be put out to grass. The tribunal clearly wants to move away from graphs that are based on historic subjective opinion, settlements and/or tribunal decisions. The Gerald Eve Graph, although a fine old war-horse that has served the Prime Central London market well, is likely gently to fade away unless subjected to an overhaul to reflect the economic and market changes which suggest that relativities should be lower. The tribunal has accepted that it most likely overstates current relativities in PCL and probably has done so for a number of years. We shall see if Savills take up the challenge to perfect their 2015 hedonic regression model so that it becomes accepted as the best evidence of real world relativities in the London market. If that happens, then the issue of the value of rights will once again come to the forefront as valuers look to adjust their real world comparables for that value [18].  We may not yet have the final answer, but the Upper Tribunal has done what the RICS Report on Relativities singularly failed to do and has given us a clear path for valuers to follow for the future in order to determine relativity. They are to be commended for that.

Damian Greenish

11th May 2016


1. In the two Sloane Stanley Estate cases, the freehold vacant possession values were also in dispute.
2. This has generally been called the “the no Act world” but this is something of a misnomer; the assumption is that the subject lease does not have rights but exists in a real world where most (if not all) leases do have rights [15].
3. It was put forward by lessees because generally it suggested a higher relativity than other methods used in the market
4. They were (i) Gerald Eve Graph (ii) Knight Frank Graph (iii) WA Ellis Graph (iv) Cluttons Graph (v) John D Wood Graph (vi) Charles Boston Graph (vii) College of Estate Management Graph (viii) Savills Enfranchiseable (2002) Graph and (ix) Savills Enfranchiseable (2015) Graph
5. The UT accepted evidence that the Gerald Eve Graph was used by the market and influenced market behaviour at the valuation dates; it was therefore legitimate to take into account that use and influence in determining market values. [142] [145]
6.  It was described as being “… a clock that strikes 13” [122]. This assessment was derived from the fact that the valuers had agreed that the FHVP of Flat 5 was £2,750,000 and that the existing lease value (with rights) was £2,000,000. Applying the Parthenia model to the agreed FHVP produced an existing lease value (without rights) of £2,232,450; a figure which significantly exceeded the agreed value of the same asset (with rights). This was not a possible outcome and a clear indication that the model was defective [121] App B [60]
7. Evidence was given as regards the compilation of the various graphs which is set out in Appendix C to the decision
8. It was noted that the real world relativities produced by this graph were in many cases lower that the without rights relativities produced by the Parthenia model; another reason to doubt the relevance of the Parthenia model.
9. The evidence was that in most cases, the price sought on the sale of a lease in the market would be influenced (if not largely determined) by the cost under the 1993 Act of extending that lease [105].
10. This was the figure contended for by the landlord
11. This was reduced to £435,000 after a deduction for onerous ground rent. The UT used two methods in considering the question of relativity. The first was application of the Gerald Eve Graph. The second was to apply the Savills Enfranchiseable (2002) Graph to the FHVP value and then to make a deduction for rights from the resultant figure. The deduction made for Flat 3 was 20%. The UT then took the lower of the figures (that determined by the Gerald Eve Graph) “as the least unreliable” on the evidence.
12. The figure contended for by the landlord was £600,000
13. This was reduced to £360,000 after a deduction for onerous ground rent. Again, the UT used two methods in considering the question. The first was application of the Gerald Eve graph. The second was to apply the Savills Enfranchiseable (2002) Graph to the FHVP value and then make a deduction for rights from the resultant figure. The deduction made for Flat 11 was 10%. The UT then took the lower of the figures (that determined by the Gerald Eve Graph) “as the least unreliable” on the evidence.
14. The landlord’s evidence, that there should be a 10% deduction from the agreed existing lease value (with rights) of £2,000,000, was accepted. This produces a relativity below the Gerald Eve Graph (67.06%)
15. Set out in [163] to [170]
16. A particular example put forward by the tribunal was that weight given to any particular graph might change over time
17. Primarily, the failure to meet certain diagnostic tests.
18. A description of the benefits that the Act confers are set out in [127].