Brickfield Properties Limited v Soil Miah Ullah & Ors

This decision deals with an interesting valuation conundrum. Where a property has been sold twice within the space of two weeks at radically different sale prices, which transaction, if either, presents a more reliable indication of its market value?

The issue arises in an appeal from a decision of the First-tier Tribunal (Property Chamber) (“the FTT”), which on 20 April 2021 determined that the premium payable by the enfranchising tenants of 42 Queens Court, Rocklands Drive, Harrow (“the appeal property”) should be £128,774. The element of the valuation resulting in that figure that is in issue before us is the value of the very short (16.23 years) existing leasehold interest in the appeal property.

The competent landlord of the appeal property, Brickfield Properties Limited (“Brickfield”) appeals against the FTT’s decision, the Tribunal having granted permission to appeal by way of a review on 5 August 2021.

In a short hearing at the Royal Courts of Justice on 26 January 2022, the appellant was represented by Carl Fain of counsel. The respondent tenants, Soil Miah Ullah, Gurmack Singh and Jagir Alam, were represented by Robert Denman of counsel. We are grateful to them both.

The appeal property is a purpose built first-floor flat within one of several two storey blocks on the east side of Kenton Lane, Harrow. Accessed from a communal area shared with three other flats, it has an entrance hall, two bedrooms, reception room, bathroom, and kitchen, all in around 600 sq ft. It was held under a lease dated 11 July 1958, for a term of 99 years less three days from 29 September 1936. The annual rent was fixed at 12 guineas or, in new money, £12.60.

The unexpired leasehold interest of some 16 years was marketed by Ellis and Co between March and June 2019 at a guide price of £129,950. On 4 July 2019, the vendors served a notice under section 42 of the Leasehold Reform, Housing and Urban Development Act 1993 as amended (“the Act”), claiming a new lease which would have the effect of extending the term by 90 years. In the notice, the vendors suggested a premium of £49,950. On the same day, the short leasehold interest subject to the notice was sold to a property company, Birdview Homes Limited, for £112,000.

Birdview wasted no time in immediately putting the property into the next Allsop auction on 18 July 2019, at a beguiling guide price of £80,000. The respondents were the successful bidder, at £175,000. There is no evidence before us of how many, if any, other bids were made.

On 18 November 2019, Brickfield served a counter-notice under section 45 of the Act, contending for a premium of £198,620.

The value of a tenant’s existing leasehold interest forms part of the calculation in assessing the premium to be paid by the tenant claiming a new extended lease under the Act. The tenant is required to pay a sum made up of three elements: a) the diminution in the value landlord’s existing interest; b) the landlord’s 50% share of the marriage value created (taken to be zero if the lease had more than eighty years unexpired at the valuation date), and c) compensation for any other loss the landlord will suffer (not relevant here).

To calculate the marriage value, the aggregate of the landlord’s and tenants’ interests before the extension is deducted from the aggregate value of those interests after extension. Thus, the value of the tenant’s existing interest is required as it forms part of those calculations. We should add that any effect on value of tenant’s improvements is ignored.

There are generally two methods used to arrive at the value of the tenant’s existing leasehold interest.

The first is by reference to comparable evidence – sales of short leases in the subject block or nearby. But there is a wrinkle. In order to claim a new lease, the tenant has to have been a qualifying tenant for two years, but a notice given by a qualifying tenant to claim a new lease can be transferred to a new buyer. So, the market operates by the selling tenant, as in this case, serving a section 42 notice to claim a new lease shortly before selling the lease to the buyer, who inherits the benefit of the notice.

However, the legislation (paragraph 4A of schedule 13 to the Act) imposes an artificial assumption that the tenant’s right to enfranchise – known as ‘Act rights’ – must be ignored. Thus, the values produced by any real-world sales of comparable short leasehold interests must be adjusted to remove the element of Act rights. As the Tribunal (Morgan J and Mr Andrew Trott FRICS) explained in The Trustees of the Sloane Stanley Estate v Mundy [2016] UKUT 0223 (LC) at paragraph 168:

“it is likely that there will have been a market transaction at around the valuation date in respect of the existing lease with rights under the 1993 Act. If the price paid for that market transaction was a true reflection of market value for that interest, then that market value will be a very useful starting point for determining the value of the existing lease without rights under the 1993 Act. It will normally be possible for an experienced valuer to express an independent opinion as to the amount of the deduction which would be appropriate to reflect the statutory hypothesis that the existing lease does not have rights under the 1993 Act.”

The second method is by reference to graphs of relativity produced by leading firms. In this context, relativity means the relationship between the value of a leasehold interest in a property compared with its notional freehold value; put simply, the shorter the unexpired term of the lease, the lower the relativity, while a very long lease will have relativity approaching 100%.

In Deritend Investments (Birkdale) Limited v Treskonova [2020] UKUT 0164 (LC), at [31] to [44] the Tribunal (Martin Rodger QC, Deputy Chamber President and Mrs Diane Martin MRICS) gave a helpful tour d’horizon of the developments in relativity calculations over the last few years. For the purposes of this appeal, it is common ground that the two most helpful graphs are those produced in 2016 by Savills and by Gerald Eve, each providing indications of relativity at differing lease lengths. The Gerald Eve graph is unenfranchisable - i.e. the effect on value of Act rights has been stripped out; the Savills graph displays curves on both bases, allowing the value of Act rights at various lease lengths to be assessed. It should be noted that both graphs relate to properties in Prime Central London or PCL.

As the Tribunal went on to mention in Mundy, at [169], when the two methods throw up different figures, it will then be for the good sense of the experienced valuer to determine what figure best reflects the strengths and weaknesses of the two methods which have been used.

Before the FTT, the landlord relied upon expert evidence of Mr Robin Sharp FRICS, the tenants on that of Mr Neil Miller FRICS. The experts had agreed many aspects of the valuation, including the capitalised ground rent, the deferment rate, and the long leasehold to freehold uplift of 1%. The issues remaining to be determined by the FTT were the existing and extended lease values.

In valuing the existing lease, Mr Sharp took his lead from Mundy and first considered the available transactional evidence. There were no comparable sales within the development, other than the two sales in quick succession of the appeal property. However, Mr Sharp did not consider the auction sale to represent a true reflection of the market value of the interest “because of its marketing history prior to auction and because the price paid exceeds any graph-based evidence”. Compared with his estimated freehold value of £273,232, the auction sale at £175,000 equated to a relativity of 64% with Act rights or 50.28% without Act rights.

In Deritend, in which Mr Sharp had given expert evidence before the FTT, the Tribunal adopted an average of the Savills and Gerald Eve graphs, notwithstanding the property in that case was in Sutton, Surrey, outside PCL. Mr Sharp adopted this approach in the present case – averaging the Savills and Gerald Eve graphs to apply a relativity percentage of 32.1% to his estimated freehold value of £273,232 (derived by applying a 1% adjustment to his long leasehold value of £270,500) to arrive at an existing leasehold interest value of £87,707.

Mr Miller also considered Mr Sharp’s approach – applying the average of the two unenfranchisable graphs (like Mr Sharp, he calculated the average to be 32.1%) to his assessment of freehold value of £261,287, to arrive at £83,873. However, Mr Miller preferred to use the auction sale as a reasonable starting point. He discounted the first sale; Birdview had recognised a deal when they saw one, and had bought the property at below market value. They would not have done so had they not considered that a substantial profit could have been made. The auction price would have reflected competitive bidding and provided a reliable indication of market value.

Mr Miller made two adjustments to the auction sale price. The first was to deduct £22,000 for the value of work he thought constituted tenant’s improvements. The second was to make an adjustment for Act rights. Comparing the Savills enfranchisable and unenfranchisable curves, Mr Miller calculated Act rights at an unexpired lease term of 16.23 years at 21.49%. Applying this to the auction sale price net of improvements produced a value of £120,100. He accepted that this figure represented a relativity of 45.96% of freehold value, and was 13.86% above the average graph figure, but was satisfied that it was soundly based on actual market evidence.

The FTT preferred Mr Sharp’s evidence on long leasehold value, adopting his £270,500, and thus a notional freehold value of £273,232. As regards the short leasehold value, the FTT found that a ‘contemporaneous bona fide auction sale’ is the best evidence of value and is preferable to an analysis based on graphs, relying on the Tribunal’s decision in Allen v Leicester City Council [2013] UKUT 016 (LC).

After allowing £2,500 for tenant’s improvements, the FTT adopted Mr Miller’s approach of adjusting the auction sale price by 21.49% for Act rights, to arrive at a short leasehold value of £135,430. The resulting premium was therefore determined at £128,774.

In subsequently declining to review the decision, and refusing permission to appeal to the Tribunal, the FTT reiterated that it was entitled to conclude that an open market sale at auction on 18 July 2019 was the best evidence of value, following Allen; that the guide price evidence showed no relation to sale prices; and that there were no witness statements from those directly involved in the private treaty sale for £112,000 on 4 July 2019, whereas ‘a sale at a public auction speaks for itself’.

This is a review of the FTT’s decision, confined to its determination of the value of the existing leasehold interest. There is no appeal against its findings as to long leasehold value at £270,500, notional freehold value at £273,232, or the value of tenant’s improvements at £2,500.

In carrying out a valuation, it is necessary to have regard all available evidence, and then to stand back and consider how it all fits together. More weight might be given to the elements of evidence which seem to fit a pattern, especially when, as here, some pieces of the jigsaw are blurry.

Before us, there are three pieces to that valuation jigsaw. The first is the private treaty sale on 4 July 2019 at £112,000. It has the advantage of being a sale of the subject property at the valuation date – ordinarily an unimpeachable starting point for a valuation. There was no dispute that the property had been on the market for three months before being sold with the benefit of the s42 notice.

The second piece of the jigsaw is the auction sale at £175,000. As a sale of the subject property only two weeks after the valuation date, again it would normally be a sound basis for valuation. There was no evidence before the FTT that such an increase in value could be explained by a dramatic general rise in residential values.

The third piece is the valuation by reference to relativity graphs. It is common ground that for an unexpired lease term of 16.23 years, the Savills graph suggests an Act rights adjustment of 21.49%, and an average of the Savills and Gerald Eve unenfranchisable graphs suggests a relativity of 32.1%.

If we take the first sale price at £112,000, allow for tenant’s improvements at £2,500, and adjust by 21.49% for Act rights, we arrive at £85,982. That would represent a relativity of 31.46%, which seems to us to compare consistently with the average graph relativity figure of 32.1%. Those two pieces seem to us to fit together. We note that £85,982 also falls between Mr Sharp’s figure of £87,707, and Mr Miller’s figure – if he based it on the first sale – of £83,873.

Instead, the FTT chose the jigsaw piece which fitted with neither of the others – the auction sale, based on which it arrived at a short leasehold value of £135,430 (which, incidentally, was higher than that contended for by either expert). This figure equated to a relativity of 49.57%. On both the Savills and Gerald Eve unenfranchisable graphs, that would represent a lease with an unexpired term of around 28.5 years.

The FTT was influenced by the Tribunal’s decision in Allen, which the panel introduced to the proceedings itself, and in its decision quoted paragraph 45 where Mr Trott FRICS said this:

“An auction is a recognised method of disposal to achieve open market value. Bids are made openly. It differs from an informal or formal tender where prospective purchasers make their offers in confidence and where there is scope for misjudging the market and offering considerably more than the other bidders.”

That is entirely uncontroversial. But Allen was a compulsory purchase case in which the Tribunal was asked to decide between an auction sale on the one hand, and on the other a basic residual method, which was noted as being a method of last resort that the Tribunal adopts only with reluctance. The Tribunal was not suggesting that an auction sale must prevail in all circumstances, in the face of all competing evidence.

We do not consider that either of the sales, in isolation, is reliable. There was insufficient evidence before the FTT to illuminate the circumstances of either. It is possible that the leasehold interest was snapped up by Birdview at less than market value, despite having been openly on the market for some months. It is equally possible that the respondents overpaid at auction. Given the guide price, it seems unlikely that there were no underbidders, so it is difficult to say by how much. It might have been helpful to the FTT for there to have been evidence from the respondents themselves, or other witnesses of fact. We note, however, that the auction room would probably have been aware of the original s.42 notice suggesting a premium of £49,590, but Brickfield’s counter notice at £198,620 post-dated the auction by some months. Were the respondents persuaded to pay a price thinking that the eventual premium would be modest? We just don’t know; neither did the FTT.

We think the FTT fell into error in focusing solely on the auction sale. Mr Denman argued, rightly, that we cannot ignore the auction price. Indeed, we cannot and do not ignore either sale price. But they are so far apart that any attempt to find a middle ground between them would be arbitrary. What we do have, however, is the additional evidence of the graphs produced by two separate specialist firms, and the consistency of the first sale price with the graphs indicates to us that the first sale price is more likely to reflect the market value than does the auction sale.

We accept that the property was immediately “turned”, and that there might have been another bidder prepared to pay something like the price which the respondents paid. It is possible that there might have been two or more excited bidders, or the respondents and a canny auctioneer, but this is all unknown.

In our view, the evidence surrounding the auction is sufficiently opaque to not displace the two other jigsaw pieces which fit together. We think the FTT should have used the first sale, and adjusted it for tenant’s improvements and Act rights, then checked against the enfranchisement graphs, as a better method of determining the existing lease value for the purposes of the enfranchisement calculation.

We do not consider it proportionate to remit this single point to the FTT to recalculate when we have sufficient material before us to carry out the exercise.

Our calculation in paragraph 30 above gives a value for the existing leasehold interest of £85,982. As a market transaction this might be susceptible to rounding, but since there is no challenge to any of the other elements of the FTT’s calculation which involve precise figures, we do not do so.

The calculation of the premium therefore becomes, adopting the FTT’s appendix:

We therefore determine the premium payable at £153,498, and substitute that for the figure of £128,774 decided by the FTT.

We are aware in coming to this figure that the amount the respondents will in the end have paid (being the auction price of £175,000 and now the premium of £153,498) exceeds the value of the long lease by a considerable distance. But in the absence of any evidence from the tenants we cannot comment on that, and can see no reason why the appellant landlord should be penalised for the tenants’ decision to bid that amount.