Chifley Holdings Ltd (BVI) v The Commissioners For HMRC

The dispute in this appeal is not for want of evidence – there were available to the experts 20 transactions on 18 houses around Chester Square. In their statement of agreed facts, the experts narrowed these down to five transactions upon which they both placed weight, and they each relied on an additional two (Mr Alderton additionally relied upon the sale of a house in South Eaton Place, some distance from Chester Square, but given the amount of evidence in the square I have not placed any weight on that). Apart from no.s 13 and 80 Chester Square, they are all inner-terraced houses.

Shown on the plan above (with the appeal property shown edged red) they are as follows. The five joint comparables, with their dates of exchange of contracts and prices, were:

In addition to the above, Mr Alderton relied upon the following transactions:

Mr Nesbit considered there were three transactions which were his ‘principal and key’ evidence. In addition to no.13, above, he relied on the following two sales:

The valuation date is 1 April 2017. In his initial report, Mr Alderton said he had placed most weight on comparable evidence that would have been available to the valuer at the valuation date but, in line with RICS Guidance and the decision of the Tribunal (Mr A J Trott FRICS) in Allen v Leicester City Council [2013] UKUT 22 (LC), he had also had regard to transactions in the same calendar year, post valuation date, to help to confirm the state of the market in April 2017.

Mr Nesbit said that he had considered all evidence of transactions in Chester Square within a window either side of the valuation date – 3.5 years pre-date and five years post-date (September 2013 to March 2022). However, his three pieces of ‘principal and key evidence’ were the sales of no.13 (July 2016, nine months before the valuation date), no.2 which he gamely described as being ‘in 2018’ [that is true, just – contracts were exchanged on 21 December 2018, so 20 months after the valuation date], and no.80, in November 2018, 19 months after the valuation date.

The RICS Guidance which Mr Alderton referred to was originally published as a 2013 Guidance Note “Comparable evidence in real estate valuation” (therefore in force at the valuation date) before being updated in 2019 as a Professional Statement, and again in 2023 as a Professional Standard. Despite the changes to the document’s nomenclature, it appears that the guidance remained the same throughout:

“3.3.2

Retrospective and projected valuations

Valuers may sometimes be asked to provide an opinion of value on a specified date in the past when market conditions will probably have been significantly different from those at the date of instruction. This situation can arise, for example, in valuations for taxation purposes or those used as evidence in a court case.

If the valuation is an audit of another valuer’s work, the valuer will need to work with the evidence used in the valuation subject to the audit. In other cases, the valuer should consider the following.

• Comparable evidence should only be used if it would have been available to a valuer on the date of valuation.

• Viewed with the benefit of hindsight, comparable evidence can be much clearer than it would have appeared to a valuer at the date of valuation. The valuer valuing retrospectively needs to place themselves in the position of someone reviewing the available evidence on the valuation date, and then make a judgement on the extent and nature of the evidence that could reasonably be expected to have been available at the time.”

In his second report, Mr Alderton referring to a decision of the First-tier Tax Tribunal, which held that post-valuation date transactions should carry less evidential weight than those transactions before the date. Mr Nesbit’s interpretation of the RICS guidance was that it was aimed at the auditing of another valuer’s report. The intention, he said, was not to limit the scope of post valuation date evidence in other cases. There was no logic, to Mr Nesbit’s mind, as to why evidence in August 2014 which he said was the earliest of the transactions upon which Mr Alderton relied not having the same weight as that in December 2019, equidistant from the valuation date. Mr Nesbit thought that there had been a drafting error in the RICS document, and that the principle applied only to auditing scenarios.

In one of many instances where he made similar allegations, Mr Nesbit thought that Mr Alderton’s choice of comparable selection raised questions of even handedness and consistency and was ‘a deliberate attempt to find a reason not to include post-valuation date evidence’. This theme was developed in Mr Nesbit’s later written evidence, making further accusations, and voicing ‘grave concerns as to whether Mr Alderton’s selection of evidence undermines the entirety of Mr Alderton’s evidence’, and accusing Mr Alderton of seeking out evidence ‘in a vain and somewhat panicked attempt to justify an unjustifiable stance…’.

Mr Nesbit’s intemperate criticisms, upon which I comment later, are to my mind entirely unfounded and are a misreading of Mr Alderton’s evidence, which I have summarised at paragraph 37. Mr Alderton placed most weight on transactions that had occurred before the valuation date but used limited post-valuation date evidence in the months after the valuation date to confirm the state of the market. That seems to me to be sensible in principle.

It is important to distinguish between events and evidence. In Castlefield Property Limited v National Highways Limited [2023] UKUT 217 (LC) the Tribunal (Martin Rodger KC, Deputy President, and Mr M Higgin FRICS) commented [21] that

“As a general rule, without an express contractual or statutory instruction to do so …. it would always be wrong to value land as if with knowledge of matters which were not known, and could not have been known, at the valuation date.”

That was in a claim for compensation for compulsory purchase, but the general tenet is applicable to other spheres. The learned authors of the Handbook of Rent Review suggest that the principle was equally applicable to the field of rent review, as regards post-valuation date events. It is equally applicable here. Events which might affect market conditions after 1 April 2017 must of course be disregarded, for instance the Prime Minister’s ill-fated decision on 18 April to hold a general election.

But the position as regards post-date transactions is more nuanced, at least those in the months immediately after the valuation date. Returning for the moment to the world of rent review, the Handbook refers to the decision of Staughton J in Segama v Penny Le Roy [1984] 1 EGLR 109 that:

“If rent of comparable premises had been agreed on the day after the relevant date, I cannot see that such an agreement would be of no relevance whatever to what the market rent was at the relevant date itself. If the lapse of time before the agreement for comparable premises becomes greater then, as the arbitrator said, the evidence will become progressively unreliable as evidence of rental values at the relevant date. The same is no doubt true of rents agreed some time before the relevant date; but nobody suggested to me that those should be excluded….”

In Allen, the Tribunal was faced with the sale of the subject property at auction some nine months after the valuation date, in the context of s.5A of the Land Compensation Act 1961, which provided that the valuation for compensation purposes must be made at the relevant valuation date, but that ‘no adjustment is to be made in respect of anything that happens after the valuation date’. The Tribunal was satisfied that it was permissible to take account of the subsequent sale of the actual property being valued. As Mr Alderton quoted, the Tribunal went on, [39]:

From all of the above I take these points. First, events after the valuation date, which would not have been known to either the valuer or (because every valuation begs the question of what a property would have sold for) the hypothetical buyer and seller, should not be taken into account.

Secondly, what can provide retrospective assistance is what comparable properties were selling for around or some months after the valuation date, especially in the case of a transaction on the subject property, or very similar properties in close proximity. In conjunction with pre-date transactions, these can help inform the valuer as to the state of the market in the months around the valuation date. I accept that there may be a tension between those two principles, to the extent that some transactions might have been affected by impermissible events, and in that respect the valuer must be alive to outliers or sudden changes in market levels that might have been affected by post-date events, for instance a general election or a change to interest rates. The valuer must make a judgement as to whether the market, at the valuation date, anticipated a future event, which would be legitimate to take into account. With the passage of time, some market expectations become reality and some do not, and it is necessary to consider with any post-valuation date evidence whether something has changed in the intervening period which undermines the reliability of the evidence as a measure of value of the subject property at the valuation date. As ever, valuation is an art.

Thirdly, for these reasons, as regards post-review transactions there can be no doubt that the further one moves away from the valuation date the less weight should be applied to a transaction, because it becomes ‘progressively unreliable’, as in Segama. I accept, as did Staughton J, that the same might be said for historic transactions, but the difference is that, while historic, they are known to the market at the valuation date, and sensible adjustments can be made on the basis of ‘known knowns’.

How do these conclusions align with the basis of the RICS Professional Standard? In his later evidence Mr Alderton pointed out the difference in the definitions section of the document between ‘must’ and ‘should’. The Standard uses ‘should’, which is not mandatory but is nevertheless best practice, and that ‘it is recognised that there may be acceptable alternatives to best practice that achieve the same or a better outcome.’ That does not seem to me a terribly satisfactory answer. Neither do I accept Mr Nesbit’s assertion that there is a drafting error in the Professional Standard, which survived two changes in status under the auspices of an experienced and eminent committee. But the general approach of the courts, arbitrators, and of the Tribunal, slightly departs from the statement in the Professional Standard in the ways I have mentioned above.

In the end, as Mr Alderton says, both he and Mr Nesbit use post-valuation date evidence; the difference between them is the length of time after the valuation date within which comparables are referred to. As I have said, Mr Alderton was in principle justified in his choice of comparables. As regards Mr Nesbit’s two (very) late 2018 sales, it is puzzling, in the light of other evidence, that he could consider them to be ‘principal and key’ pieces of evidence, particularly when one of the properties, 80 Chester Square, is a significantly larger and prominent corner property, and there are transactions closer to the valuation date.

I have no doubt that those beyond the end of 2017, which include two of Mr Nesbit’s three ‘key’ transactions, should not be afforded any weight. There is sufficient evidence available being the five joint comparables plus Mr Alderton’s further two, to enable a reliable valuation of the appeal property to be undertaken. I have therefore used those seven properties as the comparable evidence.

Disappointingly, the experts have been unable to carry out a joint inspection of the appeal property and there is no agreed floor area.

The current mandatory standard for the measurement of property is the RICS Property Measurement (2^nd edition) Professional Statement (effective 1 May 2018), which incorporates the International Property Measurement Standards. However, at the valuation date of 1 April 2017, the sixth edition of the RICS Code of Measuring Practice was in force. As regards residential valuation, it says this (para. 25.0):

“There is no single accepted practice for measurement of residential property for valuation purposes… If reference to property area is required then the alternative approaches are [gross external area], [gross internal area], [net sales area] or [effective floor area] The basis of those areas should be stated in the valuer’s report.”

In so far as relevant to this appeal, gross internal area (‘GIA’) is defined as the area of a building measured to the internal face of the perimeter walls at each floor level. It includes garages (para. 2.16) plant rooms (2.9) and ‘areas with a headroom of less than 1.5m (para 2.14 - except in non-domestic rating), but excludes, among other things, ‘external open-sided balconies’ (2.19).

Net sales area (‘NSA’) is defined (para 26.0) as the GIA excluding, among others, areas with headroom of less than 1.5m where the dwelling does not have usable space vertically above (26.5), garages (26.6)…. and external open-sided balconies (26.8).

As the Code is careful to explain, it is not a code of valuation, and the value to be attributed to any particular floor area of special characteristic is part of valuer judgment. Quite. What is important is that, whichever floor area basis is adopted, the valuer values as he or she devalues. Mr Alderton adopts GIA as his method of devaluation of the comparables and the valuation of the appeal property. Mr Nesbit prefers to adopt NSA, because he argues that garages do not attract the same value per sq ft as the main space of a house.

Whichever floor area basis is adopted, it should be perfectly possible for the actual floor areas to be agreed by two chartered surveyors, and here they are not.

Mr Alderton inspected the appeal property in March 2020 and again in October 2023 with Mr Passmore (the appellant’s former expert), following which the GIA was agreed at 4,277 sq ft including the garage and the area of the vaults over 1.5m in height (in fact, as mentioned above, even areas below 1.5m in height should be included for residential properties). A breakdown of his floor area was provided.

Mr Nesbit inspected the appeal property on 31 January 2024. His GIA is 4,176 sq ft, comprising 3,808 sq ft of living accommodation and 386 sq ft of garage space. He measures the external roof terrace at 342 sq ft. No further breakdown was provided.

The appellant had the appeal property on the market in 2022, without success. The experts agree that floor layout at the valuation date is the same as shown on the plan attached to the 2022 sales particulars. This ‘Lonres’ plan indicates an ‘approx Gross Internal Area’ of 4,437 sq ft. This appears to exclude the separately stated ‘vault area’ of 116 sq ft, and an ‘under garage storage area’ of 282 sq ft (neither expert includes this).

Mr Alderton exhibited a valuation report from JSRE Partners, dated October 2019, in which the GIA, taken from a ‘measuring exercise’, is stated as 4,194 sq ft including vaults, plant, and garage, but excluding areas under 1.5m.

Taking the above together, I prefer Mr Alderton’s floor area of 4,277 sq ft. It is supported by a breakdown, had been agreed with the appellant’s previous expert, and I have adopted it.

One of the main disputes in the experts’ reports was how the integral garage should be valued. Having rejected no.s 2 and 80 Chester Square as reliable evidence, the point recedes somewhat, as only two of the other seven properties do not have a garage.

Mr Nesbit’s view was since it was of lower specification, the floor area of a garage should attract a lower rate per sq ft than the rest of a house, especially in a high value area - hence his preference to value to net sales area. At the appeal property, Mr Nesbit said, the garage represented 8.81% of the total floor area, and to include it at main space rate would ‘result in an artificially high valuation’ especially compared with a house without a garage. He said that garages do not, for example, have air conditioning, or are fitted out with the same audio/visual or heating facilities.

In Mr Nesbit’s valuation, he applied a capital value rate of £2,400 per sq ft to his floor area of 3,808 sq ft to arrive at £9,139,200; he applied £500 per sq ft to the garage space of 368 sq ft, to produce a value of £184,000. Taken together, his value of the appeal property was £9,323,200, say £9,325,000.

Mr Alderton devalued and valued to gross internal area. He considered the likely purchaser of houses in Chester Square would be a high-net worth individual who would value their privacy and security and would look for off-street parking or a private garage. He made no differential between the garage and the main house. Six of his eight comparable transactions were of houses with integral garages and no adjustment was required. The seventh (no.72) was a development project, and the eighth (no.20) has living accommodation where the garage would be. In his experience, the Prime Central London market works on a gross internal basis and makes no reduction in the value attributed to the garage space. He pointed out that had Mr Nesbit applied his £2,400 to the garage, a value of £9,560,700 would be arrived at, which was only £60,000 higher than the price which the appellant paid for the property six years previously, and before they carried out an extensive refurbishment. Mr Alderton also pointed to the 2015 and 2017 sales of 70 Chester Square, where the developer added a garage, and that the appellant did not convert the garage of the appeal property to ‘more valuable’ living accommodation.

In their later reports, Mr Nesbit was immoveable from his position. Mr Alderton produced evidence of the value of parking spaces in Princes Gate, which ranged from £135,000 to £315,000 depending on size and position. These included a variety of ‘add-ons’, including valet parking, and ‘handmade peach skin vehicle covers giving further protection for your car’ (really). The value of others ranged between £150,000 and 280,000. The level of detail provided prevented proper analysis, but the overall picture was that substantial values were attributed to secure parking spaces. This must be the case to a greater extent if the space is integral to the house, and I have little difficulty in rejecting Mr Nesbit’s view. I accept Mr Alderton’s evidence that the market (and, I should say, many previous decisions of the Tribunal) works on a gross internal basis, and that there is no evidence that the presence of an integral garage is anything other than value neutral. It follows that I agree with Mr Alderton that the comparables should be devalued, and the appeal property valued, to gross internal area including garage space.

In order to adjust comparable sales to their equivalent at the market conditions at the valuation date, both experts applied a form of indexation, to account for changes in value over time. They could not agree which index was most appropriate.

Mr Nesbit used the Land Registry’s House Price Index for terraced houses in the City of Westminster between January 2013 and December 2022. He said that the index showed a high point for the market in April 2017 – an average price of £1,735,935 from 254 transactions.

Mr Alderton preferred to use the Savills’ Prime London Residential Statistical Supplement Index, pointing out that both JSRE Partners and Gerald Eve also used it. Mr Nesbit had more confidence in the Land Registry data as ‘the ultimate authoritative source’, being unconvinced that the Savills data is any more granular and being concerned as to the sample size of the data.

Both indices are perfectly respectable; the question is which one is more appropriate in this valuation. Mr Alderton’s criticism of the Land Registry Index is that it uses data across the Westminster Borough, taking in a diverse housing stock.

The specific Savills index which Mr Alderton used was that for houses in Central London (which in addition to Belgravia, comprises Bayswater, Chelsea, Earls Court, Holland Park, Kensington, Knightsbridge, Marylebone, Mayfair, Notting Hill, Pimlico, South Kensington and Westminster). This index shows a high-water mark in June 2014 with the market cooling (apart from a slight rally in June 2015) by the valuation date and continuing to do so until June 2020.

So we have two indices which show peaks in the market at different points in the property cycle – the Westminster Land Registry at April 2017, and the Savills Prime index some three years before. Which is more appropriate in this case? Some assistance can be gained from the market commentaries.

Mr Alderton said that changes to SDLT rates and a 3% surcharge for second homes all fed into the decline from the peak in 2014, although areas such as Belgravia weathered the storm somewhat better than the general market. He referred to Knight Frank’s Belgravia 2017 market review, which indicated that ‘pragmatic sellers were reflecting on their asking prices and…setting levels which reflect the true value as opposed to historic hope’.

According to Savills in April 2017,

In my judgment the Savills Central London House index provides a more reliable guide to changes in houses in the same general price bracket as the appeal property, as evidenced by the market commentary above. I am satisfied that the general Westminster market, encompassing a wide range of property types across the borough, was at a different point in the property cycle at the valuation date. This is also evidenced by the average price shown by that index at the valuation date of £1.7m, compared with the parameters in this appeal which are either side of £10m. Accordingly I have adopted the Savills index in analysing the comparable evidence.

Mr Alderton produced slightly differing versions of the Savills index in his first and second reports, possibly owing to the date they were written (the index is updated from time to time). The calculations in his first report were based on an older version, whereas the version used in his second report showed the same indices as in the version produced by Mr Nesbit. The changes are relatively small, for example for September 2014, Mr Alderton’s first version showed an index figure of 245.6, but in his second the figure was 245.8. Of the other five changes, the most relevant was that for March 2017 - the month before the valuation date – where his first report showed 216.5, and his second, 216.3. That skews the way in which he interpolated for intermediate months, including for the valuation date.

The later Savills index showed March 2017 at 216.3, and June 2017 at 213.5. The point in the month at which the figure is taken is not clear, and it would be tempting to adopt the March figure as a proxy for the valuation date of 1 April. However, from the evidence before me it seems that the most appropriate way of interpolating is to assume a straight line between the quarter dates. That would produce March 2017: 216.3, April 215.4, May 214.4 and June 213.5. Accordingly, the index at the valuation date to which the comparables should be adjusted is 215.4, rather than the 216.3 which Mr Alderton used.

It follows from the above that the starting point for the valuation of the appeal property can be based upon the transactions on seven properties, which took place before, or in the months immediately after, the valuation date. They are to be devalued on a gross internal area basis. Their sale prices are to be adjusted for changes in the market by reference to the Savills Central London House Index, and such adjustment is to be made before any other adjustments, which I deal with later.

This is a long leasehold property on the opposite side of the square to the appeal property. It sold twice, in April 2015 for £10,500,000, and again in August 2017 for £20,671,295.

Both experts discussed these sales at length, but in reality the sales are of limited assistance.

In his first report Mr Nesbit relied on the two sales as being the best example of how an upgrade in specification can ‘move the needle on value considerably, analysing the sales of ‘the same house with the same sq ft’ as adding £1,422 per sq ft after allowing for indexation. This, he said, was ‘key empirical evidence on Chester Square as adjusting for time and knowing the floor plates remained consistent one can be confident that other factors are organically isolated meaning that there is a quantification to the specification parameter applied in my analysis’. He used £1,422 per sq ft for benchmarking, assuming that figure from poor to high specification, and applying 50% to adopt £750 per sq ft as low to average and average to high adjustments elsewhere in the evidence.

Mr Nesbit’s confidence was misplaced. As Mr Alderton said in his second report, when the property sold in April 2015, the main house fronting Chester Square and the rear mews house on Ebury Mews were not linked. The Lonres floor plans for that sale show a floor area of 6,894 sq ft. After refurbishment, and indeed therefore extension, the floor area had increased to 7,100 sq ft, but of more importance the two blocks were linked to allow internal access front to back.

Accordingly, Mr Nesbit’s analysis was defective. Of more concern, despite having Mr Alderton’s second report when drafting his two later reports, Mr Nesbit made no comment on this error, and how it might affect his analysis and valuation.

I place no weight on the first sale of no.70, which was sold unmodernised as a development opportunity of two unlinked buildings. The second sale is also of limited assistance. This was more than the sale of a freehold house with vacant possession; it was as Mr Alderton described, a ‘turnkey’ opportunity. The later sales particulars say that ‘the property’s furniture is bespoke and British-built, with art work curated in-house from selected galleries and artists comprising paintings, photography and sculpture including specially commissioned pieces..’. Neither expert has attempted to isolate those elements from the true freehold value of the building, and their effect on value is unknown. I place no weight on this comparable.

This is on the same north-west side of the square as the appeal property but on the south side of Eccleston Street. Mr Alderton has the gross internal area at 4,941 sq ft, Mr Nesbit 4,947 sq ft. To the very limited extent that it matters, I think Mr Alderton is correct, as his figure is the same as quoted on the sales particulars. There is no garage. There is no lift. It sold in October 2013 for £12,500,000, and after refurbishment again in August 2014 at £15,300,000. The 2014 sale, being closer to the valuation date and when the property had been refurbished is of more use than the earlier sale.

Assuming an index at the valuation date of 215.4, the August 2014 sale (assumed index of 245.9) would equate to £13,398,387.

Indexation for time was the only adjustment which Mr Alderton considered necessary. Mr Nesbit made two adjustments – 7% for the potential outside space which he thought No 20 could accommodate, and £400 per square foot for its better specification.

As regards outside space, Mr Nesbit referred to a lightwell which he considered could potentially provide further space to the ground, raised ground and first floors, and which afforded natural light to the existing space. There are several objections to that being a sensible approach. First, as regards natural light, the lightwell is simply providing light to a media room and sixth bedroom on the lower ground floor, to the kitchen and a study on the upper ground floor, and to the drawing room on the first floor. In my judgment all it is doing is putting those rooms into a state similar to that of the comparable evidence, or at best marginally lighter, in circumstances where rates per sq ft are being compared like for like. Had some space (apart from perhaps the media room where one might imagine darkness would be welcome) had no natural light, a discount might be argued for, but I am not persuaded that the lightwell at no.20 is adding much.

But there is a more fundamental point which is this. From the floor plans provided, I estimate that three floors which could be expanded into the lightwell would amount to around 15 sqm, or say 160 sq ft. Mr Nesbit’s allowance of 7% would equate to over £1 million, or something in the order of £6,700 per sq ft – for space that has yet to be built, before any build costs and in circumstances where he values built space at a tone of around £2,400 per sq ft. Put in that way, to me Mr Nesbit’s addition proves nonsensical.

Mr Nesbit also referred to the rear balcony/terrace. From the plans provided it appears that this was accessed from the main staircase – in the same way as the subject property before this was changed as noted above. If anything, the external space of no. 20 is inferior to that at the appeal property. I make no adjustment for external space.

As for condition, having reviewed the photographs of both, I am not persuaded that any adjustment is warranted when comparing no.20, once refurbished, with the appeal property.

Accordingly, the adjusted sale price of £13,398,387 would equate to £2,712 per sq ft at the valuation date.

This is next-door-but-one to the right of the appeal property. It sold in October 2014 at £15,000,000. Mr Alderton had a gross internal area of 4,674 sq ft. Mr Nesbit did not provide much evidence or comment on the property, which was puzzling since it was two doors’ away. In his schedule he had the floor area at 4,391 sq ft plus a garage of 181 sq ft, totalling 4,572 sq ft. This excludes a vault area (which as I have explained above is included in gross internal area) which has an area of 102 sq ft. Together, they would equate to Mr Alderton’s area.

Mr Alderton’s application of the Savills index resulted in an equivalent value of £13,253,676 at the valuation date. I calculate £13,263,309 (assumed index in October 2014 of 243.6, and at April 2017, 215.4). He made no other adjustments.

Aside from valuing the garage differently, as I have previously outlined, Mr Nesbit made two further adjustments. First, an adjustment of 3%, amounting to £450,000, for ‘Square Position’. In my judgment, no such adjustment is warranted given the close proximity to the appeal property. Secondly, he made an adjustment to reflect No 10 having a superior specification, of £400 per sq ft, amounting to £1.756 million. This was unsupported by any particular narrative, but was probably informed by his analysis of no.70, above.

The sales particulars refer to a ‘fully air conditioned home…having been developed by Finchatton as a private client project about 6 years ago’. Shortly after the property was sold, planning and listed building consent was granted for what was described in the June 2015 Heritage Statement in support of the application as encompassing ‘structural alterations to reinstate stability to the floor structures, remedial works to the roof structure, installation of a new internal lighting system, and small scale changes to the already permitted third and fourth floor alterations.’ The statement went on to explain that the structural alterations were required in part as a result of ‘existing inadequate structural interventions’; the existing main and mansard roof are in a poor state of repair; and that repairs and refurbishment was required to stucco, timber windows and rainwater goods.

The photographs in the sales particulars appear on the face of it to show a high specification, but clearly all was not well with the property. I do not consider Mr Nesbit’s adjustment for quality is justified. It might be argued that No 10 was in worse condition than the appeal property, but taking account of the above I make no adjustment for specification.

It seems to me that this is one of the most relevant comparables, in terms of propinquity and physical similarity to the appeal property. The time-adjusted value of £13,263,309 at the valuation date would equate to £2,838 per sq ft.

This immediately adjoins the appeal property to the left and is an end-of-terrace building with return frontage to Eccleston Street. Both experts base their figures on a gross internal area of 5,597 sq ft.

The property sold in July 2016 for £15,000,000.

While having a Chester Square address, the main entrance to the property is on Eccleston Street. It is externally more ornate, probably designed to ‘bookend’ the southern end of the terrace. The photographs in the sales particulars show a relatively plain internal specification with a dated kitchen. The property had a lift, which the experts agree should be reflected by a 3.75% adjustment.

Mr Nesbit considered the property’s end of terrace position to be advantageous, giving a ‘significant quantum of additional natural light to depth’, and being unopposed by a neighbour to one side. His quantum adjustment was in part informed by a conventional retail valuation where an addition of 5-10% is commonly added for ‘return frontage’. He allowed 7.5%. Mr Alderton considered the dual aspect which the property enjoyed was balanced by the front door being on the, by common ground, busier Eccleston Street. Some windows were bricked up, limiting the extra natural light which the property enjoyed when compared to the appeal property.

I am mindful that Eccleston Street is busier, and that there might be something to be said by being within the terrace and slightly away from the traffic. Later in his first report Mr Nesbit referred to the appellant having run tape around the inside of window frames ‘to prevent the effect of noise’. Clearly this would be of more concern if much of the property faces Eccleston Street. However, I agree with Mr Nesbit that the advantages of having an end of terrace property, with more natural light and having one neighbour, should be reflected in an allowance.

As for condition, Mr Alderton reflected the more basic condition of no.13 compared with the appeal property by allowing £150 per sq ft, or £839,550. Mr Nesbit made no allowance, his schedule saying that the specification was ‘similar’, but with little comment in his report. The photographs show a property in reasonable condition, but empty and in effect a blank page. The kitchen looked again looked to be in reasonable condition, but plain and probably dated.

From the evidence it seems that most buyers refurbish to their own taste and style, and in my judgment there is a danger of making a level of adjustment which in reality would not be reflected in the market. I allow £500,000 for specification.

No 13 was one of Mr Nesbit’s three ‘key comparables’; indeed, he considered no.13 to be the ‘standout evidence’. In his schedule he noted it as ‘significant evidence given location’. (and yet as I indicate above, he was more circumspect when it came to no.10, three doors away).

As for open space, Mr Nesbit said that no.13 had ‘two terraces and three patio areas’ and had the extra balcony to Eccleston Street. The main rear outdoor space is not dissimilar to that of the appeal property. The other ‘terrace’ is a balcony area above the front door, and which is accessed via a landing on the stairwell. It is of limited utility and seems to simply have some potted topiary on it. Two of the areas which Mr Nesbit describes as ‘patios’ are below pavement level in what are the lightwells to the lower ground floor, over which there are security bars. They may have some limited value, but I do not accept that a hypothetical buyer would pay notably more than had they not been there. As above, they simply allow the lower ground floor rooms to have natural light, and this has already been factored into a main space rate being applied to the whole gross internal area. The third ‘patio’ area is potentially of more use, accessed from the kitchen and a breakfast room, but again simply lets light into those rooms. Mr Nesbit’s 8% adjustment, amounting to £1.2 million, is again excessive. For both end of terrace and limited better open space, but reflecting more road noise, I allow 5%.

This is the first comparable where I have both percentage and spot rate adjustments to make. But in what order should they be made? Originally, Mr Alderton made his indexation adjustment before applying other allowances for condition etc. Later in the exchanges of evidence, he agreed with Mr Nesbit that indexation should be applied last.

In The Earl Cadogan v Faizapour and Stephenson [2010] UKUT 3 (LC) [at 32] the Tribunal (George Bartlett QC, President, and Mr A J Trott FRICS) explained that where adjustments are commutative (all based on percentages or multiples) it matters not the order in which the adjustments are made because the end result will be the same (Footnote: ¹ ) . But where some adjustments are percentages, and some are fixed or spot figures, the Tribunal’s preference was to adjust for non-physical factors (time and relativity) before adjusting for physical factors. There was no dispute about the order in which adjustments to reflect those physical factors should be made, and the Tribunal accepted adjustments for condition, location and lateral layout (the properties concerned had been laterally combined), in that order whilst noting that had the adjustment for condition been made last, the end result would have been different.

So, the adjustment for time, using the indices, is made first. Assuming an index of 227 in July 2016 (June was 229, and September 223), at the valuation date the sale price of £15,000,000 would be adjusted to £14,231,278. The question of whether to adjust for the lift (-3.75%) and location/outdoor space (-5%), before or after condition (+£500,000) is largely academic. £14,231,278 x 96.25% (for the lift) x 95% (location) = £13,012,724, plus £500,000 = £13,512,724, or £2,414 per sq ft. Or, £14,231,278 + £500,000 = £14,731,278 x 96.25% x 95% = £13,442,291, or £2,401 per sq ft. Consistent with Cadogan, I take condition before lift and location/outdoor space, and therefore analyse the sale to produce £2,400 per sq ft.

This is an inner-terraced house, located diagonally across the square from the subject property, and to the south of Eccleston Street. Neither expert made an adjustment for location. It sold in July 2017, some months after the valuation date, at £14,500,000.

Mr Alderton’s gross internal area is 6,069 sq ft, Mr Nesbit’s is 5,529 sq ft plus a garage of 436 sq ft, totalling 5,965 sq ft. Neither expert explains the source of their figures, and the very brief sales particulars supplied do not help. I have taken an average of 6,017 sq ft. Whilst on any view this is notably larger than the appeal property’s 4,277 sq ft, neither expert made any form of quantum allowance. The property had accommodation over six floors, with a garage and roof terrace. There was no lift.

It was common ground that the property was sold in an unmodernised state. The internal photographs show a fairly tired specification. Mr Alderton reflected this by allowing £300 per sq ft, amounting to £1.82 million, Mr Nesbit £500 per sq ft, or £2.76 million. Consistent with my comments at paragraph 85 above, and in the light of Mr Nesbit’s £500 per sq ft being informed by his analysis of no.70, I have taken the lower figure.

Mr Alderton makes no further adjustment. Mr Nesbit again adjusted for private outside space by reducing the sale price of no.62 by 7%. But there was little if any narrative in support of that calculation, and it was not clear which areas he suggested were superior or capable of adding value. I make no adjustment.

My devaluation is as follows. £14,500,000 at July 2017 (index 212.9) is adjusted to £14,670,294 at the valuation date (index 215.4) plus say £1.8m for condition = £16,470,294. This would equate to £2,737 per sq ft.

This is in the same terrace as the appeal property, but further to the north-east. It does not face the square, instead being opposite the side elevation of no.80. The upper elevations are brick faced. It has a gross internal area of 4,215 sq ft over four floors.

The date of exchange of contracts (which has been used throughout for the purposes of indexation) is not known, but completion of the sale was on 5 April 2017, within a week of the subject valuation date. The sale price was £11,500,000.

Mr Alderton made no adjustment for time, but reflected the inferior location by adjusting the price by 2.5%, analysing at £2,797 per sq ft.

Here, we have what on any view is a very useful comparable transaction, the sale completing within days of the valuation date. Mr Nesbit makes no reference to it in any of his reports, nor does it feature in his schedule of evidence. That is simply baffling.

While the date of exchange of contracts is unknown, it seems likely that it would have been a month or two before completion (the evidence varies from the same month to three months). I have assumed exchange of contracts in February 2017 (Index 216.9), which after 2.5% for location equates to £11,704,171, or £2,777 per sq ft.

This is diagonally across the square from the appeal property. It sold in November 2017, for £12,800,000, unmodernised ‘as a development project’, having a gross internal area of 4,824 sq ft. Mr Alderton, after adjusting for time, and allowing £300 for refurbishment, analysed the sale at £3,017 per sq ft. Mr Nesbit made little comment.

I am not satisfied that this sale, seven months after the valuation date and requiring substantial valuation adjustment, can safely be relied upon as an indicator of value for the appeal property, and I place no weight upon it.

Adjusting for time, for condition and for other factors can only provide a guide to the valuation of the appeal property. The more adjustment that is required, the more susceptible the analysis is to error. Taking the evidence chronologically, we have the following:

In my judgment 13 Chester Square, while adjoining the appeal property, is less useful as a comparable transaction, because of the number of adjustments that were required to be made to it. The better comparables are no 10 – two doors away and physically very similar to the appeal property, but requiring indexation over a period of two and a half years making it susceptible to inaccuracy, and no 87 – transacting around the valuation date and requiring only adjustment for location. No 62, selling some months after the valuation date, gives some comfort that the analysis of 87 is reliable.

In regard to what is really at stake between the parties, whether the appeal property’s value is above or below £10,000,000, all of the transactions above when applied to it would produce a value in excess of that level (the lowest, £2,400 per sq ft, producing just over £10,250,000).

I also place some weight on the appellant’s purchase of the appeal property in August 2011 at £9,500,000, following which £500,000 was spent on refurbishment. Assuming a June 2011 exchange of contracts, the Savills index was 208.6. This would equate to £9,809,683 at the valuation date before any reflection for improvements. Indexing over such a long period is unsatisfactory, but this would also point to a value in excess of £10 million at the valuation date.

A value of the appeal property at 1 April 2017 based on the sale of 62 Chester Square at £2,737 per sq ft, and 87 Chester Square at £2,777 per sq ft applied to the gross internal area of 4,277 sq ft would produce a value range of £11,707,404 to £11,877,299.

In Mr Alderton’s opinion the value of the appeal property at the valuation date was £11,750,000. In my judgment he was correct.

I therefore determine the market value of the appeal property at 1 April 2017 for the purposes of s.98(8) of the Finance Act 2013 was:

I have outlined above some of the many instances when Mr Nesbit criticised Mr Alderton’s conduct, questioning his professionalism and motives in selecting evidence. For the reasons I give above, these were without justification and were entirely unnecessary. Seldom is this tactic helpful to the decision maker, especially in the setting of an expert Tribunal.

It also misunderstands the role of the expert witness. As the RICS Practice Statement, Surveyors acting as Expert Witnesses 4^th edition (amended February 2023) says (at 17.8.c):

“Expert witnesses would not generally be expected to… include comments that are in the nature of advocacy submissions about an opposing expert’s evidence. You may find yourself at greater risk of slipping into ‘advocacy mode’ at the rebuttal stage of presentation of evidence, when the focus of your evidence shifts from explanation of your own opinion to a more critical role in dealing with the expert witness report of your counterpart.”

In my judgment, throughout his later reports Mr Nesbit succumbed to the risk foreseen by the Practice Statement, and the credibility of his own evidence was the weaker for it.