These are six applications in the form of appeals referred to the Tribunal under Section 1(3A) of the Lands Tribunal Act 1949. The subjects of appeal are representative of retail units at a designer outlet village which was opened early in 1999. The appeals were against the rateable values proposed by the Assessor for the 2000 Revaluation. The issues involved the correct application, in the light of certain features at this development, of valuation principles in the analysis of lease agreements at the development for the purpose of arriving at a Net Annual Value rate. Parties were also unable to agree on a scheme of quantum allowances. In summary, the Tribunal has decided as follows:-
In the light of these decisions, we have fixed a basic rate of £195/per m2 applicable to each of the subjects, and calculated the resultant net annual values accordingly.
The subjects of appeal are Units 3/4, 10/11, 15, 28, 29 and 36 at Sterling Mills Designer Outlet Village, Devonvale, Tillicoultry. This retail development comprises approximately 40 units, and was completed and opened in 1999. Agreement was then reached on the values entered in the new entries on the Roll at that time, but appeals in respect of the assessments for the 2000 Revaluation in respect of most of the units were lodged by both the landlords and many of the tenants. These appeals were referred to the Tribunal under Section 1(3A) of the Lands Tribunal Act 1949, as amended. The six units were selected as representative.
The subjects require to be valued in accordance with Section 6(8) of the Valuation and Rating (Scotland) Act 1956, which is in the following terms:-
“6(8) … the net annual value of any lands and heritages shall be the rent at which the lands and heritages might reasonably be expected to let from year to year if no grassum or consideration other than the rent were payable in respect of the lease and if the tenant undertook to pay all rates and to bear the cost of the repairs and insurance and the other expenses, if any, necessary to maintain the lands and heritages in a state to command that rent.”
In terms of the Valuation Timetable (Scotland) Order 1995, as amended, the valuations are required to be made on the basis of the level of rents prevailing as at 1 April 1998 and on the basis of the physical circumstances of the properties as at 1 January 2000.
Parties were agreed that (apart from considerations of quantum) all the units had to be valued at the same basic rate per square metre, regardless of their location within the development, and that that value had to be arrived at by analysis of the information which was available about lease agreements entered into in respect of about 30 units at or very shortly after the opening date. The value would be expressed as an overall rate, without zoning. The appellants contended for a basic rate of £190/m2. The assessor defended at the hearing a rate of £216/m2.
The appellants also contended for a scheme of end allowances in respect of quantum at two rates, 40% in respect of three of the units at the development and 20% in respect of another three units. The assessor proposed to apply quantum at 17.5% to the largest unit and 10% to two other units.
The appeals were heard together at a hearing on 15 to 17 October, with closing submissions on 25 October 2002. The appellants were each represented by Mr A D D MacIver, Advocate, instructed by Messrs Culverwell and Company, Chartered Surveyors. The assessor was represented by Mr S L Stuart, Advocate, instructed by the Assessor for Central Scotland.
Both sides lodged substantial documentary productions in addition to leading oral evidence. Much of the basic detail was agreed, although there were certain very minor discrepancies in some of the tables of analysis.
The appellants called four witnesses, viz.:-
(i) Stephen L Innes, MRICS, a Director of Culverwell and Company, who was involved as a letting agent in respect of the Sterling Mills development and spoke to the letting experience;
(ii) Nicholas P Draper, FRICS, MIRRV, an associate Partner at Dalton Warner Davis, Chartered Surveyors, London, who spoke to his involvement on behalf of the British Retail Consortium in negotiations with the Valuation Office in relation to a number of factory outlet centres in England and Wales and also to a negotiation with the Assessor for Dunbartonshire regarding the Loch Lomond Factory Outlet Centre at Alexandria;
(iii) Charles Bell, FRICS, principal of Charles Bell Associates, Quantity Surveyors, who spoke to the preparation of estimates of the costs of fitting-out the units at the development;
(iv) John K McClimens, MRICS, a Surveyor with Culverwell and Company, who spoke to the appellants’ valuations.
The assessor called one witness, Ian A Ballance, MRICS, MIRRV, a member of the Scottish Association Executive of the Institute of Revenues, Rating and Valuation, Depute Assessor for Central Scotland Valuation Joint Board, who spoke to the assessor’s valuations.
The Tribunal also carried out a site inspection of the subjects, during which they were shown round the development.
The Tribunal is obliged to counsel and all the witnesses for their handling of these appeals.
The appeal subjects form part of a retail development constructed in 1999 and comprising over 40 units, one of which is a restaurant, and ancillary accommodation including a management suite, public toilets, childrens’ play area and car parking. There are over 7,900 m2 of retail space and over 540 car parking spaces together with 10 coach parking spaces.
This retail development is of a distinctive kind known as factory outlet centres which are of recent origin. These centres allow manufacturers and retailers to dispose of surplus and obsolete stock. The unit leases usually contain a condition that goods must be sold at a certain percentage below recommended retail price. This kind of shopping attracts visitors from a wide catchment area, typically an hour’s drive, particularly where leisure activities are also available. Because of their special character these centres are normally not associated with other retail centres.
The development is located just outside Tillicoultry on the A908 to Alloa. It is within about one hour’s drive from the fringes of Glasgow, Edinburgh and Dundee a catchment area which includes Stirling, Perth, Dunfermline and Falkirk. However, it is not directly located to the central Scotland motorway network, being just off the A91 which connects the M9 at Stirling through the Hillfoots villages with the M90 at Kinross. On the other side of the A908 there is Sterling Warehouse, a well established outlet specialising in furniture and one of the largest of its kind in Scotland. This kind of outlet may also be expected to draw from a large regional hinterland.
While factory outlet centres have tended to be developed in more remote locations, away from full retail price stores, this has changed recently and the BAA McArthur Glen centre at Livingston, opened in October 2000, is located adjacent to Almondvale Shopping Centre and Livingston Retail/Leisure Park.
The centre is inward orientated and laid out with parades of shop fronts lining a number of internal pedestrian walkways. A large restaurant unit (M3) occupies the greater part of a block within the middle of an internal square. At the eastern extremity the largest retail unit (28) is to be found. Across the pedestrian walkway from that unit and surrounded on all four sides by walkways is unit 29 let to USC.
The whole development comprises single storey buildings of steel framed construction with concrete block and brick fill, enclosed by pitched roofs clad with metal sheeting. The shell units provided by the landlord have shop fronts supplied; plastered walls, a floor screed and toilet and sink facilities are provided. The units have been fitted out by the tenants. Such tenant fitting-out typically includes internal partitioning, lowered ceilings, lighting, heating and ventilation systems with associated electrical circuitry.
Altogether there are 41 units separately let in the centre together with public toilets and a management office. The units vary in size from 84m2 to 838m2 (unit 28). The majority have areas between 100m2 and 300m2. Parties have agreed that no particular location within the centre is superior to any other and that a single rate per square metre overall is applicable to all the units, including the restaurant unit (M3), for the purpose of determining rateable value, subject only to adjustment for quantum.
Because of the nature of factory outlet centres, where the tenants are more often manufacturers rather than conventional High Street retailers, the owner of the centre takes a more active management role than would normally be expected. The owner has an interest in immediately establishing a fully let centre in order to attract custom and because he typically charges turnover rent in excess of an agreed base rent. The tenant, and particularly the manufacturer with little retailing experience, seeks reassurance of the promotion of the centre and guidance from the owner on retailing matters. He may also be less inclined than in normal retailing circumstances to commit himself to a long lease given the uncertainty about the success of his own operation and the success of the newly established centre as a whole.
This unusually close landlord and tenant relationship results in leases which themselves are different in certain respects from normal commercial leases for retail outlets. They are characterised by individual negotiations resulting typically in a package including a rent composed of a base rent and a percentage of turnover, lease length, tenant/landlord break options, and certain financial inducements. These negotiations have resulted in distinctive individual packages reflecting different concerns and, no doubt, negotiating strengths.
The rental evidence analysed by both parties covered 30 units at the subjects. The evidence was drawn from leases mainly dating from February 1999 to March 1999. There were twenty two leases agreed in that period. Only one was agreed after 1999, in February 2000. Nineteen of these were in respect of 10 year leases, nine in respect of 5 year leases, one in respect of a 15 year lease and one (unit 29) had a lease of 15 months.
Some of the 10 year leases have provision for a rent review at the fifth anniversary of the date of entry. The revision is to be determined by reference not to market rents but to the Retail Price Index.
The leases are characterised by having provision for breaks at the instigation of either party. For example, in a 10 year lease, the landlord may be entitled in terms of the lease to terminate the agreement in any year, subject only to two months notice during the 2 month period following the end of the relevant accounting period, if the tenant’s turnover had not exceeded the minimum figure specified in the lease. The tenant may be entitled to break on the third, sixth or ninth anniversary of the date of entry, on giving two months notice. Tenant break options have been granted in seventeen of the original thirty lettings at the subjects.
Base rent could be fixed for the term or could be stepped. Excluding the larger units the range of base rent is about £20,000 to about £50,000 per annum.
‘Turnover rent’ is defined in the leases as the amount by which the turnover percentage of gross turnover (as defined in the lease) exceeds the base rent payable in the relevant period. The provision for the percentage of turnover to be paid to the landlord by the tenant varied from 5% to 12% of gross turnover. Aside from the larger units minimum turnover figures are in the range of about £250,000 to £500,000. A unit with a base rent of £30,000 a gross turnover of £300,000 and an obligation to pay 10% turnover rent over a minimum of £300,000, would pay no rent additional to base rent. If gross turnover rose to £400,000 the rent received by the landlord would be the base rent of £30,000 plus 10% of £100,000 or £10,000 giving a total of £40,000. Effectively the 10% turnover rent provision would secure to the landlord a 25% increase in rent for the unit. This gearing effect means that a successful centre with successful tenants results in a disproportionately favourable outcome for the landlord. If the turnover never reached or fell below the minimum turnover figure of £300,000 the tenant might exercise a break option to terminate the lease where that was provided for. The landlord would then be left with a void unit to re-let or the prospect of accepting an underperforming tenant just to secure the base rent.
The necessary strategy for the centre owner therefore is to establish the success of the centre, so far as he is able to do so, by letting as much of the space as possible and by actively promoting the centre. In order to achieve a successful launch of the subjects the owner, as landlord, has made considerable concessions to tenants in respect of not only lease periods and break clauses but also financial inducements.
Within the leases there are examples of straight financial inducements to the tenant to undertake the lease. For example the tenant of unit 3/4 was paid such an inducement in the sum of £125,000 to enter into a 10 year lease at a base rent of £60,000. Sums were paid by the landlord ranging from £30,000 to £160,000 to the tenants of various units to assist with their shop fitting costs. Rent free periods, beyond the conventional short period (which parties agree does not exceed three months) were also commonly offered. These incentives were not necessarily mutually exclusive and can be found in combination. Fourteen of the original thirty lettings had an incentive package which equated to more than a three months rent free period.
It may be convenient here to refer to an extreme example. Unit 29 was let to USC Group plc from November 1999 for a term of 15 months with no base rent. Rent was agreed to be 8% of turnover with no minimum turnover figure. The landlord himself fitted out the unit. It was explained to the Tribunal by Mr Innes that unit 29, which is a separate building and which is glazed on all four sides, was difficult to shop fit and had been difficult to let. However, it was opposite the largest unit occupied by an important brand name tenant and the landlord was keen for this prominently located unit to be occupied. USC as a quality retailer was an attractive tenant. Hence this unusual bargain, heavily biased in favour of the tenant, had been struck.
The variety of lease packages negotiated during this process has given considerable difficulty in the rental analysis necessary to determine a net annual value. The extent to which information from Unit 29 was relevant to such an analysis is one example.
Parties agree the gross internal areas of the units and the factual information from the various leases such as the base rents at 1999, the quantity of financial inducements or the extent of rent free periods. Assessor’s production 5 set out the agreed facts which are used in the parties’ respective analyses – in appellants’ production 6, and assessor’s production 17 (as amended by assessor’s production 24). They have also agreed that it is appropriate to use 6% as a de-capitalisation rate and that adjustment from the date of rental agreement to the tone date of 1 April 1998 can be made by a reduction of 3.5%. There were some very minor matters on which parties were not agreed.
The principal differences between the parties are the adjustment to be made to rents in respect of:
Mr Innes confirmed that at the date of opening of the Sterling Mills centre about 60% of the space was let. In the following twelve months this rose to 80%, and subsequently a 90% letting of space has been achieved. However, he believed the optimum marketing environment for the subjects had now passed, and he predicted that future letting would be difficult. He attributed this to an increase in the number of factory outlets in Scotland from the two that had been in existence when the subjects were developed (Freeport, West Calder and the Loch Lomond Factory Outlet Centre, Alexandria) to the six now in operation. He considered this to be over provision and that only two or three factory outlet developments in Scotland would be successful in the longer term. The proliferation of factory outlets meant that tenants could be selective in deciding on location, and landlords were being increasingly forced to offer incentives to such potential tenants.
To his knowledge early lettings at the subjects had been granted at base rents of £18-20 per square foot (£193.75 – 215.28/m2) on a 5/10 year lease with a turnover top up provision at 10% of gross turnover. Indicators of performance such as turnover rent had been disappointing at Sterling Mills. In the first year of trading, ending April 2000, 13 out of 31 occupiers paid a turnover rent amounting to £96,000. In the year ending April 2001 only 8 out of 36 traders contributed turnover rent, in the amount of £82,000.
Lettings concluded 12 to 18 months later had been made on terms much more favourable to the tenant both in terms of reduced base rent and reduced turnover percentage together with increased financial incentives. He described ‘stock supply’ arrangements which had been entered into whereby the retailer provided the stock and the landlord shop fitted the unit and hired staff. This device was to demonstrate to retailers that there was a demand for their product at this location. If so satisfied the retailer might then be persuaded to take over the sales operation and run it themselves.
Mr Draper has experience with dealing with factory outlet developments in England and Wales where there are now more than 30 such shopping centres. He was able to speak to the approach taken to the valuation for rating purposes of these centres by the Valuation Office Agency. He referred to a report which he had prepared for the Property Committee of the British Retail Consortium (BRC) following his negotiations with the Valuation Officer in respect of the valuation for rating of the Clacton Common Factory Outlet Centre. He had also been involved with the valuation for rating of the Loch Lomond Factory Outlet Centre at Alexandria.
He explained that, in his experience, it was normal to find turnover rent provisions in addition to base rents in these centres. It was customary valuation practice to treat capital payments as financial inducements to the tenant to undertake a lease and to deduct the annual equivalent from the rent and discount it back to the appropriate valuation date. Whereas Valuation Officers had initially sought to decapitalise contributions over the whole term of the lease they now accepted much shorter periods with factory outlet centres. He did not know of any decapitalisation period less than 5 years but thought that the period should be such as to avoid calculations producing a negative value.
Because these centres were often fitted out with shop fronts, toilets and floor finishes by the landlords there was a limited amount of finishing to do by the tenant in comparison with the shell rents in conventional shopping centres. There was no consistency of approach to this facet of lease analysis. In some centres fitting out was ignored, in others additions of as much as 5% were made.
So far as an adjustment for the turnover element in the rents of units in factory outlet centres was concerned, he said that it was normal throughout England and Wales to deduct 50% for the established turnover element. He referred to the BRC report which sated that:-
“In accordance with agreements reached on the outlet centres, the [Valuation Office Agency] has accepted that approximately 50% of the amount by which the adjusted average turnover rent exceeds the adjusted base rent should be adopted as being a fair and reasonable addition to the base rent in terms of rateable value”.
Mr Draper said that it had been accepted both by valuation tribunals and also by valuation officers that the adjusted turnover rents did not represent rateable value because part of the turnover element represented a recompense for the landlord’s acceptance of part of the operating risk. The 50% was appropriate because only the base rent was guaranteed to the landlord at the start of the lease. There was a risk that turnover rent would not be received. It was not guaranteed.
This practice had been widely adopted in the 1995 Revaluation after the decision of the Valuation Tribunal dated 22 April 1997 (unreported) in respect of Bicester Village Factory Outlet. He referred to settlements at Cheshire Oaks, Swindon and Doncaster. The Swindon settlement had been arrived at after a valuation tribunal hearing. He had agreed rateable values for the Clacton Common Factory Shopping Village on the same basis for the 2000 re-valuation.
Mr Draper described the process by which he had agreed rateable values with the assessor for the Loch Lomond centre on a valuation prepared by him making the same 50% adjustment. He had become involved at quite short notice following a change of ownership of that centre and had written to the assessor explaining the approach adopted in England and how it would be applied in these appeals. The agreement reached with the assessor was, he said, on the basis of his approach.
Mr Draper said that turnover rents were not a universal feature of factory outlet centres. To his knowledge 8 out of 42 such centres had less than 30% of the units showing turnover rents. In these situations the element of turnover rent was ignored in lease analysis for rating purposes. In his opinion, where a centre with a turnover lease element displayed an actual receipt by the landlord of such income from less than 30% of the leases then the turnover element should be ignored as a factor in analysing the leases for the purpose of rating valuation. Such a low incidence of turnover rents was indicative of a centre which was performing poorly against expectations.
Mr Bell gave evidence of estimated shop fitting costs he had made for units at the subjects. He had inspected seven units at the centre and had taken measurements. The level of specification and standard of shop fitting at these units had varied very widely. From this analysis he proceeded to prepare schedules of typical shop fitting costs for three different sizes of unit. These costs were estimated using rates current in September 2002:-
|Floor area (m2)||175||205||290|
These figures are exclusive of VAT. Mr Bell agreed that allowance should also be made for professional fees at a maximum of 10%, in line with the scale of fees recommended by the Royal Institution of Chartered Surveyors. He said that these costs should be backdated to the date of entry in line with the Tender Price and Building Cost Index published by the Department of Environment, Transport and Regions.
He was critical of the approach that had been adopted by the assessor to determine tenants’ shop fitting costs because in looking at the returns made by the actual tenants to the assessor he observed a number of significantly odd entries. For example, although shop fronts had been supplied by the landlord many tenants had shown a cost attributable to this item; many items were entered at figures which were well in excess of that which could be considered reasonable; professional fees were often recorded at rates considerably in excess of those to be expected or recommended by professional bodies; and, it was not clear whether tenants had returned costs inclusive or exclusive of VAT. He was concerned that tenants had not properly understood what was required or had been careless in filling in these returns particularly where certain figures were well outside the range of that which might be expected.
Mr McClimens grouped financial inducements both of cash payments and rent free periods together. He noted that about 50% of the initial lettings featured packages of financial inducements equivalent to greater than three months rent free periods. Since these incentives were essential to the conclusion of a bargain to occupy between the landlord and tenant he thought it necessary that the rents should be adjusted for these incentives to arrive at a proper rental figure.
Mr McClimens considered that the appropriate period over which financial inducements should be treated should be no greater than the period up to the first rent review or tenant break option whichever was the earlier. Because in certain cases the tenant had to demonstrate a certain level of turnover to claim a break option Mr McClimens took the occurrence of the second break option as the termination date of the lease. He said that this approach was consistent with the guidelines 2000 Revaluation – Report No 1 – Adjustment of Rents produced by the Scottish Assessor’s Association, and lodged as a production. That document was supported by the guidelines contained in the Red Book published by the Royal Institution of Chartered Surveyors. These Guidelines indicated that where deductions were to be made for capital sums or other inducements the time period over which allowance should be made was the period up to the tenant’s break option, as being the actual period during which the tenant has been ‘locked in’ to the terms of the lease. Mr McClimens observed that this centre had many more break options than would be found in a conventional shopping centre. These options avoided a tenant being tied in to a unit and the landlord therefore only had security of income over the period to the tenant break. At that point he was faced with the possible loss of a tenant and a void period or a reduction in rental and possibly the payment of additional incentives. That prospect for the landlord would necessarily have to be allowed for in a capital valuation of the landlord’s interest in the property and the same principle should apply in the rating context.
Mr McClimens proceeded to adjust the rents of those units where the tenant had incurred shop fitting expense by applying the capital cost for each element proposed by Mr Bell for each of the three sizes of unit.
Mr McClimens was of the opinion that an understanding of the nature of factory outlet centres was vital to an appreciation of the way in which turnover rents were handled for valuation purposes. There was a level of management distinctive from, and additional to, conventional facilities management to be found elsewhere, and the degree of promotion required for these.
Turnover rents by their nature were insecure and volatile. Where turnover rental returns were absent it meant that sales volume was insufficient to cover base rent meaning that occupational costs were a greater proportion of sales than should be the case. Either sales volumes were too low or base rent was too high. Either way the short and flexible leases at factory outlet centres meant that tenants could either vacate or negotiate a lower base rent. These were significant differences from conventional shopping centres and represented an unusual degree of risk to the landlord.
He contended that this aspect of factory outlet centres was well understood by the Valuation Office Agency in England and Wales which had longer and more extensive experience of this type of retail development. It was accepted practice there to adopt only part of this income when valuing for rating purposes, particularly where a low proportion of units contributed such income.
Given these considerations he considered it appropriate to take one third of the turnover income.
Mr Ballance also thought it appropriate to make adjustments to the rents paid for those properties where financial inducements were evident. He chose to treat financial inducements in the form of capital payments and rent free periods separately. This was his conventional approach to the analysis of other shopping developments. In respect of rent-free periods he adopted the same criterion as Mr McClimens and adjusted the rents of those leases exhibiting a rent-free period of greater than three months. He also adjusted for capital payments made to the tenant as an inducement to undertake the lease. Although he considered these concessions as effectively causing a lowering of the rent rather than a reduction of the amount of rent received by the landlord, the net effect was exactly the same.
For both these classes of financial inducement Mr Ballance considered it proper to adjust the rent by an annual amount determined by reference to the term of the lease. He ignored break options in the leases because it was his normal practice to analyse financial benefit in other shopping developments over the term of a lease.
Mr Ballance preferred to use the information provided by tenants in the pro forma returns that they had competed. He discarded irrelevant or erroneous information such as the cost of fitting a shop front. He grouped his shop fitting costs under three headings – heating costs, electrical costs and ceiling costs. To these he added the cost of professional fees as stated by tenants on the returns.
Mr Ballance was of the opinion that where a payment was made by a tenant to a landlord in respect of turnover it constituted rent and should be added to the base rent. Since it was de facto rent it could not be disregarded. Any rent that the landlord actually received in this way was riskless and there was no reason to discount the value of the amounts received for that reason.
The net base rent is arrived at after adjusting the base rent for financial inducements to the tenant to enter into the lease. In the present case this involved both rent free periods and the payment of capital sums.
Mr McClimens treated both of these inducements as capital payments. He calculated the capital value of the rent free period by multiplying the number of months granted by one twelfth of the base rent. He then added the financial inducement to that amount and found the annual equivalent of that total by dividing it by the year’s purchase for the period to the tenant’s break at the agreed rate of 6%.
Mr Ballance treated the two financial inducements separately. He adjusted for the rent free period inducement by determining the present value of the lease, taken to its term but with the rent free period excluded, and deferring the commencement of the receipt of income to the end of the rent free period. He then calculated an equivalent annual rent by dividing the present value by the year’s purchase for the term of the lease. He deducted that amount from the base rent. The resultant figure was added to the amount of any monetary incentive. He then determined the annual equivalent value of both of the financial inducements over the term of the lease at the agreed rate of 6%.
Mr McClimens at this stage chose to identify a base rate per square metre representative of the whole centre. The further adjustments dealt with below, as additions to the base rent, he dealt with on a rate per square metre basis. Mr Ballance proceeded with his analysis on the basis of individual adjustments to all the units in the sample, and then sought to identify a representative base rate per square metre at the conclusion of the adjustments.
In order to find his base rate per square metre for the whole centre based on the information available to him at this stage, Mr McClimens identified a core range from £178.27/m2 to £214.03/m2. These amounts formed ‘shoulders’ to the range beyond which there was a noticeable step change to the next value in the range. He observed that 13 out of the 23, or 57% of the standard units in the sample fell within this core range. These 23 units represented the sample less three units with negative returns at this stage and the remaining large units above about 300m2. The average for all the rents of all the standard units was £187.47/m2. He felt justified therefore in seeking an average from the representative range, which he found to be £194.27/m2.
He considered that the nature of the lease of unit 29 to USC was relevant at this stage of the analysis. That tenant paid no base rent but had received a premium of £72,000; with £17,500 returnable if certain fixtures were removed by the tenant at the expiry of the 15-month lease. Although paying no base rent the tenant was contributing to rent by payment of turnover rental. The standard units within Mr McClimens’ preferred range have areas from 107.74m2 to197.06m2. He considered unit 29, at 216.47m2, also to be a standard unit and he thought it proper to adjust the base rate for the standard units to take into account this premium. He brought the net premium of £54,500 back to tone date value by applying a rate of 6% over the 15-month term of the lease. He then applied a fraction of 13/23, representing the ratio of standard units in his range to the total number of standard units. The resultant calculation produced a base rate of £180.48/m2 which he adopted as the net base rate for subsequent adjustment.
Following the determination of net base rent, parties were agreed that adjustments then had to be made in respect of the addition to the base rate of two items. The first of these was the fitting-out costs borne by the tenant, so far as these contributed to an improvement in the value of the heritage; the second was an addition to reflect the turnover rental element in the leases.
Mr McClimens said that in his experience of valuing shops in the Central Scotland area the assessor had not added any value to shell rents for fitting out costs. He referred to figures from various shopping centres which he claimed proved that shell rents were equivalent to and often higher than rateable values. This proved, in his opinion, that the assessor had not added any value for heritable shop fitting in these higher grade centres. For his part Mr Ballance produced his analysis of agreed rateable values in shopping centres showing allowances for shop fitting. Shop fitting costs varied from £6.45/m2 in retail parks to £30.4/m2 in town centre developments.
To calculate the appropriate rate per square metre Mr McClimens adopted the capital cost for each element of shop fitting proposed by Mr Bell. He then calculated the annual equivalent depending upon the expected useful life of each element at a de-capitalisation rate of 6%. Depending on the type of element the range of useful life varied from 5 years to 25 years. To each aggregated total he added professional fees at 9.75% and the cost of obtaining a Building Warrant, equated over 10 years at £200 pa. He then took an average of the three rates per square metre so derived as follows:-
|Floor area (m2)||175||205||290|
|Rate (£/m2)||13.12||12.92||12.82||Average = 12.95|
These figures were derived from costings made by Mr Bell in September 2002. Using the Building Cost Information Service Index to deflate these figures to the tone date of 1 April 1998 he applied the ratio of 150:177 to obtain an average annual equivalent rate of £11.00/m2.
Mr McClimens accepted that adjustment should be made to the base rate for fitting out, but considered that, in fairness to the ratepayers in setting a level of value, that 50% only of his rate of £11.00/m2 should be added. He therefore made an adjustment to his base rate by adding £5.50/m2.
As already explained, Mr Ballance had grouped actual shop fitting costs under three headings, plus professional fees. He relied on the standard approach which he used to analyse groups of figures in rental analysis. He divided the amount under each head by the floor area of the relevant unit to derive a rate per square metre for each entry in the group. He then calculated the average of all the entries in the group. To take account of extreme entries he excluded the lower and upper quartiles of the group and then calculated the average of the inter-quartile range. In each case the mean of the inter-quartile group was lower than the mean for the whole group and he adopted that to make his calculation of the overall average rate for shop fitting costs at the subjects. He conceded that in using the figure for professional fees from the returns there would be included a greater number of shop fitting items other than the relevant heritable items he had identified. The result of his calculation was as follows:-
|Heritable item||Heating costs||Electric costs||Ceiling costs||Total|
|Inter-quartile mean (£/m2)||30.32||46.67||15.65||92.64|
|Professional fees: inter-quartile mean (£/m2)||(13.95%)||12.92|
|Back dated using BCI Index [1999(ii) = 149: 1998(ii) = 147] = 147/149 =||0.9866|
Having established the capital cost of fitting out each square metre of the units in the development Mr Ballance then discounted the rate using the Tender Price and Building Cost Index. Since the majority of costs occurred in the 2nd quarter of 1999, he applied the same discount, to 2nd quarter 1998, to all units.
He then de-capitalised the cost per square metre over the term of the lease for each individual unit at a rate of 6%, giving a rate of £18.18/m2.
Mr McClimens’ method of dealing with turnover rent was to aggregate the amount received from all units paying such sums in the centre, including Unit 29. He then applied a factor of 33.33% to that income. Finally he discounted it back to tone date by applying the change over that period in the Retail Price Index. That index had been chosen, rather than any rental index, because the amount of rent was directly attributable to retail sales.
For the first year of trading this produced a net base rate adjustment of £5.22/m2 and in year two a rate of £3.99/m2. This gave an average rate over the two years of £4.60/m2. He applied a rate of £4.25/m2.
Mr Ballance took the turnover figures, for the first year only, and applied a straight reduction of 3.5% to bring their levels back to the tone date.
It was at this final stage that Mr Ballance sought to establish a common base rate for the subjects. Having worked the adjustments to base rent through on an individual unit basis, including an adjustment for quantum, he then applied standard statistical methodology to the determination of the base rate. In doing so he excluded unit 29 for the reason that it was the only one to produce a negative result on his analysis.
He determined that the average rate per square metre for the whole sample was £214.38/m2. He removed the lower and upper quartiles and took an average of the inter-quartile range which he found to be £216.48/m2. He considered the latter to be more representative and chose that as his preferred rate.
The equivalent rate for Mr McClimens was derived from his average net base rate with allowance for the subsequent additions as follows:-
|Net base rate||£180.48/m2|
|Add for fitting out||£5.50/m2|
|Add for turnover||£4.25/m2|
Mr McClimens identified the six units in the centre above about 300m2 as potentially suitable for an allowance in respect of size. He grouped the six units into three ‘bands’. By reference to the average rent/m2 he established that Unit 28 had a rental rate approximately 60% of the rate for the standard units. Units M3 and 3+4 had a rate about 20% of that of the standard units and Units 8/9, 10/11 and 14 had a rate about 75% of that of the standard units. Considering these figures he felt justified in applying a 40% reduction to Unit 28, and also to Units M3 and Unit 3+4; and. 20% to units 8/9, 10/11 and 14.
Mr Ballance took a different approach. He equated unit size with the individually adjusted base rates for each unit in the centre. He concluded that only three of the largest units departed from an otherwise close relationship between the two series of data. However, he failed to find any statistical relationship that would enable him to determine appropriate levels of allowance in respect of unit size.
In the absence of any firm basis of analysis he chose to adopt the percentage allowances which had been agreed by negotiation for the previous re-valuation. These percentage allowances had been based on the same rents. Accordingly he adopted 17.5% for Unit 28 and 10% for Unit 3+4.
As a result of their respective approaches to the analysis of rents in the centre and their calculations of base rates the following valuations were produced.
Mr MacIver reminded the Tribunal of the importance of the statutory test under Section 6(8) of the 1956 Act, in particular that the Net Annual Value is the rent at which the subjects might reasonably be expected to let. He referred to the method used by Mr McClimens in analysing the rents and summarised the appellants’ position on the four areas of dispute, which were essentially valuation issues.
Mr MacIver submitted that the appellants’ treatment of the financial inducements, taking these together with the rent-free periods, was correct. There had been inducements in approximately 50% of the initial lettings and tenant break options in 17 out of 30. There was no security of continuing receipt of the base rental income beyond the date of any tenant break option, particularly when the market was not evenly balanced. It was entirely reasonable to devalue the inducements up to the first or second break option. Further, it was appropriate to reflect the premium or inducement paid to USC for Unit 29 by spreading its effect over the thirteen standard subjects in the core range which Mr McClimens’ analysis had identified. Mr McClimens’ choice of his core range of subjects was to be preferred to an interquartile or a median approach. A valuer seeking to establish a basic rate for the standard unit throughout the development had to take into account, and not ignore as the assessor did, the considerable inducement made available to the tenant of Unit 29.
The turnover element of the rentals was, on the evidence, entirely unsecured and could potentially vary widely upward or downward. Mr Draper’s evidence, particularly in relation to the approach adopted at Clacton Outlet Centre, established acceptance by both valuation tribunals and the Valuation Officer that adjusted turnover rents do not represent rateable value, because part of the turnover element represents a recompense for the landlords’ acceptance of part of the operating risk. Approximately 50% had been adopted as a fair and reasonable addition to the base rent, but that adjustment might be larger or smaller according to the proportion of tenants paying turnover rent. On the evidence, the assessor dealing with the Loch Lomond outlets had not disagreed with Mr Draper’s approach. Mr McClimens’ approach of adding 33.3% of the turnover rents for both the year to April 2000 and the year to April 2001 should be adopted. Taking one year’s turnover did not provide the basis for a reasonable expectation, particularly in the first year of a new development. The landlord was bearing an economic risk in relation to this element of the rental. Additionally, the landlord was required to take a much more integral part in the running of the designer outlet village business. The Valuation Office Agency in England had a clear understanding and knowledge of the nature and value of designer outlet developments.
On the treatment of fitting-out costs, Mr MacIver submitted that the appellants’ approach both as to estimating the capital cost and as to the period over which these items should be devalued, was to be preferred. The evidence pointed to serious flaws and discrepancies in the information provided by tenants to the assessor, and there were other criticisms of the assessor’s calculations. The estimated costs provided by Mr Bell were more reliable. Further, the costs should be devalued over the lifetime of the fittings: writing them off over the period of a short lease was not in accordance with normal practice and did not reflect a fair addition to the rental level. Mr MacIver also relied on Mr McClimens’ analysis of valuations of retail premises elsewhere, which he said demonstrated that assessors had not added any value for heritable shop fitting at these other locations. Further, there was evidence of re-lettings at Sterling Mills where rental value was, he said, not being paid by the tenants for the benefit of heritable shop fitting. Application of 50% of the annualised costs represented a fair and proper approach.
Mr MacIver submitted that the analysis carried out by the appellants in relation to quantum allowances was soundly based. Mr McClimens had appropriately selected and analysed the rents of the six largest units. The adjusted rental rate of Unit 28 was approximately 40% less than those of the standard units, and that figure should also be applied to Units M3 and 3/4. The units of around 300m2 produced an average rate of approximately 75% of the standard units and allowances of 20% in respect of them were appropriate and preferable to the assessor’s approach.
Mr MacIver referred to two unreported decisions in relation to factory outlet centres. In a decision relating to Westwood, West Calder, dated 28 June 2001, the Lothian Valuation Appeal Committee decided inter alia that the whole of the turnover rents should be taken into the rental analysis. They had not been satisfied that there was any settled practice in England and Wales on that matter, but the Tribunal had had the benefit of the evidence of Mr Draper and Mr McClimens on that. Further, while indicating that the actual rent was the best evidence, they had agreed that the first year’s rent was probably unreliable. Also, at the time of that appeal, 84% of the tenants were paying turnover rent. The other decision was that of the Valuation Tribunal relating to Bicester Village Factory Outlet, dated 22 April 1997. There, it had been the Valuation Officer’s proposal, accepted by the ratepayers, to take 50% of the turnover uplift.
Mr Stuart first submitted that references to the remoteness of the location of the subjects were irrelevant, particularly because it was accepted that that factor would be reflected in the rents. He then compared the parties’ methodologies in the rental analysis. Mr McClimens’ method involved exclusion of some rents from the outset, but Mr Ballance’s approach of analysing through all the units and only excluding anomalies, such as Unit 29, at the end was to be preferred.
On rent free periods and inducements, Mr Stuart submitted that there was no justification for including as part of an incentive a rent free period of less than three months where on its own, as parties were agreed, it would not be taken into account. Mr Ballance’s approach of annualising the incentives over the period of the lease was preferable in the interests of consistency and to avoid the risk of negative base rents resulting from the adjustment. On the evidence, it was not accepted that the market was not evenly balanced in relation to the opening period. Mr McClimens’ treatment of the incentive paid in respect of Unit 29, resulting in a reduction of £13.79 over the units in his core range, distorted the picture and was inappropriate in an analysis of the rents of each of the units to arrive at net annual value. Mr Ballance had not ignored Unit 29 but merely excluded it at the end because the result was a negative adjusted rent. But for this spreading of the Unit 29 premium, the difference between the parties before consideration of fitting out and turnover was in fact negligible.
On fitting-out costs, Mr Stuart submitted that it was preferable to proceed on actual costs where these could be ascertained, rather than estimates. Many of Mr Bell’s criticisms of costs returned to the assessor related to figures which the assessor had, by his quartiling method, excluded. The assessor’s figure was consistent with the figure for the Freeport development at West Calder. The appellants’ approach of annualising over the estimated useful life expectancy was flawed in ignoring the actual term of the lease. The tenant could not assume that he would occupy for any longer period and could not offset these costs over a longer period. The appellants’ proposed reduction of this figure by 50% on grounds of fairness could not be justified: Mr Ballance had consistently adopted the same approach at shopping centres and retail warehouses. The material submitted in relation to re-lets was of very little weight, being bargains struck some period into revaluation. It was also not known whether these units were re-fitted.
On the issue of turnover rents, Mr Stuart first submitted that the Tribunal should not import English methodology simply because there may be no comparable authority in Scotland (c.f. Forest Hills Trossachs Club v Assessor for Central Region 1991 SLT (Lands Tr) 42 at 48H-K, reversed on other grounds at 1992 SLT 295). Neither the Freeport case nor the Loch Lomond settlement supported a reduction from the full amount of the turnover rent. On principle, a turnover rental payment remained a payment of rent and as such should not be disregarded. There was no basis for figures of 50% or 30%. In terms of the leases, the tenants paid a contribution to a promotion fund which would cover the cost of the promotional activities by the landlord referred to. If the landlord, additionally, adopted a ‘hands-on’ approach, that was a matter for him. The uncertainty whether any turnover rent would be payable, and as to the amount of it, were known at the outset and there was accordingly no risk. There was no basis for applying a discount to take account of the volatility of income flow. The level of turnover income during the year to April 2002 did not disclose a reliable pattern and was of little assistance to a 2000 revaluation. Rentals should not be adjusted because of variations in throughput between revaluations. Mr Stuart referred to Occidental Petroleum (Caledonia) Ltd v Assessor for Grampian Region, LVAC, 25.3.1988, unreported.
In relation to quantum, Mr Stuart submitted that the whole range of subjects should be examined before identifying quantum cases. Analysis of the three largest units in fact showed an inverse quantum effect, indicating that the actual rents should be disregarded as a measure of quantum. The percentages proposed by the appellants were therefore not justified, and Mr Ballance’s proposed scheme, repeating the allowances granted by agreement on the basis of the same rents when the subjects were first entered in the Roll, should be preferred.
Mr Stuart submitted that the Freeport outlet village provided the best comparison for the appeal subjects. Loch Lomond, he said, was not purpose built, and it was not accepted that there had been any concession there on turnover rent. The treatment of turnover rents in the Bicester Village case appeared to proceed on agreement rather than on the tribunal’s decision.
As we understand it, the parties did not take issue on any point of law relating to these appeals. We have to determine, in accordance with the familiar statutory test, the rents at which the subjects might reasonably be expected to let from year to year. We should assume that the parties to this hypothetical tenancy are informed, i.e. have the benefit of the information given in evidence in so far as it assists in relation to valuation at the statutory valuation date. Valuation is on the basis of analysis of a body of rental evidence relating to similar subjects at the location, although there are complications in relation to lease terms which have a bearing on the rental level. Parties agree that in that situation the task of the valuer is to analyse that rental evidence and then, applying valuation expertise, arrive at a basic rate per square metre representative of market value. Essentially, we have to decide on the competing analyses of two experienced expert valuers who have at most points applied very similar methods. The basic rates at which they have arrived are not very far apart but they differ in relation to certain adjustments and aspects of the method of analysis of rental data to reach the respective figures.
We also have before us some competing contentions on comparative material. We do not consider that that comparative material assists in considering, or checking, the value level in this case, where there is adequate rental evidence at the location and where, in any event, the parties’ competing figures do not differ very substantially. The material does, however, have to be considered for such bearing as it has on the treatment of the particular issues in dispute in this case. The material includes decisions of two local tribunals, one in Scotland and one in England, in relation to factory outlet centres, and we require to consider to what extent the decisions and reasons in these cases assist. In relation to practice in England, we accept that we are not required to follow such practice but on the other hand certainly find it of assistance in our consideration of issues which have not yet been fully or consistently worked through in Scotland in relation to this type of subject.
There is a slight difference in the methods adopted by Mr McClimens and Mr Ballance in their analyses of the rents. Mr McClimens was prepared to filter out inappropriate data at the stage when the base rates were adjusted for financial inducements. Mr Ballance rather took a comprehensive approach, leaving any filtering out to arrive at a representative rental rate until the last stage. On balance, we find the assessor’s approach preferable for reasons which will be given later.
There is, however, a more significant problem in relation to one particular unit, Unit 29. This, as we have noted, is a unit of standard size, but the 1999 lease agreement has several a-typical features: it is of very short duration, 15 months; a very substantial capital inducement was paid to the tenant; and there is no basic contract rent, rent being paid only on the basis of turnover. On any analysis, the result is a negative rental figure. Despite the parties’ agreement that, subject to quantum, all the subjects at the centre should have the same value level, the evidence was that Unit 29 was difficult to let and because of its location at the entrance to the centre the landlords were very keen that it should be occupied. Both valuers in fact excluded this rent from their analysis of rents, but Mr McClimens sought nevertheless to reflect the capital inducement by spreading its effect across other standard units. He contended that this inducement was relevant to the rental levels of the other standard units. In our view, this was a mistaken approach. This unit, for whatever reason, attracted a completely un-typical lease agreement which did not fit within the range, however selected, of rents reflected in the analysis. In our view, it follows that no individual feature of that lease agreement can be incorporated into any part of the analysis. This unit should simply be excluded and neither the capital inducement nor the turnover rent applicable to it should be reflected in the value of other subjects. The fact that parties are agreed that the basic value level arrived at for the centre applies to this unit, which is one of the subjects of appeal, as well as to all the other units at the centre, does not affect the treatment of this a-typical rent.
Parties were agreed that the rents analysed were affected by certain inducements or concessions made by the landlords. Three issues arose in this area:-
There was also a very slight difference between the parties’ approaches to the actual calculation of the annual equivalent of the relevant amounts, but we were not addressed on the difference in principle involved. In so far as the results differ, although very slightly indeed, we have taken the assessor’s approach as he apparently followed the method indicated in the Scottish Assessors Association Report and presumably applied elsewhere.
On issue (i), we prefer the assessor’s approach. The conventional treatment of rent-free periods, on which parties were basically agreed, involves ignoring rent-free periods of under three months. This is presumably on the basis that short rent-free periods at the outset of leases, reflecting the period of fitting-out and preparing to trade, are so widespread as not to affect the rent paid for the subjects once the fitting-out is completed and trading has begun. Somewhat inconsistently, however, it is also conventional, and again parties were agreed on this, that where the rent-free period exceeds three months the equivalent rental for the whole of the rent-free period, not just the excess over three months, is regarded as an inducement for which adjustment has to be made. On that scheme, as it seems to us, periods under three months are simply not regarded as affecting the ongoing rental value and it follows that a rent-free period of under three months should be ignored even where there is also a monetary payment. The appellants’ argument to the contrary is that this concession should be seen as part of an overall inducement package, but the logic of that would be that account should be taken of it even if it stood on its own. We do not see why the characterisation of a two-month rent-free period should differ according to whether or not there is also an actual monetary inducement.
Issue (ii), as to the time period over which the inducement is taken to have effect, is more significant in principle. Mr McClimens based his argument on tenant break options, although he recognised that these were often qualified under reference to turnover and therefore in most cases took the second rather than the first break option. Mr Ballance, on the other hand, argued simply for the whole period of the lease.
Here, we prefer the appellants’ approach, which, we note from Mr Draper’s evidence, appears to coincide with the practice adopted in England. It seems to us correct in principle, and in accordance with the guidance given in both the Scottish Assessors Association Report and the R.I.C.S. Red Book, Guidance Note 4, to consider in each case what truly is the period over which the inducement has effect. If the tenant is being locked in over the whole period of the lease that would normally be the correct period, although as the Report indicates that may not be very likely in a market in which inducements are considered necessary. Mr Ballance frankly accepted that Mr McClimens’ approach was not wrong, but relied on the consistency of his approach on this with his approach on fitting-out costs. He also pointed out that the leases here were in any event relatively short, being mostly for five or ten years. Neither of these points seems to us to address the principle involved. The attraction of consistency in the treatment of inducements and fitting-out costs is superficial: each should be considered on a principled basis for its effect in the hypothetical rental market. It is not disputed that these are inducements which affect the rental and not merely artificial elements affecting only the calculation of rental yield. There is no reason to doubt the reality of the tenant break options, which of course may induce new rent negotiations even if the tenant remains. It is therefore in our view correct to base calculations on the periods before the breaks and not on the full periods of the leases. We also accept the discriminating way in which Mr McClimens carried out this exercise, in the light of the qualified nature of the tenant break options.
On issue (iii), we have already indicated that we do not regard the lease terms of Unit 29 as appropriate for inclusion in the analysis. The appellants’ attempt to have the substantial monetary inducement for this unit spread over the standard units whose analysed rents have been used to arrive at the basic rate, whilst at the same time accepting that Unit 29 could not be so used, is inconsistent and unacceptable. Once it is recognised that even although it is a unit of a standard size and the parties agree that its basic value level is the same, the actual rental agreement which is found is so far out of line with the others as to be unusable, the temptation to reflect the inducement in its case must be resisted. To take parts of lease agreements in this way is to take material which is irrelevant to the analysis of those rents which are used to find the average rate. We have therefore not accepted this part of the appellants' calculation of the annualised value of the inducements.
The monetary value of both the rent free periods and the financial inducements are shown on the appendix to this opinion. These capital amounts are converted to their annual equivalent by applying the agreed rate of 6% to the period of time to the tenant break suggested by Mr McClimens or to the first opportunity for review of the rent whichever is the earlier. The annual equivalent of the financial inducements is then deducted from the contract rent to produce the rental equivalent, consequential on taking into account these inducements. The equivalent rent is finally backdated to 1 April 1998 by reducing that equivalent rent by 3.5% to allow for the one-year difference between the tone date and the date of the leases.
In this area, there is agreement that at least some of the cost of fitting out the units with heritable items, again appropriately annualised, should be added to the lease rents. Again, there appear to be three issues:-
On issue (i), the assessor’s approach involved carefully identifying the types of relevant costs included in the returns and seeking to eliminate extreme figures by using an inter-quartile approach. This seems to us an entirely reasonable approach which has obviously reached a reasonable result. However, reference to the returns does support the appellants’ contention that some of the figures returned (and as to which, quite understandably, there was no further enquiry) appear slightly dubious, and one or two of these figures found their way into even the inter-quartile analysis. For example, some of the figures for professional fees appear problematic. In each of the four areas of costs taken in by the assessor, there is a wide range of figures within the inter-quartile range.
In the circumstances, we think that where a professional quantification of typical costs has been obtained and not in itself seriously questioned, we should adopt that as giving a more reliable estimate. Equally we consider the appropriate method of back-dating the amount so determined is by the application of that index which is specific to the work being costed.
Issues (ii) and (iii) are, we think, best approached together. The appellants did not suggest that there should not be any addition for fitting-out costs, although Mr McClimens claimed to have established that analysis of the assessments at a number of retail locations in Scotland, in the light of the rental evidence, demonstrated that nothing had in fact been added. Mr Ballance asserted that in his valuation area these costs had been added, and produced examples suggesting that his figure of £18.18/m2 (which falls to be slightly reduced in the light of our decision on issue (i)) fell reasonably into the range of figures found at other retail locations. We note that the Scottish Assessors Association Report directs that the actual costs, adjusted for any elements of unremunerative expenditure or non-heritable items, should be divided by the years’ purchase for the whole term of the lease and the resultant annual equivalent added to the rent. It appears that although there has been a widespread practice in situations involving 25 year lease patterns (which Para. 5.1.2 of the Assessors Association report appears to assume) to take either the actual or a broad estimate of the fitting-out costs and decapitalise these over the whole period of the lease, there have also been situations in which this item has simply been taken as a quite modest percentage of the rent. As far as designer outlet centres are concerned, Mr Draper’s evidence was that in England and Wales additions were ‘sometimes’ made, on a very arbitrary basis, of up to 5% of the basic rate. Although this issue was canvassed in the Freeport case, and determined on a percentage basis it is not clear to us from the decision whether this was a percentage addition to the basic rate or rather a decapitalisation percentage of some of the costs.
In principle, we prefer the appellants’ approach to this issue. There appears to us to be no good reason for measuring the additional rental value of a retail unit created by a physical item on the basis of the pattern which prevails at the particular location as to the duration of leases.
The evidence provided in this case of different lifespans for different elements of the fitting out shows that a general 25 year period may be on the generous side. We suspect that such a period may simply have been generally accepted as reasonable, perhaps bearing in mind a feeling that fitting out at retail locations would not always add to the rental value to the hypothetical tenancy. The latter consideration would be more accurately addressed by considering whether any particular items represent unremunerative expenditure (as indeed the Assessors Assocation Report directs), but it would not be at all surprising if a more general and approximate approach were followed in practice.
What is clear to us is that we can find no support either in principle or in any rental evidence or equity comparison for calculating annual equivalents over the much shorter lease periods which are found at Devonvale. There is no suggestion that the Assessors Association Report, which was indeed produced by the appellants, has the status of an agreement: if its instruction to take the period of the lease was intended to apply to shorter leases and if that were based on rental evidence, we have had no such evidence produced to us.
The question for us then is whether to adopt the 25 year period which appears to have become conventional, (but which neither side asked us to do), or to adopt the appellants’ more accurate approach. While there is quite a lot to be said for the more general approach, which in effect takes a conservative overall period, we think that since in this particular case the appellants have, by considering the lifespans of the various types of works, provided a more accurate estimate of the value, which we should adopt.
That does, however, leave issue (iii), the appellants’ claim for some discounting of this addition to the basic value. We have difficulty in finding a proper basis for this. There is no suggestion of unremunerative expenditure, the costs in question being, as we have accepted, based on estimates of typical, standard, heritable fitting-out. As we understand it, although contending only for a discount of 50%, the appellants relied on two strands of evidence which they suggest demonstrates that there is in fact no additional value. Firstly, they sought to show by analysis that in the assessment of various retail locations, no addition had been made. Secondly, they refer to four re-lettings at Devonvale, in three of which the new lease terms did not show any rental increase. They then argue for the discount on the basis of the requirement for fairness to the ratepayers. We are not persuaded by either of the two strands of evidence. We accept Mr Ballance’s evidence that such additions on the basis of the (25 years’) lease duration are regularly made, in accordance with the Assessors Association Report, except, no doubt, where there is unremunerative expenditure. The evidence of the four re-lettings relates to the period in which, we were told, letting conditions had become more difficult, so that the mere fact that the rents were no higher than the initial rents does not in our view establish the position at revaluation. That does not leave any basis for a discount.
We have not overlooked the assessor’s reliance on the figures for Thistle Marches (£30.41), the retail parks (£6.45) and Freeport (£17.86). We do not think that comparison with modern shopping centres, in which the lease are of basic shells, is of much assistance, and we have not been provided with any evidence as to the extent of fitting-out at Freeport, although no doubt it may be expected to be similar. Mr Draper’s evidence suggests a rather lower figure in England and Wales, although again we do not have a sufficient basis for proper comparison. If indeed there was no addition at Loch Lomond, it should be remembered that this is a totally different type of building, being adapted industrial premises.
The result of our consideration of fitting-out costs is that, applying Mr Bell’s estimated costs, back-dated, and annualising on the basis contended for by the appellants but not allowing any discount, the addition for fitting out costs is £11/ m2, compared to the appellants’ £5.50 and the assessor’s £18.18. The resultant amounts are shown in the appendix.
In the Bicester Village case, the Valuation Officer’s representative is recorded as observing that turnover rent was “a concept far removed from the definition of rateable value”. It did not provide a guaranteed income for the landlord, and it fluctuated with the retail market. We are not sure that Mr MacIver went so far as to adopt this, but at all events we do not accept it as a general proposition. We appreciate that there is something slightly uncomfortable about rent levels of shops being established by reference to the turnover of tenants’ particular businesses, but in our view the method of calculation of rent does not, as a matter of general principle, affect the categorisation of an established level of payment for the occupation of heritable subjects as rent. Nor, in our view, does reference to landlord’s risk assist in categorising payments made under leases. No rent is completely free of risk. Rents subject to review each year would be at similar risk, but would be nonetheless rents. Turnover rents might go up as well as come down.
We therefore agree in general terms with the assessor’s approach to this issue, and to the extent that the Bicester Village decision proceeded on the basis of general principle, which we must say is not clear to us from our reading of the reasons given, we do not follow it. In so far as the Valuation Appeal Committee considering the Freeport case expressed a similar view in principle, we agree with that committee.
We also reject the submission that the landlord’s additional involvement in the operation of this factory outlet centre justifies some arbitrary deduction from the turnover element of the rent. We do not doubt that this landlord is involved more than landlords at ordinary retail locations on the retail, as opposed to the property management, side, and this seems to be a feature of factory outlet locations. We could see an argument that the cost of a service to tenants which is not envisaged in the statutory hypothesis should be deducted in the analysis of all the rents, but in the present case there is specific provision in at least most of the leases for that cost to be reimbursed. To the extent that the landlord goes beyond provision of services stipulated in the leases, we agree with Mr Stuart that the landlord does that in his own interest. At all events, we do not see this as relevant to the question whether account should be taken of the turnover rent.
Matters do not, however, in our view end there. We have to determine a rental value representative of the market rent. In many situations, a turnover rent, whether comprising the whole rent or, more commonly, a small proportion added to a fixed contract rent, provides a reasonable estimate of what the market will pay. This is the way it is envisaged in the Scottish Assessors Association Report, which refers at Para. 10.0 to base rents fixed at a level of around 75% to 80% of the full market value. As we have said, we have no difficulty in principle with the idea that market value can be established by the inclusion of turnover rent where it can reasonably be concluded on the evidence that the base rent falls short of the market value. Where there is a reasonable pattern of turnover rents establishing a level of market value above base rent, that higher level does not require adjustment.
That situation might be confirmed by consideration of levels of market value in the surrounding area, or by evidence to the effect that the base rent was fixed below market value, or simply, often enough, by looking at the pattern established by a body of rents.
The particular difficulty in the present case is that we are dealing with opening rents at a centre of a type and at a location where it is not possible to look at established levels. The base rents obviously provide a floor, but can we be satisfied in the particular circumstances that the average, properly analysed, of the turnover rents establishes an additional component to be added to the base rent to give market value?
In our view, both the pattern and the level of turnover payments in the first year of operation of this centre suggest that the level arrived at primarily from a small handful of turnover ‘top-ups’ is not, without modification, a safe amount to add to determine the market level for this location.
As to the pattern, we first note that in the whole centre (including unlet units) only 13 subjects, or about 30%, paid rent comprising basic rent plus turnover rent. Within the sample of subjects analysed Unit 28 should be excluded, since there was no agreement for any turnover payment in that case. We have also excluded Unit 29 as untypical in various respects (and even if it was to be taken in for this purpose, it would be misleading simply to take its turnover rent, because there was no base rent). That leaves 28 actual rental agreements available for analysis. The ‘top-up’ was achieved in only 12, or about 43%, of these. Therefore the prevailing pattern, both in the centre as a whole and in the sample, is that turnover is not established to be a normal component of rent in this centre.
As to the level of turnover payments in the first year, we note (again excluding Unit 29) that such payments totalled some £78,000, or roughly 6% of the market rental value estimated by the parties, of which sum around £49,000 (or 63%) came from just three subjects (Units 8/9, 16 and 24). The distribution of turnover rental payments therefore is skewed to the extent that an average figure will not be representative of the expected performance of a typical standard unit in this centre.
In these circumstances, while seeing no objection in principle to taking the whole amount of turnover rents where these, added to the contract rent, establish the market value, we consider that in the present case taking in the full amount of the turnover payments, the majority of which emerge from just three of the relevant subjects, does not give a safe view of the market level of rent.
We do not consider it appropriate in the circumstances of this case to take into consideration the second year’s turnover income, which came in at a rather lower level than the first year. In general, we do not necessarily exclude evidence which would not have been available to the hypothetical parties when the rent was fixed: parties are taken to be reasonably informed. We think it appropriate to take the actual figures, which only emerged very shortly before the hearing, in preference to earlier estimates. There is some attraction in having more than just an opening year’s results, and we would not necessarily exclude evidence from outside the relevant quinquennium if that assisted in reaching a view as to the market level at the particular date. However, the second year’s figures, i.e. the figures for the year to 30 April 2001 seem to us to be possibly unreliable, bearing in mind the evidence which we heard of some deterioration in market conditions starting in mid- to late 2000. In Occidental Petroleum, Lord Clyde observed:-
“While it was not suggested that at least by way of a check it was incompetent to look at evidence as late as 1986 the Committee were clearly entitled to disregard it if they considered it to be so long after the tone date that it could be affected by circumstances which would not have affected it at that date”.
We next have to decide whether any of the turnover rent should be taken in, and if so how much. We think it would be extreme and unjustified to ignore all of these payments, which are clearly rent. The vast majority of tenants at the centre, all but Unit 28, have agreed to rents based on consideration of turnover, and in the first year almost one third of the tenants have actually paid rent related to turnover. The appellants did not seek to persuade us to reject this element altogether. While there might be more refined ways of arriving at a figure (for example, discounting to some extent the three turnover ‘top-ups’ which seem untypically high), we have decided to take 50% of the relevant turnover rent. It seems to us that the best assessment we can make is that the hypothetical landlord and tenants would consider the probability as to whether the anticipated turnover would contribute to the rent payment or not. On the evidence available in this particular case, we think that an equally divided view of that probability reflects a reasonable market view of the rental level which the landlord and the tenant of a standard unit would arrive at, at the outset of this new and somewhat speculative development.
Since this element of rental value is arrived at on the basis of retail turnover, and since we have no local information about value movements during the period in question, we consider it appropriate to follow the appellants’ suggestion that back-dating of this element should be based on the Retail Price Index at 1.6%.
The practice in England and Wales appears, on the evidence, to have been that 50% of the turnover rent has generally been taken but that there may possibly be a move towards a higher level, or even 100%, at successful centres where high levels of turnover payments are established. The Loch Lomond settlement may possibly be in line with that. While we do not necessarily agree with the rationale as it was explained to us, such a practice appears consistent with our approach. It should become easier, with the passage of time, to establish to what extent turnover rents reflect actual market rental values at particular locations.
The method of analysis of the rental information by both sides, to determine the basic rate per square metre for a standard unit in the development, has already been explained. The base rent is modified by deducting the annual equivalent, at the tone date, of monetary incentives in the form of either rent free periods or cash payments, to produce a net base rent. We will call that the first stage of the analysis. That adjusted rate then has added to it annual equivalents of both the capital cost of fitting out and the amount paid in respect of turnover. That is the second stage. In the absence of any positive means of determining what the market value of the units should be from external evidence this normative approach, by careful estimation of each of the relevant factors, enables a calculation to be carried out from which the resultant figure should approximate to market value.
Given the considerable variety of lease arrangements, both in the combination and extent of each of the factors, there is a wide range of results for the units making up the shopping centre. It might also be expected that differences in size between units, at either end of the size range, would also be a variable relevant to differences in the rate per square metre obtained. This can be seen in the last column of the appendix, where the equivalent rent for each unit is expressed as a rate per square metre relative to its gross internal floor area. Excluding negative results the range is from about £85 to over £300. The differences in result within the whole sample are so great, that both parties accept that taking a simple average of the rate per square metre for the whole sample will give an unsatisfactory result. Some means of identifying the rate most likely to be applicable to a typical standard unit, within a size range for such a unit, requires to be found. They agree that a more reliable rate can be found by discarding extreme results from the range of outcomes and by attempting to define a shorter range on either side of the median result. Both agree, for example that Unit 29, which gives a negative result at the end of both the first and second stages, should be ignored, (although, as we have seen, Mr McClimens thought it proper to include the amount of the factor of financial inducements into his remaining range of units, and into his calculation of the amount to be added in respect of turnover rent).
Mr McClimens chose to make his selection of the more representative range of units at the end of the first stage of the process. He did so by identification of a core range by observation. He saw a range of thirteen standard-sized units which he judged sufficiently close together and sufficiently distinctive from the next result at each end, to justify selection. The difference between the top and bottom of the range of 13 units is only £36, while the gap between the next unit at each end is over £20. In effect Mr McClimens has removed the lower 40 and upper 17 percentiles of the whole sample and selected a range of 43%. The average of that range is £194.27.
In our opinion this judgmental approach has merit. The additional amounts added are analysed with respect to the whole sample, or all units potentially liable to pay turnover rent, and are therefore not limited by the selection of core units. This is a reasonable enough attempt to find a reliable answer, in circumstances where there are no agreed approaches for analysing a large number of disparate leases in these circumstances.
The assessor adopted a comprehensive approach but was criticised for taking too mechanical an approach to the determination of a more representative range by simply removing the lower and upper quartiles in the sample. However, we do not consider this to be an unthinking approach. Applied by an experienced surveyor we consider this to be an equally reasonable attempt. The removal of unit 29 at the end of the second stage was a deliberate and appropriate action carried out to ensure a best approximation to a representative figure.
In effect both parties achieve much the same result by filtering out extreme results and attempting to identify a range of results around the median which were representative of the rental value per square metre. But because of the differences in approach at stage two of the process Mr McClimens’ core sample and Mr Ballance’s inter-quartile range comprise different units. The final result is therefore different.
We are conscious that we are dealing with unusual circumstances where solutions may have to be pragmatic and should not be regarded as of general application. We prefer Mr Ballance’s approach. Given the heterogeneity of the individual leases in this development it appears sensible to analyse each effect of the factors on the sample in the aggregate before filtering out results that dilute the distillation of a basic rate, by whatever technique is deemed appropriate. Further, Mr McClimens’ method of dealing with the turnover rent means that he includes that attributable to unit 29. For reasons given earlier we are of the opinion that unit 29 should not be included in the analysis. Mr Ballance rejected its inclusion at the last stage.
Given the remarkable range of outcomes of the analysis of individual leases, we consider it prudent to include the whole range of results (apart from Unit 29) in the final analysis from which the inter-quartile range will then be drawn, even although the application of our decisions actually results in two other leases (Units M3 and 26) analysing at (small) negative figures (£-36.16 and £-8.26 respectively): these are part of the overall picture.
The result produced by the application of the preferred method is £194.47/m2, as the average of the inter-quartile range of the subjects (apart from Unit 29) in the final column in the appendix. We adopt this rate (called £195) for the purpose of deriving a net annual value for each of the subjects of this reference.
We are of the clear view, in agreement with the parties, that the three largest subjects, Units 28, M3 and 3/4 are sufficiently outside the main range of subjects as to attract, in the hypothetical market, allowances for quantum. We note that as there is no zoning in these valuations there is no other way in which their size would be reflected. These three units are respectively 838, 546 and 534m2, the next largest being at 309m2. While their analysed rents do not show a consistent pattern, they are each measurably below even the slightly descending straight line shown on the assessor’s graph of unit size and rent before quantum. This picture does not change as a result of the final figures produced as a consequence of this decision.
Beyond that, we do not find the problem of quantum to be readily soluble in this case, because the pattern of rents, analysed after the various adjustments, is very unclear, particularly among the larger subjects. In the final analysis, the rental rates of the six largest subjects, in descending order of size, are £112, £85, -£36, £183, £158 and £114. The rates for six units of the same size (131.25m2) range from -£8. to £260. Although it is possible to draw a straight line, which descends slightly, across the graph of all the subjects, the linear correlation between area and rent is very weak. The line does indicate that rent tends to decline relative to increase in size. But it is not possible to apply any statistical tool to identify the appropriate degree of quantum allowable. In this situation, we have simply to do the best we can, as a matter of judgment and degree.
The first conclusion we have reached is that the appellants have failed to satisfy us, on the rental evidence, that the next three subjects, Units 10/11, 8+9 and 14 have attracted, or would attract, lower rental rates by reason of size. These three subjects range in area from 296 to 309m2. Their rents range from £114 to £183, well below the average rate for a standard unit, prima facie indicative of quantum allowance, but also within the range of rates for middle size standard units. We accept, for the reasons given above, that both individual and collective rents are poor indicators of allowances for quantum in the circumstances of this development. It also appears to us that, while it is possible to draw a line which descends (very slightly) with size, there is very broadly a range of subjects between 100 and 300m2 in which there is no very clear quantum effect. We note that in introducing this argument for these three subjects, as well as the largest three, to have allowances, Mr McClimens referred in his precognition to “the six largest units in the development, those of more than 300m2 or thereby”. These three subjects are, however, “300m2 or thereby”. We are not persuaded that any quantum effect is demonstrated in relation to them, or that they are necessarily deserving of such an allowance.
Coming back to the three largest units, we agree with the assessor that the derived rental rates at this end of the quantum scale cannot be relied on in relation to quantum. We do not, however, agree with his decision simply to revert to the quantum allowances agreed in the running roll appeals in relation to the short period, in the previous quinquennium, after the centre was opened. There was no indication of any particular basis for these figures and we therefore did not consider that they are of any real value in these revaluation appeals. We are left simply to make the best assessment we can, and in doing so have considered the extent by which these units exceed in size the top end of the standard range, i.e. around 300m2. The allowances at which we arrive are:-
We have calculated, and shown in the appendix, the effect of our decisions in relation to the rental analysis, resulting in a rate of £195/m2. The resultant net annual values of the six subjects of appeal, after applying quantum allowances where appropriate, are as follows:-
We accordingly allow the appeals to the extent of directing that the above Net Annual Values be entered in the Roll.