Chartbrook Ltd v Persimmon Homes Ltd

On October 16, 2001 an agreement for the development of Nos 1, 3, 5, 7 and 9 Hardwicks Way, Wandsworth (“the Agreement”) was entered into between (among others) Chartbrook Ltd (“Chartbrook”) and Persimmon Homes Ltd (“Persimmon”). The other parties were Persimmon plc as guarantor of Persimmon’s liabilities, and Mr Stephen Vantreen (a shareholder and director of Chartbrook) in his capacity as adjoining owner.

The site was owned by Chartbrook. Under the Agreement Persimmon was to obtain planning permission, construct a mixed residential and commercial development on the site (with commercial premises below, flats above and basement car parking), and then sell the flats. Persimmon did not acquire any interest in the land, but (although contracting to sell on its own account and not as agent for Chartbrook) was able to require Chartbrook to grant leases to complete the flat sales.

The principal persons involved, or concerned with, the negotiations leading up to the Agreement were these. On the Persimmon side, Mr Nicholas Pendlebury, who was an employee of Persimmon Homes (South East) Ltd and later Land Director, carried out the day to day negotiations under the supervision of Mr Brendon O’Neill, who became Managing Director of Persimmon Homes (South East) Ltd prior to the conclusion of the Agreement. Persimmon’s solicitor was Mr Robin Assael, a partner in the well-known firm of Hammonds Suddards Edge. The negotiations on behalf of Chartbrook (a single purpose vehicle) were conducted by Mr Stephen Vantreen and Mr Bob Reeve, who were experienced property dealers (but not property developers), and were its directors and its ultimate owners. Chartbrook’s solicitor was Mr Sean Skelly, a very experienced specialist in conveyancing and a partner in Skelly and Corsellis.

By clause 11 of the Agreement: “The Price shall be calculated and paid in accordance with Schedule 6.” By clause 1.1: “‘Price’ means the sums calculated in accordance with Schedule 6.”

In Schedule 6, paragraph 1, “‘Price’ means the aggregate of the Total Land Value and the Balancing Payment.” The elements of the Price are also defined in paragraph 1.

The first element, Total Land Value, means “the aggregate of the Total Residential Land Value the Total Commercial Land Value and the Total Residential Car Parking Land Value.” Those terms are defined as follows:

“Total Residential Land Value” shall be £76.34 per square foot multiplied by the Residential Net Internal Area (less the Section 106 Money and less the Rights of Light Money and less the Sub-Structure Assumptions Additional Cost)

“Total Commercial Land Value” shall be £38.80 per square foot multiplied by the Commercial Net Internal Area plus VAT.

“Total Residential Car Parking Land Value” shall be £3,024 multiplied by the Residential Car Parking Spaces

The second element in the price is “the Balancing Payment” which is defined in paragraph 1 of Schedule 6 as “the Additional Residential Payment,” where:

“Additional Residential Payment” means 23.4% of the price achieved for each Residential Unit in excess of the Minimum Guaranteed Residential Unit Value less the Costs and Incentives.

“Residential Unit” means each of the flats forming private residential accommodation for which Planning Permission is granted

“Minimum Guaranteed Residential Unit Value” means for each Residential Unit the Total Residential Land Value divided by the number of Residential Units for which Planning Permission is granted.

“Costs and Incentives” means the aggregate of all costs and incentives provided by the Developer for the purchasers of the Residential Premises and the Residential Car Parking Spaces including the cost or allowance given to purchasers for enhancements or variations to the specification for such premises.

By paragraph 3.1(a) of Schedule 6 the Total Land Value element was to be payable in instalments on the earlier of the date of sale of the units or the last day of a series of specified months set out in paragraph 3.1(a) from the “Implementation Date” (in effect nine months after planning permission: clause 1.1 of the Agreement), starting with the end of month 25 from the Implementation Date and ending with month 52 (i.e. in effect between about 3 and 5 years from planning permission).

By paragraph 3.1(b) the Balancing Payment was to be made on the earlier of the date of the later of the sale of the last Residential Unit and the last Residential Car Parking Space, or 6 months following completion of the Development. In the event that they were not all sold within 6 months of the completion of the Development, the Balancing Payment was to be determined by agreement or by a fair open market valuation pursuant to a dispute resolution procedure set out in clause 6 of the Agreement. That procedure was not triggered. Paragraph 3.3 is only relevant for a point on construction of the Additional Residential Payment to which I shall revert. It provided for payment of ground rent value to be payable on the earlier of the date of realisation of the ground rent value “or the date of payment if any of the Balancing Payment.”

The dispute turned on the construction of the definition of Additional Residential Payment, which for convenience I repeat:

“Additional Residential Payment” means 23.4% of the price achieved for each Residential Unit in excess of the Minimum Guaranteed Residential Unit Value less the Costs and Incentives.

In common with the judge, I refer to the Additional Residential Payment as the “ARP”, the Minimum Guaranteed Residential Unit Value as “the MGRUV”, to the price achieved for each Residential Unit as “the Unit Price” and to the Costs and Incentives as “the C&I”.

Chartbrook’s case was that it was entitled to a 23.4% share of the net proceeds of sale of each Residential Unit in excess of a minimum guaranteed amount (being the unitised Total Residential Land Value of £76.34 per square foot of Residential Net Internal Area), i.e. its stake in the residential part of the development was to be the whole of the first £76.34 per square foot of net sales value, and 23.4% of the surplus. Since the sales proceeds were bound to exceed the minimum amount by a long way this was bound to throw up a substantial extra payment in any event. This payment was referred to for convenience in this case as “super overage.”

Persimmon’s case was that Chartbrook was to receive either a fixed percentage (23.4%) of the sales revenue or the minimum guaranteed amount, whichever was the greater. There would only be a Balancing Payment if 23.4% of the net sales proceeds of the flats exceeded the minimum amounts. This had the effect of giving Chartbrook a share of the upside if sales exceed expectations (i.e. a “sales overage”). Chartbrook was to receive an additional payment only if 23.4% of the net sales price amounted to more than the Minimum Guaranteed Residential Unit Value, i.e. Chartbrook’s stake in the residential part of the development was whichever was the greater of: (1) 23.4% of the net residential sales price; and (2) the guaranteed minimum of £76.34 per square foot of Residential Net Internal Area.

The difference had a major impact on the amounts payable to Chartbrook:

Briggs J after a trial of the action and counterclaim held in favour of Chartbrook’s construction and rejected a counterclaim by Persimmon for rectification.

Chartbrook acquired Units 7 and 9 in January 2000 for £850,000. In January 2001 it acquired an option to purchase Unit 5 for £450,000, plus a figure for “Enhanced Value” which was to be calculated on a formula designed to give the vendor 30% of the increase (attributable to planning permission) in the value of the site. Mr Vantreen expected the increase to be between £155,000 and £200,000. In July 2001 it acquired an option to purchase Units 1 and 3 for £800,000.

In early July 2000 Cathedral Group plc offered £4.1 million for Units 7 and 9, subject to “planning overage” at £50 per square foot of net internal development area in the event that the residential element for which planning permission was granted exceeded 50,000 sq ft. In late September 2000 Cathedral Group offered £5,180,000 for Units 5, 7 and 9 conditional on planning permission.

Cathedral Group hoped to sell on the property at a profit (or “turn” it) to Persimmon, but by August 2000 Persimmon were considering a purchase direct from Chartbrook.

The important point in this period is that Persimmon eventually offered, as part of a proposed sale and purchase agreement, what was called a “sales overage” which was to be a supplemental payment to Chartbrook of 30% of net sales revenue over a defined trigger value of £345 per square foot. The trigger value was a high value (based on the then projection for a penthouse flat), and consequently in commercial terms what Persimmon was offering Chartbrook was a 30% stake if there were a substantial increase in the anticipated sale proceeds of the flats.

Persimmon’s first offer to Chartbrook, made in writing on September 29, 2000, was for the purchase of Units 5, 7 and 9. The enclosed Heads of Terms proposed a planning overage/underage clause, whereby the purchase price was to be adjusted, upwards or downwards, at the rate of £80.00 per square foot if planning permission for residential development exceeded or fell short of the anticipated 58,500 square feet. A minimum floor area of 45,000 square feet was stipulated. The minimum price payable under this offer for the residential space would have been £3.7 million.

On November 7, 2000 Persimmon made a revised offer of £4.3 million. This offer also contained a proposal for planning overage of £80 per square foot, but it was to be triggered by the grant of planning permission for more than 53,750 square feet of net internal residential floor area, subject to a maximum of £1 million.

The offer was accepted in principle and solicitors were instructed, Mr Skelly for Chartbrook and Mr Assael for Persimmon.

In January 2001 Persimmon told Chartbrook that an increase in Persimmon’s estimated build costs of approximately £750,000 necessitated a revised offer.

Following a meeting on January 16, 2001, on the next day Persimmon made, in a letter from Mr Pendlebury to Mr Reeve, a revised purchase offer for Units 5, 7 and 9. The basic price was reduced from £4.3 million to £3.8 million, but there was an improved planning overage, and the introduction at Chartbrook’s request of a sales overage.

The planning overage was to be improved by the removal of the £1 million cap and the increase of the overage payment rate from £80.00 to £82.00 per square foot.

The sales overage was proposed in the following terms:

“In addition, you asked us to consider some form of sales related overage, that would enable you to share in any sales uplift that is experienced during the course of the development of the project. We have given careful considerable (sic) to what we anticipate both build costs and sales inflation to be over the course of the project, and can confirm that we are prepared to offer you 30% of all net sales revenue achieved above a defined trigger level.

In terms of the calculation of this trigger, I suggest it is done by way of a multiplication of the total net internal area of the private sale residential element, (I suggest the same definition is used as that to define the planning related overage) and £345.00 per square foot. As with the calculation for the planning overage, we will be able to define exactly what the trigger is at the point when we obtain our Detailed Planning Permission. …I would, however, state that the above mentioned trigger is exclusive of any revenue we achieve for the disposal of the car parking places. At the present time we have costed these at a value of £10,000 per space and do not envisage experiencing any growth on this value”

The judge found that Mr Pendlebury pitched the trigger level for the sales overage a little beyond the highest average level which he anticipated obtaining on the sale of the residential units. The sales value in the most optimistic appraisal ranged from £291 per square foot to £345 per square foot, but that higher figure related only to a penthouse flat. Mr Vantreen’s and Mr Reeve’s evidence suggested that their perceptions of the sale value per square foot of residential units for which planning permission was likely to be granted on the site as at January 2001 lay within a broadly similar, but slightly higher, range. Neither Persimmon nor Chartbrook built into their expectations any assumption that the residential property market would either rise or fall during the marketing and construction of the development.

On January 18, 2001 Mr Skelly wrote to Mr Assael:

“Our clients apparently agreed last minute amendments to the transaction and I have set out below my understanding of these:-

1.

The initial purchase price to be reduced - £500,000.00.

2.

The primary [overage] is to be without any capping. Therefore the provision of £1,000,000.00 will be removed.

3.

There will be a secondary [overage] which will not take into account any sums paid under the primary [overage]. The secondary [overage] from my understanding is to be 1/3 (one third) of the sale price of each residential unit in excess of £345.00 per square foot for which the formula would appear to be:-

Residential unit and sale price

_______________________ - 345:-1/3(one third)

Square footage of unit

My clients do want confirmation in writing that the above does reflect both parties agreement and intention.”

Mr Assael sent Mr Skelly a draft Put and Call Option Agreement. Clause 20 under the heading “Net Sales Overage” provided that in addition to the purchase price (which was to be the base value of £3.8 million plus the enhanced value resulting from any increase in the extent of the planning permission over 53,750 square feet) “the Seller may be entitled to a further payment in addition to the Purchase Price calculated in accordance with Schedule 6.”

The additional payment was to be calculated following the sale of the last private residential unit forming part of the Development, by reference to a formula found in a series of definitions in paragraph 1:

1.1

“Aggregate Total Net Proceeds” means the aggregate sum of money (after the deduction of Costs and Incentives) received by the Buyer from purchasers of the Net Internal Area which exceeds £345.00 per square foot of the Net Internal Area and excluding any sum of money received by the Buyer for any car parking spaces constructed on the Development.

1.2

“Costs and Incentives” means the aggregate of all costs and incentives provided by the Buyer for the purchasers of the private residential units forming part of the Development and which costs and incentives will be reasonable and usual for similar developments in the locality of the Development.

1.3

…..

1.4

“Gross Sale Proceeds” means the gross sum payable for each private residential unit forming part of the Development including the price achieved for any car parking space sold with such flat and the Costs and Incentives paid or allowed in respect of such flat.

1.5

…..

1.6

“Net Sales Overage” means 30% of the Aggregate Total Net Proceeds.

1.7

“Net Sale Proceeds” means the sum of money received by the Buyer for each private residential unit forming part of the Development excluding any monies received for any car parking spaces sold with such flat and excluding the Cost and Incentives paid or allowed in respect of such flat.

On January 19, 2001 Mr Skelly had a meeting with his clients. The judge found that Mr Vantreen and Mr Reeve were broadly content with Schedule 6, and understood that it provided for a sales overage payable at the rate of 30% of net residential sales revenue less C&I, in excess of £345 per square foot. Having regard to their then perception that the market value of residential units for which planning permission was likely to be granted lay in a range which did not significantly exceed that level, the judge was satisfied that they appreciated that the sales overage then on offer was one which would depend upon a significant improvement in the residential market, if it was to give rise to an additional payment to Chartbrook over and above the aggregate of the basic sum of £3.8 million and the planning overage.

In a letter to Mr Assael of January 19, 2001, Mr Skelly said:

“I assume the trigger figure is that produced by multiplying the total net internal area by £345. Once that figure has been reached, my client is entitled to 30% of the aggregate net proceeds thereafter.”

Mr Skelly’s attendance note records that Mr Vantreen and Mr Reeve stated to Mr Skelly their assumption that the enhancement payable to the vendor of Unit 5 under the formula referred to above (para 16) could be as much as £200,000 “though this was unlikely”. Mr Vantreen and Mr Reeve confirmed in the witness box that the Enhanced Value of £200,000 reflected an assumed land value for Units 5, 7 and 9 together after the grant of planning permission in the region of £5 million. The judge accepted therefore, that in informing Mr Skelly of their expectation that the Enhanced Value payable to the vendor of Unit 5 could be as high as £200,000, Mr Vantreen and Mr Reeve both had it in mind that the value of Units 5, 7 and 9 with the benefit of planning permission could be as high as £5 million, but was unlikely to be higher, in the absence of any rise in the market.

Consequently the position when this phase of the negotiations ended was that (a) the base price was £3.8 million; (b) if sales reached a trigger level of £345 per square foot Chartbrook was to have a 30% share in the enhancement; (c) Chartbrook’s view of the maximum value, in the absence of a rise in the market, was £5 million.

Persimmon acquired Beazer Homes Limited at the beginning of 2001, and as a result did not have the funds for an outright purchase of the Hardwicks Way site.

On January 30, 2001 Mr Pendlebury sent Mr Reeve a fax enclosing Persimmon’s latest development appraisal for the site. The appraisal assumed planning permission for 81 residential units, a gross development value of £17,943,000, costs of £9,352,411, a gross margin of £5,096,350 based on 28.4% of the gross development value and a net land value of £3,205,724. In the margin Mr Pendlebury had written in an alternative calculation based upon a 24% gross margin, producing a net land value of £3,930,522. In his fax, Mr Pendlebury explained the calculations in the margins on the basis that if he was buying on the open market, he would look for a gross margin of 23-24%, increasing the residential land value to £3,930,522, and generating a gross land value including the commercial element of £4,745,632.

He said that he hoped shortly to be able to make both a building licence and a joint venture proposal, and on February 1, 2001 Mr Pendlebury set out both those alternative proposals in a letter to Mr Reeve.

The building licence proposal formed the basis of all the subsequent negotiations leading to the making of the Agreement. The letter began with a restatement, adjusted slightly downwards, of what Mr Pendlebury regarded as the open market value of the site (that is Units 5, 7 and 9) if he were acquiring it with the benefit of planning permission for 80 residential units and 10,000 square feet of commercial space. The 80 residential units assumption was described in the development appraisal sent to Chartbrook as comprising 56,442 square feet of floor area. Mr Pendlebury’s revised land value figure was £4,663,000 for residential and commercial combined, but he noted that if planning for 68,000 square feet could be obtained the land value would be significantly higher.

The letter said:

It is plain that what Persimmon was offering was 29.8% of the net proceeds of the residential area.

There was a meeting between the parties on February 6, 2001, which was followed up by a letter from Mr Pendlebury to Mr Reeve setting out “as promised” more detailed terms of a proposed building licence. The main terms of the proposed building license were set out in a series of bullet points, the last of which reads as follows:

“Upon receipt of the purchase monies, the revenue will be apportioned to Chartbrook, on the basis of 29.8% of the net revenue achieved from the disposal of the private sale residential units and 45% of the net revenue from the disposal of the commercial units. In addition, we are prepared to provide you with guaranteed backstop dates and minimum payments that will be made regardless of the actual performance of the project both in terms of timescales and cost.”

There was appended to the letter a schedule of guaranteed payments in 8 stages amounting in aggregate to £5,760,000, each payable either on the sale of 10 plots or upon specific dates starting 25 months and finishing 46 months after the implementation of the Planning Consent.

The letter continued:

“Based on the current scheme for 80 units, and 9,020 sq ft of commercial floor space, the minimum land value we are prepared to pay to Chartbrook on the disposal of each residential unit is £67,000, together with a further minimum payment of £400,000 on the disposal of the commercial unit. If as a result of improvement in the market, Chartbrook are entitled to more than the minimum payments I suggest that an equalisation calculation takes place following the disposal of the last unit.

As mentioned above, the figures contained herein are based upon our uplift scheme and we would obviously need to adjust the land value and guarantees depending upon what the actual outcome of planning is. Within the contract, I therefore suggest that a formula is included whereby the land value is calculated using the following inputs:

Private Sale Residential Accommodation (NIA) - £94.96/ sq ft

Affordable Housing Accommodation (NIA) - £0/sq ft

Commercial Floor Space (NIA) - £44.34/per square foot

Once the total land value has been calculated, a simple formula can then be applied to divide the land values by the number of units, in order for us to calculate the guaranteed payments that you will receive on the sale of each plot. I suggest that the guaranteed backstop dates for the receipt of these payments, together with the percentage of open market value that you are entitled to, remains the same regardless of the outcome of the Planning Application. ”

Mr Pendlebury wrote a letter to Mr Reeve on February 12, 2001, the purpose of which was, without altering the overall financial effect of the proposed building licence, to split the component parts of the residential elements of the pricing structure between the units and their associated car parking spaces.

The letter contained the first of two tables seeking to analyse in numerical terms the various components of value to be afforded to Chartbrook under the building licence.

The letter went on:

“You will see that, while I have broken the elements down further, the total land value payable to yourselves, still remains unchanged at £5,760,000. We are obviously also prepared to continue with the minimum guaranteed payment dates and I enclose a revised schedule with this letter. As a result of the re-apportionment of the revenues, the multiplier to calculate the land value for the Private Sale Residential element is changed to £92.92 per sq ft NIA. All the other multipliers remain unchanged. ”

The judge found that it was evident from the study of the table, the text of the letter and the accompanying schedule of guaranteed payments that the “total land value” meant the total guaranteed minimum land value, rather than any higher value that might be derived from the application of the percentages in the left column of the table to the sales revenue ultimately achieved.

Shortly after Mr Pendlebury’s letter of 12 February, 2001 Chartbrook’s agreement in principle to the building licence proposal was communicated to Persimmon.

Mr Assael prepared the first draft of what became the Agreement, which he e-mailed to Mr Skelly in two parts, on February 28, and March 1, 2001. The second e-mail contained Schedule 10, which was the predecessor of Schedule 6 to the Agreement.

Subject to one exception, all the definitions in what became Schedule 6 which are relevant to the meaning of ARP and the Balancing Payment appear verbatim in the first draft of Schedule 10. The only differences were: first, all the numerical amounts, including the percentages, changed; secondly, the Balancing Payment originally meant the aggregate of ARP, the Additional Commercial Payment and the Additional Car Parking Payment. The second and third of those three items were at successive stages in the drafting removed from the definition of Balancing Payment, and in the Agreement as executed, the Balancing Payment is defined as meaning ARP, which the judge observed was on its face an apparently redundant piece of duplicative definition, explicable only by reference to the process of drafting.

The judge found that it followed that if the definition of ARP failed to reflect a common intention as to its meaning about which the parties had reached agreement in principle in mid-February 2001, it did so as the result of a drafting mistake by Mr Assael, or a failure by Mr Pendlebury to communicate that common intention to him in February, which went undiscovered by Persimmon during the eight months which elapsed before the Agreement was executed.

Mr Assael prepared a second draft of the Agreement and circulated it both to his clients and to Mr Skelly on March 13, 2001. The judge found that manuscript notes on Persimmon’s copy of the second draft showed that what had by then become Schedule 7 (relating to the Price) had been carefully checked, probably by Mr Pendlebury. Mr Skelly took Mr Vantreen and Mr Reeve through the second draft in detail at the meeting on March 20, 2001. His attendance note contains the following paragraphs:

“We then dealt with the Agreement by going through it line by line, flagging up those matters which would need discussion tomorrow with Persimmon.

The last point left was the Seventh Schedule relating to calculation in price, as I have stated they have negotiated this with Persimmon and had Persimmon’s letters relating its calculation and they agreed they would themselves go through that Clause to confirm that the construction of the price and payment was agreed.”

The judge accepted Mr Skelly’s evidence that he had taken Mr Vantreen and Mr Reeve line by line through the Schedule, explaining its meaning in the terms for which Chartbrook contended, and that during the meeting Mr Vantreen and Mr Reeve confirmed by way of instructions to him that the Price was agreed to be as he had explained by reference to the Schedule.

On March 28, 2001 Mr Pendlebury sent a copy of Schedule 7 to Mr O’Neill, and said:

“… please find attached Schedule 7 from our proposed Purchase Agreement that sets out exactly how the land value is to be calculated.

In summary each of the individual elements of the scheme, i.e. sale residential, private sale residential car parking spaces, commercial and affordable housing, is each calculated by taking the total NIA of that particular element that obtains Planning Permission and is multiplied by an agreed land value per square foot or per space. This formula will calculate the minimum guaranteed land value, although we have agreed to make a balancing payment, which is calculated upon a percentage of the net sales revenue.”

After Chartbrook was able to acquire Units 1 and 3, Persimmon needed to carry out a due diligence and to reconsider the planning potential of the larger site, and to recalculate its residual appraisal.

An offer was made by letter from Mr Pendlebury to Mr Reeve dated May 24, 2001, following a further meeting.

The terms offered included a structure identical to that offered by Persimmon’s letter of February 12, 2001, but with substantially revised financial amounts and percentages, necessitated by the addition of Units 1 and 3 to the development site, and (according to Persimmon) by a further and more pessimistic review of anticipated build costs:

“Option 1

This Option is consistent with our previous agreement, whereby Chartbrook will not receive any form of premium on either exchange of the Building Licence Agreement or alternatively on the receipt of the Detailed Planning Permission, but rather take all of the Purchase Price as deferred payments dependant upon the performance of the project , albeit with guaranteed backstop dates and minimum sums.

Based upon this proposal, Persimmon Homes (South–East) Limited are prepared to offer a total land value of £7,191,947 (seven million one hundred and ninety one thousand, nine hundred and forty seven pounds) on the basis that our respective Companies enter into a Building Licence Agreement and that the necessary level of security is afforded to Persimmon Homes through a First Charge as well as a Power of Attorney to grant leases directly from Chartbrook to our purchasers.

The table below sets out the minimum guaranteed land values that you will receive for the respective elements of the scheme, together with the percentage of sales revenue that you will also be entitled to if the project performs better than is currently anticipated.

	Percentage of Sales Revenue	Minimum Value per Plot	Number of Plots	Total
Residential Apartments	23.4%	£53,333	105	£5,600,000
Residential Car Parking Spaces	30.24%	£3024	80	£241,947
Commercial	45.02%	£1,350,000	1	£1,350,000
TOTAL				£7,191,947

In addition to the above guaranteed payments, we are also prepared as with our previous proposal, to provide you with guaranteed backstop dates when these payments will be made regardless of the performance of the actual project. The attached schedule sets out our proposals in respect of this matter.

Given that the contract will be conditional upon planning, it is obviously not possible at this stage to finalise the exact land value for each of the individual element of the scheme and hence calculate the minimum guaranteed payments. I therefore suggest that the contract includes a ratchet mechanism incorporating a formula that multiplies that net internal floor area for each of the respective elements by an agreed land value in order to calculate the total land value for the whole scheme. The table below sets out the proposed land value for each of the elements.

	Land Value per Sq ft NIA
Private Sale Residential Accommodation	£76.34 per sq ft
Residential Car Parking Spaces	£3,024 per space
Commercial Accommodation	£38.80 per sq ft
Affordable Housing Accommodation	£0 per sq ft

You will see from the above table that if it is necessary to provide affordable housing on the scheme it is placed into the equation at nil value. This is consistent with our previous agreement.”

One of the copies of the letter (probably sent by Mr Reeve to Mr Vantreen) contained an important annotation in Mr Vantreen’s handwriting. At the top of its second page, which contained the two tables, he wrote in manuscript “£325.p.f.s” and, a little to the right, “£228,000.” I shall revert to this document, but I should mention now that the judge said that it was a very relevant calculation if Persimmon’s construction of ARP represented the parties’ common intention, but of no obvious relevance on the construction advanced by Chartbrook.

Shortly after May 24, 2001, either Mr Reeve or Mr Vantreen communicated Chartbrook’s oral acceptance in principle of the revised offer, notwithstanding the substantial reduction in the percentage of Chartbrook’s share of the sale value of the Residential Units from 29.87% to 23.4%, which Persimmon had sought to attribute to an anticipated increase in build costs. In due course, the revised financial amounts and percentages proposed in the May 24 letter found their way by amendment into what became Schedule 6 to the Agreement.

On June 27, 2001 Mr Pendlebury wrote to Mr Reeve enclosing a revised offer dated June 25, 2001 and based on the February 12, 2001 letter but with the 23.4% figure. In his covering letter Mr Pendlebury appears to record that Mr Reeve had asked him for the letter for the purposes of Mr Reeve’s negotiations with the owner of Unit 5. Persimmon says that, if Chartbrook had then thought that the offer meant what it now says it meant, Chartbrook would not have wished the owner of Unit 5 to know about it.

On July 10, 2001 Chartbrook entered into a put and call option agreement with the owners of Units 1 and 3 at a purchase price of £800,000. Thereafter, Mr Vantreen and Mr Reeve turned their minds to the question how to finance the project during the lengthy period which would precede any payment to Chartbrook by Persimmon under the building licence. Chartbrook had already charged part of the site in connection with funding the purchases of Units 5 to 9, and needed to redeem that charge so as to be able to give a first charge of the site to Persimmon as security for its obligations under the building licence, as contemplated by the Agreement then in draft. Accordingly, Chartbrook sought to re-finance on the security of the Agreement itself, backed by Persimmon’s covenant, a company which had after its takeover of Beazer Homes become the largest house builder in the United Kingdom. Approaches were made to HSBC, to the Clydesdale Bank and to the Royal Bank of Scotland. They focused on the minimum guaranteed payment under the Agreement, variously described as £7.1 million, £7.3 million and “around” £7.4 million as security for finance in a range between £2.5 and £3.6 million. No mention was made in the written approaches to the three banks of any specific monetary figure expected to be obtained by way of Balancing Payment.

Meanwhile, on August 30, 2001 Mr Pendlebury reported by email to Mr O’Neill. The email said:

“…. the Vendors are seeking to ensure that they are able to obtain the maximum return from the Building Licence, and specifically from the potential sales overage that is incorporated within the Agreement. As with all licences, the payment Chartbrook receive is either a minimum guaranteed sum or 23.4% of the net sales revenue, whichever is the gr[e]ater”

The Agreement was signed on October 16, 2001.

Negotiations with the Royal Bank of Scotland continued. Eversheds were asked by the bank to review the Agreement and in their report to the bank dated December 7, 2001 they said:

“In addition to the above Chartbrook are to receive the Total Residential Car Parking Land Value which will be £3,024 multiplied by each of the residential car parking spaces for which planning has been obtained plus a balancing payment of 23.4% of the price achieved for each residential unit in excess of the minimum guaranteed residential unit value.”

A copy of that report was sent to Mr Skelly.

At some time shortly after December 10, 2001, Mr Reeve wrote to Mr Carl Wright, of CPFC, mortgage brokers:

“It is apparent that the guarantee from Persimmon after deduction of the commercial payment will be more than sufficient to repay the borrowings and of course this takes no account of the balancing payment on completion of the development. This amounts to 23.45% of the price achieved for each residential unit in excess of the minimum guaranteed unit value. This figure cannot be quantified until we have planning permission but if the submitted application is successful it would mean a guaranteed residential unit value of approximately £65,500-00 and if we assume average sales at £200,000-00 per unit the balancing payment will be in excess of £3 million.”

The letter from Eversheds to the Royal Bank of Scotland and Mr Reeve’s letter to Mr Wright are the only contemporary communications which reflect what Chartbrook maintained was its intention.

On June 27, 2002, Mr Vantreen produced a calculation of the amount that the deal was likely to realise overall, which showed a total overage of £4,170,715. On July 8, 2002, Mr Wright wrote to the Royal Bank of Scotland describing the overage as “clearly substantial”. A spreadsheet prepared by Mr Vantreen in February 2003 showed overage substantially in excess of £3 million.

This is an unusual and difficult case, and I understand well why the judge came to his conclusion on rectification “after considerable difficulty” (paragraph 163). Mr Robert Miles QC, for Chartbrook, rightly reminded the court that there was a full trial before the judge, in which all of the individuals concerned with the conclusion of the Agreement gave evidence, and that this court should not in effect re-try the case on the documents.

Yet this is a case in which, if one puts aside the drafts of the Agreement, every contemporary document prior to the conclusion of the Agreement, and every piece of paper which throws light on the commercial purpose of the provision, supports Persimmon’s case that the deal which was on the table was Persimmon’s offer to Chartbrook of either a fixed percentage of the sales revenue or the minimum guaranteed amount, whichever was the greater.

The case is on the borderland between construction of a written agreement, the admissibility of prior negotiations as an aid to construction, and the equitable remedy of rectification, based on common or unilateral mistake. I propose to deal with the issues in that order.

On the basis that the prior negotiation materials are not admissible

The judge found that the ordinary meaning of the definition of ARP pointed clearly towards Chartbrook’s rather than Persimmon’s construction.

ARP meant 23.4% of the excess of the price achieved for each Residential Unit over the MGRUV, less C&I. Persimmon’s construction required the words “the amount by which” to be added after the word “means”, and the word “is” to be added before the phrase “in excess of”.

He was particularly impressed by the point that the positioning of C&I in the formula did not work on Persimmon’s construction. C&I had a linear relationship with the amount of the price achieved for each Residential Unit: thus Persimmon might agree the sale of a flat for £250,000 after incurring C&I (such as payment of the purchaser’s legal fees or stamp duty, or the installation of special features such as wooden floors) of say £50,000 or, with the same commercial consequence, sell the same flat for £200,000 but incur no C&I. Persimmon’s construction would deduct the C&I from the 23.4% of the price achieved for the Residential Unit before the net amount was compared with the MGRUV, to ascertain whether there was any excess. That calculation would magnify the negative effect of C&I by a factor of more than 3 in comparison with the positive effect of the increase in the Residential Unit Price attributable to the C&I. Although Persimmon sought to overcome this problem by amendment in its rectification claim to remove the phrase “less the Costs and Incentives” from the end of the definition and re-inserting it after the phrase “each Residential Unit,” it made no similar amendment to its case on construction. Its inability to offer an interpretative route which did not produce a commercial absurdity in the deduction of the C&I was a telling indication that its construction was wrong.

The judge referred to pointers claimed by Persimmon to support its construction. First, Mr Nugee QC relied on the fact that paragraph 3.3 expressly contemplated that there might be no Balancing Payment at all (“or the date of payment if any of the Balancing Payment”). The judge regarded that as a significant pointer in favour of Persimmon’s construction: since that outcome would, on Chartbrook’s construction require a market collapse in excess of 66% below the level which the parties perceived it to be at the time of the Agreement, before the Residential Unit price fell below the MGRUV (even ignoring C&I), that was such an unlikely outcome that the words “if any” would make no sense on Chartbrook’s construction. By contrast it was perfectly possible that 23.4% of the Residential Unit price would be less than the MGRUV, so that the phrase “if any” was in harmony with Persimmon’s construction.

Second, Mr Nugee argued that the combination of the phrases “Balancing Payment”, “Additional Residential Payment” and “Minimum” in the phrase MGRUV in contrast to the word “Total” in “Total Residential Land Value” and “Total Land Value” all lay more comfortably with the concept of ARP being a possible bonus payable if the development went better than expected, than with its forming a significant part of the Price in all but the most extremely adverse of market circumstances. The judge concluded that this submission carried little weight. It might have been of some force if any of those words or phrases had been used for their independent ordinary meaning, rather than as, or as part of, a label with its own definition. The whole purpose of the attribution of a detailed definition to a word or phrase was to replace what readers might otherwise have thought to be the natural meaning of that word or phrase with its contractually defined meaning. The word or phrase is stripped of its natural meaning, and the same result could have been achieved if an algebraic letter or expression were used as the label with the same detailed definition.

The overall conclusion was that the definition of ARP was that contended for by Chartbrook. The contest was between all the natural consequences of the way in which the constituent parts of the formula were constructed in the definition on the one hand, and on the other hand the use of the expression “if any” when referring to the Balancing Payment in a quite different context, in paragraph 3.3 of Schedule 6. To that contest there could be only one outcome. ARP meant 23.4% of an amount calculated by deducting both the MGRUV and the C&I from the price achieved for each Residential Unit.

Persimmon concentrated its attack on the judge’s refusal to take account of the correspondence. But in its notice of appeal and Mr Nugee’s skeleton argument Persimmon also took issue with the judge’s textual analysis and Chartbrook responded to those criticisms.

Persimmon’s case is that Chartbrook’s construction gives Chartbrook a windfall of some £4 million more for its land than Persimmon ever offered, and bears no relation to what Persimmon offered and Chartbrook agreed; it was not what, objectively considered, the parties were trying to do; and it makes no commercial sense.

Persimmon repeats the point on the phrase in paragraph 3.3 “payment if any of the Balancing Payment” which the judge found to be a pointer in its favour. It also argues that it is striking that the MGRUV is described as (i) a minimum and (ii) a guaranteed amount. Together these indicate what purpose the MGRUV was intended to serve. Persimmon was agreeing to a guaranteed payment that Chartbrook would be entitled to receive regardless of the success or failure of the development; this was the minimum that Chartbrook would get whatever the flats sold for. The language strongly suggested that there might be circumstances in which that was all that it might get, and militated strongly against a construction in which a very large additional payment would inevitably arise. If this were the case it would be very odd to describe the MGRUV as a “minimum guaranteed” value; it would be merely an “initial” payment on account, not a minimum and not a guarantee. The phrases “Balancing Payment” and “Additional Residential Payment” also tended to point in the same direction. Both had the flavour of a top-up or extra payment rather than the payment of what on Chartbrook’s construction would inevitably be a very substantial part of the overall Price.

The judge was wrong to find that that these points carried little weight because the whole purpose of the attribution of a detailed definition to a word or phrase was to replace what might otherwise be its ordinary meaning. On Persimmon’s construction there was no need to give an artificial or strained meaning to the words “Minimum Guaranteed”; whereas on Chartbrook’s construction they were inapposite words.

On Persimmon’s construction the commercial purpose of schedule 6 was to deliver guaranteed minimum payments to Chartbrook (and with a guaranteed timetable for payment) together with the possibility of an extra payment if sales exceeded expectations. By contrast it was impossible to discern what the commercial sense behind Chartbrook’s construction was. On Chartbrook’s case the MGRUV appeared to be entirely arbitrary – it made no sense to specify it as the figure of 23.4% of a realistic price and then provide for Chartbrook to receive 23.4% of the excess not over that price but over the MGRUV.

Chartbrook’s reading was the natural and obvious one. Persimmon’s approach required the reading of extra words into the clause, so that it would read “ ‘Balancing Payment’ means the amount if any by which 23.4% of the price achieved for each Residential Unit is in excess of the Minimum Guaranteed Residential Unit Value less the Costs and Incentives.” This re-writing of the clause would change its meaning, so that instead of applying 23.4% to the excess of the net price achieved over the MGRUV, Persimmon says that a Balancing Payment arises only to the extent that 23.4% of the price achieved was in excess of the MGRUV. On Persimmon’s version there is no grammatically viable way of applying the profit share both to the price achieved and the C&I.

The words “if any” in paragraph 3.3 of Schedule 6 could not materially affect the meaning to be derived from the definitions themselves. The judge was over-generous to Persimmon in saying that these words were a significant pointer. The words used in the label did not help as they were consistent with each side’s case. The words “minimum” and “guaranteed” were consistent with Chartbrook’s case – there was to be a minimum fixed amount plus a further variable amount, and so it was a minimum and it was guaranteed. The word “balancing” was consistent with Chartbrook’s reading.

The clause as interpreted by Chartbrook made sense: it simply involved a higher price than Persimmon contends for. If the purpose of the provisions was to provide Chartbrook with a share of the price achieved (which is all it purported to do) then there was nothing uncommercial at setting the trigger at one level rather than another.

The starting point is not controversial. Interpretation is the ascertainment of the meaning which the document would convey to a reasonable person having all the background knowledge which would reasonably have been available to the parties in the situation in which they were at the time of the contract. Subject (a) to the requirement that the background should have been reasonably available to the parties, and (b) the exclusion of previous negotiations of the parties and their declarations of subjective intent, it includes anything which a reasonable man would regard as relevant and which would have affected the way in which the language of the document would have been understood by a reasonable man: Lord Hoffmann in Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896 at 912; BCCI v Ali [2002] 1 AC 251, 269; and also Mannai Investment Co Ltd v Eagle Star Life Assurance Ltd [1997] AC 749, 779.

If a semantic analysis of words in a commercial contract leads to a conclusion which flouts business common sense, it must be made to yield to business common sense: Antaios Cia Naviera v Salen Rederierna AB [1985] AC 191, 201, per Lord Diplock. This does not mean, however, that the language can be rewritten in order to make the language conform to business common sense: Co-operative Wholesale Society Ltd v National Westminster Bank plc [1995] 1 EGLR 97.

I repeat here for convenience the principal provisions at issue. Schedule 6 provides (paragraph 1) that the Price is to be made up of two parts, “the aggregate of the Total Land Value and the Balancing Payment.”

The Balancing Payment is defined in paragraph 1 as “the Additional Residential Payment,” where:

“Additional Residential Payment” means 23.4% of the price achieved for each Residential Unit in excess of the Minimum Guaranteed Residential Unit Value less the Costs and Incentives.

“Minimum Guaranteed Residential Unit Value” means for each Residential Unit the Total Residential Land Value divided by the number of Residential Units for which Planning Permission is granted.

“Total Residential Land Value” shall be £76.34 per square foot multiplied by the Residential Net Internal Area (less the Section 106 Money and less the Rights of Light Money and less the Sub-Structure Assumptions Additional Cost)

Paragraph 3.3 provides for payment of ground rent value to be payable on the earlier of the date of realisation of the ground rent value “or the date of payment if any of the Balancing Payment.”

It seems to me that three factors lead to the conclusion that the judge’s decision was wrong. The first factor is that I do not consider that the judge was right to consider that it makes a decisive difference that the words appear in a definition clause. The process of interpretation is the same. The Agreement is not a statute. The fact that the terms appear in a definition section is simply one of several factors in the exercise of construction.

The second factor is that it would then follow that the judge was wrong not to give weight (which he said that he would have done had they not been contained in the definition clause) to the point that the MGRUV is described as (i) a minimum and (ii) a guaranteed amount. I accept the submission for Persimmon that these words indicate what purpose the MGRUV was intended to serve. Persimmon was agreeing to a “guaranteed” payment that Chartbrook would be entitled to receive regardless of the success or failure of the development; this was the “minimum” that Chartbrook would get whatever the flats sold for.

The third factor is that it is very difficult (and probably impossible) to discern the commercial sense behind Chartbrook’s construction. In my judgment it is permissible to take into account as background the anticipated selling prices of the Units. A guaranteed minimum indicates that there might be circumstances in which this was all that it might get (“if any”), and militates strongly against a construction in which a very large additional payment would inevitably arise. Mr Reeve’s and Mr Vantreen’s expectations of residential unit sales revenue were in the region of £200,000 per unit. The MGRUV was anticipated to be about a quarter of this, and it was in fact about £47,000 per flat. On Chartbrook’s construction, there would be bound to be a very substantial ARP (Chartbrook said it expected about another £3 million). Chartbrook’s case would still produce substantial additional payments even in the event of a catastrophic market fall. Only if the market fell by 66% (thereby reducing unit values to something in the region of £65,000) would there be no additional payment.

I accept Persimmon’s submission that Chartbrook’s interpretation made no sense of the level of MGRUV. On Persimmon’s case the MGRUV was the equivalent of 23.4% of a realistic anticipated aggregate price, so that was what Chartbrook would receive if the flats sold at or below that price, and it would receive 23.4% of the excess if they sold for a higher price. But on Chartbrook’s case the MGRUV was arbitrary, since it made no sense to specify it as the figure of 23.4% of a realistic price and then provide for Chartbrook to receive 23.4% of the excess not over that price but over the MGRUV.

In my judgment these factors lead me powerfully to the conclusion that Persimmon’s construction is the right one. That would leave as the only countervailing factor the role of C&I in the formula. It is common ground that, on Persimmon’s construction, C&I would be deducted from the 23.4% of the price achieved for the Residential Unit before the net amount was compared with the MGRUV, to ascertain whether there was any excess. That calculation would make no commercial sense. In my judgment the definition should be read so that the phrase “less the Costs and Incentives” referred to the amount payable rather than the MGRUV. In my judgment this is a case where a syntactical analysis must yield to business common sense (Antaios at 201).

Admissibility of negotiations

As I have said (para 69, above) it is a striking feature of this case that, if one leaves aside for this purpose the Agreement and its drafts, the documents from February 2001 up to the conclusion of the Agreement support the conclusion that Persimmon’s construction is correct and was the commercial intention of the parties. The February 2001 and May 2001 letters (supported by Mr Vantreen’s calculations on the copy of the May 2001 letter) would leave no room for doubt that Persimmon’s construction is what was intended.

If I am right on the construction of the Agreement without the aid of the negotiating documents, then it is unnecessary to decide on the limits of the exclusionary rule and its possible exceptions.

The judge did not express a view on what effect the admissibility of the pre-contract negotiations would have had. The judge referred to the well known authorities (to which I shall revert) and concluded that the adverse effect on third party rights was a compelling reason for not admitting evidence of prior negotiations. If the parties’ negotiations were to be routinely admissible as an aid to contractual construction, then third parties reading, dealing with or having transferred to them rights or obligations under the contract could not make any safe assumptions about its meaning without themselves carrying out an inquiry as to those negotiations, so as to put themselves in the same state of knowledge as the parties to the contract.

Even if there were an exception to the basic rule in “private dictionary” cases (The Karen Oltmann [1976] 2 Lloyd’s Rep 708) it did not extend to any case in which the word, phrase, clause or term is itself the subject of an express definition in the contract itself. The private dictionary principle should not be used for the construction of words, phrases or terms which are already defined in the Agreement. The parties themselves may, by using an express definition, reasonably be taken to have agreed that the definition should prevail over any competing definition arrived at during the course of their negotiations, but then not adopted in the contract. The existence of a definition is a clear signal to any third parties seeking to construe the contract that the meaning of the word, phrase or term in question is to be found within, rather than without, the four corners of the contract itself. The very existence of detailed definition in a commercial contract is a pointer towards the understanding of the parties to it that others who, unlike them, need education as to its finer meaning, will be expected to use those definitions, rather than to investigate the parties’ negotiations.

Persimmon says that the judge adopted an unnecessarily restrictive view of the law, and that had he applied the legal principles correctly, he should have given weight to the fact that he had himself found (judgment, paras 110 and 125) Chartbrook had communicated to Persimmon its acceptance of the February and May 2001 offers.

Evidence of pre-contractual negotiations is admissible to establish a clear consensus between the parties on the point at issue: The Karen Oltmann [1976] 2 Lloyd’s Rep 708; Jones v Bright Capital [2006] EWHC 3151 (Ch). Once the terms of the February and May offers are admitted, the construction of ARP in Persimmon’s sense is obvious. There is no reason in logic or principle why the admissibility of (as opposed to weight given to) pre-contractual material should in any way depend (as the judge thought) upon whether the ambiguity in question is to be found within the confines of a defined term. Definitions must be construed like any other part of a contract. If they are ambiguous those ambiguities must be resolved in the same way as any other part of the contract.

Chartbrook’s argument that on the judge’s findings of fact the parties did not in fact reach agreement at the points in time suggested and the evidence does not assist the process of construction in this case overlooks the point that the interpretation of a contract is an objective exercise, and the admissible facts are those reasonably available to the parties.

Chartbrook supports the judge’s reasoning on the law and its application to the facts, but also emphasises on this appeal that it is impermissible to ignore the terms of the drafts of the Agreement, which were themselves part of the negotiations and that it is unfair to rely on the documents of one side only when the Chartbrook directors did not create their own paper trail. Until parties reach agreement on the written contract their positions remain apart or divergent.

The Karen Oltmann was limited to pre-contract agreements about the very words used in the final contract. Persimmon’s approach would bring in evidence of negotiations where entirely different words were used, but there is alleged to be some earlier “consensus” or agreement “in principle” in point of substance, rather than about the meaning of the words actually used in the final agreement. On Persimmon’s test the court would have to distinguish between a case where the parties remained divergent on a particular point and the case where they are said to have “agreed”. There will often be intense controversy about this (as there is in the present case). To decide the question the court will have to hear evidence about the negotiations. The reality is that in almost every case the full negotiating history would have to be traversed. The principle in Prenn v Simmonds would therefore be swallowed by the exception.

There was no suggestion in the February letters that they were an offer capable of acceptance. The February letters did not contain any of the words or phrases used in the price formula in the draft contract. Mr Pendlebury’s letter of May 24, 2001 made some changes to the numbers, but was written on the footing that the terms of the agreement would be as set out in the draft contracts. The parties did not reach “consensus” either in February or May 2001. They merely agreed that the solicitors should get on with the drafting so that the parties could then see whether a contract could be reached.

There is no doubt about the starting point, which is that “for reasons of practical policy” the law excludes from the admissible background the previous negotiations of the parties: Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896 at 913. In Alexiou v Campbell [2007] UKPC 11, para 15, Lord Bingham referred to “ … the principle, vouched by such compelling authorities as Prenn v Simmonds … and Investors Compensation Scheme Limited v West Bromwich Building Society … that evidence of prior negotiations not leading to any antecedent agreement is inadmissible to construe a contract.” See also Moschi v Lep Air Services [1973] AC 331, 354; Schuler v Wickman Machine Tools [1974] AC 235, 261.

The policy reasons have not been fully articulated. In Prenn v Simmonds [1971] 1 WLR 1381, 1384-5, Lord Wilberforce said:

“ … The reason for not admitting evidence of these exchanges is not a technical one or even mainly one of convenience (though the attempt to admit it did greatly prolong the case and add to its expense). It is simply that such evidence is unhelpful. By the nature of things, where negotiations are difficult, the parties’ positions, with each passing letter, are changing and until the final agreement, though converging, still divergent. It is only the final document which records a consensus. If the previous documents use different expressions, how does the construction of those expressions, itself a doubtful process, help on the construction of the contractual words? If the same expressions are used, nothing is gained by looking back; indeed, something maybe lost since the relevant surrounding circumstances may be different. And at this stage there is no consensus of the parties to appeal to. It may be said that previous documents may be looked at to explain the aims of the parties. In a limited sense it is true: the commercial, or business object, of the transaction, objectively ascertained, may be a surrounding fact … The words used may, and often do, represent a formula which means different things to each side, yet may be accepted because that is the only way to get ‘agreement’ and in the hope that disputes will not arise. The only course then can be to try to ascertain the ‘natural’ meaning. Far more, and indeed totally, dangerous is to admit evidence of one party’s objective - even if this is known to the other party. However strongly pursued this may be, the other party may only be willing to give it partial recognition, and in a world of give and take, men often have to be satisfied with less than they want. So, again, it would be a matter of speculation how far the common intention was that the particular objective should be realised…

In my opinion, then, evidence of negotiations …ought not to be received, and evidence should be restricted to evidence of the factual background known to the parties at or before the date of the contract, including evidence of the ‘genesis’ and objectively the ‘aim’ of the transaction.”

In Britoil plc v Hunt Overseas [1994] CLC 561, 573 Hobhouse LJ said (in the context of a claim for rectification): “The process of negotiation and progressing towards a complete and formalised agreement is one which may contain many ambiguities. The purpose of the final document is to remove those ambiguities and to define authoritatively and clearly what the parties’ respective rights and obligations are to be.”

The reasons for the exclusionary rule are by no means self-evident. The American Law Institute, Restatement Second: Contracts, section 214, states that “…. negotiations prior to …the adoption of a writing are admissible in evidence to establish … the meaning of the writing …” In the United States, contract law is generally a matter of State law, and the courts of the majority of States retain the “plain meaning rule,” namely that where the meaning of the language taken in context is clear, evidence of prior negotiations cannot be used in its interpretation. But there is a division of authority as to whether evidence of prior negotiations can be used to show whether contract language lacks the required degree of clarity: Pacific Gas and Electric Co v G W Thomas Drayage and Rigging Co, 442 P.2d 641 (Cal. 1968). See Farnsworth, Contracts, 4^th ed. 2004 para 7.12. See also Corbin On Contracts, 1998, paras 24.7 et seq; Williston on Contracts (4^th ed 2007, Lord) Vol 11, para 33.42.

Nor is the exclusionary principle accepted in international instruments dealing with private law contracts, such as the Unidroit Principles of International Commercial Contracts (Art 4(3)) and the UN Convention on Contracts for the International Sale of Goods (1980). That material was referred to by Thomas J in the New Zealand Court of Appeal in an important criticism of the exclusionary rule in a decision in which he expressed the view that his construction was supported by provisions in earlier drafts of the contract: Yoshimoto v Canterbury Golf International Ltd [2001] 1 NZLR 523, paras 59 et seq. Thomas J’s view was that the general exclusionary rule should stand, but that there should be flexibility in its application where cautious use of the pre-contract material would enable the court to arrive at a meaning of the contract which accorded with the ascertainable intention of the parties: at [69], [76]. The point was not addressed on appeal to the Privy Council: [2002] UKPC 40, [2004] 1 NZLR 1.

The Unidroit Principles and the UN Convention were also relied on in a lecture by Lord Nicholls arguing for a more flexible approach, reprinted as My Kingdom for a Horse: the Meaning of Words (2005) 121 LQR 577. He said (at 583)

“… there will be occasions where the pre-contract negotiations do shed light on the meaning the parties intended to convey by the words they used. There will be occasions, for instance, when the parties in their pre-contract exchanges made clear the meaning they intended by language they subsequently incorporated into their contract. When pre-contract negotiations assist in some such way, the notional reasonable person should be able to take that evidence into account in deciding how the contract is to be interpreted.

This would not be a departure from the objective approach. Rather, this would enable the notional reasonable person to be more fully informed of the background context. This would recognise that pre-contract negotiations are themselves part of the background of a contract and that, like other background material, they may be relevant when interpreting a contract. They differ from other background material in that, unlike other background material, they may afford direct evidence of the parties’ actual intentions. That is not a reason for banning their use. That would be perverse. That would mean that in deciding the meaning intended to be conveyed by the language chosen by the parties the notional reasonable person would always be barred from having regard to what may be the best evidence of all. He must always conjecture, he must never know. The preferable approach is to recognise that pre-contract negotiations are relevant and admissible if they would have influenced the notional reasonable person in his understanding of the meaning the parties intended to convey by the words they used.”

Lord Nicholls suggested ((2005) 121 LQR at 587-588) that the policy reasons put forward for the proposition that pre-contract negotiations were inadmissible were as follows: (1) increased uncertainty and unpredictability in dispute resolution; (2) adverse effect on third party rights; (3) the use of the evidence would be unhelpful (Lord Wilberforce’s reason); (4) subversion of the objective approach. Lord Nicholls accepted that these were important practical considerations, but that they were not conclusive. To that I would add a further one, which is that without such a rule sophisticated and knowledgeable negotiators would be tempted to lay a paper trail of self-serving documents.

The judge in this case was particularly impressed by the second of Lord Nicholls’ reasons, the effect on third parties. But, as Lord Nicholls recognised (at 588), the same objection would apply (although perhaps with less force) to the admissibility of other background material, which might be equally unavailable to third parties. The Law Commission, Law of Contract: The Parol Evidence Rule (Law Com No 154, 1986), when concluding that the parol evidence rule did not have the effect of excluding evidence which ought to be admitted if justice was to be done between the parties, did not consider that assignees would be prejudiced: para 2.43. It is also significant that the effect on assignees does not prevent (as I have indicated) the admissibility of pre-contract negotiations for the purposes of interpretation in the practice of the world’s greatest capitalist nation, the United States.

In Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896, at 913, Lord Hoffmann accepted that the boundaries of the exception were not clear. In BCCI v Ali [2001] UKHL 8, [2002] 1 AC 251, at para 31 Lord Nicholls added that whether the reasons of practical policy referred to by Lord Hoffmann still held good had been increasingly the subject of debate, and he said that he desired to keep the point open for consideration on a future occasion.

A potentially relevant exception which is open to this court derives from The Karen Oltmann [1976] 2 Lloyd’s Rep 708. In that case the owners let the vessel for “for a period of 2 years 14 days more or less.” The charter provided that the charterers would have the option to redeliver the vessel “after 12 months trading subject to giving 3 months’ notice.” After trading for 19 months, the charterers gave 3 months notice of their intention to redeliver. The owners said that the redelivery provision meant that the charterers had the option to redeliver on the expiry of 12 months from the inception of the charter. The charterers said that it meant that they could terminate at any time after 12 months, and that their notice was therefore a good one.

The striking point about the case is that telexes between the parties showed that when they used the word “after” they meant “on the expiry of” and not “at any time after the expiry of.”

Kerr J accepted, first, that pre-contract telex exchanges were inadmissible to interpret the contract. Second, he accepted that the telex exchanges could be looked at to determine whether there had been an estoppel by representation, but held that they did not support the allegation of an estoppel.

His final conclusion (at 713) was that as a matter of impression the contract on its face meant “on the expiry of”, but said that if he should be wrong about that the construction which he preferred, he found that it was confirmed by the way in which the parties were themselves understanding and using the word.

His reasoning on the latter point has given rise to much discussion and some misunderstanding. Having accepted the general exclusionary rule, he said:

“...the principle can be stated as follows. If a contract contains words which, in their context, are fairly capable of bearing more than one meaning, and if it is alleged that the parties have in effect negotiated on an agreed basis that the words bore only one of the two possible meanings, then it is permissible for the Court to examine the extrinsic evidence relied upon to see whether the parties have in fact used the words in question in one sense only, so that they have in effect given their own dictionary meaning to the words as the result of their common intention. Such cases would not support a claim for rectification of the contract, because the choice of words in the contract would not result from any mistake. The words used in the contract would ex hypothesi reflect the meaning which both parties intended.”

He went on (at 713):

“However, on the basis that the word ‘after’ … is capable of bearing two meanings as a matter of construction, I do not think that there is any authority precluding the Court from examining the pre-charter-party exchanges in order to see whether the owners can make good their contention that the parties were in agreement in using this word in only one of its two senses, and having in effect both given it the same dictionary meaning to the exclusion of the other meaning. Having then considered the pre-charter-party exchanges on this basis I find that this contention is established. In these circumstances it seems to me that the charterers cannot now depart from this common meaning by asserting that this word has the opposite meaning in the charter-party.”

Kerr J referred to the negotiation on “an agreed basis that the words bore only one of two possible meanings” and used the expression “their own dictionary meaning.” But it is plain from the facts that there had been no actual agreement on the meaning of the word “after”. What happened was that the parties were negotiating on a common understanding of what the point in contention between them was, namely the period within which the notice had to be given so as to expire at the end of a period which the owners wanted to be of maximum duration and minimum flexibility where the charters wanted the precise opposite. There was no agreement on the word “after” except in this sense: because they were negotiating on a common understanding of what each was endeavouring to achieve they were using the word in the same sense.

On analysis, in my judgment, The Karen Oltmann does not lay down any special “private dictionary” exception. All that Kerr J was saying was that the negotiations could be looked at to see whether the parties had negotiated on an agreed basis that the words bore only one of two possible meanings. I doubt whether this differs in any material respect from admitting evidence of prior negotiations in construing a contract.

Kerr J suggested (at 712), obiter (because there was no claim for rectification) that private dictionary cases

“would not support a claim for rectification of the contract, because the choice of words in the contract would not result from any mistake. The words used in the contract would ex hypothesi reflect the meaning which both parties intended.”

Briggs J was right to point out that cases such as Re Butlin’s Settlement Trust [1976] Ch 251 show that rectification may be available even if words have been deliberately used, but where it was mistakenly considered that they bore a different meaning from their correct meaning as a matter of true construction. But I do not think he was right to suggest that The Karen Oltmann should have been decided on that basis. In my judgment, the problem in The Karen Oltmann was that the words in the contract were capable of bearing more than one possible meaning. There was no mistake either as to the words used or as to the consequence of the words used.

The decision of Kerr J, a very experienced and eminent commercial judge, has been referred to with apparent approval by this court, although its reasoning does not appear to have been the ratio of any decision: see, e.g. Ham v Somak Travel Ltd, February 4, 1998, unreported (CA); IRC v Botnar [1999] EWCA Civ 1652, 72 TC 205; Proforce Recruit Ltd v The Rugby Group Ltd [2006] EWCA Civ 69.

In Proforce Recruit Ltd v The Rugby Group Ltd [2006] EWCA Civ 69 the relevant phrase was “preferred supplier status” in a service cleaning agreement. The question for the Court of Appeal was whether an otherwise unarguable case on construction could be saved (from being struck out or from summary judgment being granted) by reference to the meaning that it was alleged that the parties had in their negotiations placed on that phrase. In concluding that it might, and therefore in allowing the appeal against the striking out of the action, Mummery LJ, after quoting from The Karen Oltmann, said (at [31]):

“… as stated in Chitty on Contracts para 12.119, evidence of facts about which the parties were negotiating is admissible to explain what meaning was intended and evidence of what the parties said in negotiations is admissible to show that the parties negotiated on an agreed basis that the words used bore a particular meaning.”

Concurring in the result, Arden LJ said (at [55]):

“In this case, the parties have used a very unusual combination of words (‘preferred supplier status’). These words are undefined and they are not introduced or accompanied by any words of explanation. In those circumstances it is in my judgment reasonably arguable that on their true interpretation those words bear the meaning that the parties in common gave them in their communications leading up to the signing of the agreement. In admitting evidence as to those communications, the court would be hearing that evidence not with a view to taking the parties’ subjective intent into account for the purposes of interpretation (a purpose precluded by the principles laid down by Lord Hoffman in the ICS case) but for the purpose of identifying the meaning that the parties in effect incorporated into their agreement in circumstances where the court was satisfied that on their true interpretation the terms of the agreement were to have this effect.”

Richards LJ said that it was not appropriate at that stage to express any concluded view on the extent to which pre-contract negotiations can be taken properly into account in that case for the purpose of ascertaining the meaning of the contract.

At trial Cresswell J found that there had been no agreed meaning: [2007] EWHC 1621 (QB).

The other most directly relevant decision at first instance is Jones v Bright Capital Ltd [2006] EWHC 3151 (Ch). The issue was as to the meaning of a paragraph in a defence in a claim about pension rights, which had been incorporated as a term of the settlement of those proceedings. Sir Andrew Morritt C referred to the argument for the claimant that correspondence was admissible to show that the parties had used words in a special sense or to show the commercial business genesis or object of the agreement. He cited Prenn v Simmonds, Investors Compensation Scheme Ltd v West Bromwich Building Society, and The Karen Oltmann, and went on (at [24]):

“In my view each of the letters in question is admissible on the issue of construction. They show the genesis and subject matter of paragraph 38(3) of the defence which became a term of the Compromise. They show the connection between the actuary’s calculations and that paragraph and explain the figures and other terms which appear in it. None of them is relied on as indications of subjective intention and on the face of them they are not objectionable on that account. The mere fact that they were written in the course of inter-solicitor correspondence seeking to agree a redundancy package is not, in my judgment, a sufficient objection. ”

He then went on to say that none of the letters added a great deal on the issue of construction (at [25]), but (at [41]) used them to determine what he described as “the subject matter of discussion” to show that what the parties were negotiating about was extra scheme pension (attracting pension increases) rather than a flat rate annual payment. A literal reading of the document would not have led to that result. This was not a “private dictionary” case. The use of the correspondence was justified on the basis that it showed the genesis and object of the provision and provided a ground for treating the parties as having negotiated on an agreed basis.

My conclusions on this aspect of the appeal are as follows. First, this court is bound to accept the exclusionary rule as the starting point. Second, the policy reasons for a strict application of the rule are not compelling, and I do not accept the judge’s emphasis on the effect on third parties as a strong policy reason for the rule. Third, the “private dictionary” and “agreed basis” exceptions established in The Karen Oltmann are open to the court. Fourth, for the same reasons as I gave in the previous section of this judgment, I do not consider that the application of these exceptions is excluded simply because the words at issue are themselves contained in a definition section. Chartbrook makes the point that the judge has found in the rectification claim that there was no common intention, and that it would be wrong by introducing the documents (and only the documents) in the guise of interpretation, to find that there had been an agreement. That in itself would not be conclusive because subjective intention is not relevant under the head of construction, and in any event all that the judge did was to conclude, with evident hesitation, that Persimmon had not proved its rectification claim with sufficient clarity.

The pre-contract material points very strongly in favour of Persimmon’s construction, and if there were no limitations on its admissibility, it would not fall short of being determinative on the construction issue. But I do not consider that this is a case for the application of the “private dictionary” or “agreed basis” exceptions because the pre-contract negotiations do not establish a sufficiently clear evidential basis for them. There is no possible basis for a “private dictionary” approach, and to look for “an agreed basis” in the correspondence and the oral negotiations would be to eliminate any distinction between construction and rectification.

There is no dispute on the applicable legal principles. The burden is on the party seeking rectification, and he must put forward convincing proof.

The court will rectify a contract if the evidence is clear and unambiguous that a mistake has been made in the recording of the parties’ intention, what that intention was, and that the alleged intention continued in both parties’ minds down to the time of the execution of the agreement: Swainland Builders Ltd v Freehold Properties Ltd [2002] 2 EGLR 71 at 74, [33].

The burden is particularly onerous where there have been prolonged negotiations between the parties eventually assuming the shape of a formal instrument on which they have been advised by skilled lawyers: Crane v Hegeman-Harris Co Inc [1939] 1 All ER 662 at 664-5; and also Snamprogetti Ltd v Phillips Petroleum Ltd [2001] EWCA 889 at [36].

In cases where common mistake is alleged, it is necessary to establish not merely a continuing common intention, but also some outward expression of that prior accord: Joscelyne v Nissen [1970] 2 QB 86 at 97-98. The requirement of an outward manifestation of accord has been said to be an evidential factor rather than a strict legal requirement: Beasley v Munt [2006] EWCA Civ 370 at [36], per Mummery LJ. The claimant does not have to meet more than the civil standard of balance of probabilities, but convincing proof is required to counteract the cogent evidence of the parties' intention displayed by the instrument: Thomas Bates and Sons v Wyndham's Ltd. [1981] 1 WLR 505, 521 (CA).

Where only one party is mistaken, a species of equitable estoppel precludes a person who knows of the other party’s mistake from resisting rectification: Snell, Equity, 31^st ed 2005, para 14-15(2). Slade LJ described it as a “drastic remedy” in The Nai Genova [1984] 1 Lloyd’s Rep 353 at 365. In George Wimpey UK Ltd v VIC Construction Ltd [2005] EWCA Civ 77, [2005] BLR 135 Peter Gibson LJ referred to the “exceptional jurisdiction to rectify for unilateral mistake” (para 51). The requisite knowledge is actual, including for that purpose “blind eye”, knowledge: Commissioner for the New Towns v Cooper (Great Britain) Ltd [1995] Ch 259; George Wimpey UK Ltd v VIC Construction Ltd [2005] BLR 135, per Blackburne J at [79] (close to dishonesty).

I should also emphasise at this point, because so much of the judge’s conclusions rest on his assessment of the witnesses, that it has often been said that when a trial judge’s estimate of a witness forms any substantial part of his judgment, his conclusions of fact should be left alone by the appellate court. It is only if satisfied that the trial judge has not taken proper advantage of having seen and heard the witnesses that an appellate court is justified in interfering: e.g. Assicurazioni Generali SpA v. Arab Insurance Group [2002] EWCA Civ 1642, [2003] 1 WLR 577 at [14]-[23], approved in Datec Electronic Holdings Ltd v United Parcels Services Ltd [2007] UKHL 23, [2007] 1 WLR 1325, at [46].

It is important to preface the summary of the judge’s conclusions by his assessment of Chartbrook’s witnesses. The judge found that Mr Reeve was an impressive witness, and said that he would be firmly disposed to accept his evidence as honest unless the documents or the probabilities compelled the judge to do otherwise. The evidence of Mr Vantreen was on its face realistic and credible, and the judge found himself inclined to accept him as an honest witness, unless compelled by documents or probabilities to do otherwise. Mr Skelly was a careful and realistic witness, and his evidence was honest, reliable and helpful. The judge was less impressed by Persimmon’s witnesses, especially Mr Assael, but as there is no challenge to his factual findings with regard to Persimmon, his assessment of those witnesses is not material on this appeal.

The essential steps in the judge’s findings are these. First, the judge accepted Persimmon’s case that it always intended that the essential structure of the price formula was that Chartbrook should receive either a fixed percentage of the net sales revenue, or a minimum guaranteed amount, whichever proved to be the greater. It was supported by the evidence of all the witnesses called on its behalf, and by a fair reading of the contemporaneous documents generated by Mr Pendlebury, in particular his letters to Mr Reeve and his reports to his superiors.

Second, both Mr Pendlebury and Mr Assael made an extraordinary mistake in failing to appreciate that the definition of ARP in all the drafts of what became Schedule 6 provided clearly for something very different from that which they intended.

Third, the evidence of Mr Vantreen and Mr Reeve was that they believed in February 2001 (i.e. before they saw the first draft of the Agreement in March 2001) that the super overage would produce a further £3 million or more in excess of the guaranteed minimum even if the market remained flat throughout the development period, and indeed even in the event of a catastrophic market fall. But it was difficult to identify where they got that idea from. It was difficult to see how such a belief could have been derived from any of Mr Pendlebury’s three February 2001 letters, or from their effect in aggregate.

Fourth, Mr Vantreen gave evidence (which the judge seems to have found persuasive) that he could not possibly have believed that (assuming a flat market) Persimmon were offering only £1.1 million over their stated opinion of current market value as the quid pro quo for Chartbrook having to wait four or more years for its money. But there was no obvious answer to the point that if Persimmon had been offering any more by way of uplift, it would have been advertised in no uncertain terms in the February letters.

Fifth, Mr Vantreen and Mr Reeve either thought in February that the super overage was on offer, or they understood the offer as the judge found that Persimmon intended. There was no credible middle ground.

Sixth, the ultimate question was whether Persimmon had proved its case, rather than whether Chartbrook had proved the contrary.

Seventh, it was difficult to imagine how Mr Reeve and Mr Vantreen could for the first time have stumbled upon the true meaning of ARP between exchange of contracts on October 16 and mid December 2001. In particular, the absence of any reference to the super overage in the correspondence with banks in August 2001 was not a significant pointer to Chartbrook then being unaware of the true meaning of ARP.

Eighth, Mr Vantreen’s manuscript notes on Mr Pendlebury’s letter of May 24, 2001 constituted a persuasive piece of evidence towards a conclusion that Chartbrook had not by May 2001 focused upon the true meaning of ARP. But two manuscript figures, for which it would be unlikely for anyone now to recall the purpose, afforded a slender basis for firm conclusions as to Chartbrook’s then intentions, in the absence of corroboration, and in the face of persuasive evidence to the contrary.

Ninth, the letter of May 24, 2001 afforded some support to an inference of a perception on Chartbrook’s part that Persimmon’s approach had not changed from February, notwithstanding the terms of the draft definition of ARP.

Tenth, there was force in Mr Vantreen’s evidence that if he had shared Persimmon’s intention as to the formula for the Balancing Payment in May, then his and Mr Reeve’s acceptance of the reduction in the percentage from 29.8% to 23.4% would be even more extraordinary than their acceptance of the February offer as intended by Persimmon. Although the amount offered by way of minimum guaranteed payment had risen, this was attributable to the addition of Units 1 and 3, which Chartbrook had to buy in. The value of the price per square foot was being substantially reduced.

Eleventh, he accepted Mr Skelly’s evidence that on March 20, 2001 he had explained the first draft of what became Schedule 6 was explained to Mr Vantreen and Mr Reeve, and that was the most probable date upon which, if later than in February, Chartbrook first became aware that an additional payment of the type reflected in Chartbrook’s construction of the Agreement was on offer.

Twelfth, the consequence would have been that if that was the date when Mr Vantreen and Mr Reeve first became so aware: (1) they must have realised that Persimmon, or at least its solicitor Mr Assael, had made a serious mistake, contrary to Persimmon’s interest, and that Mr Pendlebury had not spotted it. (2) Thereafter they must have concealed their realisation of Persimmon’s mistake not only from Persimmon, but also from their own solicitor Mr Skelly, from their finance broker and from RBS when referring to their expectation of the amount of the balancing payment in December 2001. (3) The whole of their evidence would be revealed as not merely the consequence of imperfect recollection, but as a tissue of lies.

The judge said that he was faced with only two alternatives. The first was that two apparently upright and experienced businessmen made an extraordinary mistake when reading Persimmon’s letters of offer in February (and again in May), for reasons which they cannot now remember, despite their best honest endeavours. The second was that they were cool risk takers, secretive from their own lawyers, dishonourable to their business counterparties for their own financial gain, and skilled and sophisticated liars on oath. In short, thoroughgoing rogues.

Neither of those alternatives was, viewed separately, probable, but there was no convenient middle ground, and none was offered. Forced to choose between two improbable alternatives, the requirement in rectification cases for convincing proof was of particular importance. The judge concluded:

“I have after considerable difficulty come to the conclusion that Persimmon has not proved either of its alternative rectification cases with sufficient clarity. I have real difficulty in understanding how Messrs Reeve and Vantreen made the mistake of thinking that Persimmon were offering in February what the first draft definition of the ARP clearly offered in March, but I have had to overcome similar difficulty in understanding how both Mr Assael and Mr Pendlebury failed to understand what that definition meant when they respectively framed and checked it.

But those difficulties are of lesser force than the difficulty of persuading myself that Mr Reeve and Mr Vantreen are the rogues which they would have to be if the necessary factual basis for Persimmon’s rectification case were to be accepted. I had the advantage of seeing both of them cross examined at considerable length and with great skill. I have not been persuaded that their evidence should be rejected as incredible, where to do so would involve a conclusion as to their honesty and probity wholly at variance with the impression which they made on me. In this Mr Skelly was an equally powerful support, and each of them gave evidence in the absence of hearing the evidence (or reading the transcript of the evidence) of those who preceded them. It follows that I dismiss the counterclaim.”

The main points made by Persimmon are these. First, the judge should have rejected the evidence of Mr Reeve and Mr Vantreen that they understood Persimmon to be offering the super overage in February 2001, from the outset and before they saw Mr Assael’s draft. Neither Mr Reeve nor Mr Vantreen could explain where they got the super overage idea from.

Second, neither of them could give a plausible account as to how they had come to believe before the end of February 2001 that Persimmon was offering them super overage. The only explanation that was offered by them (that they had somehow carried over the idea from an overage offered in January) was no explanation at all, as the structure and effect of the January overage was radically different from the supposed February super overage.

Third, they had no explanation for the implausibility of Persimmon offering them another £3 million without flagging it up to them anywhere.

Fourth, not a single document from the contemporaneous documentation prior to the signing of the Agreement referred to or in any way reflected a belief or understanding by Mr Reeve or Mr Vantreen that what was being offered was super overage. All the contemporaneous documents were consistent with Mr Reeve and Mr Vantreen not having so understood, and some were only readily explicable on this basis.

Fifth, in particular Mr Vantreen’s calculations on a copy of the letter of May 24, 2001 for which he had no explanation. They were only capable of one explanation. If Mr Vantreen was calculating the trigger level for the overage at £325 per square foot in May, he and Mr Reeve could not have believed that Persimmon was offering super overage in February.

Sixth, the obvious answer is that they got the idea of super overage from the unfortunate way in which Mr Assael drafted the definition of ARP and so they could not have got to it before March 2001 at the earliest. If they had the super overage independently from the drafting mistake by Mr Assael, that would be an incredible coincidence.

Seventh, the suggestion that Mr Reeve and Mr Vantreen went through the overage provisions of the draft agreement with Mr Skelly at the meeting of March 20, and that he specifically confirmed their understanding, is inconsistent with his attendance note which makes it clear that he left them to check the price against Persimmon’s letters.

Persimmon relies also on these matters: (1) Mr Skelly and Mr Pendlebury recorded a concern on the part of Mr Reeve and Mr Vantreen that discounted bulk sales by Persimmon might remove altogether Chartbrook’s overage. This concern is consistent with Persimmon’s analysis of the overage and wholly inconsistent with Chartbrook’s under which they would obtain a super overage even if the flats sold for half their anticipated price or even less. (2) Mr Vantreen asked Mr Pendlebury to prepare a letter for him to assist in negotiating the Enhanced Value payment due under the option to the owner of Unit 5. If the letter was understood by Mr Vantreen to be offering a huge additional overage payment, he would not have asked Mr Pendlebury to repeat them in the form he did including the percentages of sales revenue. (3) Chartbrook wrote to three banks in August 2001 seeking funding on the basis of the sums due under the agreement, but to none of them did it make any reference at all to the expectation of super overage. (4) In late August someone took the step of suggesting that the words “if any” be inserted into clause 3.3 of Schedule 6 to recognise the fact that the “Balancing Payment” might not occur. This amendment is consistent only with the parties sharing Persimmon’s intention in August 2001.

Chartbrook says that that there are no grounds for interfering with the judge’s conclusion. It relies in particular on these matters. First, it would have been inconceivable to Mr Reeve and Mr Vantreen that Persimmon might seriously have offered an uplift of less than a £1 million in return for waiting for up to 5 years. Second, the evidence showed that Mr Vantreen and Mr Reeve did not know that Persimmon did not mean to contract on the very terms set out in the contract. Third, they understood the first draft of the contract as producing a significant overage at the time they went through it with Mr Skelly on March 20, 2001, and there were no significant changes to the wording thereafter. Fourth, if they realised that Persimmon was labouring under a mistake, they would have had to deceive Mr Skelly from March onwards. Fifth, the manuscript notes made by Mr Vantreen on the letter of May 24, 2001 do not show that he was trying to work out the trigger. The calculation is equally consistent with him trying to see what Persimmon expected to achieve for the flats. The judge concluded that this was a slender basis for rejecting the rest of the evidence. But this was in fact unduly favourable to Persimmon.

Persimmon had, in my judgment, a very powerful case for rectification. But this was not a case where the judge overlooked the points in their favour. This was a conspicuously careful and full judgment dealing with all the important points put forward on either side.

Persimmon’s strongest points were these. There was never any adequate explanation of where Mr Reeve or Mr Vantreen got the super overage idea from, or how it was they understood Persimmon offering them another £3 million without flagging it up to them anywhere. There was no document apart from the drafts of the Agreement which reflected what they said was their understanding.

These are powerful points, but the most powerful point in my judgment arises from Mr Vantreen’s calculations on a copy of the letter of May 24, 2001 for which he had no explanation. As I have said, it contained an important annotation in Mr Vantreen’s handwriting. At the top of its second page, which contained the two tables, he wrote in manuscript “£325.p.f.s” and, a little to the right, “£228,000.”

The judge’s findings on this evidence are important. In cross examination, Mr Vantreen accepted, although he could not recall doing so, that he probably derived each of those figures from, respectively, the Private Sale Residential Accommodation Land Value per square foot of £76.34 and the Minimum Value per plot of the Residential Apartments of £53,333. In each case he arrived at his manuscript figures by grossing up the two stated amounts in the tables from 23.4% to 100%. It was put to him in cross examination that he did so in order to calculate the sale price which the residential units as a whole would have to exceed in terms of pounds per square foot and the sale price which each flat would, on average, have to achieve, before any Balancing Payment would become available to Chartbrook above the minimum guaranteed amounts. While he did not accept this, Mr Vantreen could not think either in cross-examination or re-examination of any other reason for his having made that calculation.

The judge said that it was a very relevant calculation if Persimmon’s construction of ARP represented the parties’ common intention, but of no obvious relevance on the construction advanced by Chartbrook.

There is therefore very considerable force in Persimmon’s submission that the figures were only capable of one explanation. If Mr Vantreen was calculating the trigger level for the overage at £325 per square foot in May, he and Mr Reeve could not have believed that Persimmon was offering super overage in February. The judge said that although a persuasive piece of evidence, this was a “slender basis” for rejecting Chartbrook’s case in the face of persuasive evidence to the contrary (para 155).

I have no doubt that, whatever impression he may have got from the witnesses in the witness box, this was a very important piece of evidence and that he gave it insufficient weight. But for one matter to which I have to revert, I would have found that this was sufficiently important, taken together with the other matters on which Persimmon relied, to put this case among those rare cases where an appellate court is entitled to interfere with the judge’s findings on credibility. If the evidence of Mr Vantreen and Mr Reeve were to be rejected, then a plausible hypothesis might have been that they got the idea in December 2001 from Eversheds’ report to the Royal Bank of Scotland, which was sent to Mr Skelly and which shortly preceded Mr Reeve’s first assertion of what is now Chartbrook’s case on December 10, 2001, and then convinced themselves that that had been the intention all along.

But there is a stumbling block to that conclusion which is, in my judgment, decisive. As I have said, the judge found that Mr Skelly was a careful and realistic witness, and that his evidence was honest, reliable and helpful.

The evidence was that Mr Assael prepared a second draft of the Agreement and circulated it both to his clients and to Mr Skelly on March 13, 2001. At a meeting on March 20, 2001 Mr Skelly took Mr Vantreen and Mr Reeve through the draft in detail, but his attendance note suggested that he did not take them through the Schedule. I have already set it out, but I repeat it for convenience:

“We then dealt with the Agreement by going through it line by line, flagging up those matters which would need discussion tomorrow with Persimmon.

The last point left was the Seventh Schedule relating to calculation in price, as I have stated they have negotiated this with Persimmon and had Persimmon’s letters relating its calculation and they agreed they would themselves go through that Clause to confirm that the construction of the price and payment was agreed.”

The judge’s findings about this meeting are important. He recorded that Mr Reeve did not recall the meeting in sufficient detail to be able to say whether the Schedule had been considered line by line. In his witness statement Mr Vantreen said that this did happen, but in cross-examination he was not sure whether Mr Skelly took him through the Price schedule earlier than in October, shortly before execution.

But the judge recorded that Mr Skelly was in no doubt that he did take both of them line by line through the Schedule, explaining its meaning in the terms for which Chartbrook contended. He said that during the meeting Mr Vantreen and Mr Reeve confirmed by way of instructions to him that the Price was agreed to be as he had explained by reference to the Schedule. Mr Skelly said this both in his witness statement and in cross examination.

Mr Skelly explained the paragraph in his attendance note which described the Schedule as being left for his clients to go through, not as suggesting that he did not explain its meaning, but as recording that, because of its importance to Chartbrook, his clients should check carefully before the meeting next day with Persimmon, against the letters from Mr Pendlebury (which had by then been copied to Mr Skelly), whether the meaning which he had explained was what had been agreed.

The judge refused an invitation by Persimmon to reject that evidence, as being irreconcilable with the contemporaneous attendance note. He was not persuaded on that ground to reject the clear evidence of a careful and impressive witness. Although the attendance note was contemporaneous, it was (as Mr Skelly said) designed as an aide memoire, ahead of a meeting with Persimmon and its lawyers the next day, rather than as a precisely formulated statement of evidence. Having heard Mr Skelly, the judge considered that the import of the relevant paragraph in the note was to remind Mr Skelly of a piece of unfinished business to be carried out by his clients, namely to check, with the assistance of the correspondence, whether they were able to confirm that the important terms of the Schedule were in all respects as agreed. The unfinished business did not include his clients going through the Schedule on their own for the first time, to work out what it meant.

If that holding stands, in my judgment it disposes of the case on both common and unilateral mistake, because if it is believed, then for there to remain a case on mistake, the court would have had to accept the unlikely hypothesis that Mr Vantreen and Mr Reeve must not only be taken to have known that Mr Skelly’s interpretation was inconsistent with the intentions of both parties to the deal, but also to have kept up a pretence until its conclusion.

That is why to succeed on rectification Persimmon must persuade the court to overturn that holding. It seeks to do so by an argument that there was nothing in Mr Skelly’s attendance note to support that account; and that it is understandable that a busy solicitor for whom this was one transaction among many might after some six years be unable to have a clear recollection of a particular meeting. Persimmon says if he went through the price schedule with Mr Reeve and Mr Vantreen as he thought he had, it was more likely in the light of the other evidence to have been later rather than earlier.

I could well understand if the judge had decided that the combination of the wording of the attendance note and the lapse of time since the meeting justified a conclusion that Mr Skelly was simply mistaken (but not lying) about the meeting. But the judge expressly rejected a finding to that effect. He dealt with Persimmon’s argument that Mr Skelly’s evidence was inconsistent with his attendance note. He found that the note was simply an aide memoire, and was intended to remind him to check with his clients that they were able to confirm that the important terms in the Schedule were as agreed by the parties. The judge made a specific finding that Mr Skelly explained the Schedule in the sense for which Chartbrook has contended, and he did so in the light of his finding that Mr Skelly was a credible witness who took great trouble to distinguish between actual recollection and reconstruction from documents. This is in my judgment not a finding with which this court, in accordance with established principle, should interfere.

I would therefore allow the appeal on the limited ground which I have identified.

Since Tuckey and Rimer LJJ have come to a different view on the construction point, the appeal will be dismissed.

I have read Lawrence Collins LJ’s judgment in draft and gratefully adopt his account of the facts, issues and submissions. He has explained why, differing from the judge, he accepts Persimmon’s suggested interpretation of the definition in Schedule 6 to the October 2001 agreement of the “Additional Residential Payment.” With respect, I am unable to agree. In my judgment, the judge was correct to reject that interpretation.

The definition provides:

“’Additional Residential Payment’ [ARP] means 23.4% of the price achieved for each Residential Unit in excess of the Minimum Guaranteed Residential Unit Value [MGRUV] less the Costs and Incentives [C&I]”

There is nothing unclear, uncertain or ambiguous about that. It is clear, certain and unambiguous and its arithmetic is straightforward. The phrases which are elsewhere defined are accepted as meaning what their definitions provide, and no other word or phrase is said to bear any meaning other than its ordinary one. Assuming a sale for £400,000 with C&I of £75,000 of a residential unit whose MGRUV is £50,000, ARP = 23.4% x (400,000 – 50,000 – 75,000), or £64,350. If there are no C&I and the unit is sold for £325,000, ARP = 23.4% x (£325,000 – 50,000), or £64,350, the same result.

Persimmon asks for the definition to be read as if it were drafted thus:

“[ARP]” means the amount (if any) by which 23.4% of the price achieved for each Residential Unit is in excess of the [MGRUV] less the [C&I].”

That change of language fundamentally distorts the meaning and arithmetic of the definition. Applying those changes to the first of the above examples, ARP = (23.4% x 400,000) – 50,000 – 75,000, or minus £31,400. In the second example, ARP = (23.4% x 325,000) – 50,000, or £26,050. Thus, in addition to the main change, the payment to Chartbrook will vary according to whether C&I are offered: Persimmon could reduce the share payable to Chartbrook by offering C&I and increasing the price by the same amount. As the judge said, one would expect Chartbrook’s profit share to be unaffected, one way or the other, by Persimmon’s decision as to whether to sell at a low price without incentives or at a higher price with incentives. Persimmon had no answer to this consequence of its re-writing of the definition.

I can see no basis for re-writing the agreement as invited by Persimmon. I accept that it is legitimate in interpreting the October 2001 agreement to take account of the land values in 2001, which is part of the background against which the agreement was made. On that basis I recognise the commercial force of the argument that Persimmon attaches to the “if any” point and to the points based on the use of various words (“minimum” and “guaranteed”) in the Schedule 6 definitions, to which Lawrence Collins LJ has referred. But I am quite unable to regard these points as doing the fundamental interpretative work that Persimmon requires of them, not least because the use of the relevant language is strictly consistent also with Chartbrook’s construction.

I would reject any suggestion that this is a case in which it is legitimate, as part of the construction exercise, to have recourse to the pre-contract negotiations. The basic rule is that they are out of bounds (Prenn v. Simmonds [1971] 1 WLR 1381; Investors Compensation Scheme Ltd v. West Bromwich Building Society [1998] 1 WLR 896, at 913 per Lord Hoffmann). I acknowledge that Lord Hoffmann there recognised that the boundaries of this exception are in some respects unclear. But to the extent that any inroads into it may legitimately be made, I have not been persuaded that any have been made good in this case. If, and to the extent that, recourse can legitimately be had to prior negotiations to unearth any use by the parties of a “private dictionary” said to be explanatory of their contractual language, I agree with Lawrence Collins LJ that there is no scope for it here. I also agree with him that there can be no justification for an attempt to invoke the course of the negotiations for the limited purpose of identifying “an agreed basis” for the transaction. Persimmon’s purpose in going into the archaeology of the transaction is not to derive assistance in the interpretation of the ARP definition, for which there is no need. It is to seduce the court into accepting that the parties’ subjective intentions with regard to the ARP calculation were different from what the ARP definition in the agreement actually provides, and then to invite the interpretation of that definition in a way that is in line with the alleged intentions. In short, the bid is to have recourse to the negotiations for the purpose of rectifying the ARP definition under the guise of interpretation. It in fact has nothing to do with the interpretation of that definition as included in the October 2001 agreement. As Lord Hoffmann made clear in Investors Compensation Scheme, pre-contract material of the nature on which Persimmon seeks to rely can legitimately be invoked only for the purposes of a claim for rectification.

Perhaps the most that can be said on the issue of construction is that, looking at land values as at 2001, the contractual terms seem improbable ones for Persimmon to have signed up to. If so, the explanation is either (i) that it made a bad bargain, or (ii) that it may have made a sensible one but the written agreement recorded it wrongly. If the former, Persimmon is stuck with its bargain, and it is not the court’s function to reform it. If the latter, Persimmon may have a claim to have the agreement rectified.

Persimmon did ask to have the agreement rectified. It had what appears to me to have been a powerful case. The judge, however, dismissed it for the reasons he gave; and, despite Mr Nugee’s extremely skilful and cogent challenge to that decision, to which I would pay my tribute, Lawrence Collins LJ has given his reasons why he would dismiss the appeal against it. I respectfully agree with those reasons, to which I can add nothing useful of my own.

I would dismiss the appeal.

Like Rimer LJ I agree with Lawrence Collins LJ’s conclusion that the appeal against the judge’s rejection of the claim for rectification should be dismissed and his conclusions about the admissibility of negotiations for the reasons he has given. However they disagree about construction.

Having considered the matter afresh in the light of this disagreement I remain of the view which I had formed by the end of the hearing that the judge’s construction was right for the reasons which he and now Rimer LJ have given.

The words used to define ARP are entirely clear. Persimmon’s construction requires the definition to be re-written, not only as Rimer LJ has shown, but also to deal with the striking anomaly about the impact of C+I which he illustrates in para 185 above. So the clause would now have to be read as if it had been drafted:

ARP means the amount (if any) by which 23.4% of the price achieved for each Residential Unit less the [C+I] is in excess of the [MGRUV] less the [C+I].

Short of a good claim for rectification I do not think it is possible to make such radical changes to the clear words used in the agreement by invoking the forces of commercial good sense and hints from other parts of the agreement that Chartbrook would not inevitably have been entitled to ARP. They simply are not up to the major interpretive task for which Persimmon invokes them.

For these reasons I agree with Rimer LJ that Persimmon’s appeal should be dismissed in its entirety.